AIDS 2006 prelim matter 15/8/06 15:44 Page i
5th International Conference
29 August – 1 September 2006 Amsterdam, Netherlands
Abstract book
The correct citation for an abstract is: Authors. Title. Antiviral Therapy 2006; 11 (Suppl 2): Abstract number.
Photograph of ‘Gables of Amsterdam Canal Houses’ provided by Tim Killam, www.cityboek.nl AIDS 2006 prelim matter 15/8/06 15:44 Page ii
ANTIVIRAL THERAPY
EDITORS-IN-CHIEF EMERGING VIRUSES AND BIODEFENCE PRE-CLINICAL HIV Clarence J Peters Amalio Telenti Joep MA Lange University of Texas Medical Branch Institute of Microbiology, CHUV Center of Poverty-related Communicable Diseases 3.146 Keiller Building 1011 Lausanne, Switzerland Academic Medical Centre 301 University Boulevard Fax: +41 21 314 4095 University of Amsterdam, Meibergdreef 9 (T-125) Galveston, TX 77555-0609, USA 1105 AZ Amsterdam, the Netherlands Fax: +1 409 747 2429 PAPILLOMAVIRUSES Tel: +31 20 314 9300 William Bonnez Fax: +31 20 314 9399 HEPATITIS VIRUSES Infectious Diseases Unit - Box 689 E-mail: [email protected] Stephen Locarnini University of Rochester Medical Center Victorian Infectious Diseases Reference 601 Elmwood Avenue Douglas D Richman Laboratory Rochester, NY 14642, USA University of California, San Diego WHO Collaborating Centres for Virus Reference Fax: +1 585 442 9328 Departments of Pathology and Medicine 0679 and Research, and Biosafety 9500 Gilman Drive, La Jolla 10 Wrekyn Street, North Melbourne 3051, RESPIRATORY VIRUSES CA 92093-0679, USA Victoria, Australia Frederick G Hayden Tel: +1 858 552 7439 Fax: +61 3 9342 2666 Department of Internal Medicine Fax: +1 858 552 7445 Box 473, University of Virginia E-mail: [email protected] CLINICAL HIV Health Sciences Center, Charlottesville Peter Reiss VA 22908, USA National AIDS Therapy Evaluation Centre Fax: +1 434 924 9065 SECTION EDITORS Academic Medical Centre University of Amsterdam, Pietersbergweg 9 HERPESVIRUSES 1105 BM Amsterdam, the Netherlands Richard J Whitley Fax: +31 20 314 9399 University of Alabama at Birmingham School of Medicine, Division of Clinical Virology Joe Eron 616 Children’s Hospital University of North Carolina at Chapel Hill 1600 7th Avenue South, Birmingham 211a W Cameron St, CB 7030, APCF AL 35233-0001, USA Chapel Hill, NC 27599-7030, USA Fax: +1 205 934 8559 Fax: +1 919 966 1576
EDITORIAL BOARD
Fred Y Aoki (Winnipeg, Canada) José M Gatell (Barcelona, Spain) Luc Perrin (Geneva, Switzerland) Lawrence Banks (Trieste, Italy) Lutz Gissmann (Heidelberg, Germany) Thierry Poynard (Paris, France) Karen K Biron (Research Triangle Park, NC, USA) Scott M Hammer (Boston, MA, USA) Jonathan M Schapiro (Tel Hashomer, Israel) Michael Bray (Bethesda, MD, USA) Martin Hirsch (Boston, MA, USA) Raymond F Schinazi (Atlanta, GA, USA) David M Burger (Nijmegen, the Netherlands) Stanley M Lemon (Galveston, TX, USA) Robert T Schooley (Denver, CO, USA) Jeffrey I Cohen (Bethesda, MD, USA) Yun-Fan Liaw (Taipei, Taiwan) John J Treanor (Rochester, NY, USA) Ann Collier (Seattle, WA, USA) Susan J Little (San Diego, CA, USA) Stefano Vella (Rome, Italy) Robert B Couch (Houston, TX, USA) Anna Lok (Ann Arbour, MI, USA) Patrick Yeni (Paris, France) Janet Englund (Houston, TX, USA) Michael Peter Manns (Hannover, Germany) Delia Enria (Pergamino, Argentina) Patrick Marcellin (Clichy, France) Michael Gale (Dallas, TX, USA) Julio SG Montaner (Vancouver, Canada)
EDITORIAL OFFICE
MANAGING EDITOR International Medical Press Tel: +44 20 7398 0700 Stephen Cameron 36 St Mary at Hill Fax:+44 20 7398 0701 London EC3R 8DU E-mail: [email protected] MEDICAL EDITOR UK http://www.intmedpress.com Oliver Childs PRODUCTION EDITORS Emdadur Rahman, Will Fox
ii Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page iii
CONTENTS
Sponsors iv
Committees v
Programme at a Glance vi
Abstract Listing ix
Abstracts 1
Notes
AIDS Vaccine 2006 iii AIDS 2006 prelim matter 15/8/06 15:44 Page iv
Main Sponsors The organizing committee of the AIDS Vaccine 2006 Conference would like to acknowledge the following sponsors for their support:
National Institutes of Health/Office of AIDS Research www.nih.gov
Bill & Melinda Gates Foundation www.gatesfoundation.org
National Institute of Allergy and Infectious Diseases www.niaid.nih.gov
Dutch Ministry of Foreign Affairs www.minbuza.nl
The World Health Organization (WHO) and the Joint United Nations Programme on HIV/AIDS (UNAIDS) through their WHO-UNAIDS HIV Vaccine Initiative (HVI) www.who.int/vaccines/givs
Centers for Disease Control and Prevention www.cdc.gov
Other Sponsors AIDS Fonds, The Netherlands www.aidsfonds.nl
French National Agency for Research on AIDS and Viral Hepatitis www.anrs.fr
International AIDS Vaccine Initiative (IAVI) www.iavi.org
Sanofi Pasteur www.sanofipasteur.com
F. Hoffmann-La Roche Ltd www.roche-hiv.com
Municipality of Amsterdam www.amsterdam.nl
GlaxoSmithKline www.gsk.com
Elizabeth Glaser Pediatric AIDS Foundation www.pediaids.org
The Wellcome Trust www.wellcome.ac.uk
Crucell N.V. www.crucell.com
The Merck Research Laboratories www.merck.com/mrl
iv Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page v
ORGANIZING COMMITTEE
Ben Berkhout Deborah Birx Susan Buchbinder José Esparza Jaap Goudsmit Jonathan Heeney Peggy Johnston Pontiano Kaleebu Michel Kazatchkine Marie-Paule Kieny Wayne Koff Joep Lange (Chair) Peter Liljeström Marta Marthas Timothy Mastro Bonnie Mathieson Frank Miedema Gary Nabel Helen Rees Praphan Phanuphak Hanneke Schuitemaker Rafick-Pierre Sékaly Mitchell Warren Yiming Shao
SCIENTIFIC PROGRAMME COMMITTEE
Alash'le Abimiku Rafi Ahmed Brigitte Autran Gunnel Biberfeld Dennis Burton Sal Butera David Cooper Larry Corey Anthony DeVico Clive Gray Barbara Ensoli Pat Fast Thomas Folks Genoveffa Franchini Beatrice Hahn Nancy Haigwood Scott Hammer Barton Haynes Linda Klavinskis David Knipe Bette Korber Richard Koup Norman Letvin Yves Levy Jeffrey Lifson Fred Mhalu John Mascola Souleymane Mboup Francine McCutchan Julie McElrath David Montefiori Lynn Morris Saladin Osmanov Julie Overbaugh Steve Patterson Giuseppe Pantaleo Louis Picker Punnee Pitisuttithum Nina Russell Jeffrey Safrit Quentin Sattentau Gabriella Scarlatti John Shiver Charles Vitek Robin Weiss Carolyn Williamson Hans Wolf
LOCAL PROGRAMME COMMITTEE
Frans van den Boom Roel Coutinho Sam Gobin Ab Osterhaus Kees Rümke Janneke van de Wijgert
CONFERENCE COORDINATOR
Jacqueline van Tongeren
AIDS Vaccine 2006 Conference Secretariat
The Conference Organisation of the AMC, Nicolaes Tulp Institute Academic Medical Center, University of Amsterdam
P.O. Box 23213, 1100 DS, Amsterdam, The Netherlands
E-mail: [email protected] Telephone: +31 20 566 8585 Fax: +31 20 696 3228 Website: www.aidsvaccine06.org
AIDS Vaccine 2006 v AIDS 2006 prelim matter 15/8/06 15:44 Page vi
PROGRAMME AT A GLANCE
Time Tuesday 29 August 2006
13.00 – Registration (Onyx Lounge) Mounting Posters (Diamond/Topaz Lounge)
15.30 – 17.45 Opening Session (Auditorium)
17.45 – 20.30 Canal Boat trip & Welcome Reception (Maritime Museum)
Wednesday 30 August 2006
08.30 – 10.00 Plenary Session (Auditorium)
10.00 -10.30 Coffee & Tea Break (Lounge Auditorium)
10.30 – 12.30 Oral Abstract 1 Oral Abstract 2 Oral Abstract 3 HIV Transmission/ Vaccine Antigen Prophylactic Vaccine Trials Acute HIV Infection/ Selection & Design (Room B) Innate & Mucosal Immunity (Room A) (Auditorium)
12.30 – 13.30 Lunch Break (Lounge Auditorium)
13.30 – 15.30 Symposium 1 Symposium 2 Basic Findings and their Novel Vectors and Impact on Vaccine Design Vector Combinations (Room A)* (Auditorium)
15.30 – 16.00 Coffee & Tea Break (Lounge Auditorium)
16.00 – 18.00 Symposium 3 Symposium 4 Streamlining Vaccine Development T Cell and Innate for HIV and Other Diseases Immunity (Auditorium) (Room A) *
18.00 – 19.00 Poster Presentations Topic 9 (A–D) (Topaz Lounge)
*The live presentation will be in Room A, overflow in Room B
vi Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page vii
Time Thursday 31 August 2006
08.00 – 10.00 Plenary Session (Auditorium)
10.00 – 10.30 Coffee & Tea Break (Lounge Auditorium)
10.30 – 12.30 Oral Abstract 4 Oral Abstract 5 Oral Abstract 6 T Cell Immunity B Cell Immunity Clinical Trial Site Development/ (Auditorium) (Room A) Social, Ethical, Access and Regulatory Issues (Room B)
12.30 – 13.30 Lunch Break (Lounge Auditorium)
12.30 – 13.00 Special Lunch Time Lecture (Room A)
13.30 – 15.30 Symposium 5 Symposium 6 Neutralizing Antibodies Animal Models and (Auditorium) Correlates of Immunity (Room A)*
15.30 – 16.00 Coffee & Tea Break (Lounge Auditorium)
16.00 – 18.00 Symposium 7 Symposium 8 T Cell Vaccine Epidemiology/ Development with focus Clinical Trials Issues on T Cell Responses (Room A)* (Auditorium)
18.00 – 19.00 Poster Presentations Topic 1-8 & Topic 10-14 (Diamond Lounge)
Friday 1 September 2006
08.30 – 10.00 Plenary Session (Auditorium)
10.00 – 10.30 Coffee & Tea Break (Lounge Auditorium)
10.30 – 12.30 Oral Abstract 7 Oral Abstract 8 Oral Abstract 9 Immune Escape Vaccine Delivery Animal Models/Live Attenuated (Auditorium) Methods/Adjuvants Vaccines/Therapeutic Vaccination (Room A) (Room B)
12.30 – 13.30 Lunch Break (Lounge Auditorium)
13.30 – 15.30 Symposium 9 Symposium 10 Large Scale Production Therapeutic Vaccine and Delivery Issues Trials (Auditorium) (Room A)*
15.30 – 16.00 Coffee & Tea Break (Lounge Auditorium)
16.00 – Closing Session (Auditorium)
*The live presentation will be in Room A, overflow in Room B
AIDS Vaccine 2006 vii AIDS 2006 prelim matter 15/8/06 15:44 Page viii AIDS 2006 prelim matter 15/8/06 15:44 Page ix
ABSTRACT LISTING
ORAL PRESENTATIONS
Page Session Title and presenting author Abstract Opening Session
Welcome by the chair of the conference OS01 Joep Lange
Welcome by the Mayor of Amsterdam OS02 Job Cohen
Opening by the Dutch Minister of Health, Welfare and Sport OS03 Hans Hoogervorst
Overview of the fight against AIDS and obstacles to HIV prevention OS04 HRH Princess Mabel van Oranje- Nassau
3 High noon – 20 global problems, 20 years to solve them OS05 Jean-Francois Rischard
Public perceptions on vaccines OS06 David Salisbury
An HIV vaccine development: will it ever fit the classical paradigm? OS07 Anthony S Fauci
Plenary Session Day 1
4 HIV prevention: what works, what’s needed and what’s being done PL01-01 Helen Rees
4 The human HIV vaccine pipeline: an update PL01-02 Larry Corey
4 The HIV-1 envelope glycoproteins: implications for vaccine design PL01-03 John Moore
Plenary Session Day 2
6 Global epidemiology of HIV-1: influence of risk factors and social PL02-01 networks on the complexity and relationships of inter-subtype recombinant strains in Asia and Africa Francine McCutchan
6 The challenge of HIV global diversity for vaccine antigen design PL02-02 Bette Korber
7 Toll-like receptor ligands influence the magnitude and quality of PL02-03 T cell responses Robert Seder
AIDS Vaccine 2006 ix AIDS 2006 prelim matter 15/8/06 15:44 Page x
Plenary Session Day 3
8 More than half the world’s population is under 25 and are all at risk for HIV... PL03-01 Linda-Gail Bekker
8 Optimizing trail designs to accelerate HIV vaccine clinical research: PL03-02 Phase IIb test of concept trails Pat Fast
8 Overview of the global HIV vaccine enterprise: progress and priorities PL03-03 for the future José Esparza
Closing Session
Summary Session CL01
Jeffrey Saffrit (Elizabeth Glaser Pediatric Foundation, Santa Monica, CA, USA) Anne Duerr (HVTN, Fred Hutchinson Cancer Center, Seattle, WA, USA) Carolyn Williamson (Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, South Africa) Jill Gilmore (International AIDS Vaccine Initiative, New York, NY, USA)
Announcement of the 2007 Conference and Closing CL02
Joep Lange Larry Corey
Symposium Session 1
Basic Research Findings and Their Impact on Vaccine Design
Controllers of HIV: implications for vaccines S01-01 Bruce Walker
13 Novel HIV-1 neutralizing antibodies recognizing conserved S01-02 conformational epitopes on gp-41 Dimiter Dimitrov
13 Do viral subtypes and mode of transmission define the envelope S01-03 characteristics of transmitted viruses? Eric Hunter
14 Transient viremia and immune responses preceding sustained, systemic S01-04 viral replication: implications for evaluating acute HIV infection and vaccine efficacy Phillip Norris
14 HIV –containing intracellular compartments mediate HIV transmission S01-05 from macrophages and DCS Mark Marsh
Symposium Session 2
Novel Vectors and Vector Combinations
15 Strategic development of novel vectors and criteria for S02-01 advanced clinical development Gary Nabel
15 Novel Adenovirus vector-based vaccines for HIV-1 S02-02 Dan Barouch
x Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xi
Live vector vaccines S02-03 Steve Udem
16 Development of CMV vectors for an HIV vaccine S02-04 Louis Picker
Novel rBCG and Shigella rdsRN vaccine vectors S02-05 Jerald Sadoff
Symposium Session 3
Streamlining Vaccine Development for HIV and Other Diseases
Strategic overview of state-of-the-art across the vaccine fields -– S03-01 commonality and lessons learned Jerald Sadoff
17 Malaria: status of the field and the RTS, S vaccine as a case study S03-02 Joe Cohen
17 Development of HPV-specific vaccines: current status and S03-03 future developments Lutz Gissman
HCV S03-04 Tbd
17 Genital herpes vaccine: parallels with AIDS vaccines S03-05 David Knipe
Symposium Session 4
T Cell and Innate Immunity
Wrap-up of the ‘immune correlates’ workshop preceeding the conference S04-01 Jonathan Heeney and Stanley Plotkin
19 Protective signature of antiviral immunity S04-02 Alexandre Harari
19 HIV-1-specific T cell responses and viral evolution in primary HIV-1 infection S04-03 Markus Altfeld
20 Recent results from macaque studies Toll-like receptor (TLR) ligands on the S04-04 replication of SIV and influenza after mucosal transmission Chris Miller
20 Modulating vaccine responses against HIV with dendritic cells and TLRs S04-05 Bali Pulendran
Symposium Session 5
Neutralizing Antibodies
21 Update on strategies for inducing broadly reactive neutralizing antibodies S05-01 Barton Haynes
21 The anti-HIV activity of antibodies in vitro and in vivo S05-02 Dennis Burton
21 Towards a working paradigm of structure-assisted vaccine design (SVAD) S05-03 Peter Kwong
AIDS Vaccine 2006 xi AIDS 2006 prelim matter 15/8/06 15:44 Page xii
21 Computational design of non-HIV protein scaffolds presenting S05-04 conserved HIV-1 neutralization epitopes William Schief
22 Mechanics by which Antibodies block HIV Infection S05-05 Suzan Zolla-Pazner
22 The role of antibody-dependent cellular cytotoxicity in S05-06 vaccine elicited protection Marjorie Robert-Guroff
Symposium Session 6
Animal Models and Correlates of Immunity
24 Association between strong vaccine-induced SIV-specific T-cell S06-01 responses and reduced viral load after challenge virus exposure in the SIV/macaque model Christiane Stahl-Henning
24 Cellular immune responses that successfully control AIDS virus replication S06-02 David Watkins
Selected presentation from ‘immune correlates’ workshop S06-03 Tbd during ‘Immune Correlates’ Workshop
25 Experimental infection of humanized mice with HIV S06-04 J Victor Garcia-Martinez
25 Oral delivery of replicating SIV vaccines in newborn Rhesus macaques: S06-05 safety and immunogenicity of recombinant VSV-SIV-based and live- attenuated SIV vaccines Marta Marthas
Symposium Session 7
T Cell Vaccine Development with Focus on T Cell Responses
26 Feasibility of T cell vaccines S07-01 Andrew McMichael
26 Update on the Merck Ad5 Phase II trial (the STEP study-Merck V520 S07-02 protocol 023/HVTN 502) Michael Robertson
27 Immunologic characterization of a multiclade HIV-1 DNA followed by S07-03 recombinant adenovirus prime/boost vaccine in normal healthy volunteers Richard Koup
EuroVacc DNA+NYVAC trial update S07-04 Giuseppe Pantaleo
27 Construction, development and evaluation of matched clade-C/B’DNA- S07-05 and MVA-based HIV-1 Vaccine candidates David Ho
Symposium Session 8
Epidemiology and Clinical Trials Issues
Where is the HIV epidemic going? S08-01 Catherine Hankins
28 Techniques for estimating HIV-1 incidence S08-02 Bernie Branson
xii Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xiii
28 HIV incidence rates: impact on vaccine trial design S08-03 Steve Self
28 Issues regarding pregnant women/breast feeding in vaccine trials S08-04 Susan Allen
Streamlining the HIV vaccine pipeline: criteria for advancing new HIV S08-05 vaccines from preclinical to phase I trials and beyond (Phase II and III) Scott Hammer
Symposium Session 9
Large Scale Production and Delivery Issues
30 Producing prophylactic vaccines in continuous cell lines S09-01 Rebecca Sheets
The unique properties of an adenovector-based HIV vaccine: S09-02 the importance of a robust production platform to deliver at the scale of need Jaap Goudsmit
Poxvirus scale up S09-03 Tbd
Vaccine distribution S09-04 Tbd
30 The impact of an AIDS vaccine in developing countries: S09-05 the use of modeling build support for vaccine development John Stover
Symposium Session 10
Therapeutic Vaccine Trials
31 TheraVac-01: an open-label Phase I study to evaluate the safety of the S10-01 HIV-1 vaccine NYVAC-B in chronic HIV-1 infected patients successfully treated with HAART Pierre-Alexandre Bart
31 A phase II randomized, partially-blinded trial of antiretroviral therapy (ART), S10-02 HIV-specific immunizations, and IL2 cycles to promote efficient control of viral replication (ACTE A5024) Michael Kilby
Immunogenicity of the ALVAC-HIV vCP1452 CanaryPox vector in HIV-infected S10-03 patients = the ORVACS-Manon-02 trial. Brigitte Autran
32 Inactivated retroviral virions with functional envelope glycoproteins: S10-04 evaluation as a therapeutic vaccine immunogen Jeffrey Lifson
32 DermaVir patch therapeutic vaccine in chronic HIV infection: S10-05 from the concept to the clinic Julianna Lisziewicz
Special Lunch Time Lecture
33 EC initiatives on HIV: capacity building, ethics, human rights promotion SLL01 in Africa Maurizio Salvi
AIDS Vaccine 2006 xiii AIDS 2006 prelim matter 15/8/06 15:44 Page xiv
Oral Abstract Session 1
HIV Transmission/Acute HIV Infection/Innate Immunity/Mucosal Immunity
37 Adaptation of human immunodeficiency virus type 1 to host HLA upon OA01-01 transmission to a new individual M Navis
37 Gp41 HMA as a screening tool to study HIV transmission between partners OA01-02 O Manigart
38 Evidence for potent autologous neutralizing antibody titers and OA01-03 compact envs in early infection with subtype C human immunodeficiency virus type 1 CA Derdeyn
38 Neutralization profiles of HIV-1 subtype C viruses from acute infection. OA01-04 P Moore
38 Resistance to HIV-1 infection among African female sex workers is OA01-05 associated with inhibitory KIR in absence of their HLA ligands W Jennes
39 Robust multifunctional HIVgag specific CD8+ T-cell responses in OA01-06 rectal mucosa and PBMC during chronic infection BL Shacklett
39 High levels of SIV-specific CD8+ T-cells in the mucosal tissues of OA01-07 macaques immunized with MVA-based vaccines expressing SIVmac239 gag and tat G Silvestri
Oral Abstract Session 2
Vaccine Antigen Selection & Design
41 Structural basis of antibody neutralization of HIV-1: OA02-01 implications for vaccine design I Wilson
41 Identification of portable immunodominant sequences leading to OA02-02 efficient HIV epitope presentation. S Le Gall
42 Enhancing CTL-epitope diversity in polyvalent vaccines: OA02-03 potential for world-wide HIV-1 coverage W Fischer
42 Humoral immunity directed at gp41-MPR for the prevention of HIV-1 OA02-04 mucosal transmission T Mor
43 Improvement of V1/V2-deleted HIV-1 envelope glycoprotein trimers OA02-05 through forced virus evolution R Sanders
43 The importance of antibody effector function in the ability of serum IgG OA02-06 to protect against vaginal challenge with SHIV. AJ Hessell
44 Preclinical safety and immunogenicity studies of adenovirus OA02-07 serotype-35-based vectors in rabbits and non-human primate J Mascola
xiv Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xv
Oral Abstract Session 3
Prophylactic Vaccine Trials
45 A Phase I study to evaluate the safety and immunogenicity of a recombinant OA03-01 adeno-associated virus HIV vaccine N Clumeck
45 Multigene, multiclade HIV-1 plasmid DNA prime and MVA boost is safe OA03-02 and highly immunogenic in healthy human volunteers. E Sandström
46 Prevalence of HIV infections among screened volunteers participating in the OA03-03 Phase III HIV vaccine trial in Thailand N Premsri
46 ALVAC HIV/AIDSVAX B/E Prime boost, the community Phase III OA03-04 HIV-1 preventive vaccine trial: an update -2006 S Rerks-Ngarm
47 Balanced cellular and antibody responses induced by the polyvalent DNA OA03-05 prime-protein boost HIV-1 vaccine formulation dp6-001 in healthy volunteers S Lu
47 Safety and reactogenicity of a multiclade adenovector HIV vaccine at two OA03-06 different doses: a randomized, placebo-controlled clinical trial (HVTN 054) L Peiperl
48 Breadth of the HIV-specific cellular immune responses to replication- OA03-07 defective adenovirus HIV vaccines in healthy subjects D Casimiro
Oral Abstract Session 4
T Cell Immunity
49 Upregulation of PD-1 expression in HIV specific CD8 T cells leads to OA04-01 reversible functional exhaustion L Trautmann
49 PD-1 expression on HIV-specific CD8 T cells is associated with disease OA04-02 progression C Day
50 Phenotype, apoptotic potential and polyfunctional characterization of OA04-03 T cells responses associated with live-attenuated SIV vaccine mediated protection of Rhesus macaques M Genescà
50 Structural basis for T-cell receptor recognition of HIV-1 epitopes OA04-04 G Stewart-Jones
51 Relationship between perforin expression and degranulation activity OA04-05 in virus-specific CD8 T-cells A Harari
51 Positive association between HIV-1 viral load and targeting of an epitope OA04-06 presented by B*5802, an HLA allele associated with disease progression KC Ngumbela
52 Amino acid variation of targeted HIV-1 specific CD8+ T-cell epitopes OA04-07 correlates with HLA class I disease association A Bansal
AIDS Vaccine 2006 xv AIDS 2006 prelim matter 15/8/06 15:44 Page xvi
Oral Abstract Session 5
B Cell Immunity
53 HIV-1 subtype A envelope variants from early in infection demonstrate OA05-01 variable neutralization sensitivity, in part due to changes in the transmembrane domain C Blish
53 HIV multiplication in macrophages and dendritic cell is blocked by some OA05-02 non-neutralizing antibodies directed against the V3 loop or the principal immunodominant domain gp41 C Moog
54 A human anti-V2/V3 monoclonal antibody efficiently neutralizes SF162 OA05-03 by targeting the unliganded envelope M Gorny
54 Mapping the neutralizing and non-neutralizing fractions of plasmas OA05-04 from HIV infected donors J Binley
55 Llama heavy-chain antibody fragments that neutralize HIV-1 OA05-05 RA Weiss
55 Development and characterization of HIV-1 epitope-specific OA05-06 reagents for the detection of antigen specific B cells MA Moody
56 Detection of HIV-1 variable loop-specific neutralizing antibody responses OA05-07 by HIV-2/HIV-1 envelope chimeras KL Davis
Oral Abstract Session 6
Clinical Trial Site Development/Social Issues/Ethical Issues/Access Issues/Regulatory Issues
57 ELISPOT standardization for laboratories conducting immunogenicity OA06-01 testing in IAVI sponsored HIV vaccine trials P Hayes
57 Negative social impacts in preventive HIV vaccine clinical trials OA06-02 M Allen
58 The willingness of black heterosexuals and men who have sex with men OA06-03 (MSM) in three U.S. cities to participate in HIV vaccine trials BN Bartholow
58 Recruiting into Phase I/II HIV vaccine trials in Soweto South Africa – OA06-04 a model for developing countries E Vardas
59 Self-reported sexual risk behaviors of a cohort being prepared for multiple OA06-05 Phase I/II HIV vaccine trials in Soweto, South Africa K Mesesan
59 Recruiting HIV discordant couples for vaccine efficacy trials through OA06-06 couples voluntary counseling and testing centers in Lusaka, Zambia. C Vwalika
60 Clearing the path: why the best news for HIV vaccines may be a cancer vaccine OA06-07 K Fisher
xvi Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xvii
Oral Abstract Session 7
Immune Escape
61 Late escape from an HLA-B27-restricted CTL epitope in HIV-1 p24/ OA07-01 capsid is associated with a dramatic reduction in replicative capacity A Schneidewind
61 Resistance profile of HIV-1 to the broadly neutralizing OA07-02 anti-gp120 2G12 antibody D Schols
62 Capture and transfer of antibody neutralized HIV-1 to CD4 lymphocytes OA07-03 by dendritic cells and DC-SIGN expressing cells T van Montfort
62 Reversion of HLA-B57-restricted CTL epitope escape mutations in OA07-04 HIV-infected subjects in Durban, South Africa H Crawford
63 Rapid reversion of transmitted sequence polymorphisms dominates early OA07-05 HIV-1 evolution B Li
63 Correlates of neutralization resistance in heterosexual transmission pairs OA07-06 infected with subtype C human immunodeficiency virus type 1 E Hunter
64 N-linked glycosylation sites proximal to the CD4 binding site of SHIV89.6P OA07-07 Env affect both neutralization escape and viral fitness NL Haigwood
Oral Abstract Session 8
Vaccine Delivery Methods/Adjuvants
65 Age dependence of adenovirus seroprevalence in South Africa OA08-01 AR Thorner
65 Antigen- and vector-specific immune responses elicited by hexon-chimeric OA08-02 adenovirus serotype 5 vectors and rare serotype adenovirus vectors DM Roberts
66 Immunization with vaccinia virus incluses polyfunctional and phenotypically OA08-03 distinctive CD8+ T cell responses ML Precopio
66 Development of a mucosal vaccine strategy against HIV-1 using virus like OA08-04 particles and recombinant Clostridium perfringens expressing HIV-1 proteins. P Poonam
67 Depletion of CD4+,CD25+, Foxp3+ regulatory T cells enhances OA08-05 systemic and mucosal HIV-1 env specific vaccine responses in mice J Peacock
67 Immunogenicity of a multi-antigen pDNA vaccine with and without OA08-06 in vivo Electroporation in Rhesus macaques M Egan
68 Activation of innate immune responses by human immunodeficiency OA08-07 virus type-1 (HIV-1) Pr55gag virus-like particles (VLP) S Bredl
AIDS Vaccine 2006 xvii AIDS 2006 prelim matter 15/8/06 15:44 Page xviii
Oral Abstract Session 9
Animal Models/Live-Attenuated Vaccines/Therapeutic Vaccination
69 Host factors contributing to rapid disease progression in SIV-infected OA09-01 infant macaques K Abel
69 Preservation of central memory T cell subsets after SIVmac239 OA09-02 challenge following a DNA prime and adenoviral boost vaccination regime R Schulte
70 Protection against heterologous SHIV challenge following vaccination OA09-03 with HIV-1 W61D recombinant envelope in the macaque model M Page
70 Development of a novel, live-attenuated virus vaccine against HIV OA09-04 C Jurgens
71 A novel mucosal vaccinia vector for inducing high-level anti-viral immunity OA09-05 Z Chen
71 Progesterone abrogates live-attenuated SIV vaccine mediated protection OA09-06 in intravenously challenged male rhesus macaques C J Miller
72 Therapeutic vaccination with SIV DNA+IL-12, IL-15 induces protective OA09-07 T cell memory subsets in SIV infected macaques. R Halwani
xviii Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xix
Poster Session: Topic 01
HIV Transmission
75 Detection of HIV-1 intrasubtype superinfection in a cohort of high-risk P01-01 African women B Chohan
75 Cultured Enzyme-linked immunospot assay (Elispot) enhances detection P01-02 of low HIV-specific immune responses in exposed seronegative individuals in Kenya O Anzala
76 Maternal neutralizing antibodies toward a CRF01-AE primary P01-03 isolate are associated with lower mother-to-child transmission T Samleerat
76 Multiple HIV-1 variants are transmitted between mother and infant with P01-04 selection on the gp120 envelope G Pollakis
77 DC-SIGN capture of HIV gp120 is disrupted by CD4 ligation P01-05 J Arthos
77 Systematic study of potential autoreactivity of the broadly neutralizing P01-06 antibodies 2F5, 2G12 and 4E10 G M Stiegler
78 Prevalence of sexually transmitted infections in areas of low HIV P01-07 prevalence in the Asia Pacific N T Thanh Thuy
78 Low Virus (SHIVsf162p) dose intra-vaginal challenges do not conform P01-08 with the mass-action principle: implications for HIV vaccine development S Garg
79 Prophylactic and therapeutic phase I trials with the active tat protein P01-09 vaccine: results of the interim analysis B Ensoli
Poster Session: Topic 02
Acute HIV Infection
80 Generation of functional envelope clones from HIV-1 subtype C plasma P02-01 RNA for use in pseudovirion assays I Choge
80 Novel role for HIV-1 transframe protein p6* in viral replication associated P02-02 with nef function A Leiherer
81 Evaluation of novel strategies for identification of primary HIV-1 C P02-04 infection in Botswana V Novitsky
Poster Session: Topic 03
Innate Immunity
82 Blocking of inhibiting receptors on fresh NK-cells leads to increased P03-01 degranulation in HIV S Johansson
AIDS Vaccine 2006 xix AIDS 2006 prelim matter 15/8/06 15:44 Page xx
82 HIV-1 replication in human primary cells is enhanced by P03-02 macrophage Migration Inhibitory Factor (MIF) EG Regis
83 CAF-activity in long-term SIV-infected rhesus macaques P03-03 J Harbort
83 Native tat preferentially enters dendritic cells inducing tumor necrosis P03-04 factor- mediated Th1-polarized maturation and immune responses in monkeys and humans A Cafaro
84 Multiple newly identified Uridine-rich TLR7/8 Ligands within the RNA P03-05 of HIV-1 activate human CD8+ T cells A Meier
84 Broad anti-viral activity of group X secreted phospholipase A2 P03-06 J Kim
85 Robust activation of NK cells by single stranded RNA derived from HIV-1 P03-07 G Alter
85 Specific recognition of an HIV-1 CTL escape mutation by dendritic P03-08 cells leads to functional inhibition via ILT4 D Kavanagh WITHDRAWN
Poster Session: Topic 04
Mucosal Immunity
86 Intranasal delivery of DNA IL-12 during mucosal DNA-prime/ P04-01 MVA-boost immunization schemes enhanced the cellular immune response against HIV-1 Env antigen. AM Rodríguez
86 Secreted heat shock protein gp96-Ig induces mucosal immune response P04-02 E Podack
Poster Session: Topic 05
T Cell Immunity
88 Capacity of gag m-RNA electroporated dendritic cells to induce P05-01 potentially protective HIV-specific T-cell responses in vitro G Vanham
88 HLA molecules that preferentially target the p24 Gag protein are P05-02 associated with slow progression to AIDS JAM Borghans
89 The kinetics of cytolysis by SIV Gag specific CD8+ T cells is markedly P05-03 different between vaccination and infection E Rollman
89 Statistical determination of threshold for T cell division in the P05-04 CFSE-labeling assay X Jin
89 Analysis of HLA-A*1101-restricted Nef73-specific CTLs with a strong P05-05 ability to suppress HIV-1 replication H Koizumi
xx Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxi
90 Structured treatment interruption does not regenerate HIV-specific IFN-γ P05-06 and IL-2 production but enhances broader HIV-specific perforin responses in a Ugandan population J Serwanga
90 CD8 T cell recognition of multiple epitopes within two regions of Gag is P05-07 associated with the maintenance of a low steady-state viremia in HIV-1 seropositives C Geldmacher
91 Cross-reactive cellular immune responses in HIV-1 patients infected with P05-08 different clades of HIV-1 L Gudmundsdotter
91 HIV-2 differs from HIV-1 in the ability to infect dendritic cells and to P05-09 induce polyfunctional CD4+ and CD8+ T cell responses M Duvall
92 Epidemiological studies revealed that in Argentina, early predominance P05-10 of B subtype has been overshadowed by the emergence of BF recombinants. Until correlates of protection are well understood, existence and determinants of cross-clade immune responses should be studied G Turk
93 Identification of novel conserved consensus CD4+ T cell epitopes from P05-11 clade B HIV-1 whole genome sequences that are frequently recognized by HIV-1 infected patients from different disease groups SG Fonseca
93 Low CD4 T cell counts despite low viremia: HIV-1 infected patients P05-12 with discordant response to HAART versus advanced HIV-2 disease A Albuquerque
94 Myeloid and plasmacytoid dendritic cells are susceptible to recombinant P05-13 Adenovirus vectors and stimulate polyfunctional memory T cell responses K Loré
94 HIV-1 C infected Indian Subjects recognize cytotoxic T lymphocyte (CTL) P05-14 epitopes conserved across the HIV-1 subtypes M Thakar
94 Analysis of T cell Interferon-α response to genome wide peptides in HIV-1 P05-15 Subtype C infection shows Immuno-dominance of Pol, Gag and Nef antigens. R Paranjape
95 Dissection of epitope-specific CTL responses suggests a mechanism P05-16 underlying differential HLA-A, -B and -C associations with HIV disease progression PJR Goulder
95 Priming of HIV-1 specific CD8+ T cells by dendritic cells P05-17 BA Colleton
96 Maintenance of antigen-specific TEM and TCM according to antigen P05-18 exposure in HIV-infected patients: comparison of vaccinia-, PPD- and HIV-specific immune responses. B Puissant
96 The IFN-γ-producing CD8+CD45RA-CCR7-CD27+ T cells specific for P05-19 HIV-1 Gag but not for Nef define immune correlates of protection W Lu
97 Differential selection pressure exerted by CTL targeting identical epitopes P05-20 but restricted by distinct HLA alleles within the same HLA-supertype A Leslie
97 Capturing HIV viral sequence diversity in in vitro immune analyses P05-21 C Brander
AIDS Vaccine 2006 xxi AIDS 2006 prelim matter 15/8/06 15:44 Page xxii
98 Narrow T cell receptor repertoire of virus-specific CD8+ T lymphocytes P05-22 is associated with HIV control N Frahm
98 Estimating the effectiveness of SIV specific CD8+ T cells from the P05-23 dynamics of viral immune escape JN Mandl
99 Small molecule telomerase activators: a novel approach to retard P05-24 immune exhaustion and progression to AIDS CB Harley
99 Stimulation of T cells with dendritic cells loaded with apoptotic cells P05-25 infected with autologous HIV-1 C Rinaldo
100 Restoration of specific CD8 CTL responses using autologous HIV P05-26 electroporated DCs B Yassine Diab
100 CTLA-4 is overexpressed in HIV-specific Th1 CD4 cells of HIV-infected P05-27 individuals, except in most subjects who spontaneously control viremia D Kaufmann
101 Identification of a subset of SIVgag-specific CD4+ IFNγ+ central memory P05-28 T cells that correlates with control of viremia and lack of disease progression in SIVmac251 infected macaques BK Felber
101 Elite control of HIV replication is distinguished from chronic disease by P05-29 lower HIV-specific IFNγ producing CD8+ T cell and higher polyfunctional CD4+ and CD8+ T cell responses F Pereyra
102 Persistence and homeostasis of CD4+ central memory T cells is dependent P05-30 on FOXO3a phosphorylation and STAT-5a activation C Riou
102 Effective immortalization of SIV-specific rhesus macaque CD8+ T cell P05-31 clones with maintenance of primary cell characteristics JD Lifson
103 Increased expression of NK cell receptors on HIV-specific CD8+ T cells P05-32 correlates with impaired TCR signalling G Alter
103 Design of HIV-1 pseudoviruses to monitor HIV-1 specific cellular P05-33 immune responses. D Beels
104 Dual selection pressure by drugs and HLA class I-restricted immune P05-34 responses on HIV-1 protease T Harrer
Poster Session: Topic 06
B cell Immunity
105 Sequences in the V1/V2 and V3 variable domains of HIV-1 gp120 that P06-01 determine the roles of these regions as mediators C Krachmarov
105 Dynamic model of HIV-1 infection and neutralization P06-02 S.Maloveste
xxii Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxiii
106 Mechanisms of HIV-1 inhibition by monoclonal antibodies when P06-03 macrophages and iDC are the target cells V Holl
106 Neutralization sensitivity of HIV-1 after homosexual and parenteral P06-04 transmission E Quakkelaar
107 Progress towards definition of HIV-1 neutralization serotypes: P06-05 Broad and potent HIV-1 primary isolate neutralization using polyclonal antibodies from subtype C infection VR Polonis
107 Dissecting the neutralizing antibody specificities of broadly neutralizing P06-06 sera from HIV-1 infected donors H Donners
108 The antigenicity and immunogenicity of a clade c trimeric gp140 P06-07 vaccine candidate NC Sheppard
108 Antibody Responses to MPER linear epitopes occur more frequently in P06-08 HIV-2 than in HIV-1 infections T Moon
109 Neutralizing specificity mapping in complex polyclonal sera P06-09 Y Li
109 In vivo evolution of coexisting R5 and X4 HIV type 1 variants P06-10 E Bunnik
110 Neutralization phenotype of clade C human immunodeficiency virus type 1 P06-11 gp160 clones from acute/early sexually acquired infections in India SS Kulkarni
110 Immunogenicity of constrained HIV-1 gp140CF envelope oligomers that P06-12 express membrane proximal external region epitope (MPER): evidence for differential regulation of antibodies to the 2F5 and 4E10 gp41 nominal epitope H-X Liao
Poster Session: Topic 07
Immune Escape
111 Designing a vaccine strategy: implications of viral escape and SHIV-specific P07-01 CD8 T cells at transmission and during acute infection E Rollman
111 Characterization of naturally occurring 4E10 resistant viruses in a subtype C P07-02 HIV-1 infected child E Gray
112 Cytotoxic T lymphocytes and the escape mutants of HIV-1 during and P07-03 after HAART with structured treatment interruptions in acutely infected HLA-A24-positive patients J Tanuma
112 Envelope regions involved in the antigenic evolution of sequential P07-04 HIV-1 isolates. P Nyambi
113 Escape from an HLA-B57-restricted CTL response in capsid reduces HIV-1 P07-05 fitness and increases sensitivity to APOBEC3G-mediated host restriction M Brockman
AIDS Vaccine 2006 xxiii AIDS 2006 prelim matter 15/8/06 15:44 Page xxiv
113 Towards identifying correlates of protective immunity: enhanced P07-06 detection of HIV-specific cellular immune responses using peptides based on the patient’s own viral quasispecies S Norley
114 Development of broadly reactive neutralizing monoclonal antibody P07-07 KD247: implications for passive immunotherapy S Matsushita
114 Therapeutic vaccination with HIV nef decreases patient-derived nef P07-08 protein sequence variability D Hoffmann
Poster Session: Topic 08
Viral Diversity
115 Increasing genetic distance to HIV-1 subtype B and F consensus P08-01 sequences in the Brazilian epidemic: a challenge for vaccine development G Bello
115 Evolution of viral populations in individuals infected with single and P08-02 multiple HIV subtypes AL Malaza
116 Social/demographic variables associated with Recombinants and Dual P08-03 infections in Injecting Drug Users in Northern Thailand G Kijak
116 Enhanced detection of HIV-1 specific T-cells in early HIV-1 infection P08-04 using peptides designed by a novel biometric approach U Malhotra
117 Genetic diversity of HIV-1 strains among IDUs in 2002-2004 in P08-05 St. Petersburg, Russia E Dukhovlinova
117 Unique recombinant forms of HIV-1 subtypes A and G emerging from P08-06 dual infections in Cameroon R Powell
118 Global and regional distribution of HIV-1 genetic subtypes and P08-07 recombinants in 2004. J Hemelaar
118 Genetic diversity of HLA-class I alleles among HIV-infected and P08-08 uninfected individuals in a Chinese Uygur ethnic group KX Hong
119 Full-length HIV-1 subtype C sequences from individuals with acute and P08-09 chronic infection from Kwa-Zulu Natal, South Africa F Treurnicht
119 Molecular and phylogenetic analysis of HIV-1 isolates identified in P08-10 Mashhad - Iran. L Buonaguro
120 Serotyping approach as a feasible tool for assessing HIV-1 diversity in P08-11 Sao Paulo, Brazil. C A F Oliveira
120 A comprehensive molecular epidemiological study of HIV-1 P08-12 subtypes in Yunnan province of China Z Chen
121 High throughput functional analysis of HIV-1 env genes without cloning P08-13 JL Kirchherr
xxiv Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxv
121 HIV-1 prevalence and subtype distribution in HIV-1 positive volunteers deferred P08-14 from enrollment in a phase III prime-boost HIV-1 vaccine trial in Thailand F McCutchan
Poster Session: Topic 09 – A
Vaccine Concepts & Design
Antigen Selection & Design
122 Development of HIV-1 nef DNA vaccine construct from Indian subtype C P09a01 and its immunogenecity testing in Balb/C mice. P Seth
122 Conjugation of peptides bearing HIV drug-resistance mutations to the P09a02 B-subunit of recombinant cholera toxin enhances the immune response against the mutations A Boberg
123 Proteasome targeted reverse transcriptase of HIV-1 generates a balanced P09a03 T-cell response in mice after DNA-vaccination A Boberg
123 Induction of potent cross clade immune response following immunization P09a04 with HIV-1 Indian subtype C mutated and codon optimized tat DNA/MVA vaccine in mice. H Qureshi
124 Immunogenicity of HIVCON, a novel HIV-1 vaccine candidate based on P09a05 conserved regions of clades A-D S Letourneau
124 Combined single-clade candidate HIV-1 vaccines induced T cell responses P09a06 limited by multiple forms of in vivo immune interference T Hanke
125 HIV-1 Envelope on the Surface of a Lentiviral Particle Elicits Broader Immune P09a07 Responses than Soluble gp120 or Trimeric gp140 Envelopes. T Ross
125 Interactions HIV-1 - Host Cells: Immunomodulation in the HIV-1/ P09a08 Trypanosomatids Co-infection Model. V Baretto-De-Souza
126 Critical role of an NKP44 ligand in the CD4+ T depletion during HIV infection: P09a09 a new target for a vaccine candidate P Debre
126 Explaining the failure of SIV/HIV vaccinations priming CD8 T cell responses P09a10 RJ de Boer
126 A consensus clade B env HIV virus-like particle vaccine elicits broader P09a11 immunity than a polymeric clade B virus-like particle based vaccine S McBurney
127 Large-scale and highly reproducible selection of peptides to the gp41 P09a12 immunodominant epitope of HIV-1 by phage library screening with patient serum Y Palacios-Rodríguez
127 Characterisation of a tat mutant, antigen candidate for HIV-1 P09a13 therapeutic vaccination K Mayol
AIDS Vaccine 2006 xxv AIDS 2006 prelim matter 15/8/06 15:44 Page xxvi
128 Biophysical and antigenic properties of a soluble, cleaved and stabilized P09a14 HIV-1 subtype A envelope trimer S Iyer
128 Sequential cross-clade vaccination of HIV vaccines predominantly P09a15 stimulated T-cell immunity against HIV conservative epitopes J Xu
129 Characterization of HIV-1 gag specific T-cell immune responses and P09a16 correlation with plasma viremia in infected Indian individuals S Kaushik
129 Structure of the SIV envelope spike in situ P09a17 S Sourial
130 Identification of mimotopes of env immunogenic epitopes of HIV-1 P09a18 from donors with HIV-1 broad cross-neutralizing antibodies T Dieltjens
130 Chimaeric HIV-1 subtype C Pr55 gag virus-like particles with large P09a19 in-frame C-terminal polypeptide fusions as subunit analogues of a DNA vaccine EP Rybicki
131 Multivalent immune cell surface carbohydrate moieties inhibit HIV-1 P09a20 primary isolate infection of peripheral blood mononuclear cells but enhance infection of TZM-bl reporter cells A Rosa Borges
131 Mimotopes from combinatorial phage libraries as anti-HIV P09a21 vaccine candidates K Gazarian
132 Broadly neutralizing antibodies against receptor-activated epitopes on P09a22 the N-heptad repeat region of HIV-1 gp41 are elicited during natural infection and by immunization M Zwick
132 Structure-based stabilization of the 2F5 antibody epitope of the HIV-1 P09a23 gp41 envelope glycoprotein for immunogen design FJ Guenaga
132 Effectiveness of gp120 hyperglycosylation on improving the elicitation P09a24 of broadly neutralizing antibodies to HIV-1 R Pantophlet
133 Evaluation of a synthetic tetramannose (Man4) glycoconjugate as an P09a25 immunogen designed to elicit anti-mannose antibodies to an oligomannose cluster on HIV-1 gp120 R Astronomo
134 The broadly neutralizing antibodies 2F5 and 4E10: Specific anti-HIV P09a26 antibodies or autoantibodies? EM Scherer
134 2006 progress update on the GAIA HIV vaccine: broad coverage by HLA P09a27 alleles and geographic location, optimization of HIV immunogens AS de Groot
135 Elicitation of robust T cell responses in HLA DRB1*0101 transgenic P09a28 mice by MHC class II compartment-targeted and untargeted HIV T helper multiepitope molecular constructs L Moise
135 Comparison of neutralisation antibody HIV vaccine strategies based on P09a29 trimeric gp140 Envelope of clades A, B and C and peptides to conserved neutralisation epitopes; Efficacy against R5 SHIV challenge D Davis
xxvi Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxvii
136 Impact of a synthetic transdominant negative HIV-1 Gag and the Gly-Ala P09a30 repeat of EBNA-1 on the inhibition of HIV-1 replication and the survival of transduced cells for gene therapy approaches D Hammer
136 Intragenic CpG dinucleotides: A novel tool for improving plasmid based P09a31 DNA based therapeutics and vaccines D Leikam
136 Influence of transgenic CpG dinucleotides on transgene expression P09a32 A Bauer
137 Identification of HIV-related cellular proteins as novel targets for therapeutic P09a33 vaccines B Asbach
137 Improvement of innate properties of DNA vaccines and bridging innate P09a34 and adaptive immune responses by sequence modifications of the plasmid backbone D Kosovac
138 Effects of C3d fusion proteins on the expression and immunogenicity of P09a35 HIV-1 gp120 DNA vaccines V Wanninger
138 Directed nucleic acid packaging of HIV-1 subtype C Pr55 gag virus-like P09a36 particles potentially used as vaccine candidate. Z Valley-Omar
139 Removal of glycans in the inner domain of HIV-1 gp120 affects P09a37 binding and neutralization by monoclonal antibodies and plasma PS Nkosi
139 A novel HIV/AIDS vaccine based on the combination of HIV Tat and V2-Env P09a38 proteins: preclinical studies F Titti
140 Anti-tat humoral and cellular immunity in HIV-1 infected patients, including P09a39 asymptomatic individuals and LTNP, aimed at identifying immune correlates of protection which are key to vaccine development F Ensoli
140 Evidence for underestimation of the repertoire of cytotoxic T lymphocyte P09a40 epitopes in HIV-1 infection IMM Schellens
141 Single molecule force spectroscopy of 2F5 monoclonal antibody-epitope P09a41 interactions R Tawar
141 A specific gp41 conformation as immunogen P09a42 A Hinz
141 Functional and structural homology between the V3 region of HIV-1 P09a43 gp120 and the 40s loop of SDF-1a T Kimura
142 An HIV-1 peptide-based vaccine inducing cross-subtype immunity in P09a44 macaques F Diaz-Mitoma
142 Stabilization of gp120 in the liganded conformation by coupling with P09a45 CD4 mimic - AIDS vaccine application G Martin
143 Mapping epitope candidates from HIV-1 Brazilian sequences for P09a46 vaccine Development L C J Alcântara
AIDS Vaccine 2006 xxvii AIDS 2006 prelim matter 15/8/06 15:44 Page xxviii
143 Development of a new vaccine against HIV: virosomes incorporating P09a47 HIV proteins H Quendler
144 Distribution and three-dimensional structure of AIDS virus envelope P09a48 spikes by cryo electron microscopy K Roux
144 Immunogenicity of HIV non-glycosylated env trimer expressed in P09a49 Escherichia coli using Neisseria meningitidis NadA as scaffold. B Capecchi
145 A polyvalent HIV-1 gp120 candidate vaccine induces broad cellular P09a50 immunity against HIV-1 subtypes in golden hamsters A Azizi
145 Evaluation of the HIV CCR-5 Co-receptor: conformation and P09a51 immunogenicity S Phogat
146 Preclinical evaluation of epitope-based DNA and MVA vaccines in a P09a52 heterologous prime/boost regimen D McKinney-Baker
146 Induction of HIV-neutralising antibodies specific for the membrane- P09a53 proximal external region (MPER) of the transmembrane envelope protein gp41 J Denner
147 Comparative analysis of neutralizing antibody responses elicited P09a54 during infection with a SHIV expressing the SF162 Env and during immunization with SF162 Env derived gp140 proteins. L Stamatatos
147 MPR649-684-HBsAg fusion as a vaccine candidate against HIV-I infection P09a55 I Cherni
148 Modulation of immunogenicity of SIV and HIV antigens using fusion P09a56 proteins with lysosomal-associated membrane protein-1 (LAMP-1) C Bergamaschi
148 Ex vivo-primed CD8+ response to HLA-A2-restricted (A2-) HIV-1 P09a57 gag epitopes. J Kan-Mitchell
149 Glycosylation variants of the HIV-1 envelope protein: antigenicity P09a58 and immunogenicity L Xu
149 Immunogenicity of solid phase proteoliposomes containing elements P09a59 of HIV-1 envelope glycoproteins M Tang
150 Identification of envs that preferentially bind the cross-reactive neutralizing P09a60 Ab in chronic patient sera HL Robinson
150 Rational designing of novel env for generating stable Env-CD4M33 P09a61 complexes as potential vaccine immunogen E Kan
151 HIV-1 infectivity decay depends on env and can be accelerated by P09a62 neutralizing antibodies F Charifi
151 Expression of HIV-1 subtype C p24 in transgenic plants P09a63 S Andersson
xxviii Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxix
152 Biochemical and immunogenic characterization of stabilized HIV-1 P09a64 trimeric gp120 glycoproteins B Dey
152 Approaches to develop a CD4 mimetic capable of binding to, and P09a65 inducing conformational changes in the env glycoprotein of HIV-1 VA Sharma
153 Structural characteristics correlate with immune responses induced by HIV P09a66 envelope glycoprotein vaccines VA Sharma
153 Loss of a single N-linked glycan in the V2 loop of HIV-1 env results in P09a67 CD4-independence in viral infection and enhanced ability to elicit neutralizing antibody responses S-L Hu
154 Binding kinetics and energetics of broadly neutralizing anti-HIV-1 P09a68 membrane proximal external region (MPER) human monoclonal antibodies to HIV-1 peptide epitopes anchored on lipid membranes indicate an entropic barrier to Mab binding SM Alam
154 Evaluation of the immunogenic potential of gp41-based peptide P09a69 vaccines that contain the caveolin-1 binding domain (CBD) motif Y Huang
155 HIV neutralizing anti-CCR5 antibodies exert a protective effect against P09a70 disease progression in a subset of long term non progressors L Lopalco
155 Anti-CD4-gp120 complex antibodies in long term non progressors HIV-1 P09a71 positive patients play a role in slowing disease progression L Lopalco
155 Immunogenicity of a novel particle vaccine targeting HIV-1 gp41 P09a72 D Gardiner
156 DNA prime and VPL boost vaccination strategy targeting membrane- P09a73 proximal region of gp41 induces weak HIV-1 neutralizing antibodies in rabbits D Gardiner
157 AIDS vaccine integrated project (AVIP): a consortium funded by the P09a74 FP-6 EU program B Ensoli
157 Reservoir cells no longer detectable after a heterologous SHIV P09a75 challenge with the synthetic HIV-1 tat Oyi vaccine EP Loret
158 Very broadly cross-reactive neutralizing antibodies induced by HIV-1 P09a76 oligomeric gp140 in rabbits: distinction from partially cross-reactive response and basis for design of non-human primate challenge studies G Quinnan
158 Preparation of a therapeutic vaccination strategy aiming at functional P09a77 T cell responses against the early regulatory HIV-1 proteins tat, rev and nef ADME Osterhaus
159 Functional constraints and antibody recognition of the CD4-binding site P09a78 on HIV-1 gp120 PD Kwong
AIDS Vaccine 2006 xxix AIDS 2006 prelim matter 15/8/06 15:44 Page xxx
Poster Session: Topic 9 - B
Vaccine Concepts & Design
Delivery Methods
160 Evaluation of a plasmid DNA prime MVA boost vaccine regimen P09b01 intended for human use A Bråve
160 HIV-1 specific humoral, cellular and reminiscent immune responses P09b02 induced in Chinese Rhesus macaques by DNA-tiantan vaccinia combined AIDS vaccine GB Yang
161 Activity of different vaccine-associated promoter elements in human P09b03 dendritic cells T Papagatias
161 Immunological properties of oral Salmonella typhimurium-based, and P09b04 injectional forms of DNA-vaccine administered in various prime-boost combinations in mice E Dukhovlinova
161 Multivalent HIV-1 envelope vaccines delivered in the VEErep/SINenv P09b05 alphavirus replicon particle chimera Y Lian
162 Different tropisms of canarypox (ALVAC) and vaccinia viruses are P09b06 comparable with their reactogenicity profiles in human vaccine trials Q Yu
163 Molecular characterization and immunogenicity of a novel P09b07 DNA-launched VEE-based replicon DNA vaccine. K Ljungberg
164 Systemic and mucosal humoral response induced by HIV-1 p24 P09b09 protein adsorbed onto poly(L-lactic acid) nanoparticles S Munier
164 Ability of herpes simplex virus vectors to boost immune responses P09b10 to DNA vectors and to protect against challenge by simian immunodeficiency virus A Kaur
165 Priming with a rBCG or DNA vaccine expressing HIV-I subtype C gag P09b11 elicits comparable CD8+ T cell responses when boosted by a rMVA expressing gagC A.-L Williamson
165 The impact of priming on memory of clinical HIV-1 clade C NYVAC based P09b12 vaccine candidates in Rhesus macaques P Mooij
166 Evaluation of GMP grade HIV-1 clade C CN54 DNA candidate vaccines for P09b13 human phase I trials in a Balb/c mouse model J Wild
166 Equine herpesvirus (EHV) type 1 based viral vectors effectively transduce P09b14 and activate monocyte derived dendritic cells (MDDC) J Köstler
167 Accelerated and robust immune responses induced by EuroVacc DNA HIV-1 P09b15 vaccine immunogens delivered by novel intradermal DNA delivery. A Bins
xxx Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxxi
167 A novel measles virus Edmonston Zagreb vector is highly immunogenic in P09b16 laboratory animals A Zuniga
167 Electroporation of DNA vaccines encoding human immunodeficiency virus P09b17 (HIV-1) antigens in small animals and non human primates elicits potent humoral and cellular immune responses R Pal
168 Production of C-clade HIV-1 gp140 for a novel cervico-vaginal delivery system P09b18 SA Jeffs
168 Baculovirus-expressed HIV-1 virus-like particles (HIV-VLP) activate dendritic P09b19 cells and induce ex vivo T cell response L Buonaguro
169 Propagation enhancement of chimeric lentiviral vaccines P09b20 K Young
169 Immunogenicity OF ANRS LIPO-5 alone, sanofi-pasteur ALVAC-HIV P09b21 (vCP1452) alone, and ALVAC HIV Prime/LIPO-5 boost in healthy, HIV-1 uninfected adult participants S Frey
170 Enhancing expression and stability of foreign genes in recombinant P09b22 MVA vaccines L Wyatt & P Earl
170 Development of immunologically-enhanced HIV subtype C vaccines from P09b23 genetically modified MVA vectors L O’Mara
171 Characterization of a replication-defective MVA-based vaccine vector that P09b24 induces cellular apoptosis atypically during abortive infection DA Garber
171 Differential consequences of infection of dendritic cells with vaccinia virus P09b25 or its attenuated derivative, modified vaccinia virus Ankara D A Garber
172 Construction and characterization of a novel replication competent vaccinia P09b26 Tian Tan-based vector for vaccine development Z Chen
172 DNA microparticles delivered intranasally are able to prime mucosal and P09b27 systemic anti-SIV immune responses in rhesus macaques L Giavedoni
173 Robust immune responses to an attenuated antibiotic-gene-free strain of P09b28 Listeria monocytogenes that expresses HIV-gag Z Li
173 Blockade of a negative immuno-regulatory pathway increases magnitude P09b29 of HIV-1 specific CD8+ T cells induced by simian origin adenovirus vector HCJ Ertl
174 A novel Clostridium perfringens-based SIV vaccine with adenovirus P09b30 boosting induces strong systemic and gut mucosal immune responses RA Helmus
174 Long-term control of Simian immunodeficiency virus infection (SIV) in P09b31 macaques by mucosal immunisation with a bimodality vector combination C Kuate
175 Microdermabrasion of skin to target cutaneous dendritic cells for P09b32 MVA-based AIDS vaccines H Gill
AIDS Vaccine 2006 xxxi AIDS 2006 prelim matter 15/8/06 15:44 Page xxxii
Poster Session: Topic 9 - C
Vaccine Concepts & Design
Adjuvants
176 Complementary HIV specific CD4 and CD8 responses: NHP adenovirus P09c01 and adjuvanted protein prime boost N Mathy
176 Intranasal immunization with a multigene vaccine in combination with P09c02 a novel adjuvant, N3, induces systemic and mucosal immune responses A Bråve
177 Neutralizing antibody responses to subtypes B and C adjuvanted HIV P09c03 envelope protein vaccines in rabbits B Burke
177 Type I interferons as biological adjuvants for the improvement of P09c04 poxvirus HIV vaccine efficacy S Day
178 Adjuvanting Effects of dsRNA molecules Transcribed from DNA P09c05 Vaccine Vector D F Li
178 Interleukin-7 and interleukin-15 enhance antigen-specific memory P09c06 T-cell responses elicited by topical HIV-1 DermaVir vaccine S A Calarota
179 Increasing the efficiency of DNA/MVA vaccines P09c07 H Robinson
179 Cytokine and chemokine-mediated augmentation of MVA immunogenicity P09c08 D A Garbier
180 Novel strategy for generation of mucosal immune responses against P09c09 HIV-1 following systemic vaccination D Weiner WITHDRAWN
180 Multimeric soluble forms of CD40L and GITRL as molecular adjuvants P09c10 for HIV vaccines R Kornbluth
180 Mother-child immunity against HIV-1 after gagp, nef, tat, gp160/rev P09c11 plasmid DNA prime recombinant gp160 boost vaccine regimen J Hinkula
Poster Session: Topic 9 - D
Vaccine Concepts & Design
Live-attenuated Vaccines
182 Live Attenuated EIAV Vaccine Induced high Th1 Immune Responses and P09d01 provided immune protection against virulent EIAV infection X Zhang
182 Effectiveness of different chemical and physical inactivation methods to P09d02 reduce the infectivity of HIV-1 Strains. C Gil
183 Head-to-head comparison on the immunogenicity in mice of two HIV/AIDS P09d03 vaccine candidates based on the attenuated poxvirus strains MVA and NYVAC co-expressing in a single locus the HIV-1BX08 gp120 and HIV-1IIIB Gag-Pol-Nef proteins of clade B M Esteban
xxxii Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxxiii
183 Live attenuated measles virus as a candidate multivalent vaccine vector P09d04 against AIDS HY Naim
184 Improving the stability of conditional life virus by removal of the TAR region P09d05 M Centlivre
184 Conditional-live HIV-1 and SIV variants as a novel strategy toward a safe P09d06 live attenuated AIDS vaccine and as a tool to study virus replication AT Das
185 Expression of HIV-1 gag proteins from a replication competent P09d07 coxsackievirus vector J Miller
185 Head-to-head comparison of immunogenicity and efficacy in macaques of P09d08 MVA and NYVAC vectors expressing HIV-1 Env and SIV Gag/Pol/Nef P Mooij
185 Immunogenicity of recombinant measles virus Edmonston Zagreb (rMVEZ) P09d09 vaccine vector expressing multiple antigens of HIV-1 M Liniger
186 Development of covalent HIV-1 gp120-CD4 mimic complexes for vaccine P09d10 application L Martin
Poster Session: Topic 10
Animal Models
187 Immunogenicity of a multiepitope multivalent DNA vaccine against HIV-1 P10-01 in cynomolgus macaques F Martinon
187 DNA induced Th1 biased immune responses are more effective against the P10-02 pathogenic effects of SHIV challenge than the broader and stronger immune responses induced by protein G Koopman
188 Contribution of major histocompatibility complex (Mhc) class I P10-03 genotypes to set point viral load, survival time, and immunisation results in Rhesus macaques U Sauermann
188 Protective efficacy of multiprotein genetic vaccine in the SIV-macaca P10-04 animal model F Titti
189 Viral outcome of Simian-Human Immunodeficiency Virus SHIV-89.6P P10-05 adapted in cynomolgus monkeys (SHIV-89.6 Pcy243) F Titti
189 Optimization of mucosal immunization protocols involving based on P10-06 HIV-1 Virus-like Particles F M Buonaguro
190 Protection of cats by immunisation with the transmembrane envelope P10-07 protein p15E of feline leukaemia virus (FeLV) J Denner
190 The neutralizing antibody response in long-term non-progressor P10-08 monkeys infected with simian immunodeficiency virus (SIVsm) E M Fenyö
AIDS Vaccine 2006 xxxiii AIDS 2006 prelim matter 15/8/06 15:44 Page xxxiv
191 Mutagenized region within the first loop of CCR5 induce high affinity P10-09 HIV neutralizing antibodies L Lopalco
191 Systemic T cell priming in macaques exposed rectally to SIV after P10-10 local application of a nucleotide analogue reverse transcriptase M Cranage
Poster Session: Topic 11
Prophylactic Vaccine Trials
192 Out of range laboratory values: a major reason for exclusion from a P11-01 Phase I HIV vaccine clinical trial W G Jaoko
192 Vaccine-induced cytotoxic T lymphocyte responses in uninfected adult P11-02 Ugandan volunteers enrolled in a phase I double-blind multiclade HIV-1 DNA plasmid vaccine trial (VRC-HIVDNA009-00-VP) M A Eller
193 Cross sectional volunteers’ satisfaction survey in Phase III prime-boost P11-03 HIV preventive vaccine trial in Thailand T Woratanarat
193 Frequent large blood draws and decrease in hemoglobin levels among P11-04 AIDS vaccine trial participants in India S Mehendale
194 Optimization of cell separation procedures for determination of HIV-1 P11-05 vaccine-induced cell-mediated immune responses in Swedish and Tanzanian individuals. C Nilsson
194 Detection of circulating HIV vaccine plasmid DNA in immunized individuals P11-06 G Engström
195 HIV-specific lymphocyte proliferative responses determined by a P11-07 standard thymidine uptake assay and a flow cytometric assay in healthy individuals immunized with HIV-1-DNA/MVA S Aboud
195 Sources of volunteer’s perception and decision to participate in the P11-08 prime-boost HIV vaccine trial C Namwat
196 Safety and immunogenicity of an alphavirus replicon HIV gag vaccine P11-09 (AVX101) in healthy HIV-uninfected adults J Chulay
196 Screening and enrollment into a phase I HIV vaccine trial in Kigali, Rwanda P11-10 E Shutes
197 Safety and immunogenicity of HIV CTL Multi-Epitope Peptides (MEP) P11-11 vaccine in RC529-SE adjuvant with or without GM-CSF S Kalams
197 Evaluation of the safety and immunogenicity of the EP HIV-1090 DNA P11-12 vaccine in healthy, HIV-1-uninfected adults N D Russell
198 Phase I clinical trial of the first Russian candidate HIV vaccine (VICHREPOL) P11-13 is being conducted successfully E Karamov
xxxiv Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxxv
198 Abnormal laboratory values among volunteers for a Phase I vaccine trial in P11-14 the Caribbean Region (Merck 018/HVTN050) and the need to reevaluate the eligibility criteria C D Zorrilla
199 Adolescents participation in the phase III efficacy trial of ALVAC HIV-1 P11-15 vaccine priming, AIDSVAX vaccine boosting in Thailand P Pitisuttithum
199 Dose sparing with intradermal injection of HIV lipopeptides in HIV P11-16 uninfected adult volunteers: a randomized controlled study (ANRS VAC16 trial) C Durier
200 Enhanced positive responder rates in elispot assay reported when using an P11-17 optimized method for collection and isolation of PBMC: evidence supporting the use of a PBMC network. D Casimiro
200 Building productive, collegial dialogue between researchers and P11-18 communities: The HVTN Model G Broder
Poster Session: Topic 12
Therapeutic Vaccine Trials
202 Therapeutic SIV derived DNA-vaccines in the SIVmac239 rhesus macaque P12-01 animal model JZ Megede
202 Cell mediated immune response in HIV-1 infected subjects on HAART induced P12-02 by two different DNA vaccines expressing either single gene, GTU-Nef or multigenes, GTU-MultiHIV of a subtype B-HIV-1 V Blazevic
203 Comparable immunogenicity between patients on or off combination P12-03 antiretroviral therapy (CART) after immunisation with a novel therapeutic peptide-based immunotherapy candidate targeting dendritic cells (DC) D Kvale
203 Follow-up of HIV infected patients who received a therapeutic P12-04 anti-tat vaccination D Zagury
204 Randomized-controlled phase-II-study with an MVA-nef vaccine in HIV-1 P12-05 infected patients with CD4 counts > 250/µl followed by Structured Treatment Interruption (STI) – First Results T Harrer
204 DNA Vaccines against human immunodeficiency virus type 1 P12-06 Y Zadorin
205 Development of a therapeutic vaccination strategy with a recombinant P12-07 modified vaccinia virus Ankara vaccine expressing an HIV-1 gag / /multi-epitope gene (MVA.HIVA) for control of HIV-1 infection. L Dorrell
205 Control of viremia after antiretroviral treatment and therapeutic P12-08 vaccination with novel forms of DNA vaccines in chronically SIVmac251-infected macaques GN Pavlakis
AIDS Vaccine 2006 xxxv AIDS 2006 prelim matter 15/8/06 15:44 Page xxxvi
Poster Session: Topic 13A
Clinical Trial Site Development
Cohort Development
206 Potential cohort for microbicide trials, Mukuru, Nairobi, Kenya P13a01 JJ Bwayo
206 Process analysis of a three-tiered community based volunteer P13a02 recruitment model for the first AIDS vaccine trial in India S Sahay
207 The importance of thorough and continuous informed consent in HIV P13a03 cohort studies: experiences from the Kericho HIV cohort study J Cheruiyot
207 Unique challenges and lessons learned in HIV vaccine cohort P13a04 development in Kericho, Kenya F Sawe
208 Qualifying a vaccine trial laboratory in Rwanda for PBMC isolation, P13a05 cryopreservation and shipping. J Bizimana
208 Developing a cohort and evaluating of HIV prevalence and HIV P13a06 incidence rates among Injection Drug Users in St. Petersburg, Russia AP Kozlov
209 Expansion and maintenance of an HIV discordant couple cohort in P13a07 Kigali, Rwanda in preparation for vaccine efficacy trials E Kestelyn
209 Lessons learned in developing Research Counseling and Testing (RCT) P13a08 in East Africa N Bahati
210 Challenging for the enrollment in a phase I trial HIV vaccine in Haiti P13a09 SS Jean
210 Recruiting African Americans and Latinos into HIV vaccine trials P13a10 SG Arreola
211 Adolescent sexuality, stigma, and potential benefits of trial participation: P13a11 views on adolescent participation in HIV Va G de Bruyn
Poster Session: Topic 13B
Clinical Trial Site Development
Epidemiology Studies
212 HIV-1 prevalence, incidence and molecular characteristics at a feasibility P13b01 study in a voluntary counseling and testing site (VCT) at south Brazil LFM Brígido
212 Knowledge, attitudes, and practices study in marketplaces of Bamako, Mali P13b02 to inform the design of HIV vaccine trials M Koné
213 Necessity of region-specific normal reference ranges for conducting P13b03 HIV clinical trials L Eller
xxxvi Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxxvii
213 A cross sectional, observational study to establish clinical laboratory P13b04 reference ranges in adults as part of a multi-center study in preparation for HIV Vaccine efficacy trials A Kamali
214 Community Heath Forum; a method of community education in the P13b05 prime-boost phase III HIV vaccine trial in Thailand V Chaiwong
214 Further vaccine related analysis of the Buenos Aires men who have sex P13b06 with men (MSM) cohort. Argentina MM Avila
Poster Session: Topic 13C
Clinical Trial Site Development
Prevalence Studies / Incidence Studies / Transmission Studies
216 Volunteer recruitment strategies in HIV vaccine trials. An experience P13c01 from the Botswana HIV Vaccines Trials Unit (BHVTU) L Mwananyanda
216 South African community members’ perceptions of enablers and P13c02 inhibitors to participation in HIV / AIDS vaccine trails A Kagee
217 Estimating population HIV incidence in subtype C infected populations P13c03 for phase IIB/III HIV vaccine trials accurately and cheaply with the Avidity Index (AI) E Vardas
217 Web-based training to support HIV vaccine development P13c04 J Fuchs
218 EDCTP: A new partnership for HIV clinical trials in developing countries P13c05 M Karras
218 Improving recruitment for HIV vaccine trials: lessons learned in P13c06 Entebbe, Uganda E Mugisha
219 P24 antigen screening for early detection of HIV infection in P13c07 discordant heterosexual couples in Zambia J Mulenga
219 Comparison of three methods to detect recent HIV-1 infection in P13c08 specimens collected cross-sectionally in a cohort of female sex workers in the Dominican Republic S Gupta
220 Establishing a high risk HIV-negative cohort in Kilifi, Kenya P13c09 EJ Sanders
Poster Session: Topic 13 D
Clinical Trial Site Development
Measuring Efficacy Trial Endpoints
221 HSV-2 Infection in potential high risk group volunteers for HIV P13d01 vaccine trials B E Bekan Homawoo
AIDS Vaccine 2006 xxxvii AIDS 2006 prelim matter 15/8/06 15:44 Page xxxviii
221 One or two efficacy trials for HIV vaccines and microbicides? P13d02 Comparison of statistical power, sample size and costs P M Coplan
222 A small dose of HIV? HIV vaccine mental models and heuristics among P13d03 U.S. communities at risk (Project VIBE) PA Newman
222 HIV vaccine concerns, misconceptions and mistrust among vulnerable P13d04 communities in the United States (Project VIBE) PA Newman
Poster Session Topic 13E
Clinical Trial Site Development
Vaccine Preparedness
224 Assessment of willingness to participate in HIV vaccine trials in a P13e01 cohort of homosexual and heterosexual intravenous drugs users: Our experience in a resource-poor sub-Saharan African city OA Busari
224 Progress towards the first Phase I/II HIV vaccine trial in Tanzania, the HIV P13e02 vaccine immunogenicity study (HIVIS) project M Bakari
225 Enrollment and screening failure in 2,424 African volunteers recruited for a P13e03 multi-center study of laboratory reference ranges in preparation for HIV vaccine clinical trials K Kayitesi
225 Ethical aspects of adolescent participation in HIV vaccine trials in the P13e04 Nyanga district, Cape Town, South Africa NF Soka
226 Predictors of seropositivity among men at a voluntary HIV counseling P13e05 and testing site (VCT) in Curitiba, South Brazil R Rodrigues
226 Seasonal variation in clinical laboratory parameters among HIV-uninfected P13e06 adults in Kigali, Rwanda E Karita
227 Knowledge, attitude and perception of health care providers on the ongoing P13e07 phase-I vaccine trials in the country in Jaipur, the capital city of the state of Rajasthan,India M Singh
227 Willingness to participate in HIV vaccine trials among trainee artisans in P13e08 Ibadan, Nigeria O Onigbogi
228 Towards a new GCP: “Good Community Practice” in prevention research P13e09 MJ Warren
228 Increasing community involvement: model training programs P13e10 E Lee
229 Research advocacy: maximizing preparedness for HIV research P13e11 E Lee
229 Successful enrollment of adolescents in pre-HIV vaccine clinical trials P13e12 L Peralta
xxxviii Antiviral Therapy 11, Supplement 2 AIDS 2006 prelim matter 15/8/06 15:44 Page xxxix
230 Selectively willing and conditionally able: motivations, deterrents, and P13e13 barriers to participation in HIV vaccine trials among women at “high risk” of HIV infection in Philadelphia C Voytek
230 African American participation in HIV preventive clinical trials – P13e14 A preliminary examination of willingness to participate in HIV vaccine clinical trials and the impact of HIV vaccine and other clinical trial knowledge on willingness KS Washington
231 Willingness to participate in HIV vaccine trials: The impact of trial P13e15 characteristics (Project VIBE) PA Newman
231 Willingness to participate in future HIV vaccine trials among persons P13e16 at risk in South India M Suhadev
232 How do adolescents compare with adults in attitudes and outcomes P13e17 in Vaccine Preparatory Cohort Studies. L Bekker
232 Factors influencing participation and retention of healthy volunteers in P13e18 a phase I trial of a preventative candidate HIV vaccine in New York city S Vasan
233 Circumcision as a component of standard of prevention for HIV P13e19 vaccine trials: potential impact on public sector service G de Bruyn
Poster Session: Topic 14
Social Issues/Ethical Issues/Access Issues/Regulatory Issues
234 Development of national guidelines for research and development of P14-01 HIV/AIDS vaccines: The Kenyan experience WG Jaoko
234 Ethical challenges in the recruitment and informed consent process P14-02 for an HIV vaccine efficacy trial in Haiti P Joseph
235 What do stakeholders in South Africa view as major ethical challenges P14-03 in adolescent HIV vaccine trials? C Milford
235 Payment for HIV vaccine research participation: South African P14-04 stakeholders’ perceptions C Milford
236 Gender concerns in HIV vaccine research: reflections from key P14-05 stakeholders in East Africa L Nyblade
236 Building a new framework to explore global demand scenarios for P14-06 preventive AIDS vaccines G Gandhi
237 Expanding resources, expanding options: funding for HIV vaccine P14-07 research and development (R&D) between 2000 and 2005. K Fisher
237 Toward a collaborative model of HIV vaccine research P14-08 K Fisher
AIDS Vaccine 2006 xxxix AIDS 2006 prelim matter 15/8/06 15:44 Page xl
238 Speeding vaccine R&D – what does history tell us? P14-09 H Wong
238 HIV vaccine research & development: modeling the path to speedier success P14-10 H Wong
239 Community activism, good science and the politics of HIV vaccine P14-11 research in Canada: A case study of Phase IIb industry-sponsored trials with marginalized women D Elliott
239 Be part of something Big! Promoting volunteerism in HIV vaccine P14-12 clinical trials MdR Leon
xl Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 1
OPENING AND PLENARY SESSIONS Plenaries and Symposia 14/8/06 16:53 Page 2 Plenaries and Symposia 14/8/06 16:53 Page 3
Amsterdam, Netherlands 29 August–1 September 2006
OPENING SESSION
OS05
High Noon: 20 Global Problems, 20 Years to Solve Them
Jean-François Rischard
Former Vice-President Europe World Bank
This lecture reviews the main and most pressing global prob- lems, why they are not being solved by the existing interna- tional system, and what one could imagine in the way of new approaches for getting them solved faster and deeper. It’s a mind-opening, provocative lecture that leaves no one indif- ferent, and that gets people to think about the future in new, unaccustomed ways. The lecture is based on Mr. Rischard’s book High Noon: 20 Global Problems, 20 Years to Solve Them (Basic Books, NY, USA 2003).
AIDS Vaccine 2006 3 Plenaries and Symposia 14/8/06 16:53 Page 4
Amsterdam, Netherlands 29 August–1 September 2006
PLENARY SESSION 1
PL01-01 PL01-02
HIV prevention: what works, what’s needed and The human HIV vaccine pipeline: an update what’s being done L Corey H Rees University of Washington and Fred Hutchinson Cancer Reproductive Health and HIV Research Unit, University of the Research Center, Seattle, WA, USA Witwatersrand, Johannesburg, South Africa The number of candidate vaccines that are designed to elic- It is widely accepted that vaccination is the most effective it T cell responses has markedly increased in the last 24 public health intervention for the prevention of infectious months. In the HVTN system, over 15 different vaccine diseases. Twenty years ago, the scientific and public health prototypes have been evaluated. Unfortunately, several world embraced the idea of developing an HIV vaccine with prototypes have been less than optimally immunogenic as both confidence and enthusiasm. By 2006, the scientific and measured by γIFN producing T cells, while others have economic challenges that characterize these efforts are wide- elicited high levels of T cell responses. This disparity has ly recognized, and earlier enthusiasm has been tempered caused the human trialists to look at whether increased with a degree of caution. While HIV vaccine research has standardization of preclinical testing in non-human pri- slowly progressed, the rest of the HIV prevention world has mates should precede initiation of human clinical trials. continued in its efforts to develop new technologies and Two separate groups (Vaccine Research Center and interventions to expand the armamentarium of options Eurovac Foundation) have shown immunogenicity of DNA available to country programmes. Early intervention efforts, vaccines, especially for priming a subsequent Ad5 or such as condom use and safer sex messages, were based on NYVAC vector boost. Overall, replication defective aden- limited data and on common sense. In the last 10 years, the oviruses appear to elicit the most consistent immune disappointing impact of prevention programmes in some set- responses in humans, eliciting γIFN producing T cell tings has led to a resurgence of interest in new prevention responses in >75% of recipients at levels that range from technologies other than vaccines, and for new prevention 200–500 spot forming cells/106 PBMC. Immunogenicity of interventions. From a country perspective, the costs involved Ad5 vaccines is influenced by dose, HIV insert and prior in the scale up of prevention programmes has led to increas- immunity of the recipient to Ad5. Efficacy trials are under- ing need for rigor in the evaluation of the efficacy of inter- way to define if the magnitude of current responses with ventions, be they biologically or behaviorally based. Ad5 or DNA Ad5 vectors are useful for controlling This presentation will review the state of the art in HIV pre- viremia or decreasing acquisition of infection. vention. It will examine the evidence for a variety of inter- ventions including structural interventions, behavior change programmes, clinical interventions, and new prevention technologies. It will examine the challenges of designing PL01-03 studies to evaluate some of these programmes, and the limi- tations of some of the data that we have before us on which The HIV-1 envelope glycoproteins: implications decisions must be made. Specifically, the paper will cover for vaccine design behaviour change programmes including adolescent and school programmes, gender based programmes, economic JP Moore and lifestyle interventions and the role of media. The role of VCT in different contexts and the importance of a variety of Weill Medical College of Cornell University, New York, USA risk reduction messages will be assessed. Current informa- tion about new and existing technologies including male and The HIV-1 envelope glycoprotein complex is the focus of female condoms, diaphragms, microbicides and pre expo- attempts to make a vaccine wholly or partially based on sure prophylaxis will be reviewed. The paper will examine the induction of neutralizing antibodies. The properties of the state of knowledge about STI interventions from the gen- this complex create significant challenges that will be eralities of STI treatment programmes to the specifics of reviewed in this presentation. Thus, the structure of the what is known about the suppression of Herpes Simplex. Env complex has evolved to limit its immunogencity and New data on circumcision will be presented, and the pro- to minimize the exposure of binding sites for any antibod- grammatic implications discussed. Finally, the presentation ies that are induced. Moreover, the gp120, and perhaps the will make suggestions about what is the state of the art for gp41, elements of the Env complex are biologically active prevention programmes as currently understood. molecules that interact with cell surface receptors to trans-
4 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 5
Amsterdam, Netherlands 29 August–1 September 2006
duce immunosuppressive signals. The concentrations of gp120 usually delivered in vaccination regimens are capa- ble of activating such signals, which may constitute yet another immune defense mechanism. Possible ways to overcome these defenses will be discussed.
AIDS Vaccine 2006 5 Plenaries and Symposia 14/8/06 16:53 Page 6
Amsterdam, Netherlands 29 August–1 September 2006
PLENARY SESSION 2
PL02-01 tion leads to a dynamic network of complex and inter- related recombinants. CRF may emerge when strains from high-risk groups enter larger, but lower-risk, heterosexual Global epidemiology of HIV: influence of risk fac- networks, where lower dual infection rates lead to a stabi- tors and social networks on the complexity and lization of their structure. Bridges between high and low relationships of inter-subtype recombinant strains risk social networks are effective targets for interventions in Asia and Africa to limit the overall complexity of strains in the pandemic.
FE McCutchan1, GH Kijak1, S Tovanabutra1, E Sanders- Buell1, JN Chiu1, C Beyrer2, M deSouza3, M Arroyo3, M Robb1, D Birx1.* and N Michael1 PL02-02
1 US Military HIV Research Program, Rockville, MD, USA; 2 Johns The challenge of HIV global diversity for vaccine Hopkins University Bloomberg School of Public Health, antigen design Baltimore, MD, USA; 3 Armed Forces Institute of Medical B Korber1,2, T Bhattacharya1,2, C Brander3, W Fischer1, Sciences, Bangkok, Thailand. *Current address: Centers for N Frahm3, F Gao4, B Hahn5, B Haynes4, T Leitner1, Disease Control and Prevention, Atlanta, GA, USA N Letvin5, B Walker4, K Yusim1 and M Zhang1 Background: Candidate HIV-1 vaccines are being evaluat- ed in East Africa and Thailand, where inter-subtype recom- 1 Theory Division, Los Alamos National Laboratory, Los Alamos, binant forms often constitute a substantial fraction of the NM, USA; 2 The Santa Fe Institute, Santa Fe, NM, USA; 3 circulating strains. Dual infection, more common in high Partners AIDS Research Center, Massachusetts General Hospital, risk, multiply exposed individuals, is thought to be the Harvard Medical School, Boston, MA USA; 4 Duke University proximal source of new recombinants. Most recombinant School of Medicine, Durham, NC, USA; 5 Department of strains do not appear to establish population-level spread; Medicine, Beth Israel Deaconess Medical Center, Harvard Medical the majority of identified recombinants are unique, isolat- School, Boston, MA, USA ed only from a single individual. Factors that lead to the emergence of circulating recombinant forms (CRF) are Background: HIV evolves rapidly, and there is a complex incompletely understood, but important to document. global map of HIV-1 genetic subtypes and their recombi- Objectives: Analyse the structure and relationships of inter- nants. Some regional epidemics are dominated by one sub- subtype recombinant HIV-1 from different risk groups in type, and vaccine strategies can be framed in the context of Asia and East Africa, and identify factors leading to emer- this more limited diversity, however even one subtype har- gent CRF. bours extensive variation. Epidemics in other geographical Methods: Virtually complete genome sequences representing regions result from a complex array of subtypes and inter- 125 inter-subtype recombinant strains, mostly collected subtype recombinant forms. between 1996 and 2005, were tabulated with respect to com- Objectives: Our overarching objectives are to better under- ponent subtypes and breakpoints. Newly developed software stand the interplay between the host response and viral evo- (Breakpoint Blast) was used to evaluate the frequency of lution, and the limitations and extent of HIV variation. We breakpoints across the genome and to identify breakpoints then incorporate this knowledge into rationally designing common to two or more strains. The relationship of strain reagents for assessing immune responses to diverse HIV-1 complexity and shared breakpoints to date of isolation, geo- strains, and to inform decisions regarding vaccine antigen graphical region, and risk group was evaluated. design for subsequent testing. Results: Sixty-six recombinant strains from East Africa, Results & Conclusions: The circulating recombinant form exclusively from heterosexual or perinatal transmission, nomenclature is beginning to collapse as more complex and 59 from Asia, about half from injecting drug users recombinants are being found that do not readily fit into (IDU) and half from heterosexual transmission, were avail- groupings for molecular epidemiology and vaccine design. In able for analysis. In Asia, strains from IDU shared many terms of T-cell immunity, we have found that even within- more breakpoints than those from heterosexual contact. clade consensus sequences fail to detect many natural CRF maintained their structure during wide geographical responses, and we can detect more responses using a strategy spread but also shared breakpoints with other recombinant of peptide synthesis (toggled-peptides) that allows many vari- strains composed of the same subtypes in their respective ants to be included. The complexity of HIV diversity has dri- geographical regions. ven us towards design and testing of vaccine antigens that Conclusions: Recombinant strains reflect the social net- have the theoretical potential for stimulating enhanced intra- works in which they spread. In multiply exposed, high risk and inter-subtype cross-reactivity. Our first attempts compare groups, repeated cycles of dual infection and recombina- natural strains to artificial M group central antigens, and
6 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 7
Amsterdam, Netherlands 29 August–1 September 2006
small animal vaccination studies show some promise. We have gone on to design sets of artificial HIV proteins (mosaics) that maximize the coverage of epitope variants in a given population.
PL02-03
Toll-like receptor ligands influence the magnitude and quality of T cell responses
R Seder, U Wille-Reece and M Roederer
Vaccine Research Center, NIAID, Bethesda, MD, USA
T cells play a critical role in mediating protection against HIV, malaria and tuberculosis. At present, there are no clear- ly defined prospective immune correlates of protection for these infections. A mechanism by which T cells mediate their function is through production of specific cytokines. However, there can be substantial heterogeneity of T cell cytokine responses. Hence, methods to assess the functional heterogeneity of T cell responses and correlate this with pro- tection will be useful for successful vaccine development. In a series of studies we focused on how Toll-like receptor (TLR) ligands and agonists that activate distinct dendritic cell (DCs) subsets influence the magnitude and quality of Th1 and CD8+ T cell responses. NHP immunized with HIV Gag pro- tein and a TLR7/8 agonist or a TLR9 ligand (CpG oligodeoxynucleotides, CpG ODN) had significantly increased Gag-specific Th1 and antibody responses compared to animals immunized with HIV Gag protein alone. Importantly, optimizing vaccine delivery by conjugating the HIV Gag protein to the TLR7/8 agonist (Gag-TLR7/8 conju- gate) or administering the protein and the TLR 7/8 agonist in an oil-based emulsion dramatically enhanced the magnitude and influenced the quality of the T cell responses. Moreover, in NHP immunized with these optimized formulations with the TLR7/8 agonist there was a high frequency of Th1 and CD8+ T cells secreting IL-2, IFN-γ and TNFα. We speculate that the TLR 7/8 agonist imprints poly-functional cytokine producing T cells by activating all DC subsets. The genera- tion of such poly-functional T cell responses should be opti- mal for both effector function and capacity for sustained memory. These studies should inform vaccine development for HIV, Malaria and Tuberculosis.
AIDS Vaccine 2006 7 Plenaries and Symposia 14/8/06 16:53 Page 8
Amsterdam, Netherlands 29 August–1 September 2006
PLENARY SESSION 3
PL03-01 of critical importance. The commitment to initiate pivotal efficacy trials may be difficult. Such trials require support from both governments and communities. Sponsors must More than half the world’s population is under 25 create infrastructure, establish laboratory assays and large- and all are at risk for HIV… scale manufacturing methods. PhIIB or Test of Concept tri- als may be used to generate preliminary data on the effect L-G Bekker of vaccines on incidence of HIV infection or the course of infection in vaccinated persons who subsequently become Desmond Tutu HIV Centre, Capte Town, South Africa infected with HIV. Objectives: To define the circumstances in which PhIIB Test More than half of the worlds’ population is under 25 years of of Concept trials may be most or least useful. To delineate age- the largest population of young people the world has challenges inherent in such trials. ever seen. In South Africa, 54% of the population is less than Methods: A meeting was convened by WHO/UNAIDS and 24 yrs amounting to more than 15 million children in our IAVI, with, clinicians, biostatisticians, and trial sponsors country. All are at risk for HIV infection. from industrialized and developing countries.. A report will UNAIDS reports that in 2004 there was 14000 new infection be issued in order to clarify issues on which there is consen- daily and almost 100% of these occurred in the developing sus or require more attention. world. 1200 occurred in the 14-25 year olds and half of these Results: The group supported the value of Ph2B Test of were in females. Looking more carefully at the 14–24 y Concept trials, as part of a well defined product development group, two thirds occurred in Sub-Saharan Africa with plan. Such trials may be most useful when there is not a firm another quarter occurring in Central Asia and Eastern scientific basis for selecting a particular candidate vaccine, or Europe. An HIV prevalence survey carried out in nearly 12 when the relationship between short-term outcomes such as 000 South African youth and recently reported in AIDS, biomarkers or early course of infection and later clinical show a prevalence rate of 2.3% in boys and 4.1% in girls at events is not clear. If the formulation of a vaccine, the manu- age 15, increasing steadily to 5.9% in boys and a staggering facturing methods or the assays are not final, a Ph2B trial 31.2% in girls by age 21 y (Pettifor et al. AIDS 2005: may guide further product development. Where product 19;1525–534) development is nearly complete and endpoints are simple Studies show that the risk overwhelming risk factors for (e.g. infection versus no infection), such trials could create a infection in these young people is sex, with numbers most dri- lengthier and more costly path to licensure/approval. ven by heterosexual sex in young women in Sub-Saharan Conclusions: A number of challenges lie ahead once an HIV Africa. This is because HIV transmission is more efficient vaccine candidate is shown to protect against HIV infection or from men to women. In addition, youth may be vulnerable in progression to AIDS. Ph2B trials may provide one relatively their decision making around sexual practices Sexual behav- rapid and efficient method of evaluating candidate vaccines.. iour surveys have been carried out extensively in South Africa Such information could be invaluable in guiding the design of with at least 35 published articles on HIV knowledge and 75 future vaccines and vaccine trials. The greatest challenges lie in papers on sexual behaviour and risk. These studies confirm how to explain the purposes and limitations of such trials to that South African youth have an early sexual debut (<15 yrs) participants, communities, and governments and how to move and indulge in relatively high risk behaviour. This paper will forward once some degree of efficacy has been indicated. discuss the adolescent susceptibility to HIV infection: gender, age and race issues. PL03-03
PL03-02 Overview of the global HIV vaccine enterprise: progress and priorities for the future Optimizing trial designs to accelerate HIV vaccine clinical research: Phase IIB test of concept trials J Esparza and S Malone
PE Fast on behalf of the WHO/UNAIDS/IAVI Working Interim Enterprise Secretariat, Bill & Melinda Gates Foundation, Group on Ph2B Test of Concept Trials Seattle, WA, USA After more than 20 years of intense research, an HIV vaccine International AIDS Vaccine Initiative, New York, NY, USA is not yet available, and the scientific community has come to understand that developing an HIV vaccine presents one of Background: Designing clinical development plans that can the major challenges confronting biomedical research today. efficiently select HIV vaccines that will prevent HIV/AIDS is To confront this challenge, an international group of
8 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 9
Amsterdam, Netherlands 29 August–1 September 2006
scientists proposed in 2003 the creation of a Global HIV Vaccine Enterprise (Klausner RD et al. Science 2003; 300:2036–2039). The Enterprise is being developed, not as a new organization, but as an alliance of independent organi- zations (funders and implementers of research) acting as part- ners to accelerate the development of a preventive vaccine for HIV. The Enterprise does not propose to replace the creativi- ty of individual investigators, but to complement it with the structures and resources needed to harness new knowledge to accelerate the development of a much needed HIV vaccine. In January 2005 the Enterprise published its first Scientific Strategic Plan, developed in consultation with over 140 sci- entists from 15 countries (Coordinating Committee of the Global HIV/AIDS Vaccine Enterprise. PLoS Med 2005; 2:e25). The plan identifies the major roadblocks in vaccine research in six key areas: Vaccine discovery; laboratory stan- dardization; product development and manufacturing; clini- cal trials capacity; regulatory issues; and intellectual property. Several Enterprise partners are beginning to implement the recommendations of the Scientific Strategic Plan. In 2005 the U.S. National Institutes of Health launched the Center for HIV/AIDS Vaccine Immunology (CHAVI), a virtual consor- tium of 80 investigators from 35 different institutions around the world. In July 2006 the Bill & Melinda Gates Foundation committed funding to support the “Collaboration for AIDS Vaccine Discovery” (CAVD), a network of 16 vaccine dis- covery and laboratory standardization consortia grouping more than 165 researchers in 20 countries. Additional activ- ities have been, or will soon be announced in support of the Global HIV Vaccine Enterprise in US and Europe. .
AIDS Vaccine 2006 9 Plenaries and Symposia 14/8/06 16:53 Page 10 Plenaries and Symposia 14/8/06 16:53 Page 11
SYMPOSIA SESSIONS Plenaries and Symposia 14/8/06 16:53 Page 12 Plenaries and Symposia 14/8/06 16:53 Page 13
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 1: BASIC FINDINGS AND THEIR IMPACT ON VACCINE DESIGN
S01-02 S01-03
Novel HIV-1 neutralizing human antibodies Do viral subtype and mode of transmission define recognizing conserved conformational epitopes the envelope characteristics of transmitted viruses? on gp41 E Hunter 1, S Gnanakaran2, E Karita3, J Mulenga4, DS Dimitrov1, M Zhang1,2, V Choudhry1, B Vu1, JM Decker 5,6, F Bibollet-Ruche5,6, E Chomba7, BH Hahn5,6, IA Sidorov1, V Tenev1, A Choudhary3, H Lu4, AR Borges5, GM Shaw5,6, BT Korber 2, S Allen1 and CA Derdeyn1 F Cham3, GM Stiegler6, HWD Katinger6, DC Montefiori7, V Polonis5, GV Quinnan, Jr.3, S Jiang4, and CC Broder3 1 Emory University Vaccine Center, Atlanta, GA, USA; 2 Los Alamos National Laboratory, Los Alamos, NM, USA; 3 Rawanda 1 National Cancer Institute, Frederick, MD, USA; 2 SAIC-Frederick, Zambia HIV Research Group, Kigali, Rwanda; 4 Blood Transfusion Inc., Frederick, MD, USA; 3 Uniformed Services University of the Service, Lusaka, Zambia; 5 Howard Hughes Medical Institute, Health Sciences, Bethesda, MD, USA; 4 New York Blood Center, Birmingham, AL, USA; 6 University of Alabama at Birmingham, New York, NY, USA; 5 U.S. Military HIV Research Program, Birmingham, AL, USA; 7 Rawanda Zambia HIV Research Group, Rockville, MD, USA; 6 Institute of Applied Microbiology, Vienna, Lusaka, Zambia Austria; 7 Duke University, Durham, NC, USA Background: We previously reported that the subtype C HIV-1 Background: The development of vaccine immunogens that quasispecies passes through a severe genetic bottleneck during can elicit highly potent and broadly HIV-neutralizing anti- heterosexual transmission that selects for compact viruses with bodies (nAbs) remains a major challenge. Only several such high sensitivity to neutralization by antibodies in the infecting antibodies have been identified, including 2F5, 4E10 and partner’s plasma. While studies of sub-type A infections have Z13, which can bind peptides from the gp41 subunit of the also indicated that viruses with significantly shorter V1V2 HIV envelope glycoprotein (Env). However, immunogens loops are responsible for initiating infection, this does not based on these peptides have not elicited broadly nAbs. appear to be the case for early sub-type B isolates. Objectives: To identify novel cross-reactive HIV-1-neutraliz- Approximately half of all new infections in both sex workers ing gp41-specific human monoclonal antibodies and charac- and STD clinic patients appear to be the result of transmis- terize their epitopes with potential as vaccine immunogens. sion of more than one virus variant, contrasting with the Methods: Competitive antigen panning (CAP) of a phage- extreme bottleneck observed in discordant couples and argu- displayed antibody library derived from long-term nonpro- ing that other factors may influence this process. gressors with high concentrations of broadly nAbs. CAP was Objectives: To determine the diversity and nature of subtype performed against tagged (biotinylated) gp140s from C and subtype A viruses isolated from recent transmission CM243, 89.6 and R2 in presence of excess non-biotinylated pairs, and to compare the results with those from other early gp120s from the same isolates. infection cohorts. Results: Six antibodies (m43–8) were selected and tested as Methods: We have analysed by single genome analysis the IgG1s, Fabs or scFvs in four different panels of primary iso- Env variants in an additional 14 transmission pairs, including lates from different clades in PBMC/replication competent 9 subtype C and 4 subtype A transmissions. The nature of the virus and cell line/pseudovirus-based assays. Their neutraliz- genetic bottleneck and the diversity of recently infecting vari- ing activity was, on average, comparable to that of 2F5 and ants were analysed by direct nucleotide sequencing of the 4E10 and significantly higher in PBMC-based assays than in PCR amplicons. cell line-based assays likely due to the relatively low CCR5 Results: In 13/14 epidemiologically-linked heterosexual concentration on the surface of PBMCs and related slow transmission pairs (i.e. where the chronically infected partner kinetics of fusion. They bound with high affinity to gp140s transmits to their spouse), irrespective of subtype, a single from various primary isolates but not to denatured gp140 or virus appeared to establish infection since all of the recipient a panel of gp41 peptides indicating a conformational nature variants emanated from a single branch of the donor phylo- of their epitopes. All antibodies competed strongly with the genetic tree, consistent with a major genetic bottleneck at the cluster IV antibody T3, variably with each other, and variably time of transmission. This bottleneck was evident regardless and relatively weakly with 2F5, 4E10, Z13, and the cluster V of whether gag or env amplicons were analysed. In only one antibody D3. case, where an acutely infected partner transmitted to their Conclusions: Our findings indicate the existence of new spouse, did we observe evidence for transmission of two viral conserved conformational neutralization epitopes on gp41 genotypes in the newly infected partner. In contrast, one that may have potential as HIV vaccine immunogens and as newly infected partner of an unlinked transmission event har- targets for therapeutics. boured a diverse but monophyletic virus quasispecies, indi- cating transmission and propagation of more than one variant from a single donor.
AIDS Vaccine 2006 13 Plenaries and Symposia 14/8/06 16:53 Page 14
Amsterdam, Netherlands 29 August–1 September 2006
Conclusions: Transmission bottlenecks appear to be extreme responses were not elicited by transient viremia, this phe- in the case of heterosexual transmission in Zambia and nomenon could help explain immune responses measured in Rwanda, but there is a potential role for other factors and highly exposed, HIV seronegative subjects. STDs in modulating such bottlenecks.
S01-05 S01-04 HIV-containing intracellular compartments mediate Transient viremia and immune responses pre- HIV transmission from macrophages and DCs ceding sustained, systemic viral replication: implications for evaluating acute HIV infection M Marsh, A Pelchen-Matthews, M Deneka and R Byland and vaccine efficacy Cell Biology Unit, Medical Research Council Laboratory for PJ Norris1,2,3, EW Fiebig2, BL Pappalardo1, EL Delwart1, Molecular Cell Biology, University College London, UK and MP Busch1,2 Background: In macrophages HIV has been reported to bud 1 Blood Systems Research Institute, San Francisco, CA, USA; 2, 3 into organelles with some characteristics of late endosomes Departments of Laboratory Medicine and Medicine, University of (Pelchen-Matthews et al. J Cell Biol 2003; 162:443–455; California, San Francisco, San Francisco, CA, USA Raposo et al. Traffic 2002; 3:718–729) and not at the cell surface. Morphologically and functionally, this compartment Background: HIV is an RNA virus that has a high error rate resembles a compartment in dendritic cells that mediates in replication. These errors generate viral diversity that trans infection of T cells. appears to benefit the virus at the population level by allow- Objectives: To characterize the HIV assembly compartment ing rapid escape from host immune responses. During the in macrophages. period of viral transmission with an inherent genetic “bottle- Methods: Immunolabelling of cryosections and electron neck”, a high error rate could result in abortive infection if microscopy, immunofluorescence microscopy and HIV the viral reproduction ratio fell below one. infectivity assays. Objectives: To evaluate whether transient viremia could be Results: We show that a number of markers previously sug- detected prior to sustained ramp-up of HIV viral load in gested to be located in late endosomes are associated with acute HIV infection and to determine if early “blips’ of distinct cellular compartments in macrophages. In particular, viremia induced host immune responses. several tetraspanins including CD9, CD53, and CD81 are Methods: Paid plasma donors are routinely monitored for associated with a set of multivesicular endocytic organelles seroconversion to HIV and other viruses, and 15 donors that do not contain CD63, which co-localises with markers were identified who seroconverted to HIV and had HIV for late endosomes and lysosomes. In infected macrophages, PCR negative samples preceding viral ramp-up. Subjects the HIV Gag and Env proteins are not associated with the donated with a mean inter-donation interval of 4 days. bulk of the CD63-containing late endosome/lysosome com- Once subjects seroconverted, they were counseled and partment, but show significant overlap with the deferred from further donation. We re-tested archived CD9/CD53/CD81 compartment. Moreover, in these infected donor panels using a sensitive, qualitative PCR HIV detec- cells some CD63 is now seen to overlap with CD9/ CD53/CD81 and the organelles show some indication of re- tion assay. Additionally, plasma IFN-γ levels were deter- mined in seven of the patients examined. organization. Significantly, in macrophages transfected to Results: In two of 15 panels tested, no low-level viremia express a HIV envelope protein, Env appears to target the was detected. In seven subjects low-level viremia was CD9/CD53/CD81 compartment. The CD9/CD53/CD81 con- detected at the time point immediately preceding viral taining organelles are morphologically complex, with exten- ramp-up. Six subjects showed transient viremia preceding sive internal membranes, and they share a number of viral ramp-up. Low-level viremic samples were found 9 to properties with a compartment seen in monocyte-derived 25 days (median= 18 days) prior to the first sample with dendritic cells where virions internalised by C-type lectins can be sequestered during the process of trans-infection of T cells. >100 copies/ml. Each sample was separated from the ramp- up period by at least one HIV RNA negative sample. Based Conclusions: In macrophages, the components required for on replicate testing, the estimated viral load was 1 to 10 the production of infectious HIV are sorted to an endocytic compartment identified by the presence of the tetraspanins copies/ml during transient viremia. Plasma IFN-γ rose rapidly during viral ramp-up, but did not appear to be CD9/CD53/CD81, and that the virus causes CD63 to become induced by transient viremia. associated with these organelles. We suggest these organelles Conclusions: Low-level viremia is not uncommon in the peri- may target HIV to infectious synapses during cell-to-cell od prior to sustained viral replication. It is possible that early transmission of virus and may represent a pool of infectious low-level viral replication could result in abortive infection if virus that is resistant to antibody neutralization. viral fitness were low or high levels of anti-HIV immunity were induced by prior vaccination. While immediate immune
14 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 15
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 2: NOVEL VECTORS AND VECTOR COMBINATIONS
S02-01 toxicity and immunogenicity of the vectors. With more favor- able safety and efficacy profiles. Strategic development of novel vectors and crite- ria for advanced clinical development S02-02 GJ Nabel Novel adenovirus vector-based vaccines for HIV-1 Vaccine Research Center, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, D Barouch1, D Roberts1, A Nanda1, P Abbink1, MD, USA A Thorner1, A Lemckert2, L Holterman2, R Vogels2, M Havenga2, and J Goudsmit2 Background: The design of highly effective HIV vaccines requires identification of the most effective platforms for 1 Beth Israel Deaconess Medical Center, Boston, MA, USA; delivery of antigenic targets. Novel vector systems, based 2 Crucell Holland BV, Leiden, The Netherlands upon adenovirus, AAV, and MVA, are currently being evalu- ated in preclinical and clinical studies. Because considerable resources are required to move promising vaccine candidates Background: Replication-incompetent recombinant aden- towards clinical use, it is essential to develop a clearly defined ovirus serotype 5 (rAd5) vector-based vaccines for HIV-1 strategy to define vectors that minimize toxicity while opti- have shown considerable promise in preclinical studies and mizing immune responses. early phase clinical trials. However, the high frequency of Objectives/Aim: Recombinant, replication defective adenovi- pre-existing anti-Ad5 immunity in the developing world may ral serotype 5 (rAd5) vectors elicit potent cellular and prove a limitation of rAd5 vectors. To overcome this prob- humoral immune responses; however a large percentage of the lem, we have developed and evaluated a series of novel rare prospective vaccine recipients have neutralizing antibodies serotype and chimeric rAd vectors. associated with previous Ad5 infection that can potentially Methods: Recombinant Ad vectors expressing SIV Gag were mitigate the effectiveness of the vectors. In addition, admin- produced in E1-complementing cell lines. Seroprevalence in istration of rAd5 vaccines has resulted in dose-limiting sub-Saharan Africa was evaluated using luciferase-based Ad adverse responses, including transient low grade fevers. The neutralization assays. Cellular and humoral immune respons- goal of these studies is to understand the mechanism of es were assessed in mice and rhesus monkeys using tetramer, induction of immune responses and vector-associated toxici- ICS, and ELISPOT assays as well as ELISAs. ty, and to assess the ability of DNA priming to boost rAd vac- Results: We have developed rAd vector systems from six rare cines in humans with pre-existing Ad seropositivity, in an Ad serotypes from subgroup B (Ad11, Ad35, Ad50) and sub- effort to guide selection of next generation vectors. group D (Ad26, Ad48, Ad49). Seroprevalence studies demon- Methods: Ad5 primary receptor recognition and binding strated that these Ad serotypes were all rare in both pediatric sequences have been shown to reside in the adenoviral fiber and adult populations in sub-Saharan Africa. Moreover, region, To determine the contributions of specific domains these vectors were immunogenic in both mice and rhesus within the fiber to target different cell types and to under- monkeys and effectively circumvented anti-Ad5 immunity. stand their role in toxicity, and immunogenicity, we prepared Recombinant Ad26 proved the most immunogenic of these vectors with mutations in these domains and analysed tissue six rare serotype rAd vectors, although all were still less transduction and specific immune reactivity following intra- immunogenic than rAd5 vectors in the absence of pre-exist- muscular injection into mice. We also evaluated the relative ing anti-Ad5 immunity. We also constructed chimeric rAd5 efficacy of rAd5, rAd35, chimeric and synthetic rAd vectors vectors in which the seven hexon hypervariable regions in mice and non-human primates. (HVRs) were exchanged with the corresponding regions from Results: These studies demonstrated rAd5 tropism for dendrit- the rare serotype vector Ad48. These chimeric rAd5HVR48 ic cells and monocytes, identifying a distinct segment of the vectors were stable through 15 serial passages in vitro. Ad5 fiber responsible for targeting to these cells that mediates Moreover, these vectors proved comparably immunogenic to both the induction of adaptive immune responses and vector rAd5 vectors in both mice and Rhesus monkeys and also toxicity. The pyrogenic response to the administration of rAd5 effectively evaded anti-Ad5 immunity. was not associated with CAR or integrin binding. Conclusions: Based on comparative assessments of sero- Conclusion: Next generation vectors will include rAd35, prevalence, immunogenicity, and manufacturability, we chimeric rAd35, chimeric rAd5 and other rare serotype vec- have selected rAd26 as the optimal rare serotype vector tors, as well as synthetic viral protein/DNA complexes. Use and rAd5HVR48 as the optimal chimeric vector to advance of these fiber modified and alternate serotype vectors will into Phase I clinical trials as novel rAd vector-based allow for development of improved vectors that take advan- vaccines for HIV-1. tage of the targeting and specificity of rAd while minimizing
AIDS Vaccine 2006 15 Plenaries and Symposia 14/8/06 16:53 Page 16
Amsterdam, Netherlands 29 August–1 September 2006
S02-04
Development of CMV vectors for an HIV vaccine
LJ Picker
Vaccine and Gene Therapy Institute and the Oregon National Primate Research Center, Oregon Health & Science University, Beaverton OR, USA
The ubiquitous β-herpes virus CMV exhibits a number of natural adaptations – including 1) elicitation and long-term maintenance of uniquely strong (mucosally-oriented) cellu- lar and humoral immunity, 2) lifelong persistence despite this robust immune response, 3) the ability to sub-clinical- ly re-infect and immunologically boost fully immune hosts, and 4) little pathogenicity in normal hosts – that make it a provocative candidate for development as a vaccine vector for chronic, elusive lentiviral pathogens like HIV/SIV (for which protection will likely require potent, long-lasting cellular and humoral immunity). We have used the Rhesus macaque (RM) model (RhCMV/SIV) to explore the hypothesis that the special adaptations of CMV vectors (immunogenicity, persistence, ability to re-infect immune hosts) can quantitatively and/or qualitatively improve on the immunogenicity of current vaccine strategies, and as a result, safely elicit protective lentivirus-specific immunity. We have constructed optimized RhCMV vectors expressing SIV gag, env, and a (detoxified) rev/nef/tat chimera. These vectors grow in vitro with wildtype kinetics despite expressing high levels of SIV proteins. With subcutaneous inoculation, all of the recombinant RhCMV efficiently (and benignly) infects both CMV-naive and wildtype CMV- infected RM at inocula as low as 100 pfu. These infections are characterized by persistent (probably permanent) uri- nary and salivary secretion infection, and either low level, transient or no measurable viremia. All primary infections and re-infections with these vectors result in rapid devel- opment of mucosa-oriented, SIV-specific CD4+ and CD8+ T cell responses that are similar to or larger than those elicited to native RhCMV Ags in primary infection. These responses can be significantly boosted with re-application of the same vector. Moreover, RhCMV vectors with differ- ent inserts can be applied sequentially with similar induc- tion of specific T cells, indicating that modified CMVs might be developed as a unique vaccine system that is insensitive to (and perhaps even enhanced by) pre-existing anti-vector immunity. The potential application of such CMV vectors will be discussed.
16 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 17
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 3: STREAMLINING VACCINE DEVELOPMENT FOR HIV AND OTHER DISEASES
S03-02 S03-03
Malaria: status of the field and the RTS,S vaccine Development of HPV-specific vaccines: current as a case study status and future developments
J Cohen L Gissmann
German Cancer Research Center (DKFZ) Heidelberg, Germany GlaxoSmithKline Biologicals, Vaccines for Emerging Diseases & HIV, R&D, Rixensart, Belgium Since the association between the infection with certain (“high-risk”) types of human papillomaviruses (HPV) and The portfolio of malaria vaccine candidates that are today at anogenital tumors (in particular cancer of the uterine cervix) some stage of discovery or development is impressive. More has been firmly established, academic institutions and subse- than 30 antigens have been identified as potential vaccine quently pharmaceutical companies have been pursuing the candidates. Eighty vaccines are in advanced preclinical or development of prophylactic and therapeutic vaccines. early clinical development, 30 have entered or completed The first prophylactic vaccine based on virus-like particles Phase I in non-endemic or in malaria endemic regions and 2 (VLP) of four different HPV types that efficiently induce neu- have reached PoC in the main target population. tralizing antibodies has just been licensed in Mexico and the Recently, a long awaited breakthrough has occurred in this US. Yet there is ample room for improvement and investiga- field. The RTS,S vaccine, presently being developed by tors have taken up the challenges for second generation vac- GlaxoSmithKline in partnership with PATH-Malaria cines that might be effective against a wider range of HPV Vaccine Initiative (MVI) has demonstrated unprecedented types and be more suitable for use in resource-poor settings and significant efficacy against clinical malaria episodes with the worldwide highest burden of cervical cancer. (VE=35%; 95% CI: 21.6–46.6; P<0.0001), including severe In contrast to prophylactic strategies the success in HPV- forms of the disease (VE=48.6%; 95% CI: 12.3–71.0; specific immune therapy is lagging behind and more basic P=0.02) in young Mozambican children, over an 18 months work needs to be done to fully comprehend the immune period following vaccination. mechanisms that are responsible for rejection of established During the 15 years preceding this successful PoC study, HPV infections and associated diseases. It is particularly extensive preclinical and adjuvant research work was con- interesting to combine prophylactic and therapeutic capaci- ducted, a robust and scalable vaccine manufacturing ties within a single vaccine and exploit its potential for use as process was developed and a comprehensive clinical devel- post-exposure prophylaxis. opment plan was implemented. The safety, tolerability and immunogenicity of the vaccine was established successively in non-immune adults, semi-immune adults and finally in children living in malaria endemic region, down to the age S03-05 of 1 year old. Concomitantly, efficacy of the vaccine against P. Falciparum infection was consistently demonstrated in a Genital herpes vaccines: parallels with AIDS series of Phase IIa challenge studies conducted at the vaccines WRAIR and in semi-immune adults, in a Phase IIb study conducted in The Gambia. DM Knipe1, T Dudek1, D Watanabe1, A Kaur2, Over the next 4 years a clinical development plan will be con- Y Hoshino3 and SE Straus3 ducted in collaboration with multiple clinical sites in Africa and in partnership with PATH-MVI. This plan will culminate 1 Dept. of Microbiology and Molecular Genetics, 2 New England in pivotal multi-centric Phase III studies and in the submis- Primate Research Center, Harvard Medical School, Boston, MA, 3 sion of a registration file to the competent regulatory author- National Institutes of Health, Bethesda, MD, USA ities as early as 2010. I will share with the audience some of the lessons we have learned in the course of our work on this malaria vaccine. No genital herpes vaccine has been demonstrated to be gen- erally effective or licensed for use. Similar to AIDS vaccines, genital herpes vaccines face huge challenges in that they must induce effective cellular and humoral immune responses and protect against a persistent infection. Like AIDS vaccines, herpes subunit vaccines have been relatively ineffective, and it has been hard to attenuate a replication-competent HSV-2 strain to derive a safe but immunogenic vaccine strain. Studies with a replication-defective HSV-2 virus strain as a
AIDS Vaccine 2006 17 Plenaries and Symposia 14/8/06 16:53 Page 18
Amsterdam, Netherlands 29 August–1 September 2006
genital herpes vaccine have shown its ability to prophylacti- cally immunize mice and guinea pigs and protect against intravaginal challenge with a heterologous virulent HSV-2 strain. These HSV-2 vaccines can also be used for therapeutic immunization to reduce recurrent genital lesions in the guinea pig model. These studies have raised several concepts that may have implications for AIDS vaccines. First, immu- nization with subunit vaccines has not been effective against either HSV-2 or HIV, likely because CD8 cellular responses were not induced to significant levels. Immunization with replication-defective HSV-2 strains has induced better protec- tion in mouse and guinea pig systems than a subunit vaccine. Second, the more highly protective herpes vaccines induce T cell responses that are rapidly recruited to infection sites. Similarly, we have shown that in macaques immunized with HSV vectors expressing SIV proteins, control of SIV chal- lenge by the IV route correlates with rapid mobilization of SIV Tat- and Gag-specific T cell responses. Third, immuniza- tion with the replication-defective HSV-2 strain induces bet- ter neutralizing responses than the subunit vaccines. Thus, replication-defective genital herpes vaccines share some of the desired properties of an AIDS vaccine: induction of both humoral and CD4 and CD8 T cell responses, induction of durable immune responses, and induction of T cell responses that are rapidly mobilized upon viral infection.
18 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 19
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 4: T CELL AND INNATE IMMUNITY
S04-02 S04-03
Protective signatures of antiviral immunity HIV-1-specific T cell responses and viral evolution in primary HIV-1 infection A Harari, C Cellerai and G Pantaleo M Altfeld, M Lichterfeld, X Yu, H Streeck, and T Allen Lab. of AIDS Immunopathogenesis, Div. of Immunology and Allergy, CHUV, Lausanne, Switzerland Partners AIDS Research Center, Massachusetts General Hospital and Harvard Medical School Division of AIDS, Boston, MA, USA Background: Antiviral T-cell responses are functionally and phenotypically heterogeneous and it has been difficult to link Background: HIV-1-specific CD8+ T cells in primary infec- different patterns of antiviral T-cell responses with virus con- tion are associated with the dramatic decline of peak trol or to identify correlates of protection. viremia, whereas their antiviral activity in chronic infection Methods: Functional (secretion of IFN-γ and IL-2, degranu- is less apparent. lation activity, perforin expression and proliferation) and Objectives/aim: To identify the earliest immunodominant tar- phenotypic (expression of CD127, CCR7 and CD45RA) gets of HIV-1-specific T cells in primary infection and to characterization of virus-specific CD4 and CD8 T cells was investigate factors associated with the impairment of HIV-1- performed in 9 LTNP, 99 HIV-infected subjects (prior or fol- specific T cell responses with progressive disease. lowing 1 year of ART) and 69 subjects infected with CMV, Methods: HIV-1-specific T cell responses and their TCR EBV or Flu. usage were assessed using flow cytometric assays and TCR Results: Combined expression of perforin and CD127 sequencing of sorted epitope-specific populations. revealed the existence of 3 populations of virus-specific CD8 Autologous virus sequences were determined by sequencing + – T cells: CD127 perforin cells representing the large majori- of virus from plasma. – – ty of Flu-specific CD8 T cells; CD127 perforin cells that Results: The analysis of HIV-1-specific CD8+ T cell respons- – were the majority of EBV-specific CD8 T cells; and CD127 es in a large cohort of individuals primary infection revealed + perforin cells that contained a large proportion of CMV- consistent immunodominance patterns that depended on specific CD8 T cells. HIV-1-specific CD8 T cells from both host genetics (HLA class I alleles) and the sequence of – – untreated progressors were mostly CD127 perforin cells the infecting virus. Longitudinal studies demonstrated that + while LTNP had significantly more CD127 CD8 T cells. the functional activity of these HIV-1-specific CD8+ T cells + Untreated progressors had more perforin cells than LTNP. was consistently higher in primary infection than later in + – – – – Of note, CD127 perforin , CD127 perforin and CD127 chronic infection. This change of HIV-1-specific CD8+ T cell + – + – – perforin were CD45RA CCR7 , CD45RA CCR7 and function between primary and chronic infection was linked + – CD45RA CCR7 CD8 T cells, respectively. Ag-specific pro- to the loss of virus-specific T helper cell responses, as well as + liferating CD8 T cells were contained within the CD127 a preferential loss of high-avidity CD8+ T cell clones. In con- – perforin cells population. Comparison of perforin expres- trast, the maintenance of HIV-1-specific CD8+ T cell function sion and CD107a mobilization for the different viruses and the preservation of the initially recruited clonotypic pat- showed a discrepancy between these two markers. In partic- tern were associated with low-level set point HIV-1 viremia. ular, high levels of CD107a mobilization were observed for Conclusions: These data demonstrate that functionally com- Flu despite the absence of perforin expression. After 1 year of petent and high-avidity HIV-1-specific CD8+ T cell clones ART, HIV-1-specific CD4 T-cell responses were consistently directed against a number of predictable immunodominant polyfunctional. In contrast, two distinct patterns of HIV-1- epitopes are primed during primary HIV-1 infection. Viral specific CD8 T-cell responses were observed. One group evolution, impairment of T helper cell function and loss of (n=32) developed HIV-1-specific IL-2 secreting CD8 T cells high-avidity virus-specific CD8+ T cells subsequently lead to while the other group (n=67) did not. In the latter group, in the impairment of these initial responses. addition to HIV-1-specific IL-2 secreting CD8 T cells, higher percentage of HIV-1-specific single IFN-γ secreting CD8 T cells, degranulation activity and sustained proliferative CD8 T-cell responses were also found. Conclusions: Perforin and CD107a mobilization represent independent markers of CD8 T cells with potential cytotoxic capacity. Furthermore, unlike HIV-1-specific CD4 T cells, CD8 T cells are not homogeneously recovered after ART. One third of the subjects developed a better response regard- ing magnitude and polyfunctionality.
AIDS Vaccine 2006 19 Plenaries and Symposia 14/8/06 16:53 Page 20
Amsterdam, Netherlands 29 August–1 September 2006
S04-04 form, but when injected together with an adjuvant, these for- eign proteins can generate robust immunity and long-lived memory to the antigen. In fact, the nature of the adjuvant is Recent results from macaque studies of Toll-like what determines the particular type of immune response that receptor (TLR) ligands on the replication of SIV follows, which may be biased towards cytotoxic T-cell and influenza after mucosal transmission responses, antibody responses, particular classes of T-helper responses, or antibody isotypes. Clearly, the ability of a vac- CJ Miller cine to skew the response toward a particular type is of para- mount importance, because different pathogens require Center for Comparative Medicine, California National Primate distinct types of protective immunities. Central to this issue is Research Center, University of California Davis, Davis, CA, USA a rare but widely distributed network of cells known as den- dritic cells (DCs). DCs, which have been called ‘Nature’s Background: Toll-like receptors are expressed on a variety of adjuvants,’ express pathogen recognition receptors, such as innate and adaptive immune cells and they are essential for the Toll-like receptors (TLRs) and C-type lectins, which host recognition of infection with a pathogen through bind- enable them to sense and respond to microbes or vaccines. ing of certain general types of molecules whose expression is Research in the last decade has demonstrated a fundamental absent or restricted in vertebrates. Ten TLRs have been role for DCs in initiating and controlling the quality and described in humans and viral TLR ligands include double- strength of the immune response, and emerging evidence sug- stranded RNA (TLR-3), single-stranded RNA (TLR-7/8) and gests an important role for distinct TLRs and C-type lectins unmethylated CpG dinucleotides (TLR-9). Engagement of in differentially regulating the Th1/Th2/T regulatory balance, TLRs results in rapid production of pro-inflammatory and by inducing distinct intracellular signalling networks within antiviral cytokines that limit pathogen replication and initiate dendritic cells (DCs). As such, DCs and TLRs represent an adaptive immune response. attractive immune modulatory targets for vaccinologists. Methods: We tested the hypothesis that induction of innate This presentation will review the emerging themes in the biol- antiviral immune responses with TLR ligands in mucosal sur- ogy DCs and TLRs, with a particular focus on relevance for faces prior to influenza A or SIV exposure would prevent vaccine development. infection or limit viral replication. Results: Intramuscular injection of CpG induced antiviral cytokines in the respiratory tract of Rhesus monkeys. Further compared to PBS-treated controls, influenza virus replication and systemic signs of disease were blunted in the respiratory tract of CpG-treated monkeys after intra-nasal and intra- tracheal inoculation. Intravaginal application of CpG or imiqimod (TLR7/8 ligand) induced antiviral cytokines in the reproductive tract of rhesus monkeys. However, in the 6- month observation period after intravaginal SIV inoculation the TLR ligand-treated animals had significantly higher plas- ma vRNA levels than the PBS-treated control monkeys. Conclusions: The immune activation triggered by TLR lig- ands was able to suppress viral replication in the case of acute influenza virus infection but apparently contributed to viral replication in the case of SIV infection. These findings suggest that immune activating such as TLR ligands should be applied with caution to prophylactic HIV vaccination in pop- ulations at high-risk for infection.
S04-05
Modulating vaccine responses against HIV with dendritic cells and Toll-like receptors
B Pulendran
Emory Vaccine Center and Department of Pathology, Atlanta, GA, USA
The immune system is ignorant or even unresponsive to most foreign proteins that are injected in a soluble, deaggregated
20 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 21
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 5: NEUTRALIZING ANTIBODIES
S05-01 S05-03
Update on strategies for inducing broadly reac- Towards a working paradigm of structure-assist- tive neutralizing antibodies ed vaccine design (SAVD)
1 1 1 1 B Haynes , H-X Liao , MA Moody , L Verkoczy , PD Kwong, P Acharya, L Chen, YD Kwon, S Majeed, 1 1 2 3 1 J Peacock , F Gao , B Hahn , B Korber , and M Alam G Ofek, M Pancera and T Zhou 1 Duke University Medical Center, Durham, NC; 2 University of Vaccine Research Center, NIAID, National Institutes of Health, Alabama-Birmingham, Birmingham, AL; 3 Los Alamos National Bethesda, MD, USA Laboratory, Los Alamos, NM, USA Structural analysis has created a wealth of atomic-level Background: A major conundrum for HIV-1 vaccine devel- detail, for both HIV-1 envelope and recognizing antibody. opment is the finding that multiple HIV-1 envelopes (Envs) How best to use this information in vaccine design? With express epitopes to which broadly reactive neutralizing structure-based drug design, the technology so successful in antibodies bind, yet when used as immunogens, these Envs the development of therapeutically effective protease do not induce neutralizing antibodies of sufficient breadth inhibitors, the lead (e.g. a small-molecule drug inhibitor) to be protective in vivo to multiple heterologous primary and the target (e.g. an essential viral enzyme) could be isolate HIV-1 strains. Strategies to approach this problem analysed as a discrete complex. However, with vaccine include design of new structures of Env that might induce design, the lead (e.g. gp120) cannot be analysed with the desired broadly reactive antibodies, and modulation of target (the humoral immune response), and interactions immunoregulatory control of B cells that have the capacity between virus and host generate emergent properties, to make the desired anti-Env antibody specificities. which greatly complicate design. Thus, although there are Objective: To review recent progress made in design and superficial similarities, structure-based drug design has no formulation of consensus Envs for induction of anti-HIV-1 direct equivalent in vaccine design. We have instead pur- neutralizing antibodies. sued a tripartite approach: 1. characterization of the HIV- Methods: We have developed strategies using centralized 1 envelope glycoprotein to reveal mechanisms of humoral (consensus) Env genes as immunogens, combined with immune evasion, 2. analysis of neutralizing antibodies to immunization regimens that utilize Toll-like receptor uncover chinks in HIV-1’s protective armor, and 3. com- (TLR) agonist stimulation of select B cell subsets. bining envelope- and antibody-based structural informa- Results: We have developed a year 2001 group M tion to enable the creation of epitope mimics, consensus envelope, CON-S gp140 Env that induces neu- unencumbered by known mechanisms of humoral evasion. tralizing antibodies to ~30% of subtype B and C HIV-1 Whether this confluence of structural information is suffi- Env pseudoviruses. Data will be presented that will outline cient to elicit broadly neutralizing antibodies will depend new progress in inducing further increased breadth of anti- in part on the ability to iteratively optimize immunogenic HIV-1 neutralizing antibodies using CON-S Env formulat- and structural parameters of conformational mimicry, epi- ed in various TLR agonists. tope accessibility, elicited potency, neutralization breadth and target specificity.
S05-02 S05-04 The anti-HIV activity of antibodies in vitro and in vivo Computational design of non-HIV protein scaf- folds presenting conserved HIV-1 neutralization DR Burton epitopes The Scripps Research Institute, Department of Immunology, La WR Schief and DA Baker Jolla, CA, USA Department of Biochemistry, University of Washington, Antibodies have been shown to have activities against Seattle, WA, USA HIV/SIV in monkeys and in humans. These activities will be reviewed and attempts to understand their mechanisms New approaches to HIV vaccine design have been made through in vitro and antibody engineering studies will be possible by recent advances in computational protein described.
AIDS Vaccine 2006 21 Plenaries and Symposia 14/8/06 16:53 Page 22
Amsterdam, Netherlands 29 August–1 September 2006
design and structural analysis of multiple HIV-epitope/anti- opsonized virions. Individual Abs may utilize more than body complexes. Toward the development of immunogens one mechanism, and Abs specific for the same region of the that elicit broadly reactive antibodies against HIV-1, we virus envelope may work by different mechanisms depend- have developed computational methods to design non-HIV ing on the particular part of the epitope that the Ab targets. protein scaffolds presenting conserved HIV-1 epitopes. By Conclusions: A variety of mechanisms exist by which Abs displaying HIV-1 epitopes in non-HIV scaffolds, we hope block the infectivity of HIV-1. Understanding the epitopes to bypass mechanisms of humoral evasion such as the vari- targeted by these Abs and the pathways by which they able loops or reduced accessibility of epitopes, and rather block infection will aid in the development of an effective induce the immune system to make broadly-neutralizing HIV vaccine. antibodies. These “epitope-scaffolds” are designed with atomic detail to stabilize the antibody-bound conformation of the epitope on the surface of the scaffold, optimize epi- tope accessibility to the targeted antibody, and bury the S05-06 non-neutralizing face of the epitope within the scaffold. Scaffold sizes are minimized to focus the immune response The role of antibody-dependent cellular cytotox- on the HIV-1 epitope. To identify epitope-scaffolds with icity in vaccine-elicited protection promising immunological responses, multiple scaffolds of different immunogenic backgrounds are used to present M Robert-Guroff 1, RH Florese1, VR Gómez-Román1, each epitope. We are addressing several HIV-1 epitopes KKA Van Rompay2, D Venzon3, DN Forthal4, G with this approach in collaboration with a large and Landucci 4, M Mahalanabis5, N Haigwood5,6, VERSUS diverse group of established HIV investigators. Kalyanaraman7, SW Barnett8 and ML Marthas2
1 Vaccine Branch and 3 Biostatistics and Data Management S05-05 Section, NCI, Bethesda, MD, USA; 2 CNPRC, University of California, Davis, CA, USA; 4 University of California, Irvine School of Medicine, Orange, CA, USA; 5University of Mechanisms by which antibodies block HIV Washington, Seattle, WA, USA; 6 Seattle Biomedical Research infection Institute, Seattle, WA, USA; 7 Advanced Bioscience S Zolla-Pazner Laboratories, Inc., Kensington, MD, USA; 8 Novartis Vaccines, Emeryville, CA, USA New York Veterans Affairs Medical Center and New York University School of Medicine, New York, NY USA Background: Antibody-dependent cellular cytotoxicity (ADCC) is a non-MHC-restricted immune mechanism that Background: The availability of >100 human monoclonal can provide an early and rapid immune response to infec- antibodies (mAbs) specific for gp120 and gp41, the enve- tious agents if antibodies have already been elicited by vac- lope glycoproteins of HIV-1, have made it possible to cination. Since antibodies that mediate ADCC are not investigate the various mechanisms by which Abs block the restricted to neutralizing epitopes, they may provide broad infectivity of the virus. functional humoral immunity. In fact, we have shown that Objective: To summarize our understanding of how Abs to vaccine-induced anti-HIV envelope antibodies mediate different epitopes on the HIV envelope block HIV-1 ADCC broadly across HIV clades A, B, C, and E. infection. Previously, a vaccine regimen combining priming with Methods: Human mAbs to various epitopes of gp41 and replication-competent Ad-SIV recombinants and boosting gp120 have been used to assess their ability to block HIV- with envelope protein elicited potent protection against intrarectal SIV challenge. Reduced viremia during 1 infection in a variety of in vitro assays. mac251 Results: Antibodies block HIV-1 infection by several path- chronic infection was correlated with SIV-specific cellular ways. These include the ability of Abs: 1) to block the immunity, while reduced acute viremia was correlated with interaction between gp120 and either CD4 or the anti-envelope binding antibodies that did not neutralize the chemokine receptors; 2) to block the conformational challenge virus. These antibodies mediated potent ADCC changes in gp120 which are required for the infectious activity, also significantly correlated with reduced process to proceed; 3) to block the conformational changes acute viremia. in gp41 which are required to expose the N-terminal fusion Objectives: To investigate the role of vaccine-elicited domain that subsequently inserts into the membrane of tar- ADCC-mediating antibodies in protection against SIV get cells, initiating virus/cell fusion; and 4) to enhance cel- infection. lular uptake of virus via Fc receptors. Antibodies that Methods: A passive transfer experiment was conducted in block infection by the first three pathways are generally neonatal Rhesus macaques, using immune IgG purified referred to as “neutralizing Abs”, but those that work via from the vaccinated macaques prior to challenge and con- the fourth pathway, despite being referred to as “non-neu- trol IgG purified from mock-immunized animals. tralizing Abs”, can effectively block virus infectivity, most Following subcutaneous infusion of IgG, the neonates were challenged orally 2 and 4 days later with SIV and probably by enhancing endocytosis and degradation of mac251
22 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 23
Amsterdam, Netherlands 29 August–1 September 2006
subsequently monitored for infection. Results: The purified immune IgG did not neutralize SIV but possessed high ADCC antibody titers and mac251 low-level antibody-dependent cell-mediated viral inhibition activity at the time of oral challenge. Following passive transfer, no protection, assessed by viral burdens, CD4 counts, and time to euthanasia, was observed. Conclusions: Possible factors contributing to the discrep- ancy between the previous correlation and lack of protec- tion here include: the high oral challenge dose compared to the 400-fold lower intrarectal dose; the challenge route with regard to viral dissemination and distribution of infused IgG; insufficient NK effector activity and/or poor functionality in neonates; insufficient immune IgG; and the possibility that the previous correlation of ADCC with pro- tection was augmented by cellular immune responses also present at challenge. Future studies should explore addi- tional challenge routes in older, juvenile macaques using higher amounts of potent IgG preparations.
AIDS Vaccine 2006 23 Plenaries and Symposia 14/8/06 16:53 Page 24
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 6: ANIMAL MODELS AND CORRELATES IMMUNITY
S06-01 S06-02
Association between strong vaccine-induced SIV- Cellular immune responses that successfully specific T-cell responses and reduced viral load after control AIDS virus replication challenge virus exposure in the SIV/macaque model DI Watkins C Stahl-Hennig1, YS Suh1,2, R Schulte1, KS Kim2, 2 1 1 1 KS Park , U Sauermann , M Franz , S Sopper , University of Wisconsin-Madison, Madison, Wisconsin, USA G Hunsmann1 and YS Sung2 Background: Our laboratory is interested in defining cellu- 1 Deutsches Primatenzentrum, Göttingen, Germany; 2 POSTECH lar immune responses that control replication of highly Biotech Center, Pohang University, Pohang, South Korea pathogenic AIDS viruses. Recently we have defined the cel- lular immune responses in two rare cohorts that exhibit Objectives: To investigate the efficacy of a bimodality control over viral replication: elite controllers and success- prime/boost vaccine regime in the SIV/macaque model. To ful vaccinees. analyse effects of mucosal immunization, one study arm com- Objectives: To define immune correlates of successful con- bined systemic and mucosal immunization. trol of replication of highly pathogenic AIDS viruses. Methods: Three groups of Rhesus monkeys with 6 animals each Methods: Elite controllers effectively control HIV/SIV were used. Group 1 was immunized i.m. three times of over 16 replication and may facilitate definition of the attributes of weeks with plasmid DNA expressing structural and regulatory a successful immune response. We have identified six SIV genes plus human IL15 (hIL15) DNA followed by two macaques that have spontaneously controlled replication intramuscular (i.m.) immunizations in escalating doses with of the pathogenic AIDS virus SIVmac239 for 1–5 years. To recombinant adenovirus expressing the same SIV genes (Ad- determine which lymphocyte populations were responsible SIV) and hIL15. Group 2 was identically immunized as group 1 for this control, we transiently depleted their CD8+ T-cells except that Ad-SIV was first given by oropharyngeal spray fol- in vivo. Additionally, we have recently investigated lowed by i.m. injection. All vaccinees plus controls (group 3) whether vaccine-induced cellular immunity, in the absence were challenged orally with pathogenic SIV 12 weeks after final of any Env-specific antibodies, can control viral replication immunization. SIV-specific immune responses and T-cell sub- following multiple low-dose challenges with the highly populations before and after challenge were regularly deter- pathogenic SIVmac239 isolate. Eight Mamu-A*01+ Indian mined. Viral loads in blood were measured after challenge. Rhesus macaques were vaccinated with SIV Gag, Tat, Rev, Results: Both immunization strategies induced strong SIV-spe- and Nef using a DNA prime adenovirus boost strategy. cific T-cell responses as determined by ELISpot and prolifera- More than one year ago, these vaccinees along with eight tion assay, but with different kinetics. In group 1 these Mamu-A*01+ control macaques were challenged with responses were detectable after first Ad immunization, but were repeated low dose SIVmac239. delayed in group 2 until after the second Ad application. At the Results: Virus recrudescence was seen in the CD8+ cell- time of challenge immune responses of group 2 were signifi- depleted elite controllers and this only diminished coinci- cantly higher than those of group 1. Ad-specific neutralizing dent with the return of CD8+ cells targeting Nef and Vif. titres appeared in group 1 after the first Ad-SIV immunization This control of viral replication was also accompanied by and in group 2 only after the second one. Following challenge a massive CD4 response, largely targeting Gag. In the vac- viral RNA loads in plasma were tenfold reduced at peak cination trial, control of viral replication was seen in the viremia in both vaccine groups compared to the controls. This vaccinees out to 1 year post-challenge. Furthermore, loss of difference was observed up to 12 weeks post challenge. CD4+ memory cells was ameliorated in the vaccinees. Furthermore, both vaccine groups better conserved the central Conclusions: CD8+ T cell responses targeting Gag, Nef and memory T-cell population. Vif accompanied by strong Gag-specific CD4+ T cell Conclusions: The marked increase in cellular immune respons- responses can control replication of the highly pathogenic es after Ad immunization suggests a priming effect by DNA SIVmac239 isolate. Furthermore, control of viral replica- immunization. The combination of mucosal and systemic Ad tion mediated by cellular immune responses can occur in immunization is superior to Ad given i.m. alone in terms of level the absence of Env-specific neutralizing antibodies. of immune responses probably due to a weaker vector-specific immunity seen in group 2. Despite these differences, both vac- cine groups show a similar level of viral load reduction. The overall strong SIV-specific immune responses observed might be due to inclusion of IL15 in the vaccine.
24 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 25
Amsterdam, Netherlands 29 August–1 September 2006
S06-04 S06-05
Experimental infection of humanized mice with HIV Oral delivery of replicating SIV vaccines in new- born Rhesus macaques: safety and immunogenicity J Victor Garcia of recombinant VSV-SIV-based and live-attenuated SIV vaccines University of Texas Southwestern Medical Center, Department of Internal Medicine, Dallas, TX, USA ML Marthas 1, K Abel 1, KK Van Rompay 1, P Earl 2, K Schmidt 1, R Colón 1, E Blackwood 1, J Eastlick 1, J Moore 1, Background: Because of the limited species tropism of HIV, L Buonocore-Buzzelli 3, B Moss 2, N Rose 3, J Rose 3 and there are few models where potential clinical interventions MB McChesney 1 and vaccines can be evaluated. Currently, the macaque model of SIV infection is considered the best available. However, 1 California National Primate Research Center, Davis, CA, USA; macaques are not susceptible to HIV infection. 2 National Institute of Allergy & Infectious Diseases, Bethesda, Objectives: The long-term goal of our laboratory has been to MD, USA; 3 Yale University, New Haven, CT, USA develop and implement novel alternative approaches to inves- tigate HIV transmission, pathogenesis, vaccines and innova- Background: An ideal neonatal HIV vaccine would be safe, tive clinical interventions. efficacious and orally administered; however, few current Methods: Human CD34+ stem cells were used to reconstitute HIV vaccines are designed for oral delivery. the hematopoietic system of immunodeficient mice in the Objectives: To evaluate safety and immunogenicity of immu- context of human thymic epithelium. nizing newborn Rhesus macaques: (1) orally with attenuated, Results: Transplantation of autologous human hematopoi- replication-competent, recombinant vesicular stomatitis etic CD34+ cells into NOD/SCID mice previously implant- viruses expressing SIV Gag, Pol, Env antigens (VSV-SIV), (2) ed with thymic and liver tissues results in long-term, orally with VSV-SIV plus intramuscularly (IM) with recom- systemic human T cell homeostasis. In addition, these mice binant modified vaccinia Ankara expressing SIV Gag, Pol have systemic repopulation with human B cells, mono- and Env antigens (MVA-SIV), (3) orally and intravenously cytes/macrophages and dendritic cells (DC). T cells in these (IV) with nonpathogenic SIVmac1A11. mice generate human MHC Class I and Class II restricted Methods: Groups of four or eight infant Rhesus macaques immune responses to viral infection and are activated by per group were immunized. Group A and B animals human DCs to mount potent T cell immune response to received VSV-SIV (107 pfu) orally at birth; Group B ani- super antigens. Administration of the super antigen toxic mals also were given MVA-SIV (108 pfu) IM at 2 weeks of shock syndrome toxin-1 (TSST-1) resulted in the specific age. Group C animals received MVA-SIV (108 pfu) at birth systemic expansion of human V 2+ T cells, release of β and 2 weeks of age. Group D (vector controls) received human pro-inflammatory cytokines and localized specific VSV-measles (107 pfu) at birth and MVA-influenza (108 activation and maturation of human CD11c+ dendritic pfu) IM at 2 weeks of age. Group E animals received cells. BLT mice are susceptible to infection by HIV-1 and SIVmac1A11 (105 TCID50%) orally and IV at 0 and 2 infection results in plasma antigenimia, production of HIV- weeks of age. To assess vaccine safety, visual inspection for specific human antibodies and progressive depletion of oral or dermal lesions, formula intake and body weights human CD4+ cells from the peripheral blood with a pro- were monitored daily. Vector virus-specific, hepatitis B and portional increase in the percentages of human CD8+ T SIV-specific plasma antibodies and SIV-specific T cells cells. Analysis of different internal tissues confirms sys- responses in mononuclear cells in lymphoid tissues and temic depletion of CD4+ T cells in all organs tested includ- blood were surrogate measures of vaccine immunogenicity. ing gut and lung. Analysis by in situ hybridization Results: No adverse events were observed in immunized confirmed the presence of virus in all tissues examined. infants. SIV-specific plasma antibodies were detected in Virus was readily rescued by co-culture with activated Group B, C and E animals, but not in Group A animals; this human PBMC. finding is consistent with results for VSV-SIV immunized Conclusions: BLT mice are an excellent model to investigate older macaques, where SIV antibody responses were not multiple aspects of HIV infection in vivo. detected until later or after MVA-SIV boosting. SIV-specific T cell responses were detected only in animals immunized with VSV-SIV and MVA-SIV. Importantly, SIV-specific T cell responses were localized in oral tissues. Conclusions: An oral-VSV-prime plus IM-MVA-boost SIV vaccine strategy is safe in neonates and by 4 weeks of age elic- its broader and more consistent SIV-specific humoral and cel- lular responses in oral and systemic lymphoid tissues compared to a prime/boost with any single vaccine. Funding: PHS grants Ad 62518 (MLM), Ad 40357 (JR); RR 00169 (California National Primate Research Centre).
AIDS Vaccine 2006 25 Plenaries and Symposia 14/8/06 16:53 Page 26
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 7: T CELL DEVELOPMENT WITH FOCUS ON T CELL RESPONSES
S07-01 S07-02
Feasibility of T cell vaccines Update on the Merck Ad5 Phase II Trial (the STEP Study—Merck V520 Protocol 023/HVTN 502) A McMichael M Robertson1, D Mehrotra1, S Buchbinder2, D Fitzgerald3, MRC Human Immunology Unit and CHAVI, Oxford University, A Duerr4, and D Lawrence5 Oxford, UK 1 Merck Research Laboratories, West Point, PA, USA, 2 University Experiments in primates show us that vaccines that stimu- of California San Francisco, CA, USA, 3 GHESKIO, Port-au-Prince, late CD8 T cell immunity offer a significant degree of pro- Haiti, 4 HIV Vaccine Trials Network, Seattle, WA, USA, 5 NIAID, tection against disease caused by subsequent challenge NIH, Bethesda, MD, USA infection with SIV or SHIV hybrid viruses. However, unless mucosal T cell responses can be activated and expanded Background: Thus far, efforts to develop HIV vaccines capa- from memory T cells within 5–7 days of infection, there is ble of eliciting broadly cross-reactive neutralizing antibodies little chance of aborting HIV infection. More likely vaccine have been unsuccessful; therefore, the focus has shifted to prepared T cell immune responses will ameliorate the vaccine-induced HIV-specific cell mediated immune (CMI) effects of HIV-1 infection in humans. Ideally such virus responses. While CMI responses cannot prevent infection by control would be equivalent to that seen for EBV, CMV or HIV, they may abort infection before it becomes established, HIV-2 and would greatly reduce transmission. Although or contain the virus. Such vaccines have shown promise in such control is clearly possible, HIV-1 variability and some animal models. Such an effect could have an enormous capacity to escape immune responses has to be overcome impact on progression to disease and the likelihood of sec- and will be the major obstacle ahead. Finally, if good ondary transmission. In Phase I studies, the MRKAd5 HIV-1 immune control is what T cell vaccines can achieve, more Gag/Pol/Nef vaccine elicited CMI responses in a high pro- effort should be directed towards improving immune con- portion of volunteers at a dose with an acceptable safety pro- trol in people already infected, for example by therapeutic file. This safety and immunogenicity profile merited moving vaccination. This would be a much more efficient way of it to a test-of-concept trial. controlling the pandemic than using T cell vaccines to par- Objectives: To test the concept that a vaccine capable of elic- tially protect huge numbers of uninfected people who are iting HIV-specific CMI responses could prevent persistent at varying level of risk. infection and/or delay disease progression. Methods: Collaborative study between Merck, the HIV Vaccine Trials Network, and National Institute of Allergy and Infectious Diseases of the US NIH. Phase II prospective, randomized, multi-center, double-blind, placebo-controlled study of 3000 HIV seronegative volunteers at high risk of acquiring HIV in regions of the world where clade B pre- dominates (N. American, S. America, Australia and the Caribbean). Participants are randomized (1:1) to receive 3 injections of either MRKAd5 HIV-1 gag/pol/nef or placebo. Randomization is pre-stratified by gender, baseline Ad5 titer, and study site. Participants will be tested ~every 6 months for acquisition of HIV. If a study participant becomes HIV infect- ed, plasma HIV viral load and CD4 cell counts will be mea- sured frequently. The final analysis will occur when ~100 new infections have occurred. Safety and efficacy are being monitored by an independent DSMB. Results: Enrollment began in Dec 2004 and is ongoing. Over 1700 volunteers were randomized by June 2006; of these ~30% are women and ~65% are non-Caucasian. Thus far, the study vaccine has been generally safe and well tolerated. Conclusions: Safety and reactogenicity is similar to that seen in Phase I trials. Data gathered on seroincidence and clinical course among seroconverters will allow assessment of whether vaccine-induced HIV-specific CMI responses can prevent persistent infection and/or delay disease progression.
26 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 27
Amsterdam, Netherlands 29 August–1 September 2006
S07-03 S07-05
Immunologic characterization of a multiclade HIV-1 Construction, development, and evaluation of DNA followed by recombinant adenovirus prime/ matched clade-C/B+ DNA- and MVA-Based HIV-1 boost vaccine in normal healthy volunteers vaccine candidates
RA Koup, BS Graham, M Roederer, R Bailer, PL Gomez, DD Ho M Nason, JE Martin, ME Enama, J Mascola and GJ Nabel Aaron Diamond AIDS Research Center, The Rockefeller The VRC Clinical Trials Core, The VRC Vector Core, The VRC University, New York, NY, USA Vaccine Production Program and The VRC Immunology Core Lab. Vaccine Research Center, NIAID, NIH, DHHS, Bethesda, MD On behalf of colleagues at the Aaron Diamond AIDS Research Center, International AIDS Vaccine Initiative, and Background: Induction of HIV-1-specific T cell and anti- University of Rochester, I will present data on the clinical body responses to diverse subtypes is a major goal of cur- development of two HIV-1 vaccine candidates. The first is a rent vaccine efforts. We have tested heterologous DNA DNA-based vaccine, termed ADVAX, consisting of 1:1 mix- prime-recombinant adenovirus vector boost approaches, ture of two dual-promoter plasmids that collectively contain priming subjects with 3 doses of either 4 or 6 DNA plas- five viral genes (gag, pol, env, tat, and nef) from a circulating mids expressing clade B gag, pol, and nef, and clades A, B, strain (CRF007, clade C/B′) that is dominant in the Yunnan and C env and boosting with 4 recombinant adenovirus province of China. The second is a recombinant modified serotype 5 vectors (rAd5) expressing matching genes except vaccinia Ankara (MVA), termed ADMVA, carrying essential- for the absence of nef. ly the same five HIV-1 genes. Over the course of the past five Objectives: To characterize the immunogenicity of a multi- years, these vaccine candidates were successfully constructed, clade DNA prime, recombinant adenovirus vaccine boost in thoroughly evaluated both in vitro and in animals, and then normal healthy volunteers. carefully advanced into clinical development and evaluation. Methods: Fourteen subjects received three doses of 4 or 8 mg The Phase I, dose-escalation trial of ADVAX in 45 healthy, of DNA, and were boosted with 1010 PFU of rAd5 at 31 to low-risk volunteers has been completed, and a similar Phase 104 weeks. CD4 and CD8 T cell responses were measured by I, dose-escalation trial of ADMVA in 48 subjects is well IFN-γ ELISpot and intracellular cytokine staining for IL-2 or underway. Safety and immunogenicity data in humans will be IFN-γ. Epitope mapping, TCR functional avidity, T cell mem- presented at the conference, and the planning of a prime- ory phenotyping, and multifunctional assessments of the vac- boost study using these two vaccine candidates is in progress. cine-induced T cell responses were also examined. Results: There were good ELISpot responses to individual peptide pools post DNA prime, and the majority of these were increased after rAd5 boost (range 105–3682 SFU/106 PBMC). In addition, there were robust antibody responses after rAd5 boosting. Multiple clade-specific and clade cross- reactive epitopes were targeted, often at very high avidity. The responses were predominantly central memory T cells of which a substantial proportion were polyfunctional. Conclusions: With a variable interval between DNA prime and rAd5 boost, significant T cell and antibody responses are induced that exceed the highest values seen to date in subjects immunized with DNA or rAd5 alone. This heterologous prime/boost strategy induces a complex functional T cell response that targets multiple HIV-1 clades.
AIDS Vaccine 2006 27 Plenaries and Symposia 14/8/06 16:53 Page 28
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 8: EPIDEMIOLOGY AND CLINICAL TRIAL ISSUES
S08-02 S08-03
Techniques for estimating HIV-1 incidence HIV incidence rates: impact on vaccine trial design
B Branson SG Self
Centers for Disease Control and Prevention, Atlanta, GA, USA The Statistical Center for HIV/AIDS Research & Prevention (SCHARP) Fred Hutchinson Cancer Research Center, Seattle, Conventionally, HIV incidence has been measured in longitu- WA, USA dinal cohort studies; back-calculated from AIDS case report- ing data coupled with estimates of the incubation time from Background: The amount of information accrued in a trial initial infection to AIDS; and modelled using prevalence and about vaccine efficacy is proportional to the observed num- survival data. The capacity to estimate HIV incidence from a bers of incident HIV infection endpoints. Failure to accrue single cross-sectional survey has advantages for surveillance, the number of endpoints specified by the study design will vaccine trials, and intervention studies. A number of assays compromise the ability to meet the trials primary objectives. have now been developed for detecting recent HIV-1 infec- HIV seroincidence is a key trial design parameter that links tion and for estimating incidence using specimens from cross- the study operational plan to its scientific objectives. sectional studies. These assays define the duration of a Objective: To evaluate sources of discrepancies between pre- transient state related to early infection, either viremia before dicted incidence used in trial design and that actually seroconversion (RNA or p24 antigen tests) or characteristics observed and describe strategies for minimizing these differ- of the initial antibody response (antibody titer, proportion, ences and their impact on trial integrity. specificity, isotype, or avidity). The duration (“window peri- Methods: A brief survey of approaches to estimating end- od”) of this early state is determined, and the frequency of point incidences and trial monitoring will be described. this transient state in the at-risk population divided by its Differences between predicted and observed endpoint rates window period gives an estimate of incidence (new infections will be qualitativels compared to the empirical bases for the per person per unit of time). Serologic techniques that use predicted rates. modified enzyme immunoassays (EIAs), include the sensitive- Results: Predicted HIV incidence rates based on even the less sensitive EIA testing algorithm for recent HIV serocon- highest quality of pre-trial epidemiological data can overesti- version (STARHS), the BED capture EIA, and the avidity mate observed rates in vaccine trials by two-fold or more. index. Window periods were calibrated using specimen pan- Decisions to modify trial operational plans for remediation of els from recent seroconverters. Preliminary validations com- such discrepancies will be made with a retrials weak imperi- paring the accuracy of incidence estimates produced by these als basis but may have a large impact on trial costs. assays with an independently observed measure suggest that Conclusions: Vaccine trial designs should adopt a very conserv- the assay methods tend to overestimate incidence. Key factors ative approach to the prediction of HIV infection endpoint that affect the estimates include differences in window peri- rates. Explicit plans would be made at the design stage for mon- ods with different HIV-1 subtypes, rapidly changing inci- itoring endpoint rates and options for remediation. Such plans dence in the tested population, misclassification of should be tied to resources available on contingency. longstanding or late infections, and effects of antiretroviral therapy. The ability to accurately identify recent infections in individuals requires optimization different from that needed to produce accurate population-based incidence estimates. S08-04 This presentation will describe the principles upon which cur- rent assays are based, summarize preliminary results from validations with different HIV subtypes, and examine alter- Issues regarding pregnant women/breast feeding natives for improving assay-based incidence estimates. These in vaccine trials include using confirmatory algorithms of sequential testing with assays that rely on different principles, modifying assay S Allen and the Rwanda Zambia HIV Research Group cutoffs, and calculating adjustments for misclassification based on imputed sensitivity and specificity. Rollins School of Public Health, Emory University, Atlanta, GA, USA
Pregnancy is an exclusion criterion for participation in many clinical trials of biomedical interventions. In the case of HIV vaccine candidates, efficacy trials are likely to be conducted in areas of high HIV prevalence and incidence. Several research sites in Africa are conducting early Phase trials of safety and immunogenicity. Larger scale ‘proof of concept’ or ‘Phase IIb’
28 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 29
Amsterdam, Netherlands 29 August–1 September 2006
studies are planned in the coming 1–2 years. Fertility rates in Africa are comparatively high, with the average woman bearing 5–8 children in her lifetime. Between the ages of 20 and 45, most women progress sequentially through preg- nancy, breast-feeding, and short-term contraception to facilitate spacing. The return of menses in lactating women is irregular; many breastfeeding women do not realize they are pregnant again until 8–12 gestation. The table below summarizes data among HIV discordant couples enrolling in prospective cohort studies in preparation for vaccine trials (Table 1). Pregnancy incidence ranges between 20–30% in these cohorts. Small-scale vaccine trials in Africa have had difficulties enrolling women because pregnancy and breastfeeding are so common. Education about and provision of modern contracep- tives in addition to condoms (‘dual method use’) does reduce pregnancy incidence, but not to the low levels desired for a 2 year follow-up in an HIV vaccine trial. If women are to be included in larger trials, options include: 1). enrolling women in the post-partum period; 2). not excluding breast-feeding women; and/or 3). encouraging -or requiring-effective contra- ceptive use (implant or IUD) for the duration of the study. The latter option is likely to be welcomed by many post-partum women who desire adequate spacing independent of the trial. An HIV vaccine destined for use in developing countries must be safe for pregnant and breastfeeding women. Regulatory traditions exclude pregnant and breastfeeding women from trials but in the broader public health context this tradition should be revisited.
AIDS Vaccine 2006 29 Plenaries and Symposia 14/8/06 16:53 Page 30
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 9: LARGE SCALE PRODUCTION AND DELIVERY ISSUES
S09-01 S09-05
Producing prophylactic vaccines in continuous The impact of an AIDS vaccine in Developing cell lines Countries: the use of modelling build support for vaccine development RL Sheets J Stover1, L Bollinger1, R Hecht2 and E Roca2 Captain, US Public Health Service, NIH/NIAID, Vaccine Research Center and Division of AIDS, Bethesda, MD, USA 1 Futures Institute, Glastonbury, CT, USA; 2 International AIDS Vaccine Initiative, New York City, NY, USA Background: Since cell culture was developed for the prop- agation of viruses to manufacture prophylactic vaccines Background: A number of modelling studies have investigat- and since previously unknown and oncogenic viral conta- ed the potential impact of AIDS vaccines among specific pop- minants were discovered in the 1950’s, concerns about ulations in particular locations. These studies show the continuous cell lines (CCL) have proscribed their use. significant impact that is possible and explore the effects of These concerns hinge on the potential for unsuspected vaccine characteristics (especially efficacy in reducing the sus- viruses that might have elicited the CCL phenotype and on ceptibility to infection of the vaccinated individual) on cellular DNA. However, HIV vaccine candidates based on impact. However, they do not provide a global estimate of novel viral vector systems and need for rapid production of vaccine impact. emergent pandemic influenza vaccines are pushing against Objectives: To develop, through modelling, estimates of the this barrier. potential impact of an AIDS vaccine, at both the global and Objectives: To rationally manage risks of cell culture for national level, to illustrate to donors and governments the production of prophylactic vaccines. benefits of investing now in vaccine development, and to Methods: In addition to routine testing that would be per- provide a model for use by national experts. formed for any vaccine cell substrate, strategies to address Methods: A simulation model has been developed that con- theoretical concerns include in vivo assessments of tumori- siders the adult population of a country disaggregated by sex genic potential of CCL and oncogenic potential of their and six risk categories (no risk; low, medium and high het- components, including lysates (to release potential unsus- erosexual populations; injecting drug users and men who pected viruses) and cellular DNA. Additionally, assess- have sex with men). Population behavior is described in ments of the PrP gene to address whether CCL would terms of age at first sex, number of partners, condom use, cir- contain more mutations that might support propagation cumcision status and drug use. The model calculates the of prions. number of new HIV infections and tracks people through pri- Results: Investigational vaccines are being produced in mary infection, asymptomatic and symptomatic stages CCL, such as Per.C6 (derived from human embryonic Vaccines coverage can be specified by year and risk group. retinoblasts by transformation with Ad5 E1 genes), 293 Vaccine characteristics include effects on susceptibility, infec- and its derivatives (derived from human embryonic kidney tiousness and disease progression as well as duration of pro- by transformation with Ad5 E1 genes), HeLa (human cer- tection and cost. The model has been applied to seven vical carcinoma caused by HPV-18), and Madin-Darby countries (Nigeria, South Africa, Brazil, Mexico, China, India Canine Kidney (MDCK). Regulators are permitting the and Russia) and the results scaled-up to estimate impact by clinical testing of vaccines produced in such cell lines. region and for the developing world. Additionally, vaccines produced in avian cell lines such as Results: We examined three scenarios of vaccine implementa- DF-1 or EBx cell lines are approaching clinical testing. It is tion with programs starting in 2015, reaching coverage levels likely that influenza vaccines made in Per.C6, MDCK, or of 20%–40% and reductions in susceptibility and infectious- Vero cells will be licensed in the coming years. ness from 30% to 70%. We assume scale up of other pre- Conclusions: Pressed by the need to develop HIV vaccines vention and treatment programs to reach international based on novel viral vectors that cannot be produced in targets by 2015. Between 2015 and 2030 the vaccination more conventional vaccine cell substrates and for rapid programs would avert from 6–28 million new infections production of emergent pandemic influenza vaccines; (11–56%) and avert from 5–21% of AIDS deaths occurring through rational and systematic consideration of the theo- during that period. retical risks posed, strategies have been developed and Conclusions: The potential benefits of an AIDS vaccine are implemented to permit the safe use of continuous cell lines clearly large, even if international goals for prevention and to produce investigational vaccine candidates. treatment scale up are reached. Investing now to develop effective vaccines will produce large benefits in the future. Country experts can apply the model to their settings to explore benefits to their countries.
30 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 31
Amsterdam, Netherlands 29 August–1 September 2006
SESSION 10: THERAPEUTIC VACCINE TRIALS
S10-01 and including week 6 will be presented during the conference. Conclusions: Preliminary results indicate the recombinant HIV-1 clade B gag-pol-nef and env NYVAC vaccine is safe TheraVac-01: an open-label phase I study to and well tolerated. evaluate the safety of the HIV-1 vaccine NYVAC-B in chronic HIV-1 infected patients successfully treated with HAART S10-02 P-A Bart1, J Vermeulen2, A Harari1, G Tapia1, F Wit2, B Autran3, J Lange2 and G Pantaleo1 A Phase II randomized, partially-blinded trial of antiretroviral therapy (ART), HIV-specific 1 Centre Hospitalier Universitaire Vaudois, Lausanne, immunizations, and IL-2 cycles to promote Switzerland; 2 Academic Medical Center and IATEC, Amsterdam, efficient control of viral replication (ACTG A5024) The Netherlands; 3 Hôpital Pitié-Salpêtrière, Paris, France JM Kilby1, RP Bucy1, D Mildvan2, M Fischl3, J Santana- Background: The introduction of HAART has led to a dra- Bagur 4, J Lennox5, C Pilcher6, A Zolopa7, J Lawrence8, matic decline in HIV-related morbidity and mortality. RB Pollard9, R El Habib10, D Sahner11 ,L Fox12, E Aga 13, However, HAART is associated with metabolic complica- RJ Bosch13 and R Mitsuyasu14 for the ACTG A5024 tions and increased cardiovascular risks, therapy failure and Protocol Team accumulation of drug resistance mutations over time. For these reasons, the EU financed program “Evaluation of a new 1 UAB, Birmingham, AL, USA; 2 Beth Israel Med Ctr, New York, NY, therapeutic vaccination strategy for HIV infection” USA; 3 Univ Miami, Miami, FL, USA; 4 Univ Puerto Rico, San Juan, (TheraVac) was established to develop therapeutic HIV-1 vaccines that would limit exposure to HAART. The aim of Puerto Rico; 5 Emory Univ, Atlanta, GA, USA; 6 UNC-CH, Chapel TheraVac is to evaluate whether the administration of the Hill, NC, USA; 7 Stanford Univ, Palo Alto, CA, USA; new HIV-1 (clade B) recombinant poxvirus vaccines 8 UCSF, San Francisco CA, USA; 9 UC-Davis, Sacramento, CA, USA; NYVAC-B or MVA-B, developed by the EuroVacc 10 Sanofi Pasteur, Paris, France; AIDS, 11 Chiron Corporation, Consortium, can restore a strong protective immune control Emeryville, CA, USA; 12 NIAID, NIH, Bethesda, MD, USA; 13 of HIV-1. TheraVac-01, sponsored by the EuroVacc Harvard, Boston, MA, USA; 14 UCLA, Los Angeles, CA, USA Foundation, is the first of two Phase I clinical trials evaluat- ing NYVAC-B and MVA-B respectively. Background: Strategies are needed to delay or limit life-long Objectives: The primary objective is to evaluate the safety of dependence on antiretroviral therapy (ART). the recombinant poxvirus NYVAC-B. The secondary objec- Methods: Eighty-one subjects well-controlled on ART were tive is to assess HIV-1 specific immune responses following randomized to 4 arms involving an HIV-specific immunogen the vaccination with NYVAC-B. (ALVAC vCP1452 or placebo; wks 0, 8, 16, 24, 48) or open- Methods: NYVAC-B, containing HIV-1 clade B gag-pol-nef label interleukin-2 cycles (4.5 MIU sc bid ×5 days for 8 polygene and env gene, is administered during concomitant weeks) as adjuncts to ART for a year: A) ALVAC placebo; B) successful HAART to 10 HIV-1 infected patients with CD4 ALVAC; C) IL-2 + ALVAC placebo; D) IL-2 + ALVAC. An 3 counts ≥200 cells/mm prior to HAART and plasma HIV-1 analytical therapeutic interruption (ATI) endpoint (viral RNA load below 50 copies/ml for 3 months prior to inclu- rebound 12 weeks after ART interruption) was utilized to 7.4 sion. 1ml of NYVAC-B (10 pfu/ml) is injected by IM route compare effectiveness of these interventions. on day 0 and 28 in respectively the left and right deltoid mus- Results: Among 52 subjects reaching an ATI endpoint, there cle. Patients are observed at the study site for 2 h after vacci- was a 0.5 log lower viral load rebound following ALVAC (B 10 nation. Follow-up visits are scheduled at days 1, 7, 14, 21 versus A; P=0.033). Modest secondary endpoint advantages and 28 (= second vaccination), and at weeks 6, 8, 12, 24, 36 were also seen for ALVAC (B and D) versus non-ALVAC (A and 48. Specific vaccination-related local and systemic reac- and C) arms. IL-2 recipients (C and D) had larger CD4 count tions are recorded until week 12. Standard safety parameters increases (P<0.001) but virological rebound comparable to (vital signs, clinical symptoms and routine lab), plasma HIV- placebo, and more commonly experienced adverse events. + + 1 RNA and CD4 /CD8 T cell counts are assessed at every Significant mean changes from baseline were not detected for visit. HIV-1 specific immune responses are evaluated at base- HIV-specific lymphoproliferative responses in any arm. line, day 7, 14, 28, and week 6, 8, 12, 24, 36 and 48. Conclusions: While this exploratory trial did not demon- Results: The study is ongoing. Currently nine patients were strate definitive advantages of ALVAC therapeutic immuniza- enrolled and received the first injection, and six of them also tion, it provides proof of concept that an HIV-specific received the second one. Only grade 1 and 2 local injection immunogen may impact host immune control as measured by site adverse events (AE) and few systemic grade 1 and 2 AE viral rebound after ART withdrawal. were observed. The safety and immunogenicity results up to
AIDS Vaccine 2006 31 Plenaries and Symposia 14/8/06 16:53 Page 32
Amsterdam, Netherlands 29 August–1 September 2006
S10-04 S10-05
Inactivated retroviral virions with functional DermaVir patch therapeutic vaccine in chronic envelope glycoproteins: evaluation as a HIV infection: from the concept to the clinic therapeutic vaccine immunoge J Lisziewicz1 and F Lori2 JD Lifson 1 Genetic Immunity, Budapest, Hungary & Washington DC, USA; AIDS Vaccine Program, SAIC Frederick, Inc., National 2 RIGHT, Pavia, Italy & Washington DC USA Cancer Institute, Frederick, Frederick, Maryland, USA Background: The concept of DermaVir vaccine therapy is Background: Treatment of retroviral virions with 2,2′- originated from a case demonstrating that the wild-type HIV dithiodipyridine (Aldrithiol-2, AT-2) eliminates infectivity by induced long-lasting immune control during primary infec- preferential covalent modification of the free sulfhydryl tion (Lisziewicz et al, N Engl J Med 1999). To explain this groups of the cysteines of internal viral proteins (NC, CA, observation we hypothesized that expression of HIV genes in IN), sparing the disulfide-bonded cysteines on the virion sur- dendritic cells (DC) can expand the memory face, thereby rendering the virions non-infectious but with T cell pool in HIV-infected individuals. structurally and functionally intact envelope glycoproteins. Objective: To investigate the safety and the antiviral effec- AT-2 treated virions are being evaluated as a candidate tiveness of a vaccine therapy that specifically expresses plas- vaccine immunogen. mid DNA-derived HIV antigens in lymph node DC. Objectives: As regulatory considerations dictate that the ini- Methods: We have developed the DermaVir Patch therapeu- tial clinical evaluation of AT-2 inactivated virions will likely tic vaccine that involved (1) the construction of plasmid DNA be in the setting of therapeutic vaccination, we are evaluating antigen that safely expresses the majority of HIV (or SHIV) this candidate vaccine immunogen in preclinical studies in genes yet is incapable of replication or integration; (2) the non-human primates and establishing methodologies to facil- formulation of a mannosylated nanoparticle vaccine target- itate the transition to initial clinical evaluation. ing DC and protecting the DNA from endosomal degrada- Methods: Indian subspecies Rhesus macaques infected with tion; and (3) the development of the DermaPrep Patch for SIVmac239 were followed as untreated, unvaccinated con- topical vaccine administration (Lisziewicz et al, J Infect Dis trols, or were placed on combination antiretroviral treatment 2005). (ART) (tenofovir, emtricitabine) beginning 10 weeks post- Results: We demonstrated in chronically SIV-infected inoculation. After 16 weeks on treatment, half of the treated macaques that repeated topical DermaVir(SHIV) treatment animals were vaccinated with AT-2 inactivated SIV virions, resulted in immunological, virological and clinical benefits pulsed on autologous dendritic cells, while the other half (Lisziewicz et al. AIDS 2005). Preclinical toxicity studies in were not vaccinated; treatment continued for both vaccinat- monkeys, rabbits and swine demonstrated only transient, ed and non-vaccinated animals. ART was stopped 6 weeks mild local reactogenicity in the absence of systemic toxicities. after the final immunization, with continued monitoring of We obtained similar safety and tolerability results in chroni- virological and immunological parameters. cally HIV-infected individuals during a Phase I clinical trial Results: ART provided variable suppression of viral replica- with the DermaVir Patch (a single dose of 0.1, 0.4 or 0.8 mg tion (nadir plasma SIV RNA values from <10 copy Eq/ml to plasmid DNA). Repeated DermaVir dosing is currently being 104 copy Eq/ml); animals with incomplete suppression investigated in a placebo-controlled trial in antiretroviral showed the K65R mutation in RT associated with tenofovir drug-treated individuals (ACTG 5176). resistance. Vaccinated animals showed increased IFN-γ Conclusion: The pre-clinical and preliminary Phase I results ELISPOT responses to viral antigens, and in some cases, to date demonstrate the safety and feasibility of DermaVir improved control of plasma viremia after ART was stopped, Patch treatment during chronic HIV infection and suggest although this effect was transient. that DermaVir Patch may complement antiretroviral drugs to Conclusions: Therapeutic vaccination with AT-2 inactivated sustain suppression of virus replication. SIV is safe, well tolerated, induces enhanced virus specific immune responses and is associated with at least transient improved control of viral replication in the absence of ART. A new non-human primate therapeutic vaccination study, using an improved ART regimen is in progress. A novel cul- ture system for production of high levels of autologous patient derived HIV from autologous monocyte-derived macrophages may facilitate eventual clinical evaluation of AT-2 inactivated HIV as a vaccine immunogen. (NCI Contract N01-CO-12400).
32 Antiviral Therapy 11, Supplement 2 Plenaries and Symposia 14/8/06 16:53 Page 33
Amsterdam, Netherlands 29 August–1 September 2006
SPECIAL LUNCH TIME LECTURE
SLL01
EC Initiatives on HIV: capacity building, ethics, human rights promotion in Africa
M Salvi
EGE Secretariat, European Bureau of Policy Advisors (BEPA), European Commission, Brussels, Belgium
The Global dimension of science and technology which is fully incorporate in FP6 and FP7 proposal, and reflects the current regulatory framework of the EU (Dir 20/2001 for example) implies a main role of local research ethics commit- tees in the scientific activities carried out in different regions of the globe. Since the EU promotes responsible research and requires that EU financed research activities to be conform to the main ethics frame stated in the Charter of Fundamental Rights of the EU (Nice 2001) and international agreements such as UN guidelines or documents prepared by relevant International Organizations (Council of Europe, CIOMS, WMA, OECD etc.), major efforts are being taken to facilitate capacity building on ethics in developing countries (CBEDC). The EU plays then a proactive role not only to promote the values indicated in the Charter above but also to respect sub- sidiarity and local socio-cultural features. This approach can therefore be summarized as follow: 1) The Commission supports CBEDC initiatives in respect of nation- al autonomies while supporting the values stated in the European Charter of fundamental rights. 2) EC regulations (GCP Directive, for example) and FP6 ask ethics to be inte- grated in research practices. 3) Facilitate discussions at International level on embedding ethical and legal require- ments into research practices while keeping the highest stan- dards of protection of human rights. 4) Finance projects to facilitate CBEDC. In my presentation I will report on recent political decision on the role of the EU with regard the fight against HIV as well as current and future actions on ethics and CBEDC. I will also report on the main elements to be taken into account on ethics and clinical trials in developing countries indicated by the European Group on Ethics in Science and New Technologies (EGE) to the European Commission, an inde- pendent, pluralist and multidisciplinary body which advises the European Commission on ethical aspects of science and new technologies (Opinion N. 17, Ethical aspects of clinical research in developing countries).
AIDS Vaccine 2006 33 Plenaries and Symposia 14/8/06 16:53 Page 34 Oral Abstracts Sessions 14/8/06 16:40 Page 35
ORAL SESSIONS Oral Abstracts Sessions 14/8/06 16:40 Page 36 Oral Abstracts Sessions 14/8/06 16:40 Page 37
Amsterdam, Netherlands 29 August–1 September 2006
TOPIC 1: HIV TRANSMISSION/ACUTE HIV INFECTION/INNATE IMMUNITY/ MUCOSAL IMMUNITY
OA01-01 OA01-02
Adaptation of human immunodeficiency virus Gp41 HMA as a screening tool to study HIV type 1 to host HLA upon transmission to a new transmission between partners individual O Manigart1,2, D Boeras1, C Vwalika1,2, P Ahues1, M Navis, F Koning, C Koevoets, N Kootstra and B Chatora2, J Mulenga1, S Allen1 and E Hunter 1 H Schuitemaker 1 Emory University, Atlanta, Georgia, USA; 2 Zambia-Emory HIV Sanquin Research at CLB and Lansteiner Laboratory of the Research Project, Lusaka, Zambia Academic Medical Centre, Amsterdam, The Netherlands Background: The study of HIV heterosexual transmission is Background: Cytotoxic T lymphocyte pressure that selects crucial to the design of better HIV vaccines. We have fol- for escape mutations in CTL epitopes in HIV-1, is a major lowed a cohort of discordant couples in Lusaka, Zambia for determinant driving evolution of HIV-1. Although CTL more than twelve years, in order to optimize prevention escape mutations may allow the virus to evade host cellular approaches to decrease transmission among couples. For immunity, at least some mutations may come at a fitness those couples in whom transmission occurred despite three- cost for the virus. The presentation of viral peptides to CTL monthly counseling and condom provision, the determina- is restricted by the host human leukocyte antigen (HLA) tion of epidemiological linkage of viral strains between type, which varies among individuals. Therefore it is partners is critical for further virological studies. thought that HIV-1 evolves through adaptation to the host Objective: Development of a real time, rapid and easy to use specific HLA types, thereby preventing the presentation of tool to determine linkage of viral strains between partners. its peptides to the immune system. Due to the large varia- Subject and Methods: The goal of this study was to deter- tion in HLA types between individuals HIV-1 needs to re- mine the feasibility of using the heteroduplex mobility assay adapt upon transmission to each subsequent host with a (HMA) to perform real time linkage analyses on transmis- different HLA type. This should be reflected in the muta- sion pairs. HMA was applied to 400 bp fragments PCR tions that occur in CTL epitopes upon transmission, both in amplified from the PBMC DNA of each partner from 12 reversion of mutations that came at a fitness cost in the transmission pairs. After denaturation and renaturation of donor and in new escape mutations that allow escape from gp41 env DNA amplicons, hybridized DNA was elec- CTL pressure in the new recipient. trophoresed on a polyacrylamide gel. A comparison of the Objectives: To test this hypothesis we analysed HIV-1 evolu- HMA profiles from one individual (autologous) with the tion in five therapy naive horizontal transmission couples one obtained by mixing the gp41 PCR amplified DNA from from the Amsterdam Cohort Studies on HIV and AIDS (ACS) both partners (heterologous) was done. A slower migration with known HLA types. In addition, replication kinetics of of heterologous heteroduplexes in comparison with autolo- viruses were tested in vitro and correlated with proviral DNA gous ones was scored as an unlinked transmission as it sequences to study whether reduced viral replication is indeed depicts a diversity of more than 5% which is rarely found a trade-off for CTL escape. in gp41 sequences of linked transmissions. Results obtained Methods: Biological virus clones were isolated from both were compared with direct sequencing. donors and recipients at time points before (donors only) and Results: Twelve transmission pairs were randomly selected after the moment of transmission. Gag and nef genes were for PBMC DNA amplification and analysis by HMA and sequenced and analysed between partners for changes in epi- sequencing in the gp41 region. Among these couples, 4 were topes of donor and recipient HLA type, and between early demonstrated as unlinked by gp41 HMA and 8 as linked. and late viruses in the recipient. These results were confirmed by sequence analysis. In two Results: Viruses with escape mutations in gag and nef were cases, band heterogeneity was identified by HMA in the transmitted (90% of gag escape mutations and 72% of nef autologous samples, consistent with hypermutation observed escape mutations) and subsequent reversion of escape muta- in sequences of the same samples. Nevertheless, these samples tions to WT was found in a considerable amount of epitopes were correctly identified as linked transmission events. restricted by both common and rare HLA alleles. Conclusion: HMA in the gp41 region is an effective field-site Conclusion: Escape from CTL restricted even by non-protective method to study in real-time the epidemiological linkage of HLA types may come at a fitness cost for the virus. HIV in transmission pairs.
AIDS Vaccine 2006 37 Oral Abstracts Sessions 14/8/06 16:40 Page 38
Amsterdam, Netherlands 29 August–1 September 2006
OA01-03 1 AIDS Virus Research Unit, National Institute for Communicable Diseases, Johannesburg, South Africa. 2 Centre for the AIDS Evidence for potent autologous neutralizing anti- Programme of Research in South Africa, University of KwaZulu body titers and compact envs in early infection with Natal, Durban South Africa. 3 Institute of Infectious Disease and subtype C human immunodeficiency virus type 1 Molecular Medicine University of Cape Town, Cape Town, South Africa CA Derdeyn1 , B Li 1, JM Decker 2, 3, RW Johnson1, F Bibollet-Ruche 2, 3, X Wei 2,3, J Mulenga 4, S Allen1, Background: Recently transmitted viruses may show dis- E Hunter 1, BH Hahn3, GM Shaw 2,3 and JL Blackwell 1 tinct antibody neutralization profiles. Therefore character- ising recently transmitted subtype C viruses may be 1 Emory University, Atlanta, GA, USA; 2 Howard Hughes Medical important in the design of candidate vaccines expected to Institute and 3 University of Alabama at Birmingham, target transmitted viruses. Birmingham, AL, USA; 4 Zambia Blood Transfusion Service, Objectives: To evaluate the neutralization profiles of HIV-1 Lusaka, Zambia subtype C viruses from acute infections. Methods: Cloned envelopes from 10 patients recruited with- Background: Information about neutralizing antibody in 2 months of infection (CAPRISA 002 Acute infection responses in subtype C infected individuals is limited, even study, South Africa) were used in pseudovirus neutralization though this viral subtype causes the majority of AIDS cases assays by co-transfection with pSG∆Env. Assays were per- formed in JC-53bl cells and neutralization IC values calcu- worldwide. 50 Methods: The course and magnitude of the autologous neutral- lated. Multiple clones were generated from 2 patients. izing antibody (Nab) response against viral envelope (Env) gly- Neutralization assays were performed with 6 heterologous coproteins present during acute/early infection with subtype B plasma samples and monoclonal antibodies. For 7 patients, and C HIV-1 was compared. Nab responses were evaluated in development of an autologous neutralising response was 6 subtype B infected and 11 subtype C infected subjects over a investigated using sera from 2–15 months post-infection. mean evaluation period of 25 months using a pseudovirus Results: Multiple clones from recently infected individuals reporter gene assay. All subjects in the C cohort were infected showed little difference in neutralization sensitivities, indicat- through heterosexual contact, while 5 of the 6 subjects in the B ing that in early infection a single clone is representative of the population. The IC of clones from 10 patients (median cohort were infected via male-to-male contact. 50 Results: The kinetics and magnitude of the Nab responses var- duration of infection: 35 days) ranged from 1:16 to 1:451 ied among subjects in the B and C cohorts; however, the medi- with 5 patients showing a high to moderate level of resistance to neutralization. Seven of 10 were sensitive to IgG1b12 (IC an IC50 titer reached by antibody in the plasma of subtype C 50 <50µg/ml) and 9 of 10 to IgG-CD4 (IC <25µg/ml). Six of 7 infected subjects, overall, was 3.5-fold higher than in the sub- 50 type B infected subjects (P=0.06). The higher titers of Nabs in patients developed an autologous response within 6 months the C cohort were associated with viruses having significantly of infection with 2 patients showing a response within 3 shorter amino acid length (P=0.002) in the V1-V4 region of the months. However none of these sera showed activity against surface Env glycoprotein, gp120, compared to the B cohort. any other HIV-1 viruses. No correlation was observed Despite the potency of the autologous subtype C Nab response, between rate of disease progression and appearance of an it was not directed against cross-neutralizing epitopes. autologous response. Conclusions: These data demonstrate that subtype C Envs Conclusions: Early in infection, a single Env clone is repre- elicit a potent yet restricted Nab response early in infection sentative of neutralization profiles of the overall popula- that frequently reaches IC50 titers in excess of 1:1,000 and tion. Transmitted HIV-1 subtype C viruses may be suggest that clade-specific differences may exist in Env relatively resistant to heterologous neutralization. immunogenicity or susceptibility to neutralization. However, an autologous neutralising response developed rapidly in subtype C infected individuals and in some cases was detectable within 3 months of infection. OA01-04 OA01-05 Neutralization profiles of HIV-1 subtype C viruses from acute infection Resistance to HIV-1 infection among African
1 1 1 1 2 female sex workers is associated with inhibitory PL Moore , E Gray , I Choge , N Leseka , K Mlisana , KIR in absence of their HLA ligands SS Abdool Karim2, C Williamson3 and L Morris1 and the CAPRISA 002 study team W Jennes1*, S Verheyden 2*, C Demanet 2, CA Adjé-Touré 3, B Vuylsteke 1,3, JN Nkengasong 3,4 and L Kestens 1
38 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 39
Amsterdam, Netherlands 29 August–1 September 2006
1 Department of Microbiology, Institute of Tropical Medicine, and chronic infection. The intestinal lamina propria is also an Antwerp, Belgium; 2 HLA laboratory, Academisch Ziekenhuis- important site of effector CD8+ T-cells containing cytolytic Vrije Universiteit Brussel (VUB), Brussels, Belgium; 3 Projet granules that are high in granzyme A/B but low in perforin. RETRO-CI, Abidjan, Côte d’Ivoire; 4 Division of HIV/AIDS Little is known of the antigenic specificity or the antiviral Prevention, National Center for HIV, STD and TB Prevention, effector functions of these cells. Centers for Disease Control and Prevention, Atlanta, GA. Objectives: To assess the specificity and functionality of HIV- specific CD8+ T-cells from rectal mucosa of chronically *These authors contributed equally to this work. infected individuals, and compare these cells to their coun- terparts from peripheral blood. Background: Natural killer (NK) cell activity is regulated in Methods: In a cross-sectional study, blood and rectal biopsy part by killer immunoglobulin-like receptors (KIR) which tissue were obtained from 24 HIV+ individuals. To assess recognize human leukocyte antigens (HLA) on potential tar- antiviral effector functions, fresh PBMC and rectal lympho- get cells. Both KIR and HLA loci are highly polymorphic. cytes were stimulated with an HIVgag peptide pool and KIR/HLA genotypic combinations have been reported to assessed for production of IFNγ, TNFα, and CD107a by mul- influence disease outcome after HIV or HCV infection. tiparameter flow cytometry. Second, to assess epitope recog- Objective: To study the effect of KIR/HLA combinations on nition, PBMC and mucosal cells were polyclonally expanded protection against HIV infection among HIV-exposed and tested in an ELISPOT assay for reactivity to 123 HIV seronegative (ESN) female sex workers (FSWs) in Abidjan, Gag peptides, organized in 23 pools in a matrix format. Côte d’Ivoire. Results: HIVgag-specific CD8+ T-cells from rectal mucosa Methods: Twenty one ESN and 20 HIV-1-seropositive (SP) produced multiple cytokines and CD107a following chal- FSWs were studied. Genomic DNA was extracted from lenge. In general, the total percentage of virus-specific PBMC. Inhibitory NK cell receptors KIR2DL1, KIR2DL2, responder cells was high in rectum compared to PBMC KIR2DL3, and KIR3DL1, and the activating NK cell recep- (P<0.0001, MANOVA), while the breakdown of cells into tor KIR3DS1, were typed using PCR with sequence specific functional categories (i.e., 1, 2 or 3 functions) was similar primers (PCR-SSP). HLA-B and -C typing was performed between tissue compartments. Patients with viral load <5,000 with PCR and sequence specific oligonucleotides (PCR-SSO). tended to have robust polyfunctional responses Results: ESN FSWs more frequently possessed inhibitory KIR (IFNγ+/TNFα+/CD107a+) in mucosa, while those with VL genes in the absence of their cognate HLA ligand genes: >30,000 tended to have less complex mucosal responses. KIR3DL1-homozygosity in the absence of HLA-Bw4 and a CD107a dominated the mucosal response to HIVgag, despite trend towards KIR2DL2/KIR2DL3 heterozygosity in the low perforin expression. In terms of epitope recognition, absence of HLA-C1. These genotypic combinations poten- responses in blood and mucosa were similar in breadth and tially favor NK cell activation. HP FSWs were characterized specificity as measured by ELISPOT, and immunodominant by corresponding inhibitory KIR/HLA ligand pairings: responses were generally concordant between tissues. KIR2DL3-homozygosity together with KIR2DL1 in the Conclusions: Analysis of mucosal CD8+ T-cell responses to presence of heterozygous HLA-C1/C2 and a trend towards HIVgag demonstrated that these cells are capable of multiple KIR3DL1/HLA-Bw4-homozygosity. These combinations and complex responses, including degranulation. Thus, rectal potentially result in more NK cell inhibition. mucosa is a site of vigorous HIV-specific T-cell responses dur- Conclusions: The data suggest that resistance to HIV-1 infec- ing chronic infection. We are currently extending multipara- tion is associated with absence of inhibitory KIR signals to meter flow cytometry studies to include additional cytokines NK cells. Our findings underscore the role of KIR/HLA inter- (MIP-1β, IL-2, IL-10). Further work will focus on defining actions in innate immune responses against viruses. the relationship between these responses and clinical status.
OA01-06 OA01-07 Robust multifunctional HIVgag specific CD8+ T-cell + responses in rectal mucosa and PBMC during High levels of SIV-specific CD8 T-cells in the chronic infection mucosal tissues of macaques immunized with MVA- based vaccines expressing SIVmac239 gag and tat BL Shacklett 1, JW Critchfield1, D Lemongello1, DH 1,2 1 1 2 Young1, MM Morris 1, JC Garcia 2 and RB Pollard 3 G Silvestri , BS Sumpter , M McQuoid , MR Betts , JC Engram1,2, JD Altman1, MB Feinberg1,3 and DA Garber 1 1 Department of Medical Microbiology and Immunology, University of California, Davis, CA, USA; 2 Division of 1 Emory Vaccine Center, Emory University School of Medicine, Gastroenterology, University of California, Davis, CA, USA; Atlanta, GA, USA; 2 University of Pennsylvania, Philadelphia, PA, 3 Division of Infectious Diseases, School of Medicine, University USA; 3 Merck Vaccine Division, West Point, PA, USA of California Davis, Davis, CA, USA Background: Several recent studies have shown that pathogenic HIV/SIV infections are associated with a rapid and dramatic Background: The gastrointestinal mucosa is an important site virus-induced depletion of mucosal memory CD4+ T-cells that is of HIV replication and CD4+ T-cell depletion during acute
AIDS Vaccine 2006 39 Oral Abstracts Sessions 14/8/06 16:40 Page 40
Amsterdam, Netherlands 29 August–1 September 2006
thought to play an important role in the pathogenesis of AIDS. As such, a desirable property of a candidate AIDS vaccine would be to generate high numbers of HIV-specific mucosal memory CD8+ T-cells that may limit this early loss of CD4+ T-cells. Objective: To assess the level of SIV-specific CD8+ T-cells in mucosal tissues of rhesus macaques (RMs) immunized with two MVA-based candidate AIDS vaccines. Methods: Two groups of five MaMu-A01 RMs were inocu- lated intramuscularly (im) and intradermally (id) with 2 ×108 PFU of a parental MVA vector or an MVA vector in which the uracil-DNA-glycosylase (udg) gene was deleted (MVA∆UDG); both vectors expressed SIVmac239 gag and tat. All animals received three immunizations six weeks apart; examined tissues included peripheral blood, lymph node, rectal mucosa (via rectal biopsies, RB) and lung (via broncho-alveolar lavage, BAL). Lymphocytes isolated from these tissues were examined by flow cytometry; Gag- and Tat-specific CD8+ T-cells were assessed by tetramer staining. Results: In all studied animals both MVA and MVA∆UDG vectors induced detectable levels of Gag-specific CD8+ T-cells, with levels at day 7 after the third immunization of 0.25–2.2% of total CD8+ T-cells in blood; 0.03–0.55% in lymph nodes; 0.03–1.09% in BAL; and 0.33–16.2% in RB. Tat-specific CD8+ T-cells were also detectable in all tissues of all animals, albeit at lower levels. Interestingly, the highest levels of SIV-specific CD8+ T-cells were found in the rectal mucosa, with most of these cells showing a phenotype (CD28-CD95+) indicative of full differentiation to “effector” or “effector-memory” cells. Conclusions: Repeated id/im immunization with MVAs expressing SIV-gene products results in levels of SIV-specific “effector” CD8+ T-cells in GALT that are higher than those observed in peripheral blood, lymph nodes and BAL. SIV challenge studies are in progress to determine whether and to what extent these mucosal SIV-specific CD8+ T-cells may con- fer protection from the early depletion of MALT CD4+ T-cells described in unvaccinated, SIV-infected RMs.
40 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 41
Amsterdam, Netherlands 29 August–1 September 2006
TOPIC 2: VACCINE ANTIGEN SELECTION & DESIGN
OA02-01 OA02-02
Structural basis of antibody neutralization of HIV-1: Identification of portable immunodominant implications for vaccine design sequences leading to efficient HIV epitope presen- tation. IA Wilson, D Calarese, R Cardoso, RL Stanfield, M Zwick, F Brunel, H Katinger, C-H Wong, P Dawson and S Le Gall, P Stamegna and BD Walker DR Burton Partners AIDS Research Center, Massachusetts General Hospital The Scripps Research Institute, La Jolla, CA, USA and Harvard Medical School, Boston MA, USA
Background: Antibodies that can potently neutralize a broad Background: Inducing protective HIV-specific CD8 T cell spectrum of HIV-1 primary isolates are extremely rare, but responses will likely be a critical arm of HIV vaccine design. are invaluable in consideration of strategies to design an The success of this vaccine relies not only on the identifica- effective HIV-1 vaccine. tion of the proper epitopes but also on the capacity of vacci- Objective: To elucidate the antigen specificity of these HIV-1 nated tissues to produce and present these epitopes to the neutralizing antibodies as a means towards rational vaccine immune system. Defining predictable events leading to effi- design. cient epitope presentation will be critical to selection of vac- Methods: Crystal structures for four prototypic, HIV-1 neu- cine sequences for this highly diverse pathogen. The tralizing antibodies [b12, 2G12, 447-52D, and 4E10] have MHC-I-restricted antigen processing pathway includes the been determined in a variety of forms. degradation of proteins by the proteasome, cytosolic and Results: Anti-gp120 antibody 2g12 recognizes a cluster of endoplasmic reticulum (ER)-resident peptidases and precedes high-mannose sugars on the surface of gp120. 2g12 com- epitope presentation to CTL. plexes with 1,2-di-mannose at 1.75Å and with Man 9 Objective: To define intracellular events contributing to effi- GlcNAc at 3.0Å illustrates the mechanism for the unexpect- 2 cient epitope presentation. ed high affinity for a carbohydrate epitope. The 2g12 Fab Methods: We have developed novel techniques to follow arms dimerize via exchange of their V domains so as to form H the degradation of HIV sequences into epitopes in extracts a multivalent binding surface for carbohydrates. Recent stud- from human primary cells. Long HIV peptides are incubat- ies with a variety of oligomannoses have revealed the fine ed with PBMC extracts and peptides produced during specificities of the carbohydrate response that could be used degradation are identified and quantified by mass spec- to design a carbohydrate-based vaccine. 4E10 is the most trometry and by HPLC profile analysis. In order to mea- broadly neutralizing of all HIV-1 antibodies and is effective sure the antigenicity of the mix of purified digested against all clades and subtypes of HIV-1. The 4E10 structure peptides, they were used to pulse HLA-matched B cells in a with a gp41 peptide showed a helical conformation for the 51Cr release assay using epitope-specific CTL. epitope and constrained helical peptides have now been Results: Antigen processing contributes to immunodominance designed as possible vaccine candidates. of HLA-A3-restricted p17 RK9 by producing more peptides Conclusions: Structural studies of these four more broadly encompassing only RK9 and causing a progressive loss of pep- neutralizing anti-HIV-1 antibodies have then not only eluci- tides containing the overlapping subdominant KK9. We dated how each antibody interacts with its respective anti- hypothesized that sequences flanking a dominant epitope could genic site in either gp120 or gp41, but also have given be transposed to a subdominant epitope to increase subdomi- fascinating insights into how the immune system is able to nant epitope presentation. We designed a reporter HLA-A3 epi- evolve strategies to overcome challenges in accessing deeply- tope (ATK9 in RT) flanked with its own sequences buried epitopes (b12), epitopes of low antigenicity (2g12), (WT-ATK9-WT) or that of a dominant A3 restricted epitope, epitopes that vary in sequence (447-52D), and transiently- (DOM-ATK9-DOM) or of KK9 (sub-ATK9-sub). The degrada- accessible epitopes (4E10). The novel methods of antigen tion of DOM-ATK9-DOM yielded 4 times more ATK9 than recognition used by each of these rare, but highly effective, WT-ATK9-WT or sub-ATK9-sub. Accordingly, digestion prod- antibodies have provided a plethora of new ideas for the ucts of DOM-ATK9-DOM elicited the strongest ATK9 CTL design of novel HIV-1 immunogens to elicit such antibody response. Specific N-flanking residues responsible for increased responses and are being harnessed in a retrovaccinology or decreased ATK9 production were identified. approach for HIV-1 vaccine design. Conclusions: Our results present the first identification of portable flanking sequences that modulate expression of MHC-I-restricted epitopes. Such sequences could increase the production and antigenicity of an irrelevant epitope. Defined changes in HIV epitope flanking sequences may be used to modulate CTL responses in vaccine vectors.
AIDS Vaccine 2006 41 Oral Abstracts Sessions 14/8/06 16:40 Page 42
Amsterdam, Netherlands 29 August–1 September 2006
OA02-03 to medium-frequency 9-mers, and exclude extremely rare 9- mers. For relatively conserved HIV-1 proteins, globally-opti- Enhancing CTL-epitope diversity in polyvalent mized mosaics cover individual clades nearly as well as do vaccines: potential for worldwide HIV-1 coverage clade-optimized mosaics; therefore, a mosaic vaccine with worldwide coverage may be feasible. WM Fischer 1*, S Perkins 1*, J Theiler 1, T Bhattacharya1,2, K Yusim1, R Funkhouser 1, C Kuiken1, B Haynes 3, NL Letvin4, BD Walker 5, BH Hahn6 and BT Korber 1,2 OA02-04 Humoral immunity directed at gp41-MPR for the prevention of HIV-1 mucosal transmission 1 Los Alamos National Laboratory, Los Alamos, NM, USA; 2 Santa Fe Institute, Santa Fe, NM, USA; 3 Duke University School of TS Mor 1, N Matoba1, BC Geyer 1, A Alfsen2, CJ Arntzen1, Medicine, Durham, NC, USA; 4 Department of Medicine, Beth M Bomsel 2 Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA; 5 Infectious Disease Division, Massachusetts 1 School of Life Sciences and The Biodesign Institute, Arizona General Hospital, Harvard Medical School, Boston, MA, USA; and State University, Tempe, AZ, USA; 2 Department de Biologie 6 Department of Medicine, University of Alabama at Cellulaire, Institut Cochin., Paris, France Birmingham, Birmingham, AL, USA *These two authors contributed equally to this study Background: Because transmission of HIV-1 mainly occurs at the mucosal surfaces of the vagina and the lower gastroin- Background: An effective HIV-1 vaccine must address viral testinal tract, a prophylactic HIV-1 vaccine should aim at pre- diversity, and will probably need to induce CTL-responses. venting the establishment of chronic infection by protecting Current vaccine proposals (natural-sequence cocktails, the mucosa. Of the various mechanisms that were invoked to inferred ancestral sequences, and consensus sequences) fall explain mucosal transmission of HIV-1, transcytosis across well short of optimal diversity. the epithelium through enterocytes and M-cells may be espe- Objectives: We present an optimization method using natur- cially important at the monostratified lining of the intestines, al sequence data to design “native-like” vaccine antigens with rectum and endocervix. Recently, an important role in the improved coverage of CTL epitopes. process was demonstrated for the highly conserved mem- Methods: We developed a genetic algorithm to design poly- brane proximal region (MPR) of gp41. valent “mosaic” antigen sets. These protein sequences, gener- Objectives: To develop an effective immunization regimen for ated by recombination of natural sequences, are generic CTB-MPR a translational fusion protein consisting of HIV-1 proteins, identical to no particular isolate, but sum- 649–684 the cholera toxin B subunit and a 36-residue peptide corre- marizing the diversity of a set of isolates. Each set of mosaics sponding to MPR. approaches maximal coverage of potential T-cell epitopes (as Methods: Bacterially-produced CTB-MPR was intranasal- 9-mers); rare and unique-to-strain 9-mers are excluded. 649–684 ly and/or intraperitoneally administered to investigate several Results: We have generated mosaics for Gag, Nef, Pol, and prime-boost heterologous route immunization regimens. Env. Mosaic antigen sets resemble natural sequences, but Results: Mucosal priming with the adjuvant cholera toxin elicit- have better coverage of viral diversity (per-site and per- ed significant levels of vaginal IgA and serum IgG specific to sequence) than natural- or consensus-sequence vaccine candi- MPR. Systemic boosting after mucosal priming enhanced the dates. For example, for world-wide Gag sequences, a levels of serum and mucosal antibodies. Systemic priming previously-proposed A+B+C subtype cocktail exactly match- induced a strong serum anti-MPR IgG response, which was effi- es 55% of potential epitopes [partially (≥8/9) matching ciently recalled and augmented by either systemic or mucosal 79%]; our 3-antigen mosaic set perfectly matches 70% of boosting. However, this regimen was less effective in inducing potential epitopes (matching 85% at ≥8/9). Increasing the anti-MPR sIgA. The serum anti-MPR IgG subtype profile size of the set increases coverage: a 4-antigen mosaic set per- revealed that both IgG1 and IgG2a were induced in all the fectly matches 74%, and a 6-antigen mosaic set, 78%, of immunization regimens, and that mucosal co-administration of global Gag sequences. For within-clade and global-versus- cholera toxin shifted the bias to the latter subtype. Most impor- single-clade comparisons, mosaics provided 4–8% better cov- tantly, mucosal antibodies elicited by this regimen significantly erage than the best coverage possible with a given number of inhibited HIV-1 transcytosis in a human tight epithelium model. natural sequences. Mosaics have no unnatural 9-mers, and We also report on the expression in plants of this chimeric pro- fewer unusual 9-mers than other vaccine candidates: a 3-anti- tein and discuss its oral immunogenicity. gen mosaic set has 1/4 as many <1% frequency 9-mers as an Conclusions: Of the immunization regimens examined, optimal natural-sequence set, and 1/10 as many as a previous mucosal priming with adjuvant followed by systemic boost- 3-subtype cocktail. ing exhibited the best response in respect to either systemic or Conclusions: Our method generates a small number of mucosal anti-MPR Abs. To our knowledge, CTB-MPR native-like proteins that provide diversity coverage compara- 649–684 is the first immunogen based on the gp41 membrane-proxi- ble to thousands of individual peptides. Mosaic sets have mal region that reported to elicit transcytosis-neutralizing more balanced diversity than sets of natural strains or con- Abs against a primary HIV-1 isolate. Plant production of sensus sequences: they are less redundant, include more low- mucosally-targeted immunogens could be particularly useful
42 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 43
Amsterdam, Netherlands 29 August–1 September 2006
for immunization programs in developing countries, where OA02-06 desirable product traits include low cost of manufacture, heat stability, and needle-free delivery. The importance of antibody effector function in the ability of serum IgG to protect against vaginal challenge with SHIV OA02-05 AJ Hessell1, L Hangartner 1, M Hunter 2, G Landucci4, Improvement of V1/V2-deleted HIV-1 envelope D Forthal 4, PWHI Parren3, PA Marx 2 and DR Burton1 glycoprotein trimers through forced virus evolution 1 Departments of Immunology and Molecular Biology, The 1 1 1 1 1 RW Sanders , I Bontjer , G Pollakis , E Verkade , K Tuin , Scripps Research Institute, La Jolla, CA, USA; 2 Tulane Regional 1 1 2 2 J van Uriaan , C Baldwin , E Michael , JP Moore , Primate Center, Tulane University, Covington, LA, USA; 1 1 WA Paxton and B Berkhout 3 Genmab, Utrecht, The Netherlands; 4 University of California, Irvine School of Medicine, Irvine, CA, USA 1 Dept. of Human Retrovirology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands; 2 Dept. Background: In previous studies we have shown that the of Microbiology and Immunology, Weill Medical College, Cornell HIV-1 neutralizing monoclonal antibody IgG1 b12 provides University, New York, NY, USA protection to macaques challenged with the SHIV virus. 162P This model for heterosexually transmitted human HIV-1 Background: The HIV-1 envelope glycoprotein (Env) complex infection was used to study protection against vaginal SHIV challenge by two F -effector function crippled ver- is the principal focus of neutralizing antibody-based vaccines. 162P c The functional Env complex is a trimer consisting of six indi- sions of IgG1 b12. The first of these mutated IgG1 b12 vari- vidual subunits (3 gp120’s, 3 gp41’s). The individual subunits ants, K322A (KA), was deficient in complement activation and had a slightly reduced affinity for F Rs. The second b12 have proved unsuccessful as vaccines presumably because they c mutant, L234A, L235A (LALA), did not bind F Rs and dis- do not resemble the functional Env complex. In addition, vari- c able domains and carbohydrates shield vulnerable neutraliza- played no complement activation. Objectives: Investigation of the role of F R- and complement- tion epitopes on the functional Env complex. Deletion of c variable loops has been shown to improve gp120’s immuno- mediated effector functions of neutralizing antibodies in pro- genicity, however, problems have been encountered when tection against viral infection. introducing such modifications in stabilized Env trimer con- Methods: Each group of 9 female macaques was treated structs. Thus, loop deletion can cause problems that are not with 25 mg/kg of the corresponding antibody preparation apparent in the context of monomeric gp120. administered by intravenous injection one day prior to vagi- nal challenge with SHIV . Four control animals were Objectives: To generate variable loop deleted Env trimers 162P3 using virus evolution. treated with a human IgG1 isotype control. Blood viremia, Methods: Viruses containing deletions in the variable transferred antibody levels, and neutralizing activity of domains were passaged for prolonged time and screened for serum were monitored. the appearance of faster replicating variants and the acquisi- Results: In the isotype control treated group, 4 of 4 animals tion of secondary reversions. Secondary reversions were became infected, indicating a very reliable viral challenge. remade or cloned in an expression vector for stabilized Env Only 1 of 9 animals treated with wild-type IgG1 b12 became gp140 trimers and the resulting Env proteins were analysed infected, confirming reliable protection with the antibody. by SDS-PAGE, BN-PAGE and western blot. Similarly, only 1 of 9 animals treated with the complement Results: After prolonged virus culture we identified faster activation-deficient KA-variant of IgG1 b12 became infected. replicating viruses. Compensatory mutations included addi- Yet, in both of these viremic animals, some degree of protec- tion and/or removal of specific carbohydrates, changes in the tion was observed in terms of an attenuated onset and mag- nitude of viremia. In contrast, transfer of the F R non-binding disulfide bonded architecture of the V1/V2 stem and changes c in distal parts of gp120 and gp41. Some secondary reversions and complement activation-deficient LALA-variant of IgG1 appear to be specific for the design of the V1/V2 deletion. b12 left 4 of 9 animals unprotected, and these animals expe- Other reversions appear to be generic ∆V1/V2 compensa- rienced similar viremia as control animals. Furthermore, SHIV -specific antibody-dependent cell-mediated virus tions, since they were found in several different ∆V1/V2 162P3 backgrounds. The functional analysis of these reversions is in inhibition (ADCVI) could only be detected in IgG1 b12 and progress. Selected variants were cloned in recombinant stabi- KA-variant treated animal sera, but not in the sera of LALA- lized Env trimer constructs and improve folding, trimeriza- variant treated animals. Conclusions: This data clearly demonstrated that F -medi- tion, expression and secretion. c Conclusions: These results show that we can indeed select for ated effector functions played a significant role in the abili- functional loop-deleted Env variants through forced virus evo- ty of neutralizing antibodies to confer protection against lution. Furthermore, these evolved Env variants improve pro- HIV-1. Loss of complement activating ability alone had no tein folding expression and trimerization and may be useful as effect on protection. However, loss of both complement and F R binding had a marked detrimental effect. vaccine antigens. Immunogenicity experiments are underway. c
AIDS Vaccine 2006 43 Oral Abstracts Sessions 14/8/06 16:40 Page 44
Amsterdam, Netherlands 29 August–1 September 2006
OA02-07
Preclinical safety and immunogenicity studies of adenovirus serotype-35-based vectors in rabbits and non-human primates
J Mascola1, C Cheng1, J Gall 2, RL Sheets1, J Stein1, N Letvin1,3, S Rao1, CR King 2, P Gomez1, C Andrews1, V Dang1, B Butman 2, VK Haque 2 and G Nabel1
1 Vaccine Research Center, NIAID, Bethesda, MD, USA; 2 Genvec Incorporated, Gaithersburg, MD, USA; 3 Harvard Medical School, Boston, MA, USA
Background: Recombinant adenovirus serotype 5 (rAd5)–based vaccines have demonstrated immunogenicity in human subjects, even those with pre-existing Ad5 immunity. However, the high seroprevalence of naturally occurring Ad5 immunity in the developing world may limit the utility of these vaccines. Adenovectors based upon the less prevalent aden- ovirus serotype 35, may evade Ad5 neutralizing antibodies. Objectives: To evaluate the immunogenicity and reactogenic- ity of adenovectors constructed with Ad5 and Ad35 compo- nents in preclinical studies. Methods: Three groups of Rhesus macaques (6/group) were first vaccinated with either 1x10e10 particle units (PU) repli- cation–defective rAd5, rAd35 or a chimeric rAd35 construct containing the shaft of Ad5 [rAd35/Ad5(shaft)], all expressing a gp140 clade A HIV-1 envelope. Animals were then boosted at week 12, with a heterologous adenovector as follows: group 1 rAd5, followed by rAd35 boost; group 2 rAd35, followed by rAd5 boost, and group 3 rAd35/Ad5(shaft), followed by rAd5 boost. Immune analyses of peripheral blood lymphocytes included ELIspot assays and serum was assayed by anti-HIV-1 Env ELISA. In other experiments, rabbits were inoculated once (1×10e12 PU) with adenovectors described above (but express- ing a luciferase protein gene rather than gp140) and body tem- perature changes were monitored daily for 5 days post-injection. All injections were delivered intramuscularly by needle-and-syringe. Results: The strength of effector T-cell response to gp140, as measured by ELISpot assay in monkeys, ranked rAd5>rAd35>rAd35/Ad5(shaft) at 2, 4 & 8 wks after the ini- tial priming. After boosting, the highest overall T-cell responses (~1500SFC/1x10e6 cells) were observed in group 2 rAd35–immunized monkeys, boosted with rAd5. Pre-boost ELISA titers (3 & 6 wks post-injection) ranked rAd5>rAd35>rAd35/Ad5(shaft). All adenovectors showed mild fever in rabbits, in the first 24 hours p.i., resolving with- in 48 hours in most animals. Conclusions: A heterologous rAd priming and boosting combination regimen can induce augmented HIV-specific immune responses in monkeys compared to a single shot regimen. Vaccines designed around the rAd35 backbone will undergo further preclinical and Phase I clinical evalua- tion, in order to determine if vaccination with alternative Ad serotypes are safe and immunogenic in the background of pre-existing Ad5 antibodies.
44 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 45
Amsterdam, Netherlands 29 August–1 September 2006
TOPIC 3: PROPHYLACTIC VACCINE TRIALS
OA03-01 increased with higher doses of vaccine. Preliminary analysis of data from European volunteers showed that a single A Phase I study to evaluate the safety and immuno- intra-muscular injection of tgAAC09 at the doses tested did genicity of a recombinant adeno-associated virus not elicit significant immune responses. Conclusions: Vaccination with tgAAC09 appears to be safe HIV vaccine and well-tolerated. Immunogenicity data after single injec- tion (Indian volunteers) and boost vaccination (European 1 2 3 4 N Clumeck , S Mehendale , J van Lunzen , E Vets , volunteers) will be presented. Future clinical trials will be 5 1 2 2 J Rockstroh , K Kabamba , S Sahay , M Thakar , directed at determining the safety and immunogenicity of RS Paranjape 2, T Glaunsinger 3, P Johnson6, J Gilmour 7, higher doses of tgAAC09 and determining the optimal L Wilson8, P Anklesaria8, C Schmidt7, J-L Excler 9, interval for boost vaccination. A Heald8 and P Fast7
1 St. Pierre University Hospital, Brussels, Belgium; 2 National OA03-02 AIDS Research Institute, Pune, India; 3 University of Hamburg, Hamburg, Germany; 4 SGS Biopharma, Antwerp, Belgium; Multigene, multiclade HIV-1 plasmid DNA prime 5 Universitatsklinikum, Bonn, Germany; 6 The Children’s and MVA boost is safe and highly immunogenic Hospital of Philadelphia, Philadelphia, USA; 7 International in healthy human volunteers AIDS Vaccine Initiative, New York, USA; 8 Targeted Genetics Corporation, Seattle, USA; 9 International AIDS Vaccine E Sandström1, B Wahren2,, B Hejdeman1, C Nilsson2, Initiative, New Delhi, India A Bråve2, G Bratt2 , M Robb4, J Cox 4, T VanCott8, M Marovich4, R Stout 6, S Aboud3, M Bakari 3, Background: Intra-muscular delivery of HIV genes enclosed K Pallangyo3, B Moss 7, P Earl7, N Michael4, D Birx 5, within recombinant adeno-associated virus (rAAV) protein F Mhalu3 Wahren B 2, G Biberfeld2 for the HIVIS Study capsid has been shown to be a potent inducer of both anti- Group bodies and T-cell responses in animal studies. tgAAC09, con- sisting of single-stranded DNA from Clade C HIV-1 genes for the gag, protease and part of the reverse transcriptase proteins 1 Venhälsan, Dept Infectious Diseases, Karolinska University enclosed within a rAAV serotype 2 protein capsid, was devel- Hospital, 2 Swedish Institute for Infectious Disease Control oped as one component of a multi-component HIV vaccine. (SMI), Karolinska Institute, Stockholm, Sweden; 3 Muhimbili Objective: To evaluate the safety and immunogenicity of University College of Health Sciences, Dar es Salaam, Tanzania; tgAAC09 in healthy, HIV-seronegative volunteers. 4 Walter Reed Army Institute for Research, Rockville, MD, Methods: In this dose-escalation study, 80 healthy, HIV- 5 Global AIDS Programme, Centers for Disease Control, Atlanta, uninfected volunteers (50 in Europe and 30 in India) GA, 6 Bioject, Portland, OR, 7 Laboratory of Viral Diseases, received a single intra-muscular injection of tgAAC09 at National Institutes Allergy and Infectious Diseases, NIH, 9 10 doses of either 3×10 DRP (n=16), 3×10 DRP (n=24), Bethesda, MD, 8 Advanced BioScience Laboratories, Kensington, 3×1011 DRP (n=24) or placebo (n=16). In addition, 21 of 50 MD USA volunteers in Europe received a boost vaccination of tgAAC09 at a dose of 3×1011 DRP or placebo a median of 57 weeks after first vaccination (range 40 to 82 weeks). Background: DNA priming with MVA boosting has been Data collection and analysis are ongoing. reported to be poorly immunogenic in human HIV-1 vaccine Results: The vaccine appears to be well tolerated after both trials, despite good results in non-human primates. The cur- initial and boost vaccination. Mild to moderate local reac- rent study aims to optimize this mode of immunization. togenicity, consisting of pain and tenderness immediately Objectives: To investigate the safety and immunogenicity of after vaccination, were experienced by approximately 20% plasmid DNA boosted with MVA, both expressing HIV genes. of volunteers. Systemic reactogenicity, consisting of mild Methods: 40 healthy HIV negative volunteers, 7 women and chills, malaise, myalgia, headache, nausea and/or fever, 33 men, were randomized to 4 groups [injected id or im with were experienced by approximately 15% of volunteers. Biojector +/- rGM-CSF]. They were primed with 3 injections There was no evidence of vaccine shedding through blood, with 7 DNA plasmids produced by KI/SMI and Vecura; plas- nasal secretions, saliva, urine, semen or vaginal secretions mids containing gp160 of HIV-1 subtypes A, B, C and rev B up to six months after vaccination. At baseline, 48% of were given in the left arm (with or without adjuvant rGM- European volunteers and 95% of Indian volunteers had CSF, sargramostim) and plasmids containing p17/p24 gag A, detectable neutralizing antibodies against AAV2 capsid. As B and Rtmut B in the right arm at months 0, 1 and 3. The expected, the proportion of volunteers with four-fold or same volunteers were re-randomized to a single boost with greater rise in anti-AAV2 capsid neutralizing antibodies MVA with HIV-1 genes env, gag, pol of CRF01A_E produced
AIDS Vaccine 2006 45 Oral Abstracts Sessions 14/8/06 16:40 Page 46
Amsterdam, Netherlands 29 August–1 September 2006
by NIAID and WRAIR at month 9, either with 107 pfu id or prevalence of infection in females (P=0.02) but not in birth- with 108 pfu im. Here we report results for IFN-γ ELISPOT place (P=0.97). Married and previously married volunteers on fresh cells. Peptide pools representing HIV-1 p17B, p24A, had a higher HIV prevalence than the single volunteers p55A, gp120A/B, gp120 B, gp41B and 2 RTB pools were (P=<0.001). Higher educated volunteers tended to have a used for response evaluation. Criteria for positive ELISPOT lower infection rate (Chi-squared for trend P<0.001). were >55 spots/106 PBMCs and 4 times the background. Differences in prevalence among occupations were observed These results were supported by IL-2 Elispot and antigen spe- 9.3% for male/female commercial sex workers, 5.2% for cific lymphoproliferation assays. fishery workers, 2.9% for workers in the entertainment Results: By March 2006, 27 volunteers had received 3 DNA business and 2.5% for farm workers. and one MVA immunization. These immunizations have Conclusions: The prevalence of HIV infection among healthy been well tolerated. There were no safety laboratory abnor- screened volunteers is lower than that of the prior prevalence malities. rGM-CSF was associated with influenza-like grade in the cohort studies conducted in both provinces during 3 adverse events in 2 subjects, one of whom only received one 1999–2000 (4.0–4.8%). The difference may reflect the con- DNA injection. One volunteer defaulted after his first injec- tinued decrease in the incidence of HIV over time in tion. 2 weeks after the third DNA injection 11/38 had devel- Thailand. It also may reflect the effect of community educa- oped positive IFN-γ ELISPOT reactivity and 2 weeks after the tion. This information is very useful for both the provincial MVA injection 22/24 so far analysed had new and/or boost- prevention and control plans for HIV-AIDS. ed IFN-γ ELISPOT responses. These results were supported by IL-2 ELISPOT and antigen specific lymphoproliferation assays. The study is still blinded. OA03-04 Conclusions: Three injections with HIV-1 plasmid DNA as prime with a single HIV-1 MVA boost gave strong IFN-α ALVAC HIV/AIDSVAX B/E prime boost, the com- Elispot reactivities 2 weeks after the last injection in over munity Phase III HIV-1 preventive vaccine trial: 80% of healthy individuals. an update - 2006
S Rerks-Ngarm1, P Pitisutthithum2, S Nitayaphan3, OA03-03 J Kaewkungwan 2,J Kim3, M Benenson3, S Gurunathan4, M Gurwith5, AE Brown6, N Michael 6, C Khamboonroeng1, P Prevalence of HIV infections among screened vol- Thongcharoen1 and P Kunasol1 unteers participating in the Phase III HIV vaccine trial in Thailand 1 Department of Disease Control, Ministry of Public Health, Thailand; 2 Faculty of Tropical Medicine, Mahidol University, N Premsri1, C Namwat1, S Rerks-Ngarm1, S Chunsuttiwat1, Bangkok, Thailand; 3 Armed Forces Research Institute of Medical S Nitayaphan2, C Eamsila2, J Kaewkungwal3, Sciences, Bangkok, Thailand; 4 Sanofi Pasteur, Swiftwater, S Wattana4, W Wiriyakijja5, M Benenson2 and J Kim2 Pennsylvania, USA; 5 VaxGen Inc., Brisbane, CA, USA; 6 Walter Reed Army Institute of Research, Rockville, MD, USA 1. Ministry of Public Health–Thai AIDS Vaccine Evaluation Group (MOPH-TAVEG), Thailand; 2. Armed Forces Research Institute of Background: In 2003 the Ministry of Public Health launched Medical Sciences (AFRIMS) Thai and US Components; 3. Faculty a Phase III HIV vaccine trial with the ultimate goal of having of Tropical Medicine, Mahidol University; 4. Chon Buri Provincial an HIV vaccine for Thailand. Health Office, Thailand; 5. Rayong Provincial Health Office, Objectives: To determine if this prime-boost vaccine strategy Thailand prevents HIV infection among young adults and if it can alter disease course after breakthrough infection, Methods: Recruitment teams from communities were trained, Background: The Prime-Boost Phase III HIV Vaccine trial has community leaders were informed, friend referred friend and been conducted in Thailand since September 2003. We selec- various local media were used. Screened volunteers were ran- tively screened healthy Thai adults aged 18–30 years in Chon domized to receive ALVAC HIV canarypox vaccine (or place- Buri and Rayong Provinces to participate in the trial. bo) prime at 0, 4, 12, 24 weeks, with an AIDSVAX B/E gp120 Screening activities were completed by December 2005 with (or placebo) boost at 12, 24 weeks. Follow up for safety, HIV over 16,000 volunteers enrolled into the vaccine trial. status, counseling and social impact events will be performed Objective: To describe preliminary epidemiological data of at 6 month intervals for 3 years. those screened volunteers who were HIV infected at the time Results: 26,675 potential volunteers were screened, and of screening. 16,402 were enrolled in the vaccine trial. The major reasons Methods: As of December 2005, 26,675 volunteers had of not enrolled were having preexisting medical illnesses, fail- been screened using the standard EIA and confirmatory ure of the test of understanding, inconvenience to comply tests. Demographic data on the volunteers was obtained with vaccine protocol, and being HIV positive. The average using standardized interview questionnaires. number of enrollees was 481 per month in the first year, Results: There were 419 confirmed infections among which gradually increased to 656 per month for the second screened volunteers (1.6%). There was a significantly higher
46 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 47
Amsterdam, Netherlands 29 August–1 September 2006
year due to expanding the eligible pool of participants to anti-Env titers reached ∼ 1:105. Immune sera were able to rec- nearby health centers. 60% of volunteers were male and ognize primary HIV-1 Env antigens with very diverse gene about half came from other provinces. Half were married and sequences from many subtypes. Low titer but positive neu- had education ≥ than junior secondary school. 56% reported tralizing antibody activities were detected in majority of the that other participants were their referral source to the pro- volunteer sera after protein boosts, against pseudotyped virus- ject. 36%, 22% and 27% reported that the project staff, es expressing not only the homologous but also several het- other participants and information from screening video erologous primary Env antigens. High-level cell mediated respectively were the source of a decision to participate. A immune (CMI) responses have been detected against pooled total 57,857 injections have been administered without Gag or Env peptides by an interferon-γ ELISPOT assay. notable safety or reactogenicity concerns. A total of 953 seri- Protein boost appears effective in further maintaining the Env ous adverse events have been reported thru 1 April 2006. The specific CMI. Over one third of the volunteers developed type majority of were due to motor cycle accidents and were not IV hypersensitivity-like skin reactions shortly after the protein related to vaccination. So far, 382 participation impact events boost, supporting the effects of DNA prime. On going intra- have been reported. The majority (83%) were due to person- cellular cytokine staining data further demonstrated the pres- al relationship types. 94% of the participation impact events ence of vaccine specific CD4+ and CD8+ T cell responses have been resolved satisfactorily. expressing not only IFN-γ but IL-2 as well. Conclusion: Full enrollment of 16,402 volunteers was Conclusions: Therefore the DNA prime and protein boost achieved. The vaccine appears to be safe and follow-up for approach is effective in inducing balanced CMI and antibody three years is planned. responses in healthy human volunteers.
OA03-05 OA03-06
Balanced cellular and antibody responses induced Safety and reactogenicity of a multiclade adenovec- by the polyvalent DNA prime-protein boost HIV-1 tor HIV vaccine at two different doses: a random- vaccine formulation DP6-001 in healthy volunteers ized, placebo-controlled clinical trial (HVTN 054)
1 2 3 2 4 S Lu1, J Kennedy1, S Wang1, K West1, P Goepfert3, L Peiperl , Z Moodie , C Morgan , H Li , N Russell and 5 D Montefiori 4, S Coley1, FA Ennis1, R Pal2 and B Graham P Markham2 1 University of California School of Medicine, San Francisco, CA; 1 Department of Medicine, University of Massachusetts Medical 2 SCHARP, Seattle, WA; 3 HVTN, FHCRC, Seattle, WA; 4 Bill and School, Worcester, MA; 2 Advanced BioScience Laboratories, Inc., Melinda Gates Foundation, Seattle, WA; 5 Dale and Betty Kensington, MD; 3 University of Alabama Birmingham Medical Bumpers Vaccine Research Center, NIAID, NIH Center, AL; 4 Duke University Medical Center, Durham, NC, USA Objective: Evaluate safety and reactogenicity at two doses Objectives: Continued immunogenicity analyses have been of an adenovector HIV vaccine in adenovirus-naive, HIV- conducted for a Phase I clinical trial testing a polyvalent negative volunteers. DNA prime/protein boost AIDS vaccine formulation DP6- Methods: 48 participants were randomized to receive a single 001 in healthy human volunteers. dose of adenovirus type 5 (Ad5) vaccine VRC-HIVADV-014- Methods: The study includes two dosing levels of DNA vac- 00-VP (consisting of 4 replication-deficient adenovectors cines as prime followed by one standard dose of protein expressing HIV subtype B Gag-Pol-Nef and subtypes A, B, boost in adjuvant QS-21. Groups A and B received three and C Env) versus formulation buffer placebo. 24 of these DNA immunizations (1.2 mg total DNA divided equally participants were randomized to receive a single injection of among one clade C gag and five gp120 env DNA vaccines of 10^10 PU (n=20) versus placebo (n=4) by intramuscular which two were from clade B, one each from clade A, clade injection. Following a satisfactory safety review, the other 24 C and clade E) at weeks 0, 4 and 12 by either i.d or i.m, participants were randomized to receive 10^11 PU (n=20) respectively. Group C received a higher dose DNA prime (7.2 versus placebo (n=4). mg divided equally among the same six DNA vaccines) deliv- Results: All participants had Ad5 antibody titer <1:12 at ered by i.m. at the same interval. Protein boosts containing screening, received their injection, and completed follow-up the five recombinant gp120 proteins (75 mg each) matching through day 168. Injection site symptoms post vaccination gp120 DNA vaccine prime were administered to all groups at included pain (40% mild, 15% moderate), tenderness (52% weeks 20 and 28. mild, 23% moderate), and erythema and/or induration Results: Balanced HIV-1 specific antibody and T cell respons- (21%). These symptoms were more common in vaccine arms es were detected in the immunized volunteers. Although posi- than in the placebo arm, and induration was more frequent tive Env-specific antibody responses were rare following DNA at the 10^11 PU dose than at 10^10 PU (P=0.03). Systemic vaccination alone, 100% of subjects had robust antibody signs and symptoms of malaise and/or fatigue (69%), myal- responses after just one protein boost. The peak gia (52%), headache (54%), chills (25%), arthralgia (27%),
AIDS Vaccine 2006 47 Oral Abstracts Sessions 14/8/06 16:40 Page 48
Amsterdam, Netherlands 29 August–1 September 2006
and temperature >37.6 C (21%) were mostly mild or moder- were not necessary due to its high degree of sequence conser- ate, and each occurred with greater frequency and severity at vation. About 66–68% of 168 vaccinees with responses to the 10^11 PU dose than at 10^10 PU (P<0.025 for each the vaccine CAM1 HIV-1 gag (clade B) peptide pool cross- sign/symptom). Four participants, all of whom were in the reacted with either clade A gag or clade C gag sequences. group receiving 10^11 PU or placebo, reported one or more 37% (25/68) receiving the MRKAd5 trivalent vaccine with signs or symptoms of severe reactogenicity: malaise/fatigue responses against JRFL HIV-1 nef (clade B) were able to (in 4), severe myalgia (2), grade 3 fever to 39.4 (1), headache respond to the clade A nef pool, while 19% (13/68) cross- (1) and severe chills (3). All experienced these events within reacted with clade C nef. PBMCs were also assayed against a one day following injection and improved noticeably within series of smaller pools (minipools) each consisting of 8 one day. No significant differences in liver enzymes, bilirubin, sequence-consecutive 9-aa peptides from gag, pol, and nef to blood cell counts, creatinine, or CPK were found between assess the breadth of the T cell responses and to determine the groups at day 14, 28, 84, or 168. extent epitopes mapped to regions of sequence conservation. Conclusions: In volunteers without prior antibody immunity Conclusions: The MRKAd5 Clade B vaccine provides to Ad5, the VRC Ad5-based HIV vaccine was associated with immune responses which are potentially cross-reactive with greater incidence and severity of injection site and systemic viruses of same or different clades primarily by virtue of the reactions at the 10^11 PU dose than at the 10^10 PU dose. gag and pol component. Of the vaccine-associated events rated severe, all involved systemic reactogenicity showing improvement within one day of onset.
OA03-07
Breadth of the HIV-specific cellular immune responses to replication-defective adenovirus HIV vaccines in healthy subjects
D Casimiro, S Dubey, T Tobery, L Kierstead, J Condra, A Finnefrock, R Isaacs, M Robertson, R Leavitt, E Quirk, R Mogg, D Mehrotra and J Shiver
Merck HIV Vaccine Team, and V520 Merck Study Group, Merck & Co, West Point, PA, USA
Objective: We are developing replication-defective aden- ovirus type 5 (MRKAd5)-based vaccines that express reason- ably conserved viral antigens (gag, pol, nef) of the B clade origin in several Phase I studies. The study described below aims to assess the properties of this vaccine to provide immune coverage against viruses across multiple clades in an effort to comprehend the limitations of this vaccine. Methods: Peripheral mononuclear cells (PBMCs) were obtained from healthy subjects in several Phase I studies, each evaluating a different MRKAd5-based vaccine. MRKAd5 HIV-1 gag and the trivalent MRKAd5 gag/pol/nef vaccines were independently tested in healthy adults at low risk of HIV infection (18–50 yrs of age). HIV-specific T cell responses were evaluated by intracellular cytokine staining (ICS) and IFN-γ ELISPOT against full clade B peptide pools, a series of smaller pools from each clade B antigen, and full peptide pools representing near-consensus sequences of clade A and C antigens. Results: The MRKA5 vaccines elicited predominantly anti- gen-specific CD8+ T cells; a smaller fraction (20–30%) of patients with positive ICS responses contained detectable lev- els of virus-specific helper T cells. Cross-clade gag- and nef- specific responses to varying degrees were detected using the ELIspot method; cross-clade reactivity measurements for pol
48 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 49
Amsterdam, Netherlands 29 August–1 September 2006
TOPIC 4: T CELL IMMUNITY
OA04-01 OA04-02
Upregulation of PD-1 expression in HIV specific CD8 PD-1 expression on HIV-specific CD8 T cells is T cells leads to reversible functional exhaustion associated with disease progression
1 1 1 1 L Trautmann , L Janbazian , N Chomont , B Bessette , CL Day1,2,3, P Kiepiela2, S Chetty 2, S Reddy2, JA Brown4, 2 2 1,2 1,2 R Boulassel , J-P Routy , E K. Haddad and R-P Sekaly Q Eichbaum3, AJ Leslie1, HM Coovadia2, PJR Goulder 1,2,3, P Klenerman1, R Ahmed5, GJ Freeman4 and BD Walker 2,3,6 1 Laboratoire d’Immunologie, Université de Montréal, Département de Microbiologie et Immunologie, Montréal, 1 Nuffield Department of Medicine, The Peter Medawar Building Canada; 2 Immunodeficiency Service and Division of for Pathogen Research, Oxford University, Oxford, United Haematology, Royal Victoria Hospital, McGill University Health Kingdom; 2 HIV Pathogenesis Programme, Doris Duke Medical Centre, McGill University, Montreal, Quebec, Canada Research Institute, University of KwaZulu Natal, Durban, South Africa; 3 Partners AIDS Research Center, Massachusetts General Background: The intricate balance of the immune system is Hospital and Division of AIDS, Harvard Medical School, Boston, governed by stimulatory and inhibitory signals. The engage- Massachusetts, USA; 4 Department of Medical Oncology, Dana- ment of one such inhibitory signal is elicited by Programmed Farber Cancer Institute, Department of Medicine, Harvard Medical death 1 (PD-1) to its ligands (PD-L1 and PD-L2). PD-1, a School, Boston, Massachusetts, USA; 5 Emory Vaccine Center and member of the Immunoglobulin gene superfamily, is highly Department of Microbiology and Immunology, Emory University up regulated on activated lymphocytes and monocytes and is School of Medicine, Atlanta, Georgia, USA; 6 Howard Hughes known to regulate T-cell responses. Specifically, PD-1 recep- tor crosslinking inhibits anti-CD3 mediated proliferation and Medical Institute, Chevy Chase, Maryland, USA cytokine production. Blocking the PD-1/ PD-L1 pathway by the administration of PD-1 or PD-L1 specific antibodies into Background: Functional impairment of T cells is a character- mice chronically infected with LCMV restored exhausted istic feature of chronic mouse and human infections. In mice, CD8 T cells to undergo lymphocyte proliferation, cytokine programmed death 1 (PD-1), a negative regulator of activat- secretion, cytotoxic activity and decreased viral load. ed T cells, is significantly up-regulated on the surface of Objectives: During chronic HIV infection, virus-specific CD8 exhausted virus-specific CD8+ T cells, and blockade of this T cells are clonaly exhausted and functionaly impaired. They pathway restores T cell function and reduces viral load. show reduced capacity to produce cytokines and to prolifer- Objectives: The objective of this study was to investigate the ate. The upregulation of PD-L1 was previously reported in role of PD-1 in a chronic human viral infection, and to deter- HIV infection. Therefore, we addressed the role of PD-1 in mine if there was any evidence of enhanced functionality of CD8 dysfunction in HIV infection. virus-specific CD8 T cells following blockade of the PD- Methods: We used tetramers directed against HIV-specific 1/PD-L1 pathway. CD8 T cells and compared to other chronic infection such as Methods: We examined PD-1 expression on HIV-specific EBV and CMV in the same patients. We measured the PD-1 CD8 T cells in 65 clade C infected persons who were naive to expression on tetramer positive cells and assessed the anti-HIV treatments, using a panel of 10 HLA class I impaired function of the PD-1+ T cells by measuring their tetramers specific for frequently targeted epitopes. capacity to produce cytokines and proliferate. Results: We find that PD-1 is significantly upregulated on Results: Here, we show that PD-1 is upregulated in HIV-spe- HIV-specific CD8 T cells and positively correlates with plas- cific CD8 T cells; PD-1 expression is significantly correlated ma viral load (P<0.0001) and inversely with CD4 count with both viral load and reduced capacity of cytokine pro- (P=0.0150). Furthermore, PD-1/PD-L1 pathway blockade duction and proliferation. Of note, CMV and EBV specific results in increased proliferation of HIV-specific CD8 T CD8 T cells from the same donors do not upregulate PD-1 cells (P=0.006). and maintain the production of high levels of cytokines upon Conclusions: These data provide a long elusive link between antigen specific triggering. Blocking PD-1 engagement to its HIV-specific CD8 T cells and viral load, and indicate that the ligand rescues HIV-specific CD8 T cells from clonal exhaus- PD-1/PD-L1 pathway is operative during a naturally tion and allows them to regain their capacity to proliferate acquired persistent viral infection in humans. Moreover, this and survive. demonstration of reversible T cell impairment provides a Conclusion: Our results not only demonstrate the major role potential approach to enhance function of exhausted T cells of PD-1 in the impairment of HIV-specific CD8 T cells, but in chronic HIV infection. further show the reversion of the exhausted cells upon block- ade of PD-1/PD-L1 interaction.
AIDS Vaccine 2006 49 Oral Abstracts Sessions 14/8/06 16:40 Page 50
Amsterdam, Netherlands 29 August–1 September 2006
OA04-03 OA04-04
Phenotype, apoptotic potential and polyfunctional Structural basis for T-cell receptor recognition of characterization of T cells responses associated with HIV-1 epitopes live-attenuated SIV vaccine mediated protection of rhesus macaques G Stewart-Jones1, T Dong2, R Hamer1, G Gillespie 2, J Bell 2, S Rowland-Jones 2, A McMichael 2 and Y Jones1 M Genescà1,2, T Rourke1,2, MB McChesney2 and CJ Miller1,2 1 Division of Structural Biology, Wellcome Trust Centre for Human 1 Center for Comparative Medicine, University of California, Davis, Genetics, University of Oxford, UK; 2 Weatherall Institute of CA, USA; 2 California National Primate Research Center, Molecular Medicine, University of Oxford, John Radcliffe Hospital, University of California, Davis, CA, USA Oxford, UK
Background: SIV-specific T cell responses significantly con- Background: T-cell responses eliminate infected cells from the tribute to the control of virus replication. Further, polyfunc- body by recognising fragments of the viral proteome presented tional T cell responses have been associated with control of in MHC molecules. Viral escape at sites of immune selection virus replication during chronic HIV infection. Therefore, the pressure hamper the ability for the cellular immune response to quality and the strength of T cell responses may be a deter- function effectively and control replication. minant of its effectiveness. Objectives: In order to understand the structural basis for T- Objectives: Define the quality of epitope specific T cell cell receptor recognition that may exert immune pressure on responses of protected and unprotected SIV-vaccinated rhe- the viral epitopes we aimed to characterize the structural sus macaques using polychromatic flow cytometric assays. basis of peptides presentation in MHC class I molecules and Methods: MHC class I haplotypes and Gag-epitope specific T specific binding by T-cell receptors, to understand any rules cell responses of 4 SHIV89.6-vaccinated and SIVmac239- to design T-cell based vaccines. challenged rhesus macaques (2 protected and 2 unprotected) Methods: We used protein refolding, purification and crys- were defined using a matrix of peptides in an IFN-γ ELISPOT tallization to produce diffraction-quality crystals of HIV-1 assay. Frozen cells were used for epitope-specific stimulation specific pMHC/TCR complexes that were capable of dif- in flow cytometry assays using two different 8 colour-panels fracting to high resolution. that included: apoptosis versus survival signals (Bcl-2, Results: We present X-ray crystal structures of three HIV- Caspase-3, 7-AAD) and intracelullar cytokines (IFN-γ, TNF- peptide MHC-TCR complexes with three different peptides α, IL-2, CD107). When possible, different dilutions of a spe- and three TCRs. These are: HLA-B8-EIYRKWII-Vβ cific peptide were used to measure the avidity of the response. 22/Vα9.1 (2.3Å), HLA-B8-FLKEKGGL-Vß6/Vα14.1 (2.8Å) Further, responder T cells were phenotyped and tetramer spe- and HLA-B5703-KAFSPEVIPMF-Vβ17/Vα15.1 (2.4Å). The cific responses were assessed. structures help understand how T-cell epitopes are capable of Results: The vaccinated-protected animals maintained a high- escaping the immune system, either by mutating MHC er ratio of survival signals (Bcl-2) versus death signals anchor residues or T-cell receptor contact residues. The pep- (cleaved caspase-3) in CD4+ T cells and to a lesser extent in tide position 1 side-chain is always solvent exposed in all the CD8+ lymphocytes when stimulated with the specific pep- complexes. Therefore all positions within an epitope, besides tide. Moreover, plasma viral load was highly correlated with position 1, are involved structurally in MHC anchoring or the Bcl-2/caspase-3 ratio in specific T cells. Higher frequen- TCR recognition, with the central region of the peptide inter- cies of polyfunctional, SIV-specific CD8+ and CD4+ T cell acting most with the TCR CDR3 loops. responses were found in the protected animals. Further a Conclusions:This provides a structural perspective to ratio- decline in frequency of SIV-Gag specific naive and central nalising why immune escape mutations to any region of the memory and an increase in the frequency of effector memory peptide, except at P1, are likely to compromise antigen pre- T cell subsets was associated with the lack of protection. sentation or TCR recognition. Different TCRs utilise differ- Discussion: Control of viral replication in a live-attenuated ent recognition mechanisms, implying that specific SIV vaccine context is associated with the induction of sur- alterations at the TCR recognition interface will affect differ- vival signals in T cells and a polyfunctional cytokine profile ent responding T-cells in different ways. Thus diversity in the in response to Gag epitopes. An increase in late effector mem- TCR repertoire may counteract variation at TCR recognition ory T cells more susceptible to apoptosis and less capable of points by having differential dependencies on different polyfunctional cytokine production may play a role in vac- regions within peptide epitopes. cine failure in live-attenuated lentivirus vaccines.
50 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 51
Amsterdam, Netherlands 29 August–1 September 2006
OA04-05 OA04-06
Relationship between perforin expression and Positive association between HIV-1 viral load and degranulation activity in virus-specific CD8 T-cells targeting of an epitope presented by B*5802, an HLA allele associated with disease progression A Harari, C Cellerai, G Tapia and G Pantaleo KC Ngumbela1, P Kiepiela1, CL Day1, 2, DG Kavanagh3, Lab of AIDS Immunopathogenesis, Lausanne, Switzerland H Coovadia1, PJ Goulder1,2,3, BD Walker1, 3 and the HPP Study Group* Background: Perforin expression and degranulation activity, as measured by CD107a mobilization, are currently used to 1 Nelson R Mandela School of Medicine, University of KwaZulu define cytotoxic memory CD8 T-cells. The relationship Natal, South Africa; 2 Department of Paediatrics, Nuffield between these markers is unclear. Department of Medicine, Oxford, UK; 3 Partners AIDS Research Objectives: to investigate the relationship between degranu- Center, Harvard Medical School, Boston, Massachusetts lation activity and perforin expression. Methods: A variety of virus-specific CD8 T-cell responses Background: In HIV+ persons, some class I alleles are associ- including CMV (n=14), EBV (n=7), Flu (n=5) and HIV-1 ated with effective control of viremia, whereas others are from both progressors and LTNP (n=11) were identified associated with rapid disease progression. Among the most using tetramer complexes or peptide stimulation and divergent clinical outcomes are the relatively good prognosis analysed for proliferation capacity, expression of perforin in HLA-B*5801 expressing persons and the poor prognosis and CD127 (i.e. IL-7Rα) and CD107a mobilization. CCR7 with HLA-B5*802. These two alleles differ by only 3 amino and CD45RA were used to assess T-cell differentiation. acids in the regions involved in HLA-peptide binding recog- Results: Comparison of perforin expression and CD107a nition. The impact of minor differences in HLA sequences in mobilization on CD8 T-cells for the different viruses showed epitope recognition have to be studied. a discrepancy between these two markers. In particular, high Objectives: To determine whether the clinical outcome differ- levels of CD107a mobilization were observed in Flu-specific ences associated with HLA-B*5801 and HLA-B*5802 CD8 T-cells despite the absence of perforin expression expression are associated with differences in the HIV-1 spe- (P=0.001). Furthermore, combined expression of perforin cific CD8+T cell response. and CD127 revealed the existence of 3 populations of virus- Methods: Blood was collected and high resolution HLA typ- specific CD8 T-cells: CD127+perforin- cells that represented ing was performed on 1068 HIV-infected therapy naive sub- the large majority (90%) of Flu-specific CD8 T-cells; CD127- jects from 3 clinic sites in Durban, South Africa. Plasma viral perforin- cells that were the majority (64%) of EBV-specific load and CD4+ counts were measured using the Roche CD8 T-cells; and CD127-perforin+ cells that were dominant Amplicor assay and Becton-Dickinson FACSCalibur Flow for CMV-specific CD8 T-cells (43%). HIV-specific CD8 T- Cytometer, respectively. Four hundred and ten 18-mer pep- cells from progressors were mostly CD127-perforin- cells tides, spanning the entire HIV genome, were used to screen while LTNP had significantly more HIV-specific CD127+ study subjects for HIV-specific T cell responses by IFNγ CD8 T-cells (21 versus 7%, P<0.01). Progressors had more elispot assay. perforin+ cells than LTNP (10 versus 4%). Combined staining Results: Overall breadth and magnitude of HIV-1 specific with CCR7 and CD45RA showed that CD127+perforin, CD8+T cell responses were similar in these two groups, but CD127-perforin- and CD127-perforin+ were CD45RA- epitope presentation by HLA-B*5802 contributed signifi- CCR7+, CD45RA-CCR7- and CD45RA+CCR7- CD8 T-cells, cantly less to overall response as compared to B*5801- respectively. Ag-specific proliferating CD8 T-cells were con- restricted CD8+T cells. Moreover, the viral load in tained within the CD127+perforin- cells population. Of inter- HLA-B*5802 positive persons was actually higher and CD4+ est, 7 days after in vitro Ag-specific stimulation, there was a count significantly lower when this allele contributed to over- substantial increase in the CD127-perforin+ cell subset for all all CD8+T cell response, which occurred exclusive through a virus-specific CD8 T-cell responses: Flu: from 2 to 42%; EBV: single epitope in Env, providing clear evidence for differential from 3 to 45%; CMV: from 43 to 60%; HIV: from 10 to antiviral activity of CD8+T cell responses. 25%; (all P<0.05). Conclusions: These data indicate that minor differences in Conclusions: These data indicate that perforin and CD107a HLA sequence can have major impact on epitope recogni- mobilization represent independent and discordant markers tion, and HLA B5802 is at least ineffectual if not actively of CD8 T-cells with potential cytotoxic capacity and that adverse in containment of viremia. These data provide exper- only perforin expression correlated with the stage of differ- imental evidence that not all epitope-specific responses con- entiation of memory CD8 T-cells. Of note, the balance tribute to immune containment, a better understanding of between proliferative/precursor and cytotoxic/effector sub- which is essential to shed light on mechanisms involved in sets of virus-specific CD8 T-cells is substantially influenced HIV disease progression. by the Ag load/exposure.
AIDS Vaccine 2006 51 Oral Abstracts Sessions 14/8/06 16:40 Page 52
Amsterdam, Netherlands 29 August–1 September 2006
OA04-07
Amino acid variation of targeted HIV-1 specific CD8+ T-cell epitopes correlates with HLA Class I disease association
A Bansal1, L Yue1, J Conway1, K Yusim2, RA Kaslow1,3, CM Wilson1,4, J Kappes1 and PA Goepfert1,5
1 Department of Medicine; 2 Los Alamos National Laboratory, Los Alamos, NM, USA and the Adolescent Trials Network for HIV/AIDS intervention; 3 School of Public Health; 4 Department of Pediatrics and 5 Department of Microbiology, University of Alabama at Birmingham, Alabama, USA
Background: Among HLA Class I alleles, B*57 and B*27 have been associated with a slower progression while HLA B*35 and B*53 have been associated with a rapid progres- sion to AIDS. Objectives: To comprehensively analyse HIV-1 specific CD8 T-cell responses in individuals whose HLA Class I alleles are associated with differential rates of progression to AIDS. Methods: In a cross-sectional study, T-cell responses in sub- jects expressing either HLA-B*57(7), B*35(7) or B*53(9) were compared with 29 subjects with other HLA Class I alleles. Eighteen subjects expressing either B*57, B*35 or B*53 were followed longitudinally for over 4 yrs. The quantity and the quality of the T-cell response was mea- sured in an IFN-γ ELISpot and polychromatic flow cytome- try assays, respectively. Results: Cross-sectionally, HLA-B*57 carriers had a lower plasma viral load (log VL of 2 versus 3.6 RNA copies/ml; p value <0.01), lower magnitude and breadth of CD8+ T-cell responses (P-value for each <0.03) compared to B*35/B*53 or the rest of the cohort. In addition, HLA B*57 subjects tar- geted Gag and p24 whereas B*35/B*53 carriers targeted Nef predominantly. This focus of response was also maintained in the 18 subjects followed over a 4 year period. In the longitu- dinal study, the median viral load increased from 110 to 225 in B*57 subjects; for B*35/B*53 carriers this increase was from 3,900 to 14,094 RNA copies/ml. Viral sequencing demonstrated that mutations in key Nef and Gag epitopes were associated with a loss of virological control in B*35/53 and B*57 subjects, respectively. The quality of HIV specific CD8 T cell response to immunodominant epitopes did not differ significantly between the B*57 and B*35/B*53 group as measured by peptide avidity, secretion of IFN-γ, IL-2, and TNF-α cytokines, or expression of CD107(degranulation) and CD127(memory) markers. Conclusions: Targeting of CD8 T-cell epitopes in Nef by B*35/B*53 carriers may be responsible for their poor HIV prognosis. Factors other than the quality and efficiency of the Nef response may be involved in the rapid progression to AIDS in B*35/B*53 carriers. It is possible that the potency of the Nef CTLs for killing in vivo is much lower compared to CTLs targeting p24.
52 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 53
Amsterdam, Netherlands 29 August–1 September 2006
TOPIC 5: B CELL IMMUNITY
OA05-01 OA05-02
HIV-1 subtype A envelope variants from early in HIV multiplication in macrophages and dendritic infection demonstrate variable neutralization cell is blocked by some non-neutralizing antibod- sensitivity, in part due to changes in the trans- ies directed against the V3 loop or the principal membrane domain immunodominant domain gp41
CA Blish1,2, M-A Nguyen1,3, K Mandaliya4 and J Overbaugh1 C Moog1, V Holl1, M Peressin1, T Decoville1, S Schmidt1, S Zolla-Pazner 2, and AM Aubertin1 1 Division of Human Biology, Fred Hutchinson Cancer Research Center, Seattle, WA, USA; 2 Division of Allergy and Infectious 1 INSERM U778, University Louis Pasteur of Strasbourg, Institute Diseases, University of Washington School of Medicine, Seattle, of Virology, France; 2 Veterans Affairs and NYU Medical Centers, WA, USA; Departments of Biochemistry and Chemistry, New York, USA University of Washington, Seattle, WA, USA; 4 Coast Provincial General Hospital, Mombasa, Kenya Background: During HIV infection, CD4+ T lymphocytes are the major targets for HIV-1. In addition to CD4+ T lympho- Background: An effective HIV-1 vaccine must block the cytes, immature dendritic cells (DC), present at mucosal transmitted virus variants that initially establish a new infec- sites, are among the first cells infected by HIV. These DCs, tion. Thus, it is critical that such viruses from prevalent sub- together with macrophages, could participate in HIV-1 types be isolated and characterized as they can provide transmission, dissemination and persistence. insights into both vaccine efficacy and immunogen design. Objectives: The aim of this study was to analyse the mecha- Objectives: To evaluate the neutralization profile of HIV-1 nism of HIV inhibition mediated by antibodies when envelope variants from early in infection from individuals macrophages or immature DC were the targets. infected heterosexually with subtype A HIV-1. Methods: CD4 T lymphocytes obtained from PHA-stimu- Methods: Fifteen full length HIV-1 envelope clones from 28- lated PBMC and monocyte-derived macrophages (MDM) 75 days post infection (Long et al., AIDS Res Hum or -derived DC (MDDC) generated by differentiation of + Retroviruses 2002; 18:567–576) were used to generate CD14 monocytes were infected with primary isolates; two pseudoviruses for use in neutralization assays. The suscepti- subtype B (BaL and Bx08 and a subtype C (TV1) described bility of these viruses to autologous and heterologous plas- to be relatively “resistant” to antibody neutralization. mas, including pooled plasma, and the monoclonal Inhibitory activity of monoclonal antibodies (mAbs) was antibodies b12, 2G12, 2F5, and 4E10 was examined. Site analysed using a neutralization assay that determines, by directed mutagenesis was used to engineer mutations found flow cytometry, the percentage of infected cells by detection in neutralization sensitive variants in order to define residues of intracellular viral p24 important in determining neutralization susceptibility. Results: We found that mAbs directed against various HIV-1 Results: Pseudoviruses with subtype A HIV-1 envelopes from epitopes and polyclonal IgG purified from the sera of HIV- early in infection demonstrate a broad range of neutralization infected patients have increased neutralizing activity when sensitivities to both autologous and heterologous plasmas. MDM or MDDC were the targets compared to CD4 T lym- However, neutralization by the monoclonal antibodies b12, phocytes. We showed that for MDM and MDDC, the inhibi- 2G12, 4E10 and 2F5 was generally poor. One virus was par- tion of HIV infection by Abs can occur by two distinct ticularly sensitive to both plasma and monoclonal antibodies. mechanisms, the one consisting of the neutralization of infec- Two amino acid changes in the transmembrane region were tivity (common for T lymphocytes, MDM and MDDC) and identified that conferred an increased sensitivity to neutral- the other based on the inhibition, via FcγR, of opsonized HIV ization. Interestingly, these changes altered recognition by particles certainly by endocytosis and degradation. Moreover, antibodies that recognize epitopes within the transmembrane some mAbs directed against epitopes to the principal immun- as well as within the surface unit of the envelope protein. odominant domain (PID) of gp41 or to V3-loop, which have Conclusions: The envelope variants identified here will be no detectable neutralizing activities on T lymphocytes, effi- useful as a subtype A virus panel for screening vaccine candi- ciently inhibit the infection of MDM or MDDC. For this dates for the ability to elicit relevant neutralizing antibody non-neutralizing inhibitory Abs (NNiAbs), the second mech- responses. Small changes in the transmembrane region of anism involving FcγR was implicated HIV-1 may have an important role in regulating neutraliza- Conclusions: We found that NNiAbs, that have no inhibito- tion sensitivity to antibodies directed to diverse epitopes. ry activity in the “conventional” assay analyzing the infec- tion of T lymphocytes, inhibit HIV infection on MDM and MDDC. We proposed that these NNiAbs could participate in vivo in the protection of mucosal HIV transmission.
AIDS Vaccine 2006 53 Oral Abstracts Sessions 14/8/06 16:40 Page 54
Amsterdam, Netherlands 29 August–1 September 2006
These results, demonstrating new categories of protective OA05-04 Abs, may open new perspectives in the development and design of vaccines. Mapping the neutralizing and non-neutralizing fractions of plasmas from HIV infected donors
OA05-03 JM Binley 1, PL Moore 1,2, ET Crooks1, J Robinson3, M Franti 4, L Morris 2, D Richman5 and D Burton6 A human anti-V2/V3 monoclonal antibody efficiently neutralizes SF162 by targeting the 1 Torrey Pines Institute for Molecular Studies, San Diego, CA, unliganded envelope USA, 2 National Institute for Communicable Diseases, Johannesburg, South Africa, 3 Tulane University Medical Center, 1 1 2 1 2 MK Gorny , T Kimura , X-H Wang , B Volsky , K Revesz , New Orleans, LA, USA, 4 Progenics Pharmaceuticals, Inc., 1 1,2 C Williams and S Zolla-Pazner Tarrytown, NY, USA, 5 University of California, San Diego, CA, USA, 6 The Scripps Research Institute, La Jolla, CA, USA 1 New York University School of Medicine; 2 Veterans Affairs Medical Center, New York, NY, USA Background: Considering the modest progress in formulating vaccines to elicit neutralizing antibodies (nAbs), it is becom- Background: A human monoclonal antibody (mAb), 2909, ing clear that we need to understand the relationship between which was selected on the basis of its neutralizing activity, antigen conformation and the repertoire of Ab specificities recognizes a complex epitope containing elements of V2 and they induce. Accordingly, researchers are developing new V3 that are present exclusively on the surface of intact viri- methods to map polyclonal antisera. ons but not on the soluble viral glycoproteins (Gorny et al., Objectives: To map the patterns of neutralizing and non-neu- J Virol 2005; 79:5232). tralizing Abs directed to Envelope proteins (Env) in human Objectives: To test the hypothesis that mAb 2909 mediates HIV+ patient plasmas. neutralization by binding to the V2/V3 epitope in the unli- Methods: To investigate a panel of neutralizing and non- ganded conformation of the envelope. neutralizing human plasmas, we developed a variety of Methods: Viral stocks of two pseudoviruses were produced mapping techniques, based on the JR-FL virus prototype, by cotransfection of HEK293 cells with pNL4-3.Luc.R-E- including: 1) Modified neutralization assays; 2) Virus cap- and with plasmids carrying either the wildtype SF162 Env or ture competition assays in which a plasma is titrated a mutated form of the SF162 Env (SF162mut). The against virus to determine its ability to inhibit capture by SF162mut Env expression vector was constructed by replac- monoclonal antibodies (mAbs); 3) Peptide binding and ing nine residues in the N-terminal heptad repeat region of neutralization interference assays. gp41 with alanine residues using a Quik Change mutagenesis Results: HIV+ plasmas were split into those with neutralizing kit. Intact virions were subjected to PAGE analysis under IC50s greater or less than 1:100. In modified neutralization non-reducing conditions followed by Western blot. The bind- assays, initial evaluation of mAbs indicated that V3 loop and ing of human mAbs to intact virions was determined in a cap- CD4-induced (CD4i) mAbs neutralized effectively post-CD4, ture assay and by immunoprecipitation. and mAbs against the gp41 membrane proximal region Results: Binding of 2909 to the wildtype SF162 pseudovirus (MPR) neutralized post-CD4/CCR5. HIV+ plasmas all neu- expressing trimeric Env spikes, but not to the SF162mut tralized post-CD4, but not post-CD4/CCR5, indicating a pseudovirus expressing monomeric Env spikes, indicates that presence of CD4i and/or V3 Abs and an absence of MPR Abs. 2909 is specific for trimers. The data are consistent with the Virus capture competition indicated that all plasmas con- recognition of a complex V2/V3 epitope which, according to tained high levels of Abs against the CD4i and gp41 immun- the model of Chen et al. (Nature 2005; 433: 834), can only odominant loop epitopes. Competition against b12, a be formed by the V2 and V3 regions from adjacent neutralizing mAb overlapping the CD4 binding site (CD4bs) monomers in the unliganded conformation. This model sug- was sporadic and did not consistently correlate with neutral- gests that the ability of sCD4 to inhibit 2909 binding to ization. Competition against non-neutralizing CD4bs mAbs SF162 virions is caused by disruption of the 2909 epitope as was more effective, but again did not correlate with neutral- the result of the CD4-induced movement of the V3 loop. ization. Competition against 2G12 was surprisingly effective Thus, the mechanism of 2909 neutralization appears to be in many cases, but also did not correlate with neutralization. based on locking the V2 and V3 loops in their positions in the Intriguingly, competition for V3 loop capture was more effec- unliganded state and thus blocking the formation and expo- tive in neutralizing plasmas. However, a lack of appreciable sure of the CCR5 binding site. JR-FL neutralization by V3-directed mAbs and peptide Conclusions: MAb 2909 appears to neutralize SF162 by pre- ELISAs both suggest that neutralization does not derive from venting the movement of the V2 and V3 loops. Locking the V3-specific Abs. Instead, plasma nAbs may be sensitive to V3 envelope in the unliganded conformation would block the loop conformation. formation and/or exposure of the CCR5 binding site on the Conclusions: Presently, it appears that neutralization traces to virus surface and would prevent infection. Abs overlapping the b12 epitope and/or Abs that are sensitive to the V3 loop. We are modifying assays to investigate further
54 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 55
Amsterdam, Netherlands 29 August–1 September 2006
and characterizing animal vaccine sera for comparison with OA05-06 Ab specificities in infections. Supported by NIH grants AI49566, AI58763 and The AIDS Development and characterization of HIV-1 epi- and Infectious Disease Science Center of The Torrey Pines tope-specific reagents for the detection of antigen Institute for Molecular Studies specific B cells
MA Moody1,2,4, LK Verkoczy 1,3,4, SM Alam1,3,4 and OA05-05 BF Haynes1,3,4 Llama heavy-chain antibody fragments that 1 Duke Human Vaccine Institute; 2 Department of Pediatrics; neutralize HIV-1 3 Department of Medicine; 4 the Center for HIV/AIDS Vaccine Immunology, Duke University Medical Center, Durham, NC, USA R Weiss1, A Forsman1, W Koh1, M Aasa-Chapman1, E Beirnaert2, V Tack2, Á McKnight1 and H de Haard2 Background: The origins of B cells that produce broadly neu- tralizing antibodies are unknown. Identification of these B 1 Division of Infection and Immunity, University College London, cell precursors is an important step in the design of vaccine London, UK; 2 Ablynx NV, Ghent, Belgium candidates that elicit broad neutralization. Objectives: To develop reagents to identify and characterize Background: Members of the Camelidae family produce HIV-1 antigen-specific B cells using the reactivity of surface immunoglobulins devoid of light chains. We have character- immunoglobulin in the B cell receptor and to develop meth- ized the variable domains of these heavy-chain antibodies, ods to reliably characterize those reagents. called VHH or Nanobodies™, from llamas immunised with Methods: A peptide specific for the broadly reactive human HIV-1 envelope proteins in order to identify neutralizing neutralizing antibody 2F5 was synthesized and coupled to VHH that block HIV entry into cells. biotin. Similarly, a HIV-1 Env V3 loop biotinylated peptide Objectives: To identify a set of HIV-neutralizing VHH as was prepared as were versions of each containing scrambled tools for vaccine design. amino acid sequences as controls. Peptides were reacted with Methods: Two llamas were immunised with monomeric sur- fluorochrome-labelled streptavidin to create B cell tetramers. face glycoprotein gp120 from a Chinese HIV-1 subtype B/C Polystyrene beads were coated with 2F5 and anti-V3 Mabs chimeric isolate, CN54. An additional llama was immunised specific for the tetramers as well as a control immunoglobu- with oligomeric glycoprotein gp140 from the same HIV-1 lin, P3X63. Murine and human hybridoma cell lines making isolate. Upon screening against envelope proteins from dif- anti-2F5 and V3 epitope antibodies, respectively, were used ferent HIV-1 clades, VHH targeting either the CD4 receptor in flow cytometry for specificity analysis of tetramer versus. binding site of gp120 or gp41 epitopes were characterized in scrambled tetramer binding. binding, competition and neutralization studies. Results: Tetramers were created using various fluorochromes. Results: Potent VHH were identified with 90% inhibitory All showed highly specific binding to beads coated with the concentration (IC90) values in the range of 0.012–25 µg/ml corresponding antibody and no binding to beads coated with depending on the HIV-1 isolate tested. Different VHH exhib- other antibodies. The hybridoma cell lines expressed it variable neutralization of clade A, B and C isolates and detectable surface immunoglobulin and bound to tetramers envelope pseudoviruses of HIV-1; some VHH show cross- with high specificity. Molar excess of unlabeled tetramer clade neutralization between clade C and clade B. The VHH completely inhibited the binding of tetramer to tetramer+ B selected to recognize the CD4-binding site bind on gp120 cells. Comparison of 2F5 peptide monomer versus. tetramer were characterized by ELISA and in BIAcore assays and com- showed a markedly increased apparent affinity of the petition for binding to gp120 with sCD4. The VHH selected tetramer as seen by a slowing of the off-rate in surface plas- to bind to gp41 by ELISA were characterized for neutralizing mon resonance assays. Mixing experiments using tetramer+ properties and for competition with mab 2F5. and tetramer- B cell lines showed the ability of the tetramers Conclusion: We have identified VHH that neutralize HIV-1 to easily identify <0.1% of surface immunoglobulin-positive subtype A, B and C viruses by blocking the CD4-binding site. antigen-specific cells by flow cytometry. Experiments using The results indicate that VHH heavy-chain antibody frag- Mab-coated beads in combination with B cell receptor+ cells ments have a potential use for immunogen selection in vac- suggest that the Mab-labelled beads can be used as an inter- cine research. Since VHH are stable and can be produced at nal standard for tetramer binding. a relatively low cost, they could also be considered for micro- Conclusions: Antigen-specific B cell tetramer reagents specif- bicide development. ic for HIV-1 epitopes will be useful for the identification and characterization of anti-HIV-1 B cells that are capable of making anti-2F5 and V3 antibodies cells from normal and HIV-1 infected patients.
AIDS Vaccine 2006 55 Oral Abstracts Sessions 14/8/06 16:40 Page 56
Amsterdam, Netherlands 29 August–1 September 2006
OA05-07
Detection of HIV-1 variable loop-specific neu- tralizing antibody responses by HIV-2/HIV-1 envelope chimeras
KL Davis1, F Bibollet-Ruche1, H Li1, O Kutsch1, BH Hahn1, JA Hoxie 2, PD Kwong3, and GM Shaw1,4
1 University of Alabama at Birmingham, Birmingham, AL, USA; 2 University of Pennsylvania, Philadelphia, PA, USA; 3 Vaccine Research Center/NIH, Bethesda, MD, USA; and 4 Howard Hughes Medical Institute, Birmingham, AL, USA
Objective: To detect HIV-1 variable loop-specific neutralizing antibody (Nab) responses in sera from infected or vaccinated subjects as a novel strategy for discerning correlates of immune protection. Background: Previously, we used HIV-2 as a molecular scaffold on which to display HIV-1 CD4-induced (CD4i) or membrane proximal external region (MPER) Env epitopes in order to iden- tify and quantify the respective epitope-specific Nab responses (JEM 2005; 201:1407; CROI 2006; abstract 113). Here, we test the feasibility of this approach for detecting HIV-1 variable loop-specific Nabs. Methods: Full-length provirus and env genes were cloned from HIV-2 KR and HIV-2 KR/MN V3 infected Molt 4/8 cells, kindly provided by D. Looney. Recombinant viruses were sequenced and characterized for infectivity, replication competence, and susceptibility to neutralization by Env-spe- cific ligands and antibodies in a single-round infectivity assay (Nature 2003; 422:307). Results: Forty-two HIV-2 KR proviral clones, 42 HIV-2 KR env clones, and 20 HIV-2 KR/MN V3 env clones were character- ized. Examples of each were infectious. Sequence analysis of bulk PCR amplification products and individual env clones revealed four adaptive mutations in HIV-2 KR/MN V3 env that facilitated efficient viral replicationn of the HIV-2/HIV-1 V3 Env chimera. These included: 1) a glutamic acid to lysine change in the bridging sheet of the HIV-2 scaffold (E203K); 2) an alanine to threonine substitution in the MN V3 loop (A329T); 3) a thre- onine to alanine change that eliminated a potential N-linked gly- can near the CD4 binding region (T358A); and 4) a tyrosine to histidine substitution in the MPER (Y681H). HIV-2 KR/MN V3 Env chimeras were sensitive to neutralization by soluble CD4 (IC50=20 nM), the V3 mAb 447–52D (IC50 <0.05 µg/ml), and HIV-1 patient plasma (IC50=0.0001–0.05) but were resistant to neutralization by the non-V3 specific mAbs b12, 2F5, 17b, 19e, 21c, as well as normal human plasma. Conclusions: This study describes an HIV-2 Env scaffolding strategy for presenting autologous and heterologous HIV-1 V3 sequences as a means to evaluate the breadth and poten- cy of HIV-1 V3-specific Nab responses, thereby providing a new approach for assessing correlates of immune protection. The study also reveals adaptive mutations in the HIV-2 Env scaffold associated with enhanced infectivity of HIV-2/HIV- 1 Env chimeras important for structure-function analyses of the glycoprotein.
56 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 57
Amsterdam, Netherlands 29 August–1 September 2006
TOPIC 6: CLINICAL TRIAL SITE DEVELOPMENT/SOCIAL ISSUES/ETHICAL ISSUES/ ACCESS ISSUES/REGULATORY ISSUES
OA06-01 and rank the PBMC sample CMV and FEC responses. Comparison of pre- versus operator-coated plates showed ELISPOT standardization for laboratories conduct- lower IFN-α responses using pre-coated plates. No difference ing immunogenicity testing in IAVI sponsored HIV was seen between operators within or across sites when com- paring FEC and CMV responses. vaccine trials Conclusion: This proficiency panel demonstrates the ability of six laboratories, located across three continents, to process 2 2 2 2 3 P Hayes , D Gill , T Tarragona , L Seamons , J Birungi , PBMC samples and to rank volunteers with differential mag- 3 3 4 4 4 PK Kato , A Ssemaganda , O Anzala , B Farah , S Ogola , nitudes of IFN-α ELISPOT responses concordantly. This 4 5 5 6 J Indangasi , P Mhlanga , N Ndzamela , M Thakar , panel illustrates the ability to standardize the IFN-α A Pujari 6, B Purandare 6, S Mishra6, N Goonetilleke7, ELISPOT assay across multiple laboratories when common S Moore7, Abdul Mahmoud7, L Dally 8, G Stevens1, training, reagents and SOPs are adopted. These results are MJ Boaz1 and J Gilmour 1 encouraging for the HIV vaccine field in general.
1 IAVI, New York, US; 2 IAVI Core Laboratory, London, UK; 3 Uganda Virus Research Institute, Entebbe, Uganda; 4 Kenya OA06-02 AIDS Vaccine Initiative, Narobi, Kenya; 5 Contract Laboratory Services, Johannesburg, South Africa, 6 National AIDS Research Negative social impacts in preventive HIV vaccine Institute, Pune, India; 7 Centre for Clinical Vaccinology and clinical trials Tropical Medicine Oxford University, Oxford, UK and 8 EMMES Corporation, Rockville, US MA Allen1, B Metch 2, C Lau1, H Israel 3, K Rybczyk4, R Holt2 and M Keefer 5 Introduction: The IFN-α Elispot assay is used routinely to evaluate potency of candidate HIV vaccines. In order to com- 1 National Institute of Allergy and Infectious Diseases, Bethesda, pare candidates, and pool data across multiple trial sites, val- MD, USA; 2 Fred Hutchinson Cancer Research Center, Seattle, idated standardized methods must be applied across WA, USA; 3 Saint Louis University School of Medicine, St. Louis, laboratories. In addition, a proportion of all samples should MO, USA; 4 Vanderbilt University School of Medicine, Nashville, be sent to one central core laboratory for independent QA TN, USA; and 5 University of Rochester School of Medicine, testing. To standardize across sites participating in IAVI tri- als, the core laboratory provides SOPs, training and stan- Rochester, NY, USA dardized quality assured (QA) reagents. Furthermore, IAVI-sponsored vaccine trial laboratories participate in profi- Background: Participants in preventive HIV vaccine trials ciency panels as part of a comprehensive EQA program - key may experience negative social impact (NSI) of study partici- to monitoring assay performance plus identifying and resolv- pation including stigmatization and problems due to vaccine- ing issues promptly. induced HIV antibody, yet few published reports analyse data Method: Six IAVI-sponsored trial sites participated in the on this issue. panel. Each team had been previously trained at the IAVI core Objectives: To analyse NSI events, and identify if reporting laboratory London using standardized SOPs. At each labora- NSI is associated with participant gender, race, age, HIV risk tory, two operators independently processed identical panels behavior, sexual orientation, time on study, and study site. containing frozen PBMC samples from three different donors Methods: From December 2000 through November 2005, using four blinded stimuli. PBMC recovery, viability after 1,599 HIV Vaccine Trials Network study participants from overnight rest and IFN-α ELISPOT assay performance was 13 US and 7 non-US sites underwent standardized scheduled assessed, including the use of pre-coated and operator-coated assessments for NSI of study participation. IFN-α plates. Analyses of the IFN-α ELISPOT assay included Results: 153 (9.6%) trial participants (8.4% of US and comparison of responses observed between operators within 16.2% of non-US volunteers) reported 187 NSI events. Most and across sites with regard to overall assay performance, use commonly reported events (n=148, 79%) were negative reac- of pre- versus operator-coated IFN-α plates, and ability to tions of family, friends, and co-workers to the volunteer’s rank differential responders to CMV alone or a combination study participation. Less than 10% of events pertained to of flu, EBV and CMV (FEC) peptides. employment (n=17), medical/dental care (n=8), life insurance Results: All sites demonstrated good performance in PBMC (n=3), or housing (n=2). HIV testing was performed in only thawing and resting: recoveries (median 96%, 60–154%) and 4 events. No participants reported problems with travel, viabilities (median 96%, 80–100%). IFN-α ELISPOT results immigration, education, health insurance, or military service. demonstrated the ability of each site to correctly distinguish Most events were considered by participants as resolved
AIDS Vaccine 2006 57 Oral Abstracts Sessions 14/8/06 16:40 Page 58
Amsterdam, Netherlands 29 August–1 September 2006
(81%), with minimal impact on quality of life (QOL) (74%). with demographics, knowledge of the Tuskegee Study, geno- Of the 19 events with major impact on QOL, 14 were from cidal beliefs, negative perceptions of health care, altruism, non-US sites. Multivariate logistic regression analysis of US perceived/actual risk of HIV, motivation to decrease HIV participants revealed that younger age (P=0.001), study site risk, network support for trial participation and ambiguity (P=0.004), and longer time on study (P<0.001) were statisti- tolerance was assessed separately for heterosexuals and MSM cally significant predictors of reporting NSI. Gender, race, using ordinal logistic regression. sexual orientation, and HIV risk behavior were not associat- Results: Overall, 55% of participants expressed some WTP ed. At one non-US site younger age was associated with in trials. For heterosexuals, factors related to WTP were: reporting NSI (P=0.04) and at another non-US site women altruism (P<0.001), to learn about reducing HIV risk were more likely than men to report NSI events (P=0.04). (P<0.001), perception of risk if in a trial (P<0.001), HIV risk Conclusions: The results of this analysis are reassuring and behavior (P<0.001), ambiguity tolerance (P<0.001), per- demonstrate the utility of standardized assessments in NSI ceived support of sex partners for trial participation data collection and analysis. Informing potential study par- (P<0.001), and knowing someone with AIDS (P<0.01). For ticipants about the risk of NSI, and providing participants MSM, perceived support of sex partners for trial participa- with assistance to prevent or resolve events may be important tion (P<0.0001), altruism (P<0.001), reduced perception of factors in the relatively low incidence of problems reported risk if enrolled in a trial (P=0.02), and HIV risk behavior and few reports of major events. The finding that younger participants in some settings are more likely to report events OA06-03: Table 1. warrants further investigation. WTP in Trials (%) Not at all Might be Probably Definitely Women OA06-03 High Risk 36 32 22 10 Low Risk 53 26 16 5 The willingness of black heterosexuals and men Heterosexual Men who have sex with men (MSM) in three US cities High Risk 46 23 24 7 to participate in HIV vaccine trials Low Risk 48 28 17 7 MSM BN Bartholow 1, J Kibler 2, AT Young3, C Dillard-Smith4, High Risk 36 36 20 8 M Durham1, M Ma5, S Washington3 and G Lockett 4 Low Risk 47 30 16 7
1 Centers for Disease Control and Prevention, Atlanta GA, USA; 2 Nova Southeastern University, Fort Lauderdale, FL, USA; 3 Community Education Group, Washington, DC, USA; ( 58 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 59 Amsterdam, Netherlands 29 August–1 September 2006 Background: In conducting HIV vaccine trials, reliable Background: Accurate assessment of sexual risk behavior is mechanisms for effective recruitment and retention of trial necessary both for initial screening of volunteers wanting to volunteers are necessary. A protocol for participant enroll- participate in HIV vaccine trials as well as to ensure behavioral ment was established in 2002. After receiving voluntary test- disinhibition does not occur for those eventually enrolled into ing and counseling (VCT), HIV negative clients enter a clinical trials. “Prescreening Protocol” which includes vaccine discussion Objectives: To assess risk behavior in a heterosexual cohort groups (VDGs) and a physical exam, a behavioural risk undergoing prescreening for early Phase I/II HIV vaccine trials. assessment and an assessment of understanding. The dura- Methods: A survey collecting self-reported data in a cultural- tion of this prescreening process has decreased from 8 to 4 ly appropriate manner was developed for use during pre- weeks since the start of the program. An analysis of the suc- screening for vaccine trials in Soweto. HIV-negative potential cess of the long (8 week) and short (4 week) prescreening volunteers were recruited from VCT into a prescreening pro- program was done to inform future decisions for enrollment tocol which included assessment of sexual risk. This analysis into Phase I/II HIV vaccine trials. represents data from risk behavior surveys administered at Methods: Analysis of the enrollment process of participants the initial clinic visit. into the Prescreening protocol was undertaken in an effort to Results: Of 488 participants (mean age 26.8 years), most maximize trial cohort populations. Data from VCT recruit- (89%) were single. Only 24% were employed and 51% of ment, VDG sessions, for both 8-session and 4-session proto- households had a per capita income below poverty level. cols, and prescreening processes were examined to determine Males reported higher rates of any incidence of heavy (>5 the efficiency and significant exclusion factors at each stage. drinks/day) alcohol (62% versus. 33%; P<0.001), marijuana Results: The prescreening program was 35.1% efficient in (19% versus. 3%, P<0.001) and other recreational drug use placing HIV negative participants into VDG’s. Primary exclu- (4% versus. 0.4%, P<0.01). Males reported more sex part- sion factors were level of education (32.6%), current employ- ners than females in the previous 6 months (P<0.001), with ment (31.9%), and student status (10.5%); these factors 22% of men and 2% of women reporting >2 sex partners. Of together comprise 75% of exclusions. Examination of the those sexually active in the previous 6 months (N=408), VDG 8-session and 4-session protocols (n=254, 364 respec- males reported significantly more casual/anonymous partners tively) indicated the 4-session process was more efficient (29% versus. 8%, P<0.001) and one-night stands (14% ver- (P=P=0.00355), with primary exclusion factors of loss to fol- sus. 1%, P<0.001). Multivariate analyses for higher-risk out- low up (22.0%, 67.8% respectively), loss of interest (24.2%, comes in the previous six months revealed: <100% condom 2.0%), employment (25.8%, 61.6%) and level of education use with known/suspected HIV-positive partners was not pre- (0.0%, 13.7%). The prescreening process was 66.2% dicted by any covariates tested; having casual/anonymous (n=455) efficient for a cohort of participants being ready for partners was predicted by male gender [OR=3.7; 95% CI entry into Phase I/II HIV vaccine trials. M were investigator (2.0, 6.7)] and heavy alcohol use [OR=2.7; 95% CI (1.4, decision (45.16%) based on preexisting medical conditions 4.9)]; and, having >2 partners was also predicted by male or inaccessible veins, loss to follow up (23.87%), and other gender [OR=8.3; 95% CI (3.4, 20.4)] and heavy alcohol use (21.94%) including participants or family decisions. [OR=3.3; 95% CI (1.6, 7.0)]. Discussion: The prescreening process is a good model for effi- Conclusions: This represents the first characterization of the cient recruitment for HIV vaccine trials in Southern Africa. sexual risk behavior of a cohort undergoing prescreening for This analysis illustrates the benefit of continuous program- Phase I/II HIV vaccine trials in South Africa. For this popu- matic evaluation, and highlights the need for public educa- lation, male volunteers may need increased risk reduction tion and flexible clinic hours, accommodating those counseling during Phase I/II trials and additional recruitment employed or in school. methods may be necessary to identify high risk female volun- teers for Phase III efficacy trials. OA06-05 OA06-06 Self-reported sexual risk behaviors of a cohort being prepared for multiple Phase I/II HIV vaccine Recruiting HIV discordant couples for vaccine effi- trials in Soweto, South Africa cacy trials through couples voluntary counseling and testing centers in Lusaka, Zambia K Mesesan1, RM Van Niekirk2, LM Niccolai1, 2 1 2,3 1,2 ON Mlungwana2, IM Holdsworth2, M Bogoshi2, C Vwalika , A Tichacek , E Chomba , O Manigart , 2,3 4 4 1 JA McIntyre 2, GE Gray 2 and E Vardas 2 S Lakhi , M Krebs , P Fast , S Allen and the Rwanda Zambia HIV Research Group 1 Yale University School of Medicine, Department of Epidemiology & Public Health, New Haven, CT, United States; 1 Emory University, Atlanta, GA, USA; 2 Zambia Emory HIV and 2 Perinatal HIV Research Unit, University of the Research Project, Lusaka, Zambia; 3 University Teaching Hospital, Witwatersrand, Johannesburg, South Africa Lusaka, Zambia; 4 International AIDS Vaccine Initiative, New York, NY, USA AIDS Vaccine 2006 59 Oral Abstracts Sessions 14/8/06 16:40 Page 60 Amsterdam, Netherlands 29 August–1 September 2006 Background: Cohabiting couples are the largest HIV risk to the HIV/AIDS vaccine field. group in Africa, and couples voluntary counseling and testing Methods: We reviewed the current state of the HPV trials and (CVCT) reduces transmission within couples by 60%. HIV discussed key issues with clinicians, researchers and commu- discordant couples retain relatively high rates of seroconver- nity advocates directly involved to draw lessons for the HIV sion despite counseling. HIV vaccines may reduce transmis- vaccine field. sion further, and large cohorts of high risk individuals are Results: Although there are significant differences between needed for efficacy trials. HPV and HIV vaccines, HPV vaccine development neverthe- Methods: In Lusaka, cohorts of discordant couples have less has the potential to dispel some prevailing assumptions been recruited by the Zambia Emory HIV Research Project about HIV/AIDS vaccine development and eventual delivery. since 1995. Recruitment and testing efforts scaled up in For example, the HIV/AIDS vaccine field has long assumed 2002, when HIV vaccine and other clinical trials required that its candidate vaccines could not practically be tested in expansion of discordant couple cohorts. Outreach is a mix- adolescents before showing efficacy in adults. HPV vaccines ture of word-of-mouth and invitation by trained outreach were tested in trials with girls as young as ten and surveys of workers. After testing, interested discordant couples who parents indicate the potential for widespread acceptance have cohabited at least three months are enrolled for fol- should a vaccine be licensed. Another assumption is that a low up. The HIV negative partner is followed quarterly, partially effective HIV vaccine, like the HPV vaccine which and couples receive reproductive outpatient care. Couples targets strains that account for only 70% of cervical cancers, are recruited from established cohorts for HIV prevention will not be accepted, or will interfere with prevention mes- trials, including vaccine trials. sages. Careful study of HPV vaccine roll-out will provide Results: From 2002–2005, over 118,000 CVCT invitations clues about how to develop vaccination programs that rein- were distributed by Influence Network Agents and 2082 dis- force prevention and testing messages. Finally, the introduc- cordant couples (909 with HIV+ men, 1173 with HIV+ tion of an HPV vaccine will provide invaluable insight into women) were identified. Of those, 1620 (78%) met eligibili- the barriers and opportunities for swiftly ensuring access to ty criteria for enrollment and 1130 (70%) of those were new vaccines in resource-poor settings around the world. enrolled in the open cohort. In December 2005, 860 couples Conclusions: The experience of HPV vaccine development were in active follow up, including 208 couples enrolled and (assuming licensure) delivery, will provide critical before 2002. Retention of couples enrolled with follow up is insights for the HIV/AIDS vaccine field. Historically, it has 80% at 12 months. 147 seroconvertors (84 women and 63 been the case that years and even decades have elapsed from men) have been identified (HIV incidence rate 8.3/100 py, access in the developed to access in the developing world. 95% CI 7.4–9.3). In 2006, participants on ARV were Cervical cancer rates are highest, and arguably the need for released from the cohort to consolidate health care at the dis- cervical cancer prevention is greatest, in the developing trict clinics and to focus cohort retention on couples eligible world. We propose a research and action agenda to ensure to participate in vaccine efficacy trials. Screening and enroll- that the HIV/AIDS vaccine field learns from the lessons of ment for a Phase II HIV vaccine trial will begin in April 2006. this closely-related field. Conclusions: It is possible to recruit and retain sufficient numbers of HIV discordant couples for large scale clinical tri- als. This work will require substantial expansion of CVCT services, and the cohort maintenance run-in design ensures excellent retention in the clinical trial. Eligibility criteria at CVCT and during cohort maintenance can be tailored to meet the needs of specific clinical trials. OA06-07 Clearing the path: why the best news for HIV vaccines may be a cancer vaccine K Fisher, E Bass, E Lee and M Warren AIDS Vaccine Advocacy Coalition, New York, NY, USA Objectives: Lack of a clear scientific pathway is now the chief stumbling block to a HIV vaccine. Quick and univer- sal access to an HIV vaccine will require a clear path to licensure and access once a scientific pathway is identified. The experience of the human papillomavirus virus (HPV) vaccine submitted for licensure by Merck and GSK in the US and Europe, will provide insights of utmost importance 60 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 61 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 7: IMMUNE ESCAPE OA07-01 OA07-02 Late escape from an HLA-B27-restricted CTL Resistance profile of HIV-1 to the broadly neutral- epitope in HIV-1 p24/capsid is associated with a izing anti-gp120 2G12 antibody dramatic reduction in replicative capacity D Schols, D Huskens, K van Laethem, K Vermeire and A Schneidewind1, MA Brockman1, RI Adam1, C Brander1, J Balzarini A Kelleher 2, BD Walker1, and TM Allen1 Rega Institute for Medical Research, Katholieke Universiteit 1 Partners AIDS Research Center, Massachusetts General Leuven, Leuven, Belgium Hospital, Boston USA; 2 National Center in HIV Epidemiology and Clinical Research, Sydney, Australia Background: The mAb 2G12, a carbohydrate binding mAb that specifically interacts with gp120, is described as a potent Background: HLA-B27 is associated with control of HIV-1 human mAb with inhibitory activity against a broad array of infection. The immunodominant B27-restricted CTL epitope HIV-1 strains. KK10 (KRWIILGLNK ) is located in p24/capsid. Viral Objectives: To evaluate the effect of 2G12 mAb on various 263-272 escape in this epitope consists primarily of an early L268M HIV-1 isolates in different cell-types and to investigate mutation at position 6 (P6), followed by an R264K anchor sub- in vitro resistance and viral escape mechanisms to 2G12 mAb. stitution at P2 that is typically observed many years after infec- Methods: MT-4 cells, CEM cells and PBMCs were infected tion. An unpublished putative compensatory mutation S173A is with HIV-1 in the presence of 2G12 mAb. The T cell lines temporally associated with R264K. We hypothesized that the were incubated until a cytopathic effect was observed, in protective effect of HLA-B27 may be related to the selection of PBMCs p24 viral Ag production was determined. For the escape mutations in KK10 that result in reduced viral fitness. resistance studies, HIV-1 NL4.3 and HIV-IIIB were cultured Aims: 1). To examine the impact of CTL escape mutations in in the presence of increasing concentrations of 2G12 mAb. B27-KK10 on the replicative capacity of HIV-1; 2). To assess Then, proviral DNA or viral RNA was extracted using the the ability of compensatory mutations to rescue fitness QIAamp blood mini kit or the QIAamp viral RNA mini kit. defects; and 3). To explore mechanisms for the late escape in Sequencing was performed using the BigDye terminator this immune response. v3.1 cycle sequencing kit and the reactions were run on an Methods: A panel of escape and putative compensatory ABI3100 genetic analyser. The resistant virus strains were mutations was constructed by site-directed mutagenesis of evaluated for their sensitivity to various other classes of HIV-1 strain NL4-3. Viral stocks were prepared by transfec- entry inhibitors. tion of 293T cells and soluble p24 was quantified using Results: 2G12 mAb inhibits the infection of HIV-1 strains (i.e. IIIB and NL4.3) at IC ranging from 0.02–0.2 g/ml. ELISA. Viral infectivity and replication was monitored by 50 µ flow cytometry using a GFP-reporter T cell line. However, isolates from various HIV-1 subtypes (i.e. clade C, D, A/E, F, O) were not inhibited by 2G12 mAb (IC 20 Results: The early L268M mutation had no effect on viral 50 > replication and appears to represent a partial CTL escape µg/ml), as well as several other clade B variants (such as mutation. In contrast, the late R264K mutation, alone or in MN and RF). 2G12 mAb pressure in HIV-1 IIIB- and combination with L268M, dramatically reduced viral infec- NL4.3-infected T cell cultures rapidly selected for resistant tivity to 4% of WT, and both viruses failed to spread in 7-day viruses containing 1 to 3 N-glycosylation site deletions in in vitro cultures. Inclusion of the upstream S173A mutation gp120. Especially the amino acids N295 and N392 proved restored R264K/L268M viral replication and infectivity to crucial for resistance development. The 2G12-resistant IIIB 88% of WT. Interestingly, less common escape mutations in virus strains keep their full sensitivity to various mannose- KK10 (R264G, R264T, or R264Q) displayed intermediate specific lectins and other entry inhibitors (such as replication phenotypes. AMD3100, DS-5000 and T-20). Moreover, we observed Conclusion: A dramatic fitness cost is associated with the KK10 that the NL4.3-2G12-resistant virus, containing the N295K escape mutation R264K on helix 7 of capsid. Replication is res- mutation, became significantly more sensitive to mannose- cued by an S173A substitution, which is structurally adjacent to specific lectins. R264 but is located on helix 2. The necessity for concomitant Conclusions: 2G12 mAb easily selects for resistant HIV-1 mutations for HIV-1 to remain replication competent may strains containing glycan deletions in gp120. This is, to our explain the late kinetics of escape from the KK10 CTL knowledge, the first report showing that a resistant virus gen- response. The dramatic defects seen for variants encoding muta- erated against a neutralizing mAb, has increased sensitivity to tions in this region indicate that capsid stability may be altered, another class of HIV entry inhibitors. leading to suboptimal viral entry. This study was supported by NIAID grants AI054178 and AI067078 AIDS Vaccine 2006 61 Oral Abstracts Sessions 14/8/06 16:40 Page 62 Amsterdam, Netherlands 29 August–1 September 2006 OA07-03 OA07-04 Capture and transfer of antibody neutralized HIV- Reversion of HLA-B57-restricted CTL epitope 1 to CD4 lymphocytes by dendritic cells and DC- escape mutations in HIV-infected subjects in SIGN expressing cells Durban, South Africa T van Montfort1, AA Nabatov 1, TBH Geijtenbeek 2, H Crawford1, C Thobakgale 2, S Reddy 2, W Mphatswe2, G Pollakis 1 and WA Paxton1 A Leslie1, I Honeyborne1, J Prado1, JI Mullins 3, C Brander 4, P Kiepiela2, BD Walker 4 and PJR Goulder1, 4 1 University of Amsterdam, Amsterdam,the Netherlands; 2 Vrije Universiteit Medical Center, Amsterdam, the Netherlands 1 University of Oxford, Oxford, UK; 2 University of KwaZulu- Natal, Durban, South Africa; 3 University of Washington School Background: Neutralizing antibodies (nAbs) are considered of Medicine, Seattle, USA; 4 Massachusetts General Hospital, to have an important role in controlling progression to AIDS Boston, USA in HIV-1 infected individuals and will undoubtedly be required to be induced in effective prophylactic as well as Background: HLA-B57 is the HLA class I molecule most therapeutic vaccines. strongly associated with immune control of HIV, despite Objectives and Methods: We aimed to identify whether HIV- escape in multiple HLA-B57-restricted CTL epitopes. One 1 in the presence of nAbs could be captured by either den- mechanism proposed to explain this control is that escape dritic cells (DCs) or a cell line expressing DC-SIGN results in a significant cost to viral fitness. In vivo evidence (Raji-Dc-SIGN cells) and undergo subsequent transfer to shows the loss of these mutations in the absence of B57- CD4 lymphocytes. This enhancement could be either through associated immune pressure. interaction with C-type lectins, such as DC-SIGN, or via the Objectives: To determine the frequency and rate of rever- interaction of the immune-complexes with the Fc receptor on sion of B57-associated CTL escape mutation to wildtype in the surface of the cells. two p24 Gag epitopes, KF11 (Gag 162-172) and TW10 Results: We demonstrate that gp41 as well as gp120 directed (Gag 240-249). nAbs, which can efficiently block HIV-1 entry into CD4 lym- Methods: We sequenced the whole gag gene in a cohort of over phocytes, loose the capacity to block HIV-1 infection when HIV- 500 naive HIV-infected subjects from Durban, South Africa. 1 transfer occurs via either DCs or Raji-DC-SIGN cells. We Results: Sequence data from >200 HLA-B57-positive subjects demonstrate using GFP fluorescent labelled viruses that in the revealed the range of mutations (‘B57 footprints’) associated presence of gp41 directed antibodies (2F5 and 4E10) HIV-1 with these two p24 Gag epitopes. Over three hundred B57- binding to DCs and Raji-DC-SIGN cells can actually be negative subjects were studied, the majority of which showed enhanced over non-neutralized virus. With the Raji-DC-SIGN no B57-associated mutation. Two B57-negative subjects carry- cell and the Raji control cell line we demonstrate that this ing the B57 footprint were longitudinally studied to determine enhancement is mediated via the DC-SIGN receptor and not via the rate of reversion in the two epitopes. Reversion in TW10 Fc. The observed enhancement to transfer was shown to be (242N→T) was observed PS114-child 8 months after trans- restricted by the genotype/phenotype of the virus envelope with mission of the virus from a B57-positive mother. In the second R5/X4 and X4 viruses being preferentially enhanced over R5 subject, SK233, we observed reversion in KF11 (163G→A), as viruses. We further demonstrate through confocal microscopy well as in TW10 (242N→S), at 8 and 11 months respectively. that HIV-1 is internalized and the antibody is dissociated from In a high proportion of B57-positive subjects, A163G in KF11 the virus particles. The same phenomenon of virus transfer via is associated with another mutation, S165N, in the same epi- DCs or Raji-DC-SIGN has been shown with an array of anti- tope. However, despite reversion of 163G→A in B57-nega- bodies directed against the gp120 molecule. tives, 165N remains, suggesting that this is not an escape Conclusions: We demonstrate that neutralized HIV-1 can mutation per se but potentially a compensatory mutation. be captured by both DCs and Raji- DC-SIGN cells and suc- Conclusion: Reversion is most frequently observed in the previ- cessfully transferred to CD4 lymphocytes. This has impli- ously documented T242N mutation within the TW10 epitope. cations for viral escape from both naturally as well as Additional reversion was described for the first time within the vaccine induced neutralizing antibody responses with sig- KF11 epitope at A163G. A putative compensatory mutation in nificant consequences for HIV-1 transmission and dissemi- KF11 was observed. HLA-B57-positive subjects carrying Gag nation in the presence of such antibodies. We have also mutant virus appear to progress at the same rate as those infect- demonstrated an enhancement of HIV-1 binding to DC- ed with wildtype virus, suggesting that B57-restricted immune SIGN in the presence of gp41 directed nAbs pressure on p24 Gag has a dramatic effect on viral fitness. This is likely to contribute to the viral control observed in B57-posi- Funded partially under the European Union “EuroVac-II” tives. It is important to study the fate of escape mutations to (QLK2-CT-2001-01316). determine their relevance to vaccine design. 62 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 63 Amsterdam, Netherlands 29 August–1 September 2006 OA07-05 OA07-06 Rapid reversion of transmitted sequence polymor- Correlates of neutralization resistance in hetero- phisms dominates early HIV-1 evolution sexual transmission pairs infected with subtype C Human Immunodeficiency Virus type 1 B Li1, ADAD Gladden1, M Altfeld1, J Kaldor 2, D Cooper 2, TA Kelleher 2, BDBD Walker1 and TMTM Allen1 E Hunter1, CA Derdeyn1, R Rong1, G Gnanakaran2, JM Decker3,4, F Bibollet-Ruche 3,4, JL Mokili 2, 1 Partners AIDS Research Center, Massachusetts General M Muldoon5, J Mulenga 6, S Allen1, BH Hahn3,4, Hospital, Boston USA; 2 National Center in HIV Epidemiology GM Shaw 3,4, JL Blackwell1 and BT Korber 2 and Clinical Research, Sydney, Australia 1 Emory University, Atlanta, GA, USA; 2 Los Alamos National Background: The high replicating error rate of human Laboratory, Los Alamos, NM, USA; 3 Howard Hughes Medical immunodeficiency virus type 1 (HIV-1) generates mutations Institute, Birmingham, AL, USA; 4 University of Alabama at and sequence diversity at a nearly saturating rate. It is a cru- Birmingham, Birmingham, AL, USA; 5 University of Manchester cial challenge for vaccine development to define the stability Institute of Science and Technology, Manchester, UK; 6 Blood of different mutations and the pattern of HIV-1 evolution in Transfusion Service, Lusaka, Zambia different stages of infection. The importance of CTL-driven forward escape mutations in HIV-1 evolution is well docu- Background: Potent autologous antibody-mediated neutraliz- mented, but recent reports have begun to highlight reversions ing responses are produced against HIV-1, but are thwarted by of transmitted CTL escape mutations and their ability to successive escape variants that remain one step ahead of the reduce viral fitness. humoral response. Resistance to neutralization is achieved Aims: To identify the extent to which transmitted mutations through adaptations that occur in the envelope (Env) glyco- may be reducing early viral replication by characterizing the proteins as the viral quasispecies expands. We previously dynamics of reversions across HIV-1 genomes in early infection. reported that the HIV-1 quasispecies passes through a genetic Methods: We examined sequence polymorphisms arising bottleneck during heterosexual transmission in Zambia that across non-Env regions of the HIV-1 genome from samples selects for compact viruses with high sensitivity to neutraliza- derived longitudinally from seven subjects followed up to one tion by antibodies in the infecting partner’s plasma. This sug- year from primary infection. Two to five full genome ampli- gests a potential relationship between the topology of the cons derived from proviral DNA from each sample were newly transmitted Envs and their neutralization sensitivity. pooled and population sequenced. Objectives: We investigated whether the sequence diversity, Results: A large number of mutations (42%) represented length, or glycosylation of the gp120 V1-V4 region was cor- polymorphisms reverting towards the Clade B consensus related with neutralization sensitivity. sequence, and they occurred significantly faster than forward Methods: A statistical analysis was performed to determine mutations (P=0.007). Nearly one third of them were mapped whether sequence adaptations in the V1V4 region of gp120 onto defined CD8 epitopes that are not restricted by contem- were linked to neutralization sensitivity. V1V2 and α2 helix porary host HLA. Notably, these rapid reverting mutations chimeras were constructed to probe the role of these regions are primarily located within structurally conserved regions in neutralization phenotype. (P=0.005), strongly indicative of their role in restoring viral Results: Sequence divergence and acquisition of length in fitness. A recent publication examining viral evolution during V1V4 were correlated with a decrease in sensitivity to neu- early HIV-1 infection has illustrated that a large number of tralization. Exchange of the V1V2 domain between neutral- reverting mutations are also arising within Env (Herbeck et ization resistant donor Envs and a neutralization sensitive al, 2006), supporting these observations and the importance recipient Env confirmed that expanded V1V2 domains con- of viral fitness in driving HIV-1 evolution. fer neutralization resistance. A mutual information analysis Conclusions: Our data support that both adaptive and puri- revealed 9 amino acid positions in V1V4 that were signifi- fying selection pressures are driving early HIV-1 evolution cantly associated with neutralization phenotype. Five of these in non-Env regions primarily under cellular immune pres- 9 positions were located in a helical structure (α2) on the sures. Despite evading immune responses, the incurring fit- outer domain of gp120 that is under strong positive selection ness cost of some CD8 escape mutations may be in subtype C infection. However, exchange of the α2 helix contributing to the control of viral replication, and thus domain between neutralization-resistant donor Envs and a merit more consideration in vaccine design. neutralization-sensitive recipient Env did not alter the neu- This study was supported by NIAID grants AI054178 and tralization sensitivity phenotype. the AIEDRP. Conclusions: These results argue that the adaptations that facilitate resistance against neutralization in subtype C virus- es, such as increased V1V4 length, may diminish the capaci- ty to establish a new infection. Some sequence adaptations, particularly those in α2, appear to play a role in decreased sensitivity to neutralization but do not directly confer this AIDS Vaccine 2006 63 Oral Abstracts Sessions 14/8/06 16:40 Page 64 Amsterdam, Netherlands 29 August–1 September 2006 phenotype. The influence of the α2 helix on neutralization Conclusions: The enhancement of binding and fusion versus sensitivity suggests that subtype C viruses evade the humoral escape from NAbs are competitive forces on Env. Alterations immune response through a different mechanism than sub- to charge in variable regions and rearrangements of key type B viruses. PNGs proximal to the CD4 binding site may balance these competing pressures. OA07-07 N-linked glycosylation sites proximal to the CD4 binding site of SHIV89.6P Env affect both neutral- ization escape and viral fitness NL Haigwood1,2 W Blay 2, L Misher 1, T Kaspryzyk 2, S Gnanakaran3, B Foley 3, N Doria-Rose 2 and B Korber 3 1 Seattle Biomedical Research Institute, Seattle, WA, USA; 2 University of Washington, Seattle, WA, USA; 3 Los Alamos National Laboratories, Los Alamos, NM, USA Background: The Envelope protein (Env) of HIV-1 mediates viral binding and fusion and possesses an elaborate network of carbohydrate that enables the virus to escape neutralizing antibodies (NAbs). Mechanisms for the development of NAbs have not been elucidated, and the role of sequence divergence in this process remains unclear. Objectives: To evaluate the role of HIV-1 Env mutations, including potential N-linked glycosylation (PNG) sites, in viral fitness and escape from NAbs in eleven macaques infect- ed with the same SHIV isolate. Methods: Proviral env gp120 sequences were obtained from 11 macaques over 1 year of SHIV89.6P infection. Sequences from multiple time points were analysed for nucleotide divergence and diversity, ratio of non-synonymous to synonymous (dN/dS) changes, charge and other amino acid changes, and gain or loss of potential N-linked glycosylation (PNG) sites. Expression plasmids were created from 24 representative vari- ants, and the 21 variants shown to be functional for both fusion and infection were used to create pseudoviruses. Infectivity (TCID50) was scored and correlated with neutral- ization, variable region V1V2 charge changes, and sensitivity to neutralization by soluble CD4 (sCD4) and monoclonal anti- body IgG1b12. Env core and variable loops were modelled. Results: The location of 19/23 PNGs remained highly con- served over the first year of infection. Envs from seven macaques underwent significant divergence from the inoculum with similar changes to PNGs proximal to the CD4 binding site, comprising four additions and two shifts. All six PNG changes have significantly higher dN/dS, indicative of selection pressure. These convergent changes occurred in multiple macaques irrespective of autologous NAbs (Blay et al, J Virol 2006). In macaques that developed sustained Nabs, certain PNG variants medi- ated escape from neutralization by autologous NAbs, het- erologous NAbs, sCD4, and IgG1b12. TCID50 correlated positively with Env content (P=0.04) and negatively with neutralization (P=0.03) and CD4 sensitivity (P=0.009). Neutralization correlated positively with increased acidity in V1V2 (P=0.005). Neutralization sensitivity to IgG1b12 and sCD4 were unlinked. 64 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 65 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 8: VACCINE DELIVERY METHODS/ADJUVANTS OA08-01 OA08-02 Age dependence of adenovirus seroprevalence in Antigen- and vector-specific immune responses South Africa elicited by hexon-chimeric adenovirus serotype 5 vectors and rare serotype adenovirus vectors AR Thorner 1, R Vogels 2, J Kaspers 2, L Holterman2, AAC Lemckert 2, A Dilraj 3, LM McNally4, PM Jeena4, DM Roberts1, A Nanda1, MJE Havenga2, P Abbink1, P Abbink1, A Nanda1, PE Swanson1, AT Bates 1, DM Lynch1, BA Ewald1, A Thorner 1, PE Swanson1, KL O’Brien1, MJE Havenga2, J Goudsmit2, DH Barouch1 AAC Lemckert 2, L Holterman2, B Chen3, J Goudsmit 2 and DH Barouch1 1 Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA; 2 Crucell Holland, Leiden, The Netherlands; 1 Division of Viral Pathogenesis, Beth Israel Deaconess Medical 3 South African Medical Research Council, Durban, South Africa; Center, Harvard Medical School, Boston, USA; 2 Crucell Holland 4 University of Kwa-Zulu Natal, Durban, South Africa BV, Leiden, The Netherlands; 3 Children’s Hospital, Harvard Medical School, Boston, USA Background: Assessing adenovirus (Ad) seroprevalence and neutralizing antibody (NAb) titers in sub-Saharan Africa is Background: The immunogenicity of recombinant adenovirus critical for determining the utility of both common and rare serotype 5 (rAd5) vaccine vectors may be limited by the high serotype Ad vaccine vectors. Ad seroprevalence studies to prevalence of pre-existing anti-Ad5 immunity in the developing date have been conducted only in adults, but pediatric popu- world. Two potential solutions are to utilize capsid chimeric lations will likely be the ultimate recipients of a prophylactic rAd vectors or rare serotype rAd vectors. We have recently HIV-1 vaccine. shown that chimeric rAd5 vectors containing the seven hexon Objectives: To assess seroprevalence and NAb titers to Ad5 hypervariable regions (HVRs) of Ad48 effectively circumvented (subgroup C); Ad11, Ad35 and Ad50 (subgroup B); and anti-Ad5 immunity. Ad26, Ad48 and Ad49 (subgroup D) in pediatric populations Objectives: To compare antigen- and vector-specific immune from South Africa. responses elicited by HVR-chimeric rAd5 vectors and rare Methods: We obtained serum and plasma samples from serotype rAd vectors. neonates (n=42) and infants and children 6 months to 18 Methods: C57/BL6 mice with or without anti-Ad5 immunity years of age (n=633) from South Africa. Samples were divid- were immunized with varying doses of rAd5-Gag, ed into the following age strata: 6 months-1 year, 1–2 years, rAd5HVR48(1-7)-Gag, rAd35-Gag, or rAd48-Gag. Gag-spe- 2–6 years, 6–12 years, and 12–18 years. We performed cific CD8+ T lymphocyte responses were assessed by Db/AL11 luciferase-based Ad neutralization assays and calculated 90% tetramer-binding assays. Gag-specific humoral responses Ad NAb titers for each sample. were assessed by ELISA. Ad5-, Ad5HVR48(1-7)-, and Ad48- Results: Ad5 seroprevalence and NAb titers were high in specific NAb titers were measured by luciferase-based virus neonates (95% seropositive; 47% with titers >1000) and com- neutralization assays. parable with adult levels, presumably reflecting passively Results: In naive mice, rAd5-Gag and rAd5HVR48(1-7)-Gag acquired maternal antibodies. Interestingly, by 6 months of age, elicited more potent Gag-specific cellular and humoral Ad5 seroprevalence waned considerably (13% seropositive; 2% immune responses than the rare serotype vectors rAd35-Gag with titers >1000). Ad5 seroprevalence and NAb titers then and rAd48-Gag. In mice with anti-Ad5 immunity, the rapidly increased in individuals over age 2 and approached immunogenicity of rAd5-Gag was selectively abrogated, and adult levels in subjects over age 6. In contrast to Ad5, sero- rAd5HVR48(1-7)-Gag proved significantly more immuno- prevalence and NAb titers to Ad11, Ad35, Ad50, Ad26, Ad48 genic than all three intact serotype vectors. Vector-specific and Ad49 were substantially lower in all age groups and NAb responses elicited by the three intact serotype vectors increased slowly with age (8–25% seropositive; 0–2% with were type-specific, demonstrating that rAd5, rAd35, and titers >1000 for each serotype in individuals over age 6). rAd48 vectors are serologically distinct. Mice injected with Conclusions: These data demonstrate a critical age dependence rAd5-Gag generated high titers of Ad5-specific NAbs but 10- of Ad5 seroprevalence in South Africa. There is a “window” of fold lower titers of Ad5HVR48(1-7)-specific NAbs, confirm- low Ad5 seroprevalence from approximately 6 months to 2 ing that the majority of Ad5-specific NAbs are directed years of age, followed by a rapid rise to nearly adult titers. In against the hexon HVRs. Conversely, mice injected with contrast, Ad11, Ad35, Ad50, Ad26, Ad48 and Ad49 have low rAd5HVR48(1-7)-Gag generated low titers of Ad5- but high seroprevalence and NAb titers in all age groups. These data sug- titers of Ad5HVR48(1-7)-specific NAbs. Interestingly, these gest that a potential vaccine regimen could involve priming with mice also generated high titers of Ad48-specific NAbs, indi- a rare serotype Ad vector shortly after birth and boosting with cating that the HVRs in this chimeric vector are presented in an Ad5 vector at approximately one year of age. conformations similar to those in wild-type Ad48. AIDS Vaccine 2006 65 Oral Abstracts Sessions 14/8/06 16:40 Page 66 Amsterdam, Netherlands 29 August–1 September 2006 Conclusions: Chimeric rAd5HVR48(1-7) vectors are more was required for detection of vaccinia virus-specific CD8+ T immunogenic than rAd5, rAd35, and rAd48 vectors in the cells. Responding CD8+ T cells also exhibited an unusual phe- presence of anti-Ad5 immunity, demonstrating that these notype (CD45RO–CD27intermediate). chimeric vectors afford unique benefits that cannot be pro- Conclusions: The unique phenotype and high degree of poly- vided by either common or rare serotype rAd vectors. functionality induced by MVA or Dryvax® may be at least Exchanging the hexon HVRs effectively swapped the sero- partially responsible for the profound efficacy of these vac- logic properties of this vector from Ad5 to Ad48, demon- cines in protection against smallpox, and serve as a bench- strating that the HVRs essentially define the rAd vector mark against which other vaccines can be evaluated. serotype. OA08-04 OA08-03 Development of a mucosal vaccine strategy against HIV-1 using virus like particles and Immunization with vaccinia virus induces polyfunc- recombinant clostridium perfringens expressing tional and phenotypically distinctive CD8+ T cell HIV-1 proteins responses P Poonam1, Y Chen1, SP McBurney 2, TM Ross 2 and ML Precopio1, MR Betts2, J Parrino1, DA Price1,3, E Gostick3, P Gupta2 R Bailer1, BS Graham1, M Roederer1, and RA Koup1 1 Department of Infectious Diseases and Microbiology, Graduate 1 Vaccine Research Center, NIH, Bethesda, MD, USA; School of Public Health, University of Pittsburgh, Pittsburgh, 2 Department of Microbiology, University of Pennsylvania, USA; 2 Department of Medicine, Division of Infectious Diseases, Philadelphia, PA, USA; 3 Weatherall Institute of Molecular University of Pittsburgh School of Medicine, Pittsburgh, USA Medicine, University of Oxford, Oxford, UK Background: The mucosa is an important portal for HIV-1 Background: Vaccinia virus immunization provides protec- transmission.Clostridium perfringens has been used as a tion against variola virus, the causative agent of smallpox, vehicle to deliver SIV proteins in large quantity to the ter- and stands as the classic example of a successful vaccine. minal ileum in order to induce strong mucosal immune While vaccinia virus-induced antibody responses have been responses. A mucosal immunization strategy using C. per- shown to be necessary and sufficient for protection against fringens should be able to induce potent mucosal immune monkeypox virus, vaccinia virus-specific T cell responses gen- responses against HIV-1. erated at the time of vaccination may contribute to effective Objective: To develop a mucosal immunization strategy protection. Virus-specific CD8+ T cells contribute to viral against HIV-1 using recombinant C. perfringens expressing control by directly killing virus-infected cells, secreting antivi- HIV protein and HIV-1 virus like particles (VLPs) carrying ral factors, and secreting factors that recruit other cells of the similar antigen. immune system. While virus-specific CD8+ T cells are often Methods: We constructed a recombinant noncytotoxic (cyto- measured by a limited number of parameters, such as IFN γ pathic gene deleted) C. perfringens expressing high levels of and/or IL-2 secretion, the functional profile of T cells is cer- HIV-1 proteins directed by the cytopathic enterotoxin (cpe) tainly more diverse, and the combination of functions that promoter of C. perfringens. The VLPs were obtained by confer protection from infection are uncertain. transfection of COS cell lines with plasmid DNA encoding Objectives: To characterize the functional and phenotypic HIV-1 gag gene followed by purification of the VLP particles profile of vaccinia virus-specific CD8+ T cells in a compara- from the supernatants of these transfected cells. We then tive vaccine trial of modified vaccinia virus Ankara (MVA) immunized mice intranasally with VLPs and boosted with versus. Dryvax® immunization. either VLPs intranasally or recombinant C. perfringens oral- Methods: We analysed vaccinia virus-specific CD8+ T cells in ly, along with adjuvants CpG-ODN and cholera toxin. The vaccinia virus-naive individuals enrolled in a vaccine protocol humoral immune responses in these mice were then measured designed to test whether pre-immunization with modified by quantitating HIV-1-specific IgA and IgG in mucosal and vaccinia virus Ankara (MVA) results in protection from sub- serum samples. T cell mediated immune responses were sequent challenge with the vaccine strain, Dryvax®. determined by measuring HIV-1-specific γ interferon (IFN-γ) Polychromatic flow cytometry was used to characterize the secreting T cells in the spleen, mesenteric lymph nodes functional and phenotypic profile of vaccinia virus-specific (MLN), and Peyer’s patches (PPs) of the immunized mice in CD8+ T cells. an ELISPOT assay. Results: Vaccinia virus-specific CD8+ T cells induced by both Results: The priming with VLPs followed by clostridia MVA and Dryvax® were highly polyfunctional; they degran- boosting resulted in a significant increase in HIV-1 specific ulated and produced IFNγ, IL-2, MIP1β, and TNFα after IFN- γ producing cells compared to the adjuvant only group antigenic stimulation. While one immunization with Dryvax® in spleen, MLN and PPs. However, PPs showed highest induced measurable responses, more than one dose of MVA number of HIV-1 specific IFN-γ producing cells as com- 66 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 67 Amsterdam, Netherlands 29 August–1 September 2006 pared to spleen and MLN cells in these immunized mice. We agonists to TLR-3, 4, 7 and 9 induced both increased thymic did not observe any significant HIV-1 specific humoral production and peripheral proliferation of CD4+, CD25+, immune responses in the mucosal or systemic compartments Foxp3+ T cells while TRL-2 and 5 did not. of the immunized mice. Conclusions: Taken together, these data demonstrate that Conclusions: Mucosal immunization with HIV-1 VLPs and transient depletion of T regulatory phenotype cells can recombinant C. perfringens expressing HIV-1 antigen led to a enhance both systemic and mucosal T cell responses to HIV- strong HIV-1 specific cellular immune response in both 1 Env protein and that TLR 3,4,7 and 9 agonists significant- mucosal and systemic immune compartments. These results ly impact Treg cell production homeostasis. These findings indicate the usefulness of a mucosal immunization strategy have implications for the role of adjuvant design and using HIV-1 VLPs and recombinant C. perfringens to induce modulation of Treg cells in the setting of vaccination. strong mucosal immunity and thereby could be a potent vaccine against mucosal HIV-1 infections. OA08-06 OA08-05 Immunogenicity of a multi-antigen pDNA vaccine with and without in vivo electroporation in Depletion of CD4+, CD25+, Foxp3+ regulatory T cells Rhesus macaques enhances systemic and mucosal HIV-1 Env specific vaccine responses in mice MA Egan1, A Luckay1, I Mathiesen2 JH Eldridge1 and R Kjeken2 J Peacock1, M Ventevogel1, T Kepler 2, K Owzar 2, S Feng2, B Haynes1 and G Sempowski1 1 Wyeth Research, Pearl River NY, USA; 2 Inovio Biomedical Corp., San Diego CA, USA 1 Duke Human Vaccine Institute; 2 Department of Biostatistics and Bioinformatics Objectives: to evaluate the potency of a pDNA-based vaccine encoding multiple HIV antigens in Rhesus macaques when + Background: Removal of CD25 T cells can enhance vaccine delivered by intramuscular injection or in combination with induced immunogenicity to malaria antigen. Additionally, in vivo electroporation (EP). + + CD25 , CD4 T regulatory cells (Tregs) modulate the cellular Methods: The vaccine consisted of three plasmids encoding response to infections such as Epstein Barr virus and HIV-1. HIV env, gag, pol, and a nef-tat-vif (ntv) fusion protein. A Objectives: Our goal was to determine if T regulatory cell fourth plasmid encoding rhesus IL-12 was used as an adju- depletion can enhance the systemic and mucosal immune vant. Three groups of macaques were immunized by i.m. responses to experimental HIV-1 immunogens. We hypothe- injections at week 0, 4 and 8 with either 8.5 mg empty vec- sized that T cell immune responses to HIV-1 Env could be tor DNA (control group), 8.5 mg DNA vaccine (non EP improved following the transient depletion of Treg cells. group) or 1.7 mg DNA vaccine (EP group). Methods: Rat MAb (PC61, 0.250 mg) specific for murine Results: All vaccinations were well-tolerated and no adverse CD25, was delivered by a single i.p. injection concurrent with effects were noted, though EP was associated with a signifi- immunization. A non-reactive rat MAb, Y13, was used as a cant level of lymphocytosis which persisted for at least 22 control. Mice were immunized intranasally (i.n.) with an weeks after the final immunization. HIV-1 Env peptide containing the P18 epitope of gp120 on EP led to a faster onset and stronger cellular responses versus days 0, 7, 14, and 28 using cholera toxin (1 g) as the adju- µ i.m. injection as measured by IFN-γ ELISPOT assay. vant. Mice were euthanized on day 35 and T cell immune Importantly, EP enhanced the immune response against the responses were assessed both systemically (spleen) and less immunogenic antigens (nef-tat-vif) resulting in a more bal- mucoaslly (lungs) using an IFN-γ ELISpot. anced immune response. EP enhanced both CD4 and CD8 Results: We determined that 0.250 mg of PC61 depleted dependent immune responses although a shift towards a CD8 + + peripheral CD4 , CD25 T cells for 5 weeks. We found Treg dependent response was observed. While macaques vaccinat- depletion increased the number of HIV-1 specific IFN-γ spot ed without EP showed little response to the final booster forming cells (SFC) in spleen from 164 ±18 per million cells to immunization, HIV-specific ELISpot responses continued to 506±57 IFN-γ SFC per million cells, P<0.001. Lymphocytes rise following repeated DNA treatments in the EP group. At 8 isolated from the lungs of Treg depleted mice also had signifi- and 22 weeks after the final pDNA immunization there was a cantly higher HIV-1 Env antigen-specific immunity than con- 10- and 40-fold increase in HIV-specific ELISPOT responses trol treated mice, 1,222 ±255 and 423 ±100 respectively, compared to the non-EP group translating to an apparent 50- P<0.05. We also noted that following immunization, Tregs or 200-fold increase in pDNA vaccine potency, respectively. In returned at an accelerated rate in the spleen compared to mice addition, antibody responses against env showed that EP led that did not receive immunization (Treg return on day 28 ver- to −2.5-log increase in antibody titer as compared to the non- sus day 40). To determine the mechanism of Treg induction, we EP group. To monitor pDNA persistence and association with determined the ability of a wide range of TRL agonists alone high molecular weight (HMW) genomic DNA, injection site to modulate thymus Treg production and proliferation of Tregs biopsies were collected from the macaques 70 days after the in the periphery in the setting of vaccination. We found that AIDS Vaccine 2006 67 Oral Abstracts Sessions 14/8/06 16:40 Page 68 Amsterdam, Netherlands 29 August–1 September 2006 last injection and plasmid copies were quantitated by PCR. The results indicate that EP was associated with a significant reduction in pDNA persistence and did not result in an increase in pDNA associated with HMW DNA relative to macaques receiving the pDNA without EP. Conclusions: These data demonstrate that EP was safe, well tolerated and capable of dramatically increasing the potency of a multi-antigen pDNA vaccine in rhesus macaques. These results have important implications for the development of plasmid-based therapeutic vaccines for the treatment of HIV infection. OA08-07 Activation of innate immune responses by Human Immunodeficiency Virus Type-1 (HIV-1) Pr55gag virus-like particles (VLP) S Bredl1, M Wiesel1, L Deml1, J Wild1 and R Wagner 1 1 Institute of Medical Microbiology, Regensburg, Germany; 2 Institute for Hygiene and Social Medicine, Fritz Pregl Strasse 3, Innsbruck, Austria Objective: HIV-1 Pr55Gag virus-like particles (VLPs) produced in the baculovirus expression system have been shown to rep- resent strong inducers of the humoral and CMI in mice and non-human primates. This study aimed towards investigating the molecular mechanisms underlying the strong immuno- genicity and adjuvanticity of such VLPs. Methods: Pr55Gag VLPs were produced (i) using the bac- ulovirus (BV) expression system and (ii) mammalian 293T cells coexpressing the BV envelope protein gp64. Cytokine production (release of γIFN, IL5, others) and upregulation of costimulatory molecules (CD40, 80, 83, HAL DR, CCR7) from splenic cells of naive mice and human monocyte derived DC (huMDDC) were used as a measure for the activation of innate immunity by VLP antigens. Results: Ex vivo studies on huMDDC clearly demonstrated that VLPs of BV origin, but not VLPs produced in mam- malian cells triggered MDDC maturation and activation. This lack of mammalian cell derived VLP to activate huMDDC could not be rescued by pseudotyping these VLPs with BV gp64. Comparable results were obtained when murine splenic cells were used for the ex vivo studies. However, splenic cells from TLR9 knockout mice could be stimulated neither by mammalian, nor by BV produced VLP. Conclusion: These results clearly indicated that the potency of VLP to induce innate immune responses is not an intrin- sic property of VLPs rather than mediated by BV or insect cell derived components. The BV Env protein gp64 is clear- ly not contributing to TLR triggering, may however play a beneficial role in uptake of VLP by APC. 68 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 69 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 9: ANIMAL MODELS/LIVE-ATTENUATED VACCINES/ THERAPEUTIC VACCINATION OA09-01 OA09-02 Host factors contributing to rapid disease Preservation of central memory T cell subsets progression in SIV-infected infant macaques after SIVmac239 challenge following a DNA prime and adenoviral boost vaccination regime K Abel1, K van Rompay1, D Hartigan-O’Connor 2, J Easlick1, K Schmidt1 and M Marthas1 R Schulte1, S Sopper 1, YS Suh1,2, KS Kim2, YS Sung 2, U Sauermann1, G Hunsmann1 and C Stahl-Hennig1 1 UC Davis; 2 US San Francisco, CA, USA 1 Department of Virology and Immunology, German Primate Background: Pediatric HIV infections are often characterized Center, Goettingen, Germany; 2 POSTTECH Biotech Center, by persistently high plasma viral RNA levels and a more Pohang University of Science and Technology, Pohang, Korea rapid disease progression compared to adults. As challenge outcome is determined in the first few weeks after infection, Objectives: The development of a memory T cell response a better understanding of the earliest immune responses in seems to be a critical aspect in the design of an effective HIV local and systemic tissues is needed. vaccine. Recent studies in the simian immunodeficiency Objective: Using the rhesus macaque model of SIV infec- virus (SIV) monkey model of AIDS have shown that T cell tion, the goal was to determine what host responses are memory consists of distinct populations of central memory common to infants and adults, and which factors are (T ) and effector memory (T ) cells. CM EM unique to infants and have the potential to contribute to We measured T cell memory subsets after challenge with path- faster disease progression. ogenic SIVmac239 in the framework of a SIV/IL15-DNA Methods: Infant macaques were infected orally with SIV. prime and Adeno5-SIV boost vaccination regime to analyse the Early immune responses in mucosal and systemic lymphoid alteration of memory cells by different vaccination regimes. tissues were compared to immune responses observed in Methods: Rhesus monkeys (Macaca mulatta) were immu- acute SIV infected adult macaques using flow cytometric, his- nized i.m. with SIV/IL15-DNA at weeks 0, 8 and 16 and tological and gene expression assays. boostered with Adeno5-SIV/IL-15 either i.m. or oraly at Results: Oral SIV exposure of infant macaques results in weeks 24 and 32. Immunized and control animals were chal- rapid virus dissemination with systemic infection being lenged with SIVmac239 via the tonsils on week 44. established by 1 week. Despite similar vRNA levels in local After challenge 12-parameter flow cytometry was used to mea- and systemic tissues at 1 week p.i., immune responses are sure surface markers of T cells. Blood was collected at regular strongest in tissues closest to the site of virus exposure. In intervals after challenge and whole blood was stained with 10 both infants and adults, the early antiviral response is char- different antibodies for each sample. Erythrocytes were lysed acterized by the induction of type I interferons and proin- and cells analysed using a BD LSR II flow cytometer. flammatory cytokines. In fact, frequencies of Results: Vaccination with DNA prime and adenoviral boost interferon-producing plasmacytoid dendritic cells (PDC) resulted in a prolonged reduction of plasma viral load by are higher in infant compared to adult tissues. PDC in more than one order of magnitude in comparison to control infant tissues appear fully functional as they are rapidly animals for 3 months post challenge. Cell-associated viral recruited to local lymph nodes after oral SIV infection and load in the vaccinees was reduced 10-fold during this period. correlate with increased IFN-α mRNA and protein levels in Using 12-parameter flow cytometry we could distinguish the same tissues. multiple T cell subpopulations and describe here their alter- + In contrast, CD8 T cell responses are rarely detectable in ation in vaccinated and unvaccinated animals post challenge. infant macaques. Reduced responses are associated with (1) a We could show that both vaccination regimes resulted in a lower CD4 to CD8 T cell ratio in the first few weeks of life, preservation of the CD4+ T pool for a period of 8 weeks + + CM (2) higher frequencies of regulatory CD4 CD25 FoxP3 posi- compared to a decline in control animals. Vaccinated animals tive cells, and (3) an increased activation threshold of infant showed a steady increase of CD4+ T in contrast to control EM T cells compared to adults. animals, which exhibited a fast increase followed by a Conclusion: While early innate responses are similar in decrease of T at later timepoints. EM infants and adults, infant T cell response seem to be imma- This increase of T significantly correlated with plasma viral EM ture. Thus developmental differences need to be considered in load up to week 8. the design of vaccines and therapeutic interventions against Conclusion: Our data provide evidence that the preservation pediatric and adult HIV infection. of memory T cells should be an important aim in developing an effective HIV vaccine. AIDS Vaccine 2006 69 Oral Abstracts Sessions 14/8/06 16:40 Page 70 Amsterdam, Netherlands 29 August–1 September 2006 OA09-03 OA09-04 Protection against heterologous SHIV challenge Development of a novel, live-attenuated virus following vaccination with HIV-1 W61D recombi- vaccine against HIV nant envelope in the macaque model CK Jurgens1, VJ Madden2, PR Johnson3 and RE Johnston1 M Page1, N Berry1, R Stebbings1, D Ferguson1, R Hull1, D North1, S Harron1, W Elsley1, C Ham1, G Voss2, 1 Carolina Vaccine Institute and Dept. of Microbiology and N Mathy 2, N Almond1 Immunology, 2 Dept of Pathology, University of North Carolina Chapel Hill, Chapel Hill, NC, USA 3 Children’s Hospital of 1 National Institute for Biological Standards & Control, Division Philadelphia, Philadelphia, PA, USA of Retrovirology, Potters Bar UK; 2 GSK, Rixensart, Belgium Objective: Live attenuated SIV vaccines have offered the Background: Passive transfer of potent neutralising antibod- highest degree of protection against challenge with pathogen- ies has been demonstrated to protect macaques against chal- ic SIV isolates. Our goal is to design a vaccine that incorpo- lenge with SHIV. However, the goal is to elicit these rates the advantages of live attenuated vaccines without the serological responses by vaccination. We have developed and inherent safety issues surrounding the use of attenuated applied a SHIV based model to investigate the role of sero- lentiviruses. To achieve this goal, we are utilizing the RNA logical responses in protection elicited by HIV-1 rgp120 genome of the αvirus, Venezuelan Equine Encephalitis virus based vaccines. (VEE) to express SIV/HIV structural genes gag and env. We Objectives: To investigate the breadth of protection conferred hypothesized that the structural proteins would self-assemble by HIV-1 rgp120 vaccination and to identify the corre- into lentivirus-like particles that present Gag and Env in their W61D lates of this protection. native configurations. In addition, if chimeric VEE/SHIV Methods: Two groups of 4 M fascicularis received 8 immu- genomic RNA is packaged into the particles, and if that nizations with a rgp120 HIV-1 expressed in mammalian genome is then released after entry of the particles into sus- W61D cells and formulated in the adjuvant AS02A. Two groups ceptible cells, then the particles could potentially self-propa- were then challenged 4 weeks after the last immunization gate and function as an attenuated virus vaccine. with either SHIVsbg or SHIVsf33 intravenously, along with Methods and Results: Chimeric viral genomes expressing the groups of 4 naive challenge controls. In a separate study, a VEE replicase and the SHIV89.6P structural genes gag and group of 4 naive M fascicularis received 50mls of a pool of env were constructed. Western blot, immunogold labeling immune serum collected from the rgp120 immunised TEM and RT-PCR analysis demonstrated that expression of macaques between their third and 8th immunizations. The chimeric viral RNA in primate cells directed the assembly and recipients were challenged 24 hrs later with SHIVsbg along release of lentivirus-like particles that incorporated Env and with a group of naive challenge controls. Gag and encapsidated viral RNA. Productive infection of sus- Results: Protection against SHIVsbg was observed in all 4 ceptible cells was determined by indirect immunofluorescent vaccinated animals whereas no protection was detected in the staining of VEE nonstructural proteins in syncytia that were group challenged with SHIVsf33. High neutralising antibody formed after infection, by the release of radiolabeled progeny titres against the vaccine virus (SHIVw61D) were detected in particles, and by immunogold labeling TEM. Filtered super- all groups yet, only limited or no cross neutralising activity natants from infected cells induced syncytia formation when was observed in immune serum against SHIVsbg and placed onto a new cell monolayer. SHIVsf33 respectively by T cell line and single round infec- Conclusions and Discussion: After expression of structural tivity assays. Serum transfer did not result in protection from genes gag and env from VEE RNA, lentivirus-like particles that SHIVsbg challenge in spite of the protection observed against encapsidated VEE/SHIV genomic RNA were assembled and this same virus challenge stock following vaccination. released. Particles were capable of propagation in cell culture. In Conclusions: HIV envelope vaccines may protect against het- theory, the vaccine will induce innate immune responses and erologous virus challenge. The mechanism of heterologous pro- will be cleared from the body after sufficient induction of adap- tection remains unclear but does not correlate with neutralising tive immune responses. Since VEE is an RNA virus that repli- antibody titres as with homologous protection. The lack of pro- cates exclusively in the cytoplasm of the infected cell and tection after serum transfer does not support the view that pro- replication is self-limiting, we predict that vaccination will not tection is antibody mediated. Other mechanisms of protection lead to persistent or chronic infection. This design, in theory, mediated by rgp120 HIV-1 need to be investigated. conserves the best aspects of a live attenuated virus vaccine while overcoming the inherent safety issues associated with live attenuated lentivirus vaccines. 70 Antiviral Therapy 11, Supplement 2 Oral Abstracts Sessions 14/8/06 16:40 Page 71 Amsterdam, Netherlands 29 August–1 September 2006 OA09-05 OA09-06 A novel mucosal vaccinia vector for inducing Progesterone abrogates live-attenuated SIV vaccine high-level anti-viral immunity mediated protection in intravenously challenged male rhesus macaques Z Chen1,2, Q Fang1,2*, B Lu2*, K Zhuang2*, W Yu2, W Zhu2, H Wang2, L Liu1,2, P Tien2, DD Ho1 and L Zhang1,2 CJ Miller 1,2 M Genescà1,2, D Lu2 and J Li 2 1 Aaron Diamond AIDS Research Center, The Rockefeller 1 Center for Comparative Medicine; 2 California National University, New York, NY, USA; 2 State Key Laboratory of Primate Research Center, University of California, Davis, CA, USA Virology, Modern Virology Research Center and AIDS Center, Wuhan University, Hubei, PRC Background: Treatment with progesterone decreases the efficacy of live attenuated vaccines in rhesus macaques Background: An effective HIV-1 vaccine remains the ulti- after intravaginal challenge with pathogenic SIVmac239 mate solution for AIDS. Since conventional vaccines are (Abel et al, JID 2004). Progesterone has been reported to not successful, new approaches need to be explored. We induce thinning of the cervicovaginal epithelium and believe that an effective vaccine should induce strong immunosupression, however the exact mechanisms leading immune responses at the sites of HIV-1 transmission to to abrogation of vaccine protection due to Depo-Provera prevent viral entry, eliminate viral dissemination, and clear treatment still remain to be defined. viral reservoirs. Objectives: Determine if progesterone alters the efficacy of Objectives: The goal of this study is to develop a safe and live attenuated vaccines after intravenous challenge. potent mucosal vaccinia vector for inducing specific anti- Methods: Eight SHIV89.6-vaccinated male rhesus macaques viral immunity. were challenged intravenously with SIVmac239; 4 of the ani- Methods and Results: To induce potent anti-viral mucosal mals had been previously treated with Depo-Provera. Viral + immunity, it is essential to deliver the immunogens effective- load, CD4 count, antibody production, INF-γ specific ly via mucosal routes. We generated a modified vaccinia Tian responses and T cell activation were evaluated at 1, 2, 4, 7, Tan (MVTT ) strain by genetic engineering. Our findings 12, 18 and 20 weeks after challenge. 3 indicated that MVTT offers advantages: (1) it displays an Results: 4/4 of the vaccinated Depo-Provera-naive animals 3 attenuated phenotype, which is over 240-fold less virulent controlled viral replication after IV SIV challenge compared compared to the parental strain after intracranial inoculation to 1 of 4 unvaccinated monkeys. Pretreatment of the male in mice; (2) it retains growth capacity in Vero cells, which is rhesus monkeys with Depo-Provera significantly decreased important for future GMP production; (3) it can be delivered protection from viral replication (1 of 4 animals; P=0.039). + via intraoral or intranasal routes without causing significant Vaccinated-protected animals maintained normal CD4 toxicity effects in animals. Most importantly, we did a head- counts and stable CD4/CD8 ratio throughout the post-chal- to-head comparison with the modified vaccinia Ankara lenge period despite only low level of INF-γ production in (MVA) vector to study the use of MVTT as a mucosal vec- PBMC stimulated with SIV Gag peptides. Progesterone-treat- 3 tor. We expressed the spike glycoprotein of SARS-CoV in ed animals had an increase in T cell activation markers (i.e. MVTT (MVTT -S) and MVA (MVA-S) at the equivalent CD28, CCR7, HLA-Dr, CD38, CD95, Ki-67+) in both CD4 3 3 genomic location, and tested their capability in inducing neu- and CD8 T cells. There were particularly striking differences + tralizing antibodies. We found that MVTT -S and MVA-S in the levels of the HLA-DR+CD38-CD4 and the HLA-DR- 3 + + induced comparable levels of neutralizing antibodies through CD38 CD8 subsets. intramuscular route. However, MVTT -S is capable of induc- Discussion: Vaccination with SHIV89.6 in male rhesus 3 ing over 100-fold higher levels of neutralizing antibodies than macaques conferred 100% protection from intravenous chal- MVA-S when inoculated via either intranasal or intraoral lenge with SIVmac239. However, prior treatment with prog- routes. These levels of neutralizing antibodies are substan- esterone abrogated the efficacy of the vaccine. It seems likely tially higher than that induced via the intramuscular route. that the immune activation associated with progesterone We have recently generated a MVTT -based HIV vaccine by treatment contributed to the abrogation of vaccine efficacy. 3 expressing the env gene. Our results indicated that this vac- cine also induced a high level of anti-gp120 antibody responses via the intranasal vaccination in comparison to the intramuscular route. Conclusions: MVTT offers great advantages for inducing 3 antibody and probably cell mediated responses via the mucosal routes of vaccination. Since HIV-1 continues to spread mainly via sexual transmission, our new vaccinia vector may be a candidate for an effective HIV-1 mucosal vaccine. (*equal contributors) AIDS Vaccine 2006 71 Oral Abstracts Sessions 14/8/06 16:40 Page 72 Amsterdam, Netherlands 29 August–1 September 2006 OA09-07 Therapeutic vaccination with SIV DNA+IL-12, IL-15 induces protective T cell memory subsets in SIV infected macaques R Halwani1, BY Diab1, JD Boyer2, DB Weiner2 and RP Sekaly1 1 Laboratoire d’Immunologie, Centre de Recherche du CHUM, Campus St.-Luc, Pavillon Edouard-Asselin, Montreal, Quebec, Canada. 2 Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, USA Background: DNA-based vaccines are usually, by themselves, poor inducers of antigen-specific immunity. The use of “mole- cular adjuvants” such as cytokines, provided a solution for this shortcoming. IL-12 cytokine is considered one of the most potent adjuvants for inducing cellular immune responses. IL- 15, on the other hand, is an important cytokine for the induc- tion and maintenance of CD8 memory. IL-12 and IL-15 might, therefore, enable DNA vaccines to achieve levels of immuno- genicity comparable to those reported for viral vectors. Both cytokines has been shown to improve protective vaccines effi- ciency; however, their effect on memory T cell compartments during therapeutic vaccination is poorly defined. Objectives: To evaluate the effect of therapeutic vaccina- tion with SIV DNA+IL-12 or IL-15 on the induction and maintenance of protective memory subsets in SIV infected macaques. Methods: SIV infected macaques were divided into three groups of seven animals, with one group receiving SIV DNA alone, the second DNA plus IL-12 plasmid, and the third placebo. Selected macaques from each group were also boosted with SIV DNA plus IL-15 plasmid 6 months following the first immunization. Proliferative capacity and functional characteristics of memory T cells were investi- gated by stimulating macaque PBMCs ex vivo with SIV Gag and Env peptide pools using ICS and CFSE techniques. The use of 11 colour flow cytometry allowed us to follow the characteristics of either CD4 or CD8 central and effec- tor memory cells during the course of the study. Results: Our preliminary results indicated that the incorpora- tion of IL-12 into the vaccine resulted in an increase not only in the magnitude, but also in the diversity of specific respons- es towards SIV peptides. CD8 T and T cells of macaques EM CM received DNA+IL-12 showed a much higher increase in cytokine production (3-5 folds in response to some pools) compared to macaques received only DNA or placebo. In addition, proliferative responses were significantly higher for both CD4 and CD8 cells following immunization. This increase in memory response correlated with a lower viral load and a higher CD4 count in most cases. Conclusions: Therapeutic immunization with DNA+IL-12 enhanced the level and functionality of specific CD4 and CD8 memory T cells. Further studies are going on to deter- mine the effect of IL-15 boost on the maintenance of these memory subsets. 72 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 73 POSTER SESSIONS Poster sessions_revGJ 14/8/06 16:34 Page 74 Poster sessions_revGJ 14/8/06 16:34 Page 75 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 1: HIV TRANSMISSION P01-01 P01-02 Detection of HIV-1 intrasubtype superinfection in Cultured enzyme-linked immunospot assay a cohort of high-risk African women (ELISPOT) enhances detection of low HIV-specific immune responses in exposed seronegative indi- B Chohan1,2, A Piantadosi1,2, L Lavreys3 and J Overbaugh1,2 viduals in Kenya 1 Fred Hutchinson Cancer Research Center, Seattle, WA, USA; 2 O Anzala2,3 M Muturi1, C Gichuki1, G Mutua3, E Wala3 Departments of Pathobiology, and 3 Epidemiology, University of and J Gachie3 Washington, WA, USA 1 Kenyatta University, Nairobi, Kenya; 2 University of Nairobi, Background: We previously showed that HIV-1 intersubtype Nairobi, Kenya; 3 Kenya Aids vaccine Initiative, Nairobi, Kenya superinfection occurred in 3 of 20 high-risk women exam- ined; in each case the second virus of a different subtype was Background: There is evidence that many HIV exposed, detected at about one year after initial HIV-1 infection. Little seronegative individuals generate HIV-specific cellular is known about the rate of superinfection at later times after immune responses. IFNγ ELISPOT has emerged as one of initial infection, and whether re-infection by viruses of the the widely used assay to monitor HIV-specific immune same subtype occurs during those times. responses. It is becoming the assay of choice for evaluation Objectives: To determine how commonly superinfection, and of HIV-vaccine-induced cell-mediated responses in many particularly intrasubtype superinfection, occurs during the clinical trials. However, HIV specific IFN-γ T cell responses first 5 years after initial infection. often wane with time and may eventually become unde- Methods: Prospective analysis of 37 high-risk Kenyan tectable using the direct ELISPOT method. There is need for women with extensive longitudinal follow-up was used to a more sensitive detection method to enhance the detection identify cases of HIV-1 intrasubtype superinfection. The of such responses when they fall below levels detectable by women were included if their initial infecting viruses was sub- direct ELISPOT. type A, which is the most prevalent subtype in the region. Objective: To establish if cultured ELISPOT enhance detec- Partial gag and env proviral DNA sequences from blood sam- tion of low HIV specific immune responses. ples collected soon after initial infection were compared to Method: 12 high-risk HIV negative individuals were fol- those present approximately five years later. lowed up every 3 months for 9 months. During follow-up his- Results: Three women were identified who showed evidence tory was obtained, HIV rapid tests performed and blood of superinfection with a second strain of HIV-1 from a dif- samples obtained for assessment of HIV specific responses ferent partner, indicated by separate clustering of viral using both direct and cultured ELISPOT assays. 5 discordant sequences and an unusual amount of viral divergence. Two couples were also recruited and HIV specific responses subjects appeared as probable cases of HIV-1 intrasubtype assessed by both direct and cultured ELISPOT. superinfection. Neighbor-joining phylogenetic analysis of Results: Preliminary data shows that there were no IFN-γ early and late sequences showed separate clustering of either responses detected by direct ELISPOT. 6/12 high-risk HIV gag (first case) or env (second case), and sequence divergence seronegative volunteers gave positive responses to 1, 2, 4, 9 that was greater than expected due to virus evolution. and 90 peptide pools using Cultured ELISPOT at different Interestingly, the second viral sequence was detected with time points during follow up. Out of the 6 high-risk HIV only one of these sequence markers, suggesting the possibili- seronegative individuals who responded positively, 3/6 gave ty that the first and second virus recombined by the time the responses to pool 4 and pool 9, 1/6 responded to pool 1 and later sequences were analysed. In both cases, the superinfect- 2/6 responded to pool 3. 1/5 HIV exposed seronegative part- ing strain was detected at about 4 years PI, but not around 1 ners gave positive responses to 1, 2, 4,9 and 90 peptide pools year; intermediate times are currently being analysed. The using Cultured ELISPOT. Positive responses range from 125 third subject appeared to be a probable case of intersubtype to 1934 SPU/106. superinfection with a subtype C virus. Conclusion: The results indicate that cultured ELISPOT ampli- Conclusions: Preliminary results from this cohort study sug- fied the immune responses and thus enhanced their detection. gest that among high-risk HIV-1 infected women superinfec- Cultured ELISPOT is more reliable in detecting HIV-specific tion is occurring even with closely related viral strains of the immune responses. Further investigations to determine the same subtype. Studies are underway to more precisely pin- phenotype of the proliferating cells are underway. point the timing of superinfection using banked, temporal samples from these individuals. AIDS Vaccine 2006 75 Poster sessions_revGJ 14/8/06 16:34 Page 76 Amsterdam, Netherlands 29 August–1 September 2006 P01-03 indicator strain MBA that are associated with susceptibility to neutralization might be beneficial to better characterize correlates of protection. Maternal neutralizing antibodies toward a CRF01_AE primary isolate are associated with lower mother-to-child transmission P01-04 T Samleerat1,2, G Jourdain3,4, M Braibant1, A Moreau1, N Ngo-Giang-Huong3,4, P Leechanachai2, Multiple HIV-1 variants are transmitted between S Sirithadthamrong5, V Surasaerneewongse6, B Warachit7, mother and infant with selection on the gp120 S Hotrawarikarn8, M Lallemant3,4 and F Barin1 envelope 1 Université F Rabelais, Tours, France; 2 Faculty of Associated G Pollakis1, E Baan1, PML Roelofsen1, SMF Luchters2, J Medical Sciences, Chiang Mai University, Chiang Mai, Thailand; 3 Vyankandondera3, JMA Lange and WA Paxton1 Harvard University, IRD 054, Chiang Mai, Thailand; 4 Institut de Recherche pour le Développement, UMI 174, Chiang Mai, 1 Academic Medical Center, Amsterdam, the Netherlands; 2 Thailand; 5 Banglamung Hospital, Chonburi, Thailand; 6 IATEC, Amsterdam, the Netherlands; 3 CHK, Centre Hospitalier de Bhumibol Adulyadej Hospital, Bangkok, Thailand; 7 Hat Yai Kigali, Kigali, Rwanda Hospital, Songkla, Thailand; 8Klaeng Hospital, Rayong, Thailand Background: Recent studies on heterosexual transmission of Background: Maternal neutralizing antibodies might be part HIV-1 indicate that viruses of a specific genotype are prefer- of the factors involved in protection of mother-to-child trans- entially transmitted. Mother to child transmission (MTCT) mission (MTCT) of HIV-1. A previous study indicated that of HIV-1 is common and targeting the viruses preferentially higher level of maternal neutralizing antibodies against a het- undergoing transmission may help reduce infection rates erologous CRF01_AE primary isolate was associated with amongst children. Inducing effective CTL or neutralizing lower intrapartum transmission in a Thai cohort. antibody responses targeting these viruses in the mother Objectives: To confirm our previous observation, we would therefore be beneficial at limiting MTCT. analysed the maternal neutralizing antibodies against various Objectives and Methods: Identify the viral quasi-species pref- clades B and CRF01_AE HIV-1 primary isolates in a new erentially transmitted between mothers and their infants and population selected with stringent criteria. monitor longitudinal Env and Gag sequences close to time of Methods: 45 transmitting (cases) and 45 non-transmitting labor and longitudinally follow these in the infants and in the mothers (controls) were selected from the participants of a blood and milk compartments of the mothers. clinical trial assessing various zidovudine (ZDV) treatment Results: Amongst 6 HIV-1 transmitting mother/infant pairs 3 durations for the prevention of MTCT in Thailand (no mothers gave birth to an HIV-1-PCR positive child and 3 breastfeeding). They were matched based on the two major transmitted either intra-partum or via breast-feeding. identified risk factors for MTCT: viral load and duration of Compartmentalization and a high degree of genetic variation ZDV treatment. Most of the women were infected by was found between the plasma and the breast milk samples CRF01_AE viruses. Neutralization titers towards 6 primary from the mothers. Sliding window analysis demonstrated a isolates of clades B (FRO, BIG, CHA) and CRF01_AE (MBA, significant level of virus variation attributable to recombina- LEA, C1712) were determined in the maternal sera at base- tion in both mother compartments. From the sequence analy- line by a neutralization assay using CD4+CXCR4+CCR5+ sis of the C2V3 regions in the infants we identified a gp120 HeLa cells. Envelope genes of the CRF01_AE strains were V3 charge restriction to transmission with specific V3 sequenced and compared. charged viruses being preferentially transmitted. The V1V2 Results: MTCT was associated with undetectable or low neu- variable loop lengths were also found to be significantly tralizing antibody titers against MBA (P=0.009), but was not shorter in the viruses transmitted to the children. C2V3 enve- associated with neutralizing antibodies toward the other lope sequence analyses confirmed that in most incidences strains. When restricting the analysis to intrapartum or in minor HIV-1 species were transmitted. Nevertheless, more utero transmitters and their controls, the only association extensive sequence analysis of the virus genes revealed greater that remained significant was between low level of neutraliz- genetic diversity in other regions of the genome. Our results ing antibodies to MBA and risk of intrapartum transmission showing greater sequence diversification away from the (P=0.046). A potential N-linked glycosylation site (N301) in C2V3 region of the envelope would indicate that multiple the V3 region, previously shown as involved in lower suscep- virus variants were being transmitted with subsequent selec- tibility to neutralization, was present in the MBA envelope tion of viruses with the preferential V1V2 and V3 regions gene but not in the two other strains. growing out. Conclusions: This study confirmed our previous observation Conclusions: Studying the genetic diversity of viruses under- that high levels of neutralizing antibody to MBA were asso- going MTCT suggest that multiple virus variants can be ciated with low risk of MTCT in the Thai population. This transmitted with a preferential selection for viruses with a was mainly due to a lower risk of intra-partum transmission. specific V1V2 and V3 sequence upon infection. This has To identify the molecular and biological characteristics of the major consequences for understanding virus transmission 76 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 77 Amsterdam, Netherlands 29 August–1 September 2006 and on designing vaccines aimed at preventing transmission P01-06 since multiple variants will be required to be targeted. Systematic study of potential autoreactivity of the broadly neutralizing antibodies 2F5, 2G12 and P01-05 4E10 DC-SIGN capture of HIV gp120 is disrupted by G Stiegler1, B Vcelar1, H Wolf2, W Muntean3, S CD4 ligation Mehandru4, M Markowitz4, M M Eibl2 and H Katinger1,5 J Arthos, C Cicala, D Van Ryk, E Martinelli, D Goode, 1 Polymun Scientific, Vienna, Austria; 2 Immunology Outpatient C Cruz, K Mcleod, P Schuck, P Sun, G Snyder and Clinic, Vienna, Austria; 3 Medical University of Graz, Graz, AS Fauci Austria; 4 Aaron Diamond AIDS Research Center, New York, NY, USA; 5 Institute of Applied Microbiology, Vienna , Austria National Institute for Allergy and Infectious Diseases, Bethesda MD, USA Background: The human monoclonal antibodies (mAbs) 2F5, 2G12 and 4E10 belong to the very few broadly neu- Background: DC-SIGN functions as an antigen-capturing tralizing HIV antibodies identified to date. Recently, 2F5 receptor on human dendritic cells. The HIV envelope protein and 4E10 were reported to react with several autoantigens gp120, is among the antigens it recognizes. The role of DC- including the phospholipid cardiolipin. Increased preva- SIGN in viral transmission and replication is controversial. lence of antiphospholipid antibodies (aPL) was described in DC-SIGN has been shown to facilitate trans infection of HIV+ patients but also in a wide range of other conditions, CD4+ T-cells in vitro. Alternatively DC-SIGN may promote mainly autoimmune diseases. Persistent high titers of aPL the direct (cis) infection of CD4+ dendritic cells. have been associated with increased risk of thrombosis and Objectives: 1. Delineate the mechanisms underlying high make a risk assessment for 2F5 and 4E10 necessary before affinity capture of gp120 by DC-SIGN, 2. evaluate hetero- further clinical use. geneity in DC-SIGN/gp120 among envelope proteins derived Objectives: To evaluate binding of the mAbs to cardiolipin, from different clades, 3. assess the effect of CD4-induced phosphatidyl-serin, β2-glyco-protein 1, prothrombin. conformational changes on gp120/DC-SIGN interactions. Furthermore, to evaluate the influence on coagulation pro- Methods: We carried out biochemical/biophysical analyses of files of human plasma, including activated partial thrombo- DC-SIGN tetramer interactions with several mammalian cell plastin time (aPTT), prothrombin time (PT) and Russell expressed, genetically diverse HIV and SIV gp120s. venom test (DRVVT). Analytical ultracentrifugation, circular dichroism and surface Methods: Binding activity of the mAbs to autoantigens was plasmon resonance were employed to evaluate the sto- tested by ELISA. Reactivity of plasma samples obtained pre- icheiometry of DC-SIGN/gp120 complexes, the effect of DC- and post infusions with 4E10, 2F5/2G12 or 2F5/2G12/4E10 SIGN ligation on gp120 structure, and the influence of CD4 were evaluated. Changes of aPTT, PT and DRVVT were ligation on DC-SIGN/gp120 complexes. monitored in 4 patients receiving a high dose regimen of Results: Although all of the gp120s we evaluated bound DC- 2F5/2G12/4E10 and in adult and neonate plasma spiked with SIGN with high affinity, the stoicheiometry for different the antibodies. gp120 isolates varied from 1 DC-SIGN tetramer per gp120 to Results: 2F5 and 2G12 tested negative to binding to all eval- 4 DC-SIGNs per gp120, demonstrating that gp120 sequence uated autoantigens. In contrast, 4E10 tested positive to reac- variation can influences the nature of this interaction, more- tivity to cardiolipin, phosphatidyl-serin and β2-glyco-protein over, multiple carbohydrate patches on gp120 can serve as 1 in a dose dependant manner. Detection of lupus anticoagu- binding sites for DC-SIGN. Ligation of gp120 by DC-SIGN lant could not be confirmed. Concordantly, mild prolonga- altered gp120 structure in a manner that likely reduces con- tions of aPTT and DRVVT were found in adult and neonate formational flexibility of gp120. We also determined that plasma spiked with 4E10 but not with 2F5 or 2G12. CD4 and DC-SIGN can bind gp120 simultaneously, however Positive reaction to cardiolipin was found in plasma samples CD4 ligation to gp120/DC-SIGN complexes promoted faster from patients treated with 2F5/2G12/4E10 or single dose dissociation of DC-SIGN from gp120. 4E10 but not in patients treated with the 2F5/2G12 combi- Conclusion: Genetically diverse gp120s are recognized by nation. Mild and transient prolongation of aPTT was found DC-SIGN differently, likely a result of distinct glycosylation in patients treated with 2F5/2G12/4E10. PTs remained patterns. These differences may have implications for the effi- unchanged. ciency of HIV capture by DC-SIGN expressing dendritic Conclusions: There is no evidence that 2G12 and 2F5 har- cells. We further determined that CD4 promotes dissociation bour relevant autoreactivity. Cardiolipin reactivity and pro- of gp120 from DC-SIGN, an effect that may facilitate pro- longation of aPTT could be attributed exclusively to 4E10. ductive infection (CD4 dependent) of DC-SIGN captured Clinically, the infusion of 2F5/2G12/4E10 did not lead to sig- virions. nificant side effects in 39 treated HIV+ adults. There were no thrombotic complications, including the absence of venous thrombosis at infusion and venipuncture sites. Nevertheless, AIDS Vaccine 2006 77 Poster sessions_revGJ 14/8/06 16:34 Page 78 Amsterdam, Netherlands 29 August–1 September 2006 future 4E10 infusions should be performed under stringent P01-08 safety monitoring including assessment of coagulation pro- files. Low virus (SHIVsf162p) dose intra-vaginal chal- lenges do not conform with the mass-action prin- ciple: implications for HIV vaccine development P01-07 S Garg1, J Mandl1, B Lawson1, R Regoes2, and S Staprans1 Prevalence of sexually transmitted infections in areas of low HIV prevalence in the Asia Pacific 1 Emory Vaccine Center, Emory University, Atlanta, GA, USA; 2 Institute of Integrative Biology, ETH Zurich, Switzerland NT Thanh Thuy1, E Sulivan2, E Corpus3 and B Chultemsuren4 Objectives: Most HIV transmission events are associated with small virus inocula, yet trials in macaque models to eval- 1 World Health Organisation, Western Pacific Regional Office, uate HIV preventions are usually conducted with high dose virus challenges that may underestimate their efficacy. Manila, Philippines; 2 University of New South Wales, School of Challenge models are needed to provide more physiological- Women’s and Children’s Health, Sydney, Australia; 3 Department ly relevant measures of vaccines or microbicides efficacy, that of Health, Manila, Philippines; 4 Ministry of Health, Ulaanbaatar, can provide insights into determinants of viral infection. We Mongolia developed a repeated low-dose inoculum model with CCR5- utilising SHIVsf162P that better mimics HIV transmission. Background: Sexually Transmitted Infections (STI) play a Methods: We calculated the infectious dose 50% (ID50) of critical role in facilitating HIV transmission. Recording and the virus stock by titration in 12 depoprovera treated female monitoring of STI are crucial for HIV prevention in countries pigtail macaques (PTMs) by intravaginal inoculation with of low HIV prevalence . various virus doses (25-2500 TCID50). Five PTMs were Objectives: This review provides information on specific STI repeatedly challenged with 1 ID50 at 5-week intervals until from existing studies aiming to guide effective prevention the viremia became detectable. Peripheral and localized vagi- interventions of HIV. nal viral replication and markers of immune activation were Methods: Analysis of STI prevalence is based on surveys monitored. We assessed the relationship between absolute recently conducted during 2002-2005 in a number of coun- inoculum size or the animals’ in vitro susceptibility to infec- tries of low HIV prevalence (<1% among pregnant women tion, and in vivo replication, by calculating the area under the and <5% in populations at high risk behaviors). Laboratory virus load curves (AUCVC), an index of viral replication that techniques used in surveys were RPR and TPHA tests for accounts for the time at which viremia is first detectable and syphilis and PCR for chlamydiosis, gonorrhoeae and tri- peak viremia. chomoniasis. Results: Four of 12 animals exposed to single low doses of Results: Among populations at high risk behaviors, Syphilis virus (25-100 TCID50) did not become infected. Animals seropositivity (RPR) in sex workers (SW) was 15.7% in that became infected with 50 and 100 TCID50 showed Mongolia and particularly high in the capital Ulaanbaatar detectable but transient viremia. As predicted, 3 PTMs out of (37%). Among general population such as pregnant women, 5 exposed to the determined ID50 became systemically in the Pacific, at least one STI was observed in 27.0% in infected after first challenge, 1 after 2 challenges and one Samoa, 17.7% in Tonga, 15.5% in Solomon Islands, and after 4 challenges. Some animals showed evidence of local- 14.6 % in Kiribati. Chlamydia is the most prevalent STI. It ized vaginal SHIV replication before systemic infection. was found in 18% in pregnant women of these Pacific Islands Contrary to previous findings, our data show that there is a (from 6.4% in Solomon Island to 29% in Fiji), in 9% in male significant correlation between inoculum size and AUCVC youth and 7.7% in female youth of the Philippines. Young (P=0.001). There was no relationship between in vitro sus- age is significantly associated with STI in pregnant women ceptibility of individual animals and AUCVC, nor between (P<0.05). During 2002-2004, the syphilis rate increased sig- baseline immune activation and AUCVC. Preliminary analy- nificantly among male STI clients (1% to 13%), and in blood sis shows that animals exposed to lower virus doses mounted donors (1.3% to 3.4%) in Ulaanbaatar Mongolia. However, higher cellular immune responses. during the same period, in Dakhan Mongolia, Syphilis rate in Conclusions: This model provides novel insights into the SW decreased (26% to 14%), in line with an increased con- transmission process, suggesting that infection occurs not by dom use in this group (26 to 74%). a simple mass action process, but rather that localized viral Conclusions: High prevalence and increased trends of STI spread occurs prior to systemic dissemination. Infection even among general population indicate presence of risk sex- dependence on localized virus spread may enable vaccine- ual behaviors and need of strengthened STI interventions in induced immunity to protect from infection with low doses. low HIV prevalence countries, particularly appropriate strat- egy for control of Chlamydia and condom promotion. 78 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 79 Amsterdam, Netherlands 29 August–1 September 2006 P01-09 ment. To this goal Phase IIA and then IIB-proof of concept of efficacy will be supported by the Italian Government and conducted in Italy and South Africa. Prophylactic and therapeutic Phase I trials with the active Tat protein vaccine: results of the inter- im analysis B Ensoli1, V Fiorelli1, O Longo1, S Bellino1, F Ensoli2, A Tripiciano1,2, A Scoglio1,2, M Ruiz1,2, B Collacchi1,2, V Francavilla1,2, A Lazzarin3, G Tambussi3, R Visintini3, P Narciso P4, G D’Offizi4, M Giulianelli4, M Carta5, A Di Carlo2, G Palamara2 and M Giuliani2 1 National AIDS Center, Istituto Superiore di Sanità, Rome, Italy; 2 Lab. of Experimental Immunology and Allergy, S. Gallicano IRCCS, Rome, Italy; 3 Dept. of Infectious Disease, San Raffaele Hospital, Milan, Italy; 4 Spallanzani IRCCS, Rome, Italy; 5 Dept. of Clinical Medicine, University of Rome “La Sapienza”, Rome, Italy Aim: Novel vaccination strategies aimed at blocking virus replication and disease onset have been recently developed by targeting regulatory genes/products, which are essential for virus replication and infectivity. In particular, Tat is a very early protein produced upon virus entry even prior to virus integration. Native Tat targets and enters efficiently dendrit- ic cells inducing their maturation, a property relevant for vac- cination strategies. Tat presents a high degree of immune cross-recognition among the different virus clades. Further, immune responses to Tat in natural infection correlate with lower progression to AIDS. Finally, vaccination of monkeys with the native Tat protein resulted safe and induced immune responses that contained virus replication blocking disease onset upon virus challenge. Based on these data, parallel pre- ventive and therapeutic multicentric Phase I clinical trials with the native Tat protein have been sponsored by the ISS and carried out in 4 clinical sites in Italy. Methods: The trials, both randomized, double blinded, and placebo-controlled, were aimed at evaluating the safety (pri- mary endpoint) and the immunogenicity (secondary end- point) of 5 immunizations with 3 doses (7.5, 15 or 30 µg) of the recombinant active Tat protein vaccine given subcute in Alum or intradermally with no adjuvant in HIV-negative healthy adults without identifiable risk of infection (20 vol- unteers) and in HIV-1 infected adult volunteers (27) with CD4+ T cell counts ≥400/µl and plasmaviremia ≤50,000 copies/ml, respectively. Parexel International Corporation has been the CRO for quality assurance, monitoring, data man- agement and statistical analysis. Results: The results (interim analysis at 24 weeks) indicate that the vaccine was safe and well tolerated in all subjects. All vaccinated volunteers (100%) developed specific antibodies (IgM, IgG, IgA). Sera from vaccinees neutralized Tat activity in the rescue of Tat-defective proviruses. More than 80% of all volunteers also developed specific cellular responses including lymphocyte proliferation, γIFN- and IL4-Elispot. Follow-up will continue for 3 years. Conclusions: These data indicate to proceed to Phase II stud- ies in high risk individuals and HIV infected people on treat- AIDS Vaccine 2006 79 Poster sessions_revGJ 14/8/06 16:34 Page 80 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 2: ACUTE HIV INFECTION P02-01 coreceptor for viral entry. Two CVL functional clones were also shown to use CCR5. Conclusion: We have generated pseudovirions from acute Generation of functional envelope clones from HIV-1 subtype C infected individuals. These will be useful to HIV-1 subtype C plasma RNA for use in study the properties of recently transmitted viruses including pseudovirion assays their sensitivity to both autologous and heterologous anti- body neutralization, coreceptor inhibitors and entry fitness. I Choge1, P Moore1, N Leseka1, E Gray1, F Treurnicht2, K Mlisana3, SS Abdool Karim3, C Williamson 2, L Morris1 and the CAPRISA 002 study team P02-02 1 AIDS Virus Research Unit, National Institute for Communicable Novel role for HIV-1 transframe protein p6* in Diseases, Johannesburg, South Africa; 2 Institute of Infectious viral replication associated with Nef function Disease and Molecular Medicine, University of Cape Town, Cape Town, South Africa; 3 Centre for the AIDS Programme of Research in South Africa, University of KwaZulu Natal, Durban, A Leiherer, C Ludwig and R Wagner South Africa Institute of Medical Microbiology and Hygiene, University of Background: Acute HIV-1 infection is associated with high Regensburg, Regensburg, Germany viral loads and an increased risk of transmission. Understanding the properties of viruses from this early Objective: The HIV-1 transframe protein p6*, previously stage of infection may assist in identifying transmission suggested to play a role in regulation of HIV-1 protease acti- determinants. The ability to clone functional envelopes vation, has been reported to bind to 28-kDa Nef protein, a directly from plasma and generate pseudovirions which key player of HIV-1 infectivity and progression to AIDS. This undergo a single cycle of replication has facilitated the study prompted us to clarify the proposed interaction between Nef of these early viruses. The pseudovirion assay thus allows and p6* and analyse its importance for HIV-1 replication. viruses to be characterized for phenotypic properties includ- Methods: A panel of recombinant NL4-3-derived proviruses ing sensitivity to antibody neutralization and entry either containing or lacking nef was generated comprising inhibitors as well as the ability to link these phenotypic clustered mutations throughout the p6* coding region with- properties to genetic determinants. out affecting gag and pol open reading frames nor release of Objectives: To generate functional pseudovirions to study the viral protease. Besides, the role of a central p6* domain antibody mediated neutralization. duplicated in the nef-negative virus HX10 was addressed by Methods: A highly exposed cohort of female sex workers in inserting this region into wild-type NL4-3. The effect of these Kwa-Zulu Natal, South Africa, was used to identify patients mutations on viral infectivity and replication has been exam- with acute HIV infection. Plasma samples were available ined in cell culture. from 19 women within 2-11 weeks of infection. Matched cer- Results: None of the partially mutated p6* variants signifi- vico-vaginal lavage (CVL) samples were available from 2 cantly influenced viral infectivity or replication of the corre- women. RNA was extracted from plasma and CVL; gp160 sponding viruses. However, mutation of the entire p6* region PCR amplicons were generated and cloned into pcDNA decreased viral infectivity irrespective of nef expression. 3.1D/V5-His-TOPO directional vector (Invitrogen). pEnv Whereas this fully mutated p6* delayed the replication of and a subtype B based pSGDEnv backbone were co-trans- nef-containing viruses, replication was completely abolished fected into 293T cells to generate pseudovirions. Screening in established cell lines in the absence of nef. Furthermore, for infectivity was done in JC53BL-13 cells. Pseudovirion viral infectivity of nef-positive and -negative NL4-3 viruses stocks were generated by large-scale transfection in 293T was enhanced by the HX10-derived duplication of the central cells. Cloned envelopes were sequenced and compared to the p6* region which appeared to partly rescue compromised population sequence. replication of nef-lacking viruses. Results: Functional envelope clones were obtained from plas- Conclusion: Our results suggest a new role for p6* in viral ma virus from 15 of 19 acutely infected female sex workers. replication associated with Nef function and further experi- On average 16 clones were screened with approximately ments are underway to clarify the importance of distinct p6* 50% functionality. Sequence analysis of multiple functional regions for potential Nef interaction. clones from 2 patients showed minimal intra-patient diversi- ty. Phylogenetic analysis of the C2-V5 env region confirmed that the patients were infected with HIV-1 subtype C and the V3 loop was typical of subtype C viruses with the GPGQ crown. Coreceptor analysis confirmed that all used the CCR5 80 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 81 Amsterdam, Netherlands 29 August–1 September 2006 P02-03 Evaluation of novel strategies for identification of primary HIV-1C infection in Botswana V Novitsky2, E McDonald1, M Ketunuti1, R Rossenkhan1, E Woldegabriel1 and M Essex2 1 Botswana-Harvard Partnership, Boston, USA; 2 Harvard School of Public Health/Botswana-Harvard Partnership, Boston, USA Objective: Prospective HIV-1 testing of at-risk populations is a classical approach for identifying primary HIV-1 infection. To create a cohort of primary HIV-1C infection, new approaches are necessary. Methods: A novel Voluntary Counseling and Testing (VCT) based recruitment strategy was developed as a part of an ongoing study on acute and early HIV-1 subtype C infection in Botswana. The recruitment strategy involved partnering with 16 government-run VCT centers in Gaborone, the capi- tal of Botswana. Primary HIV-1C infections were defined by either HIV-1 EIA(-) and RT-PCR(+) test results or EIA(+) and detuned EIA(-) test results. The initial testing strategy was ultimately refined by 1). designing new referral/testing algo- rithms; 2). targeting two government-run VCT clinics; 3). improving health care worker (HCW) and community edu- cation; 4). collaboration with the busiest VCT center; and 5) pairing with diverse VCT providers including NGO and pri- vate sector organizations. Information on referral sources was collected from patient self-report. Results: In 13 months, 1171 screening tests resulted in the identification of 6 acute and 42 recent HIV-1C infections. The number of identified recent infections was significantly greater after refinement of recruitment strategies (P=0.0006). Significant associations were found between numbers screened and identified (R=0.75, P=0.003). Significant differ- ences were found in screening efficiencies after refinement (P=0.005). 445 clients reported referral sources. Important sources were: health care workers referral (68%); recent exposure (16%); community presentations (4%); and educa- tional posters (4%). Conclusion: VCT based strategies and cross-sectional testing using an advanced algorithm resulted in the identification of 48 primary HIV-1 infections over 13 months including 6 cases of pre-seroconversion. Developing HCW referrals and community education appear to be promising future approaches. This provides critical opportunities for educa- tional/behavioral interventions in the populations that appear to be at a greater risk for primary HIV-1 infection. AIDS Vaccine 2006 81 Poster sessions_revGJ 14/8/06 16:34 Page 82 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 3: INNATE IMMUNITY P03-01 P03-02 Blocking of inhibiting receptors on fresh NK-cells HIV-1 replication in human primary cells is leads to increased degranulation in HIV enhanced by macrophage migration inhibitory factor (MIF) S Johansson1, 2, J Hinkula1, 3, B Wahren1, 2, K Kärre 2 and L Berg2 EG Regis1,2, V Barreto-de-Souza1, MT Bozza2 and DC Bou-Habib1 1 Department of Virology, Swedish Institute for Infectious Disease Control, Stockholm, Sweden; 2 Microbiology and 1 Department of Immunology, Oswaldo Cruz Institute/FIOCRUZ, Tumorbiology Center, Karolinska Institutet, Stockholm, Sweden; Rio de Janeiro, RJ, Brazil; 2 Department of Immunology, Institute 3 Linkoping University, Department of Molecular Virology, of Microbiology/UFRJ, Rio de Janeiro, RJ, Brazil Linkoping, Sweden Background: Cytokines are believed to play a crucial role in Background: Inhibitory Killer Ig-like Receptors (KIRs) on the pathogenesis of HIV-1 infection. The pro-inflammatory NK-cells recognize MHC class I molecules on target cells. It cytokine Macrophage Migration Inhibitory Factor (MIF) is has previously been shown that KIR-ligand mismatched NK- involved in the development of many inflammatory and cells can kill tumor targets. Not much is known about the infectious diseases. However, participation of MIF in HIV-1 effect of blocking inhibiting receptors on NK-cells when infection is still unknown. using HIV infected target cells and most of the research has Objectives: To analyse whether MIF modulates HIV-1 repli- been done on NK-cell lines and clones. cation in human monocyte-derived macrophages (MDM) Aim: We therefore aimed at investigating if blocking of KIRs and phytohemaglutinin-activated peripheral blood mononu- recognizing MHC class I molecules could upregulate degran- clear cells (PBMCs). ulation of fresh NK-cells from healthy individuals and HIV Methods: PBMCs were obtained from healthy donors by densi- infected patients when using different target cells. ty gradient centrifugation, and MDM by plastic adherence of Methods: Blood was drawn both from healthy individuals PBMCs. Cells were infected with the HIV-1 isolate Ba-L (R5- and from HIV infected patients. PBMCs were stimulated tropic), and infected cells were exposed or not to recombinant with IL-2 over night and then used as effectors in a CD107a human MIF (rhMIF) or anti-MIF antibodies (20 µg/ml). Levels FACS staining assay together with target cells. This assay of MIF and HIV-1 replication (detection of p24 Ag) were mea- stains the CD107a molecule as it is exposed on the cell sur- sured in cell culture supernatants by ELISA. Cellular expression face during the process of degranulation. Targets were MHC of CD4 and CCR5 was analysed by flow citometry. class I expressing cell lines and as controls class I defective Results: We found that HIV-1 infection of PBMCs induced an cell lines were used. PBMCs were treated with an anti-KIR augmented MIF secretion in 40% of the donors (n=8), when antibody. The effector and target cells were incubated for 4h compared to uninfected cells. The addition of exogenous at 37°C, in presence of anti-CD107a FITC antibody. The cells rhMIF (25 ng/ml) to HIV-1 infected cells doubled viral repli- were then FACS stained for CD3 and CD56. Fresh PBMCs cation in PBMCs (n=3, P=0.0055). A similar effect was also were also stained for a panel of NK-receptors. observed with other MIF concentrations. Likewise, rhMIF Results: In the CD107a FACS staining, increased NK-cell augmented HIV-1 replication in MDM in a dose-dependent degranulation in PBMC from healthy as well as HIV infected manner. The replication of HIV-1 in PBMCs was inhibited by individuals could be seen after treatment with the anti-KIR 42% (n=3, P=0.0007) when infected cells were treated with antibody. A changed expression pattern of various NK-recep- anti-MIF neutralizing antibodies. Exogenous rhMIF did not tors on NK-cells from HIV-infected patients was also detected. up-modulate cellular expression of the molecules CD4 and Discussion: The results show that it is possible to increase the CCR5, suggesting that MIF-induced augmented HIV-1 repli- degranulation of NK-cells by inhibiting the KIR - MHC class cation was not due to higher expression of HIV-1 receptors. I binding. Decrease of the activation threshold for NK-cells in Conclusions: Our results suggest that MIF up-modulates a similar manner might in the future be used as a potential HIV-1 replication in peripheral human cells, as shown by treatment for HIV infected patients. increased MIF secretion by HIV-1-infected PBMCs, augment- ed viral replication in PBMCs and MDM exposed to rhMIF, and inhibition of viral replication in HIV-1-infected PBMCs exposed to anti-MIF antibodies. Our study calls for further investigation of the mechanisms by which MIF up-modulates HIV-1 replication, and the role of MIF in the clinical progress of HIV-1 infection to AIDS. Support: PAPES/FIOCRUZ, Faperj, CNPq, CAPES. 82 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 83 Amsterdam, Netherlands 29 August–1 September 2006 P03-03 P03-04 CAF-activity in long-term SIV-infected rhesus Native Tat preferentially enters dendritic cells macaques inducing tumor necrosis factor-α mediated Th1- polarized maturation and immune responses in J Harbort, G Hunsmann and U Sauermann monkeys and humans Department of Virology and Immunology, German Primate A Cafaro, E Fanales-Belasio, S Moretti, V Fiorelli, M Center, Goettingen, Germany Pavone-Cossut, A Tripiciano, F Ferrantelli, E Olivieri, F Nappi, I Macchia, P Monini and B Ensoli Background: The infection of rhesus macaques with simian immunodeficiency virus (SIV) is a valuable model for human National AIDS Center, Istituto Superiore di Sanità (ISS), Rome, Italy AIDS research. After SIV-infection about 1 % of the macaques become long-term non-progressors (LTNP), which Background: Exogenous, bioactive and monomeric Tat pro- are characterized by low viral load, no clinical progression to tein, is very efficiently taken up by monocyte-derived den- AIDS for up to 8 years and an effective anti-SIV immune dritic cells (MDDC), in a dose-, time-, maturation- and response. A major component of this immune response may temperature-dependent fashion. After internalization, Tat + be the CD8 T cell antiviral non-cytotoxic response (CNAR), activates MDDC by inducing their maturation and augment- + mediated by the soluble CD8 T cell antiviral factor (CAF). ing recall and allogeneic antigen presentation. This factor, however, has remained yet undefined. Aim: To elucidate the mechanism(s) through which Tat binds, Objectives: In this study we want to analyse if LTNPs pro- enters, and promotes MDDC maturation/activation and duce CAF. Furthermore we want to study if CAF activity is immune responses. crossreactive in rhesus macaques of Indian and Chinese ori- Methods: MDDC preparation, analysis and Tat uptake were gin. done as previously reported (Fanales-Belasio et al, J. + Methods: CD8 T cells were isolated from SIV-infected and Immunol. 2002). In blocking studies, MDDC were cultured non-infected, Mhc class I mismatched rhesus macaques. at 4°C for 2 h in the presence of mAbs directed against sev- + CD4 T cells from non-infected healthy rhesus macaques eral integrins prior to the addition of Tat. + were stimulated for 24h with Con A. CD4 T cells were Results: The uptake of pico/nanomolar amounts of native Tat -3 infected with a MOI of 1.6× 10 SIVmac239 and coincubat- protein is selective for the monocytic lineage ed with CTLs at ratios of 1:1 and 1:2 for 13 days, respec- (MDDC>macrophages>monocytes), and negligible for lym- tively. After different time points cell supernatant was phocytes from the same donors. It occurs via a receptor-medi- harvested and RNA was isolated. Viral replication in the ated endocytosis, upon binding of the Tat absence or presence of CTLs was determined by Real Time arginine-glycine-aspartic (RGD) region to the , and α5β1 αvβ3 PCR. α β integrins. At higher concentrations Tat enters MDDC + v 5 Results: CD8 T cells from 4 of 6 LNTPs of Indian origin through both this pathway and a non-cell specific interaction reduced SIV replication by 3-6 logs in a dose dependent man- by its basic region. ner. In comparison, CTLs from healthy donors or slow pro- Tat-induced MDDC maturation and T helper 1 (Th1) + gressors reduced SIV replication by 1-2 logs. CD8 T cells cytokine production do not occur when either an oxidized from LNTPs of Indian origin inhibited SIV replication only protein, which does not enter MDDC, or a mutant (Tat ), + cys22 by 1-2 logs in CD4 T cells from macaques of Chinese origin. which enters MDDC but has no transactivating activity, is Thus, CAF produced by CTLs from Indian origin macaques used. Further, the Tat-induced Th1 polarized maturation had no significant effect on viral replication in macaques occurs only with CD1a+ MDDC, and is mediated by the from Chinese origin. induction of TNFα production by Tat, consistent with the Conclusions: Two thirds of the analysed LNTPs of Indian ori- transcriptional activation of TNF gene expression by Tat, and gin produced CAF, indicating the LNTP status can be the known role of TNFα in mediating MDDC maturation. induced by a variety of factors. Unexpectedly, in contrast to Of note, inoculation of cynomolgus monkeys with Tat but the cross-species activity described for CAF, viral replication not Tat induces a prevalent Th1 response. The same + cys22 in CD4 T cells from macaques of Chinese were unaffected by response is found in asymptomatic drug-naive infected indi- this activity. We propose that potentially polymorphic factors viduals. + of the CD4 T cells influence or modify CAF-activity. Conclusions: Native Tat selectively targets and is taken up by MDDC via RGD-binding integrins and promotes their matu- ration and activation towards a Th1 polarizing phenotype through transactivation of TNFα gene expression. This is consistent with the low anti-Tat antibodies prevalence and high γIFN Elispot responses in natural infection. The MDDC targeting and T cell recruitment and activation by Tat is rele- vant both in AIDS pathogenesis and vaccine development. Supported by Grants ICAV, ISS, Rome, Italy. AIDS Vaccine 2006 83 Poster sessions_revGJ 14/8/06 16:34 Page 84 Amsterdam, Netherlands 29 August–1 September 2006 P03-05 P03-06 Multiple newly identified uridine-rich TLR7/8 Broad anti-viral activity of group X secreted ligands within the RNA of HIV-1 activate human phospholipase A2 CD8+ T cells J Kim, B Chakrabarti, A Guha-Niyogi, L Ganesh and A Meier, G Alter, H Streeck, N Teigen and M Altfeld G Nabel Partners AIDS Research Center, Massachusetts General Hospital, Vaccine Research Center/NIAID/National Institutes of Health Harvard Medical School, Boston, MA, USA Background: Phospolipase A2 is an enzyme that hydrolyzes Background: Toll like receptors (TLRs) are a family of pat- phospholipids, leading to the generation of lysophospholipids tern recognition receptors involved in the initiation of the and free fatty acids. There are more than ten isoforms of this immune response, and TLR7/8 have been recently demon- enzyme, including several secreted A2 phospholipases strated to recognize viral single stranded RNA (ssRNA). Here (sPLA2s), some of which have been implicated in the innate we describe the identification of nine novel TLR7/8 ligands immune response because of their bactericidal effects. within uridine-rich regions of HIV-1 ssRNA. Objective: The aim of this study was to identify sPLA2s with Objective: To assess the impact of TLR ligands on CD8 T potent antiviral activity on enveloped viruses and to elucidate cells and to evaluate the potential of HIV-1 to trigger immune the mechanism involved in antiviral effect. activation via TLRs Methods: Human group II, III, V, VII, X, and XII of sPLA2s Methods: Uridine-rich regions (>50%) within the HIV-1 were cloned into mammalian expression vector and trans- genome were identified. Human PBMCs from were stimulat- fected into 293 cells. sPLA2s-containing cell culture super- ed for 6 hrs with these newly identified HIV-1-derived TLR natants were incubated with diverse replication-defective + ligands and activation of CD8 T Cells was subsequently enveloped-pseudotyped lentiviruses. The antiviral activity of analysed by flow cytometry. Furthermore, activation of puri- sPLA2s was evaluated by luciferase-reporter activity of + + fied CD8 T cells, as well as monocytes-depleted CD8 T pseudoviruses. Mutagenesis studies of group X sPLA2 cells, were studied, and trans-well experiments were per- (sPLA2-X) were performed to understand the relationship + formed to investigate the interplay between CD8 T cells and between the structure and function. The effect of sPLA2-X on monocytes in TLR-mediated activation. live wild-type HIV was assessed by single-round replica- ADA Results: Stimulation with 9 of these predicted ligands result- tion assay using p24 Gag staining in target cells. To elucidate + ed in the activation of CD8 T cells, whereas the U-to-A vari- the mechanisms of antiviral effect, pretreatment of cells or ant of these ligands induced no T cell activation (median% viruses by sPLA2-X were performed. p24 release from the + CD69 CD8 T cells with U-rich HIV-1 RNA 9.12%; range virus was analysed using density-gradient centrifuge of virus 7.1% to 18.1% versus 0.32%; range 0.23% to 0.46% with treated by sPLA2-X. + U-to-A variant, P<0.001). The activation of CD8 T cells by Results: Human sPLA2-X displayed anti-viral activity these novel HIV-1-derived ssRNA oligo-nucleotides was fur- through its ability to degrade viral but not host cell mem- thermore completely blocked with Chloroquine, an inhibitor branes. sPLA2-X showed potent anti-viral activity against of lysosomal acidification that has been shown to serve as an specific viruses, including HIV-1, Ebola and Marburg, but inhibitor of TLR 7/8 signalling, and also abrogated by the not VSV or influenza envelope pseudotyped reporter viruses. + depletion of monocytes, or the separation of CD8 T cells and The antiviral activity was specific for sPLA2-X since other monocytes in transwell experiments. catalytically active members of the sPLA2 family failed to Conclusion: These data demonstrate that uridine-rich regions exert anti-viral effects. Mutagenesis studies confirmed that of the ssRNA of HIV-1 can serve as potent immune-modula- the catalytic function of sPLA2-X was necessary for anti-viral tors and induce immune activation of T cells in the presence activity. Importantly, sPLA2-X was a potent inhibitor of both of monocytes through the ligation of TLR7/8. HIV-1-derived CXCR4- and CCR5-tropic HIV-1 replication in T-cells. TLR ligands may therefore have a role in the design of adju- Conclusions: sPLA2-X plays an important role in the innate vants and may serve as potent immune-modulators enhanc- immune response against lentiviruses and possibly other ing immunogenicity of vaccines. viruses. The induction of this activity represents an effector component that may be useful to elicit as a microbicide and as a vaccine-induced response to HIV. 84 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 85 Amsterdam, Netherlands 29 August–1 September 2006 P03-07 Robust activation of NK cells by single stranded RNA derived from HIV-1 G Alter, N Teigen, T Suscovich, A Meier, H Streeck and M Altfeld Massachusetts General Hospital, Boston, MA, USA Hypothesis: Chronic viremic HIV-1 infection is characterized by persistent immune activation of several lymphocyte sub- sets, including NK cells, but the precise mechanisms of HIV- 1-induced immune activation are not well understood. Given the important nature of NK cells in the innate immune response to viral infection, we were intrigued in this study, to elucidate the role of Toll-like receptor expression on NK cells on NK cell activation in acute HIV-1 infection. Objective: Here we assessed the ability of TLR7/8 ligands, including HIV-1 RNA, to activate NK cells in vitro. Methods: TLR mRNA expression was assessed by RT-PCR, protein was detected by Western Blot, and NK cell activity was monitored by multiparameter flow cytometry. Results: Purified NK cells expressed TLR mRNA, and were significantly activated by stimulation with HIV-1-derived uri- dine-rich single stranded RNA, but not by control RNA. NK cell activation by HIV-1 single stranded RNA depended on the presence of and direct cell-to-cell contact with CD14+ monocytes, indicating a critical role of accessory cells for ssRNA40-mediated NK cell activation, and resulted in an increased capacity of NK cells to respond to MHC devoid target cells. Conclusion: Taken together, these data demonstrate that NK cells are significantly activated in the presence of CD14+ monocytes by TLR ligands, including ssRNA derived from HIV-1 LTR. These data suggest that HIV-1 derived TLR lig- ands contribute to the immune activation of NK cells observed during chronic viremic HIV-1 infection P03-08 [WITHDRAWN] Specific recognition of an HIV-1 CTL escape mutation by dendritic cells leads to functional inhibition via ILT4 D Kavanagh1, M Lichterfeld1, S Mui1, K Williams1, R Sivamurthy1, T Miura1, R Allgaier1, F Pereyra, E Rosenberg1, R Gandhi1, T Allen1, M Altfeld1, B Walker1,2 and X Yu1 1 Partners AIDS Research Center, Massachusetts General Hospital, and Harvard University Center for AIDS Research, Boston, MA, USA; 2 Howard Hughes Medical Institute, Chevy Chase, MD, USA AIDS Vaccine 2006 85 Poster sessions_revGJ 14/8/06 16:34 Page 86 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 4: MUCOSAL IMMUNITY P04-01 lated with CT and in those in which received 100 µg of DNAIL-12. Conclusion: Mucosal DNAIL-12 administration at prim- Intranasal delivery of DNA IL-12 during ing resulted effective to improve the immunogenicity of mucosal DNA-prime/MVA-boost immunization HIV antigens in DNA/MVA schemes after the vectors are schemes enhanced the cellular immune response inoculated by intranasal route. These results are of signifi- against HIV-1 Env antigen cance due to the need to improve mucosal vaccine strate- gies against HIV. AM Rodríguez1, D Mónaco1, H Salomón1, M Esteban2 and MM Gherardi1 P04-02 1 Argentinean National Reference Center for AIDS, University of Buenos Aires, Buenos Aires, Argentina; 2 Department of Secreted heat shock protein gp96-Ig induces Molecular and Cellular Biology, Centro Nacional de mucosal immune response Biotecnologia, CSIC, Madrid, Spain 1,2 1,2 1,2 1,2 Background: As the major route of HIV transmission is ER Podack , N Strbo , M Kolber and S Pahwa through the exposure of mucosal surfaces to the virus, designing of immunization regimes aimed to induce 1 University of Miami Miller School of Medicine, Miami, FL, mucosal immune responses is highly needed. USA; 2 Center for HIV Research, University of Miami, Miami, Objectives: To analyse the mucosal action of IL-12 alone FL, USA or in combination with the subunit B of cholera toxin (CTB), applied in DNA-prime/MVA-boost intranasal Background: Heat shock protein gp96 is an important immunization (in) schemes. endoplasmic reticulum (ER) chaperone for peptides and Methods: Different groups of Balb/c mice were immunized proteins. A fraction of the gp96-associated peptides are by intranasal route with a heterologous DNA prime-MVA trimmed and loaded onto MHC I for antigen presentation boost regimen with both vectors expressing the HIV- Env to CD8 cells. By replacing the KDEL retention signal of antigen from subtype B. Some groups of mice received human gp96 with the IgG1-Fc portion we have generated DNAIL-12 (50 µg or 100 µg) at priming, in the presence or a gp96-fusion protein (gp96-Ig) that is secreted from trans- absence of 50 µg of CTB applied at prime and booster fected cells. Gp96-Ig secreted from transfected tumor cells doses. Both immunization doses were spaced by 14 days in vivo mediated strong, antigen specific CD8-CTL expan- and 10 days after boosting, mice were sacrificed and the sion and caused tumor rejection. different samples obtained to evaluate the cellular immune Objectives: To evaluate the effect of gp96 secretion by response generated. transfected syngeneic tumor cells (EG7-gp96-Ig) and allo- Results: Groups of mice in which DNAIL-12 was co- geneic cell line (NIH3T3-OVA-gp96-Ig) on the gut mucos- administered with the DNAenv vector showed the highest al immunity. specific cellular immune response in the spleen; as mea- Methods: We measured induction and migration of antigen sured by IFN-γ secreting cells in an ELISPOT assay, and by specific CD8 T cells (using ovalbumin and adoptively the levels of this cytokine in supernatants after specific re- transferred gfp-OT-I responder cells in the C57 Bl6 mouse stimulation. This enhancement appears to be dose-depen- model system). After intraperitoneal (i.p.) injection of dent since in the group in which DNAIL-12 dose was of gp96-Ig secreting cells, the OT-I response was measured in 100 µg the response was four-fold higher than in control the peritoneal cavity (PEC), spleen (SPL) and gut mucosa. mice: (DNAenv-MVAenv), while the enhancement The small intestine was removed, peyer’s patches (PP) dis- observed was two-fold superior when 50µg of DNAIL-12 sected, intraepithelial lymphocytes (IEL) obtained by was applied. The response generated in groups inoculated 1×HBSS/HEPES bicarbonate/EDTA treatment and lamina with either 50 µg or 100 µg was equivalent or even higher propria lymphocytes (LPL) by collagenase treatment. than that induced in mice co-inoculated with cholera toxin Results: Secreted gp96-OVA complexes recruited OT-I (CT) in both prime and booster doses. Co-inoculation of cells in all 3 mucosal compartments 5 days after EG7- CTB with DNAIL-12 generated a reduction, or a similar gp96-Ig immunization (2.7±1 SEM% of the CD8 cells in response, to that induced in its absence depending on the PP were gfp-OT-I, 3.8±1 SEM% in I.E. and 12.5±4.8 SEM dose of DNAIL-12 co-administered. Significant responses % in LP), but not after EG7 injection (less than 0.3%). at both genito-rectal and regional draining lymph nodes Similar results were obtained after injection of the allo- were detected in the group which received the higher IL-12 geneic cell line, NIH-3T3-OVA-gp96-Ig. After gp96-Ig dose. When specific IL-2 secretion was also determined, immunization, OT-I recruited to PP, I.E. and LP compart- significant responses were generated in animals co-inocu- ments expressed significantly higher levels of α4β7, 86 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 87 Amsterdam, Netherlands 29 August–1 September 2006 αEβ7/CD103 and CCR9 compared to OT-I before transfer. In addition mucosal OT-I exhibited effector memory phe- notype (CD62low/CD44high/CCR7low) and also expressed granzyme B. The IFNγ expression was highest in OT-I within PP and LP while OT-I in intraepithelial com- partment expressed low levels of IFNγ. Conclusions: Our results indicate that cell-secreted gp96- Ig induces both, mucosal and systemic CD8-CTL based immunity. Gp96-Ig induces migration of effector memory cells equipped with cytotoxic molecules into the mucosal compartment by inducing selective expression of gut-hom- ing markers. AIDS Vaccine 2006 87 Poster sessions_revGJ 14/8/06 16:34 Page 88 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 5: T CELL IMMUNITY P05-01 P05-02 Capacity of gag m-RNA electroporated dendritic HLA molecules that preferentially target the p24 cells to induce potentially protective HIV-specific gag protein are associated with slow progression T-cell responses in vitro to AIDS G Vanham1,2, E Van Gulck1, L Heyndrickx1 and JAM Borghans1 , C Kesmir2,3 and RJ de Boer2 ZN Berneman2 1 Department of Immunology, University Medical Centre 1 Institute of Tropical Medicine, Dept. of microbiology, Antwerp, Utrecht, Utrecht, The Netherlands; 2 Theoretical Biology, Utrecht Belgium; 2 Antwerp University Hospital, Dept. of Medicine, University, Utrecht, The Netherlands; 3 Centre for Biological Antwerp, Belgium Sequence Analysis, Technical University of Denmark, Copenhagen, Denmark Objectives: Human immunodeficiency virus type 1 (HIV-1) infection is characterized by dysfunction of HIV-1-specific T- Background: Human leukocyte antigen (HLA) molecules lymphocytes. In order to suppress the virus and delay evolu- strongly affect the rate of HIV-1 disease progression. Recent tion to AIDS, antigen-loaded antigen-presenting cells, studies have suggested that rare HLA molecules are most pro- including dendritic cells (DC) might be useful to boost and tective, because HIV-1 has evolved to escape presentation by broaden HIV-1-specific T-cell responses. the most common HLA alleles in the human population. Methods: Monocyte-derived DC from 8 untreated (“naive”) Objectives: We aimed to test the alternative hypothesis that dif- and 10 highly active antiretroviral therapy (HAART)-treated ferences in the tendency of HLA molecules to bind conserved HIV-1-infected patients were electroporated with codon-opti- parts of the HIV-1 genome explain the association between mized (“humanized”) mRNA coding for consensus HxB-2 HLA molecules and the risk of HIV-1 disease progression. (hHXB-2) Gag protein. These DC were co-cultured for 1 Methods: To test this hypothesis, we analysed the HLA- week with autologous peripheral blood leukocytes (PBL). restriction of all known CTL epitopes from the HIV-1 data- Potential expansion of specific T-cells was measured by com- base at Los Alamos. Additionally, we predicted the paring ELISPOT responses of PBL before and after co-cul- HLA-binding affinity of all 9-mer peptides from the consen- ture, using a pool of overlapping peptides, spanning the sus sequence of HIV-1B and HIV-1C for 31 different HLA HxB-2 Gag. molecules, to investigate whether there are intrinsic differ- Results: Expansion of specific PBL after co-culture was noted ences between HLA molecules in their affinity for peptides for T cells producing IFN-γ, IL-2 and perforin (Wilcoxon from the different HIV proteins. signed ranks test P<0.05, except for perforin in naive Results: By analyzing the HLA-restriction of known HIV-1 patients). Examining purified CD4 and CD8 T cells after co- epitopes we found that the most protective HLA molecules, cultures revealed that HxB2 peptides induced IFN-γ in both including HLA-B57 and HLA-B27, preferentially target the subsets, IL-2 was only secreted by CD4+ T-cells and perforin p24 gag capsid protein. In line with this finding, the predict- was dominantly secreted by CD8+ T-cells. Remarkably, the ed HLA-binding affinity of epitopes from p24 for different perforin respons in the treatment naive persons was nega- HLA molecules appeared to be inversely correlated with the tively correlated with the peripheral blood CD4 T-cell count relative hazard of HIV-1 disease progression conferred by (r=-0.8, P<0.05) and the IL-2 response in the HAART treat- those HLA molecules. Similar results were found when the ed patients was positively correlated with the CD4 T-cell analyses were based on ancestral HIV-1 sequences. count (r=0.6, P<0.05). Conclusion/Discussion: HLA molecules that preferentially Conclusions: DC from HAART treated and therapy naive target the p24 protein confer best protection against HIV dis- subjects electroporated with hHxB-2 gag mRNA have the ease progression. This difference between protective and non- capacity to induce secondary T-cell responses. In an earlier protective HLA alleles is not due to a recent adaptation of study we already demonstrated that ex vivo CD4 and CD8 T- HIV-1 to its new human host, because similar results were cells from non treated HIV-infected subjects can be directly found using ancestral HIV-1 sequences. Since p24 is one of triggered by DC electroporated with autologous mRNA[1]. the most conserved parts of the HIV genome, we propose Taken together, our results open the intriguing perspective to that HLA molecules conferring relatively good protection to develop a patient-specific immunotherapy for HIV-1. HIV disease progression do so by targeting the most con- 1. Van Gulck et al. Efficient stimulation of HIV-1-specific T strained - and thereby conserved - parts of the HIV-1 genome, cells using dendritic cells electroporated with MRNA encod- thereby leaving HIV-1 little chance to successfully escape ing autologous HIV-1 Gag and Env proteins. Blood 2006; from the immune response without severe viral fitness loss. 107:1818–1827 88 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 89 Amsterdam, Netherlands 29 August–1 September 2006 P05-03 P05-04 The kinetics of cytolysis by SIV Gag specific CD8+ Statistical determination of threshold for T cell T cells is markedly different between vaccination division in the CFSE-labeling assay and infection X Jin2, D Liu1, J Yu2, R Reichman2 and H Wu1 E Rollman1, MZ Smith1, B Zuber2 and SJ Kent1 1 Departments of Biostatistics and Computational Biology, and; 1 Department of Microbiology and Immunology, University of 2 Medicine, University of Rochester School of Medicine and Melbourne, Melbourne, Australia; 2 Mabtech AB, Nacka, Sweden Dentistry, Rochester, New York, NY, USA Background: The magnitude of HIV/SIV specific cytotoxic T Background: Proliferative capacity has become an essential cells (CTLs) is associated with the outcome of infection. Still, measurement of T cell immune responses elicited by candidate the frequencies of Interferon γ (IFN-γ) and/or tetramer posi- HIV vaccines. The combination of flow cytometer and car- + tive HIV/SIV specific CD8 T cells cannot be directly corre- boxyl fluorescent succinimidyl ester (CFSE) labeling techniques lated with a delay in disease progression. We hypothesise that has been widely used in the study of T cell proliferation: mea- an estimate of the cytolytic capacity of CTL clones would be surement of the percentage of proliferated cells; and, the num- + of greater value when assessing the role of CD8 T cells in ber of cell divisions undergone by proliferated cells. HIV/SIV infection and when evaluating vaccine responses. Objectives: To determine the smallest number of cell division Objectives: To evaluate the cytolytic phenotype of SIV Gag that represents true cell division rather than experimental + CD8 T cells in vaccinated/infected non-human primates variation, and to define a threshold that separates true cellu- using a combined tetramer/intracellular staining (Tet/ICS) lar proliferation from experimental variation. assay. Further, to examine the kinetics of activation (IFN-γ) Methods: We performed a large number of replicate CFSE and effector molecules (Granzyme B and the degranulation labeling experiments using polyclonal stimulation of primary + marker CD107) in these CD8 T cell populations and to human peripheral blood mononuclear cells, obtained the per- analyse the relationship between cytolytic profiles of CTLs centages of proliferated cells using ModFit software, and and disease progression. then analysed these data using several statistical methods. Methods: Blood samples were drawn from two vaccinated Results: Our results indicate that the threshold of prolifera- (DNA/poxvirus) and four naive pigtail macaques (Macaca tion lies between 0.071% (95% confidence) and 0.114% nemestrina) before and after intravenous challenge with the (99% confidence) of total CFSE-labelled cells under our lab- highly pathogenic SIVmac251. Fresh whole blood was co- oratory conditions. cultured with labelled anti-CD107 antibody following re- Conclusions/Discussion: We offer our methods for other + stimulation with the immunodominant CD8 epitope (KP9, investigators to calculate a threshold in their own CFSE- SIV Gag ) or media alone. After 0, 0.5, 1, 3 and 5 h of in 164-172 labeling experiments, as part of their standard operating pro- vitro re-stimulation the cells were stained with the MHC cedures for data analysis. class I specific tetramer (Mane-A*10-KP9) in combination with conjugated antibodies directed towards intracellular IFN-γ and Granzyme B and analysed by flow cytometry. Results: DNA/poxvirus vaccination induced tetramer positive P05-05 (Tet+) CD8+ T cells capable of only slow production (after 3h) of IFN- and with low frequency ( 5% cells) of CD107 γ < Analysis of HLA-A*1101-restricted Nef73-specific expression accompanied by poor Granzyme B release. Upon acute SIV infection, the kinetics of the Tet+ CD8+ T cell cytol- CTLs with a strong ability to suppress HIV-1 ysis phenotype changed dramatically with a rapid (within 30 replication min) antigen-specific up-regulation of IFN-γ and with high + + frequencies (>50%) of Tet CD8 T cells expressing CD107 H Koizumi1, M Fujiwara1, Y Kawashima1, T Ueno1, and with robust Granzyme B release. S Oka2,3, M Takiguchi1 Conclusion: SIV infection generates CD8+ T cells with much greater killing capacity compared to CD8+ T cells induced by 1 Division of Viral Immunology and 2 Division of Infectious prime/boost vaccination. The delayed killing kinetics exhibit- ed by vaccine-induced T cells may limit the ability of T cell- Disease, Center for AIDS Research, Kumamoto University, based HIV vaccines to control acute infection following a Kumamoto, Japan and 3 AIDS Clinical Center, International pathogenic lentiviral exposure. Medical Center of Japan, Tokyo, Japan Background: HLA-A11 is associated with slow progression to AIDS. It was also frequently found in highly exposed to HIV-1 but remain persistently seronegative individuals. Nef73-specific CTLs were frequently detected in these indi- viduals, implying that Nef73-specific CTLs can effectively AIDS Vaccine 2006 89 Poster sessions_revGJ 14/8/06 16:34 Page 90 Amsterdam, Netherlands 29 August–1 September 2006 suppress HIV-1 replication and provide a strong immune ART interruption (STI) would lead to the induction of pressure in HIV-1-infected individuals. responses of potential clinical benefit. Objectives: We investigated whether Nef73-specific CTLs Methods: HIV+ individuals with <200 CD4 T-cells/µl were effectively suppress HIV-1 replication in vitro and involve in initiated on ART, those attaining 300 CD4 T-cells/µl after 52 suppression of HIV-1 replication in vivo. or 76 weeks of ART were randomized into either STI (12 Method: CD4+ T cells, derived from HLA-A*1101+ healthy weeks on then 12 weeks off) (n=7), or CT (n=9) arms. HIV- donor, were infected with HIV-1 strain (NL-432 or NL- induced CD3+CD8+IFN-γ+/perforin+ and CD3+CD4+IFN- M20A: NL-432 Nef mutant strain which dawn-regulate CD4 γ+/IL-2+ release after stimulation with Gag (clades A and D), but not HLA-class I). HIV-1-infected CD4+ T cells were Nef, Vif, Rev, Tat, Vpr and Vpu peptide pools were evaluat- cocultured with HIV-1-specific CTL clones. The concentra- ed at 0 and 12 weeks of STI using intracellular cytokine stain- tion p24 Ag in the culture supernatant were measured by ing. CD4 T-cell counts were also enumerated at this point. A ELISA. Nef73-specific CD8+ T cells were detected by ex vivo positive response was determined as 2 confidence intervals tetramer assay in 20 HLA-A*1101+HIV-1-infected donors above mean background. while Nef73-specific memory CD8+ T cells were measured Results: Each of the 8 peptide pools were recognised by after Nef73 peptide stimulation for 2 weeks in vitro. We CD3+CD8+/CD4+ T-cells from at least one patient at pre-ART sequenced the part coding Nef73 epitope from plasma HIV-1 baseline with 78% (62/80) IFN-γ+, 49% (39/80) perforin+ RNA or provirus in 20 HLA-A*1101+ and 32 HLA-A*1101- and 27% (21/77) IL-2+. Staphylococcal enterotoxin B-induc- HIV-1-infected donors. tion of perforin significantly increased after 52 or 76 weeks Result: Nef73-specific CTL clones strongly suppressed HIV-1 of ART (P<0.0001), but no SEB-induction of CD3+CD8+ replication whereas HLA-A*1101-restricted Gag-specific IFN-γ+ or CD3+CD4+IFN-γ+ was observed. Concurrently, clone only partially suppressed it. Nef73-specific CD8+ T cells HIV-specific IFN-γ and perforin responses diminished follow- were detected ex vivo in 12 of 20 HLA-A*1101+ HIV-1- ing ART. After 12 weeks of STI, HIV-specific IFN-γ and IL-2 infected donors while the specific memory CD8+ T cells were responses were further diminished. Responses also dimin- found in 18 of 19 donors. There were no mutations on this ished in the CT arm sampled at the same time-point. epitope between HLA-A*1101+ and HLA-A*1101- donors. Preliminary data from a subset of patients revealed a signifi- No relation between viral load and the frequency of Nef73- cant enhancement of HIV-specific perforin breadth in the STI specific CD8+ T cells were not found in ART-free donors. arm: (5/7) STI participants responded to 4 or more peptide Conclusions: Nef73-specific CTLs had a strong ability to pools compared to (1/9) in the CT arm (P=0.035, Fischer’s suppress HIV-1 replication in vitro and they were high fre- Exact). These responses did not correlate with age or baseline quently detected in HLA-A*1101+ HIV-1-infected donors. CD4 counts at randomization. However, there was no evidence that they suppress HIV-1 in Conclusion: This first round of STI was associated with appar- vivo. These results suggest that Nef73-specific CTLs may be ent broadening of HIV-specific CD3+CD8+perforin+ responses, involved in suppression of HIV-1 replication, but they alone but did not regenerate HIV-specific CD3+CD4+ T cell respons- can not control HIV-1 replication in vivo. es. STI randomization in this study has now been discontinued due to negative results from larger STI trials. These prelimi- nary results suggest that it would have been useful to further evaluate the long-term immunological effects of STI. P05-06 Structured treatment interruption does not regen- erate HIV-specific IFN-g and IL-2 production but P05-07 enhances broader HIV-specific perforin responses in a Ugandan population CD8 T cell recognition of multiple epitopes within two regions of Gag is associated with the mainte- J Serwanga1 , P Kaleebu1, S Mugaba1, B Auma1, C Gilks3, nance of a low steady-state viremia in HIV-1 P Munderi1, H Grosskurth1, F Gotch2 and The DART trial seropositives team C Geldmacher1, JR Currier3, E Herrmann4, M Gerhardt1, 1 1 3 3 1 MRC/UVRI Unit On AIDS In Uganda, C/o Uganda Virus A Haule , L Maboko , D Birx , F McCutchan , 4 3 2 Research Institute, Entebbe, Uganda; 2 Department of A Meyerhans , J Cox and M Hoelscher Immunology, Imperial College, Chelsea and Westminster Hospital, London, UK; 3 World Health Organisation 1 Mbeya Medical Research Programme, Referral Hospital, Mbeya, Tanzania; 2 Institute for Tropical Medicine and Infectious Background: Antiretroviral therapy (ART) delivery strategies Diseases, University of Munich, Munich, Germany; 3 The US that maximise clinical benefit despite limited drug availabili- Military HIV Research Program, Rockville, USA; 4 Faculty of ty are essential in resource poor settings. We hypothesised Medicine, Saarland University, Homburg/Saar, Germany that continuous ART (CT) would not allow regeneration of HIV-specific cellular immune responses, whilst structured Background: The importance of HLA class I restricted CD8 90 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 91 Amsterdam, Netherlands 29 August–1 September 2006 T cell response in the control of HIV infection is generally infection in order to monitor responses to vaccination accepted, yet few studies have shown a correlation of the Objectives: To investigate if in vitro presentation of peptides HIV-specific CD8 T cell response with markers of HIV dis- representing a multigene multiclade plasmid DNA construct ease progression or the expression of “protective” HLA class intended for therapeutic vaccination will give rise to subtype I alleles, such as B5801, B4201 or B8101. specific or cross reactive responses in HIV-1 patients infected Objective: To study the influence of the hosts’ genetic back- with different subtypes. To characterize immunodominant ground on the HIV-specific CD8 T cell response and on viral regions within Gag p24 and investigate how the responses are control. linked to the patient’s HLA-type. Methods: The study was conducted in the region of Mbeya, Methods: Blood samples from 60 HIV-1 individuals infected South-West Tanzania in the framework of an HIV superin- with several subtypes of HIV-1 and at different stages of HIV- fection study (HISIS) that has followed a high risk cohort 1 infection were investigated. These included patients with over 4 years in 3-monthly intervals. During the 4th year of the subtypes A, B, C, D, G, CRF_01AE and CRF_02AG infec- study, HIV specific CD8 T cell responses against Gag, Nef tion. Here we report the results of IFN-γ ELISpot on fresh and Env were assessed in 56 chronically HIV infected indi- cells. Peptide pools representing HIV-1 p17B, p24A/B, viduals using an IFN-γ ELISPOT assay in combination with gp160A/B/C/D and gp41B were used for response evaluation. intracellular cytokine staining. Fresh peripheral blood Characterization of immunodominant regions within the Gag mononuclear cells were stimulated with 15-mer peptides of p24 region was assessed by a peptide matrix approach. subtypes A, C and D to determine the magnitude and breadth Results: The cross-reactivity between Gag p24 A and B was of the peptide-specific responses. Simultaneously plasma viral strong in patients infected with all the various HIV-1 sub- load and CD4 T cell counts were determined and HLA class types. In total 47 out of 60 (78%) patients responded by I alleles were typed. interferon γ secretion to peptides of Gag p24 subtypes A or B. Results: Expression of the HLA class I alleles B5801, B8101 In 87% of the responders, cross-reactivity occurred. The and B0702 was associated with a low median viral load and overall reactivity to peptides of gp120 was low and there simultaneously with a broader median recognition of Gag were no significant differences between the reactivities peptides compared to all other HLA alleles (2 fold; towards the different variable parts of Env. An immunodom- P=0.0035). We further found an inverse linear relationship inant region within the p24 was detected and the majority of between the number of Gag epitopes recognized and the plas- the patients cross-reacted within this region to peptides both ma viral load (R=-0.36, P=0.0016), whereas responses direct- from p24A and B library. The HLA sequences, not yet avail- ed against Nef or Env had no apparent effect on viral control. able in detail, may possibly aid in explaining peptide presen- Particularly, multiple targeting of two regions within Gag tation in some of the non-Gag reactive individuals. (aa001-aa075 and aa248-aa500, GagR1R3) was associated Conclusions: Gag responses clearly dominate the late natural with the maintenance of a low steady-state viremia even years T-cell response in individuals infected with various subtypes. after acute infection. Gag genes representing only a few circulating strains may Conclusions: The breadth of Gag-specific epitope recogni- therefore act to induce immunity to a very broad range of the tion, particularly within GagR1R3, is important for main- HIV-1 clades circulating world wide. tained control of viral replication during the chronic phase of HIV infection and is in part dependent on the expressed HLA class I alleles. P05-09 HIV-2 differs from HIV-1 in the ability to infect P05-08 dendritic cells and to induce polyfunctional CD4+ and CD8+ T cell responses Cross-reactive cellular immune responses in HIV-1 patients infected with different clades of HIV-1 MG Duvall1,2, ML Precopio2, DA Ambrozak2, K Loré3, WC Adams2, H Blaak4, JR Mascola2, M Roederer2, 1,2 1 3 L Gudmundsdotter , A Sjödin , B Hejdeman , R Theve- SL Rowland-Jones1,5 and RA Koup2 Palm4, D Bernasconi5, S Butto5, A Alaeus4, K Lidman4 and 1,2 B Wahren 1 MRC Human Immunology Unit, Weatherall Institute of Molecular Medicine, Oxford University, UK; 2 Vaccine Research 1 Swedish Institute for Infectious Disease Control; 2 Center, NIAID, NIH, Bethesda, MD, USA; 3 Center for Infectious Microbiology and Tumorbiology Center, Karolinska Institutet; 3 Medicine, Department of Medicine, Karolinska Institutet, Venhälsan, Department of Infectious Diseases, Karolinska Stockholm, Sweden. 4 Department of Virology, Erasmus MC, University Hospital; 4 Department of Infectious Diseases, Rotterdam, The Netherlands; 5 MRC Laboratories Fajara, The Karolinska University Hospital, Solna, Sweden; 5 AIDS Division, Gambia, West Africa Istituto Superiore di Sanità, Rome, Italy Background: The clinical course of disease with HIV-2 infec- Background: In immunotherapy, it is important to define the tion is attenuated with respect to HIV-1. Host immune AIDS Vaccine 2006 91 Poster sessions_revGJ 14/8/06 16:34 Page 92 Amsterdam, Netherlands 29 August–1 September 2006 responses are thought to play a role in control of HIV-2 Objective: To characterize HIV-1 Nef-specific T cell respons- viremia, and maintenance of HIV-2-specific CD4+ T cell help es in patients infected with clade B or BF URFs and to deter- has been shown to be a defining feature of non-progressive mine the degree of cross-reactivity. HIV-2 infection. Methods: Fourteen HIV-1 recently infected patients (docu- Objectives: To further characterize the quality of the HIV-2- mented seroconvertion within 6 months) were enrolled. specific CD4+ and CD8+ T cell response in comparison to Plasma viral load, viral subtype, CD4 count, CD8 count and HIV-1 and to evaluate potential mechanisms by which HIV- HLA haplotype were determined. T cell responses were 2-specific CD4+ T cells are preserved. assessed in an Interferon-γ (IFN-g) based ELISPOT assay using Methods: We used polychromatic flow cytometry to describe overlapping peptides arrayed in a matrix system. Positive the function of HIV-1 and HIV-2 specific CD4+ and CD8+ T responses were further confirmed at the single peptide level. cells in 16 HIV-1+ and 18 HIV-2+ Gambians. We simultane- Statistical analysis was performed based on Spearman rank ously and independently measured 5 functional parameters correlation, Wilcoxon or Mann-Whitney tests. induced by Gag peptide stimulus: IFN-γ, IL-2, TNF-α, MIP- Results: Subtyping of nef gene was used to group patients. 1β, and CD107a. Additionally, we characterized the ability of Eight out of 14 patients were Nef-B while 6 were Nef-F. primary HIV-2 isolates from non-progressors and a lab-adapt- Strong correlations in IFN-g T cell responses were observed ed HIV-2 strain to infect and induce maturation of human in both patient groups for Nef-B and Nef-F peptide pools CD11c+ myeloid dendritic cells (MDCs) and CD123+ plasma- (0.548 for B patients and 0.771 for F patients). Neither cytoid dendritic cells (PDCs) in comparison to HIV-1 isolates. inter or intra-clade statistical differences were observed for Results: HIV-2-specific T cells are more polyfunctional in total anti-Nef-B or –F responses. However, responses in F nature than are HIV-1-specific T cells. Forty percent of the patients were narrower than in B patients’ (and concentrat- HIV-2-specific CD4+ T cell response was composed of cells ed in the core domain of the protein) but higher in magni- simultaneously positive for 3, 4, or 5 functions. In contrast, tude. Surprisingly, overall entropy of BF sequences was only 15% of the HIV-1-specific CD4+ T cell response was lower than in B sequences. Number of recognized peptides simultaneously positive for 3, 4, or 5 functions. The HIV-2- in B patients ranged from 1 to 6 (mean=2.75) for autolo- specific CD8+ T cell response was also highly polyfunctional, gous peptides and 0 to 4 (mean=2) for heterologous pep- with a particularly prominent population of cells expressing tides. In F patients, positive responses ranged from 4 to 1 four simultaneous functions. Notably, unlike with R5 isolates for autologous and heterologous peptides (means=1.83 and of HIV-1, MDCs were not susceptible to infection by primary 1.5, respectively). As regards magnitude of response, medi- R5 HIV-2 isolates or an X4 isolate of HIV-2 in vitro, though ans found in B patients were 179 spot forming units these viruses are capable of infecting primary CD4+ T cells. (SFU)/106PBMC (range 50–771) and 112.5 (50–960) for Conclusions: HIV-2-infected individuals mount a functional- autologous and heterologous peptides, respectively. For BF ly superior HIV-specific T cell response characterized by infected patients, medians obtained were 776 highly polyfunctional Gag-specific CD4+ and CD8+ T cells. SFU/106PBMC (range 80–4530) and 1326 (40–4250) for Preservation of HIV-2-specific CD4+ T cells suggests a lack autologous and heterologous peptides, respectively. of preferential infection. Mechanistically, lack of preferential Conclusions: Overall cross-clade T cell responses were infection of HIV-2-specific CD4+ T cells may be explained by found between patients infected with different clades. a limited infection of dendritic cells, and perhaps less trans- Evaluation of responses at the single peptide level allowed fer of virus to antigen-specific CD4+ T cells in the setting of us to determine the existence of inter-clade differences at HIV-2 infection. the frequency of epitopes recognized, structural domains targeted and magnitude of response. P05-10 Epidemiological studies revealed that in Argentina, early predominance of B subtype has been overshadowed by the emergence of BF recombinants. Until correlates of protection are well understood, existence and determinants of cross-clade immune responses should be studied G Turk 1, MM Gherardi 1, N Laufer 2, P Cahn 2, J Cox 3 and H Salomon 1 1 Argentinean National Reference Center For AIDS - University of Buenos Aires; 2 Fernandez Hospital; 3 Henry M Jackson Foundation/US Military HIV Research Program 92 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 93 Amsterdam, Netherlands 29 August–1 September 2006 P05-11 particular HLA-DR alleles. Conclusion: In silico prediction of promiscuous epitopes led to the identification of naturally immunodominant CD4+ T Identification of novel conserved consensus CD4+ cell epitopes recognized by PBMC from a significant propor- T cell epitopes from clade B HIV-1 whole genome tion of a genetically heterogeneous patient population sequences that are frequently recognized by HIV-1 exposed to HIV-1. This combination of CD4+ T cell epitopes infected patients from different disease groups – 11 of them not described before – may have the potential for inclusion in a vaccine against HIV-1, allowing the immu- SG Fonseca1,2,3, A Coutinho-Silva1,2, LAM Fonseca2,3, nization of genetically distinct populations. Supported by AC Segurado3,4, SL Moraes1,5, O Leite4, E Gutierrez4, FAPESP, CNPq. S Santos4, HY Li4, R D’Agnolo4, LK Iwai1, EC Mairena1, H Rodrigues1, J Hammer6, E Kallás3,7, J Sidney8, A Sette8, J Kalil1,2,3, E Cunha-Neto1,2,3 P05-12 1 Heart Institute (InCor), 2 Division of Clinical Immunology and Low CD4 T cell counts despite low viremia: HIV-1 Allergy, School of Medicine, University of São Paulo, São Paulo, infected patients with discordant response to Brazil; 3 Institute for Investigation in Immunology – Millenium HAART versus advanced HIV-2 disease Institute, Brazil; 4 Casa da Aids, São Paulo, Brazil, 5 Institute of Tropical Medicine, University of São Paulo, São Paulo, A Albuquerque, R Soares, R Foxall, C Cortesão, M Garcia, 6 Department of Genomic and Information Sciences, Hoffmann- RMM Victorino and AE Sousa La Roche Inc., Nutley, New Jersey, USA, 7 Division of Infectious and Parasitic diseases, Federal University of São Paulo, São Paulo, Imunologia Clínica, Instituto de Medicina Molecular, Faculdade Brazil; 8 La Jolla Institute for Allergy and Immunology, San de Medicina de Lisboa, Lisboa, Portugal Diego, USA Background: Mechanisms underlying low CD4 T cell recovery in HIV-1 infected patients with suppression of viremia under Background: Experimental HIV-1 vaccines which underwent highly active antiretroviral therapy (HAART) are poorly clinical assessment in humans so far had limited or no suc- understood. Interestingly, HIV-2 infection is also able to reach cess, and progress in epitope constitution may help improve major CD4 depletion in spite of low to undetectable viremia. vaccine efficacy. Aim: Characterize the immunological disturbances associated Objective: To identify promiscuous and potentially protective with major CD4 lymphopenia in these two different scenarios. human CD4+ T cell epitopes in most conserved regions with- Methods: A cross-sectional study of patients with less than in the protein-coding genome of HIV-1 clade B consensus 300 CD4 cells/µl living in Portugal was performed involving sequence. 11 HIV-1 infected patients with viremia below 50 RNA Methods: We used the TEPITOPE algorithm to screen the copies/ml after at least one year of HAART (discordants); 18 most conserved regions of the whole genome of HIV-1 sub- untreated aviremic HIV-2+ patients and 27 untreated viremic type B consensus sequence to identify immunodominant, HIV-1+ patients. These groups were compared to 11 HIV-1+ promiscuous and potentially protective human CD4+ T cell patients with successful immunological and virological epitopes in HIV-1. Synthetic peptides were tested with IFN-γ responses to HAART, and 16 healthy controls. Freshly iso- ELISPOT assays on peripheral blood mononuclear cells lated PBMCs were characterized by 6 parameter flow cytom- (PBMC) from 8 uninfected controls and 32 chronically HIV-1 etry. Circulating IL-7 was measured by ELISA and proviral infected patients: 8 LTNP (more than 8 years with HIV-1 DNA and Foxp3 mRNA by real-time PCR. infection and untreated; >500 CD4+ T cells/µl and low viral Results: Low CD4 counts are associated with major naive load (VL)) and 24 HIV-1-infected patients under anti-retrovi- CD4 and CD8 T cell depletion and T cell activation, irre- ral therapy in different disease stages: 7 reconstituted patients spective of the type of infection or HAART exposure. (>500 CD4+ T cells/µl; VL <10000 HIV-RNA copies/ml), 9 However, in contrast to the untreated HIV-1 and HIV-2 partial controllers (250-500 CD4+ T cells/µl; VL <10,000 patients, the discordants exhibited an enhanced proportion HIV-RNA copies/ml), and eight typical progressors (<200 of CD4 T cells expressing CD25 (IL-2Rα) and, in particular, CD4+ T cells/µl; VL >10,000 HIV-RNA copies/ml).The actual a major expansion of memory CD25bright cells with intracel- promiscuity of HLA binding of the 18 selected peptides was lular CTLA-4. An increase in Foxp3 mRNA levels was also assessed by binding assays to 9 prevalent HLA-DR molecules. documented in discordant patients in comparison with Results: PBMC from 29 out of the 32 HIV-1+ patients (91%) healthy and HIV-1 patients with immunological recovery recognized at least one of the promiscuous peptides, while under HAART. Moreover, discordants had less increase in none of the healthy controls recognized peptides. All 18 pep- circulating IL-7, which is associated with an enhanced tides were recognized, and each peptide was recognized by at expression of intracellular Bcl-2 and a down regulation of least 7 patients; 44% of the patients recognized 5 or more IL7-Rα, suggesting increased IL-7 consumption. peptides, and 19% of patients recognized 10 or more pep- Conclusions: In spite of similar levels of CD4 depletion and tides. Similar responses were obtained in CD8+ T cell-deplet- absence of viremia, a different involvement of IL-7/IL-7Rα ed PBMC, and apparently this response was not associated to and IL-2/IL-2Rα pathways seems to occur in HIV-1 patients AIDS Vaccine 2006 93 Poster sessions_revGJ 14/8/06 16:34 Page 94 Amsterdam, Netherlands 29 August–1 September 2006 with discordant responses to HAART in comparison with P05-14 advanced untreated HIV2 disease. These results suggest an optimization of these pathways in discordant HIV-1 patients HIV-1 C infected Indian subjects recognize cyto- possibly contributing to the absence of CD4 decline and opportunistic infections that is usually observed. Moreover, toxic T lymphocyte (CTL) epitopes conserved these data strengthen the relevance of these pathways for across the HIV-1 subtypes complementary immune-based HIV therapy in order to achieve T cell immunity. MR Thakar, S Bichare, S Suragaonkar, and RS Paranjape P05-13 National AIDS Research Institute, Pune India Myeloid and plasmacytoid dendritic cells are sus- Background: Different HLA specificities seen in different ceptible to recombinant Adenovirus vectors and populations and genetic diversity of HIV poses the challenge stimulate polyfunctional memory T cell responses for development of immunogen that will be effective against divergent HIV-1 strains. HIV-1 CTL epitopes that are con- K Loré1,2, WC Adams1, M Havenga3, ML Precopio1, served across the subtypes may be useful in devising such an L Holterman3, J Goudsmit3 and RA Koup1. immunogen. Response to such conserved epitopes covering heterogeneous MHC class I alleles was studied in asympto- matic HIV-1 infected Indian subjects with CD4 counts above 1 Vaccine Research Center, NIH, Bethesda, MD, USA; 2 Karolinska 500/cmm for two years. Institutet, Stockholm, Sweden; 3 Crucell Holland BV, Amsterdam, Objective: To identify the conserved CTL epitopes recognized the Netherlands by HIV-1 C-infected persons from India. Methods: Epitopes with >80% conservation across HIV-1 Background: While replication-incompetent adenovirus subtypes were identified from Los Alamos database and type 5 (rAd5) is a potent vaccine vector for stimulating T using the MotifScan software program. These included 26 and B cell responses, high seroprevalence of Ad5 within epitopes from Gag; 10 from Pol, 6 from Env; and 8, 3 and human populations may limit its clinical utility. Therefore, 2 peptides from Nef, Vpr and Vif, respectively. CTL alternative adenovirus serotypes have been studied as responses were determined in 19 HIV-1 subtype C individ- vaccine vectors. uals using Interferon-γ secretory ELISPOT responses to a Objectives: In this study, we characterized the ability of rAd5 matrix of 55 peptides. and rAd35 to infect and induce maturation of human Results: Fourteen (74%) participants responded to at least CD11c+ myeloid DCs (MDCs) and CD123+ plasmacytoid one or more HIV peptides, with majority (10/14) responding DCs (PDCs), and their ability to present rAd-derived antigens to at least one of the gag protein (p17 and/or p24). Only 4 of to activate antigen-specific T cells. 10 Pol peptides were recognized. Gp120 and Vpr antigens Methods: MDCs and PDCs were directly sorted from healthy were poorly recognized. Five p24, one gp41, three p17 and donors and exposed to rAd5 or rAd35. Uptake of rAd, phe- one Vif peptide were the most frequently targeted peptides notypic differentiation, cytokine production and stimulation followed by three Nef peptides. Overall, 44 of the 55 select- of autologous antigen-specific T cell responses were mea- ed peptides were recognized by subtype-C infected PBMCs. sured using polychromatic flow cytometry. Fifty percent of the responders recognized more than 8 pep- Results: We found that MDCs were significantly more sus- tides with one responding to as many as 15 peptides. ceptible to both rAd5-GFP and rAd35-GFP than were PDCs. Conclusions: Indian patients exhibited polyclonal response to Both DC subsets were more susceptible to rAd35 than to different HIV proteins, recognizing multiple conserved epi- rAd5. Neutralizing anti-CD46 antibodies effectively blocked topes from Gag, Env, Nef and Vif. This information may pro- rAd35 but not rAd5 infection. In addition, rAd35 induced vide clue to development of immunogen for universal high levels of IFNα in PDCs and phenotypic differentiation in application. both DC subsets. Both MDCs and PDCs exposed to either rAd5 or rAd35 encoding for CMV pp65 were able to present + pp65 and activate CMV-specific memory CD8 T cells in a P05-15 dose dependent manner. MDCs stimulated the highest fre- quencies of pp65-specific T cells. Responding T cells (responding to DCs exposed to either rAd5 or rAd35) Analysis of T cell interferon-γ response to genome expressed multiple functions including degranulation wide peptides in HIV-1 subtype C infection shows (CD107a) and production of IFNγ, IL-2, TNFα and MIP-1β. immuno-dominance of Pol, Gag and Nef antigens Conclusions: Thus, the ability of rAd5 and rAd35 to natural- ly target important DC subsets, induce their maturation, and R Paranjape, S Bichare, S Suragaonkar, M Ghate and appropriately present antigen to T cells, may explain their M Thakar potency as vaccine vectors. National AIDS Research Institute, Pune, India 94 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 95 Amsterdam, Netherlands 29 August–1 September 2006 determine the effectiveness of each anti-HIV CTL specificity. Background: The identification of epitopes recognized in dif- Methods: The HIV-specific CTL response was determined in ferent HIV-1 proteins may help to understand relevance of a cohort of >500 chronically infected subjects using a panel these proteins for inducing immune responses that are help- of 410 overlapping 18mer C clade consensus-based peptides. ful in controlling HIV infection. In order to identify proteins In 579 HAART-naive adult subjects the viral loads (VL) of that are preferentially targeted by immune response, genome subjects responding to each of the 160 defined epitopes was wide search was made using overlapping peptides. compared with the VL of subjects sharing the relevant HLA Methods: T cell Interferon-γ response was measured in 31 but not responding to that epitope. Full-length virus sequenc- HIV infected individuals (17M/14F, mean age 34 years) ing was undertaken in >200 subjects to determine selection using peptide matrix ELISPOT representing the entire pressure on the virus mediated by all 160 CTL responses. genome through 681 overlapping peptides. Ten patients had Furthermore, a novel statistical method was employed to dis- CD4 count <350 cells/cumm and 21 pts had CD4 count tinguish responses driving VL, and those driven by VL. >500 cells/cumm. Results: Gag-specific CTL responses were associated with sig- Results: 28 (90%) of 31 patients responded to at least one of nificantly lower VL than failure to target the same epitopes. the antigen. The magnitude of the response measured as spot In contrast, VL was not reduced by targeting Pol epitopes. forming units (SFUs)/million cells ranged from 140 to 1150 Targeting Env, or Nef/Tat/Rev/Vif/Vpr/Vpu, was associated SFUs/million cells. Breadth of the response was considerably with significantly higher VL than not targeting these epi- high with recognition of 20 to 100 peptides from different topes. No HLA-C-restricted responses were associated with antigens. Twenty-one of the responders recognized peptides effective control of HIV; even Gag-specific HLA-C-restricted from multiple proteins. Pol, Gag and Nef were the most tar- responses were associated with a higher VL (P=0.0051). geted protein eliciting response in 24, 22 and 18 participants HLA-B-restricted Gag-specific responses were most strongly respectively. Env peptides were recognized by 17 of 31 associated with reduced VL (P<0.0001; HLA-A responses: patients. Rev and Tat-specific responses were seen in five and P=0.0196), and HLA-B Env-specific responses most strongly three patients respectively associated with higher VL (P=0.0068; HLA-A: NS). Viral Conclusion: The HIV-1 C infected Indian patients showed sequence and statistical analyses indicate that HLA-B- broad recognition of various antigenic peptides. Accessory restricted Gag-specific drive strong selection pressure, and genes and regulatory genes other than Nef did not elicit sig- Env-specific responses impose minimal/no selection pressure nificant response. The responses were comparatively lower in on HIV. magnitude in patients with CD4 counts <300 cells/mm3. The Conclusions: These data reveal the diversity of HIV-specific immunodominant regions recognized in Pol, Gag and Nef CTL operating in chronic HIV infection: HLA-B-restricted, may be important in context with vaccine development. Gag-specific CTL play the strongest role in successful immune control. The large majority of CTL responses appear to play a prominent passive part that may potentially inter- P05-16 fere with the activity of normally subdominant Gag-specific specificities. Dissection of epitope-specific CTL responses sug- gests a mechanism underlying differential HLA- P05-17 A, -B and -C associations with HIV disease progression Priming of HIV-1 specific CD8+ T cells by PJR Goulder1-3, P Kiepiela1, K Bishop1, A Leslie2, A dendritic cells Rathod3 JI Mullins4, C Brander3, D Heckerman5 and BD Walker1,3 BA Colleton, XL Huang and CR Rinaldo, Jr 1 HIV Pathogenesis Programme, University of KwaZuluNatal, Graduate School of Public Health, University of Pittsburgh, Durban, South Africa; 2 University of Oxford, Oxford, UK; 3 Pittsburgh PA, USA Partners AIDS Research Center, MA, USA; 4 University of Objectives: We hypothesize that priming with HIV-1 antigens Washington, Seattle, USA; 5 Microsoft, USA expressed by mature Th1 polarizing dendritic cells (DCs) is key to inducing strong CD8+ T cell immunity to HIV-1 infec- Background: HLA class I associations with particular rates tion. Therefore, we have developed an in vitro model for of AIDS progression provide a clue to mechanisms under- assessing priming of CD8+ T cells by DCs to HIV-1. lying HIV immune control. The principal role of HLA class Methods: DCs were derived from CD14+ monocytes of HIV- I molecules to present viral peptides to CTL for recognition 1 negative donors by culture in IL-4 and GM-CSF for 5 days. and elimination of virus-infected cells suggests that the par- The immature DCs were treated with (A) IFNγ/poly-I:C/IL- ticular epitopes targeted by CTL strongly influence HIV 1β/TNFα/IFNγ cocktail, (B) trimeric recombinant CD40L disease outcome. (Amgen), or (C) CD40L+IFNγ and examined for production Objectives: To define the significant CTL responses generat- of IL-12 (p40/p70), IL-15, IL-10, and IL-2 by ELISA. The ed in a large cohort of HIV-infected subjects from Durban, phenotypic expression of DC maturation was examined by South Africa, and to employ an epitope-specific analysis to AIDS Vaccine 2006 95 Poster sessions_revGJ 14/8/06 16:34 Page 96 Amsterdam, Netherlands 29 August–1 September 2006 flow cytometry. ELISPOT and tetramer assays were used to (MVA) were used to evaluate HIV-specific responses in assess autologous CD8+ T cell specificity and priming to comparison to peptide pools. known HLA A*0201 HIV-1 CTL epitopes. Results: None of 31 HIV+ patients had vaccinia-specific Results: Primary, HIV-1 specific CD8+ T cells as determined TEM. Among 10 HIV+ patients tested, the only one dis- by IFNγ production and tetramer reactivity were routinely playing PPD-specific-TEM was the only one to have a his- induced by stimulation of naive T cells for 4 weeks with DCs tory of tuberculosis. Five of 36 HIV+ patients and 6 of 9 matured with CD40L + IFNγ, CD40L and the cocktail, in HIV+ patients displayed vaccinia- and PPD-specific TCM, descending order. Optimal priming of CD8+ T cells required respectively. By contrast HIV donors maintained vac- neg coculture with autologous CD4+ T cells. The CD40L+ IFNγ cinia-specific TEM and TCM (27% and 60%, respective- modulated DCs produced the greatest amounts of IL-12. We ly). HIV-specific responses were mainly mediated by TEM are currently examining whether IL-12 polarized Th1 cells (87% responders) compared to TCM (45% responders) help stimulate these de novo antigen specific CD8+ T cell when stimulated by HIV-r-Copenhagen and r-MVA that responses. were able to present HIV antigens to CD8 T cells in vitro. Conclusions: DCs from HIV-1 negative persons can be engi- Conclusion: CD4 immunodeficiency strongly affects persis- neered to prime naive CD8+ T cell responses to HIV-1 anti- tence of TEM and TCM to live vaccines administered in gens. This model is being used for T cell activation by DCs childhood. Effector memory T cells to such vaccines disap- loaded with autologous HIV-1 peptides and infected apop- pear during HIV infection and are not restored under totic cells, leading to a clinically relevant immunotherapy. HAART in absence of antigen exposure while circulating antigens from Mycobacteria species might allow higher level of persistence of mycobacteria-specific TCM com- pared with vaccinia-specific TCM. The almost complete P05-18 disappearance of vaccinia-specific T cells in HIV+ patients and the ability of HIV-r-vaccinia to stimulate HIV-specific Maintenance of antigen-specific TEM and TCM response underline the feasibility of using r-vaccinia as HIV according to antigen exposure in HIV-infected vaccine. patients: comparison of vaccinia-, PPD- and HIV- specific immune responses P05-19 B Puissant1, B Verrier2, N Wincker3, H Ait-Mohand3, C + - - + Katlama3, B Combadière1 and B Autran1 The IFN-γ-producing CD8 CD45RA CCR7 CD27 T cells specific for HIV-1 Gag but not for Nef define 1 Laboratoire d’Immunologie Cellulaire et Tissulaire, INSERM U – immune correlates of protection CHU Pitié-Salpétrière, Paris, France; 2 ERE CNRS bioMerieux, IFR W Lu1,4, A Schnuriger1, D Costagliola2, C Blanc1, B Da 128, Lyon, France; 3 Service de Maladies Infectieuses, CHU Pitié- Silva1, M Almeida1, C Rouzioux3, B Autran1 and the ALT Salpêtrière, Paris, France study group Objectives: The impact of CD4 T cell immunodeficiency on 1 INSERM U543, University of Pierre and Marie Curie,Pitie- the maintenance of immunity to vaccines administered in Salpetriere Hospital, Paris, France; 2 INSERM U720, Pitie- childhood has not been assessed yet. The end of vaccina- Salpetriere Hospital, Paris, France; 3 Lab of Virology, Necker tion against smallpox in 1980 and its eradication allowed investigation of long-term immune memory in absence of Hospital, Paris, France; 4 Department of Infectious Diseases, antigen while BCG might be followed by distinct patterns Pekin Union Medical College Hospital, Beijing, China of exposure to mycobacterial antigens. We analysed the maintenance of antigen-specific T cells (TEM Objectives: The mechanisms for protection of Non effector/memory and TCM central/memory T cells) in Progressors (NP) against HIV are still not elucidated Our treated HIV-infected patients in 3 models of antigen expo- study aims to compare the CD8 immune responses specific sure: live vaccinal antigens administered in childhood fol- for two immunodominant antigens, HIV-1 Gag and Nef, and lowed by reexposition (BCG/PPD) or not (vaccinia), and to explore immune correlates of protection against disease antigen responsible for persistent infection (HIV). We eval- progression. uated the capacity of HIV-recombinant vaccinia to re-stim- Methods: Frozen samples (n=55) from the French NP cohort ulate HIV-specific T cells in vitro. (ALT) were tested by ELISpot using pools of 15-mer Gag and Methods: Thirty-six HIV+ patients under HAART with Nef peptides. Intra-cellular cytokine (IFN-γ and IL-2) stain- CD4 T cells >350/mm3 and VL<200 copies/ml and 11 ing on 14 samples with simultaneous positive responses to HIV controls both vaccinia-vaccinated in childhood were Gag and Nef, combined with anti-CD27, CD45RA and/or neg studied. Vaccinia- and PPD-specific T cell responses were CCR7 stainings were performed by five-colour flow cytome- evaluated in ELISpot-IFN-γ (TEM) and proliferation assays try. Statistical analysis was done using nonparametric (TCM) using wild-type replicative Copenhagen vaccinia Spearman’s rank correlation and Wilcoxon tests and correla- and PPD. HIV-r-Copenhagen and -attenuated vaccinia tions with proviral DNA loads, HIV-1 plasma RNA loads and CD4 counts were studied. 96 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 97 Amsterdam, Netherlands 29 August–1 September 2006 Results: ELISpot frequencies of Gag-specific CD8 T cells from Kwazulu Natal, South Africa, by IFN-γ ELISpot assay. correlated with provirus DNA content but not the Nef-spe- These screens were also used to determine the frequency, cific ones on the whole 55 samples group. ICS revealed no magnitude and immunodominance of the CTL responses difference between proportions of Gag and Nef-specific restricted by these HLA. The in vivo efficacy of these CTL CD8 cells producing IFN-γ alone, IL-2 alone or IFN-g plus was determined by examining their ability to select for IL-2. The CD8+ IFN-g-producing T cells specific either for escape mutation within the promiscuous epitopes. Peptide Gag or Nef were predominantly CD45RA-CCR7-CD27- titration and TcR sequencing were employed to examine (74.9% ±13.8% and 80.3% ±10.3%), and this major sub- the effect of HLA restriction element on functional avidity set of IFN-γ-producing cells showed no correlation with and TcR usage. markers of disease progression. Interestingly, the CD27+ Results: Substantial differences were observed in the ability fractions of Gag-specific CD8+CD45RA-CCR7+ and of CTL to select for escape mutation when targeting the same CD8+CD45RA-CCR7- IFN-γ-producing cells were higher epitope but restricted by different HLA. This observation was than those specific for Nef (P=0.008 and P=0.005 respec- common to all six promiscuous B7-epitopes identified. tively). The subset of IFN-γ-producing CD8+CD45RA- Moreover, with one exception, there were no significant dif- CCR7-CD27+ Gag-specific cells, but not the Nef-specific ferences in the frequency, magnitude or immunodominance ones, negatively correlated with the HIV-1 proviral DNA of the CTL responses restricted by different HLA alleles to loads (P=0.045) while both positively correlated with CD4 explain these discrepancies. This suggests that the unique counts (P=0.003 and P=0.0002 respectively). peptide/MHC complexes generated by even closely-related Conclusion: Gag-specific but not Nef-specific IFN-γ-produc- HLA, induce CTL responses that are qualitatively different. ing CD8 T cells correlate with numbers of HIV-infected cells. This hypothesis is supported by additional differences This difference is related to a less differentiated status of observed between CTL targeting identical epitopes but CD8+CD45RA-CCR7-CD27+ Gag-specific cells, which restricted by different HLA: first, the occurrence of distinct, depends on preserved CD4 counts and might participate in HLA-specific escape mutation; second, the recruitment dis- the immune control of HIV. tinct TcR repertoires of by particular peptide/MHC com- plexes; and, third, significant differences in the functional avidity of CTL P05-20 Conclusions: Taken together, these data indicate that signifi- cant functional differences exist between CTL targeting iden- tical epitopes but restricted by different, albeit closely-related Differential selection pressure exerted by CTL tar- HLA. These findings are of relevance to vaccine approaches geting identical epitopes but restricted by distinct that seek to exploit the property of epitope promiscuity to HLA alleles within the same HLA-supertype circumvent obstacles imposed by HLA diversity. A Leslie1, D Price2, P Mkhize3, A Rathod3, H Crawford1, 1 2 3 4 I Honeyborne , D Douek , P Kiepiela , B Walker and P05-21 P Goulder1, 3 1 Department of Paediatrics, Nuffield Department of Medicine, Capturing HIV viral sequence diversity in in vitro Peter Medawar Building for Pathogen Research, Oxford , UK; 2 immune analyses Human Immunology Section, Vaccine Research Center, National C Brander1, N Frahm1, N Joijc2, K Yusim3, DC Nickle4, CH Institute of Allergy and Infectious Diseases, National Institutes Linde1, HS Hewitt1, KL Faircloth1, M Muldoon5, C Kadie2, of Health, Bethesda, MD, USA; 3 HIV Pathogenesis Programme, BD Walker1, D Heckerman2, JI Mullins4 and B Korber3 The Doris Duke Medical Research Institute, University of Natal, Durban, South Africa; 4 Partners AIDS Research Center, 1 Partners AIDS Research Center, Massachusetts General Massachusetts General Hospital, Charlestown, Boston, MA, USA Hospital, Charlestown, MA, USA; 2 Microsoft Research, Redmond, WA, USA; 3 Los Alamos National Laboratory, Los Background: HLA diversity is seen as a major challenge to Alamos, NM, USA; 4 University of Washington, Seattle, WA, USA; CTL epitope vaccines. One current approach is to focus on 5 School of Mathematics, University of Manchester, UK “promiscuous” epitopes, which are presented by multiple HLA alleles. However, this requires that all CTL targeting a Background: HIV diversity is considered a significant hur- particular epitope are functionally equivalent, irrespective of dle for global HIV vaccine design. This diversity also HLA restriction. impedes the comprehensive assessment of HIV-specific Objectives: To identify promiscuous epitopes presented by immune responses, since in vitro test sets cover only a por- the common B7-supertype and determine whether there are tion of the circulating viral population and thus miss qualitative differences between the CTL responses directed responses to more variable regions. against these epitopes but restricted by different HLA. Objectives: The development of in vitro test reagents cover- Methods: Promiscuous epitopes targeted by at least two of ing the most frequent and immunogenic antigen sequences in B7-supertypes alleles B*0702, B*4201 and B*8101 were the currently circulating viral population would allow for a identified in a cohort of 515 C-clade infected individuals more comprehensive analysis of host immune responses AIDS Vaccine 2006 97 Poster sessions_revGJ 14/8/06 16:34 Page 98 Amsterdam, Netherlands 29 August–1 September 2006 induced upon natural HIV infection, and could thus provide tope-specific CTL responses is associated with better in vivo important guidance for future vaccine design. control of HIV infection. Methods: Two complimentary approaches, one using “tog- Methods: The TCR repertoire of CTL specific for three HIV- gled” peptides and the other based on the inclusion of 10mer derived, promiscuous HLA class I restricted epitopes was peptides with maximal population coverage, referred to as assessed by flow-cytometric sorting of antigen-specific cells “epitome”, were used in in vitro immune analyses. Toggle and subsequent sequencing of at least 50 TCR transcripts per peptides incorporate diversity in the synthesis step of the test response. Additionally, the functional avidity of these epi- peptides and thus resemble a combinatorial peptide library, tope-specific responses restricted by different HLA alleles was with limited diversity in the number of variable sites as well analysed by determining the half-maximal stimulatory anti- as in the number of different amino acids occupying these gen concentration in peptide titration experiments. sites. The epitome approach aims to identify 10-mer peptides Furthermore, the ability to recognize the most commonly that are most frequently present in a pre-defined fraction of occurring epitope variants was evaluated. the viral population under investigation and to compress Results: The data demonstrate a link between narrower anti- these targets into shorter sequences for future vaccine gen-specific TCR repertoires and high CD4 counts in the test- immunogen design. ed individuals. In addition, the functional avidity of the Results: Either approach independently and significantly epitope-specific CTL population was associated with the abil- increased the detection of T cell responses in comparison to ity to recognize dominant naturally occurring epitope vari- peptide test sets based on single-gene-length computationally ants, thereby providing a possible explanation for the better derived or naturally occurring sequences. Autologous viral outcome of HIV infection in subjects with CTL responses sequence analyses explain overall increased response rates as comprising fewer clonotypes. test sequences more effectively covered autologous sequence Conclusions: These data show an association between nar- diversity. In addition, increased response rates can be row TCR repertoires and improved HIV control in vivo and achieved at massively reduced costs compared to synthesizing indicate a dynamic tripartite interaction between the context autologous peptide test sets. of antigen presentation, the nature of the engaged TCR reper- Conclusions: Both approaches tested here represent powerful toire and control of HIV replication. These findings con- tools to maximize detection of virus-specific cellular immune tribute to a better definition of immune correlates of responses in genetically diverse individuals, and represent controlled HIV infection and will have important implica- extraordinarily powerful tools for epitope identification and tions for the vaccine-mediated induction of protective T cell assessment of cross-clade reactivity. Combined, they afford a responses. significantly improved in vitro detection of virus-specific T *These authors have contributed equally to this work. cells at massively reduced costs, and may provide crucial guidance for the future design of HIV vaccine immunogens. P05-23 P05-22 Estimating the effectiveness of SIV specific CD8+ T cells from the dynamics of viral immune escape Narrow T cell receptor repertoire of virus-specific CD8+ T lymphocytes is associated with HIV control JN Mandl1, R Regoes1, DA Garber1 and MB Feinberg1,2 1 2 1 1 N Frahm *, DA Price *, CH Linde , KL Faircloth , 1 Emory Vaccine Center, Emory University School of Medicine, 2 2 2 1 JM Brenchley , TE Asher , LE Ruff , HS Hewitt , Atlanta, GA, USA; 2 Merck Vaccine Division, West Point, PA, USA DC Douek2 and Brander1 Background: CD8+ T cells appear to play a significant role in 1 Partners AIDS Research Center, Massachusetts General limiting the viremia in immunodeficiency virus infections Hospital and Harvard Medical School, Boston, MA, USA; 2 both in humans (HIV) and macaques (SIV). However, it has Human Immunology Section, Vaccine Research Center, National not been possible to directly measure the in vivo effectiveness Institute of Allergy and Infectious Diseases, National Institutes of cytotoxic T cell (CTL) killing. Hence it is unknown what + of Health, Bethesda, MD, USA their actual contribution to the death rate of CD4 T cells is and what will be required for antiviral CTLs to effectively and durably control virus replication for the purpose of an Background: The breadth of T cell receptor (TCR) reper- AIDS vaccine. toires of virus-specific cytotoxic T lymphocyte (CTL) popu- Objective: To model the in vivo effectiveness of SIV-specific lations has been shown to crucially contribute to the CTLs. functional avidity and, at least in some animal studies, to the Methods: Frequencies of Tat SL8-specific CD8+ T cells, tar- efficacy of CTL responses in vivo. get cells (Ki67+ CD4+ T cells), and SL8 mutational-escape Objective: To identify promiscuous HIV-derived CTL epi- variants were determined in SIVmac239-infected topes that can be presented in the context of multiple HLA MamuA*01 macaques that were either treated acutely with class I alleles differentially associated with HIV progression, costimulatory blockade (CS) molecules (CTLA4Ig, anti- and to test whether the TCR repertoire breadth of these epi- 98 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 99 Amsterdam, Netherlands 29 August–1 September 2006 CD40L) to impair generation of antiviral T and B cell activatory antibodies in the presence of 0.01–1 uM TAT0001 responses, or were untreated (control). Mathematical model- or TAT0002. Telomerase activity was evaluated using the ling of the selection of SL8 escape variants allowed estima- telomerase repeat amplification protocol (TRAP) assay. HIV- tion of both the in vivo killing rate (κ) per Tat SL8 specific specific secretion of IFN-γ using the ELISPOT assay was test- CD8+ T cell and the fitness cost of Tat SL8 escape. ed in CD8+ T cells expanded in culture for up to 14 days in Results: Tat SL8 escape occurred rapidly in control animals the presence of the compounds. The effect of the compounds (≥90% of clones exhibiting escape mutations by 42 days on the ability of CD8+ T cells to reduce viral levels when co- post-infection) but not in CS-treated macaques (escape vari- cultured with autologous HIV-infected CD4+ T cells was eval- ants <42%). No difference in per cell CTL killing rates were uated using a p24 ELISA. observed between control and CS blockade groups (Control: Results: Exposure of activated T cells from HIV-infected per- κ = 0.0213 day-1, 68%CI: 0.0180–0.0319 day-1; CS blockade: sons to either compound at concentrations as low as 10 nM κ = 0.0209 day-1, 68%CI: 0.0189–0.0243 day-1). Estimates of led to significant increases in both cell proliferation and the effectiveness of SL8-specific CTL killing of SIV-infected telomerase activity, an effect that was abrogated in the pres- cells indicates that, at peak CTL levels, Tat specific CD8+ T ence of ERK/MAP kinase pathway inhibitors. HIV-specific cells are 0.5–1.7 times as effective as HAART in reducing production of IFN-γ by CD8+ T cells tested immediately ex virus replication. vivo or after expansion in culture was also enhanced. Conclusions: These first quantitative estimates of the in vivo Importantly, viral production by infected autologous CD4+ T effectiveness of SIV-specific CD8+ T cells suggest that these cells was significantly reduced by CD8+ T cells that were pre- cells have a significant impact on shortening the longevity of treated with the compounds. CD4+ T cells (at least during acute infection) and may exert a Conclusions: Pharmacological telomerase activators may pressure on the virus population that approximates (or may provide a practical and effective strategy to maintain CD8+ T even exceed) that of HAART. This approach could be used to cell control over HIV, thereby preventing immune exhaustion gauge the beneficial impact of vaccines as well as the require- and possibly delaying progression to AIDS. (Supported by ments for successful control of virus replication by CD8+ T NIH & TA Therapeutics, Ltd.). cells. P05-25 P05-24 Stimulation of T cells with dendritic cells loaded Small molecule telomerase activators: a novel with apoptotic cells infected with autologous HIV-1 approach to retard immune exhaustion and pro- gression to AIDS CR Rinaldo, Z Fan, XL Huang, P Piazza, L Borowski, NC Connolly, P Gupta, TL Whiteside and SA Riddler CB Harley2, RB Effros1 SR Fauce1, OO Yang1, B Jamieson1, AC Chin2, H Ng1 University of Pittsburgh, Pittsburgh, PA, USA 1 David Geffen School of Medicine at UCLA, Los Angeles, CA, Background: We urgently need therapies that suppress HIV-1 USA; 2 Geron Corporation, Menlo Park, CA, USA replication following therapy discontinuation. Dendritic cell (DC) vaccines using inactivated autologous virus have demon- Background: Chronic HIV infection is associated with the strated ability to augment immunity to HIV-1 infection. We progressive increase in CD8+CD28- T cells, which have have previously shown that apoptotic cells infected with HIV-1 shortened telomeres and reduced anti-viral function. X4 serve as potent stimulators of HIV-1 specific T cells. Interestingly, high proportions of this same cell population Objectives: To assess the immunogenicity of DCs loaded with in elderly persons is significantly correlated with reduced apoptotic cells infected with autologous HIV-1 as an influenza vaccine responsiveness. Telomerase gene transduc- immunotherapy for HIV-1 infection. tion of HIV-specific CD8+ T cells from HIV-infected individ- Methods: Monocyte-derived DCs from HIV-1 infected sub- uals leads to delayed loss of CD28 expression, sustained jects were loaded with HIV-1-infected apoptotic cells and antigen-specific proliferation and enhanced anti-viral func- treated with either our most potent maturation inducer, i.e., tion. These proof-of-principle studies suggest that manipula- trimeric CD40L (Amgen), or with a maturation cocktail from tion of T cell telomerase constitutes a novel strategy for our clinical trials, i.e., IL-1β + IL-6 + TNFα + PGE-2. augmenting immune control over HIV. Activation of HIV-1 specific T cells was assessed for produc- Objectives: To assess the effectiveness of small molecule tion of IFNγ by Elispot and by flow cytometry for IFNγ, IL- telomerase activators in increasing immune control over HIV 2 and TNFα production and CD107a mobilization. and preventing immune exhaustion. Results: T cell responses specific for autologous HIV-1 and X4 Methods: Two small molecules, TAT0001 and a related com- or R5 lab strains of HIV-1 were induced in vitro by DCs loaded pound, TAT0002, discovered in a cell-based telomerase acti- with infected apoptotic cells, with the greatest T cell reactivity vation screen in human keratinocytes, were tested for their to autologous HIV-1. Epitope mapping showed that multiple effect on T cells from HIV-infected persons. T cells were stim- Nef and Gag epitopes were recognized; other HIV-1 specifici- ulated in cell culture with mitogen (phytohemagglutin), or ties are being tested. A Phase I immunotherapy trial is com- AIDS Vaccine 2006 99 Poster sessions_revGJ 14/8/06 16:34 Page 100 Amsterdam, Netherlands 29 August–1 September 2006 mencing to define the ability of DCs loaded with autologous Conclusions: In conclusion, we were able to generate CTL virus-infected, apoptotic cells to stimulate T cell immunity in HIV-specific CD8+ T-cells responses using autologous HIV HIV-1 infected subjects on viral suppressive therapy. mRNA electroporated DC. Moreover, a full pattern of CD8 Conclusions: DCs loaded with apoptotic cells infected with differentiation was seen after the addition of cognate CD4- autologous HIV-1 are potent stimulators of virus-specific T specific peptide allowing the establishment of memory T cell reactivity in vitro, with specificity for a broad repertoire cells, most important in the vaccination strategies. of HIV-1 epitopes. These results support use of DCs loaded with autologous virus as an immunotherapy in HIV-1 infec- tion. We are currently initiating a Phase I clinical trial at our P05-27 institution that is sponsored by the NIH. CTLA-4 is overexpressed in HIV-specific Th1 CD4 P05-26 cells of HIV-infected individuals, except in most subjects who spontaneously control viremia Restoration of specific CD8 CTL responses using 1 1 1 2 autologous HIV electroporated DCs D Kaufman , F Pereyra , D Kavanagh , J Zaunders , M Brockman1, E Mackey1, A Rathod1, A Kelleher2, B Walker1 and E Rosenberg1 B Yassine Diab1, O Yegorov1, T Baumgartner1, B Gagnon1, 2 2 3 3 R Boulassel , J-P Routy , D Healey , I Tcherepanova , C 1 Massachusetts General Hospital; 2 Centre for Immunology, St 3 1 Nicolette and R-P Sekaly . Vincent’s Hospital 1 University of Montreal, Montreal, Canada; 2 McGill Background: HIV-specific CD4 cells are critical for immune University Health Center, Montreal, Canada; 3 Argos control of HIV infection and are impaired in the majority of Therapeutics, Durham, USA infected individuals. CTLA-4 is an important inhibitory co-sig- nalling molecule and its expression in the general CD4 T cell Background: Development of a memory immune response fol- population has been shown to increase with disease progression. lowing a viral infection is a critical step in the control of the Methods: In a cross-sectional study, we investigated 36 individ- disease. In HIV infection, we and others have previously uals at different stages of HIV-1 infection, treated with antiviral shown that chronically infected patients fail to control their therapy or untreated. Nine-colour flow cytometry and ICS viremia despite high frequency of HIV-specific CD8+ T cells. It assays with Gag, Nef and CMV pp65 peptide pools and CMV is believed that this defect comes from skewed phenotypic and lysate were used to assess intracellular IFNγ, IL2, CD40L and functional maturation of the T cell memory compartment. CTLA-4 expression. Three subjects were also longitudinally fol- Objectives: Our main objective is to test if properly ex vivo lowed before and after institution of therapy. Class II tetramers matured DCs could correct the defective CD4 and CD8 HIV- were used to assess CTLA-4 expression before antigen stimula- specific T cell response, and whether strategies based on the tion. use of DCs transfected with mRNA encoding autologous Results: CTLA-4 expression in Gag-specific IFNγ+ CD4 was HIV sequences can restore HIV-specific help and induce the lower in the group of individuals with spontaneous viremia con- differentiation of anti-HIV CD8 T cells into potent cytolytic trol compared with all other groups (P<0.05). Median CTLA-4 killers able to mediate viral control. MFI for people with spontaneous viremia control: 807, range: Methods: Autologous HIV or CMV mRNA-electroporated 196 to 3970; median CTLA-4 MFI for the other subjects: 4187, DCs were co-cultured in vitro with PBMC derived from 6 range: 2162 to 17188. (P≤0.0001). The highest level of CTLA- chronically infected HAART treated patient for 6 days fol- 4 expression was observed in subjects with PHI. The same pat- lowed by analysis of T cell antigen-specificity by both CFSE terns of CTLA-4 expression were observed in IL2-secreting proliferation and cytokine secretion. CD4, although with a lower CTLA-4 expression (P=0.0008). Results: These experiments show that DC electroporated Longitudinal investigation of subjects on and off therapy with HIV (Gag, Nef, Rev and Tat) mRNA or with CMV showed that control of viremia was associated with a modest (pp65) mRNA are able to activate antigen-specific CD8+ and decrease in CTLA-4 expression. CTLA-4 levels correlated induce them to proliferate in the absence of a detectable CD4 directly with viral load and inversely with CD4 counts. IFNγ+ response. Furthermore, the use of multicolorimetric FACS and IL2+ Gag-specific CD4 did not express CD25 or FoxP3, assay allowed us to show that the majority of proliferating indicating that these cells are not classical Treg. Comparison cells have the characteristic of effectors cells (highly positive with CMV-specific CD4 showed that CTLA-4 overexpression is for the Granzyme B, CD107a and IFNγ production). HIV-specific. Class II tetramer staining demonstrate that CTLA- Moreover, the addition of exogenously CD4-specific peptide 4 is already upregulated before stimulation and further or manipulated electroporated autologous DC in the co-cul- increased upon encounter with the cognate antigen. ture shows that the CD8+ activated-specific cells (also CD4 T Conclusions: HIV-specific CD4cells express high levels CTLA-4 cell activation) have a multiple pattern of T cell differentia- in all categories of subjects investigated with the exception of the tion including the generation of central memory cells and a majority of individuals who spontaneously control viremia. diverse TCR repertoire. These results suggest that CTLA-4 overexpression may be an 100 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 101 Amsterdam, Netherlands 29 August–1 September 2006 important factor in HIV-specific T cell dysfunction. These cells H P05-29 could also have a suppressor activity. Further investigation into the effects of CTLA-4 expression may shed important insight Elite control of HIV replication is distinguished into disease pathogenesis. from chronic disease by lower HIV-specific IFNγ producing CD8+ T cell and higher polyfunctional CD4+ and CD8+ T cell responses P05-28 F Pereyra1,2, D Kaufmann1, A Rathod1, B Baker1, Identification of a subset of SIVgag-specific A Piechocka 1,2 and BD Walker 1,2 CD4+IFNγ+Central Memory T cells that correlates with control of viremia and lack of disease pro- 1 Partners AIDS Research Center and 2 Howard Hughes Medical gression in SIVmac251 infected macaques Institute, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA BK Felber, A Valentin, A von Gegerfelt, M Rosati, M Morrow, C Alicea and GN Pavlakis Background: Large cohorts of Elite controllers of HIV are rare and comprehensive analysis of CD8+ and CD4+ T cell National Cancer Institute at Frederick, Frederick, MD, USA responses is lacking. Objectives: To describe the HIV specific CD8+ and CD4+ T cell responses in HIV elite controllers compared to chronic Aim of the study: To identify specific immune responses corre- infected individuals with detectable viral loads. lating with control of viremia in SIV-infected macaques. Methods: We investigated 149 individuals with HIV-1 infec- Methods: We study 3 groups of chronically SIV- infected rhesus tion, including 64 elite controllers (VL<50 c/ml), 60 viremic macaques able to control viral replication. Macaques infected controllers (VL 50–2000) and 25 persons with chronic infec- by an attenuated, non-pathogenic Rev-independent SIV strain tion (VL>2000). Intracellular cytokine staining after stimula- control viremia for up to 7 years (group 1). Macaques in group tion with HIV peptide pools spanning all expressed HIV 2 have been infected by the attenuated SIV and were subse- proteins were used to measure HIV specific CD4+ and CD8+ quently challenged with highly pathogenic SIVmac251, and the T cell responses. Gamma interferon ELISpot assay was used animals continue to control viremia for more than 4 years. to asses HIV-specific CD8+ T cell responses. Animals in group 3 are SIV infected macaques that were treated Results: Median duration since diagnosis and viral loads with a combination of antiviral drugs (ART) for 20 weeks. were 14 years and <50 c/ml for elite controllers, 16.5 years During this time, the animals were also immunized with opti- and 770 c/ml for viremic controllers, and 7 years and mized forms of DNA vectors expressing modified forms of SIV 124,000 copies for those with chronic infection. HIV-specific antigens. CD8+ and CD4+ responses were detected in all individuals. Results: Animals in groups 1 and 2 show persistent humoral and The total breadth and magnitude of HIV-1 specific CD8+ T cellular immune responses despite undetectable (group 1, <100 cell responses was significantly lower in elite controllers com- copies/ml) or persistent low (group2, virus load of <103–105 pared to viremic controllers and chronic infected groups but copies/ml) levels of viremia. Of the animals in group 3, we found not between detectable controllers and chronic infected. that DNA immunization during ART (n=12), but not ART Among elite controllers and viremic controllers the breadth alone (n=11), led to substantial decreases in plasma viral load and magnitude of Gag-specific responses was significantly after therapy termination. Several animals (6/12) continue to higher compared to Nef, Pol and Env proteins but not among control viremia (levels <100-104 copies/ml) for more than 2 chronic infected and the relative breadth and magnitude of years. Their cellular immune responses were boosted by DNA Gag specific responses was higher in elite and viremic con- vaccination and persisted despite lower virus loads after thera- trollers compared to chronic infected. The percentage of HIV py. Importantly, 10-parameter flow cytometric analysis of specific CD4+ and CD8+ T cells that secrete both IL-2 and PBMC from animals of these 3 groups of ‘controllers’ revealed IFN γ and the CD4+/CD8+ ratio of HIV-specific T cells that preservation of central memory T cells, defined as CD3+ secrete IL-2, IFN-γ or both was significantly higher in elite CD45RA- CD28+. We further found that a subset of SIVgag-spe- controllers compared to both viremic controllers and chron- cific CD4+IFNγ+ central memory T cells was preserved in atten- ic infected groups. uated SIV infected animals (groups 1 & 2) as well as in animals Conclusions: Elite controllers represent a unique subgroup receiving DNA vaccination in addition to ART (group 3), of HIV infected individuals characterized by spontaneous whereas this population was absent in macaques with progres- strict control of viral replication. The preferential targeting sive disease. of Gag among elite controllers and viremic controllers sup- Discussion: Our results demonstrate that control of viremia and ports a potential role of immune responses to this protein lack of progression towards immunodeficiency correlate with in sustained control of HIV viremia. The lower virus spe- preservation of SIVgag-specific CD4+ central memory T cells in cific CD8 T cell responses and higher percentage of poly- SIV infected macaques. Strategies to induce and preserve these functional CD4+ and CD8+ T cells appears to be more cells may be useful to control disease progression. closely associated with the presence of viremia than with a particular clinical state. AIDS Vaccine 2006 101 Poster sessions_revGJ 14/8/06 16:34 Page 102 Amsterdam, Netherlands 29 August–1 September 2006 P05-30 P05-31 Persistence and homeostasis of CD4+ central Effective immortalization of SIV-specific rhesus memory T cells is dependent on FOXO3a phos- macaque CD8+ T cell clones with maintenance of phorylation and STAT-5a activation primary cell characteristics C Riou1, J Van Grevenynghe1, B Yassine Diab1, JD Lifson, MT Trivett, EV Barsov, CM Trubey, H L Greller3, R Somogyi3, M Cameron4, D Kelvin 4, Andersen, F Yuan, MF Princiotta, DE Ott and C Ohlen EK Haddad1 and RP Sekaly1 AIDS Vaccine Program, SAIC Frederick Inc., NCI Frederick, 1 Laboratory of Immunology, Centre de Recherche Saint-Luc, Frederick, MD, USA University of Montreal, Montreal, Quebec, Canada; 2 Royal Victoria Hospital, Division of Hematology, Montreal, Quebec, Objectives: Studies of antigen specific T lymphocytes have Canada; 3 Biosystemix, Ltd, Sydenham, Ontario, Canada; 4 typically depended on either short term analyses of freshly University Health Network, Toronto Research Institute, Ontario, isolated cells or painstakingly derived T cell lines or clones, Canada with both being limited in cell numbers and in vitro lifespan. Sophisticated assays such as MHC-peptide tetramer analysis and intracellular cytokine staining (ICS) used with polychro- Background: The molecular events that are involved in the matic flow cytometry allow detailed assessment of cellular establishment and the maintenance of CD4+ Central immune functions but can be limited by logistical issues relat- Memory (CM) and Effector Memory (EM) T cells are ed to the availability of appropriate positive control cell pop- poorly understood. ulations. A consistent source of fresh responder cells may not Aims: We investigated the molecular pathways involved in be available and supplies of cryopreserved cells are finite. To memory T cell populations survival. circumvent this limitation, we immortalized SIV antigen spe- Methods: Purified CD4+ CM and EM were sorted by flowcy- cific T cells from rhesus macaques. tometry and we performed functional assays, western blot Methods: The human telomerase RT (hTERT) gene was analysis and global gene expression profiling. cloned into xlox-PGK-LNGFR, a retroviral vector that also Results: We found that Central Memory cells are signifi- expresses C-terminally truncated (non-functional) human cantly more resistant to both spontaneous and Fas-mediat- low affinity nerve growth factor receptor (LNGFR). SIV- ed-apoptosis than Effector Memory cells. We showed that specific CD8+ T cell lines were generated by stimulating the survival of CM CD4+ T cells is mediated trough two PBMC from SIVmac239 infected Mamu A*01 positive rhe- converging pathways: 1) the stat5 pathway that is triggered sus macaques with the immunodominant A*01 restricted with a better efficiency in CM cells in response to both IL- Gag CM9 and Tat SL8 epitope peptides. After initial stimu- 7 and IL-2 as compared to EM cells and 2) the FOXO3A lation, reactive T cell lines were cloned by limiting dilution pathways. As compared to EM cells, CM cells express high- using anti-CD3 stimulation and human PBMC and B-LCL er levels of the phosphorylated form of FOXO3a, prevent- feeder cells with IL-2. Clones with SIV antigen specificity by ing these cells from expressing pro-apoptotic molecules tetramer reactivity and IFN-γ ICS were transduced with such as BIM and FASL. We demonstrated that the phos- hTERT-containing xlox-PGK-LNGFR and NGFR+ cells phorylation of FOXO3a is AKT/IKK dependent and can be were sorted using anti-NGFR mAb and magnetic beads. achieved after TCR triggering. Experiments aimed at block- Results: Transduced macaque T cell clones have remained ing FOXO3A phosphorylation confirmed the role of IL-2 dependent, and have maintained their cell surface FOXO3A in protecting Central Memory T cells from apop- immunophenotype, antigen specificity and reactivity, tosis. We further determined whether these two survival demonstrable by MHC-peptide tetramer staining and IFN-γ pathways are implicated in natural protection to HIV infec- ICS, for up to 8 months in culture. Clones have been stable tion. We found that both the FOXO3a and STAT-5a path- over time; it should be feasible to use such cells as a consis- ways to be up-regulated in HIV long term non progressors tent source of positive control responders for cellular when compared to normal progressors. immune function assays and other purposes. Immortalized Conclusion: Our results define the underlying molecular CD4+ T cells have been produced by similar hTERT trans- mechanisms responsible for the longevity and persistence of duction, following antigen non-specific activation by anti- CM CD4+ T cells. CD3 stimulation, and shown to be susceptible to SIV infection in vitro. Conclusions: In addition to control cell populations for immune function assays, this strategy thus also provides both the effector and target elements for in vitro studies of antivi- ral activities mediated by CD8+ T cells, potentially useful for studies of pathogenesis and for vaccine evaluation. (NCI Contract N01-CO-12400) 102 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 103 Amsterdam, Netherlands 29 August–1 September 2006 P05-32 P05-33 Increased expression of NK cell receptors on HIV- Design of HIV-1 pseudoviruses to monitor HIV-1 specific CD8+ T cells correlates with impaired TCR specific cellular immune responses signalling D Beels, L Heyndrickx, G Vanham and L Kestens G Alter, N Teigen, A Meier and M Altfeld Institute of Tropical Medicine, Antwerp, Belgium Massachusetts General Hospital, Boston, MA, USA Background: Currently available antigens used to measure in Background: Virus-specific-CD8+ T cells play a central role vitro anti- HIV-1 specific cellular immune responses assess in the control of viral infections by direct elimination of the immune response against a small part of the HIV virus. infected cells and secretion of a number of soluble factors. During initial screening it may be more appropriate to mea- However despite the induction of strong and broad HIV- sure HIV-1 specific cellular immune responses against a max- specific CD8+ T cell responses in chronic HIV-1 infection, imum of HIV-1 epitopes at once. these cells progressively lose critical effector functions. A Objective: To design HIV-1 pseudoviruses able to stimulate number of recent studies have shown that a significant sub- cellular immune responses against a maximum of HIV-1 epi- set of CD8+ T cells appear to up-regulate inhibitory “NK topes via natural antigen processing and peptide presentation cell receptor” (KIR and NKG2A) expression following on MHC class I and class II molecules. These pseudoviruses encounter with antigen, and that CD8+ T cells expressing should be single-cycle infectious, contain a maximum of HIV- NK cell receptors persist in chronically infected mice but 1 proteins and the pseudoviral genome should encode a max- not in mice that clear the infection. imum of HIV-1 epitopes. Objective: Given the profound inhibitory effect of these Methods: Starting from a full length HIV-1 plasmid, the enve- receptors, we aimed to characterize the expression and role of lope cleaving site was removed by site directed mutagenesis. KIR and NKG2A on the surface of CD8+ T cells in HIV-1 This cleaving site is responsible for the cleavage of gp160 into infected subjects at different stages of infection. gp120 and gp41, necessary for the formation of functional Methods: KIR/NKG2A expression as well as the enrichment trimers. The ‘envelope cleavage site deficient HIV-1 con- of these molecules on HIV-1 specific CD8+ T cells was per- struct’ (further referred to as dREKRA) was co-transfected formed by multi-paratmeter flow cytometry using tetramer with a complementary HIV-1 or VSV-G envelope construct staining. The functional capacity of KIR and/or NKG2A+ into 293T cells. In order to check for replication competent CD8+ T cells was assessed in a 6-hour intracellular cytokine recombinants, the resulting virions were cultured with secretion assay using a cognate peptide, staphylococcal- PHA/IL-2 stimulated PBMC and infection was monitored enterotoxin-B, PMA/ionomycin, or medium alone by staining using an in-house p24 ELISA. From the p24 positive cultures, for IFN-γ, TNF-α, and CD107a. proviral DNA was extracted, and the env gene was amplified Results: Here we demonstrate that both KIR and NKG2A and partial sequenced. expression are dramatically increased on the surface of Results: Co-transfection of the ‘dREKRA’ construct with a CD8+ T cells in subjects with viremic HIV-1 infection com- complementary HIV-1 envelope construct resulted in replica- pared to both aviremic subjects and uninfected controls tion competent HIV-1 virus. Sequence analysis showed that (P<0.05 for both comparisons). Furthermore, these mark- homologues recombination between the HIV-1 plasmids ers were preferentially increased on the surface of HIV-spe- occurred during transfection. Co-transfection of ‘dREKRA’ cific CD8+ T cells. Finally, the expression of these markers with a VSV-G envelope vector resulted in VSV-G pseudo- on the surface of CD8+ T cells resulted in an impaired abil- typed, HIV-1 single-cycle pseudoviruses. ity of these KIR/NKG2A+ CD8+ T cells to respond to pep- Conclusions: We have shown that the design of HIV-1 tide-stimulation and SEB but not to stimulation with pseudoviruses for the in vitro antigen presentation of a max- mitogens (PMA/ionomycin). imum of HIV-1 epitopes is restricted by homologues recom- Conclusions: This data suggests that KIR and NKG2A bination. More HIV-1 epitopes (in other genes) will have to expression on the surface of HIV-specific CD8+ T cells pre- be sacrificed in order to obtain safe single-cylce HIV-1 dominantly occurs in the context of persistent viral infection, pseudoviruses suitable for the stimulation of a maximum of and that these markers may actively suppress TCR activa- HIV-1 specific cellular immune responses. tion. Thus chronic immune activation may lead to the func- tional impairment of these cells through the up-regulation of inhibitory NK cell-receptors. AIDS Vaccine 2006 103 Poster sessions_revGJ 14/8/06 16:34 Page 104 Amsterdam, Netherlands 29 August–1 September 2006 P05-34 Dual selection pressure by drugs and HLA class I-restricted immune responses on HIV-1 protease T Harrer1,4 , SM Müller1,4, B Schätz1,4, K Eismann1, S Bergmann1, M Bäuerle1, H Walter2,4, B Schmidt2, K Korn2, H Sticht3, B Spriewald1, and EG Harrer1 1 Dept. of Medicine III, University Hospital Erlangen, Erlangen, Germany; 2 Institute of Clinical and Molecular Virology, University of Erlangen-Nuremberg, Germany; 3 Institute of Biochemistry, Emil-Fischer-Center, University of Erlangen- Nuremberg, Germany; 4 German Competence Network on HIV/AIDS Background: The development of drug resistance is one of the biggest obstacles in HIV-1 therapy, at least in countries where antiretroviral therapy (ART) is available. Being one of the major drug targets the protease (PR) is subjected to the acquisition of resistance mutations. Considering the various mutational pathways of the development of drug resistance in patients with similar or even identical therapeutic regi- mens, it seems likely that immunological host factors interact with the emergence of resistance mutations. Objectives: To determine the influence of HIV-1-specific cyto- toxic T lymphocytes (CTL) on the development of drug resis- tance mutations in the HIV-1 protease, we analysed protease sequences in a human leucocyte antigen I (HLA-I)-typed cohort HIV-1-positive individuals. Methods: Protease sequences in a HLA-I-typed cohort of 94 HIV-1-positive individuals were analysed and univariate sta- tistical analyses were performed to identify correlations between HLA alleles and drug induced mutations and poly- morphisms in HIV-1. On the basis of our univariate statisti- cal analyses we identified new CTL epitopes, which were subsequently tested in γ-IFN-ELISPOT analyses. Results: Minor and major drug mutations as well as drug- associated polymorphisms showed significant associations to HLA class I alleles and we could define eight new CTL epi- topes, which were all verified in γ-IFN-ELISPOT analyses. Several drug resistance-associated mutations in the protease acted as CTL escape mutations, especially if anchoring amino acids in the CTL epitopes were affected by the mutations. However, many patients were able to recognize the peptides containing these mutations, and analysis of CTL recognition patterns showed that several patients mounted oligoclonal CTL responses against certain epitopes targeting a variety of viral variants. Conclusions: Our results support a fundamental role of HLA class I-restricted immune responses in shaping the sequence of the HIV-1 protease in vivo. This may have important clin- ical implications both for to the understanding of drug resis- tance pathways and for the design of therapeutic vaccines. 104 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 105 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 6: B CELL IMMUNITY P06-01 sequences suggests that targeting conserved forms of such epitopes may allow the induction of potent neutralization responses that are not subject to epitope masking effects. Sequences in the V1/V2 and V3 variable domains of HIV-1 gp120 that determine the roles of these regions as mediators of epitope masking and as P06-02 potent netralization targets Dynamic model of HIV-1 infection and neutralization CP Krachmarov1, WJ Honnen1, Z Lai1, X Bu1, MK Gorny2, S Zolla-Pazner2, 3 and A Pinter1 S Maloveste1, F Charifi1, M Franti2, K Delgado1, PJ Klasse3 and P Poignard1 1 Public Health Research Institute, UMDNJ, Newark, NJ, USA; 2 NYC School of Medicine, New York, NY, and 3 New York VA 1 Centre d’Immunologie de Marseille-Luminy, Marseille, France; Medical Center, New York, NY, USA 2 Progenics Pharmaceuticals Inc., Tarrytown NY 10591, USA; 3 Background: We have shown that polyclonal and mono- Department of Microbiology and Immunology, Cornell University clonal anti-V3 antibodies isolated from infected patients Weill Medical College, New York, NY, USA possess cross-clade neutralization activity for primary virus- es with “unmasked” envelopes, and that the restricted activ- Background: Studies suggest that a vaccine eliciting neutral- ity of such antibodies for typical primary isolates is due to izing antibodies, which recognize HIV-1 envelope glycopro- epitope masking by the V1V2 domain. We have also identi- teins, may be efficient to block HIV-1 transmission. However, fied type-specific epitopes in the V2 domain that can medi- no vaccine candidate is currently able to induce potent and ate ultra-sensitive neutralization. broadly neutralizing antibodies. Objectives: To map sequences in the V1/V2 domain that Objectives: A better understanding of the mechanisms of mediate its masking effect on epitopes in the V3 domain and antibody neutralization may raise new ideas for antigen elsewhere in gp120, and to further characterize potent neu- design strategies. Dynamic aspects of the HIV-1 infection tralization epitopes in the V2 region. process have so far not been fully considered in understand- Methods: Chimeric and mutant Envs were generated by ing neutralization. Therefore, we studied the kinetics of HIV- restriction fragment exchanges and QuickchangeΤΜ site- 1 infection and how monoclonal neutralizing antibodies directed mutagenesis. Neutralization assays were performed interfere with this process. in U87-CD5-CCR5 cells in single-cycle infectivity assays Methods: To study the dynamic process of HIV-1 infection using luciferase-expressing virions pseudotyped with the Env and neutralization, we monitored the infection kinetics of proteins of interest. single round replication pseudotyped HIV-1 particles in vitro. Results: An analysis of the relative contributions of sequence Viruses were pseudotyped with primary isolates envelope gly- variation in the V1/V2 and V3 regions towards the neutral- coproteins. The level of infection was measured using a ization sensitivity of a panel of primary HIV-1 isolates luciferase reporter gene. demonstrated the dominant effect of V1/V2 masking in Results: Our results showed that the degree of HIV-1 cell restricting the neutralization sensitivity of these Envs. infection increases slowly with time and is strictly propor- Introduction of as few as four mutation in the YU-2 V1/V2 tional to the viral particles concentration, suggesting that domain increased the neutralization sensitivity of this Env to HIV-1 infection is a stochastic process. Furthermore, we a pool of anti-V3 mAbs by >2,500-fold. Several V2-specific confirmed that over time viral infectivity decreases expo- mAbs with ultra-sensitive neutralizing potencies for viruses nentially with a corresponding viral half life. The data sug- expressing the relevant epitopes were mapped to alternate gest that the degree of HIV-1 infection is the net outcome of sequences located in a small region in V2, which in each case two competing dynamic processes: the rate of infection and differed from the clade B consensus sequences at only single the spontaneous decay of HIV-1. Next, we investigated the positions. impact of neutralizing antibodies on these processes. Our Conclusions: The dominant effect of epitope masking by the results demonstrated that antibodies slow down the rate of V1/V2 domain in the resistance of primary HIV-1 Envs to infection and accelerate viral decay. The intensity of this neutralization presents a major hurdle in the development of irreversible inactivation depends on antibody concentration an effective vaccine. Whereas a few mAbs have been identi- and specificity. fied that are not affected by V1/V2 masking, the potencies of Conclusion: We suggest a dynamic model where HIV-1 neu- these mAbs are generally much lower than those of mAbs tralization is achieved through a dual mechanism with neu- directed against variable region epitopes in their unmasked tralizing antibodies slowing down the rate of HIV-1 infection forms. The identification of ultrasensitive neutralization tar- and accelerating viral decay. In this model, HIV-1 particles gets in the V2 domain that are closely related to consensus covered with neutralizing antibodies have a lower probabili- AIDS Vaccine 2006 105 Poster sessions_revGJ 14/8/06 16:34 Page 106 Amsterdam, Netherlands 29 August–1 September 2006 ty to infect due to steric hindrance and to a shorter half life. whether virus trapped in immune complexes will be inacti- This model of HIV-1 neutralization may bring new insight in vated or escape form the endosomal compartment leading to vaccine design strategy. Indeed, it may be more advantageous productive infection. to elicit antibodies that irreversibly inactivate HIV-1. P06-04 P06-03 Neutralization sensitivity of HIV-1 after homo- Mechanisms of HIV-1 inhibition by monoclonal sexual and parenteral transmission antibodies when macrophages and iDC are the target cells E Quakkelaar, F van Alphen, B Boeser-Nunnink, A van Nuenen, F Samad and H Schuitemaker V Holl, M Peressin, K Xu, T Decoville, S Schmidt, AM Aubertin and C Moog Dept. of Clinical Viro-Immunology, Sanquin Research at CLB and Landsteiner Laboratory of the AMC, University of Amsterdam, INSERM U778, University Louis Pasteur of Strasbourg, Institute the Netherlands of Virology, France Background: In newly infected HIV-1 patients, it is assumed Background: Macrophages and iDCs, present at mucosal site, that a relatively neutralization sensitive virus population is are among the first HIV-1 cellular targets infected during sex- responsible for the primary infection. After development of a ual transmission of HIV-1. FcγRs, present at the cell surface neutralizing antibody response in the infected patient, virus- of macrophages and immature dendritic cells (iDCs), play a es with escape mutations will be rapidly selected, resulting in key role in the internalization and the clearance of immune a relatively neutralization resistant virus population. This complexes (IC). We have found an increased inhibitory activ- would imply that in the absence of neutralizing antibodies, ity of 5 well known neutralizing monoclonal antibodies neutralization sensitive HIV-1 variants have a selective (NMAbs) when monocyte-derived macrophages (MDM) or advantage over neutralization resistant viruses. dendritic cells (MDDC) were used as target cells compared to Objectives: To investigate a potential role for the route of T lymphocytes. Moreover, some non-neutralizing MAbs were transmission on the neutralization sensitivity of HIV-1 during able to efficiently inhibit HIV-1 replication on MDC and primary infection, we studied the neutralization sensitivity of MDDC (referred as non-neutralizing inhibitory monoclonal biological HIV-1 clones that were obtained after parenteral antibodies: NNiMAbs) whereas other HIV specific MAbs and homosexual transmission. have no inhibitory activity on these cells. We showed that the Methods: Virus clones were isolated at different time points increased inhibitory activity was due to the participation of after infection. Neutralization sensitivity of the virus clones FcγRs present at the surface of MDC and MDDC. was determined for sCD4, IgG1b12, 2G2, 2F5 and 4E10. In Objectives: To determine the mechanism(s) of FcγR-depen- addition, we analysed envelope sequences to investigate dent HIV-1 inhibition involved in macrophages or iDCs. whether length of V1-V2 and the number of potential N- Methods: HIV-1 binding and entry on MDC and MDDC was linked glycosylation sites (PNGS) were associated with neu- analysed by confocal microscopy using HIV pseudoviruses tralization sensitivity. containing Vpr-GFP . Results: Virus clones of one out of four homosexually Results: No modifications of the percentage of the GFP pos- infected patients were sensitive to IgG1b12 and 2F5 neu- itive MDC and MDDC as measured by flow cytometry could tralization, but resistant to sCD4, 4E10 and 2G12 neutral- be detected in the presence of MAbs suggesting that 1). ization. Of the other three homosexually transmitted NMAbs were not able to significantly decrease the virus patients, all virus clones were resistant to neutralization by binding to cells may be because of the presence of alternative all antibodies tested and sCD4. No differences in neutral- receptors and 2). the capture of viruses by NNiMAbs via ization sensitivity, the number of PNGS and V1-V2 length FcγR may not be the predominant mechanism leading to were found between virus clones that were isolated early virus entry in these cells and/or HIV degradation is increased and late after transmission. when binding of HIV/Abs immune complexes to the FcγR In one of the two parenteral transmission cases studied, occurs. However, in presence of HIV specific antibodies, the increased neutralization sensitivity was observed for localization of HIV particles was different from that observed IgG1b12 of virus clones obtained from the recipient 6 in the absence of Abs. HIV/Abs immune complexes were weeks after transmission. Virus clones that were isolated at rapidly trapped in multivesicular bodies that colocalyze with a later time point in infection were slightly more neutral- FcγRs and endocytic intracellular compartment. This intra- ization resistant, indicative of intrahost evolution of the cellular localization was observed whether Abs were able to virus. The other parenteral transmission case showed equal inhibit HIV replication or not, showing that trapping of HIV high neutralization sensitivity to IgG1b12 and 2F5 for to endocytic compartment via FcγR is not sufficient per see donor and recipient virus clones. for HIV inhibition. Resistance to 4E10 and 2F5 does not correlate with muta- Conclusion: The specificity of the antibodies will determined tions in the crucial residues of these epitopes. 106 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 107 Amsterdam, Netherlands 29 August–1 September 2006 Conclusions: These results indicate a potential role for the Conclusions: Our results indicate that clade-related HIV transmission route in the neutralization sensitivity of newly neutralization serotypes may exist. Clade C immunogens transmitted viruses. may best elicit broadly cross-reactive humoral immunity to HIV-1. These data also suggest that clade B env subunit vac- cines may generate an ineffective, or non-protective, P06-05 immune response in vaccinated subjects at risk for infection with non-B subtypes. Progress towards definition of HIV-1 neutraliza- tion serotypes: broad and potent HIV-1 primary P06-06 isolate neutralization using polyclonal antibodies from subtype C infection Dissecting the neutralizing antibody specificities of broadly neutralizing sera from HIV-1 infected 2 1 1 1 VR Polonis , BK Brown , L Wieczorek , C Bermudez , donors C Bautista2, T Santelli Antonille1, M Garcia1, A Gillis1, ML Robb1, NL Michael2, DL Birx3, FE McCutchan1 H Donners1,2*, AK Gakhal1*, R Pantophlet1, W Johnson3, 5 5 6 6 4 1 Henry M. Jackson Foundation, Rockville, MD USA, 2 Walter J Decker , GM Shaw , F Lee , RW Doms , DD Richman , 2 1 Reed Army Institute of Research, Washington, DC, 3 US Centers G Vanham and DR Burton * for Disease Control, Atlanta, USA These authors contributed equally to this work Background: The use of products derived from multiple 1 The Scripps Research Institute, La Jolla, USA; 2 Institute of clades to develop a globally efficacious vaccine is a central Tropical Medicine, Antwerp, Belgium;3 Harvard Medical School, approach in several HIV-1 vaccine development programs; MA, USA; 4 Center for AIDS Research, University of California however, the relevance of genetic clades in eliciting a broadly San Diego, San Diego, USA; 5 University of Alabama, Alabama, reactive, protective immune response against HIV-1 remains USA; 6 University of Pennsylvania, Pennsylvania, USA unknown. Previous studies of cross-clade neutralization have demonstrated little or no relationship between HIV-1 geno- Background & aim: The ability to elicit neutralizing anti- type and neutralization serotype. Recent evidence using mon- bodies is considered as an important feature of an effective oclonal antibodies (mAbs) has indicated that clade-related candidate HIV-1 vaccine. To gain better understanding of the cross-neutralization patterns may exist. We have re-examined regions of the HIV-1 envelope spike that are vulnerable to this question using full-length sequenced isolates and geneti- neutralizing antibodies, the neutralizing antibody specificities cally defined plasma pools. of three asymptomatic long-term HIV-1 infected donors with Objective: To assess cross-clade neutralizing antibody patterns broadly neutralizing sera were analysed. among the six major clades and circulating recombinant forms Methods: Two donors (LT2, FDA2) are infected with clade B (CRFs) of the HIV pandemic (clades A-D, CRF01_AE, isolates and the third (ITM1) with a clade A isolate. The sera CRF02_AG) and investigate the existence of possible neutral- of these donors are being characterized by competition neu- ization serotypes using a large panel of genetically defined tralization assays and depletion experiments with recombi- reagents from patients infected with pure clades or CRFs. nant proteins. Methods: The ELISA method was used to determine binding Results: The broad neutralizing activity was mediated by the antibody titers to multiple env antigens for the six clade-specif- polyclonal IgG fraction of donor sera. A series of competition ic plasma pools. Clade-specific pools showing >50% neutral- neutralization assays were carried out using a panel of peptides ization at a 1:40 test dilution, as well as a serum pool from HIV representing portions of the V3 loop of clade B isolates and the positive patients from the US (USHIV+), sCD4, and Tri-mAb (a membrane proximal external region (MPER) of gp41. Clade B cocktail of equal concentrations of the 2F5, 2G12, and serum neutralization was not inhibited by any of the peptides, IgG1b12 mAbs), were all tested against 60 isolates in a PBMC indicating that antibodies directed against these epitopes are p24 reduction assay to determine 50% neutralization titers. unlikely to explain the neutralizing activity observed in donor Non-parametric statistics and two-dimensional hierarchical sera. Additionally, the sera did not significantly neutralize a cluster analysis were applied to the data matrix. chimeric SIV construct displaying the highly conserved MPER Results: No relationship between binding antibody titer and epitope recognized by the broadly neutralizing antibody 4E10. neutralization potency was identified. The clade C plasma Polyclonal IgG fractions were depleted on recombinant clade B pool neutralized the greatest number of viruses with the gp120 monomers and the effects on cross-clade neutralization greatest breadth, and showed significantly higher neutral- was examined. ELISA analysis of depleted samples showed that ization titers as compared to all other antibody pools com- antibodies that bind to gp120 monomer are completely bined (P<0.0001). An association was observed between removed from the polyclonal donor IgG. However, neutraliza- isolates of the same clade using two-dimensional cluster tion results indicated that only a proportion of the neutraliza- analysis, suggesting the presence of serotypes or “seroclus- tion activity is specific for the monomeric gp120. ters”. Pairs of very poorly cross-neutralizing clades were Conclusions: Cross-clade neutralization in broadly neutralizing also identified. sera may be a complex combinations of specificities. AIDS Vaccine 2006 107 Poster sessions_revGJ 14/8/06 16:34 Page 108 Amsterdam, Netherlands 29 August–1 September 2006 P06-07 P06-08 The antigenicity and immunogenicity of a clade c Antibody responses to MPER linear epitopes trimeric gp140 vaccine candidate occur more frequently in HIV-2 than in HIV-1 infections NC Sheppard1, SA Jeffs2, SM Vieira2, S Davies3, MJ Neuberger3 and QJ Sattentau1 TD Moon1, T Olivier1, JM Decker2, F Bibollet-Ruche2, PA Goepfert2, GM Shaw2, SL Rowland-Jones3, A Jaye3, CB 1 Sir William Dunn School of Pathology, University of Oxford, Hicks4, L Sutherland4, R Scearce4, BF Haynes4 and JE Oxford, UK; 2 Imperial College London, London, UK; 3 University Robinson1 of Cambridge, Cambridge, UK 1 Tulane University Health Sciences Center, New Orleans, LA, Background: The C clade of HIV-1 is responsible for >50% USA; 2 University of Alabama at Birmingham, Birmingham, AL, of infections globally, yet little is known about the antigenic- USA; 3 MRC Laboratories, Fajara, Banjul, the Gambia; 4 Duke ity and immunogenicity of clade C envelope glycoproteins University Medical Center, Durham, NC, USA (Env) as most studies have focused on clade B. Gp140 CN54 represents the trimeric extracellular Env portion from prima- Background: Human Mabs 2F5 and 4E10 exhibit broad neu- ry isolate HIV-1 . Characterization of Clade C gp140 97CN54 tralizing activity against two closely associated epitopes in molecules will assist in the development of vaccines that aim the MPER region of HIV-1. It is currently believed that anti- to induce neutralising antibody responses. body reactivity to these epitopes during HIV-1 infection is rel- Objectives: To investigate the antigenicity of gp140 ; to CN54 atively rare and that non-neutralizing antibodies predominate develop novel mAb to gp140 in immunoglobulin– CN54 in the humoral response to gp41. Little is known about the humanised (IgHu) mice; to develop novel adjuvant strategies frequency of antibody recognition of MPER epitopes during capable of inducing high-titre antibody responses. HIV-2 infection. Methods: Antigenic characterization of gp140 was car- CN54 Objectives: To compare the antibody responses to MPER epi- ried out by ELISA and surface plasmon resonance (SPR). The topes in HIV-1 and HIV-2 infection. BAB5 strain of IgHu mice expressing a human IgM repertoire Methods: Using ELISA, we determined the frequency of anti- were immunised with gp140 and IgM mAbs were isolat- CN54 body reactivity to peptides representing HIV-1 and HIV-2 ed by hybridoma fusion. The novel mAbs were characterized MPERs in a panel of unselected sera collected from HIV-1 in terms of their epitope specificity, recognition of heterolo- infected persons from the U.S. (n=121) and a panel of sera gous isolates, binding affinities and ability to neutralise or fix from HIV-2 infected persons from the Gambia (n=72). complement against HIV-1 . Immunogenicity studies in 97CN54 Results: Only 48% (59/121) reactivity was seen in the HIV-1 Balb/c mice used novel combinations of transfection reagents serum panel against either the 2F5 or 4E10 HIV-1 MPER and Toll-Like Receptor (TLR)-ligands. The affect of Env gly- peptides. By contrast, 100% (72/72) of HIV-2 positive sera cosylation pattern on immunogenicity was also investigated. showed reactivity with at least one of several HIV-2 MPER Results: Gp140 is antigenically relevant. Two IgM mAb CN54 peptides containing sequences homologous to 2F5 and 4E10 N3C5 and N03B11 were isolated, which displayed high epitopes. Both serum panels showed greater reactivity against binding affinities (K ∼1 nM), and recognised geographical- D 4E10 compared to 2F5 homologues (98% versus 88% for ly distant isolates from clades C and A. Insect cell expressed HIV-2 and 35% versus 27% for HIV-1). IgG subclass analy- Env was more immunogenic than mammalian cell material sis revealed that 30% of HIV-2 sera contained IgG-3 MPER (titre difference >2-fold, P<0.05) and combinations of CpG antibodies while the MPER antibodies in the HIV-1 sera were or Poly(I:C) with transfection reagents induced greater nearly all IgG-1. immune responses than the either component used sepa- Conclusions: These data suggest that the MPER is more rately (P<0.05). immunogenic in HIV-2 infections than in HIV-1 infections. Conclusions: Useful mAb reagents have been developed Studies are ongoing to determine the contribution of the HIV- against clade C and A, that will allow further antigenic char- 2 MPER specific antibodies to virus neutralization. acterization. The adjuvant strategies developed here could be applied to many vaccine antigens. 108 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 109 Amsterdam, Netherlands 29 August–1 September 2006 P06-09 P06-10 Neutralizing specificity mapping in complex poly- In vivo evolution of coexisting R5 and X4 HIV clonal sera type 1 variants Y Li1, B Welcher1, K Svehla1, M Tang1, Y Shu1, EM Bunnik, A van Nuenen and H Schuitemaker J Guenaga1, S Phogat1, M Connors 2, J Mascola1 and R Wyatt1 Sanquin Research and Landsteiner Laboratory of the AMC, University of Amsterdam, Amsterdam, The Netherlands 1 Vaccine Research Center, NIAID/NIH, Bethesda, MD, USA; 2 Laboratory of Immunoregulation, NIAID/NIH, Bethesda, MD, Background: In approximately 50% of HIV-1 infected indi- USA viduals, progression to AIDS is preceded by the appearance of CXCR4-using (X4) syncytium inducing virus variants Background: One of the major challenges of HIV-1 vaccine that evolve from CCR5-using (R5) non-syncytium inducing development is the elicitation of broadly and potent neutral- virus variants. We have previously shown in four subjects izing antibodies. In our previous studies, we have reported that early (R3)R5X4 variants are more sensitive to inhibi- that gp140 (-/GCN4) trimeric immunogens derived from the tion by CXCR4 antagonists AMD3100 and T22 than late primary YU2 isolate could elicit improved breadth and X4 variants, demonstrating the ongoing evolution of X4 potency of neutralization against HIV-1 isolates compared variants in vivo. to monomeric gp120 in rabbits and guinea pigs. To map the Objectives: This study was performed to examine a possible neutralizing specificity of the animal sera, we initially devel- role of humoral immunity on the evolution of R5 and X4 oped a peptide competition and neutralization assay using HIV-1 variants. peptides derived from HIV-1 Env protein (Grundner et al, Methods: We examined the neutralization sensitivity of early Virology 2005; Li et al, J Virol 2006). In most gp120-inoc- and late X4 variants of five patients for sCD4 and monoclon- ulated animals YU2 neutralization activity was inhibited by al antibodies IgG1b12, 2F5, 2G12 and 4E10. In addition, the a single peptide located at the C-terminus of V1 loop, sensitivity of the coexisting R5 variants for these neutralizing whereas the neutralizing activity elicited by gp140-GCN4 agents was analysed. trimers was not predominantly against V1, V2, V3, or Results: In contrast to the decreased sensitivity of late X4 2F5/4E10 epitopes. variants to CXCR4 antagonists observed previously, we did Objectives: To characterize the neutralizing specificity of com- not observe a difference in neutralization sensitivity between plex polyclonal sera of human or animal subjects. early and late X4 variants, or between early and late R5 vari- Methods: we developed and validated a set of novel tools ants. However, at both early and late time points the R5 vari- using variant gp120 glycoproteins conjugated to solid phase ants were less sensitive to sCD4 and IgG1b12 compared to beads with broadly neutralizing monoclonal antibodies as the coexisting X4 variants. This difference in neutralization positive control in the assay. Sera/antibodies were incubated sensitivity between R5 and X4 variants was not observed for with the gp120 conjugated beads and then subjected to neu- 2F5, 2G12 or 4E10. tralization assay. By comparing the sera/antibodies neutraliza- Conclusions: The results obtained here show that early tion activity before and after the bead treatment, the (R5)X4 variants are relatively neutralization sensitive and do neutralizing specificity can be localized to the various gp120 not evolve towards neutralization resistance against CD4 neutralizing epitopes. binding site directed agents during the course of infection. Results: With these set of tools, we found that the broadly This may indicate that for the evolution of the virus towards neutralizing specificity from a sub-set of human patient sera the usage of CXCR4 as a coreceptor, a relatively open, neu- was predominantly focused on the conformation-dependent tralization sensitive envelope conformation is needed. We epitopes on gp120, and most likely directed to the CD4 bind- speculate that a decrease in humoral pressure would thus have ing site. to occur in order to allow the appearance of X4 variants. We Conclusions: These set of mapping tools could enable us to are currently addressing this question by investigating the map neutralizing specificity toward discontinuous and con- neutralization of coexisting R5 and X4 variants by autolo- formation-dependent neutralizing epitopes, and can also be gous sera. applied to analyse the neutralizing specificity of animal/human vaccine serum samples. AIDS Vaccine 2006 109 Poster sessions_revGJ 14/8/06 16:34 Page 110 Amsterdam, Netherlands 29 August–1 September 2006 P06-11 P06-12 Neutralization phenotype of clade C human Immunogenicity of constrained HIV-1 gp140CF immunodeficiency virus type 1 gp160 clones from envelope oligomers that express membrane proxi- acute/early sexually acquired infections in India mal external region epitope (MPER): evidence for differential regulation of antibodies to the 2F5 SS Kulkarni1, H Tang2, SP Tripathy1, H Gao2, MR and 4E10 gp41 nominal epitope Thakar1, L Morris3, RC Bollinger4, RS Paranjape1 and D Montefiori2 H-X Liao1, SM Alam1, RM Scearce1, M McAdams1, LL Sutherland1, R Parks1, S Van Leeuwen1, S-M Xia1, 1 National AIDS Research Institute (NARI), Pune, India; 2 Duke J Robinson2 and BF Haynes1 University Medical Centre, Durham, NC, USA; 3 National Institute for Communicable Diseases, Johannesburg, South 1 Duke Human Vaccine Institute, Duke University Medical Africa; 4 Johns Hopkins University, Baltimore, MD, USA Center, Durham, NC, USA; 2 Tulane University Medical Center, New Orleans, LA, USA Background: HIV-1 exhibits an extraordinary degree of genetic variation that has given rise to multiple genetic sub- Background: Design of an immunogen that can elicit broad- types and circulating recombinant forms (CRFs) of the virus. ly reactive neutralizing antibodies against HIV-1 primary iso- Information on the neutralization properties of each major lates is a major goal for HIV-1 vaccine development. genetic subtype and CRFs will inform vaccine design and Objective: To evaluate immunogens containing conserved facilitate standardized assessments of vaccine-elicited neutral- epitopes for eliciting broadly reactive neutralizing antibodies. izing antibody responses. Methods: In this study, we produced constrained HIV-1 Objectives: To characterize neutralization properties of rgp140DCF -Mab A32 complexes with stably expressed CON6 acute/early sexually acquired clade C strains of HIV-1 from co-receptor binding sites, as well as MPER 4E10 and 2F5 India and to compare them with acute/early sexually binding sites, and compared these complexes as immunogens acquired clade C strains from Africa. with un-constrained HIV-1 rgp140DCF . CON6 Methods: Viruses were isolated from PBMC of seven males Results: We determined that stable enhanced expression of and four females identified from a cohort of persons at risk the coreceptor binding site in context of Env-Mab A32 com- of HIV infection who were attending STD clinics in Pune, plexes was not sufficient for induction of broadly reactive India. All viruses were isolated after a mean interval of 14 neutralizing antibody (NA). In addition, availability of the days (SD=14, Min=2, Max=54) after p24 antigen confirma- MPER neutralization epitopes in the constrained HIV-1 tion and were sexually acquired. Env-rev cassettes were PCR gp140 construct was also not sufficient for induction of amplified from cultured PBMC DNA and cloned in the broadly reactive NA. Further, we found that induction of pcDNA 3.1/V5-His-TOPO directional vector. Functional antibodies against 2F5 and 4E10 epitopes was differentially clones were used to produce Env-pseudotyped viruses by co- regulated in rabbits, guinea pigs and mice. Immunization of transfection with an env-minus backbone vector (pSG3Denv) animals with whole Env, or nominal MPER peptide epitopes in 293T cells. Neutralization phenotypes were characterized in formulations designed to break immunological tolerance in a validated TZM-bl Luc reporter assay using sCD4, South (TLR-9 agonists in oil) readily induced antibody to the 2F5 African HIV-1-positive plasma, and the broadly neutralizing gp41 region, but neutralization breadth of these sera were MAbs IgG1b12, 2G12, and 2F5 and 4E10. not enhanced over that seen with Env in traditional TLR-4 Results: The Indian viruses were significantly less susceptible adjuvants. In contrast, we found no detectable binding anti- to inhibition by sCD4 (P<0.05), less sensitive to neutraliza- body induced against the 4E10 gp41 epitope in rabbits or tion by IgG1b12 and clade C plasma samples from South guinea pigs with the same immunogens. Africa compared to the African viruses. An exception was Conclusions: Our data demonstrate that HIV-1 Env neutraliz- unusually neutralization sensitive Indian virus, neutralized by ing epitopes constrained with crosslinked A32 Mab were not all reagents except 2G12 and 2F5; reasons for this high sen- sufficient for induction of broadly reactive neutralizing anti- sitivity are being investigated. Like the African viruses, the bodies in rabbits or guinea pigs. Moreover, in all species tested, Indian viruses were neutralized poorly by 2G12 and 2F5 and antibody responses to the 2F5 versus 4E10 epitopes of gp41 were highly sensitive to 4E10. Assays with Indian HIV-1 pos- were differentially regulated, suggesting immune control of itive sera are in progress. these antibody specificities. Conclusions: This is the first systemic characterization of the neutralization phenotypes of acute/early sexually transmitted strains of clade C HIV-1 gp160 from India. The results indicate that clade C viruses from distant geographical locations share certain common neutralization determinants but differ in oth- ers. These differences suggest that optimal HIV-1 vaccine design and testing will need to consider the impact of intra- clade variability between international vaccine trial sites. 110 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 111 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 7: IMMUNE ESCAPE P07-01 P07-02 Designing a vaccine strategy: implications of viral Characterization of naturally occurring 4E10 resis- escape and SHIV-specific CD8 T cells at transmis- tant viruses in a subtype C HIV-1 infected child sion and during acute infection ES Gray1, P Moore1, I Choge1, T Meyers2 and L Morris1 E Rollman1, CS Fernandez1, MZ Smith1, CJ Batten1, R De Rose1, J Reece1, V Venturi2, MP Davenport2 and SJ Kent1 1 AIDS Research Unit, National Institute for Communicable Diseases, Johannesburg , South Africa; 2 Harriet Shezi Clinic, 1 Department of Microbiology and Immunology, University of Chris Hani Baragwanath Hospital, University of the Melbourne, Australia; 2 Department of Haematology, Prince of Witwatersrand, Johannesburg, South Africa Wales Hospital and Centre for Vascular Research, University of New South Wales, Kensington, Australia Background: The broadly neutralizing monoclonal antibod- ies (MAb) 4E10 recognizes a linear epitope in the extreme C- Background: CD8+ T cells are important in the control of terminal membrane proximal region (MPR) of gp41. This HIV/SHIV infection. Escape from virus-specific CD8+ T cells region is particularly attractive for vaccine design because it can impart a fitness cost. is highly conserved amongst HIV-1 strains. Objective: To study escape from virus specific CD8+ T cells I Objective: To study the molecular determinants within the relation to viral fitness cost. We used vaccinated, MHC- envelope glycoprotein that defines the resistant to neutraliza- matched Macaca nemestrina (pigtail macaques) to study the tion by the MAb 4E10 and 2F5. relationship between SIV Gag KP9-specific CD8+ T cells Methods: Multiple env genes were cloned from a virus iso- 164-172 with escape at the KP9 epitope. lated from an HIV-1 positive child, who had been infected Methods: Ten out of fourteen Mane-A*10+ macaques for more than 7 years and was classified as a long term non- enrolled in the study were vaccinated by DNA and a fowl progressor. Three functional envelope clones were pheno- poxvirus vector expressing SIV Gag. All fourteen macaques typically characterized with respect to their sensitivity to were challenged with an escape mutant (EM) strain of neutralization by 2F5 and 4E10 using a single cycle neu- SHIV consisting of 90% KP9 (164KRFGAEVVP) EM tralization assay in JC53-bl cells. We also sequenced 50 mn229 virus. Plasma viral RNA was cloned and sequenced across molecular clones of gp41 to calculate the frequency of the KP9 at closely spaced time-points following challenge. An mutant variants in this patient. MHC class I tetramer (Mane-A*10-KP9) was used to detect Results: Two of the three functional envelope clones were KP9-specific CD8+ T cells. resistant to 4E10 neutralization. One of these clones, Results: Variable levels of transient reversion to wild-type TM20.13, was also sensitive to 2F5 which is unusual (WT) KP9 sequence followed by re-escape were observed amongst subtype C viruses. In this clone the resistance to during acute infection. Reversion to WT KP9 at day 10, 13, 4E10 could be explained by the substitution of phenylalanine and 20 post-challenge correlated inversely with KP9-specific to leucine at position 673. F673 has been described as criti- T cells present at virus exposure (r≤-0.73, P<0.003, cal for 4E10 binding and is 100% conserved in all the HIV-1 Spearman rank test). Expansion of KP9-specific T cells 7 days group M sequences from the Los Alamos database. after challenge also inversely correlated with reversion to WT Frequency analysis showed that F673L was present in KP9 sequence at day 10 and day 13 (P≤0.04). approximately 30% of the viral variants isolated from this The reversion to WT also influenced viremia. The proportion patient and in all cases was linked to the presence of an intact of KP9 WT virus at the peak of viral load inversely correlat- 2F5 epitope. A second clone, TM20.6 was resistant to 2F5 ed with the timing of that peak (P=0.03) since the fitter WT and sensitive to 4E10 typical of subtype C viruses. The third virus results in faster viral replication and an earlier peak in clone, TM20.5, was resistant to both 4E10 and 2F5 despite viral load. the presence of the 4E10 epitope. For this clone resistance to Conclusion: Effector T cells present at transmission and 4E10 was conferred by changes in gp120 and the cytoplasmic capable of rapid expansion are vital for the prevention of domain of gp41. reversion to wild-type and control of HIV/SHIV viremia. Conclusion: Escape from neutralizing antibodies that tar- get the MPR is possible through changes in gp120 and the cytoplasmic domain of gp41 and does not necessarily involve changes in the MPR itself. This is the first report of a naturally occurring neutralization resistant variant to the 4E10 MAb. AIDS Vaccine 2006 111 Poster sessions_revGJ 14/8/06 16:34 Page 112 Amsterdam, Netherlands 29 August–1 September 2006 P07-03 P07-04 Cytotoxic T lymphocytes and the escape mutants Envelope regions involved in the antigenic evolu- of HIV-1 during and after HAART with structured tion of sequential HIV-1 isolates treatment interruptions in acutely infected HLA- A24-positive patients P Nyambi1,2, S Burda1, L Heyndrickx3, W Janssens3, G Vanham3 and S Zolla-Pazner1,2 J Tanuma1, K Teruya1, M Fujiwara2, M Takiguchi2, and S Oka1 1 New York University School of Medicine, New York, NY, USA; 2 V.A. Medical Center, New York, NY, USA; and 3 Institute of 1 AIDS Clinical Center, International Medical Center of Japan, Tropical Medicine, Antwerp, Belgium Tokyo, Japan; 2 Division of Vial Immunology, Center for AIDS research, Kumamoto University, Kumamoto, Japan Background: A polyvalent vaccine that would prevent infec- tion by various HIV strains would include shared immuno- Background: Cytotoxic T lymphocytes (CTLs) are thought to gens from different viruses. However, the possibility exists play a critical role in controlling HIV replication. Although that epitopes on viruses could evolve overtime to escape the the limited durability of immune activation of structured effect of such a vaccine. treatment interruptions (STIs) has been reported, the escape Objective: To identify HIV-1 envelope regions that may mechanism from CTLs during STIs has remained unclear. evolve over time to escape neutralization by polyclonal anti- Objectives: To evaluate HIV-1 specific CTLs and the escape bodies in heterologous plasma. variants during and after HAART with STIs in acutely HIV- Methods: Twelve broadly neutralizing plasma obtained from 1 infected patients. HIV-1 subtype B-infected individuals were tested for their Methods: Prospective study of HAART with 5 series of STIs ability to neutralize sequential primary HIV-1 subtype B was conducted in acutely HIV-1 infected patients. STIs were viruses obtained at 6 months intervals over a 3–4 year period carried out when VL decreased to <50 copies/ml and CD4 from five individuals. The neutralization assays were per- increased to >300/mm3. We examined HIV-1 specific CTLs formed with GHOST-CD4-R5 cells and plasma diluted at by tetramer analysis and sequenced an immunodominannt 1:40. Ten gp120 clones from each sequential virus were also CTL epitope in the nef gene (Nef138-10) in HLA-A24-posi- sequenced and analysed to identify amino acid changes. tive patients. Results: Viruses from the five patients obtained at the first Results: A total of 26 individuals were enrolled to our study visit were neutralized by all 12 plasma with neutralizing (24 male, mean age 33 years, median baseline VL 5.21 ± 0.93 capacities between 42% and 75%. This neutralizing pattern log copies/ml, CD4 413 ±251/mm3). Among protocol-com- remained stable for all sequential viruses in the first two years pleted 15 patients, the average VLs of every 6 months after of follow-up with neutralizing capacity as high as 95%. treatment cessation were 3.54 ±0.86, 3.74 ±0.73, and 3.90 Subsequently, virus obtained from one of the five patients in ±0.39 log copies/ml. 12 out of the 15 patients were positive the third year of follow-up became totally resistant to neu- for HLA-A2402 and studied for CTLs and escape variants. tralization by all the heterologous plasma in the panel, while Initially, 7 out of 12 patients had a Y-to-F substitution at the viruses obtained from the remaining four patients did not second position (Nef138-2F) in their nef gene at study enroll- show a significant loss of neutralization by the panel of plas- ment and wild-type (Nef138-WT) specific CTLs had not been ma. Sequence analysis of the gp120 envelope of all the virus- increased during STIs in those patients. On the other hand, es studied revealed a progressive increase in the number of Nef138-WT specific CTLs had increased in the left 5 insertions and deletions in the V1V2 regions as well as in the patients, which led to VL suppression. However, the rever- V4 region, while nonsynonymous changes progressively sion of Nef138-2F variant was observed in all the 5 patients increased in the V1V2 and C3 regions. These changes were soon after treatment discontinuation and Nef138-WT specif- more extensive in the sequences of the virus that became ic CTLs decreased thereafter. Although a slight increase of resistant to neutralization. Intriguingly, the V3 region Nef138-2F specific CTLs was observed after Nef138-2F remained constant over time in all sequential viruses tested. mutants were evolved, VLs had no longer been suppressed in Conclusion: This study suggest that mutations will occur pro- 3 patients. The cytotoxicity assay revealed Nef138-2F specif- gressively over time in certain variable (V1V2, V4) and con- ic CTLs had a less killing-activity compared to Nef138-WT stant (C3) regions of gp120 resulting in the eventual escape specific CTLs. from neutralization by polyclonal antibodies. Therefore, this Conclusions: The high frequency of Nef138-2F variant trans- study should inform vaccine design on potential epitopes that mission and the positive selection for Nef138-2F in the pres- might evolve to escape the effect of a vaccine. ence of CTLs were documented in acutely infected patients during the STI trial. These findings might contribute to know the escape mechanisms of HIV-1-specific immune stimulato- ry interventions. 112 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 113 Amsterdam, Netherlands 29 August–1 September 2006 P07-05 P07-06 Escape from an HLA-B57-restricted CTL response Towards identifying correlates of protective in capsid reduces HIV-1 fitness and increases immunity: Enhanced detection of HIV-specific sensitivity to APOBEC3G-mediated host restriction cellular immune responses using peptides based on the patient’s own viral quasispecies MA Brockman, A Schneidewind, M Lahaie, BD Walker, and TM Allen S Norley1, JL Richardt1, B Stauber1, D Schürmann2, O Hohn1 and R Kurth1 Howard Hughes Medical Institute and Partners AIDS Research Center, Massachusetts General Hospital, Boston, 1 Robert Koch-Institute, Berlin, Germany; 2 Charité, Virchow Massachusetts, USA Clinic Campus, Berlin, Germany Background/Aim: HLA-B57 is associated with control of Aims: The use of peptides based on standard or consensus HIV-1 infection and the dominant CD8 response targets epi- sequences of HIV to stimulate cells underestimates the actual tope TW10 (TSTLQEQIGW) within a highly conserved strength and nature of the immune response because many of 240-249 region of capsid p24. Recent demonstrations of the impact of these sequences will not actually exist in HIV quasispecies drug- and immune-induced mutations prompted us to exam- infecting the patient. We therefore decided to take the ‘brute ine the effects of CTL escape mutations on HIV-1 fitness. force’ approach of analysing CTL responses in a small cohort Methods: We constructed mutants in HIV-1 strain NL4-3 of therapy-naive HIV-infected volunteers using peptides encoding the primary T N escape mutation in TW10, as based on the patient’s own autologous virus sequences. 242 well as natural putative compensatory substitutions. Using a Methods: Blood was sampled at different time points, HIV FACS-based assay, the fitness of variant viruses was com- genes were sequenced directly from plasma virus and, depend- pared using GFP-reporter T cell lines. Infectivity was also ing on the individual’s MHC-haplotype, peptides based on the assessed using T cells that lack APOBEC3G expression and autologous sequences were used to enumerate peptide-specific cells transfected with this gene. Kinetics of viral reverse tran- CTLs by ELISpot. In a separate study assessing the efficacy of scription was determined by Q-PCR. immunologic pressure on HIV, amino acid variability within Results: We observed that the T N mutation reduced HIV- the regulatory genes of HIV sampled shortly after infection 242 1 replication 10-fold in Jurkat T cells. Viral growth was sub- was correlated to their location within CTL epitopes. stantially restored in the presence of arsenic trioxide (As O ), Results: Surprisingly, although each region tested was known 2 3 a compound known to inhibit host restriction activity, but to be a CTL epitope for the patient’s corresponding HLA- this defect appeared to be independent of Trim5a. In con- type, in 445 of 521 cases (85%), responses were seen to nei- trast, the T N mutation reduced HIV-1 replication up to 10- ther the standard (HXB2) sequence nor to any of the 242 fold in CEMss-derived cells transfected with APOBEC3G, autologous variants. However, although 16 (3.1%) responses while no fitness defect was seen in parental cells lacking this were detected to HXB2 peptides alone and 18 (3.5%) to both protein. Notably, natural compensatory mutations located HXB2 plus at least one autologous variant, 42 (8.1%) within the cyclophilin A (CyPA) binding loop of p24 partial- responses were detected using the autologous sequences only. ly restored viral replication in the presence of APOBEC3G. The use of autologous sequences therefore more than dou- Single-cycle infectivity assays indicated that the APOBEC3G- bled the number of positive responses that would have been mediated effect was observed during the entry phase. Viral observed using standard HXB2 peptides only. In the second DNA production by the T N mutant was reduced post- study, comparing variability within regions predicted to act 242 infection, indicating a block during reverse transcription in as CTL epitopes failed to provide clear evidence for CTL-dri- the presence of APOBEC3G. ven viral evolution. Conclusions: Escape in B57-TW10 reduces HIV-1 replication Conclusions: In studies designed to analyse in detail the in T cells by an APOBEC3G-dependent mechanism and is ongoing immune response to infection (e.g. while attempting associated with a block in viral reverse transcription. The to define the elusive ‘correlates of protective immunity’), it is T N mutation can be rescued partially by natural compen- advisable to use reagents based on the viruses actually present 242 satory changes in the CyPA binding loop. The phenotype of in the individual patient, despite the immense effort involved. T N mutant indicates that CTL escape may alter capsid The evolution of HIV in the face of vigorous T-cell responses 242 uncoating, thereby increasing viral sensitivity to is complex and multi-factorial. APOBEC3G-mediated restriction in the target cell. These results highlight the importance of p24-directed CTL for vac- cine design and suggest that immune pressure on capsid can promote the generation of HIV-1 mutants with reduced replicative fitness. Supported by the Howard Hughes Medical Institute and NIAID grants AI054178 and AI067078. AIDS Vaccine 2006 113 Poster sessions_revGJ 14/8/06 16:34 Page 114 Amsterdam, Netherlands 29 August–1 September 2006 P07-07 P07-08 Development of broadly reactive neutralizing Therapeutic vaccination with HIV nef decreases monoclonal antibody KD247: implications for patient-derived Nef protein sequence variability passive immunotherapy D Hoffmann1, J Seebach1, HM Schätzl1, FD Goebel2, S Matsushita1, J Shibata1, A Honda1, T Murakami2, A Cosma3, K Strimmer4 and V Erfle3 Y Eda2, A Koito1 and K Yoshimura1 1 Institute of Virology, Technical University of Munich, Munich, 1 Center for AIDS Research, Kumamoto University, Kumamoto, Germany; 2 Medizinische Poliklinik, University of Munich, Japan; 2 The Chemo-Sero-Therapeutic Research Institute, Munich, Germany; 3 GSF-National Research Centre for Kumamoto, Japan Environment and Health, Institute of Molecular Virology, Munich, Germany; 4 Department of Statistics, University of Background: We produced a high-affinity humanized mono- Munich, Munich, Germany clonal antibody (mAb) KD-247 by a sequential immunization with V3 peptides. KD247 effectively neutralized primary Background: With HIV persisting lifelong in infected per- clade B viruses in vitro and sterile protection was obtained by sons, therapeutic vaccination is a widely investigated concept a single administration of a high-dose of the mAb in a highly to control virus replication. CD8 and CD4 responses to such pathogenic SHIV model. immunizations have been measured, however it remains Objectives: To further investigate KD-247 as a passive unclear whether and to what extent they actually affect the immunotherapeutic agent we tested its capability to contain patient’s HIV population. the spread of a quasi species of HIV-1 from patients by an ex Objectives: To study the impact of a triple therapeutic vacci- vivo neutralization assay. We also induced HIV-1 variants nation with nef delivered by a recombinant MVA vector on escape from KD-247 in vitro using R5 virus, JR-FL and the inter-individual Nef protein sequence.variability. defined the virological properties of pseudotyped virus carry- Methods: RNA was extracted from patient plasmas, nef was ing the escape-associated mutation. amplified by PCR. Analysis of cloned sequences was done Methods: The CD8-depleted peripheral blood mononuclear with vector-specific primers. The inter-individual variability cells (PBMC) obtained from HIV-1 infected patients were of Nef protein sequences was determined in 6 chronically stimulated and cultured in the presence or absence of KD- infected persons before and after therapeutic vaccination. To 247. The culture supernatants were collected and tested for this end, protein sequences were aligned pairwise and the p24 antigen production. To select HIV-1 variants resistant to proportion of amino acid differences were calculated. KD-247 we exposed PM1-CCR5 cells to JR-FL and the virus Differences in variability were tested by an exact T-test based was serially passaged in the presence of increasing concentra- on log-transformed variance-stabilized data. tions of KD-247. Results: Before immunization the mean pairwise variability Results: The epitope of KD-247 was generally matching with of patient-derived Nef protein sequences was 0.173 (stan- the V3 sequences of various clones isolated from plasma and dard deviation 0.018; range 0.144–0.200). After vaccination PBMC of two patients. Complete or strong inhibition of viral the respective values were 0.144 (0.020; 0.113–0.179). The replication was observed when the patients’ PBMC were cul- mean variability was significantly different between the two tured with a high concentration of KD-247. We induced a time points (P<0.005). In 5 matched control persons exem- neutralization escape mutant at passage 8 in the culture plarily tested, Nef protein sequence variability did not sig- where HIV-1 was propagating in the presence of KD-247 JRFL nificantly change. (1 mg/ml) and we found an amino acid substitution, Gly to Conclusions: The presented data indicate a selective pressure Glu at position 314 (G314E), in the V3-tip. Neutralization exerted by the immune response to nef vaccination limiting resistance of the G314E mutant to KD247 was confirmed by the protein’s natural variability. To our knowledge this is the a neutralization assay using a pseudotyped virus with G314E first proof of such an immediate effect of therapeutic vacci- mutation. Unexpectedly, this mutant virus was sensitive to nations on the individual patient’s virus population. CCR5 inhibitors, RANTES, rsCD4 and anti-CCR5 mAb compared with wild type virus. We also evaluated the anti- HIV-1 interactions between KD-247 and various CCR5 inhibitors in vitro. Combinations of KD-247 with the CCR5 inhibitors showed highly synergistic interactions. Conclusions: These results may have an implication of pas- sive immunotherapy using KD247 in combination with CCR5 inhibitors. 114 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 115 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 8: VIRAL DIVERSITY P08-01 P08-02 Increasing genetic distance to HIV-1 subtype B Evolution of viral populations in individuals and F consensus sequences in the Brazilian epi- infected with single and multiple HIV subtypes demic: a challenge for vaccine development AL Malaza1, J Grobler1, L Mboko2, M Hoelscher3, F G Bello, ML Guimarães, SL Chequer-Fernandez, McCutchan4, C Williamson1 and the HISIS Study Team WA Eyer-Silva, JC Couto-Fernandez, SL Teixeira, and MG Morgado 1 Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, South Africa; 2 Mbeya Medical Laboratory of AIDS and Molecular Immunology, Oswaldo Cruz Research Programme, Mbeya Tanzania; 3 Dept Infectious Institute – FIOCRUZ, Rio de Janeiro, Brazil Diseases and Tropical Medicine, University of Munich, Germany; 4 Henry M Jackson Foundation, Maryland, USA Background: The ever increasing HIV-1 diversity at popula- tion level poses a major obstacle for an AIDS vaccine devel- Introduction: Studies on individuals infected with multiple opment. The use of artificial “central” HIV sequences HIV subtypes provide insights into correlates of protection, (consensus, most recent common ancestor, or center-of-the- as well as the potential impact of high diversity and recombi- tree) as reference strains in vaccine designs can help to over- nation on disease progression. This study forms part of the come the challenge of this high HIV-1 diversity. HIV Superinfection Study (HISIS) which monitored a group Objective: To analyse the advantages of using subtype-specif- of high risk women in Tanzania for infection and superinfec- ic consensus sequences as vaccine strains to overcome the tion over a period of three years. problem of the growing HIV-1 diversity in the Brazilian HIV- Objective: To monitor viral population dynamics in individu- 1 epidemic. als from Mbeya, Tanzania, who were either infected with a Methods: HIV-1 env V3 sequences (210nt) with a known sam- single or with more than one HIV subtype. pling year isolated from Brazilian HIV-1 positive patients Methods: A cohort of women at high risk of HIV-1 infection between 1989 and 2004 were analysed: 145 subtype B were followed three monthly over two years as part of HISIS. sequences and 64 subtype F sequences. HIV-1 env V3 sequences We obtained samples from 12 participants with chronic infec- were grouped by year of collection and the mean intra-subtype tion who had been identified by multiple probe hybrization diversity and divergence were examined at nucleotide, synony- assay (MHA) as being infected with single or multiple HIV mous, non-synonymous, and amino acid level. subtypes. The env C2C3 and vpu regions were amplified and Results: The mean intra-subtype diversity of the Brazilian intra-person diversity of total viral populations assessed over subtype B and F epidemics almost doubled in the last 15 time using heteroduplex mobility assays (HMA). PCR frag- years. The use of subtype-specific consensus sequences can ments from all time points were sequenced and samples with reduce to almost half the degree of sequence dissimilarity high diversity were cloned and each clone representing a between the vaccine strain and Brazilian circulating viruses. unique banding pattern on HMA was sequenced and phylo- Furthermore, the amino acid consensus sequence of the genetic relationships determined. Brazilian HIV-1 subtype B and F epidemics remain stationary Results: Analysis of C2C3 and vpu regions revealed five of over time, showing that both subtype epidemics are very sta- the twelve individual were dually infected. These individuals bly rooted around these “central” sequences. Despite these were infected with both subtype A and subtype C. Of the advantages, the mean intra-subtype divergence at nucleotide, seven single infections, 3 were concordant in C2C3 and vpu synonymous, non-synonymous, and amino acid level tends to for subtype C, one for subtype A and one for subtype D. Two increase over time in both Brazilian subtype populations. individuals were infected with recombinant viruses AC and Conclusion: The use of subtype-specific consensus sequences CD. Viral population evolution in individuals with dual as vaccine strains may reduce the difficulties resulting from infection showed high fluctuations in viral populations with the growing diversity of the Brazilian HIV-1 epidemic. both subtypes only being detected in between 2 and 7 time However we found no evidence supporting the view that the points in these individuals. HIV-1 intra-subtype divergence is bounded by a set dN dis- Conclusions: This study demonstrates the co-existence of tance from the subtype consensus, as has been proposed. two subtypes during chronic infection with viral popula- Rather, the non-synonymous and amino acid distances to tions exhibiting high levels of fluctuations, highlighting the subtype consensus has been continually increasing in the importance of screening for dual infection in multiple time Brazilian HIV-1 epidemic without yet reached a limit, sug- points and multiple regions of the genome. Full length gesting that the likelihood for success of vaccine strategies genome sequencing will be required to comprehensively based on “central” sequences could be also continuously characterize dual infection. decreasing over time. AIDS Vaccine 2006 115 Poster sessions_revGJ 14/8/06 16:34 Page 116 Amsterdam, Netherlands 29 August–1 September 2006 P08-03 These results highlight how molecular epidemiology can inform the design of interventions and limit the complexity of strains that must be addressed by HIV-1 vaccines. Social/demographic variables associated with recombinants and dual infections in injecting drug users in Northern Thailand P08-04 GH Kijak1, C Beyrer2, S Tovanabutra1, T Sripaipan2, Enhanced detection of HIV-1 specific T-cells in V Suriyanon3, J Jittiwutikarn4, ML Robb1, DL Birx1, early HIV-1 infection using peptides designed by DD Celentano2, and FE McCutchan1 a novel biometric approach 1 US Military HIV Research Program, Walter Reed Army Institute of Research, Rockville, MD, USA; 2 Johns Hopkins Bloomberg U Malhotra1,3, F Li2, J Nolin1, M Allison1, H Zhao4, School of Public Health, Baltimore, MD, USA; 3 Research JI Mullins3,4,5, S Self2 and MJ McElrath1,3,5 Institute for Health Sciences, Chiang Mai, Thailand; 4 Northern Drug Treatment Center, Chiang Mai, Thailand 1 Program in Infectious Diseases, Clinical Research Division, Seatlle, WA, USA; 2 Statistical Center for HIV/AIDS Research & Background: HIV-1, including strains of CRF01_AE, subtype Prevention, Fred Hutchinson Cancer Research Center, Seattle, B, and their recombinants, has been circulating among inject- WA, USA; 3 Department of Medicine, University of Washington ing-drug users (IDUs) in Thailand since early 1980s. Most School of Medicine, Seattle, WA, USA; 4 Department of recombinants have been unique, isolated from only one indi- Microbiology, University of Washington School of Medicine, vidual, but one circulating recombinant form (CRF), Seattle, WA, USA; 5 Department of Laboratory Medicine, CRF15_01B, was found in IDUs and in low-risk groups. University of Washington School of Medicine, Seattle, WA, USA Multiple HIV-1 exposures in high-risk groups are thought to foster dual infections, the apparent source of new recombi- Background: Numerous candidate T-cell based vaccines rep- nants. Emergence and spread of recombinants can complicate resentative of multiple strains are under evaluation in clinical HIV-1 vaccine development. trials. Vaccine-induced T-cells are typically identified by their Objectives/aim: To investigate the HIV-1 genetic diversity ability to produce cytokines when stimulated with peptides among IDUs in Northern Thailand, and to assess behavioral whose sequences are based on vaccine strains. Such an and social/demographic variables associated with recombi- approach is sensitive in defining the immunogenicity of the nant strains and dual infections. vaccine preparation, but does not adequately assess specifici- Methods: The Opiate-Users Research (OUR) Study was con- ties of the immune responses that potentially may be relevant ducted among 2231 volunteers in Chiang Mai, Thailand to protection against circulating strains worldwide. (1999-2000). Samples from the participants who were HIV-1 Objectives: To develop peptide sets that provide enhanced-cov- seropositive at baseline (n=347) were subtyped in 6–8 erage of potential T-cell epitopes (PTE) contained in diverse cir- genome regions with the Multi-region Hybridization Assay culating strains for the evaluation of protective immune (MHAbce v.2). The assay distinguishes subtypes B, C, and responses in vaccine recipients and for objective comparisons CRF01_AE from recombinant forms and detects dual infec- of T-cell responses to potential candidate vaccines. tions. Full-genome sequences were obtained from selected Methods: Nef-specific T-cell responses recognizing a PTE strains. Social/demographic variables and risk behavior were peptide set designed to cumulatively cover 70% of PTE assessed using a structured questionnaire administered in a among subtype B isolates were compared with responses rec- confidential face-to-face interview. ognizing the consensus (CON) peptide set by IFN-γ ELISpot Results: 96.8% (336) of the samples could be typed. 81.8% assay in 23 subjects with early HIV-1 subtype B infection. were CRF01_AE, 3.9% were subtype B, and 9.2% were Results: Based on 23 subjects, a mean of 2.7 epitope domains recombinants, mostly between CRF01_AE and subtype B. was determined for the PTE peptide set, compared to a mean Dual infections were detected in 5.1% of subjects. Subtype B of 1.7 epitope domains for the CON peptide set (P=0.0034). was more frequent among participants 30 years or older The average magnitude of response was 2,169 SFC/106 (OR=6.92, 95%CI:1.51–31.73). Dual infections were more PBMC in the PTE peptide set, compared to 1,010 SFC/106 frequent among those with a lower education level (AOR=, PBMC in the CON peptide set (P=0.0046). Overall, an addi- 6.2 95%CI: 1.66-22.82) and those who have initiated inject- tional 29 epitopic domains not detected with the CON pep- ing in the last 3 years (AOR=4.94, 95%CI: 1.30-18.69). tide set were detected with the PTE peptides. Reactive Recombinant strains and dual infection were more frequent domains of the Nef protein detected with both the CON and among those reporting frequent needle sharing in the last 3 PTE peptide sets were similar, with most responses spanning months. (AOR=4.08, 95%CI: 1.42–11.73). A novel CRF, amino acids 65–103 and 113–147. Epitope variants were fre- CRF34_01B, was recovered from 3 individuals, and it con- quently recognized even when these were not detected among stitutes the second B/CRF01_AE recombinant in Thailand. the autologous virus populations. We detected no false posi- Conclusion: Early intervention should be aimed at reduction in tive responses in 12 HIV-1 seronegative control subjects with needle sharing, especially among new IDUs, to limit the increas- either the CON or the PTE peptide sets. ing and expanding complexity of HIV-1 strains in Thailand. 116 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 117 Amsterdam, Netherlands 29 August–1 September 2006 Conclusions: The use of PTE-optimized peptide panels for P08-06 the detection of T-cell responses is a cost-effective strategy for providing augmented T-cell epitope coverage. Furthermore, Unique recombinant forms of HIV-1 subtypes A we postulate that vaccine immunogens with expanded T-cell epitope coverage will be vital to the design of effective vac- and G emerging from dual infections in Cameroon cines for protection against highly diverse viruses and rapid- ly changing virus populations. R Powell1, F Konings1, A Nanfack2, S Burda1, M Urbanski1, DR Saa3 and P Nyambi1,4 P08-05 1 New York University School of Medicine, New York, NY, USA; 2 Laboratoire de Sante Hygiene Mobile, Yaounde, Cameroon; 3 Genetic diversity of HIV-1 strains among IDUs in Ministry of Public Health, Yaounde, Cameroon; 4 VA Medical 2002–2004 in St Petersburg, Russia Center, New York, NY, USA Background: The HIV-1 epidemic in Cameroon is character- E Dukhovlinova, E Chernyaeva, A Masharsky, ized by several subtypes and a high proportion of recombi- S Verevochkin, S Gagarina, N Klimov and A Kozlov nants. While CRF02_AG is predominant, the presence of several recombinants suggests the frequent occurrence of dual The Biomedical Center, Research Institute of Pure Biochemicals, St or super infections. Petersburg, Russia Objectives: To identify individuals dually-infected with dis- tinct HIV-1 subtypes A and G and to characterize the recom- Background: The main way of HIV transmission in Russia binants emerging from such infections. and in countries of Eastern Europe is intravenous caused by Methods: Viral RNA was extracted from the plasma of two the use of mutual syringes by injecting drug users (IDU). The HIV-1 infected individuals from Cameroon that were shown predominant HIV-1 variant in Russia and in the former by direct sequencing to have discordant subtypes A and G in Soviet Union (FSU) countries belongs to subtype A. For this gag, pol, and env genes. RT-PCR of the 700bp C2V5 env reason genes of subtype A HIV-1 strain are used to develop region was performed as a preliminary determination of dual the regional HIV vaccine. infection. Samples suspect of dual infection were further Aim: To determine the sequences and genetic diversity of the analysed by amplification of gp160 (2.9kb) and/or a 1.5kb genes of HIV-1 viruses circulating in IDUs in St Petersburg, pol fragment. The amplicons were cloned, and 10–20 Russia. clones/individual were sequenced. Subtype and recombinant Methods and materials: Sera samples were collected in identities were determined by phylogenetic and breakpoint 2002–2004 from newly diagnosed HIV positive IDUs from St analysis, respectively. Petersburg. Sera were screened to reveal HIV-1 serotype using Results: Our analysis revealed clones from two individuals synthetic V3-peptides. For genetic analysis viral RNA was that contained recombinants between subtypes A and G with extracted from 18 samples. Fragments of gag, pol and env different breakpoints than CRF02_AG. In pol, all clones genes were amplified using reverse transcription (RT) and analysed from one individual were found to be a unique nested PCR. RT-PCR products were sequenced and analyzed recombinant form with a cross-over from subtype A to G, to determine HIV-1 clade and level of genetic diversity using 850bp from the start of pol. Analysis of clones from a second computer programs BioEdit and PHYLIP. individual revealed one clone as a recombinant with a Results: Results of phylogenetic analysis and serological typ- crossover from G to A, 850bp from the beginning of pol, ing showed that all HIV-1 strains studied in our research while the remaining clones from this individual clustered could be referred to subtype A or to CRF03_AB recombinant with subtype A. In env C2V5, five clones of virus from the form. The majority of samples belonged to subtype A and second individual were identified as subtype A, while three only 3 belonged to CRF03_AB. The level of intrasubtype were G, suggesting a dual infection. Further analysis of full- genetic diversity for analyzed HIV-1 strains was found to be length env found one clone with two breakpoints, one from rather low and very close to the level of genetic diversity of G to A, 1kb into the gene and the second from A to G, 800bp subtype A strains spread in FSU. downstream. The remaining clones were found to be subtype Conclusions: Among St Petersburg IDUs diagnosed to be G, further confirming that this individual was dually infected HIV positive in 2002–2004 the FSU HIV-1 variant of subtype with distinct A and G subtypes. A dominates. HIV-1 CRF03_AB also circulates in St Conclusions: This study reveals unique HIV-1 recombinants Petersburg. No other HIV-1 subtypes or recombinant forms due to infection with subtypes A and G which have different were revealed by our investigation. That confirms correctness breakpoints than CRF02_AG. CRF02_AG emerged from of our choice of the FSU variant of subtype A as a base for West Central Africa to become a highly successful viral HIV vaccine development in Russia. strain. As such, monitoring the spread of these newly-emerg- ing A-G recombinants is critical not only for understanding the epidemiology of HIV-1, but also in the design of future therapeutics and vaccines appropriate to this part of Africa, and globally. AIDS Vaccine 2006 117 Poster sessions_revGJ 14/8/06 16:34 Page 118 Amsterdam, Netherlands 29 August–1 September 2006 P08-07 P08-08 Global and regional distribution of HIV-1 genetic Genetic diversity of HLA-Class I alleles among subtypes and recombinants in 2004 HIV-infected and uninfected individuals in a Chinese Uygur ethnic group J Hemelaar1,2, E Gouws3, PD Ghys3, S Osmanov1 and the WHO-UNAIDS Network for HIV Isolation and KX Hong*, XZ Lu, JH Zhang, MM Jia, H Xing, YH Ruan, Characterization YZ Zhang, YM Shao 1 WHO-UNAIDS HIV Vaccine Initiative, World Health National Center for AIDS/STD Control and Prevention, Beijing, Organisation, Geneva , Switzerland; 2 Magdalen College, Oxford China University, Oxford, UK; 3 Epidemic and Impact Monitoring, Geneva, Switzerland Background: The genes of the human leukocyte antigens (HLA) are highly polymorphic. This polymorphism may Objective: To estimate the global and regional distribution of result in amino acid variation in the peptide-binding groove, HIV-1 subtypes and recombinants in 2004. which determines the immune response elicited. Design: A study was conducted in which molecular epidemi- Identification of HIV restriction HLA class I alleles within a ological data on HIV-1 subtype distribution in individual population is useful for the development of an effective AIDS countries were combined with country-specific estimates of vaccine and for clinical trials. The epidemiology survey the number of people living with HIV. showed that one of the main transmission routes of HIV-1 in Methods: HIV-1 subtype data were collected for 23,874 HIV- China is through intravenous drug use now and the drug 1 samples from 70 countries, which together accounted for abuse rate is high in Uygur people in Xinjiang. HIV-1 inci- 89% of all people living with HIV worldwide in 2004. The dence in IDU population in Xinjiang was high (8.8%). This proportions of HIV-1 infections due to various subtypes indicated that Xinjiang may represent a suitable site for HIV- detected in each country were combined with the number of 1 vaccine efficacy trials. HIV infected people in the respective countries to generate Objective: To explore the distribution of HLA-A, -B, -Cw regional and global HIV-1 subtype distribution estimates. alleles in a Chinese Uygur Ethnic Group in Xinjiang. Results: Subtype C accounted for 50% of all infections Methods: 119 unrelated healthy subjects and 102 HIV-1 worldwide in 2004. Subtypes A, B, D and G accounted for infected individuals from a Chinese Uygur Ethnic Group in 12%, 10%, 3% and 6%, respectively. The subtypes F, H, J Xinjiang volunteered to participate in this study. HLA-A, -B, and K together accounted for 0.94% of infections. The cir- -Cw genotypings were performed using PCR-SSP method. culating recombinant forms CRF01_AE and CRF02_AG The gene frequencies and their association with HIV-1 infec- each were responsible for 5% of cases, and CRF03_AB for tion were analysed by using PPAP and SPSS software. 0.1%. Other recombinants accounted for the remaining 8% Results: We identified 18HLA-A, 27 HLA-B and 22HLA-Cw of infections. All recombinants taken together were responsi- specificities in 119 unrelated healthy subjects from Uygur ble for 18% of infections worldwide. Chinese, with A*01, A*02, A*24, B*35, B*51, and Cw*01, Conclusion: Combining data on HIV-1 subtype distribution Cw*04, Cw*06, Cw*07 as the most common HLA-A, -B, - in individual countries with country-specific estimates of the Cw alleles, which occupied an allele frequency of 0.1134, number of people living with HIV provided a good method 0.2185, 0.1345, 0.1136, 0.1136 and 0.1261, 0.1345, to generate estimates of the global and regional HIV-1 genet- 0.1765, 0.1891 correspondingly. The results of test for ic diversity in 2004. The results could serve as an important Hardy-Weinberg equilibrium showed that the genotype dis- resource for HIV scientists, public health officials and HIV tributions in these three loci were correspondent with expect- vaccine developers. ed. Preliminary HLA-association with HIV-1 infection analysis show that B*38(P=0.04, OR=3.12>1, CI=1.01- 9.64), Cw*12(P=0.02, OR=4.203>1, CI=1.06–9.10) were significantly higher in HIV-1 infected subjects compared with healthy controls, whereas B*07(P=0.05, OR=0.35<1, CI=0.12∼0.99) was significantly higher in healthy controls. Conclusion: Multi-epitope vaccine design and clinical trial require knowledge of HLA class I distribution and HIV CTL epitope characterization in potential target population. The present study provides the data of HLA-A, -B, -Cw allele fre- quencies in a Chinese Uygur Ethnic Group in Xinjiang. HIV- 1-specific T cell responses in infected individuals are being determined. 118 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 119 Amsterdam, Netherlands 29 August–1 September 2006 P08-09 P08-10 Full-length HIV-1 subtype C sequences from indi- Molecular and phylogenetic analysis of HIV-1 iso- viduals with acute and chronic infection from lates identified in Mashhad - Iran Kwa-Zulu Natal, South Africa L Buonaguro2, RH Naderi1, M Tagliamonte2, ML FK Treurnicht1, GP Bandawe1, K Mlisana2, G Ramjee2, SS Tornesello2, M Ciccozzi3, G Rezza3, R Farid4 and FM Abdool Karim2, CM Rousseau3, JI Mullins3, C Williamson1 Buonaguro2 and the CAPRISA 002 Study Team 1 Institute of Infectious Diseases, University of Mashhad, Iran; 2 1 Institute of Infectious Diseases and Molecular Medicine, Viral Oncogenesis and Immunotherapy & AIDS Refer. Center, Ist. University of Cape Town, Cape Town, South Africa; 2 Centre for Naz. Tumori “Fond. G. Pascale”, Naples, Italy; 3 Epidemiology the AIDS Programme of Research in South Africa, University of Unit, Dept. Infectious Diseases, ISS, Rome, Italy; 4 Allergy and Kwa-Zulu Natal, Durban, South Africa; 3 Department of Immunology Dept, University of Mashhad, Iran Microbiology, University of Washington, Seattle, WA, USA Aim: Shared needles among IVDUs in correctional settings Objective: To characterize full-length genetic sequences of represent the most important risk factor for HIV-1 infection HIV-1 subtype C from individuals with acute and chronic in the HIV-1 epidemic in Iran. Information on HIV-1 subtype infection. prevalence and distribution in Iran are not available in liter- Methods: A total of 39 plasma samples were available from ature and a sentinel study on samples identified at the 32 participants with viral loads ranging from 610 to 474 000 University of Mashhad has been carried out. A phylogenetic copies/ml. Plasma samples were collected from twenty indi- analysis has been performed on env and gag subgenomic viduals within 2–11 weeks post infection (CAPRISA 002) regions, to obtain a more significant classification of the iso- and from nineteen participants who had been infected for lates and solid information on genetic distribution and origin between 5 months to 5 years (Du cohort). Long template of HIV-1 circulating in Mashhad. cDNA transcripts were generated from viral RNA prepared Results: HIV-1 gag and env regions (approx. 300 bp each) from 200 µl plasma. Whole genomes were amplified from have been PCR amplified in PBMCs from 12 seropositive cDNA using a modified limiting dilution PCR assay. Data individuals (11 IVDUs and 1 homosexual) living in Mashhad. were analysed using ChromasPro, BioEdit, ClustalW and The phylogenetic tree analyses show a single cluster within MEGA 3 software programs. Sequences from acute infections the A subtype, for both gag and env regions; a role as founder (n=19) were compared to sequences from 10 chronic infec- for the Mashhad cluster can be inferred for the A-clade tions generated from this study, as well as to 54 sequences Ugandan UG037 isolate. On the contrary, A-clade samples from the HIV Los Alamos database. from Middle-East Countries cluster together in an indepen- Results: Whole genome sequences were generated from 29/32 dent branch. Furthermore, HIV-1 env sequences previously (90.6%) participants with full genomes amplified from plas- identified in Teheran, and deposited in the Los Alamos data- ma with viral loads as low as 610 copies/ml. All viruses were base, are interdispersed in the Mashhad group, forming an HIV-1 subtype C throughout the genome with most “Iranian” cluster. sequences clustering with other South African sequences and Conclusions: The phylogenetic analysis of 12 HIV-1 Iranian one sequence clustering with Brazilian Cs. From one individ- samples, identified in Mashhad, suggests an A subtype-driven ual, a total of 32 full-length genomes were sequenced at 8.5 epidemics deriving from the African and not from the months post infection (n=10), 2 years post infection (n=12) Middle-East Countries A-subtype. The clustering pattern and and 5 years post infection (n=10). There was limited evolu- the nucleotide divergence values suggest a recent introduction tion over time in this individual with a mean pair-wise DNA of the HIV-1 infection in this community which appears to distance for gp160 over about 4 years of 2.5%. There were have common origin with the epidemic occurring in Teheran. no significant differences in length of V1/V2 and V3/V4 These first data on HIV-1 subtypes in Iran could represent the region of env gene between viral sequences from acute (n=19) starting point for a wider molecular survey to trace HIV-1 and chronic infections. epidemics in the Iranian and Middle-East region. Discussion: In this study we were able to amplify whole genomes from over 90% of individuals, even from samples with low viral loads and/or limited sample volumes. Most genome sequences clustered with other South African HIV-1 subtype C strains and one genome sequence clustered with the subtype C strains from Brazil. Analysis is ongoing to investigate if there are genetic signature sequences unique to transmitted variants. AIDS Vaccine 2006 119 Poster sessions_revGJ 14/8/06 16:34 Page 120 Amsterdam, Netherlands 29 August–1 September 2006 P08-11 P08-12 Serotyping approach as a feasible tool for assess- A comprehensive molecular epidemiological study ing HIV-1 diversity in Sao Paulo, Brazil of HIV-1 subtypes in Yunnan Province of China CAF Oliveiro, MJ Castejon, R Yamashiro, R Matsunaga, Z Chen2, Y Zhang1, L Lu1, L Ba2, L Yang1, M Jia1, Y Shi1, R Rodrigues, M Ueda and LFM Brigido W Yan1, G Chang1, L Zhang2 and DD Ho2 Instituto Adolfo Lutz, Sao Paulo, Brazil 1 Yunnan Center for Disease Control, Kunming, Yunnan, PRC; 2 Aaron Diamond AIDS Research Center, The Rockefeller Background: Different studies have suggested that HIV-1 University, New York, NY, USA immunotyping and serotyping could allow clustering other than genotypes that may provide relevant information on Background: The first HIV-1 epidemic in China was found antigens cross-reactivity. among male injection drug users (IDUs) in the Dehong pre- Objectives: To assess the reactivity of sera from HIV-1-infect- fecture of Yunnan in 1989. Since then, Yunnan province has ed pregnant women (PW) from Sao Paulo State-Brazil (SP) had the highest numbers of HIV-1 infections and AIDS-relat- against V3 immunodominant regions of circulating clades in ed deaths in China. It is also considered a site for future vac- this area; and to evaluate specificity and cross-reactivity of cine studies, but a systematic study of the molecular V3 antibodies and their frequencies in this population. epidemiology of HIV-1 in this province has not been done. + Methods: Sera from HIV PW identified at the Instituto Objectives: The goal of this study is to conduct a molecular Adolfo Lutz, SP, from 1999 to 2005 were anonymously eval- characterization of HIV-1 throughout every region of Yunnan. uated. Employing a dilutional and avidity-based IgG-EIA, Methods and Results: Using immunoassays, we cumulatively HIV-1 antibodies (V3Ab) against 15-mer biotinylated pep- identified and mapped 103,015 cases of HIV-1 infections in tides representing consensus V3 amino acid sequences (clades Yunnan by 2004. We obtained HIV-1 p17 genes from the A–G) were investigated. Available data on time from sero- serum of 321 patients using an RT-PCR technique. DNA conversion, assessed by means of STARHS, were aggregated sequences derived from the amplified products were subject- to analysis. ed to phylogenetic analyses. These patients were from Results: A total of 7109 specimens from PW were sent to IAL Yunnan’s all sixteen prefectures representing four risk for HIV diagnosis. Sixteen samples (0.2%) were confirmed to groups, eleven ethnic populations, and ten occupations. The contain specific antibodies to HIV-1, distributed as follows: majority of these infections (189/321; 58.9%) were found 7/2888 (0.2%) in 1999–2000, 7/1971 (0.4%) in 2001–2002, among the Han population. Using phylogenetic analysis we and 2/2250 (0.1%) in 2003–2005 periods. V3 serotyping have identified three major genotypes circulating in the allowed the distinction of 7 (44%) B -serotype specimens, 3 MN province including subtypes C/CRF07/08_BC (53%), (19%) B , 3 (19%) C, and 1 (6%) F-serotype specimen. One BR CRF01_AE (40.5%), and B (6.5%). For subjects with known specimen (6%) exhibited ambiguous reactivity against B and risk factors, there were two major patient populations: IDUs F peptides and 1 sample (6%) could not be typed. (51.7%) and sexually transmitted cases (45.9%). The rest High avidity A and G cross-reactive V3Ab were observed were through vertical transmission and blood transfusion among B and C reactive specimens from longstanding HIV MN (2.4%). Among IDUs 90.9% had C/CRF07/08_BC viruses infection. No cross-reaction has been observed in sera from 2 whereas 85.4% of CRF01_AE infections were acquired recent-infection specimens (both B serotype). BR through sexual transmission. Only 6.8% of HIV infections Discussion: The observed prevalence in this low risk popula- among IDUs were due to subtype CRF01_AE. tion is compatible to the estimated HIV prevalence for Brazil Geographically, C/CRF07/08_BC was found throughout the (0.45%). The identified HIV-1 serotypes are similar to HIV-1 province while CRF01_AE was largely confined to the west- genotypes described for other local populations. Recent sero- ern prefectures bordering Myanmar. conversion seemed to favor serotyping specificity, although Conclusions: This is the first report of a comprehensive, we could not rule out the possibility of STARHS misclassifi- province-wide HIV-1 molecular epidemiology study in cation due to genetic divergence of B subtype. BR Yunnan or anywhere else in China. The data are critical to Conclusion: HIV-1 V3 sero-reactivities could be evaluated as guide the design of future vaccine clinical trials. While an inexpensive tool to provide insights on broadly immuno- C/CRF07/08_BC and CRF01_AE are co-dominant in the genic epitopes from viral conserved regions. province, the discovery of many sexually transmitted CRF01_AE cases is new and suggests that this subtype may lead to a new epidemic in the Chinese population. 120 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 121 Amsterdam, Netherlands 29 August–1 September 2006 P08-13 P08-14 High throughput functional analysis of HIV-1 env HIV-1 prevalence and subtype distribution in HIV-1 genes without cloning positive volunteers deferred from enrollment in a Phase III prime-boost HIV-1 vaccine trial in Thailand JL Kirchherr1, W Kasongo2, V Chalwe2, L Mwananyanda2, RM Musonda2, BF Haynes1 and F Gao1 FE McCutchan1, GH Kijak1, S Tovanabutra1, S Rerks- Ngarm2, S Watanna2, W Wiriyakijja2, S Nitayaphan3, 1 Duke University Medical Center, Durham, NC, USA; 2 The C Eamsila3, M deSouza3, J Kim3, DL Birx1* Tropical Disease Research Centre, Ndola, Zambia 1 US Military HIV Research Program, Rockville, MD USA; 2 Background: Functional HIV-1 env clones have been widely Ministry of Public Health, Thailand; 3 Armed Forces Research used for vaccine design, neutralization assays and pathogen- Institute of Medical Sciences, Bangkok, Thailand; *current address: esis studies. Detailed mapping of the neutralization epitopes Centers for Disease Control and Prevention, Atlanta, GA, USA on the Env proteins will also have an important impact on vaccine development. However, obtaining non-recombinant Background: HIV-1 genetic diversity is considered one of the functional env clones has been a time-consuming and labor- key obstacles to an effective preventative vaccine. In Thailand, intensive process. two of the globally prevalent strains, CRF01_AE and subtype B, Objectives: To develop a high throughput method that can co-circulate, and Phase I and II vaccine trials have often includ- quickly characterize HIV-1 env gene function and deter- ed vaccines based on both of these subtypes.A Phase III trial of mine neutralization sensitivity of Env pseudotyped viruses Aventis Pasteur live recombinant ALVAC-HIV (vCP1521) prim- without cloning. ing with VaxGen gp120 B/E (AIDSVAX B/E) boosting, which Methods: The HIV-1 env genes were amplified using a single includes both subtype B and CRF01_AE immunogens, is ongo- genome amplification (SGA) method to avoid recombination ing in Rayong and Chon Buri Provinces, Thailand, with more and resampling during PCR amplification. The CMV pro- than 16,000 volunteers enrolled. During screening for enroll- moter was amplified separately. Both PCR products were ment, some volunteers were deferred due to HIV-1-infection. linked together using an overlapping PCR method, named Objectives: to describe the HIV-1 subtype distribution among pPCR. The final PCR products were purified and cotrans- volunteers deferred from enrollment, in order to provide the fected with SG3∆env backbone DNA into 293T cells to gen- best representation of the subtypes that may challenge trial par- erate pseudovirions. The env gene function was determined ticipants. by measuring luciferase activity from pseudovirion-infected Methods: Screening was conducted in Rayong and Chon Buri TZM-bl cells. Provinces, Thailand, (July, 2004–December, 2005), and more Results: Five and three SGA PCR env genes were amplified than 26,000 volunteers were evaluated. HIV-1 serostatus was from 05ZM0373 and 05ZM0374, respectively. The CMV assessed with a screening ELISA and western blot confirmation. promoter was added to all eight PCR products using the RNA extracted from seropositive plasma was used for HIV-1 pPCR method. After infecting TZM-bl cells with the genotyping with the MHAbce v.2 assay using real-time PCR and pseudovirions, four out of five 05ZM0373 env genes were subtype-specific fluorescent probes in 8 genome regions. functional and one was not, while all three 05ZM0374 env Selected samples were confirmed by sequencing of the virtually genes were functional. Two hundred micrograms of pPCR complete genome and phylogenetic analysis. products were sufficient to generate enough pseudovirions Results: The prevalence of HIV-1 seropositivity was 1.6%. for a large number of neutralization assays. Western blot Among 419 HIV-1 infected subjects, 391 plasma samples were analysis showed that the non-functional 05ZM0373 env gene available for analysis, and 376 (96.2%) could be genotyped. was due to the lack of Env protein expression. This method The subtype distribution was 89.4% CRF01_AE, 2.9% subtype avoids artificial recombination during PCR and eliminates B, 6.4% CRF01_AE/B recombinant, and 0.3% CRF01_AE/C cloning and preparation of plasmid DNA for generation of recombinant. Four individuals (1.1%) showed evidence of dual pseudovirions. Importantly, the method is both time and cost infection. Twenty-nine of the 40 non-CRF01_AE strains were effective. completely sequenced, including 10 subtype B and 19 recombi- Conclusions: We have developed a high throughput method nant strains. All subtype B and 17 of 19 recombinant strains (pPCR) to quickly identify and characterize a large number of were confirmed. HIV-1 functional env genes. The method can be used to Conclusions: CRF01_AE continues to be the predominant cir- analyse viral env gene quasispecies population from many culating strain in the eastern provinces of Thailand. HIV-1 strains. This method can serve as a powerful tool to CRF01_AE/B recombinant strains account for a larger propor- characterize functional env genes, determine neutralization tion than pure subtype B. The Phase III candidate vaccine is well profiles, and identify potential Env immunogens for vaccine suited for the circulating strains in the region and will provide development. opportunity to assess the impact of vaccination on a pure sub- type and on a variety of recombinant forms of HIV-1. AIDS Vaccine 2006 121 Poster sessions_revGJ 14/8/06 16:34 Page 122 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 9A: VACCINE CONCEPTS & DESIGN – ANTIGEN SELECTION AND DESIGN P09A-01 P09A-02 Development of HIV-1 nef DNA vaccine construct Conjugation of peptides bearing HIV drug-resis- from Indian subtype C and its immunogenecity tance mutations to the B-subunit of recombinant testing in Balb/C mice cholera toxin enhances the immune response against the mutations P Seth2 R Kohli,1 and BK Das1 A Boberg1,2, S Gaunitz3, A Bråve1,2, K Hallermalm1,2, 1 Department of Microbiology, All India Institute of Medical L Gudmundsdotter1,2, S Johansson1,2, J Hinkula1,2,4, Sciences, New Delhi, India; 2 Seth Research Foundation, New M Isaguliants1,2, N Carlin3 and B Wahren1,2 Delhi, India 1 Microbiology & Tumor Biology Center, Karolinska Institute, Background: A HIV vaccine should be formulated according Stockholm, Sweden; 2 Swedish Institute for Infectious Disease to the local circulating subtype so that the vaccine immuno- Control, Stockholm, Sweden; 3 SBL Vaccines, Solna, Sweden, 4 gen and the circulating virus share antigenic domains and University of Linköping, Linköping, Sweden induce a robust immune response in the recipients. The elici- tation of an immune response to epitopes from the early reg- Background: As more HIV-1 infected patients receive anti- ulatory nef protein from HIV-1 infected cells can lead to retroviral drug treatment the occurrence of drug-resistant termination of early virus replication. HIV variants is increasing. Therefore, research needs to Objective: 1. Construction of rDNA and rMVA vaccine can- address development of vaccines against the drug-resistant didate of codon optimized nef gene lacking myristoylation variants as well as the wild type virus. It is also believed that site and dileucine motif from HIV-1 Indian Subtype C; 2. a strong mucosal immunity is needed to lower the risk of Immunogenicity testing of rDNA/rMVA vaccine candidate in transmission of virus. We have previously shown that mutat- Balb/C mice. ed peptide sequences are equally good and often better Methods: We cloned mutated nef gene of Indian HIV-1 subtype immunogens than wild type peptides. The non-toxic B sub- C virus in the plasmid DNA expression vector, pNK14 and unit of cholera toxin (CTB) is an active substance in the oral MVA (Modified Vaccinia Ankara). Western blot and indirect cholera vaccine. It has been shown to stimulate good mucos- immunofluorescence assay were used for in vitro expression of al immunity. Therefore, by combining these findings it may the constructs. Constructs were evaluated for immunogenicity be possible to suppress upcoming mutant virus by triggering in Balb/C mice in ‘prime-boost strategy format’, where boost- an immune response to the anticipated mutants. ing with rMVA was done 2 weeks after priming with rDNA Objectives: To investigate different ways of peptide immu- (rDNA/rMVA and rDNA/rDNA/rMVA). The effect of IL2 was nization in experimental settings that may be applicable in evaluated as an immunoadjuvant (rDNA+IL2/rMVA). Vector humans. To study if peptide linkage to cholera toxin B sub- alone/MVA alone were used as control. Serum samples were unit may be such a way. collected and the mice sacrificed at regular intervals from 4 Methods: The nucleotide sequences of peptides representing weeks to 24 weeks post immunization. Humoral immune protease or RT mutants were linked to the sequence of the B- response was evaluated byELISA while cellular immune subunit of cholera toxin. These hybrid molecules expressed in response by IFN-γ ELISpot assay on spleen cells. E. coli were used to immunize mice transgenic for the human Results: Directional cloning of the nef gene was done in leukocyte antigen (HLA) A0201. Twelve days after each mammalian expression vector (NKnef vaccine construct) immunization, mice were bled and the cellular immune and MVA (MVAnef vaccine construct). X-gal staining con- response was analysed by IFN-γ ELISPOT. The response was firmed the rMVA. A band of ~27kD protein in western blot compared to that in groups of mice immunized with peptide showed the expression of both rDNA and rMVA vaccine alone or with a mixture of recombinant cholera toxin B and constructs, which is also confirmed by indirect immunoflu- the peptide. orescence assay. Immunized Balb/C mice developed robust Results: We identified strong enhancement of the immune humoral as well as cell mediated immune responses in response to the peptides fused to the B subunit of the cholera terms of magnitude, breadth and longevity. toxin. Immunization with a drug-mutated peptide also trig- rDNA/rDNA/rMVA showed higher immune response gered a cross-reactive immune recognition of the wild type (IFN-γ production) than rDNA/rMVA. The immune epitope. Studies on the long-term immunity are ongoing. response was further augmented by IL2. Conclusions: Linking peptides to the B subunit of cholera Conclusions: In vitro and in vivo studies showed that the nef toxin may be a way to stimulate stronger immune response to vaccine constructs induce a protein production and are many peptide regions than by protein/peptide immunization immunogenic. This HIV-1 nef Indian Subtype C vaccine is a alone. This may be important in the development of a vaccine good candidate to induce a robust immune response in terms targeted to drug-resistant HIV. of breadth, magnitude and longevity of immune response. 122 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 123 Amsterdam, Netherlands 29 August–1 September 2006 P09A-03 P09A-04 Proteasome targeted reverse transcriptase of HIV- Induction of potent cross clade immune response 1 generates a balanced T-cell response in mice following immunization with HIV-1 Indian after DNA-vaccination subtype C mutated and codon optimized tat DNA/MVA vaccine in mice A Boberg3, ES Starodubova1,2,3, RN Stepanenko4, AV Timofeev1, B Wahren3, MG Isaguliants3,5 and VL Karpov1 H Qureshi1, P Seth2 and BK Das1 1 Engelhardt Institute of Molecular Biology, Moscow, Russia; 1 Department of Microbiology, All India Institute of Medical 2 University of Oslo Centre for Medical Research, Moscow, Sciences, New Delhi, India; 2 Seth Research Foundation, New Russia; 3 Swedish Institute for Infectious Disease Control, Delhi, India Stockholm, Sweden; 4 Institute of Immunology, Moscow, Russia; 5 Ivanovsky Institute of Virology, Moscow, Russia Background: HIV-1 Tat an early regulatory protein is expressed soon after infection and induces Tat specific cell Background: We propose a multi-gene HIV DNA-vaccine mediated and humoral response. Presence of anti-Tat with reverse transcriptase (RT) as one of the components. immune response correlates inversely with the progression of RT is a stable protein degraded partially via the proteasome disease. Thus, this study focused on the development of tat (Starodubova et al., Vaccine 2005). By targeting RT to sole- based heterologous prime boost vaccine constructs for induc- ly proteasomal degradation, we deemed to enhance its Th1 tion of cellular and humoral response. performance as immunogen. Objective: Construction of HIV-1 Indian subtype C tat Objective: To enhance proteasomal degradation of RT by DNA/MVA vaccine and its immunogenicity testing in fusing RT to the mouse ornithine decarboxylase (ODC) a Balb/c mice. protein that is rapidly degraded by the proteasome; to Methods: HIV-1 Indian subtype C tat gene consensus examine the immunological performance of this hybrid sequence was mutated by deleting its nuclear localizing gene in DNA immunization of mice. sequence and codon optimized. This modified tat gene was Methods: ODC gene was subcloned into RT expressing vec- cloned in a mammalian expression vector pNK14 tor under the control of CMV promoter to generate the (nk14tatCo), vaccinia shuttle vector pSC65 (sc65tatCo) and hybrid gene (RT-ODC). HEK293 cells were transiently a prokaryotic expression vector pET28A (et28tatCo). For transfected with RT-ODC, and the expression was charac- construction of recombinant MVA, chicken embryo fibrob- terized in vitro. Pathway of degradation was investigated lasts cells were infected with MVA and subsequently trans- using proteasomal inhibitors MG132 and epoxomicin. fected with sc65tatCo. Virus was harvested and plaque BALB/C mice were immunized with plasmids encoding RT- assayed to select recombinant MVA (rMVAtatCo) using ODC, wtRT or empty vector. Intracellular IFN-γ synthesis BrdU and Xgal. In vitro expression of the constructs, + by CD8 T cells was assessed by FACS analysis after res- nk14tatCo and rMVAtatCo was studied by immunoblotting, timulation with RT protein or RT peptides representing immunofluorescence and confocal microscopy. Expression of aa375-389, and 528-543. Anti-RT antibodies were deter- recombinant Tat protein from et28tatCo was confirmed by mined by ELISA. Immunoblotting. Tat protein was purified using Ni2+- Results: The size of RT-ODC was as expected (120kDa). NTA affinity chromatography. Balb/c mice were immunized The fusion of RT to the ODC decreased the half-life of the intradermally with nk14tatCo (100 µg) and rMVAtatCo (107 hybrid protein, and targeted it for proteasomal degradation pfu) using various prime-boost regimens. The immune as judged by inhibition of degradation by MG132 and responses were evaluated by IFNγ ELISpot assay and ELISA epoxomicin. The RT-ODC fusion protein was found to be using subtype B and C Tat antigens. + immunogenic upon immunization of mice. CD8 T-cells Results: Both constructs, nk14tatCo and rMVAtatCo from 7/8 (87.5%) of the RT-ODC immunized mice pro- expressed Tat protein as confirmed by immunoblotting, duced IFN-γ after in vitro stimulation of the splenocytes. immunofluorescence and confocal microscopy. Purified Endpoint binding titers of anti-RT antibodies was two fold recombinant Tat protein was used as an antigen for elucidat- higher after RT-ODC immunization as compared with ing humoral response. Mice immunized with two doses of wtRT immunization (1:1600 versus 1:800 respectively). nk14tatCo construct alone developed potent clade specific Conclusions: Fusion of HIV-1 RT gene with the murine and cross clade cell mediated and humoral response. ODC gene enhanced the proteasomal degradation of the Magnitude of immune responses were increased 2–3 folds in expressed RT-ODC fusion protein in vitro. Immunization mice boosted with rMVAtatCo construct instead of + of mice with the hybrid RT-ODC gene elicited CD8 T cell nk14tatCo construct. However, mice primed with two doses response as well as a humoral response against RT protein of nk14tatCo construct and then boosted with rMVAtatCo in the majority of mice. Contradictory to our expectations, construct elicited even more robust immune responses in our results suggest targeting RT to proteasome induces a terms of magnitude and persistence. more balanced T /T response than wild type gene H1 H2 Conclusions: Generation of robust clade specific and cross immunization. clade immune responses by HIV-1 Indian subtype C tat vac- AIDS Vaccine 2006 123 Poster sessions_revGJ 14/8/06 16:34 Page 124 Amsterdam, Netherlands 29 August–1 September 2006 cine constructs suggests that it would be an important com- P09A-06 ponent in the multigene HIV vaccine development. Combined single-clade candidate HIV-1 vaccines induced T cell responses limited by multiple forms P09A-05 of in vivo immune interference Immunogenicity of HIVCON, a novel HIV-1 vac- T Hanke1, N Larke1, E-J Im1, R Wagner2, C Williamson3, cine candidate based on conserved regions of A-L Williamson3 and A McMichael1 clades A-D 1 Weatherall Institute of Molecular Medicine, University of S Letourneau, AJ McMichael and T Hanke Oxford, Oxford, United Kingdom; 2 Institute of Medical Microbiology and Hygiene, University of Regensburg, University of Oxford, Weatherall Institute of Molecular Medicine, Regensburg, Germany; 3 Institute of Infectious Disease and Oxford, UK Molecular Medicine and Division of Medical Virology, University of Cape Town, Cape Town, South Africa Background: Vaccine development efforts against HIV have been hampered because of the high mutation rate of the virus Objective: To assess vaccine induction of CD8+ T cell and the complex geographical distribution of clades. A vac- responses capable of efficient recognition of multiple HIV-1 cine capable of inducing broad T cell responses against con- clades by single-clade vaccines that have been or will be used served epitopes could overcome the dangers of both escape in clinical trials in Europe and Africa. mutations and geographical localization. We developed a Methods: Mice were immunized with single-clade A, B and C novel vaccine immunogen termed HIVCON, a fusion protein vaccines alone and in combinations, and recognition of epi- consisting of highly conserved domains of HIV proteins with tope variants by vaccine-induced T cells was assessed. no more than 6% variability between HIV-1 clades A-D. Results: Single-clade vaccines applied alone induced only lim- HIVCON was designed regardless of known epitopes. ited cross-clade reactivity and their clades affected epitope Conservation may reflect lower immunogenicity and/or func- hierarchy. However, combining single-clade HIV-1 vaccines tional conservation. into multi-clade formulations encountered multiple forms of Objectives: To show that HIVCON is immunogenic and that in vivo immune interference such as original antigenic sin and epitopes derived from these regions are presented by HIV- antagonism, which dampened or even abrogated induction of infected cells. responses to many epitope variants and reduced the breadth Methods: The vaccine gene was inserted into the pTH plas- of induced T cell responses. Administration of individual mid, recombinant Modified Vaccinia Virus Ankara (MVA) clade vaccines into anatomically separated sites on the body and Human Adenovirus Serotype 5 vectors. Immunogen alleviated antagonism and increased the number of detectable expression in human cells was assessed by histochemical epitope responses. staining. Immunogenicity studies in mice were carried out on Conclusions: Although cross-reactivity of murine CD8+ T fresh splenocytes from BALB/c or HLA-A2 transgenic mice cells does not directly translate to humans, the underlying vaccinated with a prime-boost vaccination regimen, using guiding principles of the molecular interactions involved in 51 IFNγ ELISPOT, ICS and Chromium release assays. PBMCs triggering T cell responses are the same in the mouse and from healthy volunteers or HIV-1-infected patients are cur- man. Thus, these results have important ramifications for the rently tested for responses against overlapping peptides span- design of both prophylactic and therapeutic vaccines against ning the entire HIVCON protein in a cultured IFNγ ELISPOT HIV-1 and other highly variable pathogens. assay. Results: For each vaccine vector, immunogen expression was confirmed in human cells, as well as immunogenicity in BALB/c mice. Broad and robust H-2Dd and HLA-A2 restrict- ed CD8+ T cell responses to HIVCON were observed. Responses to 5 H-2Dd restricted and 2 HLA-A2 restricted CD8+ T cell epitopes have been identified in mice. Furthermore, strong IFNγ+ T cell responses to HIVCON pep- tides have been observed so far in HIV-infected patients. Conclusions: Thus, we have successfully generated DNA, adenoviral and MVA vectors carrying the immunogen gene, and HIVCON has been able to induce broad and strong CD8+ T cell responses in mice. 124 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 125 Amsterdam, Netherlands 29 August–1 September 2006 P09A-07 1 Department of Immunology; 2 Department of Pharmacodinamics and Physiology, Oswaldo Cruz HIV-1 envelope on the surface of a lentiviral Institute/FIOCRUZ; 3 Institute of Biophisics; 4 Institute of particle elicits broader immune responses than Microbiology, Federal University of Rio de Janeiro (UFRJ), Rio de soluble gp120 or trimeric gp140 envelopes Janeiro, Brazil TM Ross, SP McBurney, KR Young, CI Nwaigwe, Background: Protozoan parasites appear as HIV-1 co- H Grieser and KA Stefano-Cole pathogens, resulting in a mutually enhancement of viral and parasite replication, and facilitating the progression of both University of Pittsburgh, School of Medicine, Department of diseases. The mechanisms by which HIV-1 induces up-modu- Medicine, Division of Infectious Diseases, Pittsburgh, PA, USA lation of protozoan replication are poorly understood. Objectives: To investigate the role of HIV-1 Tat protein in the enhancement of parasite replication in primary human Background: Virally-regulated lentiviral particles (HIV-1 and macrophages co-infected or not with HIV-1. SHIV) were expressed from DNA plasmids encoding Gag, Methods: Human macrophages were co-infected with HIV- protease, reverse transcriptase, Vpu, Tat, Rev, and Env. The 1 and either with Leishmania amazonensis or sequences for integrase, Vpr, Vif, Nef, and the long terminal Blastocrithidia culicis. In selected assays, macrophages were repeats (LTRs) were deleted. Mutations were engineered into infected only with Leishmania or Blastocrithidia and the VLP genome to produce particles deficient in activities exposed to recombinant HIV-1 Tat protein. Protozoan and associated with viral reverse transcriptase and protease. Each HIV-1 replication were analysed by endocytic index or p24 plasmid efficiently secreted particles from primate cells in ELISA assay, respectively. vitro and particles were purified from the supernatants and Results: HIV-1 infection doubled the Leishmania replication used as immunogens. in macrophages, and Tat antiserum significantly reduced the Objectives: To evaluate the immune responses elicited by exacerbated parasite replication, pointing to a direct role of trimeric native Env presented on the surface of a virus-like Tat protein released from HIV-1-infected cells in this process. particle to soluble forms of Env. Corroborating this finding, exposure of Leishmania-infected Methods: Mice (BALB/c) were vaccinated intranasally (weeks macrophages to recombinant Tat (100 ng/ml) mimicked HIV- 0, 3, 6) with purified VLPs and the elicited immunity was 1 infection. Protozoan replication diminished when compared to particles without Env (Gag ), to soluble p55 Leishmania-infected cells were treated with Tat in the pres- monomeric Env (Env ), or to trimerized Env (Env ). gp120 gp140(FT) ence of anti-TGF- 1 antibodies, showing a participation of Results: Only mice vaccinated with VLPs had robust anti-Env β TGF- 1 in the augmentation of Leishmania multiplication in cellular immunity. In contrast, all mice had high-titer anti- β macrophages. To evaluate whether HIV-1 infection deacti- Env serum antibody (IgG). However, VLP vaccinated mice vates the microbicidal activity of macrophages, HIV-1-infect- had antisera that detected a broader number of linear Env ed cells were co-infected with a non-pathogenic protozoan peptides, had anti-Env mucosal sIgA and IgG, as well as high- (Blastocrithidia), and we found that HIV-1 infection favors er titers of serum neutralizing antibodies. VLPs elicited high the survival of this trypanosomatid. Similarly, Tat or TGF- 1 titer antibodies that recognized linear regions in V4-C4-V5- β doubled the protozoan growth in Blastocrithidia-infected C5 and the ectodomain of gp41, but did not recognize the V3 macrophages. Tat induced the expression of the enzyme region. Only two linear epitopes were recognized in the COX-2 and PGE secretion, and blockage of PGE produc- V1/V2 region. Interestingly, antisera from VLP vaccinated 2 2 tion abrogated the increased Leishmania replication induced mice targeted the highly cross-reactive 2F5 and 4E10 binding by Tat. sites, as well as the Kennedy epitope in gp41. Conclusions: HIV-1 infection and Tat protein enhance the Conclusions: These lentiviral VLPs are effective mucosal multiplication of a pathogenic protozoan (Leishmania) and immunogens that elicit broader immunity against Env deter- permit survival of an otherwise non-pathogenic protozoan minants in both the systemic and mucosal immune compart- (Blastocrithidia) in macrophages. Neutralization of Tat ments than soluble forms of envelope. reduced the exacerbated Leishmania growth induced by HIV- 1 infection. Since HIV-1 Tat-based vaccines decrease virus replication, our results suggest that Tat immunoneutraliza- P09A-08 tion could be useful to controlling the growth of both, HIV- 1 and protozoan co-pathogens. Interactions HIV-1 host cells: immunomodulation Supported by: PAPES/FIOCRUZ; CNPq; FAPERJ in the HIV-1/trypanosomatids coinfection model V Barreto-de-Souza1, TX Medeiros1, GJ Pacheco4, AR Silva2, HC Caire-Faria-Neto2, PT Bozza2, MC Motta3, EM Saraiva4, DC Bou-Habib1 AIDS Vaccine 2006 125 Poster sessions_revGJ 14/8/06 16:34 Page 126 Amsterdam, Netherlands 29 August–1 September 2006 P09A-09 Methods: Although CD8+ T cells play an important role in controlling viral infections, boosting specific CD8+ T cells by Critical role of an NKP44 ligand in the CD4+ T prophylactic vaccination with SIV epitopes, hardly affects the depletion during HIV infection: a new target for a viral dynamics during acute infection. Viral replication rates, vaccine candidate and viral contraction rates after the peak viremia, hardly depend on the presence of memory CD8+ T cells. To study P Debré1, V Vieillard1, D Costagliola2 and J Strominger3 these paradoxical findings, we devise a series of mathemati- cal models parameterized for SIV/HIV infection. 1 INSERM U543, Hôpital Pitié-Salpétrière, Paris, France; Results: These models explain the failure of vaccination by 2 INSERM U720, Hôpital Pitié-Salpétrière, Paris, France; too low effector/target (E/T) ratios during the expansion phase. To explain the peak in the viral load and the subse- 3 Harvard University Cambridge, MA, USA quent immune control we require other factors limiting viral replication, e.g., limited availability of uninfected target cells, Aim: HIV infection leads to a state of chronic immune acti- around the peak viral load. The invariant down-slope of the vation and progressive deterioration in immune function, viral load after the peak can be explained with a realistic manifested most recognizably by the progressive depletion of eclipse phase of about a day. Because CD8+ T cells require CD4+ T cells. A substantial percentage of natural killer (NK) cell-to-cell contacts, immune protection requires high E/T cells from patients with HIV infection are activated and ratios at the primary site of infection. express the natural cytotoxicity receptor (NCR) NKp44. Conclusions: Unfortunately, this is difficult to achieve for Methods and Results : We show that a cellular ligand for the viruses replicating faster than the immune response, and pro- NCR NKp44 (NKp44L) is expressed during HIV-1 infection phylactic SIV/HIV vaccination is expected to remain unsuc- and is correlated with both the progression of CD4+ T cell cessful because CD8+ T cell control is only effective when depletion and the increase of viral load. CD4+ T cells express- viral growth is already controlled. ing this NKp44L ligand are highly sensitive to the NK lysis activity mediated by NKp44+ NK cells. The ligand’s expres- sion is strongly induced by the linear motif NH2-SWSNKS- COOH of the HIV-1 envelope gp41 protein, called 3S. This P09A-11 highly conserved 3S motif appears critical to the sharp increase in NK lysis of CD4+ T cells from HIV-infected A consensus clade B Env HIV virus-like particle patients. vaccine elicits broader immunity than a polymeric Furthermore, to assess the possibility that anti-3S antibodies clade B virus-like particle based vaccine might protect CD4+ T cells from NK lysis, we examined the occurrence and function of these antibodies in HIV-infected SP McBurney and TM Ross patients and show that the production of anti-3S antibodies was observed in around 30% of HIV-infected patients and University of Pittsburgh, Department of Medicine, Division of was inversely correlated with both the CD4 cells count and Infectious Diseases, School of Medicine, Pittsburgh, PA, USA NKp44L expression on CD4+ T cells, particularly, in slow progressor patients. Background: The viral diversity of HIV-1 requires that Conclusion: This study may provide one of the first demon- potential vaccines must protect against a wide range of stration that specific B and T helper memory immune HIV-1 isolates. Initial HIV-1 vaccine studies demonstrated responses can influence NK cytotoxicity directed against protection against the vaccine strain of virus, but not + autologous CD4 T cells. In addition, in linking specific adap- against heterologous virus strains. tive immunity to the innate immunity induced by the 3S Objectives: To determine if an HIV-1 virus-like particle motif, this study may have important implications for thera- containing an envelope derived from the consensus peutic vaccines against AIDS directed non against the sequence of 100 Clade B Envs (ConB VLP), elicited an pathogen, but against the pathogenesis. increased breadth and maturation of immune responses to various Envs from different clades compared to a mixture of four HIV VLPs expressing envelopes from primary iso- P09A-10 lates (Poly VLPs) or a VLP expressing Env from a single primary Env (HIV VLP). Explaining the failure of SIV/HIV vaccinations Methods: BALB/c mice were vaccinated intranasally priming CD8 T cell responses (weeks 0, 3, 6) with purified VLPs derived from primate cells. The breadth of the elicited immune responses was RJ de Boer then measured by ELISAs, neutralization, and ELISpot reactivity against a panel of HIV-1 Envs from a range of Theoretical Biology, Utrecht University, Utrecht, The Netherlands primary isolates representing clades A, B, and C. Results: Mice receiving the ConB VLP and Poly VLPs elicit- ed similar anti-Env serum antibodies (IgG) titers against Objectives: To understand why memory CD8 T cells that the panel of clade B Envs. However, the ConB VLP mice were elicited by vaccinations fail to control SIV/HIV upon elicited broader neutralizing antibodies than mice vaccinat- challenge. 126 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 127 Amsterdam, Netherlands 29 August–1 September 2006 ed with the Poly VLPs. Antibodies from ConB VLP vacci- like titers to the HIV-1 epitope than could be induced by nated mice recognized Envs from various clades, while the individual mimotopes. Poly VLPs mice were only able to recognize Env from with- Conclusion: Pre-HAART serum IgG of two out of four dis- in clade B. Similar cellular responses were also induced ease progressors selected a collection of 21 structural mimet- against clade B Env peptides using both vaccines. Again ics of the gp41 immunodominant CSGKLIC loop sequence ConB VLP vaccinated mice had stronger cellular responses with conserved CKC and variable SGLI residues. The 21 pep- against consensus group M Env and consensus clade C Env tides cloned and described here is a new generation structur- peptides. al and antigenic substitutes of the HIV-1 gp41 ID epitope, Conclusions: Both ConB VLP and Poly VLPs vaccines rec- useful for diagnosis, structural studies of the epitope, and ognize a broader range of clade B Envs than a single prima- affinity purification from patient sera of naturally elicited ry isolate Env based VLP. However, ConB VLP elicits antibody to ID epitope. immune responses that bind and neutralize viruses from var- ious clades. Therefore, the ConB VLP vaccine elicits broad- er immunity than either a Poly VLPs or an HIV VLP vaccine. P09A-13 Characterization of a Tat mutant, antigen candi- P09A-12 date for HIV-1 therapeutic vaccination Large-scale and highly reproducible selection of K Mayol, S Munier, C Terrat, B Verrier and C Guillon peptides to the gp41 immunodominant epitope of HIV-1 by phage library screening with patient IFR 128 Biosciences Lyon-Gerland, U503 INSERM, Lyon, France serum Background: Besides its major role of transactivator of viral Y Palacios-Rodríguez1, T Gazarian2, M Rowley3, A Majluf- transcription, HIV-1 Tat protein modifies the expression of Cruz4 and K Gazarian1 numerous cellular genes involved in immune response. As therapeutic vaccine against HIV is destinated to infected 1 Department of Molecular Biology and Biotechnology of immunodepressed patients, the use of Tat antigen in a thera- Institute of Biomedical Research, Mexican National Autonomous peutic vaccine cocktail would therefore be safer if the protein is defective for its impact on these cellular genes. University, Mexico City, Mexico; 2 Department of Public Health Objectives: To define a mutated form of Tat, defective for the of Faculty of Medicine, Mexican National Autonomous deregulation fonctions of the wild type protein, but still able University, Mexico City, Mexico; 3 Department of Biochemistry to induce anti-Tat immune responses in vivo. and Molecular Biology, Monash University, Clayton, Australia; 4 Methods: For this work, we used 101 residues Tat sequences General Hospital No. 1 Gabriel Mancera, IMSS from primary isolates. Jurkat cells were stably transfected with wild-type Tat (“WT”) or mutant tat Background: The HIV-1 Env subunits or their synthetic pep- C27S/K51T/R55L/G79A (“STLA”) using a retroviral vector. tide versions are used for diagnosis and vaccination purpos- The expression of MHC I, Il-2 and CD25 genes was assessed es. Epitope mimics selected in combinatorial phage-peptide in this context. In vivo, WT and STLA Tat immunosuppres- libraries by infected host antibodies may better reflect prop- sive fonctions were compared in a mouse model, after simul- erties of native viral epitopes. taneous injection of WT or mutated Tat protein with HIV Objective: Characterization of the ability of serum of AIDS p24 antigen. Finally, mapping of peptides recognized by anti- progressors to retrieve from combinatorial phage libraries Tat antibodies induced after WT or STLA Tat protein immu- peptides mimicking the HIV-1 gp41 immunodominant (ID) nization was carried out in a rabbit model. epitope functionally active, prior to and after the initiation of Results : In Jurkat cells, the capacity to inhibit MHC I expres- highly active antiretroviral therapy (HAART). sion observed for WT Tat was lost for STLA mutant. Method: We used phage display technology involving Expression of Il-2 was also repressed in cells transduced with biopanning of large random libraries with IgG of HIV-1- WT tat, but stimulation of Il-2 expression after activation infected patients, before and after the initiation of HAART. was increased in STLA tat expressing cells compared to non- We defined conditions permitting selection of large family of transduced cells. Expression level of CD25 followed the same perfect mimics of ID epitope. trend as Il-2. In mice, immunosuppression of p24 specific Results: After three selection rounds with HIV-1+ IgG, a CD4+ response was observed after injection of p24+WT Tat, library of 109 different sequences were converted into a but not for mice immunised with p24+STLA Tat. Anti-Tat homogeneous population in which 79% of peptides con- antibody titers after immunizations with WT or STLA Tat tained CxxKxxC-motif (“x” designates a non-epitope were equivalent, and these antibodies were able to recognize amino acid). The collection was shown to display principal both autologous and heterologous Tat from clade B. structural (sequence and conformational), antigenic and Interestingly, injection of STLA Tat induced a broader spec- immunogenic features of the HIV-1 immunodominant loop trum of Tat specific humoral response than immunization epitope. Notably, when the whole family was injected into with WT Tat, which suggests that new epitopes are exposed mice, it induced two-to-three-fold higher patient antibody- on STLA Tat antigen. AIDS Vaccine 2006 127 Poster sessions_revGJ 14/8/06 16:34 Page 128 Amsterdam, Netherlands 29 August–1 September 2006 Conclusions: STLA Tat is defective for immunosuppressive P09A-15 effects observed with WT Tat in vitro and in vivo. Moreover, it induces anti-Tat antibodies at the same titers as WT Tat, Sequential cross-clade vaccination of HIV vaccines that recognize heterologous Tat. Besides, it induces a wider predominantly stimulated T-cell immunity against spectrum of antibodies than WT Tat. Thereby, STLA Tat is a HIV conservative epitopes promising antigen as part of an anti-HIV therapeutic vaccine. J Xu, L Ren, C Qiu, X Huang, Y Liu, Y Liu and Y Shao P09A-14 National Center for AIDS/STD Control and Prevention, China CDC, China Biophysical and antigenic properties of a soluble, cleaved and stabilized HIV-1 subtype A envelope Background: The effort to develop an effective preventive trimer vaccine against human immunodeficiency virus 1 (HIV-1) infection has encountered the unprecedented challenge con- SPN Iyer1, M Franti1, DC Pevear1, S Beddows2, AK Dey2, stituted by the large genetic diversity of HIV-1 among differ- DN Fisch1, J Jones1, L Campbell-Gardener2, J Overbaugh3, ent isolates. A novel strategy able to enhance HIV-specific T VanCott4, PJ Maddon1, WC Olson1, and JP Moore2 immune responses targeting at conservative epitopes will be crucial for this vaccine development effort. 1 Progenics Pharmaceuticals Inc., Tarrytown, NY, USA; 2 Dept. of Objectives: To explore a new vaccination strategy “sequential cross-clade vaccination” in inducing cross-reactive T-cell Microbiology / Immunology Weill Medical College, Cornell immunity targeting at conservative epitopes in mice model University, New York, NY, USA; 3 Division of Human Biology, Fred with vaccines derived from three different clades: C, B’ Hutchinson Cancer Research Center, Seattle, Washington, USA; 4 (Thailand B) and E which account for 45%, 30% and 17% Henry M. Jackson Foundation, Rockville, Maryland, USA HIV-1 prevalence in China, respectively. Methods: Three DNA vaccines and two recombinant replica- Background: The largest number of incident and total HIV-1 tive vaccinia vaccines were constructed by inserting the syn- infections is found in sub-Saharan Africa. Kenya is a typical thetic fusion genes into plasmid vector pDRVISV1.0 and East African example, where the majority of circulating strains TianTan vaccinia, respectively. Each vaccine contains a fusion are homogenously subtype A. Therefore, there is a need to gene of gag and env derived from either C (CN54), B’ (RL42) develop an HIV-1 vaccine capable of eliciting broadly reactive or E clade (AE2F). A new strategy named as sequential cross- immune responses against members of this subtype. clade vaccination strategy was tested in mice model, which Objectives: Develop new HIV trimer as potential HIV vaccine employs vaccines derived from E, B’ and C clades for sequen- candidates. tial inoculation, respectively. ELISPOT and ICS were used to Methods: Envelopes (Env) from four East African HIV-1 iso- read out the T-cell immunity and peptides spanning Gag and lates were evaluated for their ability to form stable trimers env proteins of consensus C and B were used as stimuli. when expressed as soluble, proteolytically cleaved proteins Results: Experiments in both Balb/c and C57BL/6 showed containing a gp120-gp41 intersubunit disulfide bond and an that sequential cross-clade vaccination stimulates predomi- I559P mutation in gp41 (SOSIP gp140). Env proteins were nant T-cell immunity against conservative epitopes among transiently expressed in 293T cells and analysed for oligomer- three immunogens, including epitope SPRTLNAW in p24, ic state by Blue-Native PAGE and size-exclusion chromatog- HNAWATHACVP and NNLLRAIEAQ in gp160, whereas raphy. Antigenicity was examined by ELISA and vaccination with a single vaccine derived from B’ clade or immunoprecipitation using neutralizing and non-neutralizing with the mixture of multiple vaccines preferentially raised T- monoclonal antibodies (MAbs), and the neutralization sensi- cell immune responses directing at non-conservative or less tivity of the corresponding Env-pseudotyped virus was conservative epitopes, including AMQMLKETINE and assessed using MAbs, CCR5 inhibitors and T-20. EAMSQVTNSATIMMQ in p24, NYQHLWRWGTM and Results: SOSIP gp140 derived from the subtype A isolate SLLNATAIAVA in gp160. KNH1144 formed homogeneous, stable and fully cleaved Conclusions: We concluded that only cross-clade HIV-specif- trimers that were purified to homogeneity. KNH1144 ic immune responses against conserved epitopes will be SOSIP gp140 displayed a pattern of MAb reactivity consis- sequentially enhanced by sequential cross-clade vaccination, tent with that of native trimer. KNH1144 Env-complement- thereby dominating the total HIV-specific immune responses, ed pseudoviruses were moderately susceptible to whereas strategies including conventional vaccinations with a neutralization by MAbs. Finally, KNH1144 SOSIP trimers single vaccine or the mixture of multiple vaccines failed to do were stable in the presence of non-ionic detergents (such as so. This new strategy may have important implications for Tween-20, NP-40 and Triton X-100), but the ionic deter- the development of vaccines against pathogens containing gent, SDS, was able to dissociate the trimers into dimers and genetic diversities. monomers, at room temperature. Conclusions: KNH1144 SOSIP gp140 forms stable trimers with favorable antigenicity. KNH1144 SOSIP gp140 merits further evaluation as an HIV-1 vaccine candidate, and immunogenicity studies have been initiated in rabbits. 128 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 129 Amsterdam, Netherlands 29 August–1 September 2006 P09A-16 P09A-17 Characterization of HIV-1 Gag specific T-cell Structure of the SIV envelope spike in situ immune responses and correlation with plasma viremia in infected Indian individuals S Sourial1,G Zanetti1, JAG Briggs2, K Grünewald3, Q Sattentau4 and SD Fuller1 S Kaushik1, M Vajpayee1, N Wig2 and V Sreenivas3 1 Division of Structural Biology, Wellcome Trust Centre for 1 Department of Microbiology, 2 Department of Medicine, 3 Human Genetics, University of Oxford, Headington, Oxford, UK; Department of Biostatistics All India Institute of Medical 2 Department of Chemistry and Biochemistry, Ludwig- Sciences, New Delhi, India Maximillians-Universität, München, Germany; 3 Department of Molecular Structural Biology, Max-Planck-Institut für Biochemie, Background: India is at the epicenter of the global HIV/AIDS Martinsried, Germany; 4 Sir William Dunn School of Pathology, epidemic in Southeast Asia, predominated by subtype C University of Oxford, South Parks Road, Oxford, UK infections. Hence, it is imperative to understand the immune correlates of protection in this population, in particular the Background: X-ray crystallographic studies have provided optimal and dominant HIV-1 specific responses to control insight in the structures and the differential conformations of HIV-1 infection. monomeric subunits of HIV and SIV Env. However, little is Objectives: We aimed to characterize and define the HIV-1 known about the overall structure of the Env spike and the subtype C specific CD4+ and CD8+ T-cell responses to the distribution of these spikes on virions. preferentially targeted Gag protein and correlate with plasma Aim: To produce a three-dimensional reconstruction of the viremia in 45 HIV-1 infected ART-naive Indian individuals. SIV glycoprotein spike on the viral membrane. Methods: The HIV-specific T-cell responses in 45 HIV-1 Methods: Cryo-electrotomographic data of SIV virions was infected individuals were screened by IFN-γ ELISpot assay collected and the Env spikes were extracted and averaged to using Indian HIV-1 subtype C Gag peptides (15 mer overlap- obtain a structure of the complex in its native CD4 unbound ping 11 aa) and each positive peptide confirmed individually conformation. by Intracellular cytokine assay for IFN-γ and IL-2 production Results: The distribution pattern of the spikes appears to be on a FACS Calibur (BD Biosciences) using CellQuest and random, with some areas of local ordering. The final recon- Flowjo softwares. Plasma HIV-RNA levels were determined struction had a resolution of approximately 28Å, which was by Amplicor HIV Monitor Assay version 1.5 (Roche appropriate for the docking of published crystal structures. The Diagnostics Systems). The relationship between the HIV-1- spike exhibits a three-fold symmetry and projects ∼120Å from specific responses and plasma viremia was determined by the viral membrane and is composed of a ∼35Å wide and 50Å Spearman rank correlation. high stem capped by three globular densities which fold over the Results: Broadly targeted CD8+ T cell responses of higher stem in a right hand propeller orientation. The distance between magnitude reflecting the response to increased antigenic load the tips of the globular densities is 110Å. during chronic infection were detectable in all patients as Conclusions: Much emphasis has recently been put on develop- compared to the CD4 responses. The magnitude of the HIV- ing trimeric Env based antigens as improved mimics of the viral specific IFN-γ responses was observed to be higher than the spike. Fitting the monomeric gp120 into the complex reveals corresponding IL-2 response, Plasma viremia levels correlat- the mechanism of epitope masking and receptor mediated con- ed positively with HIV-specific CD8+ T cell responses and formational change in the assembled viral spike complex. negatively with HIV-specific CD4+ T cell responses. The Gag- p24 subunit dominated the relative magnitude and breadth of T-cell immune responses and was also the most frequently targeted protein subunit. Several peptide sequences were identified in both CD8+ and CD4+ T-cell responses. Conclusions: The results reflect the inability of HIV-1-specif- ic CD8+ T-cell responses to control viral replication during chronic infection due to the probable dysfunction of CD4T cells that lack IL-2 production required for maintenance of cell-mediated responses. The identification of HIV-1 Gag-spe- cific responses at single peptide level across HIV-1 subtype C infected Indian population and correlation with the data from Caucasian population may provide useful insight for the design of new immunotherapies and vaccines for effective control of HIV-1 infection. AIDS Vaccine 2006 129 Poster sessions_revGJ 14/8/06 16:34 Page 130 Amsterdam, Netherlands 29 August–1 September 2006 P09A-18 P09A-19 Identification of mimotopes of Env immunogenic Chimaeric HIV-1 subtype C Pr55 Gag virus-like epitopes of HIV-1 from donors with HIV-1 broad particles with large in-frame C-terminal cross-neutralizing antibodies polypeptide fusions as subunit analogues of a DNA vaccine T Dieltjens, B Willems, S Coppens, G Vanham and W Janssens EP Rybicki1,2 , RJ Halsey1,2, FL Tanzer1,2, A-L Williamson1,3 and A Meyers1,2 Institute of Tropical Medicine, Antwerp, Belgium 1 Institute of Infectious Disease and Molecular Medicine, Background: A minority of individuals has plasma antibodies University of Cape Town, Cape Town, South Africa; 2 Dept capable of neutralizing primary HIV variants of multiple sub- Molecular and Cell Biology, University of Cape Town, Cape Town, types. Neutralizing antibodies are often directed to confor- South Africa; 3 National Health Laboratory Service, Groote mational epitopes that are expressed on the free virion or Schuur Hospital, Cape Town, South Africa during the process of binding with cell receptors and entry. Mimotopes to these neutralizing epitopes can be identified Background: Virus-like particles (VLPs) made from myristy- using peptide phage display and HIV-1 specific antibodies, and lated HIV-1 Pr55 Gag in a variety of expression systems have may represent vaccine peptides. been shown to have significant potential as subunit vaccines Objectives: To identify HIV-1 positive individuals with broad against HIV infection. Gag protein also has potential as an cross-neutralizing (BCN) plasma; to select peptide phage from antigen carrier, in chimaeric VLPs, for the presentation of M13 peptide phage libraries that share reactivity with antibod- other HIV proteins. ies from BCN plasma. Objectives: We wished to determine how long a sequence Methods: Plasma samples from HIV-1 seropositive individuals could be fused to Gag while still allowing VLP formation, were screened for their capacity to neutralize the infection of and to explore the possibility of making HIV-1 subtype C PBMC by primary isolates of subtypes A, B, C, D, and Pr55Gag-based chimaeric VLPs as a protein boost to the CRF01_AE. Neutralizing plasma were screened with 7 mer, C- HIV-1C multigene DNA vaccine pTHgrttnC, which encodes 7-C, and 12 mer random M13 peptide phage display libraries. a modified Gag-RT-Tat-Nef fusion protein (grttnC; Burgers Selected peptide phage were analysed by sequencing and for WA, et al., J Gen Virol 2006; 87:399–410). binding with neutralizing and non-neutralizing plasma antibod- Methods: A range of in-frame fusions with the C-termini of ies. HIV-specificity of antibodies reacting with the selected pep- myristylation-competent p6-truncated Gag (Pr50Gag) and tide phage was analysed by line immunoassay (INNO-LIA, native Pr55Gag were made to test how the length of polypep- Innogenetics). tide and its sequence might affect VLP formation and struc- Results: Plasma were identified that potently cross-neutralized ture. These included an artificial HIV-1 subtype A 75% or more of representatives of multiple subtypes. polyepitope string (HIVA; 155 aa); the truncated protein Biopanning using peptide phage display libraries allowed iden- sequences 3′RT (113 aa) and 3′TN (169 aa), and the entire tification of peptide phage that were more reactive to antibod- RT (450 aa) and TN (322 aa) and RTTN portions (778 aa) ies from BCN plasma as compared to antibodies of of grttnC. Baculovirus-expressed chimaeric proteins were non-neutralizing plasma. Reactivity to BCN plasma was not examined by western blot and electron microscopy. uniform, but restricted to a subset of BCN plasma analysed. Results: All chimaeras except for Gag-HIVA formed regular Antibodies reacting with these selected peptide phage were eval- VLPs; VLP diameter increased with protein MW, from ∼100 uated for binding in LIA. Antibodies reacted with gp120, gp41, nm for native Pr55Gag to ∼250 nm for Gag-RTTN. HIVA or not at all. Sequencing indicated that all peptide phage bind- polyepitope fusions produced large irregular budded struc- ing LIA gp120 positive antibodies presented a V3 tip of the loop tures, up to 450 nm across. Thus, the sequence of the mimotope. Peptide phage binding LIA gp41 positive antibodies polypeptide fusion was more important for VLP formation had either a gp41 principal immunodominant domain motif or than sequence length, possibly due to protein folding con- a motif that has no obvious linear epitope equivalent in the straints. The presence or absence of the Gag p6 region did not gp41 linear sequence. The latter as well as peptide phage whose obviously affect VLP formation or appearance. reactive antibodies scored negative in LIA may express mimo- Conclusions: Our results significantly expand the known topes of conformational Env epitopes. range of sizes of tolerated fusions involving whole Gag, Conclusions: Biopanning of peptide phage display libraries with which had hitherto been presumed to be limited to additions BCN plasma may guide us to vaccine peptides that mimick Env of ∼200 amino acids. This has important implications for conformational epitopes that are associated with induction of mixed prime-boost strategies involving DNA and subunit BCN antibodies. vaccines: the ability to make particles with all or part of the protein sequence specified by a DNA or virally vectored vac- cine will allow prime-boost vaccination regimens which should significantly enhance especially cellular responses. 130 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 131 Amsterdam, Netherlands 29 August–1 September 2006 P09A-20 P09A-21 Multivalent immune cell surface carbohydrate Mimotopes from combinatorial phage libraries as moieties inhibit HIV-1 primary isolate infection of anti-HIV vaccine candidates peripheral blood mononuclear cells but enhance infection of TZM-bl reporter cells KG Gazarian A Rosa Borges1, L Wieczorek1, BK Brown1, RD Kensinger2, Mexican National University, Mexico City, Mexico A Puri3, AJ Benesi4, FC Krebs5, B Wigdahl5, R Blumenthal3, DL Birx6, CL Schengrund7 and V Polonis8 Background: many candidate HIV-specific immunogens have been tested so far some of which looked promising but even- 1 Henry M. Jackson Foundation, Rockville, MD USA; 2 Sanofi tually turned out impotent as vaccines meeting Phase III trial Pasteur, Swiftwater, PA USA; 3 NCI-Frederick, Frederick, MD USA; requirements. Their principal debility is inability to induce 4 Pennsylvania State University, State College, PA USA; 5 Drexel robust and sustained responses against rapidly modifying University College of Medicine, Philadelphia, PA USA; 6 US viral determinants. Centers for Disease Control, Atlanta, GA USA; 7 Pennsylvania Objectives: To evaluate the vaccine perspectives of mimo- topes selectable from combinatorial phage libraries by patient State University College of Medicine, Hershey, PA USA, 8 Walter serum antibodies generated against contemporary viruses. Reed Army Institute of Research, Washington, DC USA Methods: IgG purified from a panel of sera of HIV-1-infected patients with disease progression were used to screen several Background: Glycosphingolipids (GSLs) are integral com- peptide libraries; sequence and HIV-1-specificity were deter- ponents of efficient infection by human immunodeficiency mined to evaluate the antigenic relatedness to particular epi- virus type 1 (HIV-1). HIV-1-relevant GSLs, Gb and GM 3 3 topes, and immunogenic potential was tested via are major constituents on the surface of immune cells. Each immunization of laboratory animals. has a distinct carbohydrate head group shown to interact Results: Due to the variability of titers of antibodies to HIV- with gp120; GM on T-lymphocytes co-localizes with CD4 3 1 epitopes, about 70–75% of the IgG, and IgG of two HIV- and CXCR4. 1(–) subjects, failed to present in samples from post-selection Objectives: To evaluate the effect of multivalent molecules population of peptides recognizable as mimotopes of viral functionalized with the carbohydrate head groups of the GSLs known epitopes. Along with this, 25–30% of IgG yielded col- Gb and GM on HIV-1 infection in different cell models. 3 3 lections of mimotopes with reference to a group of HIV-1 Methods: Multivalent compounds composed of the corre- subtype B epitopes: gp41 CSGKLIC, gp120 V3 GPGR, Tat sponding head groups of Gb or GM , (globotriose and 3′-sia- 3 3 cysteine-rich, and smaller group of mimics of gp41 ELD- lyllactose, respectively) were synthesized by covalent KWA. The large-scale biopanning experiments revealed fac- attachment to a dendrimer scaffold. The resulting compounds tors in serum and in libraries, targeting the selection towards were tested against 14 primary HIV-1 isolates (subtypes A–D particular epitope of interest. Mimotopes exhibited antigenic and CRF01_AE) in peripheral blood mononuclear cells specificity of natural epitopes, and induced in rabbits and (PBMC), as measured by reduction of p24 production. The mice responses towards the same epitopes. In a jurkat cell-cell effects of compounds on HIV-1 pseudovirus and primary iso- fusion system, some mimotopes, e.g. GPGR-mimics, tended late infection of the TZM-bl luciferase reporter cell-line were to inhibit the fusion. also assessed. Conclusions: (1) Despite heterogeneity and individual vari- Results: Both multivalent globotriose and 3′-sialyllactose ability of serum Ig in HIV-1-infected patients, the methodol- compounds inhibited viral infection in PBMC with IC val- 50 ogy permits, based on criteria we used to discard inefficient ues between 1 and 20 ug/ml. Interestingly, in the HeLa cell- experiments, enhancing particular epitope-targeted selection derived TZM-bl model, both compounds enhanced infection of mimotope collections by polyclonal patient serum; (2) of pseudoviruses and primary viruses between 2- to 8-fold. Each mimotope collection probed by patient antibodies for Results in both PBMC and the TZM-bl cell-line were inde- their antigenic fitness under stringent selection condition rep- pendent of viral clade or tropism. resents spectrum of cross-reacting antigens specific to the nat- Conclusions: Our results highlight the important roles that ural epitope and able to induce next generation these immune-cell surface GSLs (and in particular their carbo- epitope-specific antibodies, some of with a moderate inhibi- hydrate head groups) play in primary HIV-1 infection of tion of membrane fusion; (3) Further steps are underway to PBMC, and suggest that the cell system used will significantly improve selection and make patient sera a dominant source impact the data obtained in viral infection assays. While cell of HIV-1-epitope-relevant immunogens expressing the specif- line-based approaches represent a standardized screening tool, ic features of viruses, now existing and evolving in their nat- the physiologic relevance of these assays must be considered ural milieu, infected human population. when assessing HIV-1 vaccines or antiviral drug candidates. AIDS Vaccine 2006 131 Poster sessions_revGJ 14/8/06 16:34 Page 132 Amsterdam, Netherlands 29 August–1 September 2006 P09A-22 P09A-23 Broadly neutralizing antibodies against receptor- Structure-based stabilization of the 2F5 antibody activated epitopes on the N-heptad repeat region epitope of the HIV-1 gp41 envelope glycoprotein of HIV-1 gp41 are elicited during natural infection for immunogen design and by immunization FJ Guenaga1, G Ofek1, B Schiff 2, D Baker2, P Kwong1 and MB Zwick1, JD Nelson1, R Jensen1, CA Bewley6, R Wyatt1 FM Brunel2, JM Louis5, GM Clore5, RG Mage7, PE Dawson2,4 and DR Burton1,3 1 Vaccine Research Center, NIAID/NIH, Bethesda, MD, USA; 2 University of Washington, Seattle, WA, USA 1 Departments of Immunology, 2 Chemistry and Cell Biology, 3 Molecular Biology, 4 Skaggs Institute for Chemical Biology, 5 The Background: An essential component of an effective HIV-1 Scripps Research Institute, La Jolla, Laboratories of Chemical vaccine is the elicitation of neutralizing antibodies. One of Physics 6 and Bioorganic Chemistry, 7 NIDDK, and Laboratory of the most broadly neutralizing antibodies is the 2F5 antibody Immunology, NIAID, National Institutes of Health, Bethesda USA that binds a contiguous epitope on the membrane proximal region (MPR) of the gp41 envelope glycoprotein. Various Background: HIV-1 gp41 is an important vaccine target attempts to elicit 2F5-like antibodies by different immunogen because it bears sites to which broadly neutralizing antibod- design strategies have resulted in little to no neutralizing ies and fusion inhibitors bind. However, eliciting neutralizing activity. The elucidation of the crystal structure of the 2F5 antibodies against gp41 has been problematic. In the core antibody in complex with its cognate epitope provides new structures of gp41, the N-heptad repeat (NHR) region forms information to guide immunogen design. the inner helical trimer, which is a model for the receptor-acti- Aim: To elicit HIV-1 MPR-directed neutralizing antibodies. vated (fusogenic) conformation of gp41 and a target for Methods: In collaboration with investigators at the University fusion inhibitors. Anti-NHR antibodies capable of inhibiting of Washington, we have identified and designed by computa- fusion have recently been described that were selected from tional analysis a number of non-HIV protein scaffolds that can non-immune phage display libraries using gp41-mimetics. accommodate the extended conformation of the 2F5 epitope Results: Here, we have rescued HIV-1-neutralizing anti-NHR defined by the crystal structure. 2F5 epitope scaffolding is a antibodies from immune phage display libraries that were novel approach that accounts for two aspects that might be prepared from an HIV-1 infected individual, and from b9 essential to elicitation of 2F5-like neutralizing antibodies: sta- rabbits immunized with a previously described mimetic of the bilizing the conformation of the epitope as it is bound to the NHR inner trimeric coiled-coil. The human and rabbit mon- 2F5 antibody and selective presentation of the face recognized oclonal antibodies bind to overlapping epitopes on the NHR by the antibody while occluding the unbound face. trimer, generate distinct binding profiles against a panel of Results: Here we report that selected scaffolds bind the 2F5 gp41 mimetic proteins and neutralize primary HIV-1 from antibody by ELISA and Biacore and that one protein chimera various clades with modest potency using a pseudotyped (2F5e-1KU2) elicited antibodies against the 2F5 epitope in an virus assay. An examination of the differing abilities of the initial rabbit immunogenicity experiment. This scaffold also antibodies to neutralize different viruses bearing identical showed a higher affinity for the antibody 2F5 when com- NHR sequences has revealed a particular restriction to neu- pared to the free peptide by ELISA. tralization that appears to be steric in nature, rather than Conclusions: The generation of selected 2F5 structure-based dependent on viral entry kinetics. A mechanistic analysis of scaffolds is a novel approach to elicit MPR-directed HIV neu- antibody binding to native NHR epitopes is now in progress, tralizing antibodies. including the comparison of the neutralization potency of monovalent antibody fragments versus whole IgGs. Conclusions: Taken together, the results indicate that broad- P09A-24 ly neutralizing antibodies, albeit currently of limited potency, are elicited against the NHR trimer of HIV-1 gp41 during Effectiveness of gp120 hyperglycosylation on natural infection and by immunization. For vaccine design, improving the elicitation of broadly neutralizing the newly selected antibody panel can be used in optimizing the NHR trimeric coiled-coil to favorably display neutraliz- antibodies to HIV-1 ing epitopes. R Pantophlet, RO Aguilar-Sino and DR Burton The Scripps Research Institute, Department of Immunology, La Jolla, California, USA Background: We have engineered a series of gp120 mutants (termed ∆N-mCHO, ∆N2-mCHO, ∆N-mCHO-GDMR, and ∆N2-mCHO-GDMR) containing a truncated gp120 N ter- 132 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 133 Amsterdam, Netherlands 29 August–1 September 2006 minus, up to 10 additional N-glycosylation sites and, in some P09A-25 cases, serial alanine mutations (473GDMR476◊AAAA). These modifications limit or abolish exposure of non-neutralizing Evaluation of a synthetic tetramannose (Man ) 4 epitopes on gp120, while preserving the epitopes recognized glycoconjugate as an immunogen designed to elic- by broadly neutralizing antibodies (e.g. b12). it anti-mannose antibodies to an oligomannose Objectives: Here we have characterized the antibody respons- es in rabbits to these hyperglycosylated antigens and evaluat- cluster on HIV-1 gp120 ed their potential to elicit more broadly neutralizing RD Astronomo1, HK Lee1, CN Scanlan1,2, R Pantophlet1, antibodies relative to wild-type gp120. 1 1 2 1 Methods: Rabbits (groups of 4) were injected 4× at 4-week CY Huang , IA Wilson , RA Dwek , CH Wong and DR 1 intervals with antigen (0.2 mg) in Ribi adjuvant. Antibody Burton titers were determined by ELISA and serum neutralizing activity was evaluated in a pseudovirus-based assay (1:3 1 The Scripps Research Institute, La Jolla, California, USA; 2 serum dilution). Serum antibody responses were mapped University of Oxford, Oxford, UK using a panel of anti-gp120 mAbs and a set of overlapping HIV-1 consensus subtype B envelope peptides. Background: The extensive glycosylation of HIV-1 gp120 Results: Most animals elicited reasonable ELISA antibody contributes to viral immune evasion; however, a human HIV- titers to wild-type gp120 (endpoint titers 12,800–25,600). 1 broadly-neutralizing antibody (2G12) has been described ELISA titers to the homologous modified antigen were gen- that recognizes a cluster of oligomannose glycans on gp120. erally similar as for wild-type gp120, indicating the absence This discovery raises the possibility that a carbohydrate of major serum antibodies to neo-epitopes in these animals. immunogen may be developed to elicit 2G12-like neutralizing Sera from wild-type gp120-immunized animals were only antibodies (Abs) and contribute to an AIDS vaccine. To aid in capable of neutralizing (IC50) the neutralization-sensitive designing such an immunogen, we further dissected 2G12’s viruses SF162 and HXB2. Select sera from animals immu- specificity, with a panel of synthetic oligomannose deriva- nized with the modified antigens neutralized SF162 and tives, revealing its ability to bind the D1 and D3 arms of Man GlcNAc . HXB2 also, but further neutralized the moderately-resistant 9 2 viruses JR-FL (homologous virus) and JR-CSF; neutralization Objectives: To design a D1 arm glycoconjugate that binds of these viruses was noticeably more frequent with sera from 2G12 and evaluate its ability to elicit anti-oligomannose Abs ∆N-mCHO- and ∆N-mCHO-GDMR-immunized animals. that bind gp120 Methods: Synthetic Man , analogous to the D1 arm, was Unexpectedly, the specificities in the sera from animals immu- 4 nized with these two mutants did not map to the CD4-bind- conjugated to free lysines on bovine serum albumin (BSA) for ing site. Peptide mapping suggests that sera from animals multivalent display. BSA constructs with increasing copy number of Man were assayed for their ability to inhibit immunized with modified antigens, but not sera from wild- 4 type gp120-immunized animals, contain antibodies to an epi- 2G12 binding to gp120. Rabbits were immunized with BSA- (Man ) , the construct with the highest valency, to evaluate tope at the base of V2 as well as a conserved portion of the 4 14 gp120 inner domain adjacent to the silent face, which may its potential to elicit 2G12-like Abs. The specificity of serum Abs elicited in BSA-(Man ) - immunized rabbits was exam- account for their increased breadth of neutralization. 4 14 Conclusions: Our results indicate that select hyperglycosylat- ined by direct binding and in-solution competition ELISAs ed mutants, notably ∆N-mCHO and ∆N-mCHO-GDMR, with synthetic oligomannosides. can indeed elicit somewhat better neutralizing antibodies Results: Increasing valency was shown to enhance the appar- ent affinity of 2G12 for Man (up to 1000 fold over the free than unmodified gp120. Further analyses are underway to 4 evaluate the breadth of these neutralizing antibodies and to glycan), achieved at ∼10 copies, beyond which no further enhancement was observed. Although neither BSA-(Man ) dissect their specificities. 4 14 sera nor control sera from BSA-immunized rabbits react with gp120, BSA-(Man ) sera does display strong binding to Man 4 14 4 conjugated to ovalbumin, indicating the presence of Man -spe- 4 cific serum Abs. Linear (Man and Man ) oligomannosides are 3 4 preferentially recognized by the elicited Abs relative to D-Man and branched oligomannosides, such as on gp120. Conclusions: This study validates the ability of multivalently displayed oligomannosides to elicit anti-glycan Abs with defined specificities. The preferential recognition of linear oligomannosides by the elicited Abs suggests that their anti- gen-binding sites may be groove-like structures unsuited for accommodating branched glycans. However, to elicit Abs that recognize the oligomannose clusters on gp120, a better mimic of the 2G12 epitope is likely required, which might be achieved by the presentation of branched and/or more dense- ly clustered glycans. AIDS Vaccine 2006 133 Poster sessions_revGJ 14/8/06 16:34 Page 134 Amsterdam, Netherlands 29 August–1 September 2006 P09A-26 P09A-27 The broadly neutralizing antibodies 2F5 and 4E10: 2006 progress update on the GAIA HIV Vaccine: specific anti-HIV antibodies or autoantibodies? broad coverage by HLA alleles and geographical location, optimization of HIV immunogens in pre EM Scherer, MB Zwick, L Teyton and DR Burton clinical studies The Scripps Research Institute, La Jolla, CA, USA AS De Groot1,2,3, M Goldberg2, D Rivera2, JA McMurry2, L Moise1, M Koné3,4, OA Koita4, L Marcon1, M Kutzler5, M Background: A recent study by Haynes et al. (Science 2005 Lally1, KH Mayer1, DB Weiner5 and WD Martin2 308:1906–8) suggests that the broadly neutralizing anti-gp41 antibodies 4E10 and 2F5 react with autoantigens, explicitly 1 Brown University, Providence, RI, USA; 2 EpiVax, Providence, cardiolipin (CL), calling into question the specificity of these RI, USA; 3 GAIA Vaccine Foundation, Providence, RI, USA; 4 antibodies, and accordingly, their safety and utility as tools University of Bamako, Bamako, Mali; 5 University of for vaccine design. Pennsylvania, Philadelphia, PA, USA Objective: To investigate claims that 4E10 and 2F5 are reac- tive with CL using a number of methodologies. Methods: An anticardiolipin ELISA routinely used to detect Objective: To develop an epitope-driven, DNA-prime, anticardiolipin antibodies (aCL) in patients with systemic pseudoprotein-boost HIV vaccine (GAIA vaccine) com- autoimmune disorders was employed to evaluate the relative posed of both CTL and T helper cell epitopes that are high- binding affinities of 4E10 and 2F5 to endogenous lipids and ly conserved and immunogenic over a broad range of HLA exogenous antigens. Cardiolipin reactivity was assessed in the backgrounds. Methods: All HIV sequences posted to GenBank presence and absence of β-2 glycoprotein-I (β2 gp-I), to eval- (1995–2005) were scanned for protein fragments containing uate whether 2F5 and 4E10 require β2 gp-I to bind CL, as characteristic of aCL associated with autoimmune disease. >60% of the nominal full-length sequence. The 10,199 found 4E10 or 2F5 were also tested for their ability to prolong were searched for conserved 9–10mer segments using phospholipid-dependent coagulation in Lupus anticoagulant Conservatrix. 5,494 of the most highly (>5%) conserved 9- assays, as aCL exhibiting this characteristic are strongly cor- mer sequences were analysed by EpiMatrix for affinity to six related with autoimmune disease. To characterize 4E10 and HLA class I alleles and used to create promiscuous class II 2F5 binding to lamellar CL, phosphatidylserine (PS), or phos- epitopes. Candidate HIV vaccine epitopes were ranked by phatidylcholine (PC), surface plasmon resonance (SPR) conservation, EpiMatrix Z-Score with respect to each HLA experiments were conducted on liposomes. allele and proportionally by protein. ELISpot assays were Results: 4E10 and 2F5 cross-react with CL, PS, and ovalbu- performed using blood from HIV-infected subjects in the USA min in ELISA. CL reactivity is typically an order of magni- and Mali. Experimentally verified class I and class II epitopes tude less than observed for gp41 or ovalbumin, but varies were cloned as a multi-epitope string into commercial DNA with assay conditions. Regardless, binding of 2F5 to CL is vaccine vectors and evaluated for immunogenicity in HLA only observed at the highest antibody concentration tested transgenic mice. Results: Individual epitopes selected for study were generally (150 µg/ml). CL binding does not decrease in the absence of more broadly conserved than epitopes selected for other epi- β2 gp-I, suggesting that 2F5 and 4E10 binding is not β2 gp-I dependent. In SPR experiments, 2F5 exhibits negligible bind- tope-based vaccines (>70%, compared to EpImmune’s 40%). ing to bilayers containing CL, PS or PC. 4E10 binds to bilay- A total of 147 epitopes (114 class I and 33 class II) were ers containing CL and PS, but dissociates rapidly. Neither shown to be immunogenic in assays using PBMC from HIV- 2F5 nor 4E10 prolonged coagulation in Lupus anticoagulant infected individuals. Selected sets of peptides provide assays. extremely broad coverage over time, HIV strains, and geo- Conclusions: We conclude that 2F5 does not bind CL, where- graphical regions as a result of including six HLA supertypes as 4E10 does bind CL. However, because 4E10 does not pro- and promiscuous class II epitopes. For example, we identified and validated 42 class I B7 supertype restricted T cell epi- long Lupus anticoagulant assays and does not exhibit β2 gp-I dependency, we conclude that 4E10 is not autoreactive but, topes by ELISpot and all HLA matched blood samples we rather, exhibits some cross-reactivity at high antibody con- have tested have reacted to at least one of these peptides. centrations. These 42 peptides cover 9,048, or 74% of GenBank HIV pro- tein isolates. On average approximately 2.8 peptides can be matched to each of the covered isolates. Ten prototype DNA vaccine prototypes have been synthesized and six DNA immunization experiments in HLA transgenic mice have established proof of principle. Conclusion: We anticipate that the GAIA vaccine, rich in CTL and Th epitopes that possess cross-reactivity, optimized for maximal epitope expression will be equally or more immunogenic in human volunteers as other DNA/viral vector prime-boost vaccines. 134 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 135 Amsterdam, Netherlands 29 August–1 September 2006 P09A-28 P09A-29 Elicitation of robust T cell responses in HLA Comparison of neutralization antibody HIV vac- DRB1*0101 transgenic mice by MHC class II cine strategies based on trimeric gp140 Envelope compartment-targeted and untargeted HIV T of clades A, B and C and peptides to conserved helper multiepitope molecular constructs neutralization epitopes: efficacy against R5 SHIV challenge LJ Moise1, L Marcon1,2, D Rivera3, M Lally4, M Kutzler5, JA McMurry3, WD Martin 3, DB Weiner5 and AS De Groot1,3 D Davis1, Y Sun2, E Kan2, Y Lian2, W Bogers1, W Koornstra1, H Niphuis1, V Sharma2, H Oostermeijer1, 1 TB/HIV Laboratory, Brown University, Providence, RI, USA; G Koopman1, Z Fagrouch1, E Verschoor1, I Srivastava2, 2 University of Padua Medical School, Padua, Italy; 3 EpiVax, Inc., S Barnett2 and J Heeney1 Providence, RI, USA; 4 Miriam Hospital, Providence, RI, USA; 5 University of Pennsylvania, Philadelphia, PA, USA 1 Biomedical Primate Research Center, Rijswijk, The Netherlands; 2 Chiron Corporation, Emeryville, California, USA Background: CD4 T cell responses to HIV-1 are linked to pro- tection from progression to AIDS and possibly to prevention of Background: Effective strategies to induce broad neutralising HIV infection. Stimulation of CD4 T helper cell responses can antibodies to HIV-1 are needed for the development of pro- restore CTL responses lost during chronic HIV infection. phylactic HIV vaccines aimed at preventing infection. Objective: To design an epitope-based vaccine containing Objectives: Utilizing peptides to linear or conformationally helper T cell epitopes that will restore CD4 T cell repertoire important neutralising epitopes we compared immunization and function. with trimeric Envelope gp140 of clades A, B and C in MF59 Methods: Highly conserved helper T cell HIV epitopes were + CpG alone or in combination with of peptides representing computationally identified for promiscuous binding to 8 of the specific neutralising epitopes to determine which protocol most common HLA class II alleles using the EpiMatrix algo- would give protection from a mucosal R5 SHIV challenge. rithm. EpiAssembler was used to construct “immunogenic con- Methods: Rhesus macaques were immunized three times sensus sequence” (ICS) T helper epitopes, containing extended with either a multivalent Env proteins alone, an Env pro- natural epitope flanking regions conserved across clades. To tein prime, peptide boost or a peptide prime, Env proteins experimentally confirm computational predictions, a competi- boost schedule. The immunogens were: oligomeric gp140 tion binding assay was used to measure epitope peptide affinity from three human immunodeficiency virus, HIV-1 1461 to soluble HLA DRB1*0101and ELISpot assays to measure (subtype A), HIV-1 SF162 (subtype B) and HIV-1 TV1 IFN-γ release arising from epitope peptide stimulation of periph- (subtype C); synthetic peptides representing linear HIV-1 eral blood mononuclear cells obtained from HIV-infected indi- SF162 V2 neutralizing epitope or the membrane-proximal viduals. To design a multi-epitope DNA vaccine construct, epitope recognised by the human monoclonal antibody VaccineCAD organized a linear string of 25 experimentally con- 2F5; a full length cyclised V3 subtype B or C peptide; firmed ICS epitopes to contain minimal junctional immuno- mimotopes to the conformational epitope recognised by genicity, and the optimized epitope string was cloned into IgG1 b12. Control macaques were immunized with adju- eukaryotic expression vectors for expression either as an inde- vant alone. Immune responses were monitored by interfer- pendent protein or as a LAMP-1 chimera. HLA DRB1*0101 on-γ, interleukin-2 and interleukin-4 ELISpot and transgenic mice were immunized according to a DNA intracytoplasmic staining, ELISA and pseudovirus neutral- prime/peptide boost protocol and IFN-γ secretion from murine ization assays. Macaques were challenged intra-rectally splenocytes was measured by ELISpot. with 1,800 tissue-culture infectious doses of the R5 SHIV Results: 52% of ICS epitopes bound soluble DRB1*0101 in SF162P4 (subtype B). The success of immunization was competition binding assays. Positive T cell responses in HIV- monitored by quantifying plasma viral loads. 1 infected individuals were observed for 95% of ICS epitopes Results: This adjuvant formulation successfully promoted a tested in ELISpot assays. Immunization of DRB1*0101 trans- B-cell predominant IL-4, ELISpot response to the subtype A genic mice with multi-epitope vaccine constructs elicited and B recombinant glycoproteins with all schedules. The Env robust epitope-specific T cell responses to 7 out of 24 epitope protein only schedule also induced ELISpots against subtype peptides tested, which were confirmed as robust responses to C. Neutralising antibodies were induced against the challenge each of several peptide pools. MIIC-targeted LAMP-ICS vac- virus by all three immunization regimens. Following chal- cine elicited slightly stronger and broader T cell responses lenge, statistically significant reductions in viral load were than untargeted vaccine. seen for each immunization schedule. Viral loads at the peak Conclusions: Computational tools can be used to efficiently of the acute phase were reduced by 32–280 fold. Areas under map epitopes and incorporate validated epitopes into vaccine the viral load – time plot were reduced 100 fold. constructs. Vaccine constructs encoding T helper HIV epi- Conclusions: A cocktail of multivalent HIV-1 recombinant topes elicit robust epitope-specific T-cell responses in envelope glycoproteins from three different subtypes can DRB1*0101 transgenic mice. Optimal immunogen engineer- induce protection in rhesus macaques against mucosal chal- ing and targeting can improve multi-epitope-based immu- lenge with a subtype B R5 SHIV. Immunization with peptides nization approaches to raise strong T cell responses for can substitute for the recombinant glycoproteins, either at the therapeutic and preventive immunization. priming or boosting stages. Supported by NIH 1 P01 AI48225-01A2 AIDS Vaccine 2006 135 Poster sessions_revGJ 14/8/06 16:34 Page 136 Amsterdam, Netherlands 29 August–1 September 2006 P09A-30 DNA vaccination or gene therapy and vaccination approach- es are hampered by silencing gene expression, often correlat- Impact of a synthetic transdominant negative HIV- ing with epigenetic control mechanisms such as methylation 1 Gag and the Gly–Ala repeat of EBNA-1 on the of CpG dinucleotides. Objective: This study aimed to determine the contribution of inhibition of HIV-1 replication and the survival of intragenic CpG content to protein and mRNA levels in tran- transduced cells for gene therapy approaches siently transfected cells and recombinant stable cell lines. Methods: Two humanized GFP gene variants were generated 1,2 1 2 1 D Hammer , J Wild , F Notka and R Wagner containing 60 versus 0 CpG dinucleotides. Reporter activity was quantified by FACS analyses in transiently transfected 1 Institute of Medical Microbiology and Hygiene, University of H1299 and stable recombinant Flp-In CHO and Flp-In 293T Regensburg, Germany; 2 GENEART GmbH, Regensburg, Germany cell lines. Protein expression in stable cells was regularly analysed by fluorescence quantification for more than one Background: Transdominant negative HIV-1 Gag mutants year. Protein and mRNA levels were analysed by standard (TDgag) have been shown to inhibit HIV-1 replication effec- western blot and quantitative RT-PCR analyses, respectively. tively by interfering with the assembly. The main problem of Moreover, for both GFP variants the amount of newly syn- gene therapy approaches using TDgag derivatives is the host’s thesized mRNA (nuclear run-on assay) and mRNA stability immune response to the transgenes. (ActinomycinD experiments) were determined. Objective: (i) to determine the capacity of a 24 aa Gly–Ala Results: A positive correlation of transgenic CpG content stretch derived from EBNA-1 to overcome proteasomal with protein expression and mRNA level could be demon- degradation of GA-TDgag fusion proteins and (ii) to prevent strated. The individual expression levels were constant for a recognition of transduced cells by CD8+ T cells. period of 56 weeks. The mfi (mean fluorescence intensity) Methods: PM1 cells were transduced using retroviral vectors was reduced 6- to 9-fold (CHO cells) and 10- to 20-fold and inhibition of HIV replication was determined. (293T cells) for the CpG-depleted GFP gene, respectively. Differences in mRNA levels were analysed by Light Cycler This reduction of reporter activity and expression could be and Northern blot analysis. Expression studies were per- correlated to reduced mRNA copy numbers, which we could formed to test the influence of proteasome- and translation- clearly refer to decreased amounts of newly synthesized inhibitors on protein stability. The immunological impact of mRNA. Regarding mRNA stability, splicing products or the GA stretch was tested by determining the CTL responses nuclear export rate the CpG modified constructs did not dif- after plasmid immunization and additionally by in vitro and fer significantly. in vivo cytotoxicity tests. Conclusions: Altogether this observation is contradictory to Results: GA-TDgag was shown to strongly interfere with the general understanding that removal of intragenic CpGs is virus replication, the antiviral activity was increased com- beneficial for long-term and high level protein expression. pared to TDGag. Although N-terminal fusion of the GA- Moreover, this phenomenon has been shown for a number of Linker reduced specific mRNA levels, intracellular protein unrelated genes, for example cytokines, chemokines and HIV levels were increased, which could be attributed to a dimin- derived immunogens (Gag, GagPolNef, others) supporting ished rate of proteasomal degradation. The intramuscular the validity of this observation. Therefore we propose a gen- immunization of BALB/c mice with plasmid DNA coding for eral transcription-based mechanism of activating gene Gag and TDGag with and w/o GA revealed that recognition expression via CpG dinucleotides. Thus, optimizing trans- of cells expressing GA-TDgag by CD8+T cells in vitro and in genic CpG content serves as appropriate strategy to improve vivo is dramatically reduced. DNA-based therapeutic applications by enhancing transgene Conclusion: GA fusion does positively influence the trans- expression. dominant negative properties of TDgag and counteracts immune recognition by CD8+ T cells, a prerequisite for potential in vivo gene therapy application of TD-negative P09A-32 polypeptides. Influence of transgenic CpG dinucleotides on P09A-31 transgene expression Intragenic CpG dinucleotides: a novel tool for A Bauer1, F Notka2, M Graf2, J Wild1 and R Wagner1,2 improving plasmid based DNA based therapeutics and vaccines 1 Institut für Medizinische Mikrobiologie und Hygiene, Universität Regensburg, Regensburg, Germany; 2 Geneart AG, D Leikam1,2, F Notka2 and R Wagner1,2 Regensburg, Germany 1 University of Regensburg, Regensburg, Germany; 2 GENEART Introduction: Chemokines/cytokines provide highly support- AG, Regensburg, Germany ive tools to rationally modulate and target vaccine induced immune responses. The presented study aims towards Background: Different plasmid DNA applications used for increasing the levels of cytokine expression by applying vari- 136 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 137 Amsterdam, Netherlands 29 August–1 September 2006 ous strategies of RNA- and codon-optimization. bait we used recombinant Sam68 purified from E. coli. The Furthermore, the molecular mechanisms leading to enhanced phage library consisted of all 296 human SH3-domains as chemokine/cytokine expression were elaborated. fusions with the M13-surface-protein pVIII. Methods: Cytokine genes were modified by applying selected Results: In several rounds of bio-panning we selected the multi-parameter RNA- and codon-optimization strategies highest-affinity SH3-binders of Sam68 from the SH3-pro- without changing the encoded primary amino acid sequences. teome-pool. These were in part members of the Src-family, Following transient or stable transfection of various cell lines, consistent with earlier observations by several groups. Two of the impact on transcriptional activation, RNA-stability and the top-three-binders were identified as Sam68-binders for export and translational efficacy was assessed by means of the first time. The specificity of the interactions was con- northern blot and nuclear run on analysis, western blotting firmed by pull-down assays with recombinant proteins and and ELISA. EMSAs with a known, sequence specific DNA the binding-affinities were determined by a phage-ELISA. binding protein were carried out. Conclusions: Using phage-display we could identify new Results and Discussion: The quantitative comparison of high-affinity SH3-binders of Sam68 as well as confirm chemokine expression and secretion after transient transfec- known ones, showing the general usability of the library as a tion clearly demonstrated a correlation of protein expression, tool for identifying SH3-interactions. Further detailed char- the algorithm used for gene optimization and the content of acterization of these interactions will allow us to assess their intragenic CpG dinucleotides. This bias was even more pro- importance for the role of Sam68 in the export of late viral nounced in stable cell lines and seemed to correlate with the transcripts, being the basis for the design of novel therapeu- content of CpG dinucleotides within the coding sequence. tic vaccines directed against cellular structures. Our results suggest that an improved recruitment of cellular factors binding to CpG dinucleotides contributes to enhanced chemokine transcription and subsequent cytokine expression. P09A-34 Conclusion: The possibility to achieve a more stable and enhaced chemokine/cytokine expression upon delivery of mol- ecular adjuvants via non-viral or viral vectors may impact on Improvement of innate properties of DNA vaccines the development of vaccine formulations by directly influenc- and bridging innate and adaptive immune ing the magnitude and quality of immune responses in vivo. responses by sequence modifications of the plas- mid backbone 1 1 1 1,2 P09A-33 D Kosovac , V Lütschg , J Wild and R Wagner 1 University of Regensburg, Regensburg, Germany; 2 Geneart Identification of HIV-related cellular proteins as AG, Regensburg, Germany novel targets for therapeutic vaccines Objective: The main limitation of plasmid-based (pDNA) B Asbach, C Ludwig and R Wagner genetic vaccines is low efficiency in non-human and human University of Regensburg, Regensburg, Germany primates requiring high amounts of pDNA to properly prime cellular and humoral immune responses. Herein, we tested the influence of vector backbone sequence modifications, mainly Background: The long-term efficacy of vaccine-candidates as CpG-content, on the innate immune system ex vivo in a murine well as therapeutics targeted against HI-viral proteins suffers spleenocyte model and on human dendritic cells (DCs). from the virus’ high variability. During its life-cycle HIV Methods and results: Various pcDNA5/FRT (Ref; 100% hijacks various cellular systems and interferes with the equi- CpG]) derived vector backgrounds (p∆S [50.6% CpG], librium of the cell’s signalling status in favour of viral fitness, p∆S110- [47.5% CpG]) were synthesizes from scratch and for example by downregulating CD4 expression. analysed phenotypically and functionally on murine spleeno- Consequently, there are many interactions between viral and cytes and human DCs. In contrast to Ref and p∆S110-, p∆S- cellular proteins against which therapeutic vaccines can be DNA induces the secretion of high amounts of IFNγ and IL-6 directed, that are focused on the cellular side. after in vitro stimulation of naive mouse spleenocytes at dif- Objectives: Our goal is to identify new interactions in the ferent concentrations of plasmid DNA as demonstrated by cell’s complex signalling network, which are crucial for suc- ELISPOT analyses and quantification of secreted cytokines cessful HIV replication, focussing especially on interactions indicating a strong Th1- but not Th2-polarization by p∆S- of SH3 domain-containing proteins. The src-associated pro- DNA. Elimination of five CpGs within the 110-region over- tein in mitosis of 68 kDa (Sam68), which contains numerous lapping the prokaryotic origin of replication within p∆S or potential target sites for binding of SH3-domains, has been translocation of the 110 bp fragment within the plasmid shown to be a necessary host-factor for HIV-replication, resulted in a significantly reduced stimulation of proinflam- playing an important role in the nuclear export of late viral matory cytokines suggesting a strong influence of individual transcripts. Our aim is to identify new SH3 interactors of CpGs on induced immune responses in a position dependant Sam68 and to determine their importance in modulating the manner. Whereas p∆S-DNA stimulates splenocytes from functions of Sam68 in HIV-replication. wildtype mice to produce high amounts of IFNγ and TNFα, Methods: We applied a phage display technique based on this effect was totally aborted in TLR9/- mice. Additionally, bacteriophage M13 to identify SH3-binders of Sam68. As AIDS Vaccine 2006 137 Poster sessions_revGJ 14/8/06 16:34 Page 138 Amsterdam, Netherlands 29 August–1 September 2006 in vitro stimulations of human plasmacytoid dendritic cells P09A-36 with p∆S-DNA resulted in high amounts of type I interferon (IFNα). Confirming the mouse data, deletion of the 110 bp Directed nucleic acid packaging of HIV-1 subtype fragment comprising 5 CpG residues (p∆S110-) rendered this C Pr55 Gag virus-like particles potentially used as plasmid immunosilent with respect to the induction of IFN . α vaccine candidate. Conclusion: The synthesis of CpG-reduced plasmid vectors and modifications of CpG amounts provide a basis for fur- Z Valley-Omar1,2, A Meyers1,2, FL Tanzer1,2, E Shepard1,3, ther rational development of DNA vaccines and pDNA based A-L Williamson1,3 and EP Rybicki1,2 vectors for gene therapy approaches. 1 Institute of Infectious Disease and Molecular Medicine, University of Cape Town, Cape Town, South Africa; 2 Dept P09A-35 Molecular and Cell Biology, University of Cape Town, Cape Town, South Africa; 3 National Health Laboratory Service, Groote Effects of C3d fusion proteins on the expression Schuur Hospital, Cape Town, South Africa and immunogenicity of HIV-1 gp120 DNA vaccines Background: Virus-like particles (VLPs) made from myristy- V Wanninger1, K Bieler1, F Notka2, M Graf2, J Wild1, lated HIV-1 Pr55 Gag in a variety of expression systems S Jeffs3, and R Wagner1 have been shown to have significant potential as subunit vaccines against HIV infection. HIV-1 Pr55 Gag particles 1 Institute of Medical Microbiology and Hygiene, University of selectively encapsidates psi site containing viral genomic Regensburg, Regensburg, Germany; 2 GENEART AG, Regensburg, RNA with a high specificity during assembly. This incorpo- Germany; 3 Imperial College, London, UK ration of RNA has been shown to be an important prereq- uisite for VLP assembly where RNA may play the role of a structural element (scaffolding) during assembly. Other than Objective: To determine immunomodulatory properties of C- the viral genomic RNA, HIV particles also encapsidate var- terminal C3d fusions on gp120 specific B cell responses ious cellular RNAs, which could also compensate for the Methods: DNA vaccines encoding a secreted monomeric absence of psi site containing RNA (particularly in VLP form (gp120) of a codon optimized C-clade Env, C-terminal- expression systems). The encapsidation of the foreign nucle- ly fused to three tandem copies of the human C3d sequence ic acids does therefore present the potential for the delivery were constructed and biochemically analysed. Signal and expression of foreign nucleic acids in the vaccinated sequences encoding the tissue plasminogen activator were individual. It would preferable if the RNA constituency of fused to the 5′-end of the gp120/C3d gene derivatives. The VLPs administered as vaccines could be modulated to some most promising constructs were tested in two different animal degree and gauged for the potential to transfer this RNA to models. FACS studies were made to analyse the binding of the inoculated individual. C3d to the receptor CD21 on B cells. Objectives: We wished to investigate the encapsidation of for- Results and Discussion: All fusion proteins were efficiently eign nucleic acids (particularly RNA) into our insect tissue produced and secreted from mammalian cells. Variations in culture produced HIV-1 subtype C Pr55Gag-based VLPs, and the signal peptide did not influence the yields of gp120, to determine the effect on encapsidation by targeting a whereas C3d fusion slightly impaired expression rates most known psi site containing RNA sequence for packaging and probably due to cytotoxic effects on cells. The most efficient detect expression of the packaged RNA in dendritic cell cul- gp120 secreting DNA vaccine constructs were tested in ture in VLP uptake assays. BALB/c-mice and in rabbits for the induction of reactive anti- Methods: An attempt to encapsidate a psi site tagged CAT bodies and T-cell responses. Short and long term immuniza- open reading frame was done by co-expressing the CAT open tion protocols revealed no benefit of the C-terminal C3d reading frame tagged with either an upstream; downstream fusions regarding antibody titers and induction of Env-specif- or no psi site along with a full length myristylated Pr55Gag ic T-cell responses. Different in vitro analyses are currently open reading frame. Detection of encapsidated CAT-specific employed to prove the functional integrity of C3d following RNA was done by RT-PCR. fusion to gp120. Results: Pr55Gag VLPs were successfully produced in the dual Conclusion: C-terminal fusion of C3d to gp120 reduces expression system used. The presence of the psi site at either expression and secretion of the fusion proteins in vitro. DNA upstream or downstream of the CAT open reading frame had vaccines relying on RNA and codon optimized gp120/C3d no effect on CAT expression and enhanced the probability of fusion constructs turned out to be less effective in inducing CAT RNA encapsidation from the pool of cellular RNAs. humoral and cellular immune responses compared to the Conclusions: As a known RNA species has been targeted for parental gp120 DNA vaccine. encapsidation by the Pr55Gag VLPs in a VLP production sys- tem, the protein products from this RNA species (CAT) can be tested for to determine the potential for RNA transfer from VLPs and expression within the vaccinated individual. It will also be possible to test nucleic acid inactivating agents for efficacy on these VLPs. 138 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 139 Amsterdam, Netherlands 29 August–1 September 2006 P09A-37 P09A-38 Removal of glycans in the inner domain of HIV-1 A novel HIV/AIDS vaccine based on the combina- gp120 affects binding and neutralization by mon- tion of HIV Tat and ∆V2-Env proteins: preclinical oclonal antibodies and plasma studies 1 2 1 S Nkosi *, R Wyatt and L Morris F Titti1 F Ferrantelli1, E Olivieri1, S Farcomeni1, P Leone1, I Macchia1, T Maggiorella, A Borsetti1, A Cafaro1, 1 National Institute for Communicable Diseases, Johannesburg, I Srivastava2, P Monini1, S Barnett2 and B Ensoli1 South Africa and 2 Vaccine Research Center, NIAID, NIH, USA * S Nkosi has sadly deceased 1 National AIDS Center, Istituto Superiore di Sanità, Rome, Italy; 2 Chiron Corporation, Emeryville, CA, USA Background: The glycans in the inner domain of gp120 on HIV-1 across the subtypes and SIV are conserved and ‘sit’ at Background: Recently, novel modified HIV Env proteins have the interface where gp120 subunit interacts with gp41 sub- been developed that better expose conserved epitopes to elic- unit. We hypothesized that these glycans might be ‘masking’ it broad immunity overcoming the intra- and inter-clade Env the conserved functional structures in gp120 and gp41 and variability-related issues for efficacious vaccination. On the removing them might expose these structures. other hand, in vivo studies proved that vaccines based on Objectives: To determine if the removal of glycans in the HIV regulatory proteins (Tat, Rev, Nef) are safe and immuno- inner domain of HIV-1 subtype C gp140GCN4 soluble genic in mice, monkeys, and humans, and able to contain trimers affect their binding to monoclonal antibodies virus replication and to prevent disease onset in monkeys. In (mAbs) and HIV-1 positive plasma and to test neutralization addition, immune responses against regulatory viral antigens sensitivity of pseudoviruses bearing these mutations to correlate with non progression to AIDS in humans. Of note, mAbs and plasma. regulatory viral proteins are highly conserved among differ- Methods: The glycans at positions 89, 236 and 247 in the ent HIV clades and have immunomodulatory functions that inner domain were removed in gp140GCN4 trimers and in can be exploited for vaccine design. gp150 in order to make pseudovirions using site directed Objectives: For all of these reasons, our purpose was to verify mutagenesis. The presence of the desired mutations was con- the hypothesis that a vaccine combining both regulatory and firmed by sequencing and the mutants were termed 89∆CHO, novel structural HIV antigens (combined vaccine) is superior to 236∆CHO and 247∆CHO. The binding mAbs and plasma to the single-antigen type approach, because of the induction of the mutant and wild type envelope proteins were compared immune responses to both early and late viral products. using an ELISA. Neutralization sensitivity of pseudoviruses to Methods: To this purpose, we evaluated in monkeys a vaccine mAbs and plasma was tested using a single entry pseudovirus based on the rational combination of the native form of the assay. HIV regulatory Tat protein and of a trimeric Env variant that Results: The soluble gp140GCN4 236∆CHO had higher contains a deletion in the V2 loop of the gp120 subunit ( V2 binding to IgG1b12 and b6 mAb compared to the wild type. ∆ Env), which allows a better exposure of neutralization-sensi- 89∆CHO had a slightly less binding to IgG1b12 and b6 than tive epitopes of Env. wild type whereas 247∆CHO had lower binding to IgG1b12 Results: The Tat/∆V2 Env combined vaccine was safe and and b6. The 236∆CHO pseudoviruses were more sensitive to immunogenic. Animals immunized with the combination of neutralization by IgG1b12 and sCD4 than the wild type Tat and ∆V2 Env showed strong antibody (Ab) responses whereas 241∆CHO pseudoviruses were resistant to neutral- after fewer immunizations, in comparison to animals vacci- ization. The 89∆CHO soluble trimer had higher binding to nated with the single proteins. In particular, neutralizing Ab 4E10 and pseudoviruses bearing this mutation had higher (nAb) titers generated after two immunizations were higher in neutralization sensitivity to 4E10, IgG1b12, sCD4 and to Tat/ V2 Env-vaccinated monkeys than in animals immunized plasma than the wild type. ∆ with V2 Env alone. Cross-clade nAb against primary HIV Conclusions: These data showed that the glycan at position ∆ isolates were also induced. Further studies to characterize 236 masks the CD4 binding site whereas that at position 247 humoral responses include epitope mapping of both binding is essential for the formation and function of this site. and neutralizing Abs, the assessment of virus-specific IgG, Furthermore, the carbohydrate at position 89 shields epitopes IgM and IgA, and ADCC activity. in gp120 and gp41. These mutants with exposed CD4 bind- Conclusions: Thanks to these encouraging preclinical data ing site and gp41 epitopes may be important immunogens and to the promising results of Phase I clinical trials conduct- that can be used to induce potent broad neutralizing anti- ed with the single vaccine components, the Tat/trimeric V2 bodies against diverse primary HIV isolates. ∆ Env combined vaccine will enter Phase I studies in humans. These studies were carried out within the AIDS Vaccine Integrated Program (AVIP; EU-FP6 program) and the Italian Concerted Action for the Development of a Vaccine against HIV/AIDS (ICAV). AIDS Vaccine 2006 139 Poster sessions_revGJ 14/8/06 16:34 Page 140 Amsterdam, Netherlands 29 August–1 September 2006 P09A-39 tive of a slower progression to the disease. ICAV grant, ISS, Rome, Italy. Anti-Tat humoral and cellular immunity in HIV-1 infected patients, including asymptomatic individ- uals and LTNP, aimed at identifying immune cor- P09A-40 relates of protection which are key to vaccine development Evidence for underestimation of the repertoire of cytotoxic T lymphocyte epitopes in HIV-1 infection F Ensoli1, V Fiorelli2, A Tripiciano2, A Scoglio2, B Collacchi2, M Ruiz-Alvarez2, A Fazio2, G Paniccia2, IMM Schellens1, C Kesmir2, D van Baarle1, F Miedema1 A Arancio1, F Stivali1, V Francavilla2, O Longo2, S Buttò2, and JAM Borghans1 G Poli3, B Ensoli2 1 University Medical Center Utrecht, Utrecht, The Netherlands; 2 1 Pathology & Microbiology, IRCCS S. Gallicano Hospital, Rome, Utrecht University, Utrecht, The Netherlands Italy; 2 National AIDS Center, ISS; Rome, Italy; 3 AIDS Pathogenesis Unit, IRCCS S. Raffaele Hospital, Milan, Italy Background: There is increasing evidence that cytotoxic T lymphocytes (CTL) play an important role in the control of Aims: HIV Tat represents a very early protein required for HIV infection. Studies addressing the role of HIV-1 specific virus gene expression, replication and infectivity. Based on CTL responses are complicated by the large population diver- this notion, studies have been performed in HIV-1 infected sity of HLA molecules. Nevertheless since the use of overlap- patients, including asymptomatic drug-naive individuals and ping peptide pool screening, it is believed that most conserved LTNP, to analyse the immune response against Tat in the HIV-1 specific CTL epitopes presented via intensively studied course of natural infection. HLA molecules are nowadays known. Patients and Methods: Anti-Tat immune response has been Objectives: To study if HIV-1 epitope databases are indeed cross-sectionally and longitudinally investigated in sera nearly complete, we sought to identify novel HIV-1 specific and/or PBMC from a) 302 HIV-1-infected patients at differ- epitopes presented by HLA-B27 or -B57, two HLA alleles ent disease stage; b) 37 asymptomatic drug-naive patients, c) that are intensively studied because of their well-known cor- 47 LTNP individuals; d) 84 HIV-infected patients before and relation with slow HIV-1 disease progression. during HAART e) 252 individuals with estimated dates of Methods: HIV-1 consensus subtype B sequences were seroconversion and follow-up of 14 years. analysed with peptide prediction programs based on MHC- Results: The prevalence of anti-Tat antibodies (IgM and IgG) binding, proteasomal cleavage and TAP transport. was 17.2% (52/302). Frequencies of anti-Tat IgG and IgM Recognition of the novel identified epitopes by CTL was sub- were 13.2% and 6.9%, respectively. Anti-Tat Ig were more sequently tested using the IFNγ Elispot assay. frequent in stage A patients (21.8%) as compared to stage B Results: In total 22 novel epitopes that were predicted to be (6.1%), or C (4.1%) (P=0.002), respectively. A correlation presented through either HLA-B27 or -B57 were selected. Of was found between anti-Tat antibodies and a more contained these 22 peptides tested, all 11 peptides predicted to be pre- loss of CD4 T cells (P=0.01). In contrast, cellular responses sented through HLA-B27 induced IFNγ production in at least + measured by γ-IFN Elispot were found in the majority (66%) one HLA-B27 individual (median 22% of tested individuals of stage A drug-naive patients. LTNP had the highest preva- gave a positive IFNγ response, range 11–89%), which was lence of anti-Tat antibodies [25.5% (12/47) and 14.9% also the case for 8 out of the 11 epitopes predicted to be pre- (7/47) for IgG and IgM, respectively]. Cumulative anti-Tat sented through HLA-B57 (median 20% of tested individuals IgG and IgM were 38.3% (18/47). Longitudinal analysis gave a positive IFNγ response, range 0–60%). Furthermore, revealed a more contained loss of CD4 T cells (P=0.0009) the magnitude of CTL responses towards the predicted pep- and lower levels of viral load (VL) in anti-Tat positive as tides was similar to the magnitude of responses towards compared to anti-Tat negative LTNP (P=0.02). Anti-Tat anti- known HLA-B27 or HLA-B57 restricted peptides from the bodies were also associated with significantly lower levels of Los Alamos HIV database analysed in the same HIV-1 infect- VL before HAART (P<0.02) and a more pronounced viro- ed individuals. Only two HLA-B27 restricted peptides and logical response to therapy (P=0.05). Finally, Kaplan-Mayer one HLA-B57 restricted peptide gave significantly lower analysis of a cohort of 252 individuals with accurately esti- responses, while one HLA-B27 restricted peptide even gave a mated dates of seroconversion indicated a 60% lower risk for significantly higher response (P=0.007, Mann Whitney). The anti-Tat positive, as compared to anti-Tat negative individu- latter peptide was recognized by as many as 8 out of 9 tested als of developing AIDS. individuals (P=0.033 compared to known HLA-B27 restrict- Discussion and conclusions: The results indicate that anti-Tat ed peptides, Chi-Square test). antibodies 1) are produced by a small fraction of HIV infected Conclusions: This study shows that numerous HIV-1 epi- individuals as compared to anti-Tat cellular responses; 2) are topes have not yet been identified, and that prediction pro- associated with the asymptomatic stage of the infection, lower grams are powerful tools to identify such novel epitopes. levels of virus replication and a stronger efficacy of HAART; 3) have the highest frequency in LTNP and strongly correlate with CD4 T cells preservation and low levels of VL; 4) are predic- 140 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 141 Amsterdam, Netherlands 29 August–1 September 2006 P09A-41 1 European Molecular Biology Laboratory, Grenoble Outstation, Grenoble, France; 2 Institue of Applied Microbiology, University Single molecule force spectroscopy of 2F5 mono- of Natural Resources and Applied Life Sciences, Vienna, Austria; clonal antibody-epitope interactions 3 Polymun Scientific Immunbiologische Forschung GmbH, Vienna, Austria; 4 Hygiene Institut, University of Heidelberg, RG Tawar1, S Allen2, PM Williams2 and JK Ball1 Heidelberg, Germany 1 Institute of Infection Immunity and Inflammation, School of Background: An important target for HIV vaccine develop- Molecular Medical Sciences, 2 Laboratory of Biophysics and ment is the envelope glycoprotein fusion protein subunit gp41. The primary structure of gp41 contains at its mem- Surface Analysis, School of Pharmacy, The University of brane-proximal region the epitopes of two broadly neutraliz- Nottingham, UK ing antibodies, namely 4E10 and 2F5. Peptides mimicking these epitopes interact with the mAbs but fail to induce a neu- Background: The 2F5 monoclonal antibody (MAb) recognis- tralizing immune response. Therefore other factors such as es a linear epitope ELDKWA on gp41 of the virus. Attempts the lipid environment has been speculated to play a role in the to induce primary-isolate-neutralising antibodies with differ- generation of such an immune response. ent presentations of this epitope have not been successful. Objectives: To perform rabbit immunization studies with This justifies efforts to study the interaction of 2F5 MAb with gp41 fragments reconstituted into liposomes in its mem- its epitope in a bid to aid vaccine design. brane-bound conformation. Objectives: 1. Does 2F5 interact differently with monomeric Methods: Several gp41 constructs containing a short cytoplas- and oligomeric presentations of the epitope? 2. If so, does it mic region of gp41, the transmembrane (TM), the membrane- have effect on the binding kinetics? 3. Does having extra proximal (MP) and the C-terminal heptad repeat (CHR) region amino-acids at the C-terminus affect the binding to 2F5? were expressed in E. coli. and purified in buffers containing Methods: We used an atomic force microscope to study the detergent. The purified protein was then reconstituted into dynamic force spectra in order to probe the energy landscape liposomes with a defined lipid composition. of the interaction of 2F5 with the peptide CQN- Results: The detergent soluble gp41 proteins as well as the QQEKNEQELLELDKWAS and an extended peptide CQN- gp41 proteoliposomes bind to the neutralizing antibodies QQEKNEQELLELDKWASLWNWF with the 2F5 MAb. 4E10 and 2F5. Results from micro calorimetry on binding Results: The interaction of 2F5 with the shorter peptide was will be presented. CD spectroscopy experiments reveal that marked by a K of 2.25 sec-1 and with the extended peptide off not only the TM but also the extracellular portion of gp41 by a K of 0.55 sec-1. The energy barriers(x ) were located at off ß present in the constructs contain secondary structure. The a distance of 0.37 nm, and 0.49 nm for the short and the proteoliposomes were used for immunization of rabbits and extended peptide respectively along the dissociation pathway generated reciprocal antibody titres of up to 230000 upon in the direction of applied force. In addition single, as well as three antigen injections. Potential neutralization activity is simultaneous multiple interactions were observed in both currently tested and will be presented. interactions. Conclusions: We suggest that the gp41 fragment presented Conclusions: The altered K and the energy barrier for the off here represents either part of native gp41 or a gp41 interme- two sets of interactions suggest the role of extra amino acids. diate conformation that is highly structured and immuno- We conclude that the multivalent interaction could either be genic in rabbits. because of the bivalency of the MAb or because of single anti- body molecule binding to oligomeric forms of peptide. Further work is being carried out to address the issue of mul- tivalent interaction and to define the contribution made by P09A-43 the additional amino acids to the altered K . off Functional and structural homology between the V3 region of HIV-1 gp120 and the 40s loop of P09A-42 SDF-1a T Kimura1, T Cardozo1 and S Zolla-Pazner1,2 A specific gp41 conformation as immunogen 1 New York University School of Medicine, New York, NY, USA; 2 A Hinz1, H Quendler2, B Ferko2, G Stiegler3, A Wagner3, New York Veterans Affairs Medical Center, New York, NY, USA H Katinger2,3, MT Dittmar4 and W Weissenhorn1 Background: Structural homology was previously described between the V3 region of the gp120 glycoprotein of HIV-1 and the 40s loop of the CXC chemokine SDF-1a. Objective: To evaluate the hypothesis that the structural homology reflects functional similarities between these regions. Methods: A series of chimeric Env expression vectors were constructed using PCR-based mutagenesis in which the AIDS Vaccine 2006 141 Poster sessions_revGJ 14/8/06 16:34 Page 142 Amsterdam, Netherlands 29 August–1 September 2006 crown of the V3 region of the CXCR4-tropic virus HXB2 effects. The vaccine was injected subcutaneously was replaced with portions of the 40s loop of SDF-1a, the approximately 2 cm from the inguinal area in the anterior natural ligand of CXCR4. Binding activities of anti-V3 mon- aspect of the thigh, close to the draining inguinal lymph oclonal antibodies (mAbs) to the chimeric Env proteins, nodes. Animals were vaccinated at months 0, 1, 3, 7, and 10. expressed on transfected 293T cells, were determined by Results: All animals had cellular immunity to naturally FACS. To produce pseudotyped viruses, the Env expression processed viral antigens. 3/6 animals had T helper cell vectors were co-transfected with NL4-3.R-.E-.Luc+ vector responses to Env proteins from multiple, divergent subtypes into 293T cells and the culture supernatants were collected. of HIV-1 (B, C, and E); two additional animals recognized These viruses were used to infect U87.CD4.CXCR4 cells, and multiple distinct subtype B variants of HIV-1. All animals had the infectivity was measured as luciferase activity produced activated CD8+ T cells, measured both by intracellular IFN-γ by infected cells. In addition, chimeric SDF-1a molecules staining and ELISPOT, that recognized naturally processed tagged with FLAG epitopes were constructed using PCR- epitopes delivered by vaccinia constructs encoding Env and based mutagenesis. These molecules, in which the tip of the Env/Gag/Pol proteins from multiple, divergent subtypes of V3 was substituted for a portion of the 40s loop of SDF-1a, HIV (A, B, C, D, E. and F). Noteworthy was the ability of were purified with the anti-FLAG M2 mAb. The biological sera from one animal to neutralize multiple, primary, T cell activity of the chimeric SDF-1a molecules was determined by tropic isolates of HIV-1 (30% of the isolates tested). Ca++ flux assay. Vaccination of HLA A*0201 transgenic mice with vaccine, Results: The HXB2/SDF-1 chimera Env protein expressed on followed by challenge with a recombinant vaccinia vector the surface of transfected cells was recognized by anti-V3 expressing Env/Gag/Pol from a clade E strain of virus, result- mAbs as efficiently as the HXB2 wild type (WT) Env. ed in a two-log reduction in viral titers. Pseudotyped viruses bearing HXB2/SDF-1 chimeric Env dis- Conclusions: Our strategy is based on the belief that infection played significant CXCR4-dependent infectivity. In comple- of humans with HIV-1 results in immunity that while rarely mentary experiments, the SDF-1/HXB2 chimeric chemokine protective, nonetheless creates strong negative pressure on could modulate CXCR4-dependent Ca++ flux. viral replication in vivo manifested as antigenic variation. Conclusions: The crown of the V3 region of gp120 and the This vaccine approach differs starkly from current vaccine 40s loop of SDF-1a are structurally and functionally homol- strategies that target conserved epitopes to achieve cross-sub- ogous. The data help to clarify how both gp120 and SDF-1a type immunological recognition of HIV-1. interact with the CXCR4 chemokine receptor. P09A-45 P09A-44 Stabilization of gp120 in the liganded conforma- An HIV-1 peptide-based vaccine inducing cross- tion by coupling with CD4 mimic – AIDS vaccine subtype immunity in macaques application F Diaz-Mitoma1,2, DE Anderson1, A Azizi1,2, M Ghorbani1,2, G Martin1, EC Kan2, B. Heyd1, Y Sun2, O Combes1, P C Soare1,2, A Ogrel1, S Aucoin1,2, R Frost1,2, S Ogrel1 and Kessler1, A Menez1, K Hartog2, VA Sharma2, JB Ulmer2, C JV Torres1 Vita1, SW Barnett2, IK Srivastava2 and L Martin1 1 Variation Biotechnologies, Gatineau, Canada; 2 Children’s 1 Department of Protein Engineering and Research, CEA Saclay, Hospital of Eastern Ontario, Regional Virology Laboratory, Gif-sur-Yvette, France; 2 Department of Immunology and Ottawa, Canada Infectious Diseases, Chiron Vaccines, Emeryville, CA, USA Background: The unprecedented rate of mutation of HIV-1 Background: CD4 mimics have been shown to possess CD4- and resulting antigenic heterogeneity among the HIV viruses like affinity for gp120, and to induce conformational changes circulating throughout the world poses a significant challenge in the Env leading to CD4i epitopes exposure. Animal immu- to vaccine development. This study measures the immune nizations with gp120-CD4 complexes have induced broadly response to a peptide-based vaccine against variable regions neutralizing antibody responses directed towards CD4i epi- of HIV with the belief that these regions represent areas in topes. However a significant proportion of these antibodies which the virus is susceptible to immune recognition and was directed against CD4. In the context of vaccine develop- elimination. We have synthesized and evaluated the immuno- ment, anti-CD4 antibodies may cause autoimmune disorders, genicity of a vaccine that contains a total of 176 lipidated and therefore undesired. non-lipidated peptide variants that represent 7 antigenically CD4-binding to gp120 has also been reported to increase the variable regions of the Env and Gag proteins of HIV-1. gp120 binding to chemokine receptor. This binding has been Objectives: To evaluate a vaccine strategy that targets only described as a crucial step in HIV fusion process and its inhi- antigenically variable regions of HIV-1 Env and Gag and to bition prevents HIV infection. measure cellular immunity induced by the vaccine candidates. Objectives: The focus of this study is to generate a disulfide Methods: Six cynomolgus macaques were vaccinated with the bond between gp120 and CD4 mimic leading to stabilize vaccine in Montanide ISA-51 adjuvant with no adverse gp120 in the liganded conformation exposing both CD4i epi- 142 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 143 Amsterdam, Netherlands 29 August–1 September 2006 topes and the chemokine binding site. ment, and the potential sites analyses were performed using Methods: We used an R5 isolate gp120 SF162 exhibiting an Prosite tool. Besides, we have calculated the dN/dS ratio to amino acid located in the proximal region of the CD4 bind- suggest the positive selection pressure. ing pocket mutated in cysteine and a CD4 mimic with a SH Results: The HIVbase epitope mapping showed 30 CD8 and group. The stability of the covalent complex was analysed at eigth CD4 epitopes with a high frequency. The subtype B was low concentration by fluorescence polarization. The coupling the most conserved of the subtypes on C1 and C3 epitope yield was also evaluated by size exclusion chromatography in regions, followed by subtype F. On V3loop region the most denaturing conditions and SDS-PAGE. Antigenic properties conserved was the subtype F, however none of the most fre- of each cross-linked complex were investigated with antibod- quent mutations had been associated with the loss of the N- ies directed against CD4i epitopes by surface plasmon reso- glican site at this domain. This epitope conservation suggests nance. The functionality of the covalent complex was that these regions have great importance for viral fitness and investigated in CCR5 expressing cells binding assay. also suggest that these conserved regions are not exposed to Results: At low concentration below CD4 mimic K , the the immune system pressure, in these Brazilian sequences. d covalent complex formed with the disulfide bound remained The epitope VPVWKEATTTL, associated with rapid pro- stable. In contrast, the non-covalent complex was dissociated gression allele, B35 , presented 100% of variation in the sub- at 60%. The disulfide bond between gp120 and CD4 mimic type C isolates and had low variation in non-C subtypes yielded to 60% of coupling. The covalent and non-covalent (18%), suggesting that this allele may be imposing selective complexes bound CD4i epitopes and CCR5 with the same pressure in this subtype. affinity. This cross-linked complex is being evaluated in rab- Conclusions: In this study, the functional regions responsi- bits for its ability to induce broadly neutralizing antibody ble for N-glycosylation sites mapped in HIV-1 proteins were responses. highly conserved. These sites are potentially important for Conclusions: The generation of the disulfide bond between these proteins function and their exclusion could reduce the gp120 and CD4 mimic stabilizes gp120 in the liganded con- viral fitness. An ideal vaccine must contain epitopes to cre- formation increasing its reactivity towards the chemokine ate immune pressure on virus functional regions that cannot receptor CCR5 and exposing CD4i epitopes. Thus, the cova- escape. In addition, it is very important to continue map- lent complex could possess therapeutic potential for the ping the genetic information of the circulating HIV-1 strains blockage of HIV attachment and the generation of broadly from Brazil. neutralizing antibodies. P09A-47 P09A-46 Development of a new vaccine against HIV: viro- Mapping epitope candidates from HIV-1 Brazilian somes incorporating HIV proteins sequences for vaccine development H Quendler1, A Wagner2, G Stiegler2, K Vorauer-Uhl1, LCJ Alcântara3, ATL Queiroz1, LA Santos1, T De Oliveira2, B Ferko1, W Weissenhorn3, A Hinz3 and H Katinger1,2 R Moreau 1 and B Galvão-Castro1,3 1 Institute of Applied Microbiology, University of Natural 1 LASP/CPqGM/FIOCRUZ, Salvador Brazil; 2 Oxford University, Resources and Applied Life Sciences, Vienna, Austria; 2 Polymun Oxford, UK; 3 FBDC/EBMSP, Salvador, Brazil Scientific Immunbiologische Forschung GmbH, Vienna, Austria; 3 European Molecular Biology Laboratory, Grenoble Outstation, Background: The mapping of the epitopes presented on Grenoble, France diverse virus proteins can provide information for vaccine development. The use of bioinformatics softwares to data Background: Our aim is to develop a preventive HIV vaccine analyse is an easier way to search epitopes for HIV epidemic based on virus-like particles, which contain membrane incor- study, and it will provide important information about the porated HIV proteins in their native conformation, derived HIV epidemic characteristics in our country and to develop from primary viruses or recombinant expression. an effective vaccine. Virosomes resemble the natural virus in size and protein con- Objectives: Molecular characterization of the Brazilian HIV- tent. Since these particles neither replicate nor contain the 1 env sequences, mapping into this region the possible previ- HIV genome, they cannot produce progeny virus and there- ously related epitopes from Los Alamos Database. fore avoid the safety concerns associated with whole-inacti- Methods: All sequences of Brazilian HIV-1 (3,813 sequences) vated and live-attenuated virus vaccines. Virosomes have were collected from GenBank and added in HIVbase moreover been shown to allow mucosal application and to be Database for the epitope mapping analysis as described in the able to induce humoral as well as cellular immunity. HIV Immunology and HIV/SIV Vaccine Databases 2003 Materials and Methods: Virus is derived from protein-free from Los Alamos. We selected and analysed only the env cell culture, concentrated by means of ultrafiltration, and region, comprehending 2,644 sequences. The multiple align- then solubilized, followed by a RNA/DNA degradation ment of these sequences was done using ClustalX software. step. Several different recombinant gp41 antigens Genedoc software was used to edit and translate the align- (gp41CTM, gp41int, gp41CHRTM5) have been produced AIDS Vaccine 2006 143 Poster sessions_revGJ 14/8/06 16:34 Page 144 Amsterdam, Netherlands 29 August–1 September 2006 at the EMBL, since gp41 is considered to be one of the most non-mutated full-length tail, shows only 14 ±7 Env/particle. important proteins for vaccine design as it is targeted by An analysis of the Env spike clustering distribution on HIV-1 four of the most potent monoclonal antibodies (2F5, 4E10, virions shows a significantly higher proportion of spikes Z13, Clone 3). within clusters than predicted by chance alone. We have pro- Virosomes are produced by the crossflow injection technique, duced a 3.2 nm resolution model of a SIV spike based on in which lipid vesicles are formed by detergent dilution imme- cryoEM tomogram 3D volume averages, along with provi- diately after injection of a lipid mixture into a micellar pro- sional fits of atomic models of the unliganded gp120 core tein solution containing either recombinant proteins or native structure and the HIV-1 gp41 2F5 and 4E10 epitopes within proteins obtained from virus lysate. this cryoEM tomogram derived model. The surface glycopro- Virosomes are analysed with respect to size, protein content, tein (gp120) “head” of each subunit of the trimeric SIV spike and protein orientation. Immunogenicity is primarily contains a primary mass, with two secondary lobes. The TM assessed in small rodent studies (mice, rabbits), and is later on “stalk” of each trimer is composed of three independent legs to be tested in macaques. that project obliquely from the trimer head, tripod like. Results:. Virosomes are of a homogeneous diameter in the Conclusions: The clustering of Env spikes has implications range of 130 nm, displaying HIV envelope proteins on the for viral fusion and antibody-mediated neutralization. The outer surface as was revealed in FACS studies. In a cross-link- unusual open leg configuration of TM has implications for ing assay we could show that recombinant gp41CTM and the design of future soluble Env trimer vaccine candidates. It gp41CHRTM5 were present in their trimeric state. also suggests a mechanistic explanation for the events leading Furthermore, immunization studies in mice and rabbits were up to membrane fusion wherein the three unassociated mem- carried out. In ELISA experiments sera of animals immunized brane-spanning domains (MSDs) and membrane-associated with virus derived virosomes yielded antibody titers up to elements of the membrane proximal external domains 1:40,000 against gp160. Gp41-CHRTM5 virosomes induced (MPEDs) of a single spike can, collectively, encompass and antigen specific antibody titers up to 1:230,000. The neutral- disrupt the integrity of a relatively large area of the mem- izing potential thereof is currently being investigated. brane, possibly by pulling upon and drawing the outer leaflet Conclusion: The results shown above are encouraging to fur- toward the target cell membrane upon triggering. ther evaluate our virosome approach for the development of an HIV-1 vaccine. P09A-49 P09A-48 Immunogenicity of HIV non-glycosylated Env trimer expressed in Escherichia coli using Distribution and three-dimensional structure of Neisseria meningitidis NadA as scaffold AIDS virus envelope spikes by cryo electron microscopy B Capecchi1, B Burke2, E Kan2, I Ferlenghi1, M Scarselli1, IK Srivastava2, SW Barnett2 and R Rappuoli1 KH Roux1, P Zhu1, J Liu1, J Bess Jr2, E Chertova2, 2 1 3 1 JD Lifson , H Grisé , G Ofek , KA Taylor 1 IRIS, Chiron S.r.l., Siena, Italy; 2 Vaccine Research, Chiron Corporation, Emeryville, CA, USA 1 Florida State University, Tallahassee, FL, USA; 2 SAIC Frederick, Inc., National Cancer Institute at Frederick, Frederick, MD, USA; 3 Background: The Env glycoprotein of HIV is the major tar- National Institute of Allergy and Infectious Diseases, National get of neutralising antibodies and is likely to be an impor- Institutes of Health, Bethesda, MD, USA tant component of an effective anti-HIV vaccine. The mature Env is present as trimer on the virion surface. It’s Background: Though crystal structures for Env subunit frag- composed of an exterior envelope domain (gp120) that con- ments are known, the overall structure of the Env spike and tains most of the surface-exposed epitopes and the CD4 its distribution on virions has remained obscure. binding domain, and a transmembrane domain (gp41) Objectives: To investigate the surface features of HIV-1 and made by two coiled-coil regions. The glycosylation is criti- SIV virions, to generate a 3D model of the SIV Env spike, and cal for the correct folding of the protein, however it can to reconcile this model of authentic virion trimers with avail- mask conserved neutralizing epitopes. able atomic structures of their constituent subunits. Objective: The aim of this work was to express a trimeric Methods: We have used cryoelectron microscopy (cryoEM) non-glycosylated Env protein in Escherichia coli by using the tomography to imaging unfixed and unstained HIV-1 and bacterial protein NadA as scaffold, in order to focus the SIV virions coupled with image averaging and fitting with immune response towards the conserved epitopes that are available atomic models of gp120 and gp41 fragments. generally masked by the extensive glycosylation. NadA is a Results: Mutant (truncated TM tail) SIV virions show a sur- trimeric surface-exposed protein of Neisseria meningitidis face fairly well saturated with 73 ±25 rather uniformly dis- presenting a tripartite structural organization distinctive of tributed Env spikes, in line with our previous data. In striking the novel class of oligomeric adhesins named “Oca” contrast to SIV, yet consistent with our earlier results, (Oligomerization coiled-coil adhesin). cryoEM analysis of HIV-1 (MN) virions, which have a native Methods: Chimeric NadA-Env proteins were generated by 144 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 145 Amsterdam, Netherlands 29 August–1 September 2006 fusing the N-terminal gp120 envelope region to the gp120 proteins possessed two fold higher neutralizing anti- heterologous coiled-coil domain from NadA or to its own body titers to HIV-1 MN than did sera from hamsters gp41 region, used as oligomerization domains. In addition, immunized with a single gp120 protein (P<0.05). The group the membrane region of NadA is included to allow the that received a cocktail of fourteen proteins had the highest anchorage of the chimeric proteins to the bacterial surface. proliferative immune responses to HIV-1 gp120 subtypes B, Results and conclusion: Here we describe the expression of C, and A/E compared to the other groups. Env-NadA fusion proteins in E. coli. We show that Env- Conclusions: Although polyvalent approach achieved only NadA form high molecular weight oligomers exposed on modest increase in the breadth of humoral and cellular the E. coli surface and able to bind to CD4 human receptor. immunity, qualitative change in the vaccine (14 versus 1 Moreover, following animal immunization using a prime- gp120) resulted in a quantitative improvement in vaccine- boost regiment, we show that Env-NadA chimeric proteins induced immunity. Taken together, these results suggest that are able to induce high titers of antibodies that neutralize incorporation of gp120 in a polyvalent format may prove to the homologous primary isolate, SF162 strain. All these be an efficient vaccine formulation against HIV-1. data suggest that the Env domains expressed in E. coli could be properly folded. P09A-51 P09A-50 Evaluation of the HIV CCR-5 co-receptor: conformation and immunogenicity A polyvalent HIV-1 gp120 candidate vaccine induces broad cellular immunity against HIV-1 S Phogat1, I Navratilova2, J Sodroski3, D Myszka2 and subtypes in golden hamsters R Wyatt1 A Azizi1,2, R Frost1, D Albert1, M Ghorbani1, DE Anderson3, 1 Vaccine research Center, NIAID, NIH, MD, USA; 2 Center for DC Montefiori4 and F Diaz-Mitoma1,2 Biomolecular Interaction, University of Utah, Salt Lake City, UT, USA; 3 Dana Faber Cancer Institute, Harvard School of Medicine, 1 Infectious Disease and Vaccine Research Centre, Children’s Boston, MA, USA Hospital of Eastern Ontario, Ottawa, ON, Canada; 2 Department of Biochemistry, Microbiology and Immunology, Faculty of Background: For the elicitation of broadly neutralizing HIV- Medicine, University of Ottawa, Ottawa, ON, Canada; 3 Harvard 1 antibodies, viral heterogeneity is a daunting hurdle. The Medical School, Boston, MA, USA; 4 Human Vaccine Institute, viral co-receptor, CCR5, displays limited heterogeneity and is Duke University, Durham, NC, USA an attractive therapeutic target. Here we investigate CCR5 as a therapeutic vaccine target. Although this minimizes viral Background: The antigenic variation of HIV-1 gp120 is a heterogeneity, cellular vaccine targets possess inherent diffi- major obstacle for the development of an effective vaccine. It culties of tolerance and auto-immunity. Our approach is to has been shown that mono-or bi-valent HIV-1 envelope vac- target discreet regions in CCR5 to potentially overcome cines are poorly immunogenic and induce a limited breadth issues of tolerance or auto-immunity. of reactivity. It is possible that a cocktail of HIV-1 gp120 pro- Methods: For the initial experiment in small animals, we teins containing multiple epitopes will be able to maximize have constructed a N-terminal CCR5 construct expressed as the breadth of immune responses against HIV-1. fusion peptide flexibly linked to heterologous T helper epi- Objectives: To compare and evaluate the immunogenicity of topes derived from tetanus toxoid and ovalbumin. The pep- HIV-1 vaccine containing either gp120 protein alone or in tide was expressed in HEK 293 cells, purified via a combinations of four or fourteen gp120s from different pri- C-terminal His tag and gel filtration, characterized by ELISA mary HIV-1 isolates in immunized hamsters. and tested for immunogenicity in mice and rabbits. Using a Methods: We amplified and characterized 14 different CCR5 cell line, a set of gp120 glycoproteins were analysed gp120s from primary subtype B isolates with both syn- for binding to CCR5 in a Biacore format. We then adapted cytium- and non-syncytium-inducing properties, and CCR5 binding to both an ELISA and a solid phase proteoli- expressed these proteins in Chinese Hamster Ovary cell posome (PLs) format. The CCR5 PLs possess a reconstituted lines. Purified proteins were used either alone or in combi- lipid bilayer and display properties similar to that of the nations of four or fourteen different gp120s to vaccinate native CCR5. We have immunized the CCR5 PLs, along with golden hamsters. Antibody binding, neutralizing antibody the N-terminus CCR5 peptide +T help, into rabbits. The sera titers and T-cell proliferation were tested against HIV-1 sub- are being evaluated for binding to CCR5 by a novel ELISA types B, C, and A/E. and for the elicitation of HIV inhibitory antibodies. Results: Higher levels of antibody titers to HIV-1 subtype B Results: The N-terminal CCR5 peptide raised low titer anti- gp120 proteins (MN and SF162 strains) were detected in bodies to native CCR5 expressed on the cell-surface only the group of hamsters that received a cocktail of fourteen when heterologous T-help was provided. The N-terminal primary recombinant gp120 proteins compared to the CCR5 antibodies had low to no detectable inhibitory activi- groups that received four or one gp120 proteins. Sera from ty in an HIV entry assay. Affinity-purified CCR5 could be the groups that received a cocktail of either four or fourteen captured on a Biacore chip, nanomolar binding of gp120- AIDS Vaccine 2006 145 Poster sessions_revGJ 14/8/06 16:34 Page 146 Amsterdam, Netherlands 29 August–1 September 2006 CD4 complexes and TAK779 could be observed. CCR5 received a boost with MVA-mBN32 as compared to the ani- expressed on the cell surface, by ELISA and in the PL format, mals primed and boosted with EP-1233 alone. Increased bound to the CCR5 monoclonal antibodies 2D7, 3A9 and responses to individual peptides ranged from 2 to 100 fold 45523 with similar affinity. CCR5 presented in all contexts over EP-1233 alone. Similarly, HTL responses were boosted was recognized by gp120-CD4 complexes. Inhibition of viral by MVAmBN-32 by approximately 5-fold; individual pep- entry by anti-sera elicited by the CCR5 PLs immunized into tide-specific responses were boosted by 2 to 10-fold. rabbits will be presented. Conclusions: Based on the ability to effectively stimulate both Conclusions: N-terminus CCR5 is weakly immunogenic, CTL and HTL responses with EP-1233 and MVAmBN-32 as benefits from heterologous help and raises cross-reactive anti- well as to amplify these responses through prime/boost inoc- bodies to the native CCR5. As determined by recognition by ulation make EP-1233 and MVAmBN-32 vaccines promising gp120-CD4, the CCR5 is retained in its native state in the candidates for clinical development. Clinical testing is sched- context of the CCR5 PLs and this may be essential to elicit uled to begin in early 2007. antibodies able to inhibit a variety of HIV-1 strains. P09A-53 P09A-52 Induction of HIV-neutralising antibodies specific for Preclinical evaluation of epitope-based DNA and the membrane-proximal external region (MPER) of MVA vaccines in a heterologous prime/boost the transmembrane envelope protein gp41 regimen J Denner, U Fiebig, M Schmolke and R Kurth D McKinney-Baker1, C Crimi1, A Ludvigsen2, X Shen1, L Vang1, J Vollmar2, J Hain2, P Chaplin2 and M Newman1 Robert Koch-Institute, Berlin, Germany 1 Pharmexa-Epimmune, San Diego, CA, USA; 2 Bavarian-Nordic, Background: Broadly neutralising antibodies specific for Martinsried, Germany the MPER of HIV such as 2F5 and 4E10 have been isolat- ed from infected patients. These antibodies prevented Background: The failure of DNA vaccines, when used alone, infection with SHIVs in monkeys and decreased the virus to elicit high quality immune responses in clinical trials has load in clinical trials. However, many attempts to induce been disappointing. In animal models, viral vectored vac- such antibodies failed so far. cines, used alone or in heterologous prime-boost vaccination Objectives: Based on our immunization experiments using strategies, have often been more effective in inducing cellular the transmembrane protein p15E of porcine endogenous immune responses. The Modified Vaccinia Ankara vectored retrovirus (PERV) and feline leukaemia virus (FeLV), which vaccine (MVAmBN-32) encodes two inserts containing a) 21 resulted in neutralising antibodies recognising two epitopes, cytotoxic T lymphocyte (CTL) epitopes restricted to HLA- one located in the N-terminal part (designated E1), the other A2, HLA-A3/11, and HLA-B7 supertype alleles and the uni- in the C-terminal MPER (E2) of p15E, we investigated the versal helper epitope, PADRE, and b) 18 helper T lymphocyte binding of 2F5 and 4E10 to their epitopes in gp41 and devel- (HTL) epitopes that are HLA-DR3 and/or HLA-DR restrict- oped strategies to induce 2F5/4E10-like antibodies. ed. The DNA vaccine, EP-1233, encodes a matched single lin- Methods: Binding studies were performed with 2F5 and ear arrangement of the CTL epitopes followed immediately 4E10 and synthetic peptides corresponding to their epitope by the HTL epitopes. The MVAmBN-32 and EP-1233 vac- and to the entire surface and transmembrane envelope pro- cines will be tested individually in a Phase I clinical trial and tein. Immunization was performed with a hybrid protein using a prime/boost schedule. based on p15E of PERV, containing the E1 domain and the Objective: To develop a preclinical data package to support E2 (2F5/4E10 epitope) domain of gp41 of HIV-1. Sera were Phase I clinical testing of the EP-1233 and MVAmBN-32 analysed in neutralization assays and by epitope mapping. vaccines. Results: An E1 domain was identified in the N-terminal part Methods: Individual vaccine and heterologous prime/boost of gp41 and a synthetic peptide corresponding to E1 was studies were performed using HLA-A2/Kb transgenic mice. shown to interact with E2 of HIV-1 and to increase the bind- For the prime/boost studies mice were immunized with EP- ing of 2F5 and 4E10 to their epitopes. Sera obtained by 1233 and three weeks later were boosted with MVAmBN-32. immunization with hybrid proteins neutralised HIVIIIB, Bal The control vaccine consisted of peptide pools comprised of and SF162 (12 of 20 sera, titres up to 1:20) and were bind- the HLA-A2-restricted peptides emulsified in Incomplete ing to the E2 2F5/4E10 epitope region. Freund’s Adjuvant or the HTL peptides emulsified in Conclusion: This is the first successful induction of neutralis- Complete Freund’s Adjuvant, and naive mice. Eleven days ing antibodies binding to the MPER. Such hybrid proteins after the booster immunizations, CTL or HTL responses should be used to generate a vaccine inducing broadly neu- were measured using a primary IFN-γ ELISPOT assay and tralising antibodies to prevent or curtail HIV infection. flow cytometry. Results: The cumulative CTL response, defined as the sum of all the responses, increased over 10-fold in animals that 146 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 147 Amsterdam, Netherlands 29 August–1 September 2006 P09A-54 to the continuous replication of virus and thus to the con- tinuous Ag (Env) presentation during infection; or to the Comparative analysis of neutralizing antibody continuous viral Env evolution during infection. We are in responses elicited during infection with a SHIV the process of evaluating these possibilities. expressing the SF162 Env and during immunization with SF162 Env derived gp140 proteins P09A-55 L Stamatatos1,2, NR Derby1,2, T Wrin3, E Moysi1,4, B Burke1,2 and RA McCaffrey1,2 MPR -HBsAg fusion as a vaccine candidate 649-684 against HIV-1 infection 1 Seattle Biomedical Research Institute, Seattle, Washington, USA; 2 University of Washington, Department of Pathobiology, I Cherni, BC Geyer, N Matoba and TS Mor Seattle, WA, USA; 3 Monogram Biosciences, South San Francisco, CA, USA; 4 Department of Molecular Biology, University of School of Life Sciences at Arizona State University, Tempe, AZ, Thrace, Greece USA Objectives: We aim at designing HIV Env based gp140 Background: The HIV-1 envelope protein gp41 is essential to immunogens capable of eliciting broadly reactive neutralizing transmission of HIV virus across mucosal surfaces. The mem- brane proximal region (MPR ) of the gp41 is highly con- antibodies (NAbs). In the present study we compared the 649-684 antibody responses elicited by SF162Env gp140-derived served among various HIV subtypes, harbours epitopes immunogens to those generated during infection of macaques recognized by broadly neutralizing antibodies such as 2F5 with a SHIV virus that expresses the SF162 Env gp160, and 4E10, and is the minimal portion required for binding SHIV . Macaques chronically infected with SHIV the epithelial receptor of HIV-1, GalCer, rendering MPR as SF162P4 SF162P4 develop broadly reactive neutralizing antibody responses and an attractive vaccine candidate. Historically, MPR has by comparing the antibody responses elicited during immu- proven to be poorly immunogenic when administered alone nization and infection we hope to gain information that will so to elicit more potent immune response alternative antigen assist in the development of more relevant SF162gp140- presentation strategies need to be explored. derived immunogens. Objective: We wanted to exploit hepatitis B surface antigen Methods: Rhesus macaques of Indian origin were chal- (HBsAg) ability to form virus-like particles (VLPs) and to test lenged intravenously with cell-free SHIV . Plasma the dynamics of MPR display in context of such entities. Our SF162P4 viremia was monitored by b-DNA. Macaques were also goal was to construct MPR fusion with N-terminus of HBsAg immunized with SF162gp140, DV2gp140, DV3gp140, or and to express it in whole plants and insect cell culture for the DV2DV3gp140 by the ‘DNA prime plus protein boost’ purpose of evaluating its performance as a potential vaccine immunization methodology. The titers, epitope specificity, through subsequent mice immunization studies. neutralizing potency and breadth of serum antibodies were Methods: Translational fusion was created by de novo syn- monitored with established techniques. thesis of MPR region and inserting it into an intermediate Results: During the course of chronic SHIV infection the vector already containing HBsAg gene. His-tag was added for SF162P4- animals examined here developed antibodies that could neu- purification purposes. The fusion’s coding region was further tralize not only SHIV , but also several heterologous pri- subcloned into its respective expression vectors. Transgenic SF162P4 mary HIV-1 isolates, including non-clade B isolates. The N.benthamiana plants were created through Agrobacterium breadth of serum neutralizing activity increased over time. In – mediated transformation. A series of screens was performed contrast, immunization with SF162gp140 or DV2gp140 and high-expressing transformants were selected, grown to immunogens resulted in the generation of NAbs of limited maturity and propagated under greenhouse conditions. Sf9 breadth. Immunization with DV3gp140 or DV2DV3gp140 insect cell line was transfected with the MPR-HBsAg fusion- resulted in the elicitation of only homologous NAbs. Epitope containing bacmid generated using Bac-to-Bac kit from mapping analysis of the epitopes recognized by the antibod- Invitrogen. Expression levels were independently evaluated ies elicited during SHIV -infection and during immu- through polyclonal ELISA against HBsAg and MPR. To test SF162P4 nization with SF162 derived gp140 proteins, indicated that formation of virus-like particles in tobacco plant, we per- the immunogenicity of neutralization epitopes on the SF162 formed sucrose gradient sedimentation of relevant extracts Env is different on the virus associated gp160 and on soluble followed by the Western Blot to confirm the presence of gp140 proteins. In particular, the V1 loop was highly fusion components. Clarified plant extracts were subjected to immunogenic on our gp140 proteins but not on the SHIV- affinity chromatography and purified fusion protein was virus. used to immunize mice. SF162P4 Conclusions/Discussion: The ability of SF162gp140- Results: MPR-HBsAg accumulated to ∼0.05% of total solu- derived immunogens to elicit broadly reactive NAbs is lim- ble protein in our highest-expessing line. The sedimentation ited by the highly immunogenicity of V1 on these profile for plant VLPs was similar to that of the yeast-derived constructs. The generation of broadly reactive NAbs dur- control. Purification the fusion protein and immunization ing SHIV infection could be due to differences in studies are currently underway. SF162P4 structure of soluble gpp140s and virion-associated gp160; Conclusions: We have succeeded in generating a transgenic AIDS Vaccine 2006 147 Poster sessions_revGJ 14/8/06 16:34 Page 148 Amsterdam, Netherlands 29 August–1 September 2006 tobacco plant line that expresses the MPR-HBsAg fusion. affected differently by fusion to LAMP. Therefore, the form Moreover, we observed assembly of our plant-derived fusion of the presented antigen is important for the type and mag- protein into virus-like particles. Immunogenicity of our mate- nitude of immune response. LAMP fusions shown dramatic rial needs further testing in mice model to fully evaluate the enhancement in immunogenicity should be considered for efficacy of MPR-HBsAg fusion as a prospective vaccine. further development. P09A-56 P09A-57 Modulation of immunogenicity of SIV and HIV Ex vivo-primed CD8+ response to HLA-A2- antigens using fusion proteins with lysosomal- restricted (A2-) HIV-1 Gag epitopes associated membrane protein-1 (LAMP-1) J Kan-Mitchell1, K Schaubert1, E Paul1, M Bajcz1, C Bergamaschi1, M Rosati1, A Valentin1, V Patel1, DA Price2, AK Sewell3, JM Brenchley 2 and DC Douek2 GM Zhang1, C Alicea1, P Rothl1, PR Chikhilar2, ET Marques Jr2, TJ August2, BK Felber1 and GN Pavlakis1 1 Wayne State University School Medicine, Detroit, MI , USA; 2 Vaccine Research Center, NIAID, USA; 3 Oxford University, 1 National Cancer Institute at Frederick, Frederick, MD, USA; 2 Oxford, UK The John Hopkins University School of Medicine, Department of Pharmacology and Molecular Science, Baltimore, MD, USA Background: The few A2 -epitopes in current HIV strains do not elicit early acute escape and thus, not effective in viral Background: The generation of modified antigens with control. Therefore, recruitment of previously silent (subdom- altered immunogenicity provides an approach to improve inant) determinants may be advantageous. Here we target efficacy of DNA-based vaccines. We have previously shown p24 Gag, an extremely conserved and potentially critical that SIV DNA vectors encoding viral proteins fused to either immunogen in vaccine design. MCP-3 or β-catenin alter the antigen properties and induce Objectives: To evaluate the CTL response to the immunolog- potent immune responses able to significantly lower viremia. ically silent TV9 in p24. Objectives: To investigate the immune responses to viral anti- Methods: Epitope-specific CTL will be generated from gens fused to the lysosome-associated membrane protein, healthy donors by priming circulating CD8 T cells ex vivo LAMP-1, in mice and macaques. with peptide-pulsed dendritic cells. Methods: Optimized DNA vaccine vectors encoding either Results: We reported that ex vivo-primed SL9-CTL require a HIV or SIV proteins were modified by fusion to the LAMP-1 lower threshold for activation than those specific for a variety cDNA. LAMP-fusion vectors were evaluated for in vitro of CTL epitopes and produce sufficient cytokines to support expression and for SIV or HIV specific cellular and humoral autocrine proliferation for months. In contrast, TV9 consistent- immune responses, upon intramuscular vaccination of mice ly primed multiple T cell subpopulations with discrete intensi- and rhesus macaques. ties of staining for tetramer that are IL-2-dependent. TCRB Results: We tested LAMP fusions to SIV Gag, Env, Pol, Nef, CDR3 sequence analysis of TV9 cultures revealed Distinct TCR Tat and Vif and to HIV Gag. LAMP fusion proteins were high- usage for each subpopulation. Using target cells expressing wild ly expressed and colocalized with MHC-II in the lysosomes/late -type A2 (CIR wt -A2) or an A2 variant point mutated to abro- endosomes. LAMP fusions to Gag and Pol resulted in dramat- gate CD8 binding (CIR -CD8null), SL9-CTLs were shown to be ic increases in immunogenicity. Steady state levels of a selected invariably dependent on the CD8 corecept or to effect killing. In LAMP-NefTatVif fusion protein were dramatically increased contrast, TV9-cultures showed reduced but definite lysis against compared to NefTatVif, and the immunogenicity of this DNA CIR-CD8null cells. Analysis of constituent TV9-clones sorted construct in mice was also increased. A 4-fold increase in with mAbs specific for V b genes revealed both CD8 -dependent Elispot and a 3-log increase in antibody titers were measured. and -independent CTLs. Thus, TV9 appears to elicit more func- In general, upon vaccination of both mice and macaques, tionally avid T cells than SL9. LAMP fusion proteins induced an earlier and stronger humoral Conclusions: Priming of multiple tetramer -binding popula- response. Although LAMP fusions significantly enhanced cel- tions suggests a greater T cell repertoire for TV9 as compared lular immune responses in mice, as measured by Elispot and to SL9. However, our studies suggest that the TV9-CTL intracellular cytokine staining, no significant difference was response does not undergo full clonal maturation, and may found in the magnitude of Elispot response in macaques immu- be a “true primary” response. Understanding the human T nized with LAMP-gag. The antigen specific T cells elicited by cell repertoires to various A2 epitopes may provide important the LAMP fusions showed a skewed phenotype towards IFN- insight to the design of HIV vaccines for these individuals. producing CD8 cells lacking CD28 expression, suggestive of effector memory T lymphocytes. Conclusions: Fusions to LAMP altered dramatically the localization, trafficking, stability and immunogenic proper- ties of different antigens. The fusions potently enhanced both arms of the immune system. Different antigens were 148 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 149 Amsterdam, Netherlands 29 August–1 September 2006 P09A-58 P09A-59 Immunogenicity of solid phase proteoliposomes Glycosylation variants of the HIV-1 Envelope containing elements of HIV-1 envelope glycopro- protein: antigenicity and immunogenicity teins L Xu, XS Du, R Idiart, H Chen and RG Whalen M Tang and R Wyatt Department of Infectious Diseases, Maxygen Inc., Redwood City, Vaccine Research Center, National Institute of Allergy and CA, USA Infectious Diseases, National Institutes of Health, Bethesda, MD, USA Background: The HIV-1 envelope protein (Env) is heavily glycosylated with an average of ~25 N-linked glycosylation Objectives: Structural and binding analysis of the membrane sites in the gp120 subunit. At least ten N-glycosylation sites proximal region (MPR) antibodies 2F5 and 4E10 in complex within the gp120 from both primary-isolate and lab-adapted with their epitopes indicated that antibody-epitope interac- HIV strains have been shown to influence the sensitivity of tions were intimately associated with the lipid bilayer. We the virus to neutralization by CD4 and monoclonal or poly- developed strategies to target the highly conserved gp41 clonal antibodies. MPR by generating MPR mini-proteins and sequential immu- Objectives: The principal aim of this work was to study the nization of proteoliposomes (PLs) containing the native HIV- CD4-binding characteristics, antigenicity, and immunogenic- 1 Env-glycoproteins (gp160 EnvPLs) in a lipid context for ity of a large number of glycosylation variants of Env. inducing antibodies against multiple HIV-1 isolates. Methods: One gene encoding the JRCSF gp140 envelope pro- Methods: The HIV gp41-MPR and Env-glycoproteins were tein was synthesized in which ten potential N-linked glycosy- captured on a solid-phase bead and reconstituted with syn- lation sites were mutated from Asn to Gln. This gene was thetic lipids to form PLs. Mice were immunized with 10 µg recombined in vitro with a second gene encoding the wild- of MPR-PLs or EnvPLs in a prime-boost regimen or by mix- type gp140 protein in which all ten glycosylation sites were ture of two EnvPLs followed sequential heterologous EnvPL Asn. A total of 1,024 possible glycosylation variants are thus boosts. Animals were inoculated 5 times at 4-week intervals possible. Clones were chosen at random and the DNA in CpG adjuvant and analysed for binding antibodies, neu- sequence of the env gene was determined. Clones encoding tralizing antibodies and the elicitation of cellular responses glycosylation variants were used to transfect CHO cells and by intracellular cytokine staining. the supernatants containing secreted gp140 were evaluated Results: Sera from mice immunized twice with either the for binding of CD4 and the broadly neutralizing human mon- MPR-PLs or ADAgp160 EnvPLs recognized the MPR or oclonal antibody b12. ADAgp160 proteins expressed on the cell surface. As expect- Results: Over 200 genes encoding unique glycosylation ed, much higher levels of binding MPR proteins were variants were identified including variants with between observed in mice immunized with MPR-PLs ×2 compared to 1–9 N-linked glycosylation sites in the ten sites. CD4 bind- mice immunized with EnvPLs ×2. Conversely, higher binding ing required, on average, a minimum of 4 of the 10 glyco- was observed to Env-glycoproteins in animals inoculated sylation sites, and strong CD4 binding typically required with EnvPLs compared to the MPR-immunized mice. 6–9 glycosylation sites. Many clones bound b12 more However, both groups displayed a subset of antibodies that strongly than CD4. could cross-recognize each respective immunogen, a minimal Conclusions: A large proportion of the hypoglycosylated Env prerequisite for a prime:boost strategy. Interestingly by proteins were capable of being secreted in CHO cells. ELISA, responses to the MPR could be enhanced by boosting However, for those with less than half of the ten glycosylation MPR-inoculated animals with EnvPLs containing the MPR in sites previously shown to be important for CD4 binding and a native conformation. Surprisingly, mice immunized with neutralization sensitivity, the ability to bind CD4 and b12 homologous ADAgp160PLs showed equivalent to better neu- was severely compromised. Overall, the variants character- tralizing activity against a panel HIV-1 clade-B isolates com- ized represent a spectrum of structural and function features pared to mice primed with a mixture of two EnvPLs and and will allow a clear evaluation of the impact of glycosyla- boosted sequentially with heterologous EnvPLs. FACS analy- tion on the immunogenicity of the Env. sis confirmed specific CD8+ IFN-γ T cell responses were induced by MPR-PL:EnvPL prime:boost or by serial immu- nization of the EnvPLs. Conclusions: Initial mouse immunogenicity experiments demonstrate that the MPR and EnvPLs are immunogenic. A MPR-PL:EnvPL prime:boost strategy may selectively drive MPR-focused immune responses. Sequential inocula- tion of EnvPLs did not result in increased breadth of neu- tralization compared to serial inoculation of the homologous ADAgp160PLs. AIDS Vaccine 2006 149 Poster sessions_revGJ 14/8/06 16:34 Page 150 Amsterdam, Netherlands 29 August–1 September 2006 P09A-60 P09A-61 Identification of Envs that preferentially bind Rational designing of novel Env for generating the cross-reactive neutralizing Ab in chronic stable Env-CD4M33 complexes as potential vaccine patient sera immunogen HL Robinson1, L Lai1, DC Montefiori2, and L Chennareddi1 E Kan1, Y Sun1, K Hartog1, G Martin2, V Sharma1, A Menez2, J Donnelly1, J Ulmer1, L Martin2, S Barnett1 1 Yerkes National Primate Research Center and Emory Vaccine and I Srivastava1 Center, Emory University, Atlanta, GA, USA; 2 Duke University Medical Center, Raleigh, NC, USA 1 Vaccines Research, Chiron Corporation, Emeryville, CA, USA; 2 Department of Protein Engineering, CEA, Gif-sur-Yvette, France Background: A central problem in Ag design is the identifi- cation of Envs that elicit broadly neutralizing Ab. Here we Background: HIV Env undergoes structural rearrangement identify Envs that are targets for the broadly neutralizing Ab upon binding to CD4 and exposes a co-receptor binding site. present in chronic patient sera with the goal of testing these The virus attaches itself to the host cell via the receptor and co- Envs for the potential to elicit broadly neutralizing Ab. receptor, and starts a cascade of events that leads to membrane Objective: To raise broadly neutralizing Ab for incident fusion and virus entry into the cell. We have demonstrated that HIV-1 strains. Env-sCD4 complexes are capable of inducing neutralizing anti- Methods: Sera from 15 untreated US patients were tested for bodies (Sun et al.). However, the use of full length CD4 has their ability to neutralize a panel of 12 incident isolates and potential to induce anti-CD4 responses, therefore, we sought to for the avidity with which each isolate bound the anti-Env Ab identify a CD4 mimetic that is capable of triggering the confor- in the 15 sera. mational change in Env and does not induce anti-CD4 response Results: The neutralization and avidity assays allowed the in vivo. We are currently pursuing several approaches to identi- assignment of the Envs in the incident panel into three fy such a mimic; including CD4 peptide mimetic (CD4M33) groups: (1) those for which neutralization titers and avidity and small molecules. One mimetic CD4M33 binds to Env and indices correlated with the geometric mean titers for the neu- induces a conformational change in the Env protein similar to tralization of the incident panel by the 15 test sera (GMTs), CD4. However, Env needs to be cross-linked with CD4M33 to (2) those for which neutralization titers but not avidity produce a stable complex for evaluation in animal models. indices correlated with GMTs and (3) those for which neither Objectives: We have evaluated several cross-linking strate- neutralization titers nor avidity indices correlated with the gies, however only lower concentration of glutaraldehyde GMTs. Analysis of the sensitivity of the incident isolates to was effective in partially preserving the conformational epi- the broadly neutralizing Ab IgG1b12, 2G12, 2F5, and 4E10 topes. Therefore, there is a need to evaluate other strategies and to sCD4 revealed that all of the isolates in group 1 were either for cross-linking the Env to CD4M33 or designing sin- highly resistant to IgG1b12 and sCD4 and that all in groups gle chain molecules expressing both Env and either full length 2 and 3 were sensitive to neutralization by IgG1b12 and or part of CD4 together. The focus of this work is to design sCD4. IgG1b12 targets the binding site for sCD4. Thus, the novel cross-linking strategies based on an Env and CD4M33 isolates with the best correlations with the broadly neutraliz- docking model. ing activity in chronic sera appeared to have tight CD4 bind- Methods: We have identified several amino acids in the Env ing structures. Studies on one Env revealed that mutations protein that are in close proximity to the CD4M33 binding that exposed the CD4 binding site disrupted the high avidity site, however none of them seems to be involved in direct binding of the cross-reactive neutralizing Ab. binding to CD4M33. Next, we mutated these residues in Env Conclusions: We hypothesize that the cross-reactive activity to cysteines either alone or in combination, and characterized that is present in the majority of chronic sera is an activity the resultant Env mutants for their ability to bind CD4, that cross-links targets on quaternary structures present in CD4M33 and other important mAbs such as b12, X5, 17b, Envs with tight structures for CD4 binding. This activity and 48d using surface plasmon resonance technology. could be cross linking of identical epitopes on two individual Results: Based on the binding profile, we selected an Env with gp120 monomers, or the binding of an epitope whose con- a single cysteine substitution at position 281. It appears that formation is present only on Envs with tight CD4 structures. this substitution did not affect the ability of Env to bind Cys281 to CD4, CD4M33, and the critical neutralizing epitopes rec- ognized by mAb such as b12, X5, 17b and 48d are preserved. Conclusions: We hypothesize that this cys mutant Env (Env ) will have the correct structure, and it will be readily cys281 cross-linked to corresponding CD4 miniprotein cys mutant. 150 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 151 Amsterdam, Netherlands 29 August–1 September 2006 P09A-62 P09A-63 HIV-1 infectivity decay depends on Env and can Expression of HIV-1 subtype C p24 in transgenic be accelerated by neutralizing antibodies plants F Charifi1, S Maloveste1, Per J Klasse2, P Poignard1 S Andersson1,2,3, I Lind1, I Kalbina1, N Scherbak1, S Thulin2 and Å Strid1 1 Centre d’Immunologie de Marseille-Luminy, Institut National de la Santé et de la Recherche Médicale-Centre National de la 1 Life Science Center, Örebro University, Örebro, Sweden; Recherche Scientifique-Université de la Méditerranée, Marseille, 2 Örebro University Hospital, Örebro, Sweden; 3 Department of France; 2 Department of Microbiology and Immunology, Weill Virology, Immunology and Vaccinology, Swedish Institute for Medical College of Cornell University, New York, NY, USA Infectious Disease Control, Stockholm, Sweden Background: A better comprehension of the molecular basis Background: Transgenic plant technology is well established of HIV-1 resistance to antibody neutralization may help the but has had limited use in vaccinology. The mucosal immune design of vaccine antigens capable of eliciting broad neutral- system is an important route of entry for HIV and a poten- izing antibodies. We have recently proposed a model of neu- tially powerful site for immunization. tralization that considers dynamic aspects of viral infection. Objective: To establish a transgenic plant model (in According to this model, viral infectivity spontaneous decay Arabidopsis thaliana) with stable production of p24 for oral may play a role in the sensibility/resistance of virus to anti- immunization. body neutralization. Methods and results: Transformation of Arabidopsis thaliana Objectives: To study and determine the origin of viral infec- was done with the vector Agrobacterium tumefaciens carrying tivity spontaneous decay and to study the impact of neutral- a plasmid with the gag gene region coding for the complete izing antibodies on infectivity decay. HIV-1 subtype C p24 protein. Selection of transgenic plants was Methods: Pseudoviruses were obtained by co-transfection of perfomed by inclusion of a herbicide resistance gene (BASTA) in pNL4.3Luc∆Env and a second plasmid expressing HIV-1 pri- parallel with the p24 gene. The BASTA-resistant plants were mary isolates envelope glycoproteins. We analysed the loss of grown to full size and used for the p24-expression experiments. virus infectivity over time by measuring luciferase activity in The p24 gene was detected by PCR in many of the plants. By U87-CCR5 cells following infection with pseudovirus incu- sequencing we could show that the p24 gene had been correct- bated at 37°C for various length of time. ly and completely inserted. We then performed a detection of Results: Our results confirm that viral infectivity sponta- mRNA expression of the p24 genes in three plants by Northern neously decreases over time in an exponential manner. blot, i.e. confirming the functionality of the inserted genes. p24 Calculated half-life for HIV-1 JR-FL HIV-1 IIIB and protein expression was demonstrated by Western blot (using an SIVmac239 were 16, 10 and 3.5 hours, respectively, which anti-p24 monoclonal antibody) and a commercial p24 antigen suggests that kinetics of viral decay depend on the nature of assay. Almost complete inhibition of the response in the p24 Env. Analysis of JR-FL SOS, which contains a disulfide bond antigen assay was achieved by blocking with an anti-p24 mon- stabilizing the gp120/gp41 interaction, revealed a steady oclonal antibody. infectivity over time. In contrast, HIV-1 IIIB K617A and Conclusions: Successful transfer of the p24 gene was accom- S618A mutants, that display increased gp120 spontaneous plished in a large number of plants. In a few of them expres- shedding, lost their infectivity faster than wild type virus. sion of the p24 protein was demonstrated. These plants have These results suggest that the stability of the gp120/gp41 been selected for expansion and planned animal trials. interaction influences kinetics of viral decay. We then tested the impact of antibodies on virus infectivity decay by incubating viruses with monoclonal neutralizing antibodies and calculating corresponding half-lifes. We observed that virus half-life decreases with increasing con- centrations of antibody, depending on the epitope recognized. High concentration of the CD4 binding site neutralizing anti- body b12 induced a decay of JR-FL SOS infectivity similar to the one of wild type virus, suggesting that complete gp120 dissociation may not be required for loss of infectivity. Conclusions: Our data suggest that the spontaneous expo- nential decay of HIV-1 infectivity depends on the progressive loss of functional Env trimers. In addition, we show that one mechanism of neutralizing antibodies is to irreversibly inacti- vate virus particles by shortening viral half life. AIDS Vaccine 2006 151 Poster sessions_revGJ 14/8/06 16:34 Page 152 Amsterdam, Netherlands 29 August–1 September 2006 P09A-64 P09A-65 Biochemical and immunogenic characterization of Approaches to develop a CD4 mimetic capable of stabilized HIV-1 monomeric and trimeric gp120 binding to, and inducing conformational changes glycoproteins in the Env glycoprotein of HIV-1 B Dey1, M Pancera1, K Svehla1, T Zhou1, J Sodroski2, VA Sharma1, J Cisto1, E Kan1, Y Sun1, M Connolly2, P Kwong1, J Mascola1 and R Wyatt1 L Martin3, JJ Donnelly1, JB Ulmer1, SW Barnett1 and IK Srivastava1 1 Vaccine Research Center, NIAID, NIH, Bethesda, Maryland, USA; and 2 Dana Farber Cancer Institute, Harvard School of Medicine, 1 Vaccine Research, Chiron Corporation, Emeryville; 2 Blood Boston, Massachusetts, USA Testing, Chiron Corporation, Emeryville; 3 Department of Protein Engineering, CEA-Saclay, Gif-sur-Yvette, France Background: The HIV-1 exterior envelope glycoprotein gp120 binds the viral receptors and is a major target for Background: HIV-1 envelope glycoprotein (Env) is the pri- broadly neutralizing antibodies. However, as an immunogen mary target for inducing neutralizing antibodies. Env under- gp120 fails to efficiently elicit antibodies with cross-clade goes structural rearrangement upon binding to CD4 and neutralization ability. The crystal structure of gp120 in a exposes a co-receptor binding site. Through this interaction, complex with the primary receptor, CD4, reveals two distinct the virus attaches itself to the host cell, and starts a cascade domains, which circumscribe a cavity known as the “Phe 43 of events that leads to membrane fusion involving gp41 and cavity” due to the location of CD4 residue F43 at the neck of virus entry into the cell. This process generates transient this cavity. Thermodynamic analyses have suggested that fusion intermediates that could be targets for inducing neu- upon binding CD4 induces large conformational changes tralizing antibodies. We have demonstrated that Env-sCD4 within gp120 and thereby exposes or creates the conserved complexes are capable of inducing broadly neutralizing anti- coreceptor-binding surface of gp120. Based upon the CD4- bodies possibly by targeting conformational epitopes exposed bound structure of gp120, we introduced structural modifi- in Env protein upon binding to CD4. Although Env-sCD4 cations within gp120 that would facilitate the glycoprotein to complexes can induce neutralizing antibodies, there is a attain the CD4-bound conformation in the absence of CD4 potential to induce autoimmune responses against CD4. and still maintain its affinity to bind to the receptor. Objectives: Therefore, we sought to identify a CD4 mimetic Objectives: To evaluate the degree of stabilization of these mod- capable of triggering conformational changes in Env. We are ified proteins as well as their immunogenic potential to elicit currently pursuing several approaches to identify such a broadly cross-reactive neutralizing antibodies against HIV-1. mimetic including CD4 peptide mimetics and small mole- Methods: Specific amino acid residues were altered by site- cules. directed mutagenesis. Antigenic and biophysical characteriza- Methods: We developed several assays using SPR, SEC- tions of the expressed proteins involved ELISA and ITC. HPLC, and ELISA to screen small molecules with structures Immunized rabbit sera were assessed in pseudotyped HIV-1 similar to MC34A for their ability to compete with CD4 and neutralization assays. the monoclonal antibody b12 for binding to gp120 SF162, Results: We have previously shown that substitution of serine and their ability to up regulate binding to monoclonal anti- by a bulky tryptophan residue at position 375 fills the Phe 43 bodies that recognize the CD4 inducible epitopes on gp120. cavity and partially stabilizes gp120. We now demonstrate Results: One mimetic (CD4M33) binds to Env as measured that an additional threonine to serine substitution at cavity by HPLC, BIAcore, and ELISA and induces a conformation- residue 257 relieves an internal clash and restores binding of al change in the Env protein similar to CD4. MC34A is a the HIV-1 broadly neutralizing antibody, b12, while main- small molecule that binds to Env as measured by CD4 com- taining gp120 stabilization. The degree of stabilization is petition assays (HPLC, BIAcore, and ELISA). We performed greater in the context of a gp120 trimer than in a monomer. structure based in silico screening, identified 100 compounds, Homologous as well as non-homologous neutralization was and tested them in various assays. We found three distinct slightly but quantitatively improved with stabilized trimer as classes of small molecules: i) compete for CD4 binding, and an immunogen compared to wild type monomer and trimer. up-regulate CDi-epitopes; ii) compete for CD4 binding, but Mapping studies are underway to determine if any specificity do not up-regulate CDi epitopes; and iii) small molecules that differences exist between the selected immunogens. up regulate CD4i epitopes, but do not compete for CD4 bind- Conclusions: Stabilization of gp120 in the CD4-bound con- ing. We are currently investigating the binding sites for these formation may help elicit antibodies with greater breadth of three classes of molecules using both in vitro and in silico neutralization. Additional modifications for further stabiliz- approaches, and we are developing a new library to screen ing gp120 may improve this effect. Selection of protein con- for higher affinity binders. text is an important consideration for developing improved Conclusion: In addition to yielding important structural infor- vaccine candidates. mation about the apo and liganded structure of gp120, this study will be used to create stable complexes with Env protein for further evaluation in rabbits for their ability to induce broad primary isolate neutralizing antibody responses. 152 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 153 Amsterdam, Netherlands 29 August–1 September 2006 P09A-66 P09A-67 Structural characteristics correlate with immune Loss of a single N-linked glycan in the V2 loop of responses induced by HIV envelope glycoprotein HIV-1 Env results in CD4-independence in viral vaccines infection and enhanced ability to elicit neutraliz- ing antibody responses VA Sharma1, E Kan1, Y Sun1, Y Lian1, J Cisto1, V Frasca2, S Hilt1, L Stamatatos3, JJ Donnelly1, JB Ulmer1, S-L Hu1,2, Y Li1, B Cleveland1, I Klots2, P Polacino2, SW Barnett1 and IK Srivastava1 BA Richardson1, D Anderson2 and D Montefiori3 1 Chiron Vaccines, Emeryville; 2 MicroCal, LLC, Northampton, 1 Department of Pharmaceutics and 2 Washington National MA; 3 Seattle Biomedical Research Institute, Department of Primate Research Center, University of Washington, Seattle, WA, Pathobiology, University of Washington, Seattle, WA, USA USA; 3 Duke University Medical Center, Durham, NC, USA Background: HIV envelope glycoprotein is highly glycosylat- Background: HIV envelope glycoprotein (Env) is the target ed. N-linked glycans are believed to play an important role in for inducing neutralizing antibodies. Env is present on the viral infectivity and evasion from host immune responses. virus surface as a trimer and, upon binding to CD4, a cascade Objectives: To test the hypothesis that removal of specific N- of events leads to structural rearrangement exposing the co- linked glycans on HIV-1 Env will alter viral infectivity, anti- receptor binding site and entry into the CD4+ host target genicity and immunogenicity. cells. Methods: Mutations were introduced into potential N-linked Objective: We have designed monomeric and trimeric Env glycosylation sites in V1, V2 and V3 of the env gene of HIV- constructs with and without deletion in the variable loop 2 1 89.6. Single or multiple mutants were rescued into expres- (∆V2) from SF162, a primary subtype B isolate and per- sion plasmids or chimeric virus SHIV 89.6 for infectivity and formed biophysical, biochemical and immunological studies neutralization analyses. Pig-tailed macaques (6/group) were to establish a potential structure-functional relationship. primed with two recombinant vaccinia viruses, one express- Methods: We expressed these Envs in CHO cells, purified the ing SIVmac239 Gag/Pol, and the other, HIV-1 89.6 Env proteins to homogeneity, and performed extensive biophysi- gp160 in wild-type (WT) or mutant forms. Animals were cal studies to define the binding properties to CD4, structur- boosted 12-14 mo later with SIV gag/pol DNA and the cog- al characteristics, and exposure of epitopes recognized by nate gp140 protein before challenge with SHIV89.6P. b12 and CD4i mAb (17B) on both full length and mutant Results: Among all the mutants tested, N197Q in the V2 loop HIV Env proteins. Parameters evaluated include molecular showed the most pronounced increase in susceptibility to weight, oligomerization state, purity, number and affinity of broadly neutralizing MAb, including those targeting the CD4 binding sites and affinity for b12 and 17b mAbs. CD4-binding site (sCD4-Ig, IgA1b12) and the V3 loop (447- Results: We observed one CD4 binding site per monomer and 52D), but not gp41 (2F5). Unlike WT Env, N197Q mutant three active CD4 binding sites per trimer. A forty-fold differ- mediated infection of U87-CXCR4-MAGI cells in a CD4- ence in affinity of the gp120 monomer versus the o-gp140 independent manner. Immunization with WT Env generated trimer towards CD4 was observed (Kd=58 nM and 1.5 nM, low levels of NtAb against SHIV89.6, SHIV89.6P and HIV- respectively), whereas only a 2-fold difference was observed 1SF162, whereas animals immunized with the mutant Env for the V2 deleted proteins (Kd of gp120∆V2 = 19 nM, Kd of had significantly higher neutralizing activities against the ogp140∆V2 = 9.3 nM). Monomers bound 3-fold tighter to same viruses. Sera from animals immunized with the mutant the monoclonal antibody 17b and at least 3-fold weaker to Env also neutralized a panel of subtype B primary isolates b12, and the V2 deleted monomer gp120∆V2 had weakest with greater potency and breadth than those from animals affinity for b12 (Kd= 446 nM). immunized with WT Env. After intrarectal SHIV89.6P chal- Conclusions: Affinity of CD4 binding correlated with pro- lenge, both groups of immunized animals showed significant portion of the antibodies induced against the conformational reduction of viral load compared to controls. Reduction was epitopes by the corresponding Env, and changes in mono- correlated with prechallenge NtAb. clonal antibody binding correlated with the induction of anti- Conclusion: The loss of a single N-linked glycan in the V2 bodies directed against linear epitopes. Furthermore, loop can result in CD4-independent viral infection, increased biophysical analysis reveals the V2 deletion has broad struc- sensitivity to neutralization and enhanced ability to elicit tural implications in the monomer not shared by the trimer broadly neutralizing antibody responses. and these changes are reflected in the quality of the immune response. These data suggest that biophysical characteristics of HIV Env such as affinity for CD4 and exposure of impor- tant neutralizing epitopes such as those recognized by b12 mAb may be important predictors of its in vivo efficacy, and may serve as important surrogate markers for screening Env structures as potential vaccine candidates. AIDS Vaccine 2006 153 Poster sessions_revGJ 14/8/06 16:34 Page 154 Amsterdam, Netherlands 29 August–1 September 2006 P09A-68 P09A-69 Binding kinetics and energetics of broadly neutraliz- Evaluation of the immunogenic potential of gp41- ing anti-HIV-1 membrane proximal external region based peptide vaccines that contain the caveolin-1 (MPER) human monoclonal antibodies to HIV-1 binding domain (CBD) motif peptide epitopes anchored on lipid membranes indi- cate high entropic barrier to Mab binding Y Huang1, DF Gardiner1, 2, A Leung1, S Basu1, Y Song1, J Zaharatos1 and DD Ho1 SM Alam, M McAdams, D Boren, K Plonk, L Sutherland, RM Scearce and BF Haynes 1 Aaron Diamond AIDS Research Center, The Rockefeller University, New York, NY, USA; 2 Weill Medical College of Cornell Human Vaccine Institute and the Center for HIV/AIDS Vaccine University, New York, NY, USA Immunology, Duke University School of Medicine, Durham, NC, USA Background: Due to the limited success of vaccine designs to induce HIV-1 neutralizing antibodies (Nabs), new B-cell Background: The inability of soluble HIV-1 Env with gp41 epitopes in HIV-1 Env are needed to elicit broad and potent MPER to elicit broadly neutralizing antibodies and crystal neutralizing responses. Previously, we reported that a pep- structures of mAbs 2F5 and 4E10 highlight the importance of tide derived from the gp41 envelope protein of HIV-1 IIIB an unidentified component that is lacking in the MPER (a.a. 616-632) is immunogenic in mice, and the antisera sequence when presented either as a peptide or in the form of raised against this peptide is capable of neutralizing the soluble gp41. It has been proposed that the highly hydropho- autologous virus, albeit with low titer (Ho et al. J Virology bic CDR3 loop of 2F5 and 4E10 might interact with a lipid 61, 1987). A recent study identified a CBD in a region of component on the viral surface. We have recently reported HIV-1 gp41 partially overlapping peptide 616-632 that both 2F5 and 4E10 are polyspecific and bind to the (Hovanessian et al. Immunity 21, 2004). This CBD motif is endogenous phospholipid, cardiolipin. Thus, it is important highly conserved among all available HIV sequences, and a to understand the binding mechanism of anti-MPER mAbs to peptide containing the CBD purportedly induced HIV-1 epitopes when presented in a membrane bound form. NAbs with broad and potent activity. This result raised Objectives: To understand the role of lipid reactivity and to speculation that the CBD may serve as a broad neutralizing define the binding mechanism of anti-MPER mAbs to HIV-1 epitope in HIV-1 vaccine design. Envelope (Env) epitopes, we have measured the binding Objective: We sought to reevaluate the immunogenic poten- kinetics and energetics of mAbs and Fab fragments of 2F5 tial of gp41 peptide 616–632, as well as a peptide that con- and 4E10 to peptide epitopes anchored on synthetic lipid tains the CBD motif. membrane (peptide-liposomes). Method: Rabbits were vaccinated with the peptides emulsi- Methods: HIV-1 gp41 MPER peptides were anchored to fied in CFA, followed by boosting in ICFA, and finally synthetic liposomes via an amphipathic peptide linker and boosted by peptide conjugated to KLH. Serum binding Ab the peptide-lipid conjugates were tethered a hydrophobic titers to both peptides were measured by ELISA and neu- BIAcore sensor chip. Surface plasmon resonance measure- tralizing activities were assessed against a panel of clade B ments were carried out to determine the binding kinetics HIV-1 isolates. and affinity. Binding energetics and thermodynamic para- Results: Rabbits that received peptide 616–632 produced bind- meters were calculated from rate constants measured at ing Abs to autologous sequences following an initial prime and varying temperatures. boost, whereas Ab titers in rabbits receiving the CBD peptide Results: Binding of both whole IgG and Fab 2F5 and 4E10 were lower. After the conjugated peptide boost, all rabbits to their respective epitope peptide anchored on synthetic showed high binding Ab titers to autologous peptides. Sera liposomes could be defined by a two-step conformational from rabbits that received the CBD peptide cross reacted strong- change model. The initial encounter of mAb with peptide- ly with peptide 616–632. We mapped the antibody binding site liposomes occurred with faster kinetics, which remained to the overlap region between peptide 616–632 and the CBD unaffected by temperature changes. The slower, rate limit- peptide (position 625–632). When tested against a panel of ing step of antibody docking to peptide-liposomes was tem- clade B HIV-1 viruses, only weak sera neutralization (1:10 to 20 perature dependent, being stable at lower and highly dilutions) against NL4-3 was obtained. unfavorable at higher temperatures. Conclusion: gp41 peptide 616–632 is immunogenic in rab- Discussion: These results suggested that the presence of mem- bits, and the antisera raised have weak neutralizing activity brane in peptide-liposome conjugates imposed an entropic against HIV-1 NL4-3. The immunogenic determinant on the constraint on the peptide conformation making it thermody- peptide that contains the CBD overlaps with peptide namically less favorable for antibody docking. Thus our 616–632, and the CBD motif per se is not immunogenic. We model suggests there are two potential hurdles for anti-MPER were, therefore, unable to replicate the results reported by neutralizing antibodies to bind to virions- first, a requirement Hovanessian et al. for Mab interaction with lipid, and second, a conformational constraint imposed on the MPER sequences. 154 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 155 Amsterdam, Netherlands 29 August–1 September 2006 P09A-70 Background: Exposure to HIV-1 does not necessarily result in infection and progression toward disease. Discovery of indi- HIV neutralizing anti-CCR5 antibodies exert a viduals who, despite multiple exposures to HIV remain unin- protective effect against disease progression in a fected (ESN) or do not progress toward disease (Long term Non Progressors)(LTNP), have confidently revealed that the subset of long-term non progressors better control of viral infection may be achieved through mechanisms of natural resistance. Antibodies (Abs) to 1 2 1 1 3 L Lopalco , B Weiser , C Pastori , S Ghezzi , R Longhi , G CD4/gp120 complex have been detected in ESN, which could 1 1 Gallotta and G Calori be involved in HIV protection. Objectives: To assess whether these Abs may also contribute 1 San Raffaele Scientific Institute, Milan, Italy; 2 Wadsworth to slow HIV-disease progression, we searched for anti- Center, Albany, NY, USA; 3 CNR, Milano, Italy CD4/gp120 complex Abs in 132 subjects, including 72 Long Term Non Progressors (LTNP), 30 Fast Progressors, and 30 Background: Exposure to HIV-1 does not necessarily result in seronegative donors. infection and progression toward disease. Thus suggests that Methods: Selection criteria for LTNP were: 1. Certified sero- + the control of viral infection may be achieved. Antibodies to conversion ≥7 years; 2. CD4 T cells counts ≥500 cells/ml ; 3. CCR5 have been detected in HIV-exposed but uninfected sub- Absence of any antiretroviral therapy; 4. Asymptomatic HIV- jects (ESN), thus they could be involved in HIV protection. 1 infection and good health conditions. ELISA was per- Objectives: To assess whether anti-CCR5 antibodies may also formed to evaluate anti-complex antibodies. contribute to slow HIV-disease progression, we searched for Results: We found that these antibodies are present at higher anti-CCR5 antibodies in 497 subjects, including 85 Long titers in LTNP as compared to Fast Progressors (P<0.001). In Term Non Progressors (LTNP), 70 Progressors, 135 HIV+ single patients, an association between the presence of HAART treated, and 207 seronegative donors. CD4/gp120 complex antibodies and neutralizing activity Methods: Selection criteria for LTNP were: 1. Certified sero- against R5 dependent strains was found. Noteworthy, com- conversion ≥7 years; 2. CD4+T cells counts ≥500 cells/ml ; 3. petition with soluble CD4 prevalently inhibited binding to Absence of any antiretroviral therapy; 4. Asymptomatic HIV- CD4-gp120 complex in LTNP (-46.0% [60.9–34.8 IQR]) as 1 infection and good health conditions. A competitive bind- compared to FP (-12.7% [33.7–5.4 IQR]) (P<0.001), further ing radio-assay carried out on CCR5+ CD4+ T lymphocytes suggesting a major contribution of anti-complex antibodies was performed. HIV neutralizing activity was evaluated by in the anti-CD4 activity detected in the former. primary viruses. PBMC and CCR5-CD4 transfected cell lines Conclusion: As these antibodies recognize conformational were used as target cells. epitopes within the CD4/gp120 complex, they may be Results: We found anti-CCR5 antibodies in a fraction of involved in modulation of HIV entry process. Thus, they like- LTNP (23.5%), but not in the other populations studied ly represent a marker of disease progression. Our finding (P<0.0001). These antibodies recognized a conformational could be relevant for vaccine design and therapeutics. epitope within the first extramembrane loop of CCR5 and they induced a stable and long lasting down regulation of CCR5 on surface of T lymphocytes, which inhibited HIV P09A-72 entry. In addition, CD4+ lymphocytes from LTNP having anti CCR5 antibodies are resistance to R5 strains of HIV-1. Immunogenicity of a novel particle vaccine tar- Follow-up studies showed that the loss of anti-CCR5 anti- geting HIV-1 gp41 bodies occurred in some subjects and this loss was signifi- cantly associated with a progression toward disease, DF Gardiner1,2, Y Huang1, S Basu1, A Leung1,Y Song1, whereas subjects who retained anti CCR5 Abs maintained M Tsuji1 and DD Ho1 their LTNP status. Conclusions: Induction of anti-CCR5 Abs could be relevant 1 Aaron Diamond AIDS Research Center, The Rockefeller to vaccine design and therapeutics. University, New York, NY; 2 Weill Medical College of Cornell University, Division of International Medicine and Infectious P09A-71 Diseases, New York, NY Background: A successful HIV vaccine will likely require Anti-CD4-gp120 complex antibodies in long-term broad and durable CD4 and CD8-specific cell mediated non progressors HIV-1 positive patients play a immune responses (CMI) as well as potent neutralizing anti- role in slowing disease progression body responses. Non-infectious viral-like particles (VLP) rep- resent a modality capable of inducing both these responses 1 1 1 2 1 L Lopalco , C Pastori , C Alberti , C Casoli , A Lazzarin , and have an established safety record in humans. We hypoth- C Paolucci1, D Breda1 and S Burastero1 esized that VLPs carrying modified gp41 envelope proteins would be capable of inducing antigen (Ag)-specific CMI and 1 San Raffaele Scientific Insitute, Milano, Italy; 2 University of gp41-specific antibody (Ab) responses. Parma, Parma, Italy Objectives: Examine the immunogenicity of HIV VLPs includ- AIDS Vaccine 2006 155 Poster sessions_revGJ 14/8/06 16:34 Page 156 Amsterdam, Netherlands 29 August–1 September 2006 ing modified gp41 proteins with or without a DNA prime. Objective: We hypothesized that a vaccine which presents a Methods: We created a human codon-optimized, clade B con- gp41 molecule with improved helical content and in the con- sensus gag and modified gp41 containing the membrane- text of a viral-like particle (VLP), may better mimic the native proximal sequence inserted into a mammalian bicistronic epitope and be capable of inducing 2F5/4E10-like neutraliz- vector and a baculovirus expression system for use in a DNA ing antibodies. prime, VLP boost vaccine regimen. VLP gp41 content was Methods: We constructed a DNA vaccine encoding a modi- confirmed following sucrose cushion isolation. We hypothe- fied HR2 with improved helical propensity fused to a carrier sized that inclusion of a DNA prime may augment the rabbit Fc fragment. We also generated a VLP that displays the immune responses to a VLP boost. BALB/c mice received modified gp41 on an HIV-1 Gag particle. We vaccinated rab- electroporation-augmented intramuscular priming DNA bits in a DNA prime and VLP boost combination. injections encoding 1. Sham DNA; or 2. gp41 and Gag, fol- Results: We observed that mAb 2F5 binds to the helically lowed by two intraperitoneal VLP boosts consisting of gp41- constrained HR2-MPR gp41 fusion proteins better than the Gag. ELISA was performed to measure the evolution of native protein sequences in vitro. Following a DNA prime anti-Gag and anti-gp41 Abs. Splenocytes were harvested and and VLP boost vaccination regimen, rabbits produced high IFN-g ELISpot assay performed to assess the Gag- specific binding Ab titers against the surrogate capture antigen, 5- CMI responses. Helix bundle of gp41. Detailed epitope mapping indicated Results: Baculovirus-derived VLPs were immunogenic as that these antibodies recognize the N-terminal portion of the indicated by high titer Gag-specific Ab responses following 2F5 epitope, but not the peptide that contains the epitope of VLP delivery (1x105) with or without a DNA prime. Both 4E10. In a CEM M7 based assay, we found that the post-vac- groups displayed detectable IFN-g ELISpot responses to a cination sera had weak neutralizing activities against HIV-1 Gag-specific immunodominant CD8+ specific peptide. 5- NL4-3 (IC50 of 1:10 to 20 dilutions). The IC50 of purified Helix bundle gp41 protein was used to detect gp41-specific total rabbit IgG fell between 50-100 µg/ml. No neutralizing IgG responses. Inclusion of a gp41-Gag DNA prime signifi- activity was detected against HIV-1 JRFL and VSV envelope cantly improved the IFN-g ELISpot responses as compared to viral control. VLP immunogens alone (P<0.0001). Importantly, inclusion Conclusion: These results suggest that a vaccine strategy tar- of a DNA prime also provided significantly higher gp41-spe- geting the gp41 MPR is capable of inducing weak HIV-1 neu- cific Ab titers compared to VLP vaccination alone. tralizing Ab against certain HIV strains. Efforts to induce Conclusions: We conclude that: 1). VLP-based vaccine reg- broad and potent 2F5/4E10-like neutralizing Abs will require imens are immunogenic and capable of inducing gp41-spe- more than the structural constraints and epitope viral mem- cific immune responses; 2). gp41-Gag DNA priming brane presentation in this vaccine design. enhances both the resulting gp41-specific Ab titer as well as the CD8-specific CMI responses as compared to VLP vaccination alone. This study supports the inclusion of a DNA priming arm in the analysis of humoral immuno- genicity as well as VLPs as a platform to display novel HIV antigens for vaccine design. P09A-73 DNA prime and VPL boost vaccination strategy tar- geting membrane-proximal region of gp41 induces weak HIV-1 neutralizing antibodies in rabbits DF Gardiner1,2, Y Huang1, S Basu1, Y Song1, A Leung1 and DD Ho1 1 Aaron Diamond AIDS Research Center, The Rockefeller University, NY, NY, USA; 2 Division of International Medicine and Infectious Diseases, Weill Medical College of Cornell University, New York, NY, USA Background: The membrane-proximal region (MPR) of the HIV-1 transmembrane protein is critical for envelope-mediat- ed membrane fusion and contains the targets for broadly reactive neutralizing antibodies 2F5 and 4E10. Structural analysis of these neutralizing epitopes indicates that they are likely constrained by the helix of the gp41 heptad repeat 2 (HR2) and that the epitope may include the viral membrane. 156 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 157 Amsterdam, Netherlands 29 August–1 September 2006 P09A-74 (MC), both composed of international experts. Results: AVIP Vaccine Candidates. Based on preclinical AIDS Vaccine Integrated project (AVIP): a con- results four different vaccine candidates and formulations sortium funded by the FP-6 EU program have been selected: 1. Tat ± ∆V2 Env; 2. Nef ± ∆V2 Env; 3. Multi-HIV antigens/epitopes [full-length rev, tat, nef, gag B Ensoli1, B Wahren2, R Le Grand3, R Gavioli4, CA (p17, p24) full-length antigens, and over 20 T cell epitopes from Pol, Protease and Env antigens]; 4. HIV multigene (nef, Guzman5, M Magnani6, I Stanescu7, V Erfle8, S Barnett9, rev, tat, gag, rt, env) F Gotch10, E Vardas11 , R Glashoff12, M Clerici13, G Poli14, 15 16 Conclusion: Preclinical studies of safety and immunogenicity H Holmes , and A Caputo have been concluded and the vaccines will soon enter Phase I studies in 5 countries. At the same time, virological and 1 National AIDS Center, Istituto Superiore di Sanità, Italy; 2 immunological studies are being carried out in South Africa Microbiology and Tumorbiology Center, Karolinska Institute, and the AVIP international School is fully active at imple- Sweden; 3 Laboratoire d’Immuno-Pathologie Experimentale, menting the training program. SNV/DRM/SDV/CEA, France ; 4 Department of Biochemistry and Molecular Biology, University of Ferrara, Italy; 5 Microbial Pathogenicity and Vaccine Research, GBF, Germany; 6 Institute P09A-75 of Biological Chemistry “G. Fornaini”, University of Urbino, Italy; 7 FIT Biotech Oyj Plc, Finland; 8 Institute of Molecular Virology, Reservoir cells no longer detectable after a heterol- GSF, Germany; 9 Chiron Corporation, Italy and USA; 10 ogous SHIV challenge with the synthetic HIV-1 Tat Department of Immunology, Imperial College, United Kingdom; Oyi vaccine 11 Perinatal HIV Research Unit, Chris Hani Baragwanath Hospital, University of the Witwatersrand, South Africa; 12 EP Loret1, JD Watkins1, S Lancelot1, GR Campbell3, Department of Medical Virology, University of Stellenbosch D Esquieu2, J de Mareuil1, S Opi4, S Annappa2, and Medical School, South Africa; 13 Dipartimento di Scienze J-P Salles2 Precliniche (DISP) LITA Vitalba/Facoltà di Medicina e Chirurgia - Immunologia e Immunopatologia, Italy; 14 DIBIT Unit AIDS 1 UMR Univ. Med./CNRS FRE 2737, Faculté de Pharmacie, Immunopatogenesis Unit, Italy; 15 Retrovirology Division, Université de la Méditerranée, Marseille, France; 2 SynProsis, National Biological Standards Board (NIBSC), United Kingdom; Hôtel Technologique BP 100, Technopôle de Château Gombert, 16 Department of Histology, Microbiology and Medical Marseille, France ; 3 Department of Pediatrics, Division of Biotechnology, University of Padova, Italy Infectious Diseases, University of California San Diego, CA, USA ; 4 Laboratory of Molecular Microbiology, NIAD, National Background: Vaccines based on viral structural products Institutes of Health, Bethesda, Maryland, USA (Env/Gag/Pol) alone have failed to prevent infection by HIV/SIV. More recently, vaccines based on viral regulatory Objectives: Extra-cellular roles of Tat might be the main gene products (Tat/Rev/Nef) have been shown to contain cause of maintenance of HIV-1 infected CD4 T cells or virus replication and to prevent disease onset. A vaccine com- reservoir cells. We developed a synthetic vaccine based on bining both regulatory and structural viral antigens (com- a Tat variant of 101 residues called Tat Oyi, which was bined vaccine) is likely to be superior to single modalities identified in HIV infected patients in Africa who did not since it can induce immune responses targeting both early progress to AIDS. and late viral products. The combined vaccine can also Methods: We compared, using rabbits, different adjuvants exploit immunomodulatory functions of HIV regulatory authorized for human use to test on ELISA the recognition of genes/products, which can improve immune responses (inten- Tat variants from the five main HIV-1 subtypes. A formula- sity, breadth) against structural antigens. tion was tested on macaques followed by a SHIV challenge Project objectives: 1. To conduct Phase I trials in EU with 4 with a European strain. novel combined vaccines in both HIV-infected and low risk Results: Tat Oyi with Montanide or Calcium Phosphate gave healthy individuals. The candidates will be first tested in pre- rabbit sera able to recognize all Tat variants. Five on seven clinical models (mice, monkeys) to optimize the formulations Tat Oyi vaccinated macaques showed a better control of and the vaccination protocols. Preclinical efficacy will be viremia compared to control macaques and an increase of evaluated in two novel mouse models and in monkeys; 2. To CD8 T cells was observed only on Tat Oyi vaccinated perform feasibility studies and technology transfer in DC for macaques. Reservoir cells were not detectable at 56 days future Phase II/III trials; 3. To carry out training in EU coun- post-challenge in all Tat Oyi vaccinated macaques but not in tries and DC. To this end, the AVIP International School has the controls. been established; 4. To ensure community involvement both Conclusions: The Tat Oyi vaccine should be efficient world- in EU countries and DC. wide. No toxicity was observed on rabbits and macaques. We Methods: The AVIP project is composed of 16 Centers from show in vivo that antibodies against Tat could restore the cel- 7 countries and includes the EVA program. AVIP is managed lular immunity and make it possible the elimination of reser- through a Steering Committee (SC), which is supported by voir cells. the Advisory Board (AB) and the Monitoring Committee AIDS Vaccine 2006 157 Poster sessions_revGJ 14/8/06 16:34 Page 158 Amsterdam, Netherlands 29 August–1 September 2006 P09A-76 P09A-77 Very broadly cross-reactive neutralizing anti- Preparation of a therapeutic vaccination strategy bodies induced by HIV-1 oligomeric gp140 in aiming at functional T cell responses against the rabbits: distinction from partially cross-reactive early regulatory HIV-1 proteins Tat, Rev and Nef response and basis for design of non-human primate challenge studies ADME Osterhaus1, SD Allard2, RA Gruters1, CA van 1 1 2 2 G Quinnan1, P Zhang1, F Cham1, M Dong1, A Choudhary1, Baalen , ME van der Ende , P Lacor , JL Aerts , and 2 Z Zhang1, Y Feng1, Y Shao2, L Wang1, N Mathy3, G Voss3 K Thielemans and C Broder1 1 Erasmus MC Rotterdam, The Netherlands; 2 Vrije Universiteit 1 Uniformed Services University of the Health Sciences, Brussel, Brussels, Belgium Bethesda, USA; 2 China Center for Disease Control and Prevention, Beijing, China; 3 GlaxoSmithKline Biologicals, SA, Objective: We previously showed that the presence of CTL Rixensart, Belgium against the early regulatory proteins inversely correlates with progression to AIDS. Our objective was to explore the Background: A major goal of HIV-1 vaccine development is underlying mechanism and the possibility to induce these induction of broadly cross-reactive neutralizing antibodies. CTL after vaccination. The search for an immunogen capable of inducing such a Methods: We constructed recombinant HIV’s in which an response has been difficult. epitope from the late RT protein was expressed as part of the Objectives: Evaluate the immunogenicity of a selected HIV- early Nef protein. CTL were tested to suppress the replication 1 envelope glycoprotein (gp140 or gp120) with the adju- of wild type and recombinant viruses. vant AS02A. The strain of envelope used was R2, from a Macaques were vaccinated with recombinant vectors donor with very cross reactive neutralization of primary expressing the early regulatory or late structural HIV pro- viruses. teins or empty vectors. CTL populations induced after vacci- Methods: We immunized rabbits with envelope glycopro- nation were compared for their efficacy to suppress virus tein from the HIV-1 strain R2 in the adjuvant, AS02A. replication in vivo. Groups of three rabbits each were immunized with doses of Ex vivo generated human dendritic cells (DC) were for the 30 µg of either gp120 or oligomeric gp140 at 0, 3, 6, and induction of T cell responses against the early HIV proteins. 28 weeks. Antibody responses were determined post third For this purpose we constructed a chimeric mRNA targeting and fourth doses against 47 different strains of HIV-1, the early regulatory proteins Tat, Rev and Nef to the HLA including 19 subtype B strains, 14 subtype C strains, and class II presentation pathway. Mature DC electroporated subtype A, D, AE, F, AG, H, and complex CRF envelopes. with this chimeric mRNA were used to generate functional T- The subtype B and C panels recently assembled for the NIH cell responses in vitro. repository were included. Results: A CTL population, directed against the late RT Results: Sera from controls rarely inhibited infectivity, but epitope, became more efficient in suppressing HIV replica- from gp120 and gp140 immunized rabbits inhibited infec- tion when the epitope was expressed early in the virus tivity ≥50% of 9 or 47/47 strains, respectively. There were replication cycle. 2, 12, and 33 strains inhibited by 1, 2, or 3 of the 3 gp140- After challenge with SIV all macaques became infected, irre- immune sera, respectively. Fifty percent neutralization spective of the vaccine used. However, SIV replication titers were ≥1:20 for one or more sera from gp140 immu- appeared to be more limited and shorter lived in animals with nized rabbits for 43/47 strains. Sera from gp140 immu- CTL against the early proteins, than in animals with CTL nized rabbits also neutralized the pathogenic SHIV strains against the late proteins or control animals 89.6p, SF162p3, and DH12R (Clone 7), and two of the Messenger RNA electroporated DC, generated from blood SHIV were neutralized by gp120-immune sera. Strains sen- monocytes of HIV-1 infected patients, expressed the chimeric sitive to neutralization by gp120 sera were neutralized protein and had preserved phenotypic and functional charac- more often by post dose 3 sera than other strains. Studies teristics. Co-culturing these DC with autologous T cells gen- + of selected mutants of the R2 and another strain indicated erated vigorous and broadly directed HIV-specific CD4 and + that the gp120 induced antibodies may be directed at V3 CD8 T cell responses. The T cells were endorsed with a poly- and other epitopes, and that sensitivity could be dramati- functional cytokine secretion pattern and a strong antigen- cally affected by a particular mutation in the membrane specific cytotoxic activity. proximal region of gp41. Conclusions: These observations support the hypothesis that Conclusions: The results constitute the first demonstration of CTL directed against early expressed proteins are efficient in an HIV-1 neutralizing response to immunization that is truly suppressing virus replication in vitro and in vivo. Since ex broadly cross-reactive, provides new principles for design of vivo generated DC are potent inducers of T cell responses, we non-human primate immunization and challenge studies, and propose administration of these DC expressing the early reg- establishes a model system for dissecting the basis for highly ulatory proteins Tat, Rev and Nef as a promising approach to cross-reactive neutralization of HIV-1. therapeutic vaccination in HIV-1 infection. 158 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 159 Amsterdam, Netherlands 29 August–1 September 2006 P09A-78 Functional constraints and antibody recognition of the CD4-binding site on HIV-1 gp120 PD Kwong1, T Zhou1, L Xu1, B Dey1, S Xiang2, MY Zhang3, DR Burton4, D Dimitrov3, J Sodroski2, R Wyatt1, and GJ Nabel1 1 Vaccine Research Center NIAID, Bethesda, MD, USA; 2 Dana- Farber Cancer Institute, Boston, MA, USA; 3 Center for Cancer Research and Nanobiology Program NCI, Frederick, MD, USA; 4 The Scripps Research Institute, La Jolla, CA, USA Background: Binding of the HIV-1 gp120 envelope glyco- protein to CD4 is a requisite first step for the entry to tar- get cells of virtually all primary HIV-1 isolates. This functional constraint of CD4 binding restricts antigenic variation and steric occlusion of the CD4-binding site on gp120 and may therefore be relevant to the development of broadly neutralizing antibodies. Objectives: To evaluate conformational restrictions on the CD4-binding site, their effect on CD4-binding site-directed antibodies, and the structural basis for effective CD4-binding site antibody neutralization. Methods: Iterative structure-based methods were used to cre- ate variants of gp120 stabilized in the CD4-bound confor- mation. Surface-plasmon resonance and isothermal titration calorimetry were used to characterize the binding properties of these stabilized variants to CD4 and to gp120-reactive antibodies. X-ray crystallographic analysis was used to reveal atomic-level details for effective antibody recognition of the CD4-binding site. Results: Kinetic analysis showed that initial engagement of CD4 and gp120 was not affected by conformational stabi- lization. Rather, stabilization affected CD4 dissociation. Stabilization also severely restricted the binding of all CD4-binding-site antibodies tested, except for the unique broadly neutralizing antibody b12. A stabilized gp120 was crystallized in complex with the antigen-binding fragment of b12. Structural analysis at 2.3 Å resolution revealed that the primary contact site of b12 was with the functionally critical CD4-binding loop of gp120. This loop is part of a conserved surface, which does not undergo large changes in conformation. Conclusions: Functional constraints compel exposure and accessibility of the gp120 surface involved in initial engage- ment with CD4. This same surface forms the binding epitope for the broadly neutralizing antibody b12. The difference between CD4 and b12 binding is that CD4 requires gp120- conformational change to form a stable complex, whereas b12 binds with high affinity without additional restriction. The results provide a framework by which to understand the functional constraints of CD4 binding and to explore whether these constraints can be used to elicit antibodies that neutralize diverse strains of HIV-1. AIDS Vaccine 2006 159 Poster sessions_revGJ 14/8/06 16:34 Page 160 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 9B: VACCINE CONCEPTS & DESIGN – DELIVERY METHODS P09B-01 P09B-02 Evaluation of a plasmid DNA prime MVA boost HIV-1 specific humoral, cellular and reminiscent vaccine regimen intended for human use immune responses induced in Chinese rhesus macaques by DNA-tiantanvaccinia combined A Bråve1, K Ljungberg2, B Moss3, P Earl3, P Blomberg4, AIDS vaccine R Stout5 J Cox6 and J Hinkula1 GB Yang1, Q Liu1, Q Sun2, Y Liu1, Y Liu1, HJ Li2, MS 1 Microbiology and Tumor Biology Center, Karolinska Institutet and Sun2, JJ Dai2, KX Hong1 and Y Shao1 Swedish Institute for Infectious Disease Control, Stockholm, Sweden. 2 Carolina Vaccine Institute, University of North Carolina 1 National Center for Diseases Control and Prevention, China- at Chapel Hill, USA. 3 Laboratory of Viral Disease, NIAID, NIH, CDC, Beijing R. R. China; 2 Institute of Medical Biology, CAMS. Maryland, USA. 4 Vecura, Karolinska University Hospital Huddinge, Kunming, P.R.China Sweden. 5 Bioject Medical Technologies Inc, Tualatin, Oregon, USA. 6 Walter Reed Army Institute of Research, Rockville, Maryland, USA Objectives: To evaluate the immunogenicity of DNA-Tiantan Vaccinia Combined AIDS Vaccine (pGPNEF, pGP140 and Background: We developed an HIV-1 multigene/multisubtype Tiantan Vaccinia AIDS vaccine) in rhesus macaques. vaccine that induces potent responses in small animals. As a Methods: Twelve Chinese rhesus macaques were divided into 4 strategy to further enhance immunogenicity we boost the plas- groups: the first group of 4 monkeys was primed, boosted twice mid vaccine-induced responses with a viral vector. with pGPNEF+pGP140 intramuscularly (i.m.), and then boost- Objectives: To evaluate a multigene/multisubtype plasmid ed intradermally with replication competent recombinant prime MVA boost immunization regimen. Tiantan vaccinia virus HIV vaccine (rTV). Another group of 4 Methods: BALB/c mice were primed three times with a cocktail monkeys was immunized with rTV for 2 times. Two groups of of plasmids encoding gag of subtypes A and B, envelope (env) 2 monkeys each were immunized with empty vectors for con- of subtypes A, B and C, rev and reverse transcriptase (RT) of trol. The HIV-specific antibodies and antibody titers were subtype B. Due to findings in previous experiments where mice examined by ELISA and CTL responses by ELISPOT using Env displayed a reduced response to env in the presence of gag, the and Gag polypeptides as stimulants. Reminiscent responses DNA was divided into two entities, one containing the gag and were stimulated with rTV 10 moths later. RT genes and one containing the envelope genes. The two enti- Results: HIV-1 specific antibodies were observed in all immu- rd ties were injected at separate locations intradermally using the nized monkeys after the 3 DNA immunization and were Biojector. The envelope-encoding DNA was adjuvanted by enhanced significantly after the boost immunization with recombinant murine GM-CSF. The MVA used for boosting rTV. Antibody titers reached up to 1:1024 after the final responses encodes HIV antigens of the circulating recombinant boost immunization. Env-specific responses were observed in form A_E originating from Southeast Asia. Responses elicited 3 of the 4 monkeys and Gag-specific responses were observed by DNA or MVA alone were compared to responses induced by in 2 of the 4 monkeys. Antibody and cellular responses were the prime boost schedule. The immunological readouts for the relatively weak in the group immunized with rTV alone. cellular responses include IFN-α and IL-2 ELISpot as well as Significantly increased titer of antibodies (up to 1:4048) and intracellular cytokine staining. The HIV-specific humoral increased number of interferon γ positive cells (up to more response was analysed by ELISA and the viral vector-specific than 3000/million cells) were observed more than 10 months response was analysed by neutralization of vaccinia. after the final boost immunization. Results: Both DNA and MVA induced potent immune respons- Discussion: These data indicate that DNA-Tiantan Vaccinia es by themselves. However, the responses induced by the DNA Combined AIDS Vaccine could induce humoral, cellular and prime MVA boost were superior to responses following the sin- reminiscent immune responses in all the immunized monkeys. gle modality immunizations, both in terms of cellular and In comparison to immunization with DNA or rTV alone, DNA humoral responses. Prime boost animals displayed very strong + rTV immunization induced much higher titers of antibodies cellular responses with 14% and 18% of their CD8+ cells spe- and much higher number of interferon γ secreting cells. cific for gag and envelope, respectively. A significant part of the Conclusion: Boost immunization with replication competent HIV-specific CD8+ cells was doubly positive for IFN-α and IL-2 rTV after DNA prime significantly increased both antibody secretion and cellular responses were directed against several and CTL responses. Replication competent Tiantan vaccinia subtypes. The humoral responses to HIV-antigens were also sig- vaccine boost might be a useful vaccination strategy. nificantly boosted by MVA. Conclusions: The combination of these DNA and MVA con- structs induces potent responses in small animals and is cur- rently being evaluated in a phase I clinical trial in Sweden. 160 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 161 Amsterdam, Netherlands 29 August–1 September 2006 P09B-03 ical trials of the DNA-vaccine based on 4 genes of Russian region-specific HIV-1 subtype A. Immunological properties have been studied using two forms of vaccine: one, utilizing Activity of different vaccine-associated promoter plasmid transfer mediated by Salmonella typhimurium strain elements in human dendritic cells T10, and also i.m. injectional form. Methods: Env (gp140), gag, pol (RT), nef subtype A HIV-1 T Papagatsias1, G Rozis1, P Athanasopoulos2, F Gotch1, strain (GenBank #AF413987) were used to develop recombi- G Dickson2 and S Patterson1 nant plasmids. Gene sequences were obtained with different codon optimization ranges. In vitro expression was detected 1 Imperial College London, Immunology, London, United by means of Western-blot analysis. Plasmid DNA for immu- nization was purified by chromatography with Sephacryl S- Kingdom; 2 Royal Holloway University of London, Biological 1000. Intramuscular immunization vaccine constitutes of an Sciences, Egham, United Kingdom equimolar mixture of four plasmids. The oral form of vaccine is a lyophilized tablet containing equal number mix of Background: Vaccine design approaches that target dendritic S.typhimurium strains T10 cells, transformed with each of cells (DC) usually aim at achieving high levels of transgene plasmids respectively. BALB/c mice were oral or intramuscu- expression. Careful selection of the promoter element driving larly administered with test DNA vaccines. Various prime- expression of the foreign gene is therefore important. boost combinations were studied. The levels, dose and time Aim: To evaluate the activity of different promoter elements dependencies immune responses were evaluated by CTL in the context of human DC. analysis, intracellular cytokine staining and ELISA. Methods: We have constructed adenovirus vectors carrying the Results: Plasmids with env (gp140), gag, pol (RT), nef genes gene for enhanced green fluorescent protein (eGFP) driven by HIV A subtypes were constructed and genes were humanized. three different promoter elements, CMV, CMV5 and Ubiquitin More high expression optimized genes was showed in vitro. C (UbC) promoter, and analysed their activity in different pop- The investigated vaccines have been purified and character- ulations of human DC. Cells tested included blood plasmacy- ized. The immunological properties were studied at i.m. and toid (pDC) and myeloid DC (mDC), monocyte-derived DC oral vaccination of mice, and also at double oral, triple i.m., (moDC), Langerhans (LC) and dermal type DC (dDC). We double i.m. followed by oral boost administration. have also compared our findings with results obtained upon Conclusions: Priming of mice by means of oral vaccine lead- infection of human cell lines 911HER and HeLa. ed to the production of specific CTL, CD8 IFNg, with a dose- Results: Although the CMV5 promoter was more active than dependent Th1/Th2 rate. On oral boosting the response the other two promoters in HeLa and 911HER cell lines, in switches to Th2, coinciding with decrease of CTL activity. human DC the highest level of transgene expression was seen After the i.m. priming a more efficient production of CTL with the CMV promoter. There was very low level of GFP and CD8 IFNg was observed, and also induction of Th1 expression in all cell types transduced with constructs carry- response. The following i.m. boost resulted in an increase of ing the UbC promoter. Highest expression levels were CTL, CD4 IFNg cell activity. CTL are detectable until the observed in moDC, cultured mDC and LC and the lowest lev- 10th week after vaccination. Antibodies were observed from els in pDC. Expression of GFP was augmented in all DC pop- 7th to 10th week. On i.m. priming and oral boost an increase ulations upon stimulation with CD40 ligand (CD40L). of CTL and CD8 IFNg activities, and it leads to more quick Conclusions: These findings demonstrate that the CMV pro- switching from Th1 to Th2 response, than in case of i.m. moter is the most effective of the three promoters tested in a prime-boost administration. range of different human DC populations. P09B-05 P09B-04 Multivalent HIV-1 envelope vaccines delivered in Immunological properties of oral Salmonella the VEErep/SINenv alphavirus replicon particle typhimurium-based, and injectional forms of chimera DNA-vaccine administered in various prime-boost combinations in mice Y Lian1, B Burke1, VR Gomez-Roman1, I Srivastava1, CE Greer1, G Otten1, H Liu1, H Legg1, E Kan1, S Hilt1, T EN Dukhovlinova, BV Murashev, YuP Galatchiants, Tang1, D Montefiori2, J Ulmer1, J Donnelly1, J Polo1 and IV Dukhovlinov, IV Murasheva, AA Kozlov, SW Barnett1 EV Kazenova, MS Pavlova, AE Masharsky, NA Klimov and AP Kozlov 1 Chiron Corporation, Emeryville, CA, USA; 2 Duke University Medical Center, Durham, NC, USA Biomedical Center, St. Petersburg, Russia Background: To improve the potency, breadth, and durabili- Objectives: The aim was to conduct immunological pre-clin- ty of neutralizing antibody responses against HIV-1, we AIDS Vaccine 2006 161 Poster sessions_revGJ 14/8/06 16:34 Page 162 Amsterdam, Netherlands 29 August–1 September 2006 employed prime-boost immunization strategies using 1 Hawaii HIV Vaccine Research Laboratory, Department of recombinant αvirus particle-based envelope (Env) vaccines as Medicine, University of Hawaii at Manoa, Honolulu, HI, USA; 2 primes and adjuvant Env proteins as boosts. The αvirus plat- Clinical Sciences Division, University of Toronto, Toronto, form was the previously described chimeric VEErep/SINenv Canada; 3 Department of Laboratory Haematology, Princess replicon vector derived from Venezuelan equine encephalitis Margaret Hospital/University Health Network, University of virus (VEE) and Sinbis virus (SIN) (Perri et al. J Virol 2003; Toronto, Toronto, Canada. 4 Department of Microbiology, Sanofi- 77:10348–10356.) Pastuer, Toronto, Canada; 5 Laboratory of Viral Diseases, National Objectives: To produce and evaluate in rabbit immunogenic- Institute of Allergy and Infectious Diseases, National Institutes ity studies the use of VEErep/SINenv α particles expressing envelopes from multiple subtypes and strains. To further of Health, Bethesda, MD, USA; 6 St. Michael’s Hospital, engineer the encoded env genes for optimal expression, University of Toronto, Toronto, Canada αvirus production, and immunogenicity. Methods: Envelope gene variants from subtype B SF162 and Background: Poxviruses have been the most intensively studied subtype C TV1 strains were designed and constructed. The live recombinant vectors for vaccine development, and numer- envs were cloned into the chimeric VEErep/SINenv αvirus ous studies have demonstrated their ability to induce humoral replicon vector. Expression levels were compared for the α and cellular immunity against virus infections. However, little particle preparations expressing SF162 or TV1 derived information available regarding the range of infective tropism gp160, gp140TM (transmembrance domain), gp140dV2, these virus vectors have in primary human cells. and gp140 (secreted form) with or without addition modifi- Objective: To study the tropisms of two poxviruses, ALVAC cations to enhance expression. A prime-boost study was con- and vaccinia, in primary human cells of peripheral blood and ducted with SIN/VEE-gp140dV2SF162 in rabbits and dosing bone marrow. was compared. The prime-boost regimens included DNA-α, Methods: Side-by-side comparison of ALVAC and vaccinia α-protein, DNA-protein, α alone and protein alone. In addi- virus tropisms for cells from human peripheral blood and tion, VEErep/VEEenv-gp160 form was compared to bone marrow was performed using recombinant EGFP- VEErep/SINenv-gp140dV2 in mice. expressing ALVAC or vaccinia virus and flow cytometric Results: The prime-boost study in rabbits showed that at analysis. 2wp3, the α prime-protein boost group had significantly Results: Both ALVAC and vaccinia showed a strong bias higher neutralization against SF162 than the protein or α towards monocyte infection. ALVAC minimally infected B alone groups. At 2wp4, the α-protein group showed signifi- lymphocytes and was unable to infect ex vivo NK cells and T cantly higher neutralization than the DNA-α group. We also lymphocytes, whereas vaccinia could infect B lymphocyte and saw an increased response with a 1e8 dose as compared to a NK cell populations. Vaccinia virus was also able to infect T 1e7 dose in rabbits. The expression of α particles with differ- lymphocytes at low, but detectable levels which could be ent forms of envelopes of the same strain had similar titers. enhanced upon their activation. Moreover, the level of CD14 Evaluations of the breadth of the neutralizing antibody expression on monocytes correlated with their preference to responses are ongoing. be infected with ALVAC or vaccinia virus. Both ALVAC and Conclusions: Chimeric VEErep/SINenv replicon particles are vaccinia could infect immature monocyte derived dendritic able to efficiently prime and boost for substantial neutraliz- cells (MDDCs), but only ALVAC infection induced their sub- ing antibody responses in rabbits. Modifications can be intro- sequent maturation. Infection in human bone marrow cul- duced that improve env expression and immune responses. tures showed that ALVAC infection was restricted to a Envelopes from different strains affect the production yields myelomonocytoid cell-specific CD33+ cell population, while of α particles. Finally, we plan to use combinations of α par- vaccinia virus showed a strong, but not exclusive, preference ticles expressing envelopes of different strains for further for these cells. immunogenicity studies in order to increase the breadth of Discussion: The different tropisms of ALVAC and vaccinia the induced immune responses. may explain in part the reduced reactogenicity of ALVAC when used as a vaccine in human trials. ALVAC preferential- ly infected monocytes which tend to have a shorter life span P09B-06 and barely infected long-lived B lymphocytes and did not infect T and NK cells. In contrast, vaccinia infected a larger spectrum of cell types, and in particular has been recently Different tropisms of canarypox (ALVAC) and vac- shown to maintain a productive infection in activated human cinia viruses are comparable with their reacto- T cells. As T cells can be long lived, it is possible that this genicity profiles in human vaccine trials combined feature of greater tropism and productive infection in a subpopulation of T cells may propagate infection away Q Yu1,2, B Jones2, N Hu1, H Chang3, S Ahmad2, J Liu2, M from local sites of inoculation and cause greater reactogenic- Parrington4, J Yewdell5 and M Ostrowski2,6 ity. 162 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 163 Amsterdam, Netherlands 29 August–1 September 2006 P09B-07 P09B-08 Molecular characterization and immunogenicity of Immunization with vaccinia virus induces polyfunc- a novel DNA-launched VEE-based replicon DNA tional and phenotypically distinctive CD8+ T cell vaccine responses K Ljungberg, J Thompson, A Whitmore, M Pressley, ML Precopio1, MR Betts2, J Parrino1, DA Price1,3, E Gostick3, T Moran, C Beard and R Johnston R Bailer1, BS Graham1, M Roederer1, and RA Koup1 Carolina Vaccine Institute, University of North Carolina, Chapel 1 Vaccine Research Center, NIH, Bethesda, MD, USA; Hill, NC, USA 2 Department of Microbiology, University of Pennsylvania, Philadelphia, PA, USA; 3 Weatherall Institute of Molecular Background: Genetic immunization has shown great promise Medicine, University of Oxford, Oxford, UK in small animal models, but fails to elicit robust immune responses in clinical trials. Effective dosage is considered one major impediment to efficient immunization. DNA-launched This poster is now oral abstract: OA08-03 RNA viral replicons based on αviruses express larger amounts of protein per plasmid transfected in vitro and may Background: Vaccinia virus immunization provides protec- also trigger immune enhancing innate mechanisms, and thus tion against variola virus, the causative agent of smallpox, offers a means to increase the potency of DNA vaccines. and stands as the classic example of a successful vaccine. Here, we describe a DNA-launched αvirus replicon based on While vaccinia virus-induced antibody responses have been Venezuelan Equine Encephalitis Virus (VEE-DNA). shown to be necessary and sufficient for protection against Objectives: To molecularly characterize the effects of the monkeypox virus, vaccinia virus-specific T cell responses gen- VEE-DNA expressing the HIV-1 gp160 gene on cells in vitro, erated at the time of vaccination may contribute to effective and to evaluate its immunogenicity in mice. protection. Virus-specific CD8+ T cells contribute to viral Methods: Standard molecular biology techniques were used control by directly killing virus-infected cells, secreting antivi- for cloning and expression analysis. FACS analysis of apop- ral factors, and secreting factors that recruit other cells of the totic cells was performed using PE-labelled Annexin-V. immune system. While virus-specific CD8+ T cells are often Immune responses were measured by antibody and IgG sub- measured by a limited number of parameters, such as IFNγ class ELISAs as well as IFN-γ ELISpot analysis. and/or IL-2 secretion, the functional profile of T cells is cer- Results: The VEE-DNA replicon was confirmed by sequenc- tainly more diverse, and the combination of functions that ing. Both a reporter gene (GFP) and gp160 were expressed, confer protection from infection are uncertain. and GFP-expression was consistently 10-15 fold higher than Objectives: To characterize the functional and phenotypic from a standard CMV-driven DNA vaccine plasmid (pVAX- profile of vaccinia virus-specific CD8+ T cells in a compara- 1 GFP). In preliminary studies, the VEE replicon DNA tive vaccine trial of modified vaccinia virus Ankara (MVA) expressing GFP was shown to induce apoptosis in vitro, versus. Dryvax® immunization. which was not the case after transfection of cells with the Methods: We analysed vaccinia virus-specific CD8+ T cells in pVAX-1 GFP vector. These features combined, and potential- vaccinia virus-naive individuals enrolled in a vaccine protocol ly other yet to be determined vector characteristics, con- designed to test whether pre-immunization with modified tributed to increased humoral responses to gp160. The vaccinia virus Ankara (MVA) results in protection from sub- response was approximately 100-fold higher and predomi- sequent challenge with the vaccine strain, Dryvax®. nantly of the IgG2a isotype compared to a standard DNA Polychromatic flow cytometry was used to characterize the vaccine expressing gp160. This result was also obtained functional and phenotypic profile of vaccinia virus-specific when considerably lower doses of DNA were used. Cellular CD8+ T cells. immune response remained similar between the groups. Results: Vaccinia virus-specific CD8+ T cells induced by both Conclusions: The VEE-DNA replicon expresses high levels of MVA and Dryvax® were highly polyfunctional; they degran- protein. Furthermore, it induces apoptosis in vitro and pro- ulated and produced IFNγ, IL-2, MIP1β, and TNFα after motes significantly higher Th1 type antibody responses while antigenic stimulation. While one immunization with Dryvax® maintaining cellular immune responses to gp160 in mice. induced measurable responses, more than one dose of MVA Importantly, it is effective at considerably lower doses as was required for detection of vaccinia virus-specific CD8+ T compared to a state of the art DNA vaccine vector encoding cells. Responding CD8+ T cells also exhibited an unusual phe- the same antigen. Taken together, these vector characteristics notype (CD45RO–CD27intermediate). may have significant clinical implications for the feasibility of Conclusions: The unique phenotype and high degree of poly- DNA based HIV vaccines. functionality induced by MVA or Dryvax® may be at least partially responsible for the profound efficacy of these vac- cines in protection against smallpox, and serve as a bench- mark against which other vaccines can be evaluated. AIDS Vaccine 2006 163 Poster sessions_revGJ 14/8/06 16:34 Page 164 Amsterdam, Netherlands 29 August–1 September 2006 P09B-09 P09B-10 Systemic and mucosal humoral response induced Ability of herpes simplex virus vectors to boost by HIV-1 p24 protein adsorbed onto poly(L-lactic immune responses to DNA vectors and to protect acid) nanoparticles against challenge by simian immunodeficiency virus 1 1 2 2 S Munier , Y Ataman-Önal , MH Charles , T Delair and A Kaur1*, HB Sanford1, D Garry1, S Lang1, SA. Klumpp1, 1 B Verrier D Watanabe2, RT Bronson3, JD Lifson4, M Rosati5, GN. Pavlakis5, BK. Felber6, DM Knipe2 and RC Desrosiers1 1 IFR 128 Biosciences Lyon-Gerland, INSERM U503, Lyon cedex, France; 2 IFR 128 Biosciences Lyon-Gerland, Unité Mixte CNRS- 1 New England Primate Research Center, Harvard Medical bioMérieux, ENS-Lyon, Lyon cedex, France School, Southborough, MA; 2 Dept of Microbiology & Molecular Genetics, Harvard Medical School, Boston, MA; 3 Harvard Background: Biocompatible and biodegradable poly(L-lactic Medical School, Rodent Histopathology Core, Boston, MA; 4 acid) nanoparticles are potent adjuvant to improve the effec- AIDS Vaccine Program, SAIC Frederick Inc., National Cancer tiveness of anti-HIV vaccine. PLA nanoparticles are versatile Institute, Frederick, MD; 5 Human Retrovirus Section, Vaccine carrier able to deliver antigen of different nature, encapsulated Branch, National Cancer Institute, Frederick, MD; 6 Human or adsorbed on their surface, without inducing anti-vector Retrovirus Pathogenesis Section, Vaccine Branch, National immune response. Cancer Institute, Frederick, MD, USA Objectives: To evaluate the ability of anionic surfactant-free PLA nanoparticles as protein adjuvant to induce systemic and mucosal antibodies. Objectives: The immunogenicity and protective capacity of Methods: PLA nanoparticles were prepared using the solvent recombinant, replication-defective herpes simplex virus diffusion method, a one-step and surfactant-free process. (HSV) vector-based vaccines with or without DNA priming Adsorption of HIV-1 p24 protein onto PLA nanoparticles was were examined in rhesus macaques. investigated as a function of the nanoparticle size and stability Methods: Three rhesus macaques were inoculated with a of the formulations was controlled. The immunogenicity of mixture of replication-defective HSV vectors expressing Gag, PLA-p24 complex was evaluated in BALB/c mice model in com- Env, and a Tat-Rev-Nef fusion protein of simian parison with Alum and Freund’s adjuvants. Then, dose effect immunodeficiency virus (SIV) at weeks 0, 4, 12 and 20. Three regarding p24 and nanoparticles quantities was determined and other rhesus macaques were inoculated with a mixture of we deduced the limit-dose of protein required for inducing high recombinant DNAs expressing Gag, Env, and a Pol-Tat-Nef- antibody responses. Finally, two routes of administration were Vif fusion protein at weeks 0 and 4 and boosted with the tested (subcutaneous and intranasal) and we measured both the same mixture of HSV vectors at weeks 12 and 20. At week induced systemic and mucosal humoral responses. 26, all vaccinated macaques were challenged intravenously Results: HIV-1 antigens (p24, Tat, gp120) were efficiently with wild-type, cloned SIVmac239. adsorbed onto anionic PLA nanoparticles at high yield, always Results: Anti-SIV antibody responses and neutralizing activi- beyond 50%. Moreover, the synthesis process allows the con- ty in all six vaccinated macaques were weak. Cellular trol of nanoparticle diameter between 200 nm to 1 µm. Protein responses to SIV were detected in all six macaques and were adsorption step was a function of nanoparticle size. By using the significantly higher in the DNA-primed animals. IFN-γ same solid content, increasing amount of protein could be ELISPOT responses to Gag and Env exceeded 1000 spot 6 adsorbed by diminishing the nanoparticle size. IgG antibody forming cells per 10 PBMC in the DNA-primed animals titer measured after three subcutaneous immunization with after recombinant HSV boosting. All six vaccinated PLA-p24 (1.6×106), was as high as Freund’s adjuvant (1.6×106) macaques became infected following SIVmac239 challenge. and higher than Alum (6.5×105). Moreover, antibodies titers However, the viral load at peak and 12 weeks post challenge increased with the protein and nanoparticles amounts, and 0.1 was significantly lower in the vaccinated macaques compared µg/injection/mouse was the limit-dose of p24 required to main- to 31 historical control macaques that had been inoculated tain high antibody production. Finally, we evaluated the mucos- with the same stock of SIVmac239. Two of the HSV-HSV- al immune response induced by PLA-p24 formulation after HSV-HSV vaccinated macaques and one DNA-DNA-HSV- subcutaneous injection or intranasal administrations, in com- HSV vaccinated monkey continued to maintain low viral parison with MVA (modified vaccinia virus Ankara) described loads at 24 weeks post challenge. The partially-protected as a good mucosal vaccine. We observed that PLA-p24 injected macaques had higher neutralizing activity and anti-Rev subcutaneously was the best route to induce higher systemic ELISPOT activity on the day of challenge, and a more rapid and mucosal humoral response. ELISPOT response to Tat and Gag following challenge. Conclusions: Poly(L-lactic acid) nanoparticles represent a Conclusions: The induction of robust cellular immune promising method for in vivo delivery protein, able to induce responses and the partial protection conferred against a path- both systemic and mucosal immune responses. Moreover, sev- ogenic SIV challenge support further study of recombinant eral proteins can be co-adsorbed and permitting the addition of herpesviruses as a promising vaccine approach for AIDS. immunostimulatory molecules to recruit dendritic cells near the injection sites. 164 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 165 Amsterdam, Netherlands 29 August–1 September 2006 P09B-11 P09B-12 Priming with a rBCG or DNA vaccine expressing The impact of priming on memory of clinical HIV- HIV-I subtype C Gag elicits comparable CD8+ T 1 clade C NYVAC based vaccine candidates in cell responses when boosted by a rMVA express- Rhesus macaques. ing GagC P Mooij1, S Balla-Jaghjoorsingh1, N Beenhakker1, P van AL Williamson2,3, E Shephard1, W Bourn2, H Stutz2, Haaften1, I Baak1, I Nieuwenhuis1, A Harari2, N Johnston2, J van Harmelen2, C Williamson2, G Pantaleo2, M-J Frachette3, S Heidari5, H Wolf4, N Douglass2, D Bowers2, S Galant2, Z Isaacs2, and R Wagner4, P Liljestrom5, and JL Heeney1 A Binder2 1 Biomedical Primate Research Center, Rijswijk, The Netherlands; 1 MRC Liver Research Centre, University of the Cape Town; 2 2 Centre Hospitalier Universitaire Vaudois, Lausanne, Institute of Infectious Disease and Molecular Medicine; Switzerland; 3 Sanofi Pasteur, Marcy l’Etoile, France; 4 Institut University of the Cape Town; 3 NHLS, Cape Town, South Africa für Medizinische Mikrobiologie und Hygiene der Universität Regensburg, Regensburg, Germany; 5 Karolinska Institutet, Background: DNA vaccines expressing HIV genes are known Stockholm, Sweden to prime immune responses that can be boosted with recom- binant modified vaccinia Ankara (rMVA). However, DNA Methods: The ability of each of these immunization regi- vaccines require high concentrations to be effective. A rBCG ments to elicit peptide-specific responses to each of the four vaccine expressing HIV genes may be an alternative approach encoded vaccine antigens was evaluated by three different to prime the immune system. cytokine (IFN-γ, IL-2 and IL-4) ELISpot assays in three Objective: To compare the immune response to a rBCG vac- groups of 10 outbred rhesus macaques. cine rBCG[pF10-Hi], constitutively expressing GagC, with Results: In the absence of priming IFN-γ responses were very that induced by a DNA vaccine pTHGagC and determine the low and detected in only a few animals, whereas a greater fre- effect of a boost with MVAGagC. quency of responders were detected by enumerating the anti- Methods: Mice were inoculated with rBCG[pF10-Hi] gen specific IL-2 or IL-4 secreting T-cells. In contrast, priming 7 (10 pfu, i.p.) or pTHgagC (100 ug, i.m.). On day 56, half the by either DNA or Alpha virus resulted in a dramatic increase mice in each group were sacrificed and the other half boost- in the magnitude of responses and the number of responders 7 + ed with MVAGagC (10 pfu) then sacrificed on day 78. CD8 to individual vaccine antigens. Of interest was the difference + and CD4 T cell responses were measured in the spleen using in the immune responses induced by the two different prim- an IFN-γ ELISPOT assay. Groups of mice were challenged ing regiments. DNA followed by NYVAC boosting gave a 6 with recombinant vaccinia expressing GagC (VVGagC, 10 predominant IFN-γ response which was predominantly to pfu) 2 weeks after an inoculation with either rBCG[pF10-Hi] Env, whereas Alpha virus priming gave more heterologous or pTHgagC to evaluate the protective capacity of the gag- responses with statistically greater Gag and Nef responses by + specific CD8 T cells. several different cytokines. + + Results: rBCG[F10-Hi] induced Gag-specific CD8 and CD4 Conclusions: In the absence of priming, responses declined T cell responses with a cumulative response of 370 net spot often below detection whereas those “boosted” by NYVAC forming units/10e6 splenocytes. The response to pTHgagC were high and persistent. Long-term, follow-up of prime- was 410 net spot forming units/10e6 splenocytes. The Gag boost groups revealed a shift in frequency of peptide specific + CD8 T cell response elicited by these vaccines provided com- CD4 as well as CD8 T-cell memory responses to Pol as well parative protection from a challenge with VVGagC of as Nef and were detected more than a year following immu- approximately 3 logs reduction in titre compared to controls. nization. The same clinical lots and immunization schedules + + MVAGagC boosted the cumulative CD8 and CD4 T cell have now advanced to human clinical trials. primary responses induced by rBCG[pF10-Hi] to 1729 net SFU/106 splenocytes. In comparison MVAGagC boosted the primary response to pTHgagC to 1534 cumulative net SFU/106 splenocytes. Conclusion: Comparable responses to Gag were achieved by either a rBCG or a DNA vaccine and the Gag-specific CD8+ T cells protected mice from a VVGagC challenge. In addition comparable T cell responses were achieved with rMVAGagC. These results suggest rBCG expressing HIV genes may be a satisfactory alternative to DNA vaccines in heterologous prime boost regimens. AIDS Vaccine 2006 165 Poster sessions_revGJ 14/8/06 16:34 Page 166 Amsterdam, Netherlands 29 August–1 September 2006 P09B-13 P09B-14 Evaluation of GMP grade HIV-1 clade C CN54 Equine herpesvirus (EHV) type 1 based viral vec- DNA candidate vaccines for human Phase I trials tors effectively transduce and activate monocyte in a Balb/c mouse model derived dendritic cells (MDDC) J Wild, K Bieler, H Wolf and R Wagner J Köstler1, H Hofmann1, K Boeckl1, K Tischer2, N Osterrieder3, J Wild1 and R Wagner1 Institute of Medical Microbiology, University of Regensburg, Regensburg, Germany 1 Institute for Medical Microbiology and Hygiene, University of Regensburg, Regensburg, Germany; 2 Institute for Medical Background: As shown previously, immunization with DNA Microbiology and Virology, University Medical Center Schleswig- vaccines containing RNA- and codon-usage optimized HIV-1 Holstein, Kiel, Germany; 3 Department of Microbiology and clade B genes induces strong and specific Th1-type humoral Immunology, College of Veterinary Medicine, NY, USA and cellular immune responses that contribute to the main- tainance of stable CD4 cell counts, a containment of virus Objectives: Recently we have proven in a Balb/c mouse model replication and protection from disease progression in rhesus the general capacity of EHV-1 derived vectors to inducing macaques following virus challenge. strong cellular, humoral and mucosal responses to HIV Objectives: These results justified extensions to human trials immunogens. The primary goal of the current study is to with analogous vaccine constructs derived from a clade C compare a clinical trial lot of a recombinant New-York- molecular clone (CN54) representing the prevalent virus cir- Vaccinia Virus based HIV vaccine candidate (NYVAC-C; culating throughout China. As a prerequisite, we aimed at expressing Gag/Pol/Nef) with a corresponding EHV-based evaluating the immunogenicity of the CN54 based DNA and vaccine construct (EHV-C) regarding their capacity to induce NYVac candidate vaccines. maturation and activation of monocyte derived dendritic cells Methods: In preparation of phase I trials the immunogenicity (MDDC). of the candidate DNA vaccines (DNA-C) as well as a replica- Methods: A recombinant EHV-1 C-Gag Pol Nef (EHV-C) was tion deficient vaccinia virus (NYVAC-C) encoding codon-opti- generated using BAC-technology and RED-Recombination. mized GagPolNef (GPN) and Env polygenes was evaluated in Viral mutants were verified by Immunofluorescence, Western- a Balb/c mouse model by using preGMP grade material. and Southern-Blot-analysis. MDDCs were infected with EHV- Results: Vaccination studies were performed comparing DNA- C and NYVAC-C at different MOIs. Expression of the trans- C versus. NYVac-C versus DNA-C/NYVac-C prime/boost pro- genes was monitored by FACS and Western Blot analysis. tocols. DNA-C constructs as well as high doses of NYVac-C Maturation of MDDCs was determined by FACS analysis of 7 (10 pfu) were able to induce strong cellular immune respons- differentiation markers (CD80, CD83, CD86, HLA-DR, es measured by ELISpot and FACS analysis using pools of CD40, CCR7) and in ELISA assay measuring secreted proin- overlapping peptides as well as significant Env-antibody titers. flammatory cytokine levels (IL-6, IL-10, IL-12, TNF-α) In contrast, lower doses of virus delivered only poor cellular Results: Depending on the used MOI a substantial fraction responses and failed to induce antibodies. The most potent cel- (∼40%) of EHV-C infected MDDCs displayed expression of lular immune responses were induced using DNA-C/NYVac-C significant amounts of GagPolNef immunogens. MDDCs as well as NYVac-C/DNA-C prime/boost protocols, also sup- infected with EHV-1 show various markers for DC matura- plying high Env-specific antibody titers. Besides, H-2d restrict- tion and activation as monitored by release of cytokines and ed CTL epitopes could be identified and mapped in clade-C Pol surface expression of costimulatory signals. In contrast, and Env by ELISpot analysis facilitating further immunization MDDCs infected with NYVAC-C only weakly express HIV studies in Balb/c mice. transgenes and do, upon direct infection, not support DC Conclusions: These results are underlining that prime/boost maturation or activation. protocols are the most promising HIV vaccination strategies. Conclusion: EHV derived vectors support efficient transgene A phase 1 clinical trial coordinated by the EUROVACC net- expression and provide signals required for DC maturation work was already able to confirm safety, tolerability and and activation. immunogenicity of the NYVac-C vaccine in HIV- negative volunteers. A DNA/NYVac clinical Phase I study assessing the efficacy of prime boost protocols to induce HIV-specific T-cells is already under evaluation. 166 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 167 Amsterdam, Netherlands 29 August–1 September 2006 P09B-15 Background: Measles virus (MV) vaccines are safe, efficient and induce long-lived immunity after a single vaccination dose. Accelerated and robust immune responses induced The Edmonston Zagreb (EZ) strain is a well-tolerated vaccine, by EuroVacc DNA HIV-1 vaccine immunogens administered both parenterally and mucosally, a feature partic- ularly important for the development as a vaccine vector. delivered by a novel intradermal DNA delivery Objective: To establish a robust multivalent vaccine vector method based on the safe measles virus Edmonston Zagreb vaccine. Methods: The cDNA corresponding to the commercial vac- 1 2 2 2 A Bins , B Verstrepen , C Rollier , P Mooij , cine, Berna Biotech EZ vaccine strain, MVEZbv, was cloned G Koopman2, R Wagner3, H Wolf 3, T Schumacher1, to construct the plasmid p(+)MVEZbc. A helper-cell-based J Heeney2 and J Haanen1 rescue system allowed the generation of a cloned virus (MVEZbc) from cDNA. Immunogenicity of the cloned vac- 1 Dept of Immunology, The Netherlands Cancer Institute, cine was tested in transgenice mice susceptible to MV. Amsterdam; 2 Dept of Virology, Biomedical Primate Research Multiple cloning sites were engineered between P and M Centre, Rijswijk, The Netherlands; 3 Institute of Medical genes and between H and L genes to insert single or multiple Microbiology and Hygiene, Regensburg, Germany HIV-1 transgenes into the rMVEZbc genome, and the resul- tant rMVEZbc-HIV virus was used for immunization. Results: The nucleotide sequence of the cloned vaccine was Background: Immune responses induced by DNA vaccina- identical to the master seed of the vaccine. Comparison of tion are generally considered slow and weak. Recently, we sequences of various MV vaccines, including the published reported that short-interval intradermal DNA delivery using EZ sequence and MVEZbc/MVEZbv indicate that the a tattoo device, generates robust T cell responses in mice MVEZ-bc/v diverged from the original EZ. In vitro and in within 12 days. vivo analysis of MVEZ-bc showed identical growth kinetics Objectives: Here we demonstrate proof of principle data that to the parental MVEZbv as well as to the commercial MVEZ short-interval tattoo DNA vaccination strategy is a potent (Serum Institute of India, SII). In vivo analysis of MVEZbc in immunization method in outbred primates as well. transgenic mice susceptible to MV revealed comparable Methods: Using the EuroVacc clade C gag-pol-nef + env humoral immune response to both commercially available DNA vaccine, 5 Rhesus monkeys (Maccaca mulatta) were vaccines, MVEZbv and that of SII. MVEZbc induced immunized at week 0 and week 4, each vaccination consist- significantly higher levels of MV antibodies when com- ing of a sequel of 3 tattoos at 3 day intervals. pared with an over-attenuated MV laboratory strain. Results: Blood was drawn from the animals at 4 and 8 weeks Immunohistochemical analysis confirmed the ability of after the start of vaccination. PBMC’s were isolated and MVEZbc vector to stably express the inserted HIV-1 analysed by ELIspot. In the ELIspot assay, T assay cells were transgenes. stimulated with 14-mer peptide pools covering all translation Conclusion: The results indicate that the cloned MVEZbc products of the vaccine with 11 amino-acid overlap. By 4 strain can be used to develop MVEZ-based recombinant vac- weeks, all animals had mounted robust IFN-γ responses. By 8 cines against HIV-1 (see poster: Liniger et al). weeks after tattoo vaccination, the average number of HIV-1 specific lymphocytes, producing either IFN-γ, IL-2 or IL-4, was 10 to 100 fold higher than the number of Ag specific T- cells enumerated after IM administration. The intramusclar P09B-17 dataset has been based on 25 Rhesus monkeys, previously injected with the same vaccine, in the same regimen, and Electroporation of DNA vaccines encoding human analysed using the same reagents and readouts. immunodeficiency virus (HIV-1) antigens in small Conclusions: These preliminary results urge further explo- animals and nonhuman primates elicits potent ration of this novel intradermal DNA vaccination delivery humoral and cellular immune responses strategy in different pre-clinical prime-boost vaccination sched- ules and early application in upcoming human clinical trials. R Pal1, A Cristillo1, D Weiss1, A Khan2, L Hocker1, L Hudacik1, L He1, J Suschak1, S Restrepo1, 2 1 P09B-16 R Draghia-Akli , P Markham 1 Advanced BioScience Laboratories Inc., Kensington, USA; 2 A novel measles virus Edmonston Zagreb vector is ADViSYS Inc, Houston, USA highly immunogenic in laboratory animals Background: DNA vaccines delivered by intramuscular and A Zuniga, M Liniger, S Weibel and HY Naim intradermal routes have been shown to be a safe and effective approach to prime humoral and cellular immune responses in Berna Biotech LTD a Crucell company and Etnavax, Bern, vaccinated subjects. However, the levels of immune respons- Switzerland es elicited by these routes are usually weak. Moreover, clear- ance of plasmids from the injection site is often slow due to AIDS Vaccine 2006 167 Poster sessions_revGJ 14/8/06 16:34 Page 168 Amsterdam, Netherlands 29 August–1 September 2006 the high dose of plasmid DNA used for immunizations. gen to dendritic cells. The results of this trial will inform the Objectives: To evaluate the humoral and cellular immune optimization of a potential vaccine immunogen. The model responses elicited by an HIV-1 DNA vaccine delivered by immunogen is a trimeric HIV-1 envelope glycoprotein, based electroporation in mice, rabbits and nonhuman primates. on the African isolate 96ZM651-8 (“ZM96gp140”). Methods: BALB/c mice, New Zealand White rabbits and Objectives: To provide highly-purified GMP and non-GMP cynomolgus macaques were immunized with plasmids ZM96gp140 to for all aspects of the above project. encoding HIV-1 Env and Gag antigens in saline using a Methods: Following the establishment of a master cell bank novel electroporation technology. DNA-primed mice were (recombinant cell line CHO K1 pEE14/tpa/ZM96gp140REKS also boosted with a single immunization of HIV-1 gp120 #13#47#03#46) rigorous fermentation, purification and char- and Gag protein formulated in QS-21 adjuvant. HIV-1 Env acterization methodologies have been developed to provide sta- binding antibodies were assayed by ELISA while Env- and ble, trimeric gp140 to defined specifications. Gag-specific cellular immune responses were measured by Results: The ZM96gp140 master cell bank successfully fulfils IFNγ ELISPOT and Cytometric Bead array assays following all criteria for stability, identity and freedom from adventi- DNA and protein administrations. tious agents. Fermentation strategies for GMP production Results: Electroporation of Env and Gag DNA (5 µg each) have been optimized to overcome the high growth rate of the followed by a single gp120 and Gag protein boost (15 µg cell line in suspension culture and its relatively low specific each) by intramuscular delivery elicited strong anti-Env productivity. A peak titre of 5 mg/l was achieved. For non- antibody and CMI responses in BALB/c mice. GMP production, adherent cell lines were cultured in 850 Electroporation of rabbits with DNA (200 µg) encoding cm2 roller bottles in serum-supplemented medium, reaching a HIV-1 Env elicited a strong anti-Env antibody response peak titre of 3 mg/l. Trimeric ZM96gp140 was immunoaffin- after each DNA immunization. Antibody levels were sig- ity purified with the gp41 antibody 5F3, with subsequent pol- nificantly higher using electroporation as a delivery ishing steps by size-exclusion and/or ion-exchange method as compared to the levels observed following intra- chromatography. Purified gp140 was found to meet the muscular administration (1 mg). Electroporation of Env defined product specificity with respect to appearance, con- and Gag DNA (250 µg each) into cynomolgus macaques centration, purity, oligomericity and endotoxin levels. elicited strong antibody and CMI responses that were Antibody, CD4 and chemokine receptor binding assays indi- found to be significantly greater than that observed cate that the gp140 is correctly folded and raises high titre following intramuscular administration of DNA. antibodies in mice. Conclusions: Delivery of low doses of DNA by a novel elec- Conclusions: We have developed fermentation and purifica- troporation immunization represents an effective strategy to tion methodologies to provide trimeric C-clade gp140 for a elicit robust humoral and cellular responses in animal mod- Phase I clinical trial, to support the development of con- els. Use of significantly lower doses of DNA reduces the trolled release technologies and to use as the template for a amount of immunogen requirement and may induce faster vaccine immunogen. clearance of plasmids from sites of injection. Further studies are needed to evaluate functional characteristics of the immune response elicited by this DNA electroporation deliv- P09B-19 ery in combination with other strategies. Baculovirus-expressed HIV-1 virus-like particles (HIV-VLP) activate dendritic cells and induce ex P09B-18 vivo T cell response Production of C-clade HIV-1 gp140 for a novel L Buonaguro1,2, ML Tornesello1, GK Lewis2,3, E Aricò4, cervico-vaginal delivery system F Marincola4 and FM Buonaguro1 1 1 2 3 SA Jeffs , S Abdul Hussein , R Shattock and D Katinger 1 Viral Oncogenesis and Immunotherapy & AIDS Refer. Center, Ist. Naz. Tumori “Fond. G. Pascale”, Naples – Italy; 2 Division of 1 Section of Infectious Diseases, Division of Medicine, Imperial Vaccine Research, Institute of Human Virology, University of College, London; 2 Division of Infectious Diseases, St George’s Maryland Biotechnology Institute; 3 Department of Hospital Medical School, London; 3 Polymun Scientific, Vienna, Microbiology and Immunology, University of Maryland School of Austria Medicine, University of Maryland Baltimore, Baltimore, Maryland; 4 Immunogenetics Laboratory, Department of Background: To protect against HIV, sterilizing immunity at Transfusion Medicine, NIH, Bethesda, MD, USA the point of viral entry is required. We believe that the induc- tion of sustained vaginal mucosal immunity may be best pro- Aim: A candidate HIV-1 vaccine model based on HIV-1 vided by repeated and/or sustained vaginal mucosal Pr55gag Virus-Like Particles (HIV-VLPs), produced in a bac- immunization using user-enabled, controlled-release tech- ulovirus expression system and presenting a gp120 molecule nologies. To examine this, we will use a number of delivery from an Ugandan HIV-1 isolate of the clade A (HIV-VLP s) systems in a Phase I clinical trial to target a model immuno- A has been developed in our laboratory. The present study has 168 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 169 Amsterdam, Netherlands 29 August–1 September 2006 been performed to evaluate the ability of the HIV-VLP to promoter. To enhance encapsidation of the genomic RNA A induce the maturation and activation of monocyte-derived by the VEE-SIV Gag particle, the SIV psi site (packaging sig- dendritic cells (MDDCs), and to analyse the pathway involved. nal) was engineered into the VEE-SIV Gag sequence in the Results: The baculovirus-expressed HIV-VLPs efficiently 5′ UTR or upstream of the 26S promoter. Also, the VEE C induce maturation and activation of MDDCs and are incor- fragment (75–138) (responsible for binding genomic RNA) porated into MDDCs preferentially via an actin-dependent was incorporated into the vaccine to produce a variety of macropinocytosis and endocytosis. The HIV-VLP-activated chimeric SIV Gag-VEE C self-assembling proteins which MDDCs show an enhanced Th1- and Th2-specific cytokine bind genomic RNA. production, in particular, IL-12 p70, IL-6, IL-10 and TNFα. Results: The VEE-SIV Gag vaccine has been modified to con- The effects of VLPs on MDDCs are, to some extent, mediat- tain αviral and lentiviral encapsidation enhancement ed through intra-cellular Toll-like receptors (TLR 3-5-7-9) sequences and evaluated for particle formation, specific bind- signalling. The HIV-VLP-loaded MDDCs are able to induce a ing and encapsidation of genomic RNA. All of the chimeric primary and secondary response in autologous human CD4+ vaccines express SIV Gag and produce particles except VEE- T cells, in an ex vivo immunization assay. Moreover, the 5′UTRpsi-SIV Gag. Chimeric VEE-SIV Gag particles were genomic transcriptional profile of VLP-activated DCs, by purified, normalized to Gag p56 content and evaluated for gene microarray analysis, show the upregulation of several specific encapsidation of genomic RNA. Results from these genes involved in the immune response. studies indicate that addition of EES to the VEE-SIV Gag vac- Conclusions: Our results show that baculovirus VLPs acti- cine enhances the specific encapsidation of genomic RNA by vate MDDCs and stimulate the production of cytokines the chimeric particles. involved in the Th-1 and Th-2 pathways, elucidating the Discussion: Encapsidation enhancement sequences from mechanisms triggering the cell-mediated immunity observed SIV or VEE have been engineered into the VEE-SIV Gag in VLP-immunized animals. The intra-cellular Toll-like recep- vaccine to enhance specific packaging of genomic RNA by tors appear to be involved in this process and additional sig- the chimeric particles. Enhancement of RNA encapsidation nalling pathways induced by VLPs in the MDDCs are by the chimeric particles should lead to improved propaga- currently under evaluation by gene profiling analysis. These tion and ultimately the immunogenicity of the VEE-HIV data give an insight into the mechanisms of the cellular propagating chimeric particle vaccine. immunity induced in vivo by VLPs, which may be extremely useful to optimize and modulate the immune response. P09B-21 P09B-20 Immunogenicity of ANRS LIPO-5 alone, Sanofi- Pasteur ALVAC-HIV(vCP1452) alone, and ALVAC Propagation enhancement of chimeric lentiviral HIV Prime/LIPO-5 boost in healthy, HIV-1 unin- vaccines fected adult participants KR Young, CK Jurgens and RE Johnston S Frey1, L Peiperl2, L Baden10, J McElrath3, P Wright10, P Goepfert10, M Keefer10, W Blattner10, C Harro10, Carolina Vaccine Institute, University of North Carolina, Chapel S Hammer10, J Hural3, G Tomaras4, L Baglyos5, Hill, NC, USA and Global Vaccines Inc., Research Triangle Park, J-G Guillet6, P Gilbert7, M Deers7, N Russell8, M Elizaga9, NC, USA L Corey3 and the HIV Vaccine Trials Network10 Objectives: Replicating particle-based vaccines have been 1 Department of Internal Medicine, Saint Louis University, St. generated to express lentiviral Gag and Env for the induction Louis, MO, USA, 2 University of California, San Francisco, CA, of protective neutralizing antibody and cellular immunity USA, 3 FHCRC, University of Washington, Seattle, WA, USA, 4 against HIV/AIDS. Gag and Env are expressed from a VEE Duke University Medical Center, Durham, NC, USA, 5 Sanofi replicon genome and form particles that contain a non-inte- Pasteur, Swiftwater, PA, USA, 6 ANRS, Paris, France, 7 SCHARP, grating, replicating, chimeric VEE/lentiviral genome. Seattle ,WA, USA, 8 Bill & Melinda Gates Foundation, Seattle However, the propagating chimeric particle does not have a specific mechanism to preferentially package genomic RNA. WA, USA, 9 HVTN Core Operations Center, Seattle, WA, USA, and Therefore, encapsidation enhancement sequences (EES) were the 10 HIV Vaccine Trials Network introduced into the vaccine to increase the specific encapsi- dation of genomic RNA by these particles. Background: Effective HIV vaccines remain elusive. A new Methods: Encapsidation enhancement sequences from SIV approach using lipopeptides to stimulate T cell immune or VEE were engineered into a VEE-SIV Gag vaccine to responses was tested. enhance specific packaging of VEE-SIV Gag RNA. The Objective: The main objective of the study was to test the parental vaccine, VEE-SIV Gag, contains the VEE 5′ and 3′ safety and immunogenicity of LIPO-5 when given alone and UTRs along with the nonstructural genes required for repli- in prime-boost regimens with ALVAC-HIV (vCP1452). cation. The structural genes of VEE are replaced with SIV Methods: Participants were randomized into 5 vaccine gag, which is under the control of the VEE subgenomic 26S groups with matching placebos: Lipo-5 (2.5 mg) alone; AIDS Vaccine 2006 169 Poster sessions_revGJ 14/8/06 16:34 Page 170 Amsterdam, Netherlands 29 August–1 September 2006 ALVAC-HIV alone; or three groups of ALVAC-HIV followed vidual plaques. For some recombinant viruses, we found that by ALVAC-HIV plus LIPO-5 (0.25 mg, 0.75 mg or 2.5 mg). env and/or gag expression was rapidly lost in a significant Vaccine was given at months 0, 1, 3 and 6. In the prime-boost fraction of the virus population. To identify the mechanism(s) groups, ALVAC was given at months 0 and 1 followed by of loss of expression, we isolated individual plaques and ALVAC-HIV plus LIPO-5 at months 3 and 6. LIPO-5 vaccine characterized the nature of the mutations. In some cases, spe- contains a mixture of 5 lipopeptides with CTL epitopes from cific DNA sequences with propensity to mutate by addition HIV-1 Gag, Pol and Nef. ALVAC-HIV, a canarypox vector, or deletion of a single nucleotide were identified. Generation expresses Env, Gag, Pol and Nef. HIV-specific binding anti- of such mutations could be avoided by altering codons with- bodies to Gag and Env were assessed by binding ELISA at out changing the predicted translation product. In other Days 0 and 182. IFN-γ-screening T cells were assessed by cases, loss of expression was caused by large deletions that IFN-γ ELISpot using peripheral blood mononuclear cells frequently extended into flanking non-essential MVA genes. stimulated with synthetic Gag, Env, Nef, and Pol peptide To prevent this from occurring, we constructed a new shuttle pools. IFN-γ ELISpot assays were performed at Day 0, 98 plasmid designed to direct insertion of foreign genes between and 182. two essential MVA genes. Recombination into this site Results: Safety results have been previously reported (ref: reduced deletions of the foreign DNA. In one case, however, AIDS Vaccine 2005 abstract). One hundred seventy-four, the toxicity associated with high-level HIV env expression 168, 142, and 73 participants received 1, 2, 3, and 4 of the was so severe that the selection of rare mutants still resulted planned study vaccinations, respectively. The IFN-γ ELISpot in an unstable population. Only truncation of the transmem- assay response rates were low and comparable among place- brane domain of env allowed the construction of a stable bo and vaccine arms at each time-point. Response rates for recombinant MVA. placebo controls at the 3 time-points were: day 0: 2/28 Conclusions: Instability of env and gagpol inserts is attrib- (7.1%); day 98: 2/27 (7.4%); and day 182: 2/24 (8.3%). uted to the generation of point mutations and deletions and Pooled response rates across the 5 vaccine arms were: Day 0: the growth advantage of non-expressing MVA mutants. 3/134 (2.2%); day 98: 11/133 (8.3%); day 182: 6/124 Instability can generally be reduced by codon alteration (4.8%). day 182 ELISA responses to p24 (gp120) were and/or insertion into an essential region of the MVA genome absent for LIPO alone, present at rate 8/25 (6/26) for ALVAC but env had to be altered in one case. alone, and present at rate 36/76 (29/82) for the ALVAC+LIPO arms combined, with no dose effect. Conclusions: In general, there was no appreciable cell-medi- P09B-23 ated immunity detected and antibody responses were limited. Neither ALVAC-HIV (vCP1452) nor LIPO-5 alone or in Development of immunologically-enhanced HIV combination was immunogenic. subtype C vaccines from genetically modified MVA vectors P09B-22 L O’Mara1, DA Garber1, M McQuoid1, I An1, E Hunter1 and MB Feinberg1,2 Enhancing expression and stability of foreign genes in recombinant MVA vaccines 1 Emory Vaccine Center, Emory University School of Medicine, Atlanta, GA, USA; 2 Merck Vaccine Division, West Point, PA, USA L Wyatt1, P Earl1, J Americo1, C Cotter2, J Vogt1, W Xiao1 and B Moss1 Background: MVA is a safe and promising vector for vaccina- tion of humans against HIV. However, its use as a single 1 Laboratory of Viral Diseases, NIAID, NIH, Bethesda, MD, modality vaccine is currently limited by several factors, includ- 2 Henry M Jackson Foundation, Rockville, MD, USA ing MVA’s relative antigenic complexity (that limits its ability to prime responses against heterologous HIV inserts) and Background and Objectives: We have made recombinant inability to replicate in vivo (that restricts transgene expression MVAs expressing HIV-1 env and gag pol genes from many to those cells initially infected at a site of immunization). Thus, different isolates. The stability of inserted genes after repeat- strategies to enhance the immunogenicity of MVA-based AIDS ed passage in tissue culture has proven to be variable. Here vaccines through genetic modification of the vector to reduce we (1) demonstrate that the instability represents a combina- its antigenic complexity and/or abrogate its ability to express tion of spontaneous mutation or deletion of the inserted gene poxvirus proteins with predicted immune-inhibitory effects and selection for non-expressing mutants and (2) describe should result in improved vaccine vectors that elicit quantita- novel methods for reducing instability. tively and qualitatively better cellular immune responses than Methods and Results: Recombinant MVAs expressing env are otherwise achievable with current AIDS vaccines derived and gag pol from many different isolates were constructed. from the parental strain of MVA. Each virus was subjected to repeated passages in chicken Objectives: To generate genetically-modified MVA-based vac- embryo fibroblast cells to mimic the large-scale amplification cines that exhibit enhanced immunogenicity profiles, which required for production of virus for clinical trials. Insert sta- express HIV subtype C consensus antigens and can be rapid- bility was monitored by env and gag immunostaining of indi- ly advanced into human clinical trials. 170 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 171 Amsterdam, Netherlands 29 August–1 September 2006 Methods: Derive novel MVA-based AIDS vaccine vectors that MVAαudg vectors results in rapid cellular apoptosis, are deleted for specific poxvirus immune-evasion genes abortive infection by MVA∆udg in a number of non-APC cell and/or replication genes from qualified cellular and viral sub- types of avian, rodent, and human-origin atypically induces strates to enable rapid translation to cGMP manufacture and the activation of pro-apoptotic caspases as determined by human clinical evaluation. caspase-3/7-mediated cleavage of a synthetic fluorescent pep- Results: We have developed and qualified cGMP-cell banks tide substrate and/or cleavage of poly-ADP-ribose-poly- of an immortalized CEF-derived cell line that allows for the merase (PARP). Comparative analyses of pathways of growth of MVA?udg vectors as well as numerous other apoptosis induction by MVAαudg vectors and potential recombinant MVAs, as a new cellular substrate (an alterna- mechanisms of apoptosis induction using low-density gene tive to primary CEF cultures) to enable the production of arrays will be presented. The implications of immunizing clinical-grade MVA-based vaccines. Using this cell line in a with MVA vectors that atypically induce apoptosis of infect- study conducted in accordance with good laboratory prac- ed cells to enhance presentation of vector-encoded antigens tices, we have targeted genetic modifications to the (pre-BSE) through cross-presentation pathways will be discussed. MVA-1974 strain that include deletion of MVA immune eva- Conclusions: An MVA recombinant that is genetically sion genes (encoding a soluble IL1β-receptor, IL18 binding blocked for progression through the E→L phases of the protein, TIR-domain-containing protein, and vCKBP homo- poxvirus replication cycle induces apoptosis atypically in a logue), deletion of the essential udg replication gene, and number of different non-APC cell types and may enable eval- insertion to direct expression of codon-optimized HIV genes uation of the relative importance of direct- versus cross-anti- encoding antigenic consensus subtype-C Gag and Env anti- gen presentation pathways in generating immune responses gens. Phenotypes of mutant viruses, including the genetic sta- against MVA-based vaccines. Genetic complementation bility of HIV inserts will be discussed. strategies should be further pursued to rationally develop Conclusions: Deletion of relevant poxvirus genes from MVA immunologically enhanced MVA-based AIDS vaccines. vectors remains a viable approach to rationally enhance the immunogenicity of MVA-based AIDS vaccines. P09B-25 P09B-24 Differential consequences of infection of dendritic cells with vaccinia virus or its attenuated deriva- Characterization of a replication-defective MVA- tive, modified vaccinia virus Ankara based vaccine vector that induces cellular apopto- sis atypically during abortive infection DA Garber1, A Chahroudi1, P Reeves2, D Kalman2, KA Easley3, SI Staprans1 and MB Feinberg1,4 DA Garber1, M McQuoid1 and MB Feinberg1,2 1 Emory Vaccine Center and 2 Department of Pathology and 1 Emory Vaccine Center, Emory University School of Medicine, Laboratory Medicine, Emory University School of Medicine, Atlanta, GA, USA; 2 Merck Vaccine Division, West Point, PA, USA Atlanta, GA, USA; 3 Biostatistics, Rollins School of Public Health, Emory University, Atlanta, GA, USA; 4 Merck Vaccine Division, Background: AIDS vaccine vectors derived from the highly West Point, PA, USA attenuated poxvirus MVA have, to date, exhibited limited utility in clinical evaluations that largely reflects MVA’s anti- Background: Modified vaccinia virus Ankara (MVA) is now genic complexity and inability to replicate in vivo. Thus, being evaluated as a candidate AIDS vaccine. However, pre- strategies to enhance the immunogenicity of such vaccines liminary Phase I data suggest that modifications to enhance through genetic modification of the MVA vector to reduce its the immunogenicity of MVA may need to be incorporated antigenic complexity and/or augment presentation of vector- into vaccine constructs. encoded antigens from cells initially transduced at a site of Aims: As dendritic cells (DCs) are key targets of MVA and its immunization should result in new vaccine vectors that parent, vaccinia virus (VV), we sought to characterize the exhibit improved immunological profiles. response of these professional antigen-presenting cells to infec- Objective: To characterize the fate of cells infected with a tion in order to elucidate the mechanisms by which the antivi- novel MVA vaccine vector that is rendered genetically ral immune response may be generated upon vaccination. blocked for progression through the E (early) → L (late) Methods: Deconvolution microscopy was used to study the phases of the poxvirus replication cycle. virus replication cycle in DCs and RT-PCR, flow cytometry, Methods/Results: We have generated a novel MVA vector ELISA, metabolic labeling, and proteomic analyses were used that is genetically blocked for progression through the to evaluate cell viability and mRNA and protein expression poxvirus replication cycle via deletion of the udg gene and following infection. propagation of the deletion mutant on an immortalized CEF- Results: We found that both MVA and VV abortively infect derived cell line engineered to complement this deletion in and inhibit the maturation of DCs, but that certain viral early trans. We have evaluated the fates of cells infected with this gene products are differentially expressed in MVA- versus VV- mutant in comparison to those infected with parental MVA. infected DCs. Importantly, only MVA induced (at both the Whereas infection of human DCs with both MVA and AIDS Vaccine 2006 171 Poster sessions_revGJ 14/8/06 16:34 Page 172 Amsterdam, Netherlands 29 August–1 September 2006 RNA and protein level) the production of specific immunoreg- as the reciprocal dilution by an ELISA test. The number of ulatory factors, including the Th1-polarizing chemokines IP- IFN-γ secreting T cells reaches over 350 SFU per million 10 and ITAC. Further, MVA induced a more rapid apoptosis splenocytes against a CD8-specific epitope of GFP. of DCs compared to VV, and this apoptosis was associated Conclusions: We believe that MVTT is an attractive vector for 2 with a faster inhibition of cellular protein synthesis and the development of vaccines against pathogens like HIV-1. decreased levels of anti-apoptotic Bcl-2 family proteins. Since MVTT carries a unique genetic modification, further 2 Discussion: These findings contribute to our understanding characterization of this virus in parallel with other MVTT of the increased immunogenicity of MVA compared to VV, strains generated will define viral factors which may contribute provide a rationale for the use of MVA-vectored vaccines, to the immunogenicity of these vectors. and suggest targeted modifications, such as those that affect the induction of DC apoptosis or maturation, to improve the efficacy of MVA-based AIDS vaccines. P09B-27 DNA microparticles delivered intranasally are able P09B-26 to prime mucosal and systemic anti-SIV immune responses in rhesus macaques Construction and characterization of a novel repli- cation competent vaccinia Tian Tan-based vector LD Giavedoni1, VL Hodara1, H Shen2, LM Parodi1, WM for vaccine development Saltzman3 and E Goldberg4 1,2 1* 1,2* 1 1 Z Chen , W Zhu , Q Fang , K Zhuang , W Yu , 1 Southwest Foundation for Biomedical Research, San Antonio, 1 1,2 1 1,2 H Wang , L Liu , P Tien and L Zhang TX, USA; 2 University of Washington, Washington, WA, USA; 3 Yale University, CT, USA; 4 Northwestern University, IL, USA 1 State Key Laboratory of Virology, Modern Virology Research Center and AIDS Center, Wuhan University, Hubei, PRC; 2 Aaron Background: Currently, the most effective method of induc- Diamond AIDS Research Center, The Rockefeller University, New ing antiretroviral cell-mediated immunity in non-human * York, NY, USA ( equal contributor) primates utilizes priming with naked plasmid DNA and then boosting with recombinant viral vectors both encoding Background: The parental vaccinia Tian Tan strain (VTT) was various parts of the HIV/SIV genome. used as a live vaccine against smallpox for millions of people Objectives: To study in rhesus macaques the systemic and in China before 1980. Although VTT is significantly less vir- mucosal immunogenicity of controlled release vaginal rings ulent than vaccinia WR strain, it remains neurovirulent in and biodegradable microspheres loaded with DNA encoding mice after intracranial infection and causes significant body SIV proteins applied to different mucosal surfaces. weight loss after intranasal inoculation. To generate a safer Methods: A plasmid DNA was prepared that contained the vaccine vector, we sought to construct attenuated VTT strains. whole SIVmac239 coding sequences driven by the CMV pro- Objective: To develop a potent and safe VTT-based vaccine moter (pcSIV); the plasmid lacked the SIV 5′ and 3′ LTRs and vector for inducing effective immunity against foreign antigens the RNA packaging signal, and was not replication compe- Methods and Results: In this study, we were able to generate tent. Plasmid DNA was incorporated into poly(ethylene-co- a modified replication-competent VTT (MVTT ) through vinyl acetate) (EVAc) matrices or poly(lactide-co-glycolide) 2 genetic modification by selectively deleting several viral (PLGA) microspheres. Six female rhesus macaques were used genes. Through in vitro and in vivo characterization, we in this study. Three received vaginal rings sutured in place found that MVTT displays significantly reduced cytopathic- (IVg) and loaded with 1 mg of pcSIV DNA and three received 2 ity and host cell range compared to the parental virus. 0.5 mg of pcSIV DNA loaded microspheres administered MVTT loses its replication capacity significantly in rabbit intra-nasally (IN). All of the animals were given intranasal 2 kidney RK13, pig kidney PK(15) and human cervix HeLa booster immunizations of 0.5 mg of DNA loaded micros- cells. MVTT is over 100 fold less virulent than the parental pheres in 0.5 ml of PBS at 6 weeks, 12 weeks and 22 weeks 2 virus, as determined by causing body weight loss after following the primary immunization. This was followed by intranasal viral inoculation. Moreover, MVTT displays over an intramuscular injection of 107 plaque-forming units (PFU) 2 360 fold drop of intracranial LD in infected mice. To fur- of vaccinia virus expressing SIV Gag, Env and Nef at week 50 ther explore the potential use of MVTT as a vaccine vector, 28. Two new animals, which were not vaccinated with pcSIV, 2 we placed a green fluorescent protein (GFP) gene in the dele- were included as controls for primary immune response tion region. The GFP gene is stable in MVTT genome after against VV. Overlapping peptides were used for detecting cel- 2 over 10 rounds of passages in Vero cell, a candidate cell line lular responses against several SIV genes, and ELISA was for future vaccine production. Importantly, MVTT -GFP was used for humoral responses against SIV Gag and Env. 2 able to elicit both humoral and cell mediated immune Results: Antibody and humoral responses against SIV were responses to GFP in infected animals after intramuscular not detectable after the series of DNA immunizations. inoculation. With two immunizations of 105 pfu, the anti- However, after VV inoculation animals primed with DNA GFP antibody endpoint titer reaches over 700 as determined had stronger mucosal and systemic cellular responses than 172 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 173 Amsterdam, Netherlands 29 August–1 September 2006 animals that only received VV. These anamnestic responses mucosal immune responses from memory in orally immu- were statistically higher in animals that received the first nized mice after vaccinia-gag challenge and has potential as a DNA immunization by the IN route compared to the live vaccine vector in HIV/AIDS and cancer research. macaques receiving vaginal rings. Conclusions: These studies show that delivery of DNA in biodegradable microparticles to mucosal surfaces is able to prime the immune system for responses at the mucosal and sys- P09B-29 temic levels. They also indicate that the intranasal route of pri- mary immunization is more efficient than the intravaginal one. Blockade of a negative immuno-regulatory path- way increases magnitude of HIV-1 specific CD8+ T cells induced by simian origin adenovirus vector P09B-28 HCJ Ertl and MO Lasaro Robust immune responses to an attenuated antibi- otic-gene-free attenuated strain of listeria mono- Wistar Institute, Philadelphia, PA, USA cytogenes that expresses HIV-gag Objectives: Our objective was to optimize HIV-1-specific Z Li, M Zhang, X Zhao, C Zhou, and F Frankel immune responses to adenovirus vaccine carriers. Vectors based on E1-deleted adenoviruses are among the most Dept of Microbiology, School of Medicine, University of promising vehicles for delivery of antigens of HIV-1. To cir- cumvent the expected negative effect of pre-existing immu- Pennsylvania, Philadelphia, USA nity to common human serotypes of adenovirus such as serotype 5, most commonly used as a vaccine vehicle, our Background: Cellular immunity plays an important role in efforts have focused on vectors derived from chimpanzee controlling HIV/AIDS disease progression. Listeria monocy- origin viruses. togenes is known to induce strong cellular immunity. A Methods: E1-deleted adenovirus vectors were constructed recombinant strain of Listeria monocytogenes carrying HIV- from molecular clones and tested for inhibition of T cell gag (Lmdd-gag) attenuated by virtue of deletions in the two responses in mice. genes (dal, dat) required for D-alanine synthesis has been Results: The chimpanzee-origin adenovirus vectors were shown to induce a long-lived, protective systemic and mucos- shown to induce potent CD8+ T cell responses to antigens of al immune response in mice when given in the presence of the HIV-1/SIV, which could be increased in magnitude and mod- required D-alanine. We report here a new generation strain ified in caliber by heterologous prime boost regimens. To fur- based on Lmdd designed to bypass the necessity of exogenous ther increase the magnitude of the CD8+ T cell response to an D-alanine administration without compromising the safety antigen of HIV-1 we modified the expression cassette to associated with the original strain. express the antigen as a fusion protein within an Objectives: To evaluate the immune responses to HIV-1 gag immunomodulatory sequence, which interferes with a nega- + in memory and effector CD8 T cells in mice orally immu- tive regulatory pathway. Specifically we used an antigen, i.e., nized with a new strain of attenuated recombinant Listeria the glycoprotein D (gD) of herpes virus simplex type 1 (HSV- after vaccinia-gag challenge. 1), which interferes with the herpes virus entry mediator Methods: We have introduced a compromised B.subtilis dal (HVEM) – B and T lymphocyte attenuator (BTLA) pathway. gene (which encodes racemase) along with HIV-gag into the HVEM is expressed on dendritic cells while BTLA is Lmdd chromosome to form a new attenuated strain, expressed on activated T and B lymphocytes and interactions LmBsdal +dat- HIVgag. comp between HVEM and BTLA result in down-regulation of Results: After 5 months from the last boost following two immune responses. A chimpanzee origin E1-deleted aden- oral immunizations of BALB/c mice for three consecutive ovirus vector of serotype C68 expressing gag as a fusion pro- days each time, the mice were challenged intraperitoneally tein within gD of HSV-1 induced exceptionally potent and with recombinant vaccinia virus expressing gag. At day 6, the sustained gag-specific CD8+ T cell responses that were LmBsdal +dat- HIVgag-immunized mice elicited robust comp markedly higher in magnitude compared to those induced by hi + Gag-specific CD11a CD8 T cells in splenocytes (39.0%) a vector expressing gag only. The effect depended on binding measured by tetramer staining, significantly increasing more of gD to HVEM as mutation of the binding site abrogated the than 100-fold from the memory level prior to vaccinia–gag adjuvant effect of gD. challenge (0.34%). The immunized mice also elicited strong Conclusion: Our results show that the immunogenicity of responses in MLN (6.30%) and Payer’s patches (30.0%), replication-defective chimpanzee-origin adenovirus vectors while control mice (vaccinia-gag only or Listeria vector with- can be augmented by addition of a sequence that inhibits a out expressing gag) had only background or near-background negative immunoregulative pathway. responses. The vaccinia virus load in the ovaries in the immu- nized mice were significantly lower than those in the control groups. Conclusion: These results demonstrated that this new attenu- ated strain of Listeria can rapidly induce robust systemic and AIDS Vaccine 2006 173 Poster sessions_revGJ 14/8/06 16:34 Page 174 Amsterdam, Netherlands 29 August–1 September 2006 P09B-30 P09B-31 A novel Clostridium perfringens-based SIV vac- Long-term control of simian immunodeficiency cine with adenovirus boosting induces strong virus infection (SIV) in macaques by mucosal systemic and gut mucosal immune responses immunization with a bimodality vector combination RA Helmus, P Poonam, L Caruso, Y Chen and S Kuate2, C Stahl-Hennig1, M Franz1, YS Suh1, H Stoiber3, P Gupta R Steinman4, S Norley5, KS Park6, YC Sung6, K Tenner- Racz7, P Racz7 and K Überla2 University of Pittsburgh, Pittsburgh, PA, USA 1 Department of Virology and Immunology, German Primate Background: Effective vaccination against HIV/SIV must Centre, Göttingen, Germany; 2 Ruhr-University Bochum, induce mucosal and systemic immune responses. Mucosal Bochum, Germany; 3 Institute of Hygiene and Social Medicine, immunity is difficult to prime through systemic immuniza- University Innsbruck, Innsbruck, Austria; 4 The Rockefeller tion, and mucosal tissues are inherently tolerogenic. Mixed University, New York, NY, USA; 5 Robert-Koch-Institute, Berlin, modalities of vector and route-of-delivery help overcome Germany; 6 Cellular Immunology Lab., Division of Molecular and these obstacles. Adenoviral vectors induce strong systemic Life Sciences, Pohang University of Science and Technology, immune responses. A novel recombinant Clostridium perfrin- Pohang, Republic of Korea, 7 Bernhard-Nocht-Institute, gens has been engineered to express high amounts of antigen Hamburg, Germany following oral administration. These vectors have not been tested together. Objectives: This study examines effects of priming and boost- Introduction An efficient needle-free vaccine strategy against ing combinations using intramuscular (IM) adenovirus- and AIDS is most desirable. We therefore compared mucosal ver- oral C. perfringens-based vaccines. Additionally, the mucosal sus systemic vaccine application in the SIV/macaque model by dendritic cell (DC) response to the oral vaccine was explored. means of a bimodal prime-boost regime. Methods: Mice were immunized against SIV p27 using one Objectives: We compared mucosal versus systemic vaccine dose of adenovirus encoding p27 (AdV-p27), three doses of application in the SIV/macaque model by means of a bimodal C. perfringens encoding p27 with cholera toxin and CpG prime-boost regime. oligodinucleotide adjuvants (Cp-p27), AdV-p27 with two Cp- Methods: Rhesus monkeys were primed with Single cycle p27 boosts, or two Cp-p27 primes followed by AdV-p27 immunodeficiency virus (SCIV) vaccines by either oropharyn- boost. ELISAs quantified serum IgG and IgG and fecal, geal spray application of the vaccines (group 1, n=4) or sys- 1 2a vaginal, and intestinal IgA. Spleens, mesenteric lymph nodes temic /intramuscular application (group 2, n=4). Monkeys (MLNs), and Peyer’s patches (PPs) were tested for p27-spe- were boosted with replication-incompetent adenovirus vec- cific interferon-γ via ELISpot. To further investigate the tors expressing SIV-gag/pol and -env/rev (Ad5-SIV) by either response of PP DCs to C. perfringens vaccine, maturation and route. All the vaccinees and control animals (n=4) were chal- priming abilities of PP DCs exposed to vaccine were exam- lenged orally with SIVmac239 20 weeks after the first immu- ined through flow cytometry and ELISpot. nization and the immune response monitored. Results: Mice who received AdV-p27 displayed strong Results: The vaccinees developed SIVgag and SIVenv specific humoral and cellular responses in serum, spleen, and MLN, antibodies after Ad5-SIV boost. Only systemic immunization whereas mice who received Cp-p27 had high cellular and induced SIV-specific neutralising antibodies. Ad5-specific moderate humoral responses in PP and intestinal wash. In all neutralising antibodies appeared in group 1 only after the sec- tissues, mice inoculated with two Cp-p27 primes plus AdV- ond Ad5-SIV application, but already were pronounced after p27 boost had the highest cellular responses. PP DCs exposed the single i.m. injection. T-cell responses determined by IFNγ- to vaccine upregulated costimulatory molecules, lacked ELISpot assay were observed in all vaccinees and directed phagocytosis, and stimulated interferon-γ production by p27- against accessory and structural proteins. Following chal- specific cells. lenge, peak plasma viral RNA loads of all vaccinees were Conclusions: These results demonstrate that C. perfringens about 100–200 fold lower than those of controls. At set expressing high levels of SIV antigen can effectively prime point, viral loads in the vaccinees were still about 50-fold mucosal immunity by delivering antigen to PP DCs. lower compared to controls. This reduced viral load lasted in Furthermore, stimulation of mucosal and systemic cellular group 1 for up to one year after challenge. and humoral responses against antigen is enhanced through Conclusion: A combination of SCIV and adenoviral vectors boosting with systemically delivered adenovirus vaccine. given orally was as immunogenic and efficient as systemic Induction of cellular immune responses at mucosal gut sur- injection of the same antigens leading to long-lasting reduc- faces could help control SIV infection following mucosal tion of viral replication. Experiments replacing SCIV by SIV inoculation. Ongoing studies will determine the optimal com- virus-like particles and/or using a reversed immunization bination of C. perfringens SIV vaccine with boosting vector order are under way. to induce maximum gut mucosal immunity. 174 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 175 Amsterdam, Netherlands 29 August–1 September 2006 P09B-32 Microdermabrasion of skin to target cutaneous dendritic cells for MVA-based AIDS vaccines HS Gill1, A Fedanov2, SN Andrews1, S Sakthivel2, I Williams2, F Priddy2, MR Prausnitz1 and S Staprans2 1 Georgia Institute of Technology, Atlanta, GA, USA; 2 Emory University, Atlanta, GA, USA Background: Modified vaccinia Ankara (MVA) has been shown to have tropism for dendritic cells (DCs). Since skin is an immunocompetent organ rich in epidermal Langerhans cells (LCs) and dermal DCs, it provides an attractive target for delivery of MVA-based AIDS vaccines. However, a safe and reliable method to target the skin for mass immunization is needed. Microdermabrasion is an FDA-approved cosmetic treatment that has the potential to expose the underlying epi- dermis for vaccine delivery by selectively removing stratum corneum (SC)-the topmost barrier layer of the skin. Objectives: To determine if microdermabrasion can selective- ly remove stratum corneum and to study its effect on skin dendritic cell population. Methods: Microdermabrasion was performed on human sub- jects (n=19) or rhesus macaques (RMs, n=9) at different microdermabrasion conditions using a commercial device. Skin electrical resistance before and after microdermabrasion was measured using a skin-resistance meter. Biopsies were collected from microdermabraded sites 0 or 24 h post micro- dermabrasion for histology. Histology sections were analysed for presence or absence of SC, and to count the dermal and epidermal DCs at 0 and 24 h. Control biopsies from untreat- ed sites were also collected. Skin electrical resistance mea- sures were correlated with the presence or absence of SC upon microdermabrasion. Results: Skin resistance decreased by as much as 99% after microdermabrasion in both humans and RMs. Post-micro- dermabrasion resistance correlated with extent of SC removal. At 20–30 kohms, complete removal of SC and viable epidermis was observed, while at 80 kohms the major- ity of SC was removed without apparent damage to the viable epidermis. Thus, skin-resistance can potentially act as a non- invasive predictor of SC removal. At severe microdermabra- sion conditions, blisters (separation of dermis and epidermis) were observed in the skin, while weaker conditions resulted in partial to focal SC removal without blisters. LC and DC populations increased 1.5 and 2 fold respectively, 24 hrs post- microdermabrasion (LCs: P<0.05, DCs: P<0.002). Conclusions: Microdermabrasion caused selective removal of SC and an increase in cutaneous DC populations 24 h post- microdermabrasion, offering an easy and perhaps more immunogenic route of vaccination. SC removal correlated with decreases in skin electrical resistance, providing a non- invasive indicator of SC removal. AIDS Vaccine 2006 175 Poster sessions_revGJ 14/8/06 16:34 Page 176 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 9C: VACCINE CONCEPTS & DESIGN – ADJUVANTS P09C-01 P09C-02 Complementary HIV specific CD4 and CD8 Intranasal immunization with a multigene vaccine responses: NHP adenovirus and adjuvanted protein in combination with a novel adjuvant, N3, induces prime boost systemic and mucosal immune responses A Bråve1, U Schröder2, B Wahren1 and J Hinkula1 NL Mathy1, Y Delmarcelle1, J Wilson2, M Cleveland3, A Beaton3 and G Voss1 1 Swedish Institute for Infectious Disease Control and Microbiology and Tumor Biology Center, Karolinska Institutet, 1 GlaxoSmithKline Biologicals, Rixensart, Belgium; 2 University Stockholm, Sweden; 2 Eurocine AB, Karolinska Science Park, of Pennsylvania, Pennsylvania, USA; 3 GlaxoSmithKline Stockholm, Sweden BioPharm CEDD, Stevenage, UK Background: We have developed a multigene/multisubtype Background: Accumulating evidence is suggesting that an HIV-1 vaccine that is currently being tested in two clinical effective HIV vaccine should elicit both CD8 and CD4 cellu- phase I trials. The vaccine contains genes encoding multiple lar responses in parallel with a humoral response. This com- subtypes of envelope, gag and RT. We are currently investi- plex immune response is most likely achieved by gating ways to enhance the vaccine specific mucosal multi-component immunization regimen. immune response. In order to enhance the delivery and Objectives: We have therefore investigated the immuno- immunogenicity of the plasmids we are using a novel genicity of adjuvanted protein vaccines, which induce pre- cationic lipid, N3, as adjuvant. dominantly CD4 cell and antibody responses, in Objectives: To evaluate, in mice, intranasal delivery of the combination with a potent CD8 cell-inducing recombinant genetic vaccine in combination with the adjuvant N3 and vector, NHP adenovirus C6. assess the systemic and mucosal responses elicited. Methods: CB6F1 mice were primed with adjuvanted protein Methods: Young C57/BL6 mice (2 weeks old at the start of then boosted with NHP adenovirus, or vice versa. Following the experiment) were immunized 3 times with the multigene each immunization cellular and humoral responses were vaccine intranasally or intramuscularly with or without the analysed for intracellular cytokine flow cytometry or anti- adjuvant N3. The mice received 20 µg of plasmid at each body ELISAs, respectively. immunization. The production of vaccine-specific antibodies, Results/conclusions: Immunization using the prime boost both IgG and IgA, was measured in sera as well as in vaginal regimen of adjuvanted protein and NHP adenovirus express- washes and feces. The cellular responses were assessed by ing a model HIV antigen, gp120 W6.1D, generates comple- IFN-? and IL-2 ELISpot on PBMCs. mentary CD4, CD8 and antibody responses directed against Results: After the second immunization all animals immu- the HIV antigen. The CD4 responses are predominantly nized with the addition of the adjuvant N3, either intranasal- CD4+ IL-2+ cells and are generated by the adjuvanted protein ly or intramuscularly, responded with significant titers of component of the vaccine. The CD8 responses are predomi- anti-gp160env and anti-p55gag IgG in sera. Furthermore, nantly of the CD8+ IFNγ subtype and exhibit cytolytic effec- animals immunized intranasally with addition of N3 had sig- tor cell function. Interestingly the order of priming and nificant vaccine-specific IgA titers in both vaginal washes and boosting plays a key role in the kinetics of the immune feces. Following the third DNA immunization the humoral response as well as the intensity of specific responses. These responses were boosted even further in the adjuvanted ani- data indicate that the combination of an adjuvanted protein mals. The cellular responses, as measured by IFN-γ and IL-2 vaccine and a non replicative adenovirus vector results in the reactivity of PBMCs following the third immunization, were induction of broad and complementary immune responses. directed against both gag and env with the broadest respons- es detected in the animals adjuvanted by N3. Conclusions: By immunizing with a low dose genetic vaccine in formulation with a novel cationic adjuvant, N3, we could induce broad humoral and cellular anti-HIV responses. Furthermore, intranasal immunization resulted in a mucosal IgA response directed against both envelope and gag. Our results are encouraging since the immune responses in mucos- al surfaces is believed to be highly important in order to inhibit sexual transmission of HIV. 176 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 177 Amsterdam, Netherlands 29 August–1 September 2006 P09C-03 P09C-04 Neutralizing antibody responses to subtypes B and Type I interferons as biological adjuvants for the C adjuvanted HIV envelope protein vaccines in improvement of poxvirus HIV vaccine efficacy rabbits SL Day, C Ranasinghe, IA Ramshaw B Burke, VR Gomez-Roman, Y Lian, E Kan, Y Sun, G Otten, J Ulmer, J Donnelly, I Srivastava, and SW The John Curtin School of Medical Research, Australian National Barnett University, ACT, Australia Chiron Corporation, Emeryville, CA, USA Background: Poxvirus prime-boost vaccination has proven to generate high levels of cell-mediated immunity. Inclusion of Background: Adjuvants have been shown to enhance the biological adjuvants in viral vectors may facilitate to steer the quality and breadth of immune responses against infectious immune response to provide enhanced protection, hence diseases. Investigation of novel adjuvant formulations for understanding the fundamental nature of the immune envelope (Env)-based HIV vaccines could serve to improve responses generated by poxvirus vectors that co-express viral the potency, breadth, durability, and protective efficacy of the antigens may enable the design of better vaccine adjuvants immune responses induced by these vaccines. for the future. Objectives: To evaluate the effects of Chiron’s MF59 adju- Aim: Evaluate the immunomodulatory effects of co-expres- vant and MF59 plus CpG oligonucleotides on the induction sion of murine interferonα4 (mIFNα4) and β (mIFNβ) by of neutralizing antibodies when co-delivered with Env pro- poxviral vectors with aims to enhance immune responses to teins from subtypes B and C HIV-1 strains. HIV antigens. Methods: Rabbits were immunized four times intramuscular- Methods: Recombinant vaccinia viruses co-expressing the ly at 0, 4, 12, and 24 weeks with purified o-gp140∆V2 pro- influenza haemagglutinin (HA) antigen and mIFNα4 or teins in MF59 adjuvant in the presence or absence of CpG. mIFNβ were constructed as a model system for HIV. The Rabbits were divided into six groups with ten rabbits per potential of these vectors was investigated using the prime- group. Two groups (1 and 2) received o-gp140∆V2 SF162 boost system, analysing both the specific antibody and T cell (subtype B), two (3 and 4) received o-gp140∆V2 TV1 (sub- responses. The nude mice and BALBc models were used to type C), and the final two (5 and 6) received o-gp140∆V2 study the effect of interferon expression on the pathology of from both SF162 and TV1. Even numbered groups received the viruses. CpG as well. Serum samples were collected throughout the Results: Vaccination with IFN-expressing vaccinia vectors immunization schedule and were analysed for neutralizing demonstrated that a higher number of HA-specific T cells antibody titers. could be generated by co-expression of IFNα4 when com- Results: The presence of CpG significantly enhanced the neu- pared to the control virus expressing HA alone (P=0.031). In tralizing titers against SF162 as compared to rabbits immu- addition, similar levels of HA-specific T cells were generated nized with the same immunogen in MF59 alone. by the IFNβ-expressing virus compared to the control virus. Interestingly, following the fourth immunization, the 80% Growth kinetics of the viruses in murine L929 cells demon- SF162 neutralization titers in the SF162 alone groups were strate a significant reduction in viral titres with co-expression not significantly different from those in the TV1 alone of IFNβ. Reduction in the viral growth was also observed in groups. However, the bivalent vaccines showed significantly vivo in BALBc mice, indicating the possibility that IFNγ higher neutralization titers than either protein alone. induced by the virus-encoded IFNβ may contribute to viral Immunizing with both SF162 and TV1 with MF59 and CpG clearance. However studies in both nude mice and mice lack- resulted in the highest geometric mean neutralizing titers ing the IFNγ receptor suggest that the reduction in viral against SF162. Neutralization titers against a TV1 clone were growth is T cell- and IFNγ- independent. only observed in the immunization groups containing CpG. Conclusions: Type I interferons expressed by poxviral vectors The TV1 with MF59 and CpG group yielded the most may improve vaccine efficacy and safety, by reducing viral responders and elicited modest neutralization titers against replication in vivo while still inducing strong immune this TV1 clone following the third immunization. responses to the virally encoded antigen. Studies are under- Conclusions: In the case of SF162 neutralization, the use of a way to investigate whether the enhancement of immuno- bivalent subtype B and C vaccine yielded more potent anti- genicity of the model antigen, HA, by the type I IFNs, can be body titers than either protein alone. Additionally, the o- translated to HIV antigens. gp140∆V2 vaccines described here elicit more potent neutralizing antibody responses when formulated in MF59 adjuvant with CpG. AIDS Vaccine 2006 177 Poster sessions_revGJ 14/8/06 16:34 Page 178 Amsterdam, Netherlands 29 August–1 September 2006 P09C-05 P09C-06 Adjuvanting effects of dsRNA molecules tran- Interleukin-7 and interleukin-15 enhance antigen- scribed from DNA vaccine vector specific memory T-cell responses elicited by topical HIV-1 DermaVir vaccine DF Li, Y Liu, YW Zhang, KM Zhou, T Ma, J Sun, JQ Xu, YM Shao SA Calarota1,2, A Dai1, JN Trocio2, DB Weiner1, F Lori2 and J Lisziewicz3 National Center for AIDS/STD Control and Prevention, Chinese Center for Disease Control and Prevention, Beijing, P. R. China 1 Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA, USA; 2 Research Objectives: To investigate the adjuvant effects of plasmid vec- Institute for Genetic and Human Therapy (RIGHT), Pavia, Italy & tor- derived hairpin double-strand RNA (dsRNA) molecules Washington DC, USA; 3 Genetic Immunity, LLC, Washington DC, USA generated in antigen-expressing cells. Methods: A series of novel plasmid vectors- pDS40, pDS200 Background: DNA-based vaccines offer the exclusive advantage and pDS750, each of which carries an expression cassette for of allowing repeated administrations since immunity is elicited a hairpin RNA molecule with a 9-nt loop and double-strand primarily to the encoded antigen(s). However, following naked stem, were constructed from conventional DNA vaccine vec- DNA vaccination, antigen-expressing myocytes fail to induce tor pDRVI2.1. pDS40, pDS200 and pDS750 encode for RNA robust antigen-specific memory responses. To augment such molecules with double-strand stems of 40 bp, 200 bp and 750 responses, a topical DNA-based vaccine, DermaVir, has been bp under the control of CMV promoter, respectively. At the developed to target dendritic cells and express the antigen at the same time, serving as control vector, plasmid vector pTR200 draining lymph nodes, improving antigen presentation. containing tandem repeated RNA sequence which is comple- Objectives: To compare DermaVir and naked DNA vaccines and ment with that in the hairpin dsRNA of pDS200 was con- evaluate cytokine adjuvants to increase DermaVir-induced mem- structed. The HIV-1 gp140 gene was subcloned into the ory responses in the murine model. vectors mentioned above to obtain five DNA vaccine plas- Methods: Balb/C mice were topically immunized with DermaVir mids- pDRVI2.1-env, pDS40-env, pDS200-env, pDS750-env formulated with an HIV-1 Gag-encoding plasmid (pGag) or and pTR200-env, respectively. Groups of Balb/C mice with 6 intramuscularly with naked pGag, alone or in a DNA prime fol- in each group were given two or three intramuscular injec- lowed by an HIV-1 Gag-expressing vaccinia vector boost given tions with 40 µg or 4 µg plasmid. Cell mediated immune intraperitoneally. Additionally, mice received topically DermaVir responses were determined by IFN-γ ELISPOT assay. formulated with a mixture of pGag and an interleukin (IL)- Results: Env-specific IFN-γ secreting cells induced by pDS40- encoding plasmid (IL-12, IL-7, or IL-15), alone or at DNA prim- env or pDS200-env were significantly higher (P<0.01) than ing/vaccinia boost. Antigen-specific effector memory T-cell those induced by pDRVI2.1-env when Balb/C mice were i.m. responses were evaluated by standard IFN-γ ELISPOT assay in immunized two or three times with 40µg DNA. Cellular response to peptide pools corresponding to the complete HIV-1 immune response induced by 4 µg pDS40-env was almost as Consensus B Gag sequence. Durable central memory T-cell high as that induced by 40 µg pDRVI2.1-env, which was 3- responses were evaluated by a cultured IFN-γ ELISPOT assay, fold higher (P<0.01) than that induced by 4 µg pDRVI2.1-env. using the same ELISPOT procedure after splenocytes were cul- Neither pTR200-env nor pDS750-env showed superiority to tured with the corresponding peptide pool for 6 days. plasmid pDRVI2.1-env in induction of antigen specific cellu- Results: Both DermaVir and naked DNA immunization induced lar immune responses. HIV-specific effector memory T-cell responses but only Conclusion: HIV-1 Env specific cellular immune responses DermaVir vaccination resulted in effective central memory T-cell were significantly improved by co-expressing hairpin RNA responses. Both IL-7 and IL-15 significantly improved molecule with proper length of double-strand structure in DermaVir-induced Gag-specific central memory T-cell responses, antigen expressing cells. but IL-12 did not, while IL-15 also enhanced effector memory T- cell responses. IL15 with DermaVir in the prime followed by a vaccinia boost significantly improved effector memory T-cell responses. Neither IL-7 nor IL-15 improved Gag-specific central memory T-cell responses in the DermaVir prime/vaccinia boost regimen. CD8+ T-cell depletion from splenocytes showed that CD8+ T-cells mainly mediated the antigen-specific IFN-γ responses. Conclusions: DermaVir induces robust antigen-specific central memory T-cells responses, which are likely to be more efficient in mediating protective immunity. The inclusion of IL-15 further improves long-term memory T-cell responses. DermaVir togeth- er with IL-15 might be an effective prime in a prime/boost pro- phylactic HIV-1 vaccine regimen. 178 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 179 Amsterdam, Netherlands 29 August–1 September 2006 P09C-07 P09C-08 Increasing the efficiency of DNA/MVA vaccines Cytokine and chemokine-mediated augmentation of MVA immunogenicity HL Robinson, RR Amara, L Lai and J Zhao DA Garber1, KA Marfatia1, R Chavan1, I An1 and Yerkes National Primate Research Center and Emory Vaccine MB Feinberg1,2 Center, Emory University, Atlanta, GA, USA 1 Emory Vaccine Center, Emory University School of Medicine, Background: Since the original publication of our results using Atlanta, GA, USA; 2 Merck Vaccine Division, West Point, PA, USA DNA priming and MVA boosting to protect against a SHIV- 89.6P challenge (Science, 2001) we have identified conditions Background: MVA is a safe and promising vector for vaccina- that improve the efficiency of protection. The first of these tion of humans against HIV. However, its ability to prime involves altering the immunization schedule and the second, immune responses is ultimately limited as a result of its inabili- using GM-CSF DNA. ty to replicate in vivo that restricts transgene expression to those Methods: Rhesus macaques were primed with a DNA express- cells initially infected at a site of immunization. As a result, ing 89.6 Gag-RT-Env-Tat-Rev and Vpu and boosted with a strategies to recruit and/or activate DCs/APCs by expressing rel- MVA expressing 89.6 Gag-RT and Env using different immu- evant cytokines or chemokines from MVA vectors should result nization schedules in the presence or absence of GM-CSF DNA in immunologically improved vaccine vectors. with the DNA prime. Challenge was >6 months post the final Objective: To comparatively evaluate the cellular and immunization by an intrarectal dose of SHIV-89.6P. humoral immune responses elicited by MVA recombinants Results:. The addition of a second MVA booster immunization that express immunomodulatory cytokines or chemokines in increased the titers of binding Ab by 100-fold which appeared both murine and non-human primate models. to enhance protective efficacy for the SHIV-89.6P challenge. Methods: Cellular and humoral immune responses elicited The use of GM-CSF DNA co-inoculated (either in cis or in against recombinant MVA vectors were assessed by MHC-I trans) with the DNA vaccine also improved protection. This tetramer, intracellular cytokine (TNFα, IFNγ), ELISA, and improved protection was associated with GM-CSF fostering the NAb assays. rapid post challenge expansion of antiviral CD8 T cells in the Results: Single-dose immunization of mice with MVA recom- gut mucosa and the improved protection of CD4 T cells in the binants expressing GM-CSF or MIP3α (but not Flt3L) results gut. Combining a second MVA boost and the GM-CSF adju- in 2-4-fold increases in the frequencies of splenic CD8+ T cells vant resulted in a 25-fold decrease in levels of plasma viral RNA directed against vector-encoded antigens, which persist both at peak infection and at set point. These increases in effi- through at least 30 days post-immunization, as well as >10- ciency resulted in most animals having viral loads below 100 fold enhancements of vector-specific antibody titers, as com- copies per ml of plasma by 5 weeks post challenge. Tests for cor- pared to non-adjuvanted MVA. This cytokine-mediated relations with protection in all of our trials revealed an inverse augmentation of T cell responses was observed only when correlation between peak viral load and pre-challenge neutraliz- MVA recombinants were delivered as primary, rather than ing Ab to 89.6 (this was despite the SHIV-89.6P challenge being booster immunizations, suggesting that these poxvirus- a neutralization escape variant of the SHIV-89.6 immunogens) expressed adjuvants are effective at stimulating naive T cell (n=47, r=0–0.4, P=0.005) and no correlation between peak viral responses rather than promoting recall of pre-existing memo- load and vaccine-elicited CD8 responses (measured using ry T cell responses. Single-dose immunization of mice with tetramers). MVA-MIP3α conferred enhanced protection against lethal Conclusion: The protective efficacy of DNA/MVA vaccines can intranasal VV challenge, indicating enhanced induction of be increased by using immunization regimens that increase mucosal immunity. In contrast, the magnitudes of cellular binding Ab and foster the mucosal presence of responding CD8 immune responses elicited in rhesus macaques following both T cells. primary and booster immunizations with analogous recombi- nant viruses were not significantly different from those engen- dered by parental MVA. Immunization of macaques with MVA-GMCSF or MVA-MIP3α resulted in increased titers of MVA-neutralizing antibodies, but was realized only following multiple immunizations. Conclusions: In a murine model, both cellular and humoral immune responses elicited by recombinant MVA vectors can be enhanced through GMCSF- and MIP3α-augmentation strategies; however further optimization of immunization regimens (route+dose) will likely be required to successfully translate these findings to primates. AIDS Vaccine 2006 179 Poster sessions_revGJ 14/8/06 16:34 Page 180 Amsterdam, Netherlands 29 August–1 September 2006 P09C-09 [WITHDRAWN] survival in mice with B16F10 melanoma. By adding poly- ethylenimine (PEI) to the plasmid DNA, even stronger anti- Novel strategy for generation of mucosal immune tumor effects were seen, confirming previous reports that PEI responses against HIV-1 following systemic vacci- enhances DNA delivery in vivo. Conclusions: Macaque versions of pSP-D-CD40L and pSP-D- nation with muscosally expressed chemokine GITRL have been prepared and produce immunologically immunoadjuvant active proteins in vitro. pSP-D-GITRL may be especially promising as an agent to abrogate the immunosuppressive 1 1 1 1 DB Weiner , MA Kutzler , KS Schoenly , K Muthumani , effects of CD4+CD25+GITR+ regulatory T cells. In murine RM Parkinson1, H Maquire1 and K Ugen2 tumor models, there is a marked synergy between pSP-D- CD40L and two TLR agonists, poly(I:C) and CpG ODN, and 1The University of Pennsylvania School of Medicine, Philadelphia, this synergy is being tested as part of an HIV DNA vaccine. PA, USA, 2The University of South Florida, Tampa, FL, USA P09C-11 P09C-10 Mother-child immunity against HIV-1 after Gagp, Multimeric soluble forms of CD40L and GITRL as Nef, Tat, Gp160/rev plasmid DNA prime recombi- molecular adjuvants for HIV vaccines nant Gp160 boost vaccine regimen RS Kornbluth1, V Snarsky1, S Barzee1, B Tran1, C Toppin1, J Hinkula1,2, A Bråve1, K Johansen1, U Schröder3 and F Villinger2, and GW Stone1 B Wahren1 1 University of California San Diego, La Jolla, CA and the VA San 1 Microbiology and Tumorbiology Center, Karolinska Institutet Diego Healthcare System, San Diego, CA, USA; 2 Emory and Swedish Institute for infectious disease control, Stockholm, University School of Medicine, Atlanta, GA, USA Sweden; 2 Department of Molecular Virology, Linkoping University, Sweden; 3 Eurocine AB, Karolinska Science Park, Background: CD40L and GITRL were produced as multi- Stockholm, Sweden meric soluble proteins by fusing their extracellular domains with a multimerizing scaffold from surfactant protein D (SP- Background: HIV-1 subtype B vaccine was used shown to D). Their plasmid forms, pSP-D-CD40L and pSP-D-GITRL, induce potent responses in immunized newborn animals are effective adjuvants for HIV DNA vaccines in mice. born to HIV-1 immune mothers. Further enhanced Consequently, there is a need for macaque versions of these immunogenicity by HIV-1 DNA immunization when molecular adjuvants for testing in the SIV model. Also, given mucosal adjuvants are used. the synergy between CD40 stimulation and TLR stimulation Objectives: To evaluate the importance of HIV-1 for CD8+ T cell responses, combinations of TLR agonists and plasmid/recombinant protein induced immunity in the moth- pSP-D-CD40L were tested in murine tumor models. er when immunizing pups early in life. Objectives: To prepare macaque versions of pSP-D-CD40L Methods: Female C57Bl/6 mice were immunized three times and pSP-D-GITRL, and to test for TLR agonist/pSP-D- with a combination of HIV-1 plasmids encoding gag, enve- CD40L synergy in murine tumor models. lope, rev, nef and tat of subtype B or with empty plasmid as Methods: pSP-D-CD40L and pSP-D-GITRL have been control. The females were divided into three immunization described (Stone et al., J Virol 2006; 80:1762-172), and their regimens, intramuscular, epidermal or intranasal routes of macaque versions were produced similarly. Conditioned immunization. Further each group was further divided into media containing their corresponding proteins were pro- groups with and without adjuvant in combination with the duced by transient transfection of 293T cells. To look for plasmids. The adjuvants used were N3 and PCPP. When a TLR/CD40 synergy, AB1 mesothelioma and B16F10 clear HIV-1 specific gag/env specific immunity was detected melanoma tumors were established in mice. Once the tumors in the adult mice they were made pregnant. The newborn were >4 mm, pSP-D-CD40L was injected peritumorally every pups were then allowed to breast feed for 3 weeks before they other day X 5 ±TLR agonists. were HIV-1 DNA plasmid immunized i.n.a, i.m or epider- Results: Macaque SP-D-CD40L strongly stimulated the pro- mally with or without adjuvant. The immunological assays liferation of human B cells in the presence of IL-4. Similarly, used were for the cellular responses IFN-γ ELISPOT. The SP-D-GITRL was an effective costimulatory signal for human HIV-specific humoral response was analysed by ELISA, CD4+ T cells stimulated with anti-CD3 antibody. In the stud- which was performed on serum IgG and IgA, fecal pellet IgG ies on TLR/CD40 synergy in mice, only pSP-D-CD40L + and IgA vaginal wash IgA. poly(I:C) (TLR3) or pSP-D-CD40L + CpG ODN (TLR9) had Results: HIV-1 DNA and recombinant protein induced synergistic anti-tumor effects. A triple combination of pSP-D- immunity was strong in the female adult mice when adju- CD40L + poly(I:C) + CpG ODN was even more effective and vants N3 or PCPP was used. Among the newborn pups born cured most mice of AB1 tumors and significantly prolonged to HIV-1 immune mothers a significantly stronger HIV-1 180 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 181 Amsterdam, Netherlands 29 August–1 September 2006 immunity was seen. In pups born to HIV-1 negative mothers. HIV-1 specific immunity was ewoked only if HIV-1 DNA combined with adjuvant N3 or PCPP was used. Mucosal (fecal and vaginal) immune responses was only seen in ani- mals immunized via the intranasal routes, with highest HIV- 1 specific IgA titers when adjuvant was used Conclusions: Animals born to HIV-1 immune mothers bene- fit by responding with stronger humoral responses against HIV-1. To deliver HIV-1 DNA plasmids with adjuvant intranasally enhances the humoral immune responses, sys- temically and mucosally. AIDS Vaccine 2006 181 Poster sessions_revGJ 14/8/06 16:34 Page 182 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 9D: VACCINE CONCEPTS & DESIGN – LIVE-ATTENUATED VACCINES P09D-01 immune-response and high level Th1 cytokines induced by DLV immunization may contribute to the immune protective Live attenuated EIAV vaccine induced high Th1 response against EIAV infection. immune responses and provided immune protec- tion against virulent EIAV infection P09D-02 X Zhang1, Y Wang2, H Liang1, H Li2, L Wei2, W Xiang2, R Shen2 and Y Shao1 Effectiveness of different chemical and physical inactivation methods to reduce the infectivity of 1 State Key Laboratory for Infectious Disease Prevention and HIV-1 strains Control, National Center for AIDS/STD Control and Prevention, Beijing, China; 2 Harbin Veterinary Research Institute, Harbin, C Gil, N Climent, C Hurtado, P Castro, M García, China S Sánchez-Palomino, F García, JM Miró, H Oliva, E Fumero, M Plana, T Pumarola, T Gallart and JM Gatell1 Background: The equine infectious anaemia virus (EIAV) is a member of lentivirus and shares many similarities with HIV. Hospital Clinic IDIBAPS, University of Barcelona, Barcelona, Spain EIAV donkey leukocyte attenuated vaccine (DLV) was pro- duced of the wild type EIAV LN strain by serial in vitro pas- Aim: The use of whole killed HIV-1 as antigen for an effec- sages, and has been proven effective against challenges at tive therapeutic and preventive vaccine involves the poten- fatal doses of virulent EIAV infection in horses and donkeys. tial advantage of using a wide spectrum of viral proteins to Several key mutation sites have identified in the DLV induce a strong immune response. However, it involves two sequence which may contribute to the attenuation. However, of the major obstacles such as to achieve a complete elimi- the immune protective mechanism of DLV was not clear. nation of viral infectivity and to preserve the virion integri- Objectives: To study the roles of T cell immune responses in ty. In this study we have investigated the effectiveness of the protection, we immunized horses with DLV, challenged different chemical and physical inactivation methods to them with a wild type strain LN and determined T cell reduce the infectivity of HIV-1 strains as an effort for a immune responses. future HIV-1 dendritic cell based vaccine using inactivated Methods: Fifteen horses were divided into 5 groups: I; DLV autologous HIV-1. vaccination only, II; DLV vaccination followed by virulent Methods: Heating methods (56°C during 30 min or 2 h and EIAV challenge, III; virulent EIAV challenge without DLV 60°C during 30 min), AT-2 methods (1 mM during 1 h or 2 pre-immunization, IV; the naive health horses, and V; EIAV h; 250 mcrM during 1 h or 2h), and a formaldehyde-heat natural infected horses. Blood samples were collected period- combined method (1:80 formaline for 1 h at 37°C followed ically and subjected to the measurement of CTL responses, by 3 pulses of 10 min at 62°C) were employed to inactivate lymphocyte proliferation, cytokine productions, and viral 1 ml of 5 HIV-1 viruses: the 69/7 clinical isolate, BaL, HTLV- load. IIIMN, HXB2 and SF2. Inactivated viral preparation were Results: All vaccinated horses were asymptomatic and had a filter by centrifugation, washed 3× in PBS and centrifuge at low level of viral replication (<10 copies /ml). The percentage 23 000 rpm for 1 h at 4°C. Pellets were diluted in 0.2 ml of of EIAV-specific CTL response induced in DLV immunized PBS and stored at -80°C. Titration was performed by infect- horses (10-39%) was higher than those in naive health hors- ing PBMC with 10 fold dilutions. The productive infection es. The expression level of cytokines including IFN-γ, IL-2, was determined by ELISA HIV Ag p24. IL-12 in immunized horses was 5–100 folds higher than that Results: Viral infectivity reduction (Ri) was in average 5.5, in healthy control (P<0.01). After challenge with virulent 4.3 and 6.5 log, at 56ºC during 30 min, at 60°C during 30 EIAV (LN), IFN-γ and IL-2 expression continuously min and at 56°C during 2 h, respectively. While using AT-2 increased and peak after 3 months in the DLV pre-immu- treatment (1 mM or 250 µM for 2 hrs) the Ri was in average nization group; then gradually returned. However, that in 5.1 log or 4.8 log, respectively. However in some viruses, horses without vaccination fluctuated (temporally increased residual infectivity (until 3 log) were still found using heat 10–100 folds, then decreased below normal level during (56°C during 30 min) or 250 µM of AT-2 treatments. Viral febrile episodes). DLV vaccinated horses showed no temper- infectivity was also sensitive to the exposition at room tem- ature increase except one spike in one animal, lower viral perature. AT-2 and formaldehyde-heating methods were very load, and significantly increased EIAV specific CTL response. effective reducing the viral infectivity to the limit of sensitiv- The non-vaccinated horses had several fever episodes ity 1.5E+01 TCID50/0.2 ml. Furthermore, the p24 detection (≥39°C) that coincided with higher viral load and lower by ELISA was clearly affected by 56°C treatment during 2 h, cytokines production, and died of Equine infectious anaemia. indicating that conservative epitopes can be impaired. This Conclusions: The results indicated that EIAV-specific cellular effect was not found on AT-2 treated viruses. 182 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 183 Amsterdam, Netherlands 29 August–1 September 2006 Conclusion: Our findings indicate that Heat, AT-2 and com- against the four different HIV-1 antigens. Some differences in bined formaldehyde-heating methods were highly effective to the magnitude and breadth of the immune response were reduce HIV-1 infectivity, although residual infectivity were observed between MVA-B and NYVAC-B which depend on still found in very few of the viruses tested. the vector and protocol of immunization used. P09D-03 P09D-04 Head-to-head comparison on the immunogenicity Live attenuated measles virus as a candidate mul- in mice of two HIV/AIDS vaccine candidates based tivalent vaccine vector against AIDS on the attenuated poxvirus strains MVA and NYVAC co-expressing in a single locus the HIV-1 gp120 BX08 HY Naim, M Liniger, S Weibel and A Zuniga and HIV-1 Gag-Pol-Nef proteins of clade B IIIB Berna Biotech LTD and Etnavax, Bern, Switzerland M Esteban1, CE Gómez1, JL Nájera1, E Pérez-Jiménez1, V Jiménez1, R Wagner 2, M Graf 3, MJ Franchette4, Background: Live attenuated measles vaccines provide excel- 5 6 P Liljeström and G Pantaleo lent safety and efficacy records and induce long-lived immu- nity after a single vaccination dose. Protective efficacy results 1 Centro Nacional de Biotecnología, CSIC, Madrid, Spain; 2 from combined humoral and CTL immune responses. The University of Regensburg, Germany; 3 Geneart GmbH, Edmonston Zagreb strain (EZ) produced by Berna Biotech Regensburg, Germany; 4 Sanofi Pateur, France; 5 Karolinska (MVEZbv) has proven to bea well-tolerated vaccine, particu- Institut, Stockholm, Sweden; 6 Centre Hospitalier Universitaire larly when used as an aerosol vaccine. Vaudois. Lausanne, Switzerland Objective: To assess the value of MVEZ vaccine for use as a multivalent vaccine vector against measles and other diseases including HIV. Background: Although the capacity of MVA and NYVAC Methods: A cDNA plasmid (p(+) MVEZbc) from the com- attenuated poxviruses to produce similar levels of recombi- mercially available Edmonston Zagreb strain MVEZbv was nant antigens as replication-competent viruses has been wide- used for the development of an HIV-multivalent vaccine can- ly demonstrated, to date, a comparative immune response didate. The MVEZbc genome was engineered to incorporate induced by recombinants based on MVA and NYVAC strains multiple genes simultaneously. rMVEZs were rescued by has not been done. transfection of cDNAs into a helper 293-cell line. In vitro and Objective: To generate, characterize and evaluate the in vivo analyses were performed using standard techniques immunogenicity in mice of two HIV/AIDS vaccine candidates such as plaque assay, immunofluorescence, Western blotting, based on the attenuated poxvirus strains MVA and NYVAC RT-PCR, ELISA and plaque neutralization titer (PNT) tests. co-expressing in a single locus the HIV-1 gp120 and HIV- BX08 Stability of gene expression was assessed in cell culture and 1 Gag-Pol-Nef proteins of clade B (referred as MVA-B and IIIB immunogenicity was tested in transgenic mice susceptible to NYVAC-B) using different immunization protocols. MV. Methods: BALB/c and transgenic HHD mice were inoculated Results: A variety of recombinant MVEZ expressing various using different prime/boost combinations. Splenocytes from HIV antigens (Env, Gag, Nef, Gag–Pol, and their derivatives) immunized animals were tested by ELISPOT for their ability were generated. The MVEZ genome was enlarged by 6.8 kb to secrete IFN-γ or IL-2 in response to 8 clade B peptide pools containing additional HIV genes. The levels of protein encompassing the Gag-Pol-Nef and env regions included in expression were shown to be dependent on the positioning of the immunogens. the genes in the MVEZbc genome. Cell culture proliferation Results: The immunogens MVA-B and NYVAC-B are geneti- of all rMVEZs carrying one, two or three HIV-genes was not cally stable and efficiently express the four HIV-1 antigens. In significantly impaired. Transgenic mice immunized with for- DNA prime/poxvirus boost protocols, the strongest immune mulations of various rMVEZbc-HIVs showed significant response was obtained in BALB/c mice boosted with humoral and cellular immune responses against MV and the NYVAC-B, while in HHD mice there were no differences transgenic HIV proteins. The pre-existing MV antibodies did between the poxvirus vectors. When the prime/boost was not inhibit the boost of anti-HIV antibodies. performed with the homologous or with the combination of Conclusion: The collected data proves the utility of MV vector and its poxvirus vectors, the protocols MVA-B/MVA-B and potential for the development of novel recombinant vaccines against NYVAC-B/NYVAC-B, or the combination NYVAC-B/MVA- AIDS because: (i) MV can stably express multiple genes simultane- B gave the most consistent broader immune response in both ously with the potential to activate and boost both arms of the mouse models, although the magnitude of the overall immune system, (ii) rMVEZs induce persisting immune response sim- response was higher for the DNA-B/poxvirus-B regime. All of ilar to the authentic measles vaccine and, (iii) rMVEZ-based HIV the immunization protocols induced some humoral response vaccines would be cost effective. against the gp160 protein from HIV-1 clone LAV. Conclusions: Our findings showed that the two poxvirus vec- tors induced in mice a broad cellular immune response AIDS Vaccine 2006 183 Poster sessions_revGJ 14/8/06 16:34 Page 184 Amsterdam, Netherlands 29 August–1 September 2006 P09D-05 P09D-06 Improving the stability of a conditional-live virus Conditional-live HIV-1 and SIV variants as a by removal of the TAR region novel strategy toward a safe live attenuated AIDS vaccine and as a tool to study virus replication M Centlivre, B Klaver, AT Das, and B Berkhout AT Das, B Klaver, M Vink, A Harwig and B Berkhout Department of Human Retrovirology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands Department of Human Retrovirology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands Background: Live-attenuated virus vaccines have shown promising results in macaques, but safety concerns remain Background: Live attenuated virus variants have shown great for its application against HIV, as the vaccine strain could promise as AIDS vaccine in the SIV macaque model system. revert over time to a virulent and pathogenic phenotype. A However, since continued replication after vaccination can novel genetic approach for an HIV vaccine is the construction lead to the selection of faster-replicating pathogenic virus of a conditional-live HIV-rtTA virus, whose replication is variants, similar live attenuated HIV variants are considered strictly dependent on the presence of doxycycline (dox). In unsafe for use as vaccine in humans. this new approach, we can limit virus replication to the Objectives: As a novel strategy toward a safe live attenuated extent that is needed to mount a protective immune response, AIDS vaccine, we designed a conditional-live HIV-1 variant after which replication is stopped and subsequent virus evo- that replicates exclusively in the presence of an exogenously lution blocked. To demonstrate the efficacy and safety of this administered effector. Upon vaccination with such a virus, approach in macaques, we generated a SIV-rtTA virus that replication can be limited to the extent needed to mount a replicates exclusively in the presence of dox. In SIV-rtTA, we protective immune response by transient effector administra- replaced the TAR-Tat axis with the components of the dox- tion. Subsequent effector-withdrawal will halt replication and inducible Tet-system. However, there may remain a concern prevent evolution. To demonstrate the efficacy and safety of about the restoration of the TAR-Tat axis. such a vaccine, we constructed a similar SIV variant that can Objectives: To improve the genetic stability of SIV-rtTA and to be tested in macaques. investigate possible secondary non-transcriptional functions of Methods: We constructed conditional-live HIV-1 and SIV vari- TAR in virus replication, we truncated the TAR hairpin. ants (HIV-rtTA and SIV-rtTA) in which the viral TAR-Tat sys- Methods: TAR deletions were introduced in the LTR-tetO-pro- tem for transcriptional activation was functionally replaced by moter region of SIV-rtTA. The effect of such mutations on pro- the E. coli-derived Tet-On system for doxycycline(dox)- moter activity and on virus replication was studied in transient inducible gene expression. The safety of the HIV-rtTA vaccine transfections with LTR-tetO-luciferase constructs and in virus- candidate was further improved by the incorporation of muta- cell cultures with SIV-rtTA constructs, respectively. tions that make Envelope activity dependent on T20, and by Results: We partially or completely deleted the TAR hairpin the complete removal of the TAR hairpin. in an LTR-tetO-luciferase plasmid. We tested the transcrip- Results: The designer HIV-rtTA virus was found to replicate tional activity of these constructs in 293T and C33A cells in in a strictly dox-dependent manner in Tcell lines, in primary the presence or absence of dox and/or Tat. This test indicat- blood cells and in ex-vivo cultures of human lymphoid tissue. ed that neither partial deletion of TAR nor its complete Incorporation of a T20-dependent env gene resulted in an removal impaired gene transcription. Moreover, all these pro- HIV-rtTA variant that replicates exclusively in the presence of moter variants are strictly dox-dependent and Tat-indepen- both dox and T20. Moreover, we constructed an efficiently- dent. The TAR deletions are currently being introduced in the replicating HIV-rtTA variant in which the TAR hairpin was SIV-rtTA proviral genome to produce viral stocks. Viral repli- completely removed. These T20-dependent and TAR-deleted cation kinetics will be determined in T cell lines and Rhesus HIV-rtTAs may be safer variants of the vaccine virus. A sim- macaque PBMCs. ilarly-constructed SIV-rtTA variant also replicates in a strict- Conclusions: Deletion of TAR does not affect the transcrip- ly dox-dependent manner in T-cell lines. tional activity of the dox-dependent LTR-tetO-promoter of Conclusions: These conditional-live HIV-1 and SIV variants SIV-rtTA. These results are promising for further optimiza- may represent improved vaccine strains because their replica- tion of the SIV-rtTA vaccine strain. This study should yield a tion can be turned on and off at will. Moreover, these vari- safer version of the dox-controllable SIV-rtTA vaccine strain, ants can be used as a tool to study virus biology. E.g., the which will be tested in macaques for its efficacy and safety. conditional-live SIV is an ideal tool to study the correlates of protection upon vaccination in macaques, since both the time of replication (period of doxadministration) and the level of replication (by varying the dox-dose) can be controlled. 184 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 185 Amsterdam, Netherlands 29 August–1 September 2006 P09D-07 Background/Aims: Important biological differences exist in various pox-based vaccine vectors that may have an impact Expression of HIV-1 Gag proteins from a replication on their clinical application. Human gene profiling analysis competent coxsackievirus vector of two attenuated vaccinia virus vectors, NYVAC and MVA, have revealed similarities but also clear differences in immunomodulatory genes such as IL-7, IL-1A, IL-8 and IL- J Miller, O Yang, Y Geng, H Ng, and P Krogstad 15 (increased in MVA-infected cells) while apoptotic path- ways which may favour cross-presentation are increased by University of California Los Angeles, Los Angeles, CA, USA NYVAC infection (Guerra et al, J Virol 2006). Methods: A DNA-prime (0, 4 weeks), pox-vector boost (20, Background: Nearly all infections with Human 24 weeks) immunization schedule identical to ongoing Phase Immunodeficiency Virus 1 (HIV-1) result from exposure to I/II clinical trials with a NYVAC vector expressing in a single the virus at mucosal sites, but HIV-1 vaccine candidates gen- locus clade C immunogens, was used with MVA or NYVAC erally have demonstrated poor mucosal immune responses. recombinants expressing in the same locus (the TK region) Env and Gag/Pol/Nef antigens corresponding to the SHIV Mucosal tissues are the normal site for infection and replica- 89.6p tion of coxsackieviruses and other enteroviruses, which can virus used for challenge 8 weeks post immunization. The elicit protective mucosal immune responses. This property study consisted of 3 groups of 7 animals each, with group 1: makes this class of viruses potentially attractive candidates MVA boost, group 2: NYVAC boost, group 3: controls. for use as recombinant HIV-1 vaccine vectors. Results: Immune responses were measured by homologous Objectives: To insert HIV-1 Gag protein coding sequences pools of SIV/HIV-1 peptides (15mers overlapping by 11 into the Coxsackievirus B3 (CVB3) genome and demonstrate amino acids) in IFN-γ, IL-2 and IL-4 ELISPOT assays. both Gag expression and presentation via the class I pathway. Marked boosting of IFN-γ as well as IL-2 and IL-4 HIV/SIV- Methods: A DNA plasmid containing the CVB3 genome specific responses following recombinant MVA or NYVAC (pMSK1) was used as the backbone for recombinant viruses. immunizations revealed similar kinetics induced by either HIV-1 sequences were amplified and ligated into a cloning poxvirus vector. Following challenge, IFN-γ, IL-2 and IL-4 site located just after the IRES at the 5′ end of the CVB3 responses were similar between the immunization groups. genome. Plasmids were then transfected into cell lines and Conclusions: Suppression of virus loads, preservation of cellular lysates were prepared for immunoblot analysis and CD4+ T-cells and survival (100% in both groups of vaccinees passage in culture. Chromium release assays were performed versus 30% in controls one year post-infection) were similar using CaCO cells as targets for lysis by CD8+ T effector cell with both poxvirus vectors. Subtle changes in cytokine clones derived from HIV-1-infected patients. responses in the immediate post-exposure period may require Results: Vectors containing the HIV-1 capsid protein (CA) extended follow-up. Possible species specificity of the were genetically unstable and deleted the HIV sequence after poxvirus vectors may require larger Phase II human trials to brief passage in tissue culture. Matrix (MA) protein was sta- clearly distinguish differences between poxvirus vectors with bly expressed for at least 10 passages in culture as detected by a broad panel of diverse cytokine read-outs. immunoblot. Infection of cells with the MA expressing vector rendered cells susceptible to CD8 T cell lysis by clones spe- cific for the SLYNTVATL (SL9) epitope within HIV-1 MA. P09D-09 Conclusions: These initial studies suggest the feasibility of a CVB based vector that could be delivered orally to prevent or Immunogenicity of recombinant measles virus mitigate the impact of HIV-1 infection associated with the Edmonston Zagreb (rMVEZ) vaccine vector gastrointestinal tract. expressing multiple antigens of HIV-1 M Liniger, A Zuniga, S Weibel and HY Naim P09D-08 Berna Biotech LTD a Crucell company and Etnavax, Bern, Head-to-head comparison of immunogenicity and Switzerland efficacy in macaques of MVA and NYVAC vectors expressing HIV-1 Env and SIV Gag/Pol/Nef Background: The outstanding safety and efficacy record of the MVEZ vaccine makes this virus an attractive vector for 1 1 1 P Mooij , S Balla-Jaghjoorsingh , N Beenhakker , P van the expression of heterologous antigens and thus for the Haaften1, I Baak1, I Nieuwenhuis1, CE Gómez3, JL Nájera3, development of recombinant vaccines. Using the Edmonston MV Jiménez 3, R Wagner, 2, M Esteban3, and JL Heeney1 B strain, we have previously established MV vectors that can simultaneously express multiple HIV genes. 1 Biomedical Primate Research Center, Rijswijk, The Netherlands; Objective: To proof the principle of the Edmonston Zagreb 2 Institut für Medizinische Mikrobiologie und Hygiene der vaccine strain in humans. Universität Regensburg, Germany; 3 Centro Nacional de Methods: The stability, propagation and potency of rMVEZ- HIV were tested in cell cultures. Immunogenicity was tested Biotecnologia, CSIC, Madrid, Spain in transgenic mice susceptible to MV. AIDS Vaccine 2006 185 Poster sessions_revGJ 14/8/06 16:34 Page 186 Amsterdam, Netherlands 29 August–1 September 2006 Results: Recombinant measles viruses (rMV) based on the yield was evaluated by fluorescence polarization, size Berna Biotech Edmonston Zagreb vaccine strain (MVEZbv, exclusion chromatography in denaturing conditions and see poster: Zuniga et al) expressing antigens of HIV-1 clade B SDS-PAGE. Antigenic properties of each cross-linked com- (rMVEZbc-HIV) from single or double positions of the plex were investigated with antibodies directed against MVEZ genome were efficiently stable on human diploid cells CD4i epitopes by surface plasmon resonance and (MRC-5). rMVs expressing single HIV-antigens did not show compared to non-covalent complex. any delay in their growth kinetics. The propagation kinetics Results: High proportion of cross-linkage was obtained with of recombinant viruses expressing simultaneously multiple glutaraldehyde, and cross-linked complexes were partially antigens (e.g. Env, Gag–Pol) were slightly slower, but the pro- capable to bind antibodies directed to CD4i epitopes. The use duction process of these viruses was essentially similar to the of BS3 cross-linker led to more than half coupling but cross- parental MVEZ vaccine strain and yielded comparative end linked products did not recognize conformational antibodies. titers. Different immunization regimens and formulations Photo-labeling approach yielded a cross-linking of approxi- (e.g. recombinants expressing single or multiple antigens or a mately half coupling, and it appeared that CD4i epitopes mixture of two recombinant viruses) were systematically were preserved and exposed on cross-linked complexes. All compared in transgenic mice susceptible to MV. The rMVEZ- these complexes are being evaluated in rabbits for their abil- HIV vaccine candidates induced potent humoral and cellular ity to induce broadly neutralizing antibody responses. immune responses against MV and HIV antigens. In addi- Conclusions: The generation of stable gp120-CD4 mimic tion, we show that MVEZ-HIV boosted anti-HIV antibodies complex based on CD4 mimic high affinity by photo labeling indicating that antibody production was not hampered by represents potential in AIDS vaccine application. pre-existing immunity towards the vector. Conclusion: The results indicate the potential of rMVEZbc- HIV for use as a vaccine against AIDS, while retaining its effi- cacy against measles, therefore rendering it multivalent. P09D-10 Development of covalent HIV-1 gp120-CD4 mimic complexes for vaccine application L Martin1, B Heyd1, Y Sun2, G Martin1, E Kan2, P Joly1, P Kessler1, A Menez1, JB Ulmer2, C Vita1, SW Barnett2, and IK Srivastava2 1 Department of Protein Engineering and Research, CEA Saclay, 91191 Gif-sur-Yvette, France; 2 Department of Immunology and Infectious Diseases, Chiron Vaccines, Emeryville, CA, USA Background: Scorpion-toxin mimics of CD4 have been shown to possess CD4-like affinity for gp120, and to induce conformational changes in the HIV-1 envelope (Env) that expose CD4-induced (CD4i) epitopes. Immunizations with gp120-CD4 complexes have shown to induce broadly neu- tralizing antibody responses directed towards CD4i epitopes. However it has been a challenge to demonstrate that the neu- tralizing activity is primarily because of anti-Env antibodies, and not because of anti-CD4 antibodies. In the context of vaccine development, anti-CD4 antibodies may cause autoimmune disorders, therefore undesired. Objectives: To develop stable cross-linked complexes between gp120 Env and CD4 mimic where CD4i epitopes are preserved and properly exposed. Methods: The generation of covalent complexes was per- formed with gp120 derived from SF162 an R5 isolate, and different CD4 mimics according to the coupling condi- tions. Several cross-linking approaches were used: i) glu- taraldehyde, ii) homobifunctional N-hydroxysuccinimide ester cross-linker, and iii) photo-labeling. The coupling 186 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 187 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 10: ANIMAL MODELS P10-01 P10-02 Immunogenicity of a multiepitope multivalent DNA induced Th1 biased immune responses are DNA vaccine against HIV-1 in cynomolgus more effective against the pathogenic effects of macaques SHIV challenge than the broader and stronger immune responses induced by protein F Martinon1, I Picq1, H Moussu1, F Le Goff1, R Sikut2, K Kaldma2, H Remme2, M Ustav2, I Stanescu3 and R Le G Koopman1, D Mortier1, S Hofman1, N Mathy2, Grand1 M Koutsoukos2, L.L. Thomsen3, P. Ertl3, G Voss2 and JL Heeney1 1 Service de Neurovirologie, CEA, Fontenay aux Roses, France; 2 FIT Biotech Oyj Plc Eesti AS, Tartu, Estonia; 3 FIT Biotech Oyj Plc, 1 Department of Virology, Biomedical Primate Research Center, Tampere, Finland Rijswijk, the Netherlands; 2 GlaxoSmith Kline Biologicals, Rixensart, Belgium; 3 GlaxoSmithKline Biopharmaceuticals CEDD Background: HIV vaccine candidates are expected to elicit Biology, Stevenage, UK potent T cell responses against broad spectrum of epitopes in addition to neutralizing antibodies (Ab). Plasmids encoding Objectives: HIV/AIDS vaccines should aim at inducing for HIV proteins have already been shown to induce efficient broad immune responses, including neutralizing antibodies T cell responses. as well as CD4 and CD8 responses. Here we evaluated HIV- Objectives: To evaluate the immunogenicity of a DNA vaccine specific immune response profiles elicited by either single or in cynomolgus macaques and to study the effect of skin electro- combined modality immunization with DNA and recombi- poration after intradermal (i.d.) injection of the vaccine. nant adjuvanted proteins in a rhesus monkey study. Methods: We have designed MultiHIV antigen, which is a Furthermore, the efficacy of the vaccines against SHIV 89.6p recombinant fusion protein that includes complete sequences challenge was assessed. of Rev, Nef, Tat, p17 and p24 proteins from HIV-1, and a Methods: Five groups of 6 Rhesus monkeys were immunized stretch of tandemly arranged previously described CTL epi- at week 0, 8, 16, 24 with either a DNA plasmid encoding tope-rich areas encoded by pol and env genes. The vector was HIV-1 env, tat, and SIV nef genes, via gene gun administra- introduced via i.d. injection into 8 cynomolgus macaques (1 tion, or with HIV-1 Env, NefTat, and SIV Nef proteins for- mg per animal, 3 times). For enhancement of plasmid DNA mulated in AS02A adjuvant via intramuscular uptake, electroporation of the injection site in skin was per- administration. Groups received either 4 DNA or protein formed for 4/8 vaccinated animals. Four additional macaques immunizations or 2 DNA followed by 2 protein immuniza- were vaccinated with an empty vector control under the same tions and vice versa. A control group received empty control conditions. Immune responses were analysed by using ELISA DNA plasmid plus AS02A adjuvant only. Eight weeks after to detect antigen specific Ab and interferon (IFN)-γ Elispot the last immunization, animals were challenged with 20 method to detect peptide specific T cells. MID of a pathogenic cell-free SHIV stock. 50 89.6p Results: After i.d. injections without electroporation, only 2/4 Results: IFN-γ and IL-2 ELISpot responses were detected in animals elicited Ab responses whereas, all 4 animals of the all groups vaccinated with HIV antigens. Animals having electroporated group showed Nef and Gag specific Ab received DNA only mounted such responses predominantly responses. IFN-γ T cell responses specific for all the plasmid- against HIV-1 Env and SIV Nef, while no IL-4 or antibody encoded proteins were detected in all the vaccinated animals. induction was detectable. In contrast, protein immunized ani- Electroporation induced strong T cell responses that were mals developed good IFN-γ, IL-2 and IL-4 responses as well still detected 6 months after the third injection. We have as antibodies against HIV-1 Env, Nef and Tat and SIV Nef. focused on a strong Rev specific IFN-γ T cell response. We The same pattern was seen in animals that had received both have shown that it was mediated by CD8 T cells and that the DNA and protein, and the presence of IL4 and antibody epitope was included within position 62–72 which overlaps responses coincided with protein immunizations. All protein several human CD8 T cell epitopes. immunized animals had strong neutralizing responses against Conclusions: We have designed a single plasmid that encod- the homologous SHIV-W6.1D. Some neutralization of ed several HIV proteins. Vaccination of macaques with this SHIV89.6p was seen in the two groups that had received plasmid induced Nef and Gag specific Ab and long lasting T adjuvanted proteins throughout the study or at the end. cell responses, including CD8 T cell responses. The mapping While all animals became infected after challenge, virus was of one of the recognized epitope shows that the vaccine controlled in 5 of the 6 animals that had received DNA only induced T cells were relevant with the human anti-HIV CD8 or protein/DNA. In contrast, in the two groups that had T cell responses. received protein immunizations last or only protein, control of virus replication was seen in only 2/6 animals. AIDS Vaccine 2006 187 Poster sessions_revGJ 14/8/06 16:34 Page 188 Amsterdam, Netherlands 29 August–1 September 2006 Conclusions: While protein immunization appeared to ease progression, and represent an additional valuable mark- induce the broadest immune responses, DNA seemed to er for evaluating vaccine efficacy. Vaccination can induce confer more solid suppression of virus load after long-term survival independent of the Mhc genotype and thus pathogenic SHIV challenge. independent of the expected disease outcome. P10-03 P10-04 Contribution of major histocompatibility complex Protective efficacy of multiprotein genetic vaccine (Mhc) class I genotypes to set point viral load, in the SIV-macaca animal model survival time, and immunization efficacy in Rhesus macaques F Titti1, MTeresa Maggiorella1, L Sernicola1, F Crostarosa1, R Belli1, MR Pavone-Cossut1, I Macchia1, U Sauermann1, C Stahl-Hennig1, RA Siddiqui2, M Franz1 , S Farcomeni1, PK Tenner-Racz2, P Racz2, B Ensoli 1 YS Suh3, YC Sung3 and G Hunsmann1 1 National AIDS Center, Istituto Superiore Sanità, Rome, Italy; 2 1 German Primate Center, Goettingen, Germany; 2 Leibniz for Department of Pathology Bernhard-Nocht-Institute for Tropical Age Research, Jena, Germany; 3 Pohang University of Science Medicine, Hamburg, Germany and Technology, Pohang, Republic of Korea Background: Genetic vaccines expressing structural and/or Background: SIV-infected rhesus macaques represent the regulatory genes (alone or in combination) showed promises most important animal model for preclinical vaccine trials. for combating AIDS in the macaque animal model, suggest- An improvement of these studies would be the use of well- ing that a vaccine approach combining a wide range of anti- defined monkeys showing predictable patterns of disease pro- gens (structural and regulatory) could have a better chance to gression upon infection to adjust to the diverse patterns of achieve a protective immunity like that one observed with the immune responses observed in HIV-infected humans. live attenuated virus approach Objective: To evaluate the contribution of Mhc region con- Objectives: The protective efficacy of a systemic figurations to disease progression and vaccination efficacy DNA/SFV/MVA vaccine combining structural and regulatory genes (Gag, Pol, Tat, Rev, Env and Nef) of SIV was Methods: Mhc class I genotypes were analysed in a cohort of mac251 79 naive SIV-infected monkeys and 31 immunised monkeys, investigated in the cynomolgus monkeys upon mucosal chal- lenge with SIV . most of them treated with various SIV adenovirus vaccine mac251 preparations. We identified 17 Mhc region configurations Methods: The monkeys were primed with DNAs (i.d., week encoding at least 4 to 11 expressed class I genes per configu- 0), boosted with SFVs (s.c., weeks 8 and 16) and finally ration by cloning, sequencing of cDNA clones, development boosted with MVA (i.m., week 24). DNA and viral vectors of novel typing techniques for 35 Mhc class I sequences and expressed a wide range of SIV proteins (Gag, Pol, Tat, Rev, segregation analysis. For segregation analysis another 150 Nef and Env). At week 32, monkeys (4 vaccinated, 4 control akin monkeys were typed. and 4 naive) were intrarectally challenged (50 MID50) with Results: As one configuration was present in 70% of the SIVmac251. Humoral and immune responses (Abs titers, monkeys, we evaluated disease progression of “quasi-Mhc proliferation, IFN-γ/ELISPOT) were determined. Viral loads class I identical” monkeys. We identified 8 genotypes associ- (plasmaviremia and provirus) at sistemic and mucosal sites ated with rapid, two with median and three with slow disease were quantitated by Real Time RT-, DNA-PCR and in situ progression. Two genotypes did not correlate with disease hybridization analyses. progression indicating that interaction with other host factors Results: The DNA/SFV/MVA immunization elicited IFN- may specifically influenced disease. In 70% of all 79 mon- γ/ELISPOT rather than humoral responses (low anti-SIV Abs keys, the type of disease progression was determined by Mhc titres; absence of neutralizing antibodies) that were broad- genes. In 80% of the monkeys (63) with a Mhc genotype sig- ened following the last boost with MVA. After intrarectal nificantly associated with disease progression, the phenotype challenge, all control and naive monkeys became infected. In coincided with the genotype. Thus, the Mhc genotypes could contrast, 3 out of 4 vaccinees yielded low to undetectable be used to predict disease progression. Consequently, we copy numbers of plasma viral RNA at week 2 after challenge investigated immunised monkeys. Most immunised monkeys (P=0.0002). The abrogation of virus replication coincided had comparable set point viral load as their Mhc matched with the burst of anamnestic IFN-γ/ELISPOT responses to all controls, but survival time was prolonged in many monkeys. vaccine antigens (range P=0.0004–0.010). By week 4, the Independent of the Mhc genotype, and the expected type of protected monkeys had become and persistently remained disease progression, 4 macaques had either very low viral (up to 2.5 years after challenge) plasma viremia negative. The loads slightly above the detection limit and/or about 3 times absence of proviral DNA and of active viral replication was prolonged survival. further demonstrated (in situ hybridization, virus isolation, Conclusion: The results of this well-defined cohort demon- DNA PCR) in lymph nodal (inguinal and axillary) and rectal strate that MHC configurations can be used to predict dis- biopsies. This scenario is accompained by stable level of 188 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 189 Amsterdam, Netherlands 29 August–1 September 2006 CD4+ T cells and anti-SIV plasma Abs, by the absence of ants with different levels of pathogenicity regardless to the spontaneous release of anti-SIV Abs, by the decline of T-cell monkeys species used. Our study demonstrates that SHIV- 89.6P is useful for evaluating the efficacy of AIDS vac- responses, and by the low frequency of perforin+ and cy243 granzyme B+ cells in lymph nodes (P=0.0011 and 0.036 cines in cynomolgus macaques utilizing viral load and CD4 respectively). At a later time, IFN-γ/ELISPOT responses to T cell decline as experimental endpoints. vaccine antigens become undetectable on fresh PBMCs but antigen-specific memory T-cells (mostly against Tat and Nef) were still demonstrated in the vaccinated and protected mon- P10-06 keys. In contrast, one vaccinated and progressor monkey, and five out of six progressor control monkeys died because of Optimization of mucosal immunization protocols AIDS. based on HIV-1 virus-like particles Conclusions: The observed scenario indicate a control of viral replication and indicate that this vaccine candidate 1 1 1 through further refinement can represent a strategic tool to FM Buonaguro , ML Tornesello , M Tagliamonte , J 2 3 1 prevent disease onset and dissemination of HIV infection. Hinkula , U Schröder and L Buonaguro (This work was supported through the EU grant QLK2-99- 00871 and by ICAV grant from Ministry of Health) 1 Viral Oncogenesis and Immunotherapy & AIDS Refer. Center, Ist. Naz. Tumori “Fond. G. Pascale”, Naples-Italy; 2 Swedish Inst. Infect. Disease Control, Dept of Virology, Solna, Sweden and P10-05 Karolinska Inst., Stockholm, Sweden; 3 Eurocine AB, Karolinska Science Park, Stockholm, Sweden Viral outcome of Simian-Human Immunodeficiency Virus SHIV-89.6P adapted in Aim: To evaluate and compare the immune response induced cynomolgus monkeys (SHIV-89.6 P ) in a mouse model by an anti-HIV-1 vaccine based on HIV-1 cy243 Virus-Like Particles, expressing a gp120 from an Ugandan HIV-1 isolate of the clade A (HIV-VLP ), injected by differ- F Titti, A Borsetti, S Baroncelli, MT Maggiorella, A S Moretti, L Sernicola, R Belli, B Ridolfi, S Catone, ent administration routes alone or within DNA/VLP prime- boost protocols. S Farcomeni, DRM Negri, A Cafaro and B Ensoli Results: Specific cellular immunity along with a systemic and mucosal IgG and/or IgA response are observed in mice immu- National AIDS Center, Istituto Superiore di Sanità, Rome, Italy nized with HIV-VLP s by the i.p. as well as the i.n. adminis- A tration routes. The induced antibodies show an ex vivo Background: For evaluating HIV vaccine candidates, we neutralizing activity on both autologous and heterologous developed a SHIV-cynomolgus monkey model using SHIV- field isolates. Different DNA/HIV-VLP prime-boost doses 89.6P that was obtained by passages of SHIV-89.6P in cy243, and schedules have been tested by i.n. immunization, show- cynomolgus monkeys. ing additive/synergistic effects on humoral as well as cellular Objectives: To investigated the pathogenicity of the SHIV- responses. The HIV-VLP formulated in a novel mucosal A 89.6P th cy243. adjuvant can be used at 1/10 of the original dose, without Methods: 25 naive or control macaques, which were part of loss of immunogenicity. different experimental protocols inoculated by the intra- Conclusions: The induction of an efficient mucosal response venous route were included. As a measure of pathogenicity is of high relevance given that gastrointestinal and vaginal the relationship between level of plasma viremia, CD4+ T- sites are the main port of entry for the HIV-1 infection. The cell number and disease progression was employed. DNA/VLP prime/boost immunization strategy shows an Moreover, in order to determine the molecular changes effective synergistic immunization effect. Moreover specific upon transmission to the new host, the entire genome from mucosal adjuvants allow to sensibly lowering the immuniza- SHIV-89.6P was sequenced. cy243 tion dose, with a reduction of possible in vivo side effects. Results: Our data demonstrate that SHIV-89.6P caused cy243 The observed ex vivo cross-clade neutralization of primary massive depletion of circulating CD4+ cells within 4 weeks field isolates is promising for the containment in vivo of a from the intravenous inoculum, leading to an irreversible broad spectrum of HIV-1 subtypes in animals immunized immune deficiency in a high percentage of infected animals. with the HIV-VLP s. A Furthermore, the levels of viremia in the infected animals remained detectable over 35 weeks post-infection. Sequencing analysis of the SHIV-89.6P genome revealed cy243 eight amino acids changes compared to the SHIV-89.6P. Conversely, SHIV-89.6P showed 100% identity to the cy243 sequence of the highly pathogenic SHIV-C2/1 derived from passaging of SHIV-89.6 in cynomolgus macaques. Conclusion: This result indicates that in vivo propagation of the SHIV89.6P may result in the selection of viral vari- AIDS Vaccine 2006 189 Poster sessions_revGJ 14/8/06 16:34 Page 190 Amsterdam, Netherlands 29 August–1 September 2006 P10-07 P10-08 Protection of cats by immunization with the The neutralizing antibody response in long-term transmembrane envelope protein p15E of feline non-progressor monkeys infected with simian leukaemia virus (FeLV) immunodeficiency virus (SIVsm) J Denner1, S Langhammer1, J Hübner2, O Jarrett3 and R EM Fenyö1,2, A Laurén1 and R Thorstensson2 Kurth1 1 Lund University, Lund, Sweden; 2 Swedish Institute for 1 Robert Koch Institute, Berlin, Germany, 2 Laboklin, Bad Infectious Disease Control, Stockholm, Sweden Kissingen, Germany, 3 Institute of Comparative Medicine, University of Glasgow, Glasgow, UK Objective: To explore the neutralizing antibody response in long-term non-progressor (LTNP) SIVsm-infected monkeys Background: Recently we described induction of antibodies and the relationship to viral properties such as macrophage neutralising porcine endogenous retroviruses (PERV) or FeLV tropism and CD4 independent use of CCR5. immunising rats and goats with the corresponding trans- Methods: Four cynomolgus macaques showed no signs of membrane envelope protein p15E of these viruses. All sera disease 35–50 months after infection with SIVsm. LTNP was recognised two epitopes, one located in the N-terminal part characterized by a median CD4 decline of -0.3%/month, 3 (designated E1), the other in the C-terminal membrane-prox- viral load in the range of 10 RNA copies/ml plasma and suc- imal external region (MPER, E2) of p15E. E2 is located sim- cessful virus isolation in 31% of attempts. For comparison, ilarly as epitopes recognised by antibodies broadly 12 monkeys that progressed (P) to AIDS were also included neutralising HIV and isolated from HIV-infected individuals in the study. Serum neutralizing activity was tested in the (2F5, 4E10) and there is a limited sequence homology of the plaque reduction assay on GHOST(3) cells. epitopes (FEGWFN in the case of PERV/FeLV and NWFNIT Results: The SIVsm inoculum virus and re-isolates obtained recognised by 4E10). at 2 weeks, 3 or 4 months and later than 1 year were tested Aims: To determine if vaccinating with p15E protein could with autologous sera collected at the same sampling times. protect cats from FeLV infection. All monkeys developed a neutralizing-antibody response to Methods: Cats were immunised with p15E of FeLV-A, with the inoculum virus. In P monkeys, neutralization escape the commercial vaccine Leucogen (p45), or with a combina- could be demonstrated by 3 months post-infection. tion of both and challenged oronasally with infectious FeLV- Neutralization resistant variants also emerged in LTNP mon- A. Neutralising antibodies were measured in infection assays keys, but were much delayed compared with P monkeys. using FeLV-A and uninfected FEA cells. p27Gag antige- Analysis of the different viral parameters in relation to patho- naemia was analysed by ELISA and quantification of genesis showed that early isolates are CD4 independent, provirus load in blood was determined by RTQ-PCR. Titres macrophage tropic and sensitive to neutralization. Thereafter of antibodies specific for p15E and p45 were determined by viruses from LTNP monkeys became less CD4-independent, ELISAs. Epitope mappings were performed using overlapping less macrophage tropic and remained neutralization sensitive peptides. for much longer time than P monkeys. In addition, pooled Results: In none of the cats immunised with Leucogen alone sera from LTNP monkeys showed a broad neutralizing or in combination with p15E p27Gag antigen was detected. capacity, including neutralization of escape variants. Three out of six cats immunised with p15E were p27 nega- Conclusion: The results suggest an important role for neu- tive. All of the nonimmunised cats were p27 positive. tralizing antibodies in controlling viraemia. Although this Although absence of p27 antigenemia is considered as pro- control is transient in the infected host, neutralization resis- tection, provirus load was detected using the highly sensitive tance is relative and variant viruses may be neutralized by a PCR in all animals, also in the animals immunised with broadly cross-neutralizing serum pool. Leucogen. Provirus load and virus load were shown to corre- late inversely with neutralising antibodies. Antisera from ani- mals immunised with p15E detected the E1 and E2 epitopes. Conclusion: This is the first report showing protection from retrovirus infection in vivo by immunization with a trans- membrane envelope protein. However, sterilising immunity was not detected in all animals immunised with p15E or the commercial vaccine or both. These data show that the MPER is an efficient target for immunization against different retro- viruses including HIV. 190 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 191 Amsterdam, Netherlands 29 August–1 September 2006 P10-09 P10-10 Mutagenized region within the first loop of CCR5 Systemic T cell priming in macaques exposed rec- induce high affinity HIV neutralizing antibodies tally to SIV after local application of a nucleotide analogue reverse transcriptase inhibitor L Lopalco1, C Pastori1, G De Mori2, R Longhi2, G Colombo2, A Clivio3 and R Consonni2 M Cranage1, S Sharpe2, A Cope1, C Herrera1, M Dennis2, N Berry3, C Ham3, P Anton4, I McGowan4 & R Shattock1 1 San Raffaele Scientific Institute, Milano, Italy; 2 CNR, Milano, Italy; 3 University of Milano, Milano, Italy 1 Centre for Infection, St George’s University of London, UK; 2 Health Protection Agency, Porton Down, UK; 3 Division of Background: The HIV-1 coreceptor CCR5 and antibodies Retrovirology, NIBSC, UK; 4 Center for HIV and Digestive Disease, against it are relevant for HIV-1 vaccine design. Antibodies to David Geffen School of Medicine, UCLA, Los Angeles, USA the first loop of CCR5 have been identified in HIV-exposed uninfected individuals (ESN) and in HIV-positive non-pro- gressing subjects. Thus, these antibodies may confer resis- Background: Rectal transmission is a significant route for the tance against HIV infection. acquisition of HIV driving the requirement for the develop- Objective: To better define the region recognized by CCR5- ment of rectally applied prophylactic microbicides as a pre- antibodies, a panel of synthetic peptides spanning the CCR5- vention strategy that may be complementary to vaccination. loop1 region, displaying Glycine or Alanine substitutions, Objectives: (1) To determine, in the macaque model, the was assayed for antibody binding with anti-CCR5 antibodies antiviral efficacy of the nucleotide analogue, 9-[(R)-2- from ESN. The mutagenized region of CCR5 was then used (phosphonmethoxy) propyl] adenine monohydrate (PMPA) to immunize mice and chicken. given rectally as a single dose prior to, or shortly after, Methods: Peptides were synthesized by the solid phase F-moc intrarectal challenge with simian immunodeficiency virus method. Binding of human sera was evaluated by ELISA. (SIV). (2) To determine if mucosal exposure to virus primes Mice and chickens were immunized with mutagenized pep- immunity in animals protected from overt infection by tide. Biological activity of antisera to mutagenized peptide locally applied PMPA. was evaluated by inducing CCR5 downregulation on surface Methods: Rhesus macaques received 1% PMPA gel in a sin- + of human CD4 T cells and chemotaxis inibition. Antisera gle dose by atraumatic rectal instillation. Twenty intrarectal were also tested in HIV neutralization. median infectious dose (MID ) of SIVmac32H was used for 50 Results: Aminoacidic substitutions in positions Ala95 and challenge. Peripheral blood mononuclear cells (PBMC) were Ala96 (A95-A96) were found to increase antibody-peptide examined for the presence of virus by PCR for proviral DNA binding in comparison with wild-type peptide (Phe95- and by co-cultivation at regular intervals over 20 weeks. Asp96). This activity was reproduced in mouse and chick- Results: Virus was recovered at every time point tested in 4 of en. Ala95-96 peptide was shown to induce antibodies 4 untreated macaques and 3 of 4 animals given placebo gel. displaying biological activity at very low concentrations. In contrast, 6 of 9 animals given PMPA prior to virus chal- Strikingly, chicken antibodies to Ala95-96 specifically rec- lenge were protected from virus challenge and virus detection ognize human CCR5 molecules, downregulate receptor was intermittent or delayed in 2 other animals. Virus was iso- from lymphocytes, inhibit CCR5-dependent chemotaxis lated on every occasion of testing from 2 of 3 animals where and prevent infection of several R5-viruses. Structural char- gel was administered 2 h after virus challenge. Corresponding acterization by NMR spectroscopy and molecular dynamics results were obtained by PCR analysis. Apparently protected simulations were performed to better understand the struc- animals had no evidence of viral sequestration in lymphoid tural and conformational features of the mutagenized pep- tissues and failed to seroconvert. Protection was associated tide. These studies proved the high flexiblility of isolated with the concentration of PMPA detected in plasma at the epitopes and suggested that Ala95-96 substitutions deter- time of virus challenge. Interestingly, Gag-specific interferon- mine a slightly higher tendency to generate helical confor- γ secreting T cells were detected by ELISpot in 4 of 7 animals mations combined with a lower steric hindrance of the side in which virus was unrecoverable from PBMC at frequencies chains in the peptides. ranging from 144 spot forming cells (SFC)/106 PBMC to 261 Conclusion: This finding may be relevant to induce strong SFC/106 PBMC. and efficient HIV neutralizing antibodies. The chicken model Conclusions: These data indicate that rectal predosing with proves a simple and inexpensive approach that could be PMPA has potential as part of a microbicide strategy and addi- translated into clinical practice. tionally may enable priming, or possibly boosting, of the immune system through exposure to mucosal virus challenge in the absence of infection. This suggests that it may be possible to develop synergistic microbicide and vaccination regimens. AIDS Vaccine 2006 191 Poster sessions_revGJ 14/8/06 16:34 Page 192 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 11: PROPHYLACTIC VACCINE TRIALS P11-01 P11-02 Out of range laboratory values: a major reason for Vaccine-induced cytotoxic T lymphocyte exclusion from a Phase I HIV vaccine clinical trial responses in uninfected adult ugandan volun- teers enrolled in a Phase i double-blind multi- WG Jaoko 1, GO Manyonyi 1, KM Bhatt 1, AO Anzala 1, clade HIV-1 DNA plasmid vaccine trial H Ogutu1, R Ndambuki1, S Wakasiaka1, R Malogo1, (VRC-HIVDNA009-00-VP) BM Farah1, S Ogola1, M Oyaro1, J Indangasi1, J Onyango1, K Bosire, JR Nyange1, H Thomson2, M Eller1, L Eller1, M Opollo1, B Ouma1, P Oballah1, J Ndinya-Achola1, J Gilmour 2, S Than 2, L Galley 2, M Robb2, F Wabire-Mangen1, B Graham4, W Komaroff 2, C Kambili 2, P Fast 2 and JJ Bwayo1 D Birx 2, M de Souza3 and J Cox2 1 Kenya AIDS Vaccine Initiative (KAVI), University of Nairobi, 1 Makerere University Walter Reed Project, Kampala, Uganda; 2 Nairobi, Kenya; 2 International AIDS Vaccine Initiative (IAVI), U.S. Military HIV Research Program, Rockville, USA; 3 Armed New York, USA Forces Research Institute of Medical Sciences, Bangkok, Thailand; 4 Vaccine Research Center, NIH, Bethesda, USA Background: The Kenya AIDS Vaccine Initiative (KAVI) is conducting a phase I clinical trial of a recombinant multi-clade Objectives: To determine the frequency of cytotoxic T lym- HIV adenoviral vector vaccine given either alone or as a boost phocyte (CTL) responses in Ugandan HIV seronegative following priming with a multi-clade HIV-1 DNA plasmid recipients of a multicade, HIV-1 DNA Vaccine (VRC- vaccine. Volunteers are recruited from healthy, HIV-uninfect- HIVDNA009-00-VP). ed adults, who are at low risk of acquiring HIV infection. Methods: Thirty-one subjects were enrolled in a Phase I Objectives: To identify reasons for screening failure among randomized, double blind, placebo controlled trial to assess volunteers of a phase 1 HIV vaccine trial in Nairobi. the safety of a multiclade HIV-1 DNA Plasmid Vaccine Methods: Interested volunteers are assessed for risk of HIV HIVDNA009-00-VP (Vaccine Research Center, NIH infection during one-to-one counselling sessions. The volunteers Bethesda). Two of the participants were excluded for are then asked to give their written informed consent after they missed visits; group A (n=14) received placebo, group B have passed a test of understanding, and are subsequently (n=15) received HIVDNA009-00-VP. The injection regimen screened for the trial. Screening involves medical history, physi- was at 0, 4, and 8 weeks. A standard chromium release cal examination, haematology, biochemistry, Hepatitis B, assay was performed at weeks 0, 6, 10, 24, 36, and 48 on Hepatitis C, syphilis and HIV, and urinalysis testing. Individuals freshly isolated peripheral blood mononuclear cells stimu- who are ineligible on health grounds are referred to relevant lated in vitro with vaccinia virus expressing HIV-1 (IIIB) specialist clinics for follow-up and appropriate management. Env/Gag/Pol and used as effectors with autologous EBV- Results: Eighty-seven (52 males and 35 female) volunteers have transformed B cells infected with recombinant vaccinia so far been screened for the trial. Twenty-eight (20 male and 8 viruses expressing clade B (IIIB) Env and LAI Gag antigens as targets. All assays were conducted on female) of these failed screening. Seventy-five percent (75%) of blinded samples. the screening failures were due to out-of-range laboratory Results: No CTL responses were observed at baseline. Seven results. A majority (>50%) had absolute neutropenia. These of the 15 (47%) vaccinated individuals demonstrated CTL were mostly mild (1,000–1,500 cells/µl). Three volunteers had activity to HIV B Env however no HIV B Gag responses elevated alanine aminotransferase (>2.5 times upper limit of were seen. Positive CTL responses were observed as early as reference range), 5 had positive or indeterminate HIV ELISA 4 weeks following the second injection in 2/15 (13%) par- test, and 4 were positive for Hepatitis B surface antigen. One ticipants. The greatest number of responses at one time- of the Hepatitis B surface antigen positive volunteers was also point, 5/15 (33%), occurred 28 weeks following the third pregnant. Only 2 volunteers failed screening due to medical injection. 2/15 (13%) volunteers had positive responses at history or abnormal physical examination: one had bronchial multiple time points. All CTL responses were CD8 restrict- asthma and the other had severe seborrhoeic dermatitis. One ed. No responses were seen in the placebo arm (0/14). volunteer was excluded because of recreational drug use. Conclusion: A multi-clade HIV-1 DNA vaccine induced Conclusion: Out-of-range laboratory results are the main CTL activity against HIV Env vaccine antigens in 47% of cause of screening failure for this phase 1 HIV vaccine clini- the immunized individuals and as early 4 weeks following cal trial at our site. Lack of local reference laboratory ranges the second vaccination. This study demonstrates that probably contributes to this large number of laboratory- immunization of human subjects with DNA induces HIV- based screening failures. The use of locally generated specific CTL at levels comparable to other HIV vaccine laboratory reference ranges may reduce this high number of candidates. Future strategies of second-generation DNA screening failures. A study of local laboratory reference products coupled with Adeno vector boosting may increase ranges is currently under way. the immunogenicity of this product. 192 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 193 Amsterdam, Netherlands 29 August–1 September 2006 P11-03 P11-04 Cross sectional volunteers’ satisfaction survey in Frequent large blood draws and decrease in Phase III prime-boost HIV preventive vaccine trial hemoglobin levels among AIDS vaccine trial in Thailand participants in India T Woratanarat 1, PA Morgan 1, RM Paris 1, N Sirijongdee 1, S Mehendale 1, S Sahasrabudhe 1, M Kakade 1, M Benenson1, P Singharaj 1, P Pitisutthitham 2, S Rerks- A Shrotri 2, M Thakar 1, S Sahay 1 and J-L Excler 2 Ngarm 3 and JH Kim 1 1 National AIDS Research Institute, Pune, India; and 2 International 1 Department of Retrovirology, US Army Medical Component, AIDS Vaccine Initiative, New Delhi, India Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand; 2 Faculty of Tropical Medicine, Mahidol University, Objective: Safety and immunogenicity assessments in Phase Bangkok, Thailand; 3 Department of Disease Control, Ministry of I and II vaccine trials require frequent blood withdrawal Public Health, Bangkok, Thailand. for tests on serum, plasma and cells and specimen reposi- tories. We evaluated the impact of repeated blood draws on Background: A vaccine for the prevention of HIV infection hemoglobin levels following concerns by local community remains an urgent need as part of the efforts to control the HIV and scientific/ethics reviewers on the schedule and volume pandemic. This Phase III trial using ‘prime-boost’ vaccine strat- of blood draws, especially among women, on the back- egy with ALVAC-HIV (vCP1521) and AIDSVAX gp120 B/E ground of prevalent nutritional deficiencies and morbidi- has enrolled and vaccinated 16,000 HIV uninfected volunteers ties in Indian populations. in Thailand. Volunteers’ satisfaction is believed to be one of Methods: The ongoing IAVI A001 dose-escalation Phase I key factors for achieving good retention of volunteers. Here, trial aims at evaluating the safety and immunogenicity of we conducted a site survey of study volunteer satisfaction. tgAAC09, adeno-associated virus based HIV-1 subtype C Objectives: To assess volunteers’ satisfaction with participat- AIDS vaccine [Targeted Genetics, USA] in 30 healthy adult ing clinical sites in terms of service, staff, facilities, and vol- volunteers in Pune, India. At screening 25 ml and at enroll- unteers’ attitude to the trial. Factors influencing volunteers’ ment as well as 2, 4, 12, 24, 36 and 52 weeks; 80 to 100 ml satisfaction were explored. blood is drawn. Hematinics are prescribed on noticing hemo- Methods: A cross-sectional survey was conducted at 4 of 8 globin level of below 12 gm%. We determined differences clinical sites using an anonymous voluntary sample and a and slopes using hemoglobin levels at various visits. standardized questionnaire during October–December 2005. Results: The mean baseline hemoglobin levels among 16 Data collected from each volunteer included gender, visit male and 14 female volunteers were 15 [12.7–17.2] and 13 number, a 5-point satisfaction scale scored for 4 aspects from [11.3–14.7] gm%, respectively. Average cumulative fall of 10 questions, along with comments. Descriptive statistics, hemoglobin from screening visit to each subsequent fol- parametric and/or non-parametric tests were used as appro- low-up visit was 0.14 (P=0.298), 0.56 (P<0.001), 0.85 priate, as well as multiple logistic regression analysis to (P<0.001), 0.81 (P<0.001), 0.96 (P<0.001), 1.02 explore potential factors. (P<0.001) and 1.28 (P=0.03) gm%. The range of fall Results: There were 532 participants from 4 sites located in among males was 1.5 to 3.3 gm% and 1.2 to 2.6 gm% in both study provinces. 54.8% were men, and 85% were in female participants. However, this drop was of no clinical the immunization phase of the study. All sites were evalu- significance. Cumulative slope for hemoglobin drop was ated with average scores more than 80% (range higher in males (-0.0050 versus -0.0046, P=0.506). 81.5–89.3), but there were significant differences in total Irrespective of gender, volunteers with higher baseline hemoglobin showed higher drop than those with lower val- score among sites (P<0.05) and between provinces ues. Six of the nine hematinics recipient women showed (P<0.01), consistent across most of sub-categorical scores. Regression analysis showed potential model to predict sum rise in hemoglobin levels. Among men hemoglobin values of score in each aspect varying from 68–85% (R2) includ- remained above 12 gm% throughout. ing site, gender, and several potential interactions between Conclusions: During follow-up evaluation statistically sig- variables. No difference in total score and subgroups nificant drop of hemoglobin was observed in majority of between those in the vaccination or follow-up phases was volunteers. Repeated blood draws may have contributed to detected. Interestingly, significant differences in scores this drop. Contribution of other factors such as bleeding, were strongly associated with gender throughout the analy- nutritional deficiencies, concomitant drugs and parasitic sis (P=0.03). Males were tended to give higher scores than infections did not appear clinically significant. Recruiting females in the study, which may suggest gender related dif- volunteers with high hemoglobin levels and prescribing ferences in either treatment of volunteers or perception of hematinics with emphasis on nutritional and adherence treatment of the volunteers. counseling are recommended to avoid hemoglobin level Conclusion: In conducting large-scale HIV vaccine clinical drops below 10 gm%. trials, investigators should periodically evaluate factors that may impact volunteers’ satisfaction and indirectly affect trial retention. Well-planned time series satisfaction survey should be beneficial. AIDS Vaccine 2006 193 Poster sessions_revGJ 14/8/06 16:34 Page 194 Amsterdam, Netherlands 29 August–1 September 2006 P11-05 individuals. Our data suggests that cell-mediated immune responses may be underestimated using CPT separated cells Optimization of cell separation procedures for in these populations. determination of HIV-1 vaccine-induced cell- mediated immune responses in Swedish and Tanzanian individuals P11-06 1 2 3 1 C Nilsson , S Aboud , B Hejdeman , K Karlén , Detection of circulating HIV vaccine plasmid DNA 2 2 3 1 W Urassa , F Mhalu , E Sandström and G Biberfeld in immunized individuals 1 Microbiology and Tumorbiology Center, Karolinska Institutet G Engström1, A Bråve 1, M Westman 1, and Swedish Institute for Infectious Disease Control, Stockholm, L Gudmundsdotter 1, E Sandström 2, J Albert 1 and Sweden; 2 Muhimbili University College of Health Sciences, B Wahren1 Dar es Salaam, Tanzania; 3 Venhälsan, Dept. Infectious Diseases, Karolinska University Hospital, Stockholm, Sweden 1 Swedish Institute for Infectious Disease Control, 2 Karolinska University Hospital and Karolinska Institute, Stockholm, Sweden Background: A phase I HIV-1 vaccine trial using a multigene, multiclade HIV-1 plasmid DNA prime MVA boost is ongoing Background: We are currently conducting a Phase I clinical in Stockholm, Sweden and a phase I/II trial using the same trial where forty healthy volunteers are immunized with a vaccines will start this year in Dar es Salaam, Tanzania. multigene, multiclade plasmid vaccine encoding nine differ- Excellent recovery and high viability of peripheral blood ent proteins of HIV-1. Different modes of administration, mononuclear cells (PBMCs) is essential for reliable assess- doses and use of granulocyte-macrophage colony stimulat- ment of cell-mediated immune responses. ing factor, GM-CSF, are being investigated.The study proce- Objectives: To evaluate and optimize cell separation proce- dures include a broad range of clinical laboratory tests. Two dures for IFN-γ ELISPOT. months after the third immunization, a HIV RNA quantita- Methods: Blood was collected from Swedish (n=15) and tive PCR was performed to confirm the noninfected status Tanzanian (n=20) healthy blood donors. PBMCs were of the participants. At this timepoint some of the vaccinees either purified by using heparinized blood collected in showed a slightly elevated number of HIV RNA copies in Vacutainer tubes and standard Ficoll separation or by the Roche Amplicor assay. All individuals tested negative using cell preparation tubes (CPT) according to manufac- for HIV antigen as well as in antibody-based tests. turer´s instructions (Becton-Dickinson). Cell recovery and Objectives: To investigate occurrence of vaccine plasmid in viability were recorded and the isolated cells were used for blood by a plasmid-specific DNA PCR and correlate these IFN-γ ELISPOT testing against a pool of 23 peptides findings to the positive HIV RNA analyses. derived from Cytomegalovirus, Epstein–Barr virus and Methods: Total nucleic acid was extracted from plasma or sera Influenza virus (CEF). Blood samples from five Swedish using Nuclisens Isolation kit (Bio Merieux). A nested PCR was HIV-1 infected long-term non-progressors were also tested performed using specific primers designed for detecting the vac- for reactivity to HIV-1 using Env (gp41, gp120) Gag (p17 cine plasmids. A diagnostic HIV-1 DNA PCR directed against and p24) and RT peptide pools representing the DNA vac- HIV genes not present in the vaccine was performed as well. cine sequence. Results: Samples were collected two weeks or two months Results: Mean recovery of PBMC isolated from Swedish after the third and last DNA vaccination. In 10 of 38 vacci- blood donors using standard Ficoll separation and CPT nees, receiving 0.2–0.6 mg of each plasmid, we noted reac- was 1.14 ±0.4 and 1.20 ±0.4 million cells/ml blood, respec- tions corresponding to 20–590 HIV RNA copies in the Roche tively and equivalent levels of IFN-γ ELISPOT reactivity to Amplicor assay. Env and/or Gag encoding plasmids were CEF was recorded irrespective of cell isolation procedure detected in the plasma or serum of the vaccines, but no HIV used. Recovery of PBMCs isolated from Tanzanian volun- virus. A PCR for HIV protease (the protease gene is not teers was 1.58 ±0.6 million cells/ml blood using standard included in the vaccine) was negative in all cases. Later Ficoll separation versus. 1.34 ±0.4 million cells/ml blood repeated antigen and antibody assays have confirmed that the for CPT prepared cells, P=0.0469. The viability of CPT- individuals were not infected. purified cells was low (92.6± 4.8%) compared to the via- Conclusions: Circulating DNA plasmids, used for vaccina- bility of cells processed by standard ficoll separation (95.8 tion, were identified by plasmid-specific PCR as late as 2 ±2.3%, P=0.0081). Furthermore, lower levels of HIV-spe- months after immunization. These plasmids may give positive cific IFN-γ ELISPOT reactivities were recorded for CPT- signals in a standard HIV RNA quantative assay. Other purified cells from Swedish HIV+ individuals as methods must then be used to ensure the noninfected status compared to Ficoll separated PBMCs. of trial participants. Conclusion: Using blood from healthy Swedish volunteers, cells purified by the two cell preparation procedures per- formed equally well in IFN-γ ELISPOT. The use of Ficoll sep- aration was superior to CPT for preparation of PBMCs from healthy Tanzanian blood donors and from Swedish HIV+ 194 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 195 Amsterdam, Netherlands 29 August–1 September 2006 P11-07 P11-08 HIV-specific lymphocyte proliferative responses Sources of volunteer’s perception and decision to determined by a standard thymidine uptake assay participate in the prime-boost HIV vaccine trial and a flow cytometric assay in healthy individuals immunized with HIV-1-DNA/MVA C Namwat 1, N Premsri1, C Khamboonruang 1, S Rerks-Ngarm 1, P Kunasol 1, J Kaewkungwal 2, 2 1 1 1 S Aboud , K Godoy-Ramirez , C Nilsson , K Karlén , S Wattana 3, W Wiriyakijja4, M Benenson5 and J Kim5 B Wahren1, E Sandström3, H Gaines1 and G Biberfeld1 1 Ministry of Public Health – Thai AIDS Vaccine Evaluation 1 Karolinska Institute and Swedish Institute for Infectious Disease Group (MOPH-TAVEG), Thailand; 2 Faculty of Tropical Medicine, Control, Stockholm, Sweden; 2 Muhimbili University College of Mahidol University; 3 Chon Buri Provincial Health Office, Health Sciences, Department of Microbiology and Immunology, Thailand; 4 Rayong Provincial Health Office, Thailand; 5 Armed Dar es Salaam, Tanzania; 3 Venhälsan, Dept. of Infectious Diseases, Forces Research Institute of Medical Sciences (AFRIMS) Karolinska University Hospital, Stockholm, Sweden Background: The world’s largest preventive HIV vaccine trial Background: A phase I HIV immunogenicity study (HIVIS), has enrolled 16,402 volunteers from September 2003 to using multigene, multiclade HIV-1 plasmid DNA prime and December 2005. Information sources that help potential vol- MVA boost, is ongoing in Sweden. Several assays are used for unteers understand the trial and enable them to voluntarily monitoring of immune responses. decide to participate in the project are keys for success. Objective: To determine HIV-1 vaccine-induced lymphopro- Methods: Using a standardized questionnaire, volunteers pro- liferative responses by the Flow cytometric Assay of Specific vided responses about the source of information where they Cell-mediated Immune response in Activated whole blood got information of the trial and that influenced them to join the 3 (FASCIA) and a standard H-thymidine-uptake assay. study. At the time enrollment was completed in December Methods: Volunteers were immunized three times with 2005, the sources that influenced and inspired volunteers to DNA expressing HIV-1 gag, env, rev and rt at months 0, 1 voluntarily join the trial were analysed. The proportions of and 3. They were boosted at 9 months with MVA express- sources were identified at three time points and compared. ing HIV-1 gag, env, and pol (constructed and produced by Results: Overall sources of perception were derived from “other Bernard Moss, NIH and the WRAIR HIV research group). enrolled volunteers” (55.6%), “local health staff” (32.6%), and Coded blood samples collected before vaccination (base- “recruitment team” (21.9%). Over time the proportion of line) and 2 weeks after HIV-1 MVA-vaccination were tested “enrolled volunteers” had significantly increased from 40.2% for lymphoproliferative responses against HIV-1 using to 66.6% (P<0.001), in contrast the “recruitment team” as a Aldrithiol-2 (AT-2) inactivated HIV-1 (2.5 µg/ml) (donated source decreased from 25.6% to 18.9% (P<0.001). The highest by Jeffrey Lifson). For the FASCIA, whole blood diluted 1/8 proportions, as sources of decision making, were “local health in medium was cultured for 7 days in the presence of HIV- staff” (35.6%), screening video (26.9%), and “other enrolled 1 antigen (Test), PHA (Positive Control, PC) and medium volunteers” (22.0%). (Negative Control, NC). Proliferative responses were mea- Conclusion: This finding shows us that in the beginning of the + + sured by detection of CD3 CD4 lymphoblasts and results trial, the local health staff and recruitment team will be the were expressed as percentage stimulation (%S): (100× major source of information and influence. As the number of 3 [Test-NC]/[PC-NC]). For the H-thymidine-uptake assay, volunteers enrolled increases the “enrolled” volunteers will fresh PBMCs were incubated in the presence of medium, become a major source. This data demonstrates that “word of mitogens (2 days) and HIV-1 antigen (6 days), and the mouth” from “enrolled volunteers” is an important and invisi- results were presented as stimulation index (SI). Samples ble power for volunteer recruitment to vaccine trial. from healthy blood donors (n=12) were used to obtain background reactivity levels. + + Results: The mean (±SD) HIV-specific CD3 CD4 T-cell proliferative responses detected by the FASCIA were 0.28 ±0.29%S in the base-line samples and 7.55 ±8.81%S in the 3 HIV-1 MVA post- vaccination samples. Using the H-thymi- dine-uptake assay a mean reactivity of 79.84±98.42 SI was demonstrated 2 weeks after HIV-1 MVA vaccination. The mean reactivity in samples from blood donors was 1.30 ±0.88 SI. Of the 25 vaccinees tested, 24 were reactive by the 3 H-thymidine-uptake assay (SI>5) and 21 by FASCIA (>1.15% reactivity). There was a correlation between the responses measured by the two assays (r: 0.62; P<0.01). Conclusions: Strong specific lymphoproliferative responses were detected in a high proportion of subjects following HIV- 1 DNA/MVA immunization. AIDS Vaccine 2006 195 Poster sessions_revGJ 14/8/06 16:34 Page 196 Amsterdam, Netherlands 29 August–1 September 2006 P11-09 P11-10 Safety and immunogenicity of an alphavirus Screening and enrollment into a Phase I HIV replicon HIV Gag vaccine (AVX101) in healthy vaccine trial in Kigali, Rwanda HIV-uninfected adults E Shutes1, E Karita1, K Kayitenkore 1, J Atkinson1, 1 1 1 2 7 1 2 3 J Bizimana , C Mambo , B Bekan , A Tichacek , J Chulay , D Burke , SSA Karim , N Russel , 3 3 3 3 1 M Wecker 3, M Allen4, G Ferarri 5 and P Gilbert 6 C Kambili , W Komaroff , P Fast , S Than , S Allen and the Rwanda Zambia HIV Research Group2 1 Johns Hopkins University, Baltimore, MD, USA; 2 University 1 Projet San Francisco, Kigali, Rwanda; 2 Emory University, of KwaZulu-Natal, Durban, RSA; 3 HVTN Core Operations Atlanta, GA, USA; 3 International AIDS Vaccine Initiative, New Center, Seattle, WA, USA; 4 NIH DAIDS, Rockville, MD, USA; York, NY, USA 5 Duke University Medical Center, Durham, NC, USA; 6 SCHARP, Seattle, WA, USA; 7 AlphaVax, Inc., Research Triangle Background: A Phase I HIV Vaccine Trial, Protocol V001 Park, NC, USA required the enrollment of 64 (32 from both Rwanda and Kenya sites) healthy, HIV uninfected adults at low risk for Objectives: To evaluate the safety and immunogenicity of an HIV infection with a minimum of 30% female volunteers. alphavirus replicon HIV-1 subtype C Gag vaccine (AVX101) Objective: To present the experience of enrollment into in healthy HIV-uninfected adults IAVI’s V001 study and the reasons for screening failure. Methods: Randomized, placebo-controlled, double-blind Methods: Concordant HIV negative couples were recruited study conducted at 4 sites in the US and 3 sites in southern from Couples’ Voluntary Counseling and Testing Centers Africa. Participants 18–50 years of age meeting enrollment (CVCT). Forty-five couples had been followed for up to 3 criteria received three injections of active vaccine (an years, others were referred directly from CVCT. alphavirus replicon particle [VRP] vaccine expressing non- Breastfeeding or pregnant women and those planning to myristoylated Gag at 10-fold increasing dosage levels, from conceive in the next year were screened out. Consenting was 4 8 10 to 10 infectious units [IU]) or placebo by subcutaneous a three-stage process for both partners. First, focus group injection at 0, 4 and 12 weeks after enrollment. Local and discussions on HIV vaccines were followed by group educa- systemic reactions were monitored daily for 7 days after each tion sessions including the informed consent video. Couples injection and adverse events were collected for 12 months. were told that one member of the couple would be enrolled Serum and PBMC were collected at weeks 0, 6, 14, 24 and 52 and priority would be given to the female. Finally, couples to measure antibody and T cell responses. were invited for a second viewing of the consent video, a Results: Blinded safety data are available from 132 partic- private question/answer session, assessment of understand- ipants (108 active, 24 placebo). The vaccine was well tol- ing and consent signing. Medical and laboratory screenings erated; local reactions were graded as none (58%), mild were completed to assess eligibility. (40%), moderate (3%) or severe (0%) and systemic reac- Results: 177 individuals (mostly partners) attended focus tions were graded as none (40%), mild (42%), moderate group sessions (90 women, 87 men), 115 volunteers signed (17%) or severe (2%) after one or more injections. No consent forms and were screened (46 women, 69 men) and severe adverse events were related to study product. 35 volunteers were eligible of whom 32 were enrolled (12 Blinded ELISA results for 92 participants (76 to 78 active, women, 20 men). The reasons for screening failure were 14 to16 placebo) showed a dose-dependent antibody clinically significant acute or chronic abnormality on histo- 4 5 6 7 response (0% at 10 and 10 IU, 25% at 10 IU, 67% at 10 ry and/or physical exam (n=49), unable/unwilling to com- 8 IU and 75% at 10 IU). T cell responses were modest: pos- ply for the study period (n=10), eligible males whose wives 7 8 itive interferon-γ ELISpot in 2 participants (at 10 or 10 were enrolled (n=7), positive HBsAg or anti-HCV antibody + 8 IU), positive CD8 CTL assay in 6 participants (3 at 10 IU (n=5), high blood pressure (n=3), refusal of long-acting and 3 at lower doses). contraceptives (n=2), significant lab abnormalities (n=2), Conclusions: The AVX101 vaccine was well tolerated, with failure to demonstrate consent form understanding (n=1) no significant safety issues identified in the clinical trial, and pregnancy (n=1). and induced a dose-dependent antibody response to HIV Conclusion: We had targeted to enroll 50% women and Gag. T cell responses were lower than in mice immunized expected that by involving the couple in the education and with AVX101 but consistent with recent unpublished non- consenting process, we would remove potential barriers to human primate data showing modest T cell responses in female enrollment. Thirty percent of men and women animals immunized with AVX101 but strong T cell screened were eligible for enrollment (35/115). After pre- responses in animals immunized with VRP expressing screening women for pregnancy, breastfeeding and desire myristoylated Gag. Definitive conclusions must await com- to conceive and prioritizing women in couples where both pletion and unblinding of this study and data from future partners were eligible, a substantial fraction of women studies with other VRP vaccines of potentially greater (n=12, 38%) were enrolled. immunogenicity. 196 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 197 Amsterdam, Netherlands 29 August–1 September 2006 P11-11 oped a positive response to the vaccine peptides. Five respon- ders had a positive assay 2 weeks after the second vaccina- Safety and immunogenicity of HIV CTL multi- tion, but not 2 weeks after the third vaccination. A sixth epitope peptides (MEP) vaccine in RC529-SE responder, not assessed after second vaccination, had a posi- adjuvant with or without GM-CSF tive assay after the third. Conclusions: The vaccine with the adjuvant(s) was moder- SA Kalams1, ML Elizaga2, P Spearman 3, B Metch2, ately reactogenic, with an acceptable safety profile. The 3- M Allen4, KJ Weinhold 5, SD Parker 6, MJ McElrath 2, dose regimen was minimally immunogenic. SE Frey 7, JD Fuchs8, MC Keefer 9, MD Lubeck10, JH Eldridge 10, DS Laufer 10 and L Corey 2 for the NIAID HIV Vaccine Trials Network 2 P11-12 1 Vanderbilt University School of Medicine, Nashville, TN; 2 Fred Evaluation of the safety and immunogenicity of Hutchinson Cancer Research Center, Seattle, WA; 3 Emory the EP HIV-1090 DNA vaccine in healthy, HIV-1- University School of Medicine, Atlanta, GA; 4 DAIDS, NIAID, NIH, uninfected adults Bethesda, MD; 5 Duke University School of Medicine, Durham, NC; 6 University of Alabama School of Medicine, Birmingham, ND Russell9, GJ Gorse1, LR Baden2, M Wecker3, M AL; 7 Department of Internal Medicine, Saint Louis University, St. Newman4, G Ferrari5, KJ Weinhold5, B Livingston4, D Louis, MO; 8 San Francisco Department of Public Health, San Lawrence6, I Thior7, H Li8, E Noonan8, and The HIV Francisco, CA; 9 University of Rochester School of Medicine, Vaccine Trials Network10 Rochester, NY; 10 Wyeth Vaccines Research, Pearl River, NY 1 VAMC and Saint Louis University, St. Louis, MO, USA; 2 Harvard Objectives: To evaluate the safety and immunogenicity of Medical School, Boston, MA, USA; 3 HVTN, Seattle, WA, USA; 4 HIV CTL MEP/RC529-SE GM-CSF. ± Pharmexa-Epimmune, Inc., San Diego, CA, USA; 5 Duke Background: MEP (Wyeth Vaccines Research) is a mixture of University, Durham, NC, USA; 6 Division of AIDS, NIAID, NIH, 4 synthetic peptides, each containing 1 of 3 HIV T helper epi- topes (from Env or Gag) and 1 of 4 HIV Clade B CTL Bethesda, MD, USA; 7 Botswana-Harvard Partnership for HIV “hotspots”–overlapping CTL epitopes (from Gag or Nef) Research & Education, Gaborone, Botswana; 8 SCHARP, Seattle, restricted by a number of MHC alleles. The peptides WA, USA; 9 Bill and Melinda Gates Foundation, Seattle, WA, USA (1000µg) are formulated with RC529-SE adjuvant (50 µg, ; 10 HVTN Core Operations Center, Seattle, WA, USA Corixa Corporation), a synthetic congener of monophospho- ryl lipid A. GM-CSF (250 µg) is used as a coadjuvant. Background: Cytotoxic T-lymphocyte (CTL) epitopes of Methods: In this multicenter, placebo-controlled, random- HIV-1 that bind to multiple HLA-A2, -A3, or -B7 supertype ized, double blind Phase I trial, 96 healthy, HIV seronega- allelic products, generally present in greater than 50% of sub- tive adult participants with HLA A3, B7 or B8 alleles type B HIV-1 isolates, were encoded in a DNA plasmid vac- received 1 ml IM injections of MEP/RC529-SE or cine (EP HIV-1090). MEP/RC529-SE+GM-CSF (40 participants each), or nor- Objectives: To evaluate the safety and immunogenicity of EP mal saline control (16 participants) in the deltoid, at HIV-1090 vaccine at 3 dose levels in a double-blinded, adju- months 0, 1, and 3. Immune responses were measured by vant-controlled, dose-escalation trial. interferon-γ ELISpot assay. Methods: EP HIV-1090 vaccine was a single DNA plasmid Results: Participants enrolled at 7 U.S. sites. 94% completed encoding 21 CTL epitopes of HIV-1 gag, pol, nef, rev, and the vaccine regimen. Injections were discontinued for allergic env, and the universal helper T lymphocyte (HTL) epitope, reaction (1), hypertension (2), other medical reasons (1), and PADRE, formulated in 3.4% polyvinylpyrrolidone in sterile participants unable to be contacted/off schedule (2). saline, and administered intramuscularly at each of three Moderate injection site pain/tenderness was the most com- dose levels (0.5 mg, 2.0 mg, and 4.0 mg) at 0, 1, 3, and 6 mon reaction (65% MEP/RC529-SE, 58% MEP/RC529- months. Healthy, HIV-1-uninfected adults, aged 18–40 years SE+GM-CSF, and 6% control participants). The only and confirmed HLA-A2, -A3, and/or -B7 positive, were differences in reactogenicity between vaccine arms were enrolled at three sites (14 per site), randomized to receive vac- headaches of greater severity associated with MEP/RC529- cine (n=35) or vehicle placebo (n=7), and followed for 12 SE+GM-CSF (P=0.01), and a trend toward injection site months after the last vaccination. Cellular immune responses pain/tenderness of greater severity associated with were assessed in subsets of participants by chromium-release MEP/RC529-SE (P=0.08). Severe reactions occurred only in cytotoxicity and IFN-γ ELISPOT assays using synthetic pep- vaccine recipients: 4 events of pain/tenderness and 1 event of tide pools representing the CTL epitopes and the HTL epi- malaise with MEP/RC529-SE; 1 event each of tope in the vaccine. fatigue/malaise, headache/malaise, and allergic reaction with Results: The distribution of HLA supertypes of the 42 partici- MEP/RC529-SE+GM-CSF. By IFN-γ ELISpot assay, 0/88 par- pants was: A2: 25 (60%), A3: 18 (43%), B7: 17 (40%). No ticipants tested at baseline demonstrated immune responses notable adverse events attributed to the vaccine were observed. to vaccine peptides. Subsequently, 6 participants in the vac- In the IFN-γ ELISPOT, none of the vaccine recipients had a cine arms (5 with MEP/RC529-SE) versus. 0 controls devel- AIDS Vaccine 2006 197 Poster sessions_revGJ 14/8/06 16:34 Page 198 Amsterdam, Netherlands 29 August–1 September 2006 response to HIV-1 peptides and one responded to the PADRE tralizing antibodies) in small animals models. VICHREPOL HTL epitope 2 weeks after the third and fourth 4 mg dose vac- was approved by National Regulatory Authority of Ministry cinations. Three participants from the 0.5 mg dose group had of Health of Russian Federation for vaccine clinical trial CD8+ CTL responses: 1 to both the HLA-A2 HIV-1 and B7 (phase 1). Candidate vaccine VICHREPOL was found to be HIV-1 peptide pool detected at 12 months after the final vac- well tolerated, safe and induced antibody response in the cination; and 2 to the HLA-A3 HIV-1 peptide pool, one at 3 course of phase 1 clinical trial. No significant vaccine-related months and the other at 6 months after final vaccination. toxicity has been observed. Conclusions: This first clinical trial evaluating EP HIV-1090 Conclusions: Our results indicate that phase 1 clinical trial of CTL epitope DNA plasmid vaccine demonstrated that the the first Russian candidate HIV vaccine (VICHREPOL) is vaccine was safe and well-tolerated, but only weakly being conducted successfully. Further subsequent studies are immunogenic. Vaccine-specific CD8+ CTL were only tran- needed to fully evaluate safety and immunogenecity of high- siently detected in a small subset of participants. er doses and repeated immunizations. P11-13 P11-14 Abnormal laboratory values among volunteers for Phase 1 clinical trial of the first Russian candidate a Phase I vaccine trial in the Caribbean Region HIV vaccine (VICHREPOL) is being conducted suc- (Merck 018/HVTN050) and the need to reevaluate cessfully the eligibility criteria E Karamov1, G Kornilaeva1, T Pavlova1, S Korobova2, I CD Zorrilla1, LE Santiago1, J Perez1, JW Pape 2, MM Nikolaeva2, A Chevalier2, G Goudima2 and I Sidorovich2 Deschamps2, S Jean2, Y Donastorg3, M Perez 3, C Volques3, S Buchbinder4, J Ginanni5 and S Hammer6 1 Ivanovsky Institute of Virology, Moscow, Russia; 2 Institute of Immunology, Moscow, Russia 1 UPR School Of Medicine, San Juan, Puerto Rico; 2 GHESKIO, Haiti; 3 HVTU, Santo Domingo, Dominican Republic; 4 San Background: More than 330,000 HIV1 cases were registered Francisco Department of Public Health, San Francisco, CA, USA; in Russia, estimated>1 million HIV cases. The majority of all 5 Merck Research Laboratories, West Point, PA, USA; 6 Columbia HIV infected patients are injection drug users (IDUs). While University, New York, NY, USA most HIV1 infections over the last 10 years of observation were associated with drug injections the proportion of indi- Background: The search for the AIDS vaccine requires the viduals infected through sexual contacts has been growing. participation of volunteers worldwide. The eligibility criteria The main HIV1 variant in Russia belongs to the genotype A for Phase I studies requires “normal” values as defined for (subtype A1) ~ 90%, affecting primarily injecting drug users Caucasians in the USA. People in different geographical (IDUs) and their sexual partners on the territory of Russia, regions have nutritional, cultural, environmental or genetic Ukraine, Byelorussia, Kazakhstan and Baltic countries. factors which will influence these values. Some “out of range Genetic homogeneity with the mean genetic distances in env values” do not have clinical significance and can become a region among viruses isolated in various regions of Russia is hindrance to active participation in vaccine clinical trials. being no more than 3-5%. 80% of new cases of HIV infec- Such is the case of the Caribbean region where a significant tion are accompanied by HCV co infection especially by 1b proportion of the volunteers have baseline out of range val- variant. ues for some laboratories. Objectives: Phase 1 trial was initiated in Russia to assess the Methods: Baseline laboratory values of the screened volun- safety and immunogenecity of the first Russian HIV1- vac- teers for a Phase I HIV vaccine trial were obtained from 3 cine candidate VICHREPOL, based on the recombinant pro- Caribbean sites: Haiti (n=81), Dominican Republic (n=39) tein which included the full length viral protein p24 and and Puerto Rico (n=47) and compared within sites. N-terminal region of gp41 HIV-1. Recruitment focused on healthy HIV-negative adults. Methods: In this ongoing trial in 5 dosage levels healthy HIV Results: 185 volunteers were screened, 160 maintained interest uninfected male adult volunteers were enrolled and they were and 75 (47%) were enrolled. Ninety-one (57%) of 160 had screened in accordance with complete medical history and abnormal laboratories. Forty of the 75 recruited participants physical examination. Four immunizations were carried out. had minor and not clinically significant abnormalities. The Hematology, clinical chemistry, serology and routine urinaly- most common abnormality reported for the screened volun- sis were performed to assess liver and kidney function. Sera teers in PR was elevated C-reactive protein (CRP) in 12 of 47 samples collected to evaluate the immunogenecity of (26%). In Haiti these were: elevated serum amylase in 28 of 81 VICHREPOL. ELISA and Western blot tests were used for (35%), low absolute Neutrophil count in 15 (19%), decreased sera samples analysis. WBC’s in 8 (10%), increased CK in 8 (10%) and high CRP in Results: Pre-clinical evaluation has shown that VICHREPOL 8 (17%). In the Dominican Republic elevated CRP was report- with built-in adjuvant Polyoxidonium induced potent ed for 12 of 39 (31%), and elevated serum amylase in 7 (18%). immune responses (both cellular and humoral, including neu- 198 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 199 Amsterdam, Netherlands 29 August–1 September 2006 Conclusions: The proportion of laboratory abnormalities in under 20 and those over 20 years of age. 68/ 2,540 (2.7%) the region was overall high (57%) although there were dif- reported a social impact event of which 43% had minimal ferences between countries for specific abnormalities reflect- impact to normal daily living, 84% resolved satisfactorily. 17 ing nutritional, environmental and behavior diversity. Some of these volunteers withdrew consent, eleven of which were were considered minor but not all. Elevations in CRP were students., The reason for withdrawal was the disagreement the most common abnormality (32/160) overall or 20% of with a family member. About 7 % of the participants under the potential subjects. the age 20 were lost to follow up. Recommendations: Due to the particular characteristics of Conclusion/Discussion: Participation of volunteers between the populations in specific regions of the world, eligibility cri- the ages of 18 and 20 in the vaccine trial is feasible. A com- teria for Phase I vaccine studies need to be re-evaluated. CRP, prehensive educational campaign for parents and the com- CK and amylase might not be useful as inclusion criteria but munity is an important tool to avoid conflict or harm would be useful as clinical markers. among them. P11-15 P11-16 Adolescents participation in the Phase III efficacy Dose sparing with intradermal injection of HIV trial of ALVAC HIV-1 vaccine priming, AIDSVAX lipopeptides in HIV uninfected adult volunteers: a vaccine boosting in Thailand randomized controlled study (ANRS VAC16 trial) P Pitisuttithum1, S Rerk-Ngam2, J Kaewkungwal 1, C Durier 2, O Launay 1, C Desaint 1, B Silbermann1, W Wiriyakitja 3, S Watana 4, K Palasudhi 4, C Pobsuk 3, A Jackson3, G Pialoux 4, B Bonnet 5, I Poizot-Martin 6, M Benenson5 and J Kim 5 G Gonzalez-Canali 7, L Cuzin 8, D Salmon 1, M Surenaud 3, C Guérin 9, I Bourgault Villada 3, and JG Guillet 3 for the 1 Faculty of Tropical Medicine, Mahidol University, Bangkok, Agence Nationale de Recherche contre le SIDA (ANRS) Thailand; 2 Department of Disease Control, Ministry of Public VAC16 trial group Health, Bangkok, Thailand; 3 Chon Buri Provincial Health Office, Chon Buri, Thailand; 4 Rayong Provincial Health Office, 1 CIC de Vaccinologie Cochin-Pasteur, Service de Médecine Rayong, Thailand; and 5 Armed Forces Research Institute of Interne, Hôpital Cochin, Paris, France; 2 INSERM SC10, Villejuif, Medical Sciences France; 3 Institut Cochin, INSERM U567, CNRS and Université Paris V, Paris, France; 4 Service des Maladies Infectieuses, Hôpital Background: The phase III HIV vaccine community trial Tenon, Paris, France; 5 Service de Maladies Infectieuses, Nantes, started in 2003 with the initial enrollment of participants France; 6 Service d’Hématologie, Marseille, France; 7 Service aged 20–30 years. Since there is an increasing trend of the d’Immunologie Clinique, HEGP, Paris, France; 8 Service de incidence of HIV infection among persons between 15–20 Maladies Infectieuses, Toulouse, France; 9 Pharmacie, Hôpital years of age, the protocol was amended and approved to Cochin, Paris, France include volunteers between 18 and 20 years of age who are considered not legally independent. Objectives: To assess demographic profiles, reasons for par- Background: Dose-sparing strategies that use intradermal ticipation and social impact events among participants age (ID) delivery of vaccines may be one approach for improving below 20 years in the phase III community trial. vaccines immunogenicity and reducing the cost of vaccines. Methods: The trial is a randomized, double-blind, placebo Objective: To compare the immunogenicity and the safety of controlled study. Participants aged below 20 years, HIV neg- ID immunization with HIV lipopeptides vaccine with intra- atives were enrolled as part of the modified protocol. General muscular (IM) immunization. demographic characteristics, reasons for participation, and Methods: In a randomized, open, phase 1 trial, 68 healthy social impact events were analysed and compared between 21–55 -old volunteers received at weeks 0, 4, and 12, either those volunteers under 20 years of age with those over 20 3 IM doses of 0.5 ml of LIPO-4 vaccine, containing 500 µg years of age. of each peptide, or 3 ID doses of 0.1 ml, containing 100 µg Results: 4,201 participants aged 18–20 years were screened, of each peptide (20% of the IM dose). LIPO-4 vaccine is a 2,540 were enrolled which accounts for 15% of all enrollees mixture of 4 HIV lipopeptides linked to TT 830-843. The (16,402). 22%, 26% and 20% reported their occupation as primary end-point for the safety evaluation was at week 24. factory workers, students or laborers respectively. 75% had a HIV-1 CD8 T cell immune responses were assessed by IFN-γ secondary school or higher educational level. 74% reported ELISPOT assay and TT 830-846 CD4 T cell response by lym- reasons for participating in the trial as wanting to help soci- phoproliferative assay two weeks after each administration ety or wanting to do something good. 44% reported as want- and at week 24. ing to know HIV information. The major sources of decision Results: All 68 subjects received the first and the second vacci- to participate in the trial were project staff, friends, and infor- nation, and 44 received the third (22 in each group). No severe, mation from the screening video. There were no statistically serious or life threatening adverse events were reported during significant differences in the above parameters between those the study. Local pain was significantly more common in the IM AIDS Vaccine 2006 199 Poster sessions_revGJ 14/8/06 16:34 Page 200 Amsterdam, Netherlands 29 August–1 September 2006 group than in the ID group. Signs of local inflammation were approximately 20–30% higher response rate for the samples more common among subjects in the ID group than among processed in less than 12 h, as well as at least a three-fold those in the IM group, but such reactions were mild and tran- average increase in magnitude of response to Gag, Pol and sient. The cumulative percentages of CD8+ T cell responses to Nef with no increase in background mock responses. at least 1 HIV peptide at weeks 2, 6, 14, 24 were 9%, 33%, Conclusion: These results provide evidence that the use of an 39% and 52% for ID group and 15%, 20%, 26% and 37% optimized method of collection, preparation and handling of for IM group. TT 830-846 CD4+ T cell response was induced PBMC samples as is the case with the use of a centralized net- in 9%, 46%, 54% and 63% in the IM group, 6%, 27%, work, result in increased immune response, Optimal quality 33% and 39% in the ID group, at weeks 2, 6, 14 and 24, PBMC are vital to the success of any clinical trial that relies respectively. on cell-mediated immunity as a potential immune correlate. Conclusion: In this trial, ID administration of LIPO-4 vaccine was well tolerated. Moreover, ID injections required one fifth of IM dose to elicit largest amount of HIV specific CD8+ T cell response. Finally, route of immunization influenced the P11-18 specific T cell population CD4+/CD8+ induced. Building productive, collegial dialogue between researchers and communities: The HVTN Model P11-17 G Broder, L Bull, A Lambert, S Wakefield, and Enhanced positive responder rates in Elispot G Djomand assay reported when using an optimized method for collection and isolation of PBMC: evidence HIV Vaccine Trials Network, Community Education Unit, Seattle, supporting the use of a PBMC Network WA, USA D Casimiro, B Meyer, L Kierstead, S Dubey, R Mogg, Objectives: To share techniques for creating structured V Fernandez, R Long, L Guan, C Gaunt, K Collins, researcher-community discussions that build cohesion. The D Mehrotra , N Chirmule and J Shiver goal is to facilitate constructive dialogues that focus on advancing research. Merck Research Laboratories: Vaccine and Biologics Research, Methods: The HVTN has developed a successful structured ‘template’ session for researcher/community interaction that West Point and Wayne, PA, USA enables productive discussion and can serve as the frame- work for establishing a cohesive dialogue. Initiated during an Objective: Establishment of a group of regional processing educational retreat for Community Advisory Board mem- laboratories capable of receiving whole blood samples, iso- bers, the session involves round-table discussions on chal- lating and freezing PBMC in under 12 h of phlebotomy. lenging scientific concepts. After an initial overview of a basic These regional laboratories must be located within a 4 h dri- scientific concept, each researcher involved uses ‘cocktail- ving distance in order to efficiently service a clinical site. party skills’ to lead community members in a small round- Those same laboratories must be able to temporarily store table discussion of some aspect of the concept, with PBMC at -70°C, and ship them on dry ice to Merck. community members prompting researchers to focus their Methods: Merck has contracted with Laboratory discussion on the needs of the listeners. Following this round- Corporation of America to set up 16 regional laboratories table discussion, community members partner with partici- across the US headed by a project team located in Raritan NJ. pants from different groups. Each tutors their new partners in Technicians were trained and submitted 5 qualification sam- what they have just learned, while researchers circulate to ples that were analysed for several parameters including per- answer questions. formance in a standardized IFN-γ Elispot assay. Results: When the structured discussion was instituted, all of Elispot data from a clinical trial were summarized by the the CAB members were able to explain the new concepts to PBMC processing method (>12 h versus <12 h) using the per- their peers with a high degree of accuracy. Community mem- centage of subjects responding to antigens in the Elispot bers’ evaluations reflected a new connection to and sense of assay, as well as the geometric mean. Because this was a non- respect from the researchers, leading to feelings of trust and randomized comparison, the groups were adjusted for poten- engagement. Researchers were surprised by community mem- tial imbalances in baseline Ad5 titer, gender, age, and race bers’ accuracy, and noted respect for the depth of their ques- using propensity score methodology. tions and comments, leading to connection and a clearer Results: Subsequent to the establishment of the PBMC understanding of the community’s needs. Both groups regis- Network, IFN-γ Elispot data from an ongoing clinical trial of tered a sense of unity of purpose and respect, and a feeling of an MRKAd5Gag/Pol/Nef HIV vaccine were analysed to look engagement in a wider dialogue. This small-group learning for any differences in the overall immunogenicity of the vac- model has become a flexible tool within the HVTN, allowing cine as a result of the change in procedures. Results were for continual development of a productive researcher-com- compared for the percentage of Elispot responders to Gag, munity connection. Pol and Nef, and responding to at least one antigen, at least Conclusions: Continued adaptations of this framework have two antigens, and all three antigens. These data revealed an proven useful. The direct connection and the opportunity to 200 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 201 Amsterdam, Netherlands 29 August–1 September 2006 ask questions empower community members, and they help researchers form analogies and metaphors for their work. Researchers are able to better gauge the needs of their audi- ence and establish trust. Initiating discussion around research, rather than policy or politics, allows the group to focus on shared goals. With the resulting respect and clarity, the two groups can develop a cohesive dynamic, leading to further productive dialogue. AIDS Vaccine 2006 201 Poster sessions_revGJ 14/8/06 16:34 Page 202 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 12: THERAPEUTIC VACCINE TRIALS P12-01 infected rhesus monkeys. It is not clear if the particular drug combination for ART might have contributed to the delayed Therapeutic SIV derived DNA-vaccines in the viral rebound. SIVmac239 Rhesus macaque animal model JZ Megede1, B Beer2, P Silvera2, D Kayda2, A Bowlsbey2, P12-02 D Hebblewaite2, D Sites2, L Nieves-Duran2, R Srivastava2, L Zhang3, D Rabussay3, GR Otten1, JB Ulmer1, Cell mediated immune response in HIV-1-infected SW Barnett1, and JJ Donnelly1 subjects on HAART induced by two different DNA vaccines expressing either single gene, GTU-Nef or 1 Chiron Corporation, Vaccines Research, Emeryville, USA; multigenes, GTU-MultiHIV of a subtype B-HIV-1 2 Southern Research Institute, Frederick, USA; 3 Inovio 1 2 1 1 Biomedical Corp., San Diego, USA V Blazevic , K Krohn , I Stanescu , K Laurikainen , I Piippo1, T Kärkkäinen1, M Tähtinen1, J Oksi3, M Ristola4 4,5 Background: Current treatment of HIV-1 infected patients is and A Ranki restricted to antiretroviral therapy (ART), with side-effects and high cost being the main limitations. Therapeutic vacci- 1 FIT Biotech Plc, Tampere, Finland; 2 Institute of Medical nation during ART may enable patients to maintain better Technology, University of Tampere, Finland; 3 Turku University control of viral loads and higher CD4 counts and even allow Central Hospital, Finland; 4 Department of Dermatology and for time-off from HAART to evade drug-related side-effects. Venereology and 5 Department of Infectious Diseases, Helsinki Objectives: To evaluate effectiveness of therapeutic DNA- University Central Hospital, Helsinki, Finland vaccines in combination with IL-2 in the SIVmac239/macaque model. Background: The development of an effective, safe and Methods and Results: Thirty naive animals were inoculated affordable vaccine against human immunodeficiency virus with SIVmac239 (1000 MID i.v.). All animals became 50 type 1 (HIV-1) infection is becoming an increasingly impor- infected and viral loads in the acute phase ranged from 5×105 tant goal as the epidemic caused by the virus spreads with an to 5×107 copies/ml. ART (PMPA, Stavudine, FTC) was accelerating speed all around the world. administered from weeks 13 to 41 post infection. Five ani- Objectives: To evaluate in HAART treated individuals the mals were euthanized due to rapid disease progression, six safety and immunogenicity of two DNA vaccines expressing animals removed from the study due to inefficient viral con- either HIV-1 Nef or complete sequences of Rev, Nef, Tat, trol post ART. The remaining 19 animals were randomized p17/p24 proteins, and an epitope stretch consisting of previ- into three groups with 3 MamuA01+ animals/group: Group ously identified CTL epitope-rich regions encoded by pol and 1 (n=7; ART only), Group 2 (n=6; ART+DNA), Group 3 env of a subtype B-HIV-1 isolate Han-2. (n=6; ART+DNA+IL-2). Modified forms of SIVmac239 Methods: Both candidate HIV-1 vaccines were designed using gp140Env, GagPol and TatRevNef plasmids were engineered FITBiotech’s GTU technology platform and were produced in for high-level expression and delivered intramuscularly with our GMP licensed facility. In the first clinical trial with the in vivo electroporation at weeks 22, 26, 30, and 34. Animals GTU-Nef, 14 patients received two intramuscular injections, in group3 received twice daily sc injections of Proleukin 2 weeks apart. Five patients received 1 µg of the vaccine, five (50,000 IU/kg), days 2-16 post each vaccination. CD8+ T cell received 20 µg and four received 400 µg at each immuniza- responses to SIVGag were evaluated by tetramer staining in tion. In the second clinical trial conducted with GTU- MamuA01+ animals. All groups had detectable CD8+ MultiHIV B, 13 patients were enrolled and immunized twice responses after infection and prior to ART (1–2%), that intramuscularly within 1 month apart. Three subjects waned during ART but increased post vaccination with max- received 2 µg, three 20 µg, four 200 µg and three 2000 µg of imum responses, post third for group 2 (2.7%) and group 3 the investigational vaccine at each vaccination. Humoral (3.7%). Four weeks after ART termination, CD8+ responses immune response was evaluated by measuring Nef and Gag were higher in DNA vaccinated animals (4.9% and 6.2%) specific antibodies. Cell mediated immune response was eval- compared to controls (1.7%), and remained at higher levels uated by measuring CD8+ cytotoxic T lymphocyte (CTL) for 12 weeks (latest evaluated timepoint). Interestingly, viral responses and CD4+ T helper (Th) lymphocyte responses to loads for all groups remained low and at comparable levels HIV-1 proteins. (2,418–5,660 copies/ml) 13 weeks post ART termination, a Results: Both candidate vaccines were safe and well tolerat- phenomenon we believe may be due to the triple drug thera- ed. None of the vaccines induced Nef or Gag specific anti- py used in this study. bodies. In contrast, a vigorous Nef-specific T cell response Conclusions: Our data suggest that in vivo electroporation of was induced after the vaccination. In the GTU-Nef clinical DNA-vaccines resulted in increased CD8+ responses in SIV trial more then 70% of the patients have developed Th lym- 202 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 203 Amsterdam, Netherlands 29 August–1 September 2006 phocyte responses to HIV-Nef, as measured by T cell prolif- detectable CD8+ cross-reactivities to p24, p17 and Tat pro- eration or IFN-γ production. Furthermore, 4 of 9 patients tein (P<0.02). C+ patients had significantly better prolifera- generated Nef-specific CTL responses. In the GTU-MultiHIV tive responses on CART at week 10 but with no difference clinical trial, 8 of the 13 immunized trial subjects developed during viraemia at study end compared with C-. Th and CTL responses to the trial vaccine. Conclusions: HIV-specific responses in vitro often may Conclusions: Genetic immunization with naked plasmid become reduced during viraemia, in contrast to DTH in vivo DNA is safe and able to induce cell-mediated HIV-specific (Kvale, AIDS 2005). A comparison between C+ and C- immune responses. Furthermore, it is possible to induce a patients may thus be complex. We nevertheless found com- robust immune response in HIV-infected patients even with parable immunogenicity between these two patient groups low doses of the DNA vaccine. having similar CD4+ counts, although their HIV history were different. The possibility that CART does not improve HIV- specific immune responses after vaccination should be P12-03 explored in further studies due this option’s simplification of mass immunization. Comparable immunogenicity between patients on or off combination antiretroviral therapy (CART) after immunization with a novel therapeutic pep- P12-04 tide-based immunotherapy candidate targeting dendritic cells (DC) Follow-up of HIV infected patients who received a therapeutic anti-Tat vaccination D Kvale1, A-MB Kran1, MA Sommerfelt2, J Nyhus2, I Baksaas3 and B Sørensen2 D Zagury1 and RC Gallo2 1 Ullevål University Hospital, Oslo, Norway; 2 Bionor Immuno, 1 Néovacs SA, Université Pierre et Marie Curie, Paris, France; and Skien, Norway; and 3 Mericon, Skien, Norway 2 Institute of Human Virology, University of Maryland, Baltimore Background: Optimal induction of HIV-specific responses Background: Basic and epidemiological documentation as after therapeutic immunization may theoretically require well as non human primate experimentation prompted us to CART. The first data are presented from a completed Phase I develop anti Tat therapeutic vaccine based on Tat toxoid, a clinical trial with a peptide-based pentavalent therapeutic non toxic but immunogenic HIV-1 Tat derivative. Phase I vaccine (Vacc-5q) consisting of short consensus peptides from trial conducted at the Hemophiliac Bonomi Center of Milan p17, p24, and Tat. (Pr. Gringeri) in 1997–1998 and Phase I/II trial organized by Objectives: To compare immunogenicity between patients on Aventis Pasteur showed that the Tat toxoid immunogen adju- effective CART (C+) and ART-naive patients (C-) as well as vanted with either Seppic oil (ISA51), DcChol or Alum was safety. safe and immunogenic on patients under HAART or not. Methods: Open, prospective trial with healthy C- patients Objectives: To evaluate the disease progression, patients (n=10) with CD4+ count >500/µl and CD4+ loss <150/µl/year with a therapeutic anti-Tat vaccination have been followed and healthy C+ patients (n=10) having current CD4+ count during 2 years. >200/µl were included and monitored for 40 weeks. The C+ Methods: A structured treatment interruption study (STI) group stopped CART between study weeks 10–14 and monitored according to EU guidelines was conducted at 28–40. All patients received GM-CSF (60 µg) intradermally Brussels (Pr. Clumeck) on the 31 vaccinees who received as adjuvant followed by Vacc-5q (0.1 mg/peptide) at the same either a DcChol adjuvanted Tat Toxoid (n=12), a DcChol site; 8 doses weeks 0–6 and 3 doses weeks 24–26. Vacc-5q placebo (n=8) or non adjuvanted Tat Toxoid (n=11). responses were monitored in vivo by skin induration 48 h Results: Anti-Tat Ab responders (n=9) exhibiting both high after DTH intradermal skin test, as well as in PBMC in vitro serum Ab titers (>10 pg/ml) and a serum anti-Tat neutralizing by proliferation and activation/proliferation (CD25 induc- capacity at the end of the vaccine trial remained significantly tion) after 5 days. Responses to p17, p24 and Tat were also HAART-free. By contrast in patients in whom HAART has monitored. Groups and time points were compared by been prescribed during STI, serum collected prior to treat- Mann-Whitney U-test and Fischer exact test. ment did not exercise anti-Tat neutralizing capacity. Results: At baseline, C+ and C- patients had similar CD4+ Conclusion: The 2 year therapeutic interruption analysis study counts (median, 590/µl). Overall, no qualitative response dif- suggested that vaccinees developing high titer of antibodies ferences between C- and C+ were observed; DTH+ developed inhibiting Tat activity prolonged HAART-interruption. in 90% (C+) and 78% (C-), respectively (n.s.). Quantitatively, DTH induration increased substantially (P<0.001) and similar proliferative responses as well as CD25+ responses were obtained by both groups. However, early at weeks 4-10, C- patients developed higher CD8+ T cell responses to Vacc-5q peptides than C+ (P=0.02) as well as AIDS Vaccine 2006 203 Poster sessions_revGJ 14/8/06 16:34 Page 204 Amsterdam, Netherlands 29 August–1 September 2006 P12-05 P12-06 Randomized-controlled Phase II study with an DNA vaccines against human immunodeficiency MVA-Nef vaccine in HIV-1 infected patients with virus type 1 CD4 counts >250/µl followed by structured treat- ment interruption (STI) – first results Y Zadorin and K Shevchenko T Harrer2, S Engfeld1, J Hain1, N Baedecker1, V Sobek1, National Bio Science, HIV department, Kiev, Ukraine J Vollmar1, E Harrer2, M Helm4, D Gorriahn5, L Schneider6 and H Jaeger3 Background: HIV-1 was among the original DNA vaccine targets and HIV DNA vaccines are now in human trials. Lack 1 Bavarian-Nordic GmbH, Martinsried, Germany; 2 Department of strong correlates of protective immunity makes vaccine of Medicine III, Friedrich-Alexander University, Erlangen- design difficult; however, DNA vaccines have the potential to Nuremberg, Germany; 3 Practice Dr. Jaeger, Munich, Germany; be an ideal vaccine and therapeutic approach against HIV-1. 4 Practice M. Helm, Nuremberg, Germany; 5 Practice Dr. DNA vaccines induce conformational-dependent antibodies, mimic live vaccines but without the pathogenic potential, and Gorriahn, Munich, Germany; 6 Practice Dr. Schneider, Fuerth, can easily be made polyvalent. Genes which encode impor- Germany tant CTL and antibody epitopes can be included while those that confer pathogenicity, virulence, antibody enhancement Background: There is considerable evidence that cell mediat- or represent non-conserved epitopes can be excluded. ed immunity (CMI) can effectively control HIV-1 replication Objectives: To evaluate the effect of DNA vaccines and ther- ® during acute and chronic infections. MVA-BN (Modified apeutic approach against HIV-1. Vaccinia Ankara-Bavarian Nordic), a safe viral vector inca- Methods: In our hands pre-treatment of muscles with bupi- pable of replicating in human cells, has shown in previous vacaine or cardiotoxin did not offer any advantage over no studies potent ability to induce CMI. Therefore, we per- muscle pre-treatment or gene gun inoculation of skin formed a Phase II study in HIV-1 infected patients to assess although gene gun immunization seem to favour a Th2 type immunogenicity and safety of a recombinant MVA vector response. As DNA vaccine candidates we have compared expressing HIV-1-LAI Nef. vaccines encoding native HIV MN gp160 with Rev-indepen- Methods: In this single blind, randomized study 77 patients dent synthetic genes encoding MNgp160 and MNgp120 8 ® received three s.c. vaccinations of either 1×10 MVA-BN , using mammalian high expression codons. In these experi- 8 8 1×10 or 5×10 MVA-Nef (n=25, 26 and 26, respectively) at ments the gene encoding secreted gp120 gave highest anti- week 0, 8 and 16. Patients were monitored for Adverse body neutralizing titers. High and fast antibody responses Events, viral load and CD4/CD8 counts. Antibody response could also be obtained by transferring the HIV-1 MN V3 against MVA was measured using ELISA. Patients that loop to the secreted HBsAg as a fusion gene vaccine. received all vaccinations and remained stable on HAART Results: Thus, in the case of HIV-1 MN genes encoding were offered at week 20 a STI with close monitoring at weeks secreted surface glycoproteins may be preferred instead of 24 / 26 / 28 / 32 / 40 and 52 regarding safety, plasma HIV-1- membrane bound envelopes. CTL responses were induced in RNA levels, CD4 count and immune response against HIV. all cases. However, in order to meet the high diversity of HIV Results: In this ongoing trial first data was evaluated; MVA- and HLA types our approach is to include many CTL epi- Nef proved to be safe and well tolerated except for transient topes in a multivalent minigene vaccine. We found that gene local reactions and mild systemic side effects. All patients gun DNA vaccination with minimal epitopes could induce ® seroconverted against MVA-BN after the second vaccination specific CTL. Flanking sequences influenced the CTL and an increase in CD4 and CD8 count was observed. End of response but was not needed. March 2006, 57 patients performed the study visit at week Conclusions: DNA vaccines encoding known and computer 20 and 24 of them decided to interrupt HAART. Median time predicted CTL epitopes are now being developed. on STI is currently 112 days with 2 patients restarting thera- py. Further subjects are expected to interrupt within the next 2 months, so that STI data of up to 10 months and further study results will be available at the conference. Conclusions: In this ongoing study MVA-Nef proved to be safe and immunogenic in HIV infected patients. Apart from the use as an HIV vaccine, the strong induction of an immune response against MVA-BN® confirms the potential as a safe and potent smallpox vaccine in HIV-infected subjects. Further results will allow discriminating between the MVA- specific and Nef-specific effect on immune status and espe- cially on viremia during STI. Acknowledgement: Partially funded by NIAID (contract No. N01-AI-40072). 204 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 205 Amsterdam, Netherlands 29 August–1 September 2006 P12-07 P12-08 Development of a therapeutic vaccination strategy Control of viremia after antiretroviral treatment with a recombinant modified vaccinia virus and therapeutic vaccination with novel forms of Ankara vaccine expressing and HIV-1 Gag/multi- DNA vaccines in chronically SIVmac251-infected epitope gene for control of HIV-1 infection macaques L Dorrell1, H Yang1, T Dong1, B Ondondo1, E Turnbull2, GN Pavlakis1, A von Gegerfelt1, A Valentin1, M Rosati1, A Guimarães-Walker1, N Goonetilleke1, N Winstone1, C Alicea1, P Roth1, J Bear1, J Boyer2, D Weiner2, C Conlon3, D Brown1, P Williams4, A Suttill4, K de P Markham3, P Albert4, G Franchini4 and BK Felber1 Gleria1, P Borrow2, S Rowland-Jones1, J Robert1s, T Hanke1 and A McMichael1 1 National Cancer Institute at Frederick, Frederick, MD, USA; 2 University of Pennsylvania School of Medicine, Philadelphia, PA, 1 MRC Human Immunology Unit, Oxford University, UK; 2 USA; 3 Advanced BioSciences Laboratories, Inc., Kensington, MD, Edward Jenner Institute for Vaccine Research, Compton, UK; USA; and 4 National Cancer Institute, Bethesda, MD, USA 3 Nuffield Department of Clinical Medicine, John Radcliffe Hospital, Oxford, UK; 4 Harrison Department, Genito-Urinary Background: Continuous antiretroviral treatment is associat- Medicine, Radcliffe Infirmary, Oxford, UK ed with toxicity and the emergence of resistant viral strains. Therapy must be continued indefinitely, since virus replica- Background: Therapeutic vaccination is a possible strategy to tion resumes rapidly upon treatment interruption due to virus reduce HAART use but candidate vaccines tested to date persistence in stable reservoirs and also due to continuous have shown modest immunogenicity. Furthermore, the qual- residual virus replication. Thus, additional approaches to ity of the T cell responses induced by these vaccines has not control viral propagation are needed. been studied in any depth, yet may provide insights into Objectives: We explored therapeutic immunization of ART- immune correlates of viral control. treated SIV-mac251 infected Rhesus macaques, using a new Objectives: To determine the effect of vaccination with generation of optimized DNA-based vaccine vectors that pro- MVA.HIVA, which expresses HIV-1 gag p24/p17 and 23 duce either secreted or intracellularly degraded antigens. Gag/Pol/Nef/Env epitopes, on the magnitude and quality of Methods: Macaques infected for 15–70 weeks with HIV-1-specific T cell responses in HAART-treated chronical- SIVmac251 were treated with a combination of PMPA, ddI ly infected individuals. and Stavudine for 13–23 weeks. During this time, the animals Methods: 16 subjects (under HAART for > 1 year, with plas- were vaccinated via IM route with optimized forms of DNA ma HIV-1 RNA <50 copies/ml) and CD4 cells >300/µl) vectors expressing SIV antigens and then released from ART. received two intradermal injections of MVA.HIVA 5 ×107 pfu The animals were monitored for rebounding virus and four weeks apart and were followed up for 1 year. Multi- changes in SIV specific immune responses during and after parametric analysis of HIV-1-specific T cell responses was termination of therapy. Statistical analyses were performed performed to assess magnitude, breath, proliferative and for all the animals at least partially responding to ART by functional capacity, TCR repertoire and in vitro viral sup- reducing viremia. Immunological analysis included antibody, pressive capacity. Elispot measurements, and intracellular cytokine analysis by Results: MVA.HIVA vaccination significantly amplified dom- 8-parameter FACS. inant and sub-dominant virus-specific CD8+ and CD4+ T cell Results: Macaques receiving DNA showed a significant responses, with a peak gag-specific response detected by IFNγ decrease in viral load for long periods after therapy termina- Elispot assay typically 2 weeks after the second immuniza- tion compared to controls (P<0.001, Wilcoxon rank sum tion. Vaccine-driven expansions were detected by tetramer test). DNA vaccination, but not ART alone, led to substantial staining and found to persist for at least one year. Tetramer+ decreases in viremia. Some animals (6/12) continue to control populations comprised clonally diverse CD8+ T cells with viremia levels for 2 years. Importantly, cellular immune proliferative, lytic and cytokine-secreting capacity. responses were boosted by DNA vaccination and persisted Preliminary data indicate that in vitro virus replication can be despite lower virus loads. FACS analysis revealed preserva- suppressed by CD8+ T cells from vaccines. The vaccination tion of central memory T cells. A SIV-specific central memo- + regimen was safe and well tolerated. ry subset of CD4 cells was preserved in these animals, Conclusion: MVA.HIVA efficiently boosts virus-specific CD8+ whereas it was absent in macaques with progressive disease. and CD4+ T cells with a favourable functional and phenotypic Conclusions: Vaccination with novel forms of DNAs during profile for viral control. Vaccinees are being enrolled in an ART induced an immune response able to control viremia extension phase to assess the effect of a third immunization on after ART release. Importantly, animals able to control virus plasma viral load and CD4 T cell counts during a short-term maintained this ability for two years after ART termination. supervised therapy interruption. This may enable us to identi- Protection correlated with preservation of SIV-specific central fy characteristics of protective T cell responses. memory cells. Optimized DNA vectors may be beneficial either alone or in combination with other vaccine modalities as an addition to antiretroviral treatment. AIDS Vaccine 2006 205 Poster sessions_revGJ 14/8/06 16:34 Page 206 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 13A: CLINICAL TRIAL SITE DEVELOPMENT – COHORT DEVELOPMENT P13A-01 P13A-02 Potential cohort for microbicide trials, Mukuru, Process analysis of a three-tiered community Nairobi, Kenya based volunteer recruitment model for the first AIDS vaccine trial in India JJ Bwayo and S Wakasiaka S Sahay1, M Phadke1, A Shrotri2 , T Gadgil1, B Patil1 and University of Nairobi, Department of Medical Microbiology, S Mehendale1 for the India IAVI Team2 Nairobi, Kenya 1 National AIDS Research Institute, Bhosari, Pune, India; 2 Background: HIV/AIDS is an increasing developmental crisis International AIDS Vaccine Initiative, Jor Bagh, New Delhi, India resulting from risky sexual behavior. Women are vulnerable to HIV infection due to lack power to negotiate for sex, lim- Background: Between October 2004 and December 2005, ited access to resources, preventive methods and treatment broad-based community contact model was implemented for options. There is consensus that women controlled method the first AIDS vaccine trial in India. At completion of recruit- e.g. effective microbicide is essential and ethically imperative ment in this trial, we present process analysis of decision-mak- in the control of global AIDS epidemic. This survey was con- ing by volunteers and investigators for screening and enrollment ducted to assess the feasibility of identifying potential high- in the trial. risk female population for microbicide trials. Objectives: To identify volunteers for Phase I preventive HIV Objective: To evaluate acceptability of IVR Microbicide Ring. vaccine trial in Pune, India through a three-tiered phased Methodology: A targeted baseline survey was conducted recruitment strategy. among female commercial sex workers (CSW) in Mukuru Methods: The recruitment model used multi-level community slum, Nairobi, Kenya between January to December 2005. contact process of educating the volunteers from community. Using a structured questionnaire data on sexual behavior, his- The first level community contact was made with 8383 individ- tory of STI and willingness for VCT uptake was collected. uals using group education strategy at the researchers’ initiative Approximately 350 CSW were identified through communi- to provide information on research, vaccines and vaccine trials. ty leaders and single women organizations. Subsequently, the interested potential volunteers were invited to Results: Respondents were mainly single (75%) aged between the institutional centers for second level meetings. Such sessions 20-40 years. Female workers, who double up as CSW, had [264 total] for small group [5–20 individuals] focused on fami- both regular (50%) and non-regular (95%) sex partners. ly involvement and socio-biological implications of participa- Majority (94%) admitted to past history of GUD and ure- tion. At the third level, 127 individuals were further assessed in thral discharges. Local clinics reported 30–50 cases of STDS one-to-one interviews to explain misconception, true and vol- per month. Although 90% had ever used condoms only 50% untary participation and false sense of security. Transport cost used condoms during recent sexual act with non-regular part- was reimbursed for second and third level meetings. Verbatim ners. Number of sexual partners was 1–5 (84%), 5–10 (10%) textual data was coded and analysed to understand the process and >10 (5%) per day. Nearly all (93%) exchanged money and constructs of decision to participate in the trial. for sex ranging from US dollar <1 (41%), 1–4 (57%) and Results: The step-wise process helped volunteers to take self 5–10 (1%) per day. According VCT among >200 CSW HIV decision to participate. Investigators could exclude the probable seroprevelance was 30%. ineligibles. Out of 99 informed volunteers willing to be Conclusion: This low socio-economic high-risk population is screened, 80 were actually screened, 41 were found eligible and potential for preventive trials. Studies in developing countries 30 were enrolled. Multiple reasons for screening ineligibility should be encouraged in order accelerate research and devel- included previous medical conditions (5), possibility of induce- opment of effective preventive interventions that may give ment (11), perceived compulsion to participate by spouse or women choices and retard the HIV pandemic. family (8) and non-eligibility due to non-medical reasons (4). Additionally, volunteers opted out due to subsequent lack of interest (14), fear (8) and emotional distress (4). By the time three volunteers decided to participate, the enrollment was over. Conclusion: Systematic phased pre-screening identification of groups and assessment helped in identifying volunteers with emotional distress, inducement, family/spousal pressure and fear. This prevented screening and enrollment of ineligible volunteers, possibly prevented future complications like social adverse events. Adhering to high ethical standards and maintaining transparency helped in educating community about vaccine trial, building community trust and conserving trial resources. 206 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 207 Amsterdam, Netherlands 29 August–1 September 2006 P13A-03 P13A-04 The importance of thorough and continuous Unique challenges and lessons learned in HIV vac- informed consent in HIV cohort studies: cine cohort development in Kericho, Kenya experiences from the Kericho HIV cohort study F Sawe1, L Langat1, S Kiplangat1, M Wasunna1, J Cheruiyot1, L Langat1, S Kiplangat1, M Wasuuna1, D Shaffer2, D Chuma1, R Ndirangu1, S Martin2, M Robb3, D Shaffer2, M Robb3, and D Birx3 N Michael3, and D Birx3 1 Kenya Medical Research Institute, Kericho, Kenya; 2 United 1 Kenya Medical Research Institute, Kericho, Kenya; 2 United States Army Medical Research Unit, Kenya; 3 United States States Army Medical Research Unit- Kenya; 3 United States Military Research Program, Rockville, United States Military Research Program, Rockville, United States Background: The US Military HIV Research Program is con- Background: In June 2003, the US Medical HIV Research ducting a 3-year, prospective HIV cohort study on agricultural Program began a 3-year, prospective, observational cohort plantation workers to determine HIV prevalence/incidence and study among adult tea plantation workers and dependents. subsequent vaccine research feasibility. Objective: To present unique obstacles encountered in the Objective: To share study informed consent (IC) experiences course of the study. and demonstrate ongoing assurance of volunteer consent for Methods: Community education held within the plantation study participation. to explain study objectives and procedures, risks and benefits Methods: Pre-Consenting Activities: Prior to IC for study with question and answer sessions at the end. Consent forms enrollment, “consenting team” trainings were held where IC were circulated 1 month prior to enrollment. Consent and processes were taught and attention was given to study spe- counseling was administered at one-on-one session. Follow- cific IC issues. Information sessions and “barazas” (commu- up is performed every six months. nity gatherings) were held where those interested were Results: 2801 volunteers were enrolled. Retention has educated about the study objectives, duration, procedures, remained above 77%. However, the following unique risks, and benefits. “Question-and-answer” sessions were obstacles have been encountered: Volunteer mobility: held where potential volunteers were given time to discuss Approximately 250 participants (8.9%) enrolled at base- study issues and concerns. Both Kiswahili and English IC line relocated off the plantation without notice or provid- documents were distributed to interested volunteers who ing contact information. Reasons: loss of employment due were encouraged to discuss openly with family and commu- to seasonal variations in productivity, retirement, social- nity members. legal issues (marital status, minor offenses); Legal chal- Enrollment day: Interested study volunteers were given a lenges: 13 participants (0.4%) sought local legal counsel one-on-one, detailed IC overview in either English or asserting physical/medical harm related to study participa- Kiswahili. A 10-question test of understanding covering the tion; Spousal hostility: In 32 months of follow-up, 2 female HIV study was administered with 90% correct responses participants (0.07%) reported verbal and/or physical required for study enrollment. Initiatives were put in place to assault (slapping, physically held back) by spouses regard- assure all interested volunteers could complete the IC process ing study follow-up. Neither suffered physical injury, and (e.g. thumb prints with independent witnesses for illiterate both continue in the study. subjects, consenters fluent in several local languages). Discussion and conclusions: Of those discovered to have relo- Post-consenting Activities: In addition to information shared cated, 200 (80%) were identified and returned for follow-up. in the consenting process about HIV testing, individual, com- Strategies used included use of field staff to find relocated prehensive HIV pretest counseling was conducted at each volunteers and facilitate follow-up. Two interventions visit to assure volunteers’ understanding and approval of addressed the legal issue: lawyers from the Kenya Medical HIV testing. Also, volunteers’ rights to withdraw from the Research Institute (KEMRI) became involved immediately study at any time without loss of benefits were discussed at and responded to the legal claims. No claims advanced each visit. beyond KEMRI and community lawyers interchange. Results: 2801 volunteers were recruited. One was unable to Secondly, study team proactively scheduled general HIV and complete the IC process due to a language barrier. Over 98% informed consent educational sessions for community of the volunteers scored 100% on the test of understanding. lawyers lead by CAB Chairman and KEMRI IRB Chair. Conclusions: IC is a continuous process starting prior to Assault incidents were immediately addressed by the study study commencement and continues throughout a study. PI, medical staff and counselors to assure volunteer safety. Adequate training of a “consenting team,” attention to pre- Referral to the study medical monitor is made. Volunteer consenting activities, facilitating a thorough consent process spouse education follows. Community education continues and understanding at time of enrollment, and continuous about HIV and HIV research. Such unique obstacles must be attention to the importance of volunteers continued consent promptly and judiciously addressed within the cultural study for study participation are all critical components in assuring context while adhering to local and international research adequate IC for HIV cohort study participants. guidelines in order to assure study integrity. Such lessons are valuable learning experiences for ongoing vaccine research. AIDS Vaccine 2006 207 Poster sessions_revGJ 14/8/06 16:34 Page 208 Amsterdam, Netherlands 29 August–1 September 2006 P13A-05 P13A-06 Qualifying a vaccine trial laboratory in Rwanda Developing a cohort and evaluating of HIV preva- for PBMC isolation, cryopreservation and shipping lence and HIV incidence rates among injection drug users in St Petersburg, Russia J Bizimana1, E Karita1, MJ Boaz2, T Tarragona3, J Gilmour2, J Stout2, E Shutes1, E Tekirya1, AP Kozlov1, AV Shaboltas1, OV Toussova1, V Musengamana1, BH Bekan1 E Hunter4 and S Allen4 SV Verevochkin1, I Hoffman2, B Masse3 and T Perdue2 1 Project San Francisco, Kigali, Rwanda; 2 IAVI, New York, US; 3 1 The Biomedical Center and St.Petersburg State University, IAVI Core Laboratory, London, UK and 4 Emory University, St.Petersburg, Russia; 2 School of Public Health, University of Alabama, US North Carolina, USA; 3 Statistical Center for HIV/AIDS Research and Prevention (SCHARP) Fred Hutchinson Cancer Research Background: Successful development of an AIDS vaccine Center, USA requires quality-assured, robust data pivotal for evaluating vaccine safety and immunogenicity. Background: Unsafe drug injection practices are the major route Objectives: Prior to initiation of a clinical vaccine trial with of HIV transmission in Russia since 1996. Russia experiences centralized immunogenicity testing, participating immunolo- the lack of epidemiological and other data on IDUs. No longi- gy laboratories must be competent in isolation, cryopreserva- tudinal cohort data exist. tion and shipment of PBMCs, thus ensuring vaccine Objective: The objective of NIH HPTN 033 study was to estab- immunogenicity can be assessed. lish and describe an HIV incidence cohort of IDUs in Methods: In preparation for an IAVI sponsored Phase I HIV St.Petersburg to evaluate HIV prevalence and HIV incidence vaccine trial, Projet San Francisco, Rwanda, in collabora- rates and preparedness for the future longitudinal studies. tion with IAVI, developed an immunology laboratory. Staff Methods: HIV uninfected IDU with a history of injecting ≥3 × was recruited, equipment was installed and training provid- week for the last month or sharing injection equipment ≥3 times ed on site and in London. The laboratory was enrolled into in the last 3 months were enrolled and followed for 12 months. accreditation and QA programs, part of the IAVI accredita- HIV testing, along with risk assessment survey occurred at tion procedure was evaluation of site competence in PBMC enrollment, 6 and 12 month visit. Subjects were provided unlim- isolation. An IAVI qualifying run was undertaken in which ited access to primary medical care, HIV counseling and refer- blood was drawn from four healthy volunteers and using rals to available drug treatment and needle exchange programs. IAVI Core Laboratory SOPs, PBMCs were isolated, cryop- Results: Between March and December 2002, 520 HIV negative reserved and shipped in liquid nitrogen to the IAVI Core IDUs were enrolled into the cohort from 900 screened individu- Laboratory, London. als. HIV prevalence rate at baseline was 30%. Most enrolled Results: Each of the five criteria for passing the IAVI qualify- subjects were young, male, heroin users with frequent experi- ing run was satisfied. At the site in Rwanda, the yield of ence of risky sharing behaviors. HIV incidence rate after 12 PBMC isolated per ml of blood for each volunteer ranged months of follow up period was 4.5/100 person-years (95% CI, from 0.9 to 1.5 million cells. The recovery of PBMCs in 2.7–7.0). In univariate analysis, psychostimulant use, especially London after overnight rest, determined using an automated frequent use, three or more sex partners in the last 6 months, cell counter, ranged from 59–123%, with viability of 96- and females selling sex were associated with HIV seroconver- 97%. No red blood cell contamination was evident. sion. In the multivariate analysis, psychostimulant use three or Assessment of PBMC performance in a qualified IFNγ more times per week was the only factor still associated with ELISPOT assay further showed good quality in all samples: HIV seroconversion. The most effective recruitment strategy in 6 low backgrounds (<37.5 SFC/10 ), positive responses to PHA terms of bringing participants into the cohort was social net- 6 (>949 SFC/10 ), and to the physiologic control flu, EBV and work approach (76%). Of the 520 subjects, 417 (80%) had a CMV peptides. Initiation of the IAVI V001/PAVE vaccine 12 month visit or HIV seroconverted during the follow-up peri- trial in Rwanda has continued to demonstrate the compe- od. Among 103 participants lost to follow-up 11 (10.7%) died, tence of the Rwanda site and laboratory in PBMC isolation, 31 (30.1%) were incarcerated, 3 (2.9%) were hospitalized, 3 cryopreservation and shipping. (2.9%) stopped using drugs and refused participation, and 55 Conclusion: Following training and site development with (53.4%) stopped participating for other or unknown reasons. IAVI, the Projet San Francisco site in Rwanda successfully Conclusions: The rates of HIV prevalence and HIV incidence passed the IAVI qualifying run showing competence in PBMC among IDUs in St Petersburg indicates the urgent need to pro- isolation, cryopreservation and shipping. This was one criti- mote interventions targeting psychostimulant and heroin users. cal aspect in validating the site to conduct an HIV vaccine The HPTN research site and IDU cohort established at the trial. Subsequent initiation of an HIV vaccine trial in Rwanda Biomedical center in St Petersburg along with developed effec- has provided further confirmation for the quality of this site tive recruitment/retention strategies and trained personnel are laboratory to perform within HIV vaccine trials. prepared for conducting any type of HIV prevention interven- tions including vaccine trials. 208 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 209 Amsterdam, Netherlands 29 August–1 September 2006 P13A-07 P13A-08 Expansion and maintenance of an HIV discordant Lessons learned in developing research counseling couple cohort in Kigali, Rwanda in preparation for and testing (RCT) in East Africa vaccine efficacy trials N Bahati1, J Nyange2, O Anzala2, MR Rwanyonga2, E Kestelyn2, A Tichacek1, E Karita2, K Kayitesi2, A Kamali2, H Oguto2, B Bender1, A von Lieven3, E Shutes2, J Bizimana2, E Chomba3, P Fast4 , S Allen1 E Sanders4, S Hannah1, M Price5 amd S Kalibala1 and the Rwanda Zambia HIV Research Group1 1 International AIDS Vaccine Initiative, Country and Regional 1 Emory University, Atlanta, GA, USA; 2 Projet San Francisco, Kigali, Programmes, Nairobi, Kenya; 2 Kenya AIDS Vaccine Initiative, Rwanda; 3 Zambia Emory HIV Research Project, Lusaka, Zambia; Nairobi, Kenya; 3 International AIDS Vaccine Initiative, Nairobi, 4 International AIDS Vaccine Initiative, New York, NY, USA Kenya; 4 International AIDS Vaccine Initiative, Kilifi, Kenya; 5International AIDS Vaccine Initiative, San Mateo, United States Objectives: The Global HIV/AIDS Vaccine Enterprise identi- fied “expanding access to large, well-defined populations of Background: In working towards large scale HIV vaccine uninfected people at high risk of HIV infection” as a gap in clinical trials, eight sites in East Africa and Zambia have col- the clinical trials capacity of developing countries. With the laborated on a number of research protocols, providing HIV majority of new HIV infections in Africa occuring within VCT to over 10,000 volunteers. Unlike traditional VCT, cohabiting couples, expanding cohorts of discordant couples RCT often requires multiple follow-up visits and includes to conduct HIV prevention trials is a priority. consent for data collection. Methods: Projet San Francisco in Kigali, Rwanda expanded Objectives: To assure high quality counseling, excellent data couples’ voluntary counseling and testing (CVCT) in 2003 to management, and support for counseling. recruit HIV discordant couples for prevention trials. Methods: The work describes innovative research counsel- Outreach is through word-of-mouth and invitation by ing models developed following a three-tiered process. trained outreach workers. After testing at one of three CVCT First, Focus Group Discussions (FGD) were conducted centers, discordant couples who have cohabited at least three with nurse counselors and principal investigators at each months are invited to enroll in an open cohort. The HIV neg- site. Following main challenges emerged from FGD: repet- ative partner is followed quarterly, and couples receive repro- itive VCT procedures created volunteers’ burnout; tradi- ductive outpatient care and ARV screening. Couples are tional VCT trainings did not equip counselors in long-term recruited from the established cohort for vaccine and non- follow-up counseling. Lack of support supervision resulted vaccine clinical trials. in counselors’ burn out. Exposure to high-risk cohort, such Results: From 2003 to 2005, over 51,000 CVCT invitations as CSW and MSM exposed counselors’ values in service were distributed by INAs (Influence Network Agents) and provision. Collection of behavioral data during the coun- 25,522 couples were tested. Among those tested, 3176 (12%) seling session was not linked to prior counseling research couples were discordant (1329 male HIV+, 1847 female questions. Second, RCT procedures and training plans HIV+) and 1423 (45%) discordant couples were eligible for were developed to address challenges. Third, implementa- cohort enrollment. Most discordant couples not eligible tion of RCT specific procedures and training plan is under- either did not meet cohabitation requirements, or were deter- way and monitoring systems are in place. mined not to be true couples. Among those eligible, 1190 Results: It is found that relationship-based counseling which (84%) enrolled. In December 2005 there were 1059 discor- promotes one-volunteer, one-counselor service is critical for dant couples in active follow-up, with retention >90% at 12 good data and retention in research studies. Counseling months. Since 2002, 50 seroconversions have been identified record keeping should be linked to behavioral research ques- despite counseling (HIV incidence rate 3.6 cases/100py, 95% tions. Supervision for RCT should provide counselor psycho- CI 2.6 to 6.8). In 2006, enrolled couples with the infected logical support, build their ability to provide follow-up partner on ARV were released from follow-up to consolidate counseling and reinforce behavioral research techniques. health care at the ARV clinics. Screening and enrollment for Trainings should emphasize self-awareness, and reinforce the a Phase I HIV vaccine trial began in November 2005. ethical issues key to maintaining a long-term counseling rela- Conclusions: The establishment and retention of a well- tionship. Exit interviews of volunteers by a non-study staff defined, high-risk cohort in preparation for vaccine efficacy person should be used to inform RCT service delivery. trials is possible in developing countries. Recruiting trial Conclusions: Quality HIV counseling and testing models in volunteers using a run-in design from an existing cohort research context requires research specific training approach- with a known HIV incidence can improve volunteer follow- es and QA that meet both volunteers’ psychosocial needs and up and compliance. Eligibility criteria at CVCT and during research needs. Continuous volunteer and staff feedback is cohort maintenance can be tailored to meet the needs of important to meeting that goal. specific studies. AIDS Vaccine 2006 209 Poster sessions_revGJ 14/8/06 16:34 Page 210 Amsterdam, Netherlands 29 August–1 September 2006 P13A-09 P13A-10 Challenges for the enrollment in a Phase I trial Recruiting African Americans and Latinos into HIV vaccine in Haiti HIV vaccine trials SS Jean1, G Perrin1, DH Louis1, MM Deschamps1, SG Arrealo1, C Campbell1, G Morrow-Hall1, B Jackson1, DW Fitzgerald2. WD Johnson2, PF Wright3, S Buchbinder4, E Molina1, J Sarche1, S Buchbinder1, S Oxendine1 and S Hammer5, L Corey6 and JW Pape1,2 E McLellan-Lemal1 1 Groupe Haitien d’Etude du Sarcome de Kaposi et Infections 1 HIV Research Section, San Francisco Department Of Public Opportunistes GHESKIO Centers, Port-au-Prince, Haïti ; 2 Health, CA, USA; 2 Center of Disease Control and Prevention, Division of Internal Medicine and Infectious Diseases, Weill Atlanta, GA, USA Medical College of Cornell University, New York, NY, USA; 3 The Division of Pediatric Infectious Disease, Vanderbilt University, Background: African Americans and Latinos are underrepre- Nashville. TN, USA; 4 San Francisco, Department of Public Health sented in HIV vaccine research even though they have the AIDS Office San Francisco, CA, USA; 5 Columbia University, highest prevalence and incidence rates of HIV in the United Division of Infectious Diseases New York, NY, USA; 6 University States. Although there are some indications as to why African of Washington Virology Division FHCRC Program of Infectious Americans and Latinos may not participate in clinical Diseases, Seattle, WA, USA research, we have limited knowledge on approaches for increasing their participation in HIV vaccine trials. Objectives: To examine motivators and barriers of HIV-vac- Objectives: Identification of volunteers meeting the eligibility cine trial participation among African Americans and Latinos criteria for Phase I trial is challenging particularly in low litera- toward informing new recruitment strategies for ongoing cy population requiring education of volunteers in all aspects of HIV vaccine trials locally; and to document a methodology the protocol and informed consent as well as technical training that addresses concerns of African American and Latino for proper completion of the Vaccination Report Card (VRC). communities regarding research participation. This abstract describes the challenges for the enrollment process Methods: To develop a qualitative interview guide, consulta- in HVTN 050 at GHESKIO Center, Haiti tions with 33 African American and Latino community lead- Methods: Our HIV voluntary counseling and testing center ers were undertaken. Thirty-six audio recorded individual (VCT) is the entry point for enrollment in HIV vaccine trials. interviews were then conducted in English or Spanish with Participants are referred from the VCT to the HIV Vaccine Trial African American and Latino men and women, 18 to 50 Unit (HVTU) where they are informed, educated, and trained years of age, residing in the San Francisco Bay Area. about key aspects of the protocol and informed consent process. Interviews were transcribed and coded for major themes A standardized test on their knowledge and desire to participate related to study objectives. Methodological processes were in the trial is administered by a team not involved in the train- documented in an effort to make the study methods available ing. Those who passed are allowed to sign the consent form. In and replicable in other sites. addition, potential volunteers are trained on how to fill out the Results: Barriers and motivators of HIV vaccine participation VRC. Appropriate educational materials were developed specif- centered around three themes: cultural competency (e.g., flu- ically for that purpose. ency in the participant’s language, avoidance of scientific Results: From February 2003 to March 2006, 23,648 HIV-per- terms, recognition that family concerns are primary and ben- sons, aged 18–50 years have been informed about HIV vaccine efits go beyond the individual); knowledge of HIV research trials and 9,588 were referred to the Vaccine Unit. Only 18.5% and vaccines; stigma; and HIV vaccine specific fears (e.g., (1771) actually arrived at the unit and 1239 who took the will- vaccine causes HIV). ingness questionnaire were willing to take part (70%). Of the Conclusions Themes identified in these interviews suggest first 491/1239 pre screened, 336 (68%) were excluded for var- that trust as well as comprehension and perceptions about ious reasons mostly medical (62%), non-availability (28%), safety account for limited participation of African Americans high risk activity (7%) . The remaining 155 completed the train- and Latinos in HIV vaccine trials. Recruitment campaigns ing sessions: 134/150 who took the standard evaluation test should focus on family, altruism and group cohesion using passed and 81 were screened officially. Eight withdrew their language that is relevant to African Americans and Latinos. consent. Of the 73 remaining volunteers, 53 (72.5%) had other Campaigns also need to address misinformation, stigma and abnormal laboratory values and 35 were enrolled specific HIV vaccine fears. As researchers increase their own Conclusion: The enrollment in HIV vaccine trials in developing cultural competency working with African Americans and country setting can be done but it is challenging and requires Latinos and address specific barriers, trust in research from intensive and continuous efforts. Of 491 evaluated, only 73 both groups could increase, understanding of HIV vaccine (15%) remained in the process and 35 were enrolled. Low lit- research could improve, and levels of fear might decrease so eracy populations can be successfully screened for HIV vaccine that successful recruitment of African Americans and Latinos trials, although the process is labor-intensive, and requires ade- in HIV clinical trials may occur. quate staff infrastructure and training. 210 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 211 Amsterdam, Netherlands 29 August–1 September 2006 P13A-11 Adolescent sexuality, stigma, and potential bene- fits of trial participation: views on adolescent par- ticipation in HIV vaccine trials G de Bruyn, N Skhosana, J McIntyre and G Gray Perinatal HIV Research Unit, University of the Witwatersrand, Soweto, South Africa Objectives/Aims: Under South African law, parental consent is necessary for participation of adolescents in clinical trials. Therefore, parental attitudes towards such studies and study procedures are likely to be key determinants of their willing- ness to allow adolescent participation. However, parental views on topics such as adolescent sexuality may conflict with those of adolescents. We undertook an exploratory study of the perceptions that may influence adolescent par- ticipation in HIV vaccine trials in Soweto, South Africa. We focus here on perceptions and attitudes around stigma, ado- lescent sexuality, and potential benefits of participation. Methods: Participants included adolescents and adults who are parents of adolescents or teachers. Adolescents were recruited in selected Soweto high schools; adult recruitment used snowball sampling and community-based activities. A self-administered, facilitated questionnaire was completed. Results: 343 respondents participated (54% female; 71 adults, 272 adolescents). Regarding issues of stigma, most adults and adolescents felt that potential discrimination, being avoided as a result of trial participation, being per- ceived as at high risk of HIV, or being thought to have AIDS would not influence their decision to participate or to allow their child to participate in trials. Adults and adolescents both indicated that access to regular counseling and testing, current information, and potential impacts on improving motivation to reduce risk behavior were very important for determining willingness to participate. Fewer adolescents thought that HIV is a threat to children from age 9 (43% ver- sus 70%, P<0.0001) and fewer thought that parents should give permission for adolescent HIV testing (57% versus 89%). Most adults indicated that they would want to know if their child is sexually active (91%). Discussion/Conclusions: Adults and adolescents tended to agree on whether potential trial-related stigma, discrimina- tion, or benefits of participation would influence their deci- sion to participate or to let their child participate in HIV vaccine trials. Disagreement was noted on items related to permissions for testing and risk perception. Participant responses are encouraging that recruitment of adolescents in HIV vaccine trials will find support from both adolescents and parents. AIDS Vaccine 2006 211 Poster sessions_revGJ 14/8/06 16:34 Page 212 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 13B: CLINICAL TRIAL SITE DEVELOPMENT – EPIDEMIOLOGY STUDIES P13B-01 P13B-02 HIV-1 prevalence, incidence and molecular char- Knowledge, attitudes, and practices study in mar- acteristics at a feasibility study in a voluntary ketplaces of Bamako, Mali to inform the design of counseling and testing site (VCT) at south Brazil HIV vaccine trials LFM Brígido1, R Rodrigues1, M Thomaz2, CAF Oliveira1, M Koné2, C Cohen1, A LaFaunce1, N Vu1 , M Lally1, MR Almeida2, TS Ito2, ERV Chomatas2, CM Oliveira1, D K Mayer1, N Keita3 and AS De Groot1 Harrad2, ACA Evangelista2 1 Brown University, Providence, RI; 2 FMPOS Medical School, 1 Adolfo Lutz Institute, Sao Paulo, Brazil; 2 Municipal Aids Mali; 3 University of Bamako, Mali Program, Curitiba, Brazil Context: Little is known about willingness to participate Background/aim: Health innovations testing sites in areas (WTP) in HIV vaccine trials in West Africa. Culturally appro- where multiple clades circulate may be important to evaluate priate vaccine trials are necessary in diverse locations to pro- issues as cross clade vaccine efficacy. Curitiba, a major met- duce a vaccine that works against all clades of HIV, including ropolitan area in South Brazil has a good health infrastruc- those found in West Africa. ture and a population of different ethnicities, and could Objectives: To assess knowledge and practices around HIV, become an interesting site for future Phase III trials. As part WTP in vaccine trials, and reasons for participation in the of a feasibility evaluation (www.rcp-hiv.org), we performed market-going population of Bamako, Mali. risk assessment, serological evaluation and preliminary mol- Methodology: A cross-sectional oral survey modelled after ecular characterization at the central VCT unit. prior KAP/B studies was conducted in French or Bambara Methods: A random sampling of VTC users was recruited with willing participants in 27 marketplaces in Bamako, Mali and, after informed consent, risk assessment and serology in August 2005. Malian medical and anthropology students were performed. Virtually all assessed users agreed to partic- trained in research ethics and protocol administered surveys. ipate in the study. HIV-1 seropositive samples (according to Sample size (n)= 279. national guidelines, www.aids.gov.br) were further evaluated Select Results: A high percentage of respondents were willing using BED-EIA (Calypte, USA) according to manufactures. to participate (62.2%), with higher WTP in older people Pol genomes (protease and partial RT) were sequenced using (>41) compared with younger (<21) (OR= 6.187, 95% CI nested PCR followed by dye termination, resolved in ABI [1.302–29.400]). There was higher WTP in the medium 3100. Quality and assembly was performed at socioeconomic status (SES) group versus the high SES group www.ial.sp.gov.br, resistance mutations determined at (OR=1.931, 95%CI [1.077–3.462]). Those with a high http://hivdb.stanford.edu and HIV clade analysed at school education were twice as likely to refuse participation //jose.med.kuleuven.be. Simplot 2.5 and Paup 4.02 were used than those with no education (OR=2.084, 95%CI to evaluate recombinants and additional phylogenic analyses. [0.751–5.785]). Top-cited reasons for participation included Epi6 was used for data analyses. perceived self-protection from HIV imparted by the trial vac- Results: Out of 791 patients tested, HIV-1 prevalence was 4.3% cine (73.1%) and contributing towards an efficacious vaccine (95%CI: 3–6). Using CDC consensus BED analysis, the esti- (81.0%). Fear that an HIV vaccine could cause HIV was not mated annual incidence was 2.8% (95%CI: 1.4–6). Pol clade a common consideration in deciding WTP (7.5%). was determined in 82% of infections; 67.8% B, 28.6% C and Knowledge of HIV was also assessed, with 82.8% respond- one case of CB mosaic. No principal mutations at protease were ing correctly to more than half the medical questions. observed, but one clade B sample showed the K103N RT muta- Conclusions: Marketplace visitors in Mali indicated a high tion. Half of clade C, 16% of clade B and the BC mosaic WTP in HIV vaccine trials. Participants were knowledge- showed reactivity of recent infection at BED-EIA. able about HIV and would want to participate in the effort Conclusions: These preliminary results confirm previous to find an effective HIV vaccine. Perceived self-protection observations of HIV-1 clades C and B co-circulation. HIV-1 from a potential HIV vaccine was high, and vaccine trials in C seems to be associated with incidence cases but recently Mali must carefully address this belief to ensure that vac- reported technical issues associated BED assay suggests cau- cine trial participants do not put themselves at additional tion on this interpretation. Larger sample size, and in special, risk. The SES and education trends associated with WTP sequential evaluation of these cases after maturing serological indicate that education about candidate vaccines and response will allow a better understanding of this issue. The informed consent are essential. interest of the community and the ability of the health unit to conduct the study suggest that this setting may provide an interesting environment for future vaccine trials. 212 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 213 Amsterdam, Netherlands 29 August–1 September 2006 P13B-03 P13B-04 Necessity of region-specific normal reference A cross sectional, observational study to establish ranges for conducting HIV clinical trials clinical laboratory reference ranges in adults as part of a multi-center study in preparation for L Eller1,2, B Ouma1, M Eller1,2, W Sateren2, B Bagaya1, HIV vaccine efficacy trials P Oballah1, P Naluyima1, A Natwijuka1, F Wabwire- Mangen1, P Kataaha3, M DeSouza2, N Michael4, M Robb2 A Kamali1, K Kayitenkore2, A Nanvubya3, E Shutes2, and H Kibuuka1 S Allen4, G Stevens5, O Anzala6, J Mulenga7, P Kaleebu3, E Ruzagira1, M Price8, C Kambili8, N Ketter8, LS Johnson, 1 Makerere University Walter Reed Project, Kampala, Uganda; 2 J Scott9, E Karita2, P Hughes1 and P Fast8 Henry M. Jackson Foundation, Rockville, MD, USA; 3 Nakasero Blood Bank, Kampala, Uganda; 4 Walter Reed Army Institute of 1 MRC/UVRI Uganda Research Unit on AIDS, Masaka, Uganda; 2 Research, Rockville, MD Project San Francisco, Kigali Rwanda; 3 Uganda Virus Research Institute, Entebbe Uganda; 4 Emory University, Atlanta USA; 5 Objectives: To establish region specific normal reference International AIDS Vaccine Initiative, Johannesburg, South ranges to be used for future vaccine trials in Kampala, Africa; 6 Kenya AIDS Vaccine Initiative, Nairobi Kenya; 7 Zambia Uganda. Emory HIV Research Project, Lusaka Zambia; 8 International Methods: Samples were obtained under an approved human AIDS Vaccine Initiative, New York; 9 University of Oxford use protocol from blood-bank donors in Kampala, Uganda after completing donor affidavit and undergoing bloodbank Background: HIV vaccine trials are being planned in Africa, screening processes. Excess blood remaining in the collection these studies will however require African laboratory refer- tubing was used for the analysis. Additional laboratory ence ranges to determine relevant inclusion and exclusion cri- screening was performed to exclude samples positive for HIV, teria and to interpret safety data. The objective of the study HBsAg, Hepatitis C antibody, and hCG. Serum chemistry was to establish clinical laboratory reference ranges among was performed using a Roche Cobas Integra 400 Plus ana- asymptomatic adults at multiple sites in Eastern and lyzer and hematology was performed on EDTA samples using Southern Africa. the ACT 5 diff analyzer. Data mean, median, and 2.5–97.5% Methods: A total of 2400 (50% men) adult clinically healthy ranges were calculated volunteers (18–60 years) are being enrolled across 7 sites: Results: A total of 960 samples were collected from 757 Kigali, Rwanda; Masaka and Entebbe, Uganda; Lusaka, males and 203 females. 72 (7.5%) were excluded due to lab- Zambia; two sites in Nairobi and one in Kilifi, Kenya. After oratory abnormalities (6% HBsAg, 0.8% HCV Ab, 1% HIV, informed written consent all volunteers undergo a compre- 2.6% hCG). Gender specific ranges for WBC parameter hensive medical history and physical examination and socio- ranges were (units 10ˆ3 cells/ul): WBC (M:2.8–8.2, demographic questionnaire. Urine and blood samples are F:3.2–9.0); neutrophils (M:1.1–4.4, F:0.9–3.8); eosinophils collected from all volunteers. Urinalysis testing in all volun- (M:0.04–1.7, F:0.04–1.4); lymphocytes (M:1.2–3.6, teers and pregnancy testing for females is performed. Blood is F:1.3–3.7); monocytes (M:0.2–0.7, F:0.2–0.6); basophils tested for HIV, syphilis, hepatitis B and C serology, hematol- (M:0.01–0.08, F:0.01–0.07). Additional hematology parame- ogy, biochemistry and CD4 counts. Exclusion criteria include ter ranges included: RBCs (M:3.8–6.1, F:3.3–5.3 10ˆ6 acute or clinically significant symptoms, syphilis or HIV pos- cell/ul), Hgb (M:11.6–17.1, F:9.8–16.2 g/dL), Hct itive, and evidence of Hepatitis B or C infection; pregnant or (M:33.8–49.5, F:28.3–45.8%), Plt (M:106–362, F: menstruating females are also excluded. 137.7–457.3 10ˆ3 cells/ul). Chemistry analytes included ALT Results: Enrollment has been completed at 2 sites (Kigali and (M: 7.2–43.3, F:5.3–39.9 U/L) and creatinine (M:0.6–1.2, Entebbe) and 597 volunteers (approximately 50% men) are F:0.5–0.9 mg/dL). available for analysis. The CD4 count median value (in In a recent Phase I vaccine trial conducted in Kampala, brackets are reference ranges 2.5 and 97.5 percentiles) among Uganda, 239 subjects were screened in order for 31 to be females in Kigali was 984 cells/µl (595–1634) and in Entebbe enrolled. 60 subjects were excluded as a result of haemato- 890 (597–1845). The corresponding values among males logical abnormalities based on US reference ranges. 54 of were 803 (494–1318) and 803 (397–1281) respectively. The these excluded subjects could have been enrolled if above absolute differences with “Western” values for most bio- ranges had been in place at the time of screening. chemistry and heamatology values were modest and differ- Conclusions: Region-specific reference ranges for chemistry ences in median not clinically significant. and hematology are critical for vaccine trial implementation. Conclusions: The study findings will provide valuable data International sites must develop region-specific ranges to on laboratory values that are critical in the design and moni- inform vaccine protocol inclusion/exclusion criteria. Toxicity toring of clinical trials, particularly trials of preventive tech- tables must also be adapted to reflect region-specific refer- nologies and vaccines for HIV, among African populations. ence ranges in order for clinical trials to be both operational- ly efficient and properly inclusive of community volunteers. AIDS Vaccine 2006 213 Poster sessions_revGJ 14/8/06 16:34 Page 214 Amsterdam, Netherlands 29 August–1 September 2006 P13B-05 P13B-06 Community Heath Forum: a method of communi- Further vaccine related analysis of the Buenos ty education in the Prime-Boost Phase III HIV Aires men who have sex with men (MSM) cohort. vaccine trial in Thailand Argentina V Chaiwong1, N Premsri1, C Namwat1, C Khamboonruang, MM Avila1 , M Segura1, S Maulen2, S Sosa Estani1, M S Rerks-Ngarm1, S Wattana2, W Wiriyakijja3, Fernandez 3, R Marone 2 , JL Sanchez 4 and M M Benenson4, J Kim4 Weissenbacher1 1 Ministry of Public Health – Thai AIDS Vaccine Evaluation 1 Centro Nacional de Referencia para el SIDA, Departamento de Group (MOPH-TAVEG), Thailand; 2 Chon Buri Provincial Health Microbiología, Facultad de Medicina, Universidad de Buenos Office, Thailand; 3 Rayong Provincial Health Office, Thailand; 4 Aires, Buenos Aires, Argentina; 2 NEXO Asociación Civil, Buenos Armed Forces Research Institute of Medical Sciences (AFRIMS) Aires, Argentina; 3 Hospital de Clínicas José de San Martín, Universidad de Buenos Aires, Buenos Aires, Argentina, 4 Anteon Background: A Prime-Boost Phase III HIV Vaccine trial has Corporation, Frederick, MD, USA been conducted in Thailand since September, 2003. The trial has screened over 26,000 potential volunteers and enrolled Background: MSM represent a significant proportion over 16,400 volunteers in December 2005. “Community (22.3%) of AIDS cases in Argentina. High HIV (13.8%, health forums” involving the community leaders in the study 2001–02) and HBV (37.7%, 2001–02) prevalence as well as area were used to provide understanding and support for the HIV incidence rates (6.0%, 2001–02; 3.9%, 2003–04) have trial and to foster support for volunteers’ participation. been recently documented. Objective: To demonstrate the pattern and evaluated the out- Objectives: To evaluate recruitment efforts and retention rates, come of community health forum which were held toward willingness to participate in HIV vaccine trials and willingness the end of enrollment period. to receive HBV vaccination. Methods: A 2-way communication technique was used in Methods: Gay men aged 18–60 years who self reported HIV- group discussion along the forum activities. The participants negative were recruited during February-December 2003. All were provided the opportunity to ask questions and make men were screened for HIV, HBV, HCV, and VDRL reactivity. comments about the trial. Post activities questionnaire were Anti-HBsAg testing was performed after HBV vaccination. used for evaluating the result of forums. Results: Of a total of 877 MSM screened, 66 (7.5%) were Results: There were “Seven Key Messages”, 1. Why do we found to be infected for HIV, 209 (23.8%) for HBV, 7 (0.8%) need an HIV Vaccine?; 2. What is the vaccine?; 3. How do we for HCV and 60 (6.8%) for T. pallidum. Out of the 66 HIV- develop the vaccine?; 4. Why are we doing this trial?; 5. How infected, 50 (75.8%) had a previous negative HIV serology and can we evaluate the efficacy?; 6. How can we succeed togeth- 16 (24.2%) had not been tested previously. er?; and 7. That the community and the volunteers are the Of 327 HIV-negative MSM recruited in the cohort study, reten- stake holders of the trial. These messages were developed and tion rates at 6 and 12 months were 97.2% and 91.5%, respec- used so the community would understand HIV-AIDS situa- tively. Twelve MSM seroconverted (HIV incidence rate of 3.9 tion, including the HIV-vaccine development process. These per 100 person-years; 95% CI = 2.0–6.7). messages would also stimulate discussion during the forum At the 12-month visit 291 men were interviewed and 290 activities. Twenty-two community health forums were con- (99.7%) agreed to answer the questionnaire on willingness to ducted between September 2005 and February 2006. 1430 participate in a vaccine trial. Of these 290, 60.3% responded participants, consisting mostly of community leaders and vil- that they would participate. lage health volunteers (93%) joined the forums. Other atten- 246 HBV negative volunteers (75.3%) were offered HBV vac- dees were volunteers themselves and/or their family members. cination: 6.7% were already vaccinated and among the rest, Most of participants (96%) had ever heard about the trial 92.3% accepted and 89.0% completed the three doses. A total before the forum but did not clearly understand what the vac- of 172 (93.5%) of 184 subjects who received 3 doses of HBV cine is made from, or the benefit to the community from the vaccine were tested and 146 (84.8%) were responders. trial. At the conclusion of the forums 97% of the participants Conclusion: A high level of HIV infection was noted among stated that they would support the trial activities. MSM who felt they were uninfected (7.5% were HIV positive), Conclusions: Community understanding is an essential ele- probably indicating a certain lack of knowledge and unwilling- ment in conducting a vaccine trial. Since the information of ness to know their HIV status. In a setting of highly educated, HIV vaccine trial is complicated and easily misunderstood, counseled and where prevention measures (condoms) are read- face to face communication between the researchers and ily available, high HIV incidence rates continue to be docu- community is needed. Community health forum conducted in mented. Around two-thirds of this population of gay men is this trial demonstrates a method that can used to enable the willing to participate in future HIV vaccine trials. In addition, community to understand about the vaccine trial. though HBV infection continues to be a big problem, HBV vac- cination rate was very low. However, when offered, high rates of acceptance and good immune response were documented. 214 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 215 Amsterdam, Netherlands 29 August–1 September 2006 Renewed efforts in conducting candidate HIV vaccine, and implementing HBV vaccination, need to be made to stem the threat from these two agents. AIDS Vaccine 2006 215 Poster sessions_revGJ 14/8/06 16:34 Page 216 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 13C: CLINICAL TRIAL SITE DEVELOPMENT – PREVALENCE/INCIDENCE/TRANSMISSION STUDIES P13C-01 P13C-02 Volunteer recruitment strategies in HIV vaccine South African community members’ perceptions trials. An experience from the Botswana HIV of enablers and inhibitors to participation in Vaccines Trials Unit (BHVTU) HIV/AIDS vaccine trials L Mwananyanda, T Villafana, K Sebinang, O Tabengwa, A Kagee, Z Kafaar, A Lesch and L Swartz U Mooketsi, R Hambira, J Makhema, I Thior and M Essex Stellenbosch University/ South African AIDS Vaccine Initiative, Botswana Harvard School of Public Health Partnership, Stellenbosch, South Africa Gaborone, Botswana Objectives/aim: The aim of this study was to identify the Background: Botswana, with an estimated adult HIV preva- enablers and inhibitors to HIV vaccine trial enrollment lence of 35% was the first country to begin a Phase I HIV among South African communities. We present the findings vaccine trial in the southern Africa region. Two Phase I trials of a qualitative investigation into the factors thought to of 2 experimental vaccines, a multiepitope DNA plasmid enable or inhibit participation among persons eligible to (HVTN 048) and an Alphavirus vectored vaccine (HVTN enrol in a future HIV vaccine trial in South Africa. 059) have been conducted at the BHVTU. Methods: Thirty-seven semi-structured interviews and two Objectives: To describe different recruitment strategies into focus groups were conducted with trial site community mem- HIV Vaccine trials at the BHVTU. bers who had attended HIV vaccine education workshops. Methods: Comparison of recruitment strategies for the con- An Enabler-Inhibitor quadrant model was developed to clas- ducted trials is assessed. For the first Phase I study (048), sify each sub-theme as an abstract inhibitor, abstract enabler, media played an important role in participant recruitment, concrete inhibitor or concrete enabler. These were contextu- while presentations by recruitment staff played a more alized as occurring either at the level of the individual, fami- important role for the second trial. The community education ly, community or society. team made presentations to various workplaces, fliers and Results: Abstract inhibitors: fear of illness or death, lack of posters were sent out and national commemorations days information about HIV/AIDS and HIV vaccines, and an HIV were used to educate the general public on HIV vaccine tri- vaccine trial’s association with HIV/AIDS. als. Volunteers were screened during one or more in-person Abstract enablers: participant’s reported sense of altruism study visits. Written informed consent was obtained prior to and quality of life issues, e.g., protection from becoming the conduct of any study procedures. Demographic and infected with HIV. Concrete inhibitors: monetary costs asso- behavioral eligibility was determined based on participant ciated with participation, fear of receiving test results indi- responses to a standard interview instrument. Eligibility relat- cating sero-positivity, fear of negative reactions from family ed to HIV serostatus was ascertained in accordance with the and community members, time delays/lapse between receiv- national standard in Botswana and the protocols algorithm. ing trial participation information and actual enrollment, and Results: For the HVTN 059 study the total number of con- a general mistrust of researchers. Concrete enablers; practi- tacts who showed interest after presentations and other calities and convenience, financial rewards, a safe testing sources of information is 507. Of these, 489 filled the con- environment, positive family reactions to trial participation, tacts forms with phone numbers. 18 were just walk-ins who the different levels of participation available to different did not fill the contact forms but accessed information from members of the community, the salience/vividness of HIV in other sources. Of the 507, 58 were screened of which 16 were communities, positive community reactions to vaccine trials recruited into the study (Age range 22–46). The number and the presence of role models. screened for the HVTN 048 study was 80 of which 14 vol- Discussion: Findings are presented in an explanatory quad- unteers were recruited (age range 21–39). Motivation for vol- rant model with recommendations for trial investigators at unteers to participate in the trials was mostly altruistic. the level of the individual, family, and community. Individual Conclusions: Recruitment of volunteers willing to participate level: helping people to respond to discrimination in a proso- in HIV vaccine trials using various outreach strategies is fea- cial manner, developing strategies to overcome resistance to sible in Botswana. Recruitment strategies employed in the testing, paying trial participants an honorarium, and visiting first Phase I trial indicated the need to establish mechanisms participants in their homes or place of work. Family level: to track participants from the first point of contact with the supporting families to respond to trial-related discrimination study team when possible. and routine difficulties of trial participation. Community level: highlighting the salience of the HIV pandemic to 216 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 217 Amsterdam, Netherlands 29 August–1 September 2006 encourage community support, enrolling prominent commu- P13C-04 nity members in a future trial as role-models, and allaying suspicion about the misuse of science with interactive and Web-based training to support HIV vaccine open community forums. development J Fuchs1,2,3, J Darden4, D Flood3, J-P Kraehenbuhl5, P13C-03 J Lam1, N Debard5 and P Py5,6 Estimating population HIV incidence in subtype C 1 San Francisco Department of Public Health, San Francisco, CA. infected populations for Phase IIB/III HIV vaccine USA; 2 Department of Medicine, University of California, San trials accurately and cheaply with the avidity Francisco, San Francisco, CA. USA; 3The HIV Vaccine Trials index (AI) Network, Fred Hutchinson Cancer Research Center, Seattle, WA, USA; 4 Henry M. Jackson Foundation; Office for Policy in Clinical E Vardas, A Mohlala and J D’Araujo Research Operations (OPCRO), NH, Bethesda, MD, USA; 5 EuroVacc Foundation, ISREC, and Division of Immunology and Perinatal HIV Research Unit, University of the Witwatersrand Allergy, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland; 6 Department of Medicine, University of Background: Differentiating individuals with early HIV Zurich, Zurich, Switzerland infection from those infected for longer periods is important for estimating HIV incidence in populations. Cohort studies The promise of the internet as a medium to train clinical and are expensive and time consuming so laboratory techniques, laboratory staff responsible for conducting trials of HIV vac- based on modified HIV antibody assays on a single sample, cines lies in its ability to illustrate complex concepts in novel have been developed to detect recently infected individuals. ways for diverse and often geographically dispersed audi- The Avidity Index (AI) is a cheap, easy to perform assay ences. OCTAVE (Online Collaborative Training for AIDS which uses a commercial, automated third generation ELISA Vaccine Evaluation [http://octave.bio-med.ch]) aims to deliv- test (AxSYM HIV1/2gO assay, Abbott) to estimate incidence. er high quality, on-line training to investigative teams in over Methods: Validation of the AI assay on subtype C infected 30 countries preparing for or actively conducting clinical tri- seroconverters was done. Serum samples from 23 serocon- als of vaccine candidates. verters, mean days to seroconversion of 100.7 days were test- OCTAVE was created to complement face-to-face training ed. Mean AI score was 0.609 indicating recent infection. The offered by trial sponsors. By involving multiple groups simi- collection of “dried blood spot” (DBS) specimens as whole larly charged with overseeing trial site preparedness, a com- blood on filter paper is an ideal blood specimen collection mon e-Learning curriculum is being developed that will system for developing countries. DBS specimens are not sus- minimize duplication of effort and can be used to prepare ceptible to temperature variations during transport and inad- trainees prior to on-site training visits. Like Immunology equate storage systems that adversely impact on specimen Online (http://iol.ch), a web-based curriculum in basic and quality. In 50 samples tested in parallel (serum versus 1 DBS clinical immunology for medical trainees as well as under- of 50µl), none of the samples were misclassified as old infec- graduate and graduate level biology students, OCTAVE uses tion i.e. AI>1. The mean AI variability was 0.193 suggesting a Learning Management System developed by EuroVacc to that the DBS produced more dilute specimens. Therefore the deliver customized, interactive training. For example, proto- size of the DBS sample (1.5 DBS), the duration of elution and col-based learning uses an enhanced vaccine protocol to volume of elute in assay were increased, decreasing the vari- review concepts in clinical trial methodology, HIV pathogen- ability to 0.150. esis, laboratory science, good clinical practices (GCP), and Discussion: The AI has a number of advantages when com- vaccinology. This curriculum, reviewed in advance of case- pared to other ELISA based methods used to estimate inci- based workshops, prepares learners to synthesize pre-clinical dence, particularly for resource poor countries. The use of safety and immunogenicity data to build a rationale for mov- DBS samples gives added value to the test in resource con- ing a hypothetical candidate HIV vaccine into clinical trials. strained countries. The Avidity Index can provide incidence Applied GCP training requires learners to assume the role of estimates that can be used to model projections and calculate an investigator, sponsor, IRB member, CAB member, or par- sample sizes for Phase IIb/III efficacy trials. ticipant to clarify research responsibilities outlined in regula- tory guidance from the International Conference on Harmonization (ICH) and U.S. Code of Federal Regulations. Case-based learning is used to explore adverse event evalua- tion, and a GCLP-compliant “virtual laboratory” illustrates for laboratory and quality management staff the proper per- formance of safety and immunogenicity assays to support clinical trials. OCTAVE is committed to involving investigators from resource-poor nations at all levels of program development AIDS Vaccine 2006 217 Poster sessions_revGJ 14/8/06 16:34 Page 218 Amsterdam, Netherlands 29 August–1 September 2006 including content creation, peer review, and evaluation. This trials for HIV vaccines in a genuine and innovative part- approach will help to confirm the relevance and utility of the nership between European and developing countries’ scien- curriculum in local contexts. The OCTAVE pilot was initiat- tific communities in close communication with the product ed in the fourth quarter of 2004 and curricula in development developers. will be featured at the AIDS Vaccine meeting during hands- on sessions. The project is funded through a grant from the U.S. Office of P13C-06 AIDS Research, the Eranda Foundation, and the EuroVacc Foundation. Improving recruitment for HIV vaccine trials: lessons learned in Entebbe, Uganda P13C-05 E Mugisha1, L Nielsen1, F Nakwagala1, A Nanvubya1, C Konde1, J Birungi1, A von Lieven1, P Fast2, B Johnson2, EDCTP: A new partnership for HIV clinical trials C Schmidt2 and P Kaleebu1,3 in developing countries 1 UVRI/IAVI HIV Vaccine Program, Entebbe, Uganda; 2 M Karras, C Naus and O Leroy International AIDS Vaccine Initiative (IAVI), New York; 3 MRC/UVRI Uganda Research Unit on AIDS c/o Uganda Virus European and Developing Countries Clinical Trials Partnership, Research Institute, Entebbe, Uganda EDCTP, The Hague, The Netherlands Background: The International AIDS Vaccine Initiative Background: With the overall goal to reduce poverty in (IAVI) in collaboration with the Ugandan government developing countries by improving the health of the popu- through the Uganda Virus Research Institute (UVRI) con- lations, EDCTP aims through European research integra- ducts HIV vaccine trials and other epidemiological studies. tion and in partnership with African countries to develop This abstract presents recruitment strategies and their evolu- new clinical interventions to fight HIV/AIDS, malaria and tion across three studies: the second Ugandan Phase I HIV tuberculosis. vaccine trial (DNA/MVA) in 2003, a laboratory reference EDCTP is a partnership between 16 European countries on ranges (RR) study among healthy individuals in 2005, and a one hand and African countries on the other. It aims to join Phase II HIV vaccine trial (AAV) in 2006. relevant European national research programmes and their Objectives: To compare volunteer recruitment strategies in African partnerships to develop new clinical tools against the three successive vaccine studies at one site and how they are three diseases. modified and made efficient over time. Objectives: The programme objective is to accelerate the Methods: For the DNA/MVA trial, volunteer initial recruit- development of new or improved drugs and vaccines against ment strategies were through institutions (e.g. schools, hospi- AIDS, malaria and tuberculosis, with a focus on Phase II and tals, police barracks, etc.). Due to poor uptake, recruitment III clinical trials and on sub-Saharan Africa. strategies were revised to target the general population Methods: EDCTP operates in three main activity areas: 1. through a series of community seminars. Untrained mobilis- Strengthening the African capacity in this field through (a) ers, unaware of trial details, referred community members to capacity building grants; (b) training fellowships; 2. trial information seminars in the community. Potential vol- Supporting relevant clinical trials in the DCs; 3. Networking unteers were then invited to the Vaccine Trials Unit (VTU) for and coordination of European and the activities carried out further information, screening, consenting and enrollment. in the developing countries. The RR study also recruited from the general community, but Results: In 2005 EDCTP launched a call for capacity build- it employed mobilisers who were fully trained in the rationale ing for the conduct of Phase I/II and Phase III trials of vagi- and procedures of the study. Interested individuals were also nal microbicides against sexual transmission of HIV. In 2006, referred to the VTU. As part of the RR study’s informed con- EDCTP is launching calls on (1) Prevention of mother to sent procedure, volunteers were asked if they would mind child transmission (MTCT) of HIV, (2) Microbicide trials and being invited to participate in future studies. The AAV trial (3) Capacity building in preparation for the conduct of Phase enrolled from participants in the RR study who agreed to be II or III of preventive HIV vaccine trials. approached for future studies. Conclusions: Clinical research and clinical trials for Results: During the DNA/MVA trial, 4400 were approached HIV/AIDS, not sufficiently addressed so far, has now been for recruitment, 550 (12.5%) visited the VTU, 227 (5.2%) identified as a main priority. Previously, new tools for the pre- were screened and 50 (1.1%) enrolled. In the RR study, 1219 vention and treatment of the disease risked remaining stuck were approached, 233 (18.3%) were screened at the VTU in the development pipeline. In addition, many EU member and 200 (16.4%) enrolled. The AAV trial approached 34 states and their partners in the developing countries have sub- consenting participants from the RR study, screened 33 stantial collaborative research activities in this field. (97.1%) and enrolled 20 (58.8%). Unfortunately, these programmes are under funded and lack Conclusions: Screen-to-enrol rates for HIV vaccine trials capacity in the field. EDCTP wishes to contribute in the coor- dropped from 4.5:1 to 1.7:1 by using fully trained moblizers dination and intensification of the efforts to develop clinical 218 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 219 Amsterdam, Netherlands 29 August–1 September 2006 and drawing from volunteers already participating in P13C-08 research studies. Participants from earlier studies or individ- uals who wanted to participate in earlier studies can be a Comparison of three methods to detect recent good source for future study recruitment provided they can HIV-1 infection in specimens collected cross- meet inclusion criteria. sectionally in a cohort of female sex workers in the Dominican Republic P13C-07 S Gupta1, E Koenig2, G Murphy3, C Adon2, C Beyrer4, C Celentano4, S Khawaja1, J Parry3 and W Straus1 P24 antigen screening for early detection of HIV infection in discordant heterosexual couples in 1 Merck Research Laboratories, West Point, PA, USA; 2 Instituto Zambia Dominicano De Estudios Virologicos, Santo Domingo, Dominican Republic; 3 Health Protection Agency Centre for Infections, J Mulenga1, E Hunter2, O Manigart,2,3 G Stevens4, London, England; 4 Johns Hopkins Bloomberg School of Public S Allen2 and the Rwanda/Zambia HIV Research Group Health, Baltimore, MD, USA 1 Zambia Blood Transfusion Service, Lusaka, Zambia; 2 Emory Background: Interest in estimating HIV-1 incidence using spec- University, Atlanta, USA; 3 Zambia-Emory HIV Research Project, imens obtained as part of cross-sectional surveys has led to the Lusaka, Zambia; 4 IAVI, Johannesburg, SA development of new methods to detect recent HIV-1 infection through the testing of a single anti-HIV positive specimen. Objective: Early detection of HIV infection is a critical end- These assays are based on quantitative and qualitative differ- point in HIV prevention clinical trials and important in stud- ences in anti-HIV-1 antibodies between recent and long-stand- ies of HIV immunology and pathogenesis. Strategies to ing infections. Although several different recent infection assays identify recent HIV infections include shorter, more frequent have been developed, to date there have been few studies that testing intervals and antigen detection methods. have compared their performance. Methods: Zambian couples discordant for HIV infection Methods: An ongoing vaccine preparedness study enrolled were enrolled though a Couples’ Voluntary Counseling and female sex workers in the Dominican Republic. Specimens from Testing (CVCT) center and followed at 3-month intervals women found to be HIV positive at the baseline visit were test- from July 2002 to December 2005. The HIV negative partner ed for recent HIV-1 infection using three methods in a central is serologically tested with rapid HIV tests (Abbott laboratory: 1) detuned assay 2) avidity index; and 3) BED-EIA Determine® and Trinity Biotech Capillus®), and plasma is set assay. An unweighted kappa statistic in pair-wise comparisons aside for weekly p24 antigen (p24Ag) screening (Beckman- was used to estimate the correlation of recent HIV-1 infection Coulter). HIV antibody-positive volunteers are counseled, detection by the three methods. and both partners are asked to provide additional samples of Results: Nineteen out of 482(3.9%) women were positive for blood and genital fluids. HIV-antibody negative individuals HIV-1 infection. The incidence of HIV infection was 1.8% who are p24Ag-positive are asked to return for repeat HIV- (95% confidence interval [CI]: 0.3, 6.0), 0.9% (95% CI: 0.1, antibody testing and sample collection. Due to an increased 4.4), 1.0% (95% CI: 0.1, 4.5) using detuned assay, avidity level of background the p24Ag positivity is defined as 5× the index and BED-EIA techniques, respectively. The overall agree- calculated cutoff of the test. ment between both detuned assay and avidity index, and Results: The seroconversion rate in counseled HIV discordant detuned assay and BED-EIA was 94% (κ=0.8, 95% CI; 0.3, Zambian couples is 7–8/100 person years. Of 148 serocon- 1.0). The correlation was highest between BED-EIA and avidi- vertors identified in 42 months of the study, 32 (22%) were ty index methods (100%; κ=1.0). An abbreviated table com- p24Ag+: Nine were HIV antibody-positive with two rapid paring recent infection assay characteristics is shown below tests at the time p24Ag was detected, and 23 were HIV anti- (Table 1). body-negative. All 23 were HIV antibody-positive when they Conclusions: All three methods performed similarly in detecting returned for repeat testing and sample collection (median 48 recent HIV-1 infection in this region dominated by clade B HIV- days), even though in some cases this was only 7–10 days 1 infection; however, incidence estimates were slightly higher later. Only 3 of the 23 were still p24Ag+ at re-draw (6–19 using the detuned assay method. These results provide useful days later). information for use in field studies and vaccine trials that Conclusions: In a cohort with a seroconversion rate of require accurate and prompt assessment of the incidence of new approximately 2% per 3-month interval, almost one quarter HIV-1 infections. of new infections can be identified at the p24Ag positive stage. The main obstacle to detection of recent infection is the short window of antigen positivity. Because increasing the frequency of study visits in this cohort may result in decreased retention, more frequent batching of p24Ag testing and same-day invitations for retesting are being employed to identify study participants in the earliest stages of infection. AIDS Vaccine 2006 219 Poster sessions_revGJ 14/8/06 16:34 Page 220 Amsterdam, Netherlands 29 August–1 September 2006 P13C-08: Table 1 Factor Detuned Avidity BED-EIA Type of antibody measured Anti-HIV quantity Antibody avidity Anti-HIV quantity Commercially available Yes Yes Yes Assay generation 2nd generation 3rd generation 2nd generation Special equipment required No Yes No Automated Partial automation possible Yes Partial automation possible Window period variations by subtype Yes Unknown Yes Percent of AIDS cases misclassified 2.4% Unknown 4% as recent infections Assay specificity affected by Yes No Yes antiretroviral therapy P13C-09 past 3 months. In the rural cohort, 68 DC (including 45 HIV- negative men) were recruited, with no acute HIV infections Establishing a high risk HIV-negative cohort in captured to date. Cohort retention at both locations was Kilifi, Kenya approximately 80% after 9 months. Conclusions: We identified acute HIV infections in MSM, who engage in high risk behaviour and report low condom EJ Sanders1, S Graham2, E van der Elst1, M Mwangoma1, use. Targeted education and risk reduction counselling for 1 1 1 1 3 T Mumba , S Mutimba , J Kanungi , A Davies , M Price , MSM and other risk groups are crucial. Research aiming to 3 1 2 H Thomson , N Peshu and RS McClelland identify groups with high HIV incidence should include MSM, a group previously not considered for HIV-1 vaccine 1 Kenya Medical Research Institure (KEMRI)-Kilifi, Kilifi, Kenya; 2 trials in Africa. University of Washington, Seattle, WA, USA; 3 International AIDS Vaccine Initiative (IAVI), New York, NY, USA Background: Most new HIV infections in sub-Saharan Africa occur among sero-discordant couples (DC) and in core-trans- mitters such as persons who engage in sex work. Identifying groups with highest HIV incidence is essential in preparation for an HIV vaccine efficacy trial. Aim: To capture acute HIV infections in two distinct cohorts: one recruiting DC, one recruiting sex workers. Methods: Since July 2005, we have recruited sex workers into an urban cohort at a sexually transmitted disease (STD) clinic, and DC into a rural cohort at a district hospital. Enrolled HIV-seronegative individuals receive 3-monthly HIV counselling and testing, including HIV p24 antigen test- ing, as well as STD screening. We identified female sex work- ers (FSW), male sex workers (MSW) who have paid sex with women, and men who have sex with men (MSM), the major- ity of whom also engage in sex work. Ongoing recruitment is facilitated by trained peer leaders, a drop-in centre, leaflets, and extensive community mobilization. HIV prevalence was assessed for each risk group, and incidence will be monitored over the next 2 years. Results: In the urban cohort, 38% (23/60) of MSM, 30% (46/155) of FSW, and 9% (6/69) of MSW were HIV-positive at screening. Median age (IQR) at screening was 26 (23–28), 24 (21–28), and 24 (22–28) years, respectively. Two of 37 MSM were p24-antigen positive but antibody negative at screening and sero-converted within 2 and 5 weeks, respec- tively. A third MSM sero-converted 7 weeks after enrollment. Only 16% (3/19) of MSM reported 100% condom use in past month, and 19% (7/37) had participated in group sex in 220 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 221 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 13D: CLINICAL TRIAL SITE DEVELOPMENT – MEASURING EFFICACY TRIAL ENDPOINTS 13D-01 13D-02 HSV-2 infection in potential high risk group One or two efficacy trials for HIV vaccines and volunteers for HIV vaccine trials microbicides? Comparison of statistical power, sample size and costs B Bekan Homawoo1,2, M Coldiron1,2, E Tekirya1, J Bizimana1,2, E Karita1,2, S Allen2 and Rwanda Zambia PM Coplan HIV Research Group2 International Partnership for Microbicides, Plymouth, USA 1 Projet San Francisco, Kigali, Rwanda; 2 Emory University, Atlanta, Georgia Background: The FDA requires one pivotal trial or two trials that demonstrate efficacy for licensure of vaccines or micro- Background: HSV-2 infection has been associated with trans- bicides to prevent HIV infection, but if two trials are done, mission and acquisition of HIV and a more rapid progression each must show efficacy. Developers can conduct one pivotal of HIV disease. HSV-2 is highly prevalent in sub-Saharan or two smaller efficacy trials for Phase III programs. We Africa, though it is often asymptomatic. New HIV infections assess the impact of one versus two trials on statistical power, frequently occur in discordant couples who remain a high sample size and cost of Phase III programs. Statistical Power risk group and the HIV negative partners in those couples are = sensitivity to detect efficacy. Low power equals high prob- potential volunteers for Phase III HIV vaccine trials. ability of rejecting truly effective products by chance: 80% Objective: To establish HSV-2 prevalence amongst HIV dis- power represents 20% chance of rejecting effective products cordant couples and the possible impact of that prevalence on by chance alone. A key observation is that the power of Phase the measurement of efficacy trials’ endpoint. III programs with 2 trials is the product of each trial’s power, Method: At Projet San Francisco (PSF) in Kigali, we con- e.g., two trials with 80% power each has programmatic ducted a cross-sectional study to establish the prevalence of power of 0.64 (=0.8 ×0.8). HSV-2 in 729 discordant couples using HSV-2 Focus Methods: Sample size was calculated for overall Phase III ELISA. We considered an Index Value of >3.5 positive and programs to have 90% or 80% power, assuming 50% effica- presumptive for the presence of IgG antibodies to HSV-2, cy, 1:1 randomized placebo-controlled 1-year-long trials, 3% an Index Value of <0.90 negative, and an Index Value of HIV incidence, 15% drop-out using Stata Version 8. FDA ≥0.90 but ≤3.5 equivocal. Equivocal results were not indicated P<0.005 is acceptable for one trial; P<0.001 was included in this analysis. calculated for sensitivity analysis. Costs used were US $9,000 Results: The analysis of our primary data shows that HIV per participant. positive individuals have a higher prevalence of HSV-2 (79%) Results: Required sample size and costs for Phase III pro- compared to HIV negative (59%). In addition, HIV positive grams to have 90% or 80% power are shown in Table 1. females are more likely to be HSV-2 co-infected (84%) com- Conclusions: To achieve 90% power for Phase III programs pared to HIV positive males (73%). Moreover, the HIV neg- of vaccines or Microbicides to prevent HIV infection, one ative partners of HIV/HSV-2 co-infected individuals have pivotal trial at P<0.005 provides savings of 57% in sample higher risk for HSV-2 infection compared to partners of HIV- size and $41 million versus two trials. With P<0.001, one infected individuals not HSV-2 infected (OR 4.4, 95% CI versus two trials provides savings of 27% in size and $24 mil- 1.8–10). We also looked at CD4 count as a measure of HIV lion. One efficacy trial provides significant savings compared disease progression. The mean CD4 count was lower among to two efficacy trials if statistical power for Phase III pro- HIV/HSV-2 co-infected individuals (480 cells/µl) compared grams is held constant. to HIV positive HSV-2 negative individuals (505 cells/µl), but this difference was not statistically significant. Conclusion: HSV-2 is highly prevalent in both partners of HIV discordant couples. HIV vaccine trials conducted in high risk groups should take into account the potential confound- ing effect of HSV-2 infection in both the donors and the recipients given the association between HIV AND HSV-2 infections. HSV-2 testing should be advocated as standard of care for those couples in loom of an HIV vaccine. AIDS Vaccine 2006 221 Poster sessions_revGJ 14/8/06 16:34 Page 222 Amsterdam, Netherlands 29 August–1 September 2006 13D-02: Table 1. One Trial (P<0.005) Two Trials (P<0.05 each) Power per trial Sample size Cost (million) Power per trial Sample size Cost (million) Power of Program for 2 trials 90% 90% 7,995 $72 95% 12,565 $113 80% 80% 6,431 $58 89% 9,953 $90 13D-03 13D-04 A small dose of HIV? HIV vaccine mental models HIV vaccine concerns, misconceptions and mis- and heuristics among United States communities trust among vulnerable communities in the United at risk (Project VIBE) States (Project VIBE) PA Newman1, DS Seiden2, KJ Roberts2 and N Duan2 PA Newman1, WE Cunningham2, S-J Lee2, KJ Roberts2, DS Seiden2 and N Duan2 1 University of Toronto, Toronto, ON, Canada; 2 University of California, Los Angeles, Los Angeles, CA, USA 1 University of Toronto, Toronto, ON, Canada; 2 University of California, Los Angeles, Los Angeles, CA, USA Background: Formidable challenges for future HIV vaccine dis- semination in the U.S. are suggested by suboptimal uptake of Background: HIV vaccine availability will not ensure effec- existing vaccines, disparities in vaccine coverage and HIV treat- tive dissemination, particularly among vulnerable communi- ment, AIDS stigma, and the likelihood that initial HIV vaccines ties at greatest risk for HIV/AIDS. may be only partially efficacious. Empirically-based risk com- Objectives: To identify and assess beliefs and concerns about munication strategies may support the effectiveness of first-gen- future HIV vaccines among low socioeconomic, ethnically eration HIV vaccines diverse adults at increased risk for HIV infection in Los Objectives: To explore HIV vaccine mental models and heuris- Angeles, California, US. tics among communities at increased risk for HIV infection. Methods: A mixed-method investigation included 15 focus Methods: Nine focus groups (6 in English; 3 in Spanish) were groups with adults (n=157; median age=34; 44% women; conducted with ethnically diverse, low socioeconomic adults 27% African American, 48% Latino, 20% white; 52% with (n=99; median age=33 years; 48% female; 22% African annual income 222 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 223 Amsterdam, Netherlands 29 August–1 September 2006 Conclusions: Widespread HIV vaccine concerns, miscon- ceptions and mistrust among vulnerable communities pre- sent formidable challenges to HIV vaccine dissemination, and may compromise the effectiveness of vaccines in con- trolling the AIDS pandemic. Culturally appropriate, empir- ically-based individual- and community-level educational and social marketing interventions and outreach may be vital to address HIV vaccine concerns and mistrust in order to facilitate successful dissemination. AIDS Vaccine 2006 223 Poster sessions_revGJ 14/8/06 16:34 Page 224 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 13E: CLINICAL TRIAL SITE DEVELOPMENT – VACCINE PREPAREDNESS P13E-01 P13E-02 Assessment of willingness to participate in HIV Progress towards the first Phase I/II HIV vaccine vaccine trials in a cohort of homosexual and het- trial in Tanzania, the HIV Vaccine Immunogenicity erosexual intravenous drugs users: our experience Study (HIVIS) Project in a resource-poor sub-Saharan African city M Bakari1, S Aboud1, M Janabi1, W Urassa1, F Mugusi1, OA Busari1, OO Busari2, G Oligbu1, AO Adeyemi3 and AA E Lyamuya1, J Mwami1, F Mhalu1, K Pallangyo1, Busari4 B Hejdeman2, B Wahren3, G Biberfeld3 and E Sandstrom2 1 Federal Medical Centre, Ido-Ekiti, Nigeria; 2 College of 1 Muhimbili University College of Health Sciences, Dar es Medicine, University of Ilorin, Ilorin, Nigeria; 3 Healthmatch Salaam, Tanzania: 2 Venhälsan, Department of Infectious International, Lagos, Nigeria; 4 Lifecare AIDS Foundation and Diseases, Karolinska University Hospital, Sweden: 3 Microbiology Lifecare Consultant Hospital, Ado-Ekiti, Nigeria and Tumorbiology Center, Karolinska Institute, Swedish Institute for Infectious Disease Control (SMI), Stockholm, Sweden Background: One of the critical factors for successful conduct of HIV vaccine trials is the willingness of potential participants Background: In Tanzania, the adult HIV-1 sero-prevalence is to participate and comply with trial rules and procedures. 7%, and the HIV annual incidence is 1–2%. Despite inter- Objective: This was to assess the willingness to participate ventions, the epidemic still rages especially among youths and (WTP) in HIV vaccine trials in a cohort of homosexual and more so among females. The country is thus suited for HIV heterosexual intravenous drugs users (IDUs) in Lagos, Nigeria. vaccine trials. Methods: A cohort of initially HIV negative 256 IDAs was Objective: To report on progress made with HIVIS trial. recruited and follow-up every 2 months for 1 year. They Methods: Preparations towards HIV vaccine trials begun in were recruited from slums, brothels, beaches and public 1989 within the bilateral TANSWED HIV Programme, fund- packs. After 1-year follow-up, questionnaires were adminis- ed by Sida/SAREC (Sweden), in preparation of a potential tered. Questionnaire contained closed and open-ended cohort of Police Officers (PO’s) and in capacity building in questions and was pretested. Administered by interviewers clinical, laboratory, social behavioral, epidemiology expertise who were selected and trained community members and and infrastructure. HIV counsellors. Analysis was: frequency distribution, Results: In 2001, a European Union funded HIVIS project, cross tabulation, test of significance with chi-square, level with a South-South (Tanzania & South Africa) and North- of significance P<0.05. North (Sweden, USA, Germany) collaboration was initiated Results: 159(62.1%) subjects out of the initial 256 recruited to develop and try out a DNA multi-clade candidate vaccine stayed to the end of 1-year follow-up. 131 (82.3%) of these with a poxvirus vaccine boost. A safety and tolerability Phase agreed to answer questionnaires. Mean age of respondents I trial with an HIV vaccine-containing multi-gene, multi-sub- was 22.6 ±6.4 years and males were 82.4% and females type DNA relevant for Africa (subtypes A, B, C) has been 17.6%. Fifty-three (40.5%) subjects expressed their WTP in completed in Stockholm, Sweden, with encouraging results. HIV vaccine trials while 78(59.5%) declined to participate. Currently, a further boosting among these volunteers with Comparing the gender proportions of subjects that agreed to MVA (CRF01 A_E) is on going. answer questionnaires to those that showed WTP The National HIV/AIDS Policy (2001) in Tanzania incorpo- (females:17.6% versus 33.9%; males: 82.4% versus 66.1%), rates research on HIV vaccines. The Tanzania National females were significantly (P=0.022) more willing to partici- Framework for the conduct of HIV vaccine trials is in place pate than males. since early 2005. Capacity building including training of staff The major reasons for nonparticipation were: “I would not in GCP and GLP has taken place. Advocacy has continued want to be used for experiments”(45/78(57.7%)) and “the with the Ministry of Home Affairs, police force authorities as vaccine might cause me to be positive for well as the PO’s (potential volunteers) themselves at their HIV”(69/78(88.5%))”.For those who would participate, the workstations. A Community Advisory Board has been estab- commonest reason adduced was: “I want to be part of glob- lished. The response of potential volunteers has been largely al search for effective HIV vaccine”(49/53(92.5%)). encouraging. The HIVIS study protocol having incorporated Conclusion: This study showed that majority (59.5%) of comments from the HIV Vaccine Advisory Committee of IDUs, who are potential trial participants, are not willing to WHO-UNAIDS and the African AIDS Vaccine Programme participate in HIV vaccine trials. This demonstrates a need to has received National Ethical Clearance. An application for step up public education/awareness on HIV/AIDS, HIV vac- vaccine registration with Tanzania’s Food and Drugs cines/HIV vaccine trials for this cohort particularly and Authority (TFDA) has been submitted. The Phase I/II trial Nigerians generally. employing the same antigens as used for the Phase I trial in 224 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 225 Amsterdam, Netherlands 29 August–1 September 2006 Sweden is expected to start in September 2006 in Dar es Conclusions: The background morbidity in sub-Saharan Salaam, Tanzania Africa contributes to the greatest proportion of screen fail- Conclusion: HIVIS project will be a significant contribution ures and exclusion from analysis once enrolled. Despite these for further HIV vaccine development efforts in Tanzania and factors there is still sufficient proportion to evaluate the study globally. There are challenges, but lessons have been learnt. objective. Equal proportions of men and women were analysed. P13E-03 P13E-04 Enrollment and screening failure in 2,424 African Ethical aspects of adolescent participation in HIV volunteers recruited for a multi-center study of vaccine trials in the Nyanga district, Cape Town, laboratory reference ranges in preparation for HIV South Africa vaccine clinical trials N Soka1, H Jaspan1, C Mathews2, AJ Flisher3, D Mark1, R K Kayitesi1, O Anzala2, W Jaoko2, P Kaleebu3, E Karita1, Wood1 and LG Bekker1 E Ruzagira4, J Mulenga5, G Mutua2, A Nanvubya3, M Price6, E Shutes1, C Kambili6, P Fast6 and A Kamali4 1 Desmond Tutu HIV Centre, IIDMM, University of Cape Town, Cape Town, South Africa; 2 Health Systems Research Unit, MRC 1 Projet San Francisco and the Rwanda Zambia HIV Research and Department of Public Health and Family Medicine, Group (RZHRG); 2 Kenya Aids Vaccine Initiative; 3 Uganda Virus University of Cape Town, Cape Town, South Africa; 3 Division of Research Institute; 4 Medical Research Council Uganda; Child and Adolescent Psychiatry and Adolescent Health Research 5 RZHRG; 6 International AIDS Vaccine Initiative Institute, University of Cape Town, Cape Town, South Africa Objectives: Current clinical safety laboratory values used to Background: The Cape Town AIDS Vaccine Clinical Trials screen populations for entry into HIV vaccine and prevention Consortium is preparing the Nyanga district as a megasite for clinical trials have been derived from Western populations. large-scale recruitment to HIV vaccine efficacy trials. We present preliminary results from a multi-center study con- Aims: Inclusion of adolescents in HIV vaccine trials is com- ducted to describe and establish reference ranges in healthy plex due to the vulnerability of this population. Thus, this sub-Saharan African volunteers. feasibility study is a preparatory step to vaccine trials in Methods: Healthy HIV uninfected volunteers, aged 18 to 60 Nyanga and seeks a better understanding of youth culture, years, were invited to participate in a cross sectional study in issues around parental consent, understanding of adolescent Lusaka, Zambia; Kigali, Rwanda; Masaka and Entebbe, and parental fears and stigmas. Uganda; and two sites in Nairobi, Kenya. Potential volunteers Methods: The study enrolled 100 participants, comprised of who were acutely ill, pregnant, menstruating, or had signifi- parents and adolescents. Participants were recruited through cant medical histories were not enrolled. After obtaining high schools, youth clubs, adolescent CAB, clinics, churches, informed consent, demographic data was collected, urinalysis and other community organizations. They participated in performed, and each volunteer provided blood for hematol- focus groups and were asked about their views regarding the ogy, immune function, and clinical chemistries. Enrolled vol- inclusion of adolescents in future HIV vaccine trials, parental unteers were tested for HIV, syphilis and hepatitis B & C; a consent and access to test results (STI, HIV, pregnancy) col- positive result excluded volunteers from analysis. lected during the study. Discussions were tape-recorded, tran- Results: Although enrollment continues in some sites, the scribed, translated into English and prepared for content study screened 2424 volunteers (1203 [49.6%] women and analysis. Transcripts were analysed using grounded theory. 1221 (50.4%) men) between December 2004 and March Results: Parents expressed a great amount of concern about 2006. Of these, 1975 (81.5%) were enrolled (954 [48.3%] HIV and youth risk behaviour, and thus welcomed their women and 1024 (51.8%) men) and 176 were excluded from inclusion in future vaccine trials. Adolescents were concerned the analysis after enrollment leaving an analyzable cohort of about the compulsory HIV testing, disclosure and potential 1799 (74.2%, 873 [48.5%] women and 929 [51.7%] men). stigma associated with possible positive results. They felt that The proportion of women who failed screening or were parents should give consent, but should not have access to excluded from analysis after enrollment was similar to the study test results. Whilst some parents strongly felt that they proportion of men (P=0.9). The most common reasons for would want to know about their children’s results to be able screen failures were: abnormal findings on physical examina- to offer support when necessary, others understood their chil- tion (172, 38.6%), medical history (81, 18.2%) and acutely dren’s right to privacy and confidentiality. ill at screening (58, 13.0%). Once enrolled, volunteers were Conclusions: Despite the concerns raised above, adolescents excluded from analysis due to clinical and laboratory reasons and parents are interested in participating in future HIV vac- including physical examination findings (57, 32.4%), HIV cine trials, which are viewed as an opportunity to help iden- antibodies (3, 1.7%), hepatitis B antigen (51, 30.0%), hepati- tify an effective HIV vaccine to protect South Africa’s youth. tis C antibodies (42, 23.9%), and syphilis (8, 4.5%). A vol- However, there is divergence regarding disclosure and unteer may be screened out or excluded from analysis for parental access to sensitive information. Although many multiple reasons. adolescents claim to have an open relationship with their AIDS Vaccine 2006 225 Poster sessions_revGJ 14/8/06 16:34 Page 226 Amsterdam, Netherlands 29 August–1 September 2006 parents, they may not want their test results to be disclosed HAART, but MSM orientation by itself shows no association to their parents, primarily due to lack of perceived support to seropositivity in multivariable analysis. Further studies of and fear of being rejected by family members should their local epidemic characteristics may provide necessary baseline results be positive. Clinical trial staff will need to support information and infra structure strengthening as preparation both disclosure by adolescents when requested and parents for future Phase III trials. when disclosure is not preferred. P13E-06 P13E-05 Seasonal variation in clinical laboratory parame- Predictors of seropositivity among men at a ters among HIV-uninfected adults in Kigali, Voluntary HIV Counseling and Testing site (VCT) Rwanda in Curitiba, South Brazil E Karita1, K Kayitenkore1, C Muvunyi1, E Shutes1, R Rodrigues3, M Thomaz1, MR Almeida1, A Nascimento2, C Kambili2, J Bizimana1, P Fast2, S Allen3, M Price2 and D Harrad1, ACA Evangelista1, M Chomatas1, L Ferreira1, A Kamali4 IO Silva2 and LFM Brigido3 1 PSF/Rwanda-Zambia HIV Research Group, Kigali, Rwanda; 2 1 Municipal Aids Program, Curitiba, Brazil; 2 USP, Preventive International AIDS Vaccine Initiative, New York, USA; 3 Emory Medicine, Sao Paulo, Brazil; 3 Adolfo Lutz Institute, Sao Paulo, University, Atlanta, USA; 4 MRC/UVRI Uganda Research Unit on Brazil AIDS, Masaka, Uganda Background: HIV epidemic in Brazil shows evidence for a rel- Background: Clinical laboratory reference ranges have not ative stabilization but information on the determinants of yet been established in many African countries. Studies have infection are very limited. As part of a feasibility study to shown differences in laboratory values between African and identify potential vaccine sites (www.rcp-hiv.org), we evalu- US or European populations, possibly due to endemic dis- ated risk and socio-behavioral aspects of male users of the eases, genetic, nutritional, environmental, and/or socioeco- Central VCT of Curitiba, a major city at South Brazil. nomic factors. There is seasonal variation of several diseases, Methods: A random sampling of VCT users was interviewed such as malaria that can affect laboratory values, however (February 2005 to January 2006) by a member of the study the impact of the seasonal changes on laboratory parameters team with a standard questionnaire in a private setting before has not yet been assessed in Sub-Saharan Africa. serological testing. Out of 799 users interviewed, 505 were Objective: To assess the impact of the seasonal changes on men and were analysed here. Descriptive, univariate and mul- laboratory parameters among HIV-uninfected adults. tivariate (logistic regression) were performed using the soft- Methods: 400 clinically healthy, HIV-uninfected adults (200 ware Stata9™. women and 200 men), aged 18 to 53 years were enrolled in Results: Mean age was 32 (SD=10.9) years, with 84.7% Kigali, Rwanda between December 2004 and May 2005 reporting at least 8 years of education and 68.8% a family (rainy season). After providing informed consent monthly income from US$ 150.00 to US$ 900.00. Exclusive (written/thumb print) all volunteers underwent a comprehen- heterosexual behavior was reported by 70.4%, exclusive sive medical history, a complete physical examination and homosexual by 20.7% and bisexual orientation by 8.9%. evaluation of laboratory values (chemistry and hematology). 484 subjects have had sexual relations for more than 9 years. They were then invited to come back to the clinic during the 204 (40.7%) reported active anal sex and 117 (23.3%) pas- dry season (June–September 2005). sive anal sex. Consistent use of condoms was referred by Results: The mean age was 28.5 years for females (range: 28% with fix partners and 49% with occasional partner. Use 18–50), and 32.4 years for males (range : 20-53). 37 volun- of alcohol was reported by 26% and use of additional drugs teers (19 females and 18 males) were excluded from the first (not IDU), with or without alcohol, by 52%. 27% had a analysis (rainy season), and an additional 29 (19 females and MSM partner, 10% had a HIV+ partner and 16.0% a sex 10 males) excluded from the second (dry season). The most worker partner. HIV seroprevalence was 5.8% (95%CI: frequent reasons for exclusion were hepatitis B/C seropositiv- 3.9–8.3%). Factors associated with HIV infection in multi- ity (28 cases) and pregnancy (18 cases). There were no dif- variate analysis were: passive anal sex (OR=7.4; 95%CI: ferences across seasons for creatinine and SGOT/ALT values. 2.6–21.2; P<0.001), initiating sexual activity between 10 and SGPT/AST, hemoglobin, and CD4 counts were higher in the 19 years ago (OR=3.0; 95%CI: 1.0–8.9; P=0.04) and use of rainy season, and these differences were statistically signifi- other drugs than alcohol exclusively (OR=3.3; 95%CI: cant. However, the absolute differences were modest, and 0.9–11.8; P=0.07). All subjects declare willingness to partici- their clinical significance was negligible. The median values pate in vaccine trials. for SGPT were 20 IU/l in the rainy season and 18 IU/l in the Conclusions: The HIV epidemic among male users of this dry season (P<0.0001). Corresponding values for CD4 VCT in Curitiba shows association to known risk factors and counts were respectively 932 and 877 cells/µl (P<0.0001). to an initiation of sexual activity before availability of Conclusion: There were no observed clinically significant 226 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 227 Amsterdam, Netherlands 29 August–1 September 2006 variations across seasons in laboratory parameters at this vaccine and reduction of social stigmas attached to the epi- site, suggesting that enrollment into clinical trials such as demic even within the medical fraternity. HIV vaccine trials may not need to consider season. Seasonal analyses are underway at three additional sites in Zambia, Uganda, and Kenya and may further inform these prelimi- nary results. P13E-08 Willingness to participate in HIV vaccine trials P13E-07 among trainee artisans in Ibadan, Nigeria O Onigbogi1, A Fajola1 and C Famoyin2 Knowledge, attitude and perception of health care providers on the ongoing Phase I vaccine trials in 1 University College Hospital, Ibadan, Nigeria, 2 UMass Memorial the country in Jaipur, the capital city of the state Hospital, Worchester, USA of Rajasthan Background: Large-scale field trials to determine the efficacy M Singh of HIV vaccines have been in place for a number of years in some parts of the world. Plans are in the pipeline to com- Gender Specialist and Faculty, Women’s Resource Center, Harish mence such activities in Nigeria in the near future. Chandra Mathur Rajasthan State Institute of Public Aim: This study sought to determine the willingness of Administration, Jaipur, Rajasthan, India respondents to be subjects of these trials. Methods: Four hundred and eight (418) trainee artisans aged Background: Though India remains a low prevalence country 15-30 years old were surveyed in Ibadan, Nigeria. There were with overall HIV prevalence of 0.91% i.e. less than 1% of the 315 males and 103 females comprising auto-mechanics, auto- population, it masks various sub epidemics in several geo- electricians, battery chargers, barbers, tailors and hair- graphical foci in India. Prevention is thus the most critical dressers. Data was analysed to determine the correlates of strategy for combating with HIV/AIDS and an HIV vaccine willingness to participate (WTP) in AIDS vaccine trials would be the most effective prevention tool. For the first time among them. in the history of the country, Indian Council of Medical Results: On the overall, 52.3% of the respondents will be Research, and National AIDS Control Organization, under willing to participate. Increased WTP was associated with the Ministry of Heath, Government of India signed a memo- smoking (OR=1.32, 95% CI: 1.14–1.85), illicit drug use randum with International AIDS Vaccine Initiative to con- (OR=1.46, 95% CI: 1.12–1.95), alcohol use (OR=1.24, 25% duct a vaccine trial. The Phase I vaccine trials were launched CI: 1.09–1.54), higher level of HIV/AIDS awareness (OR= in 2005 by National AIDS Research Institute. A study was 1.41, 95% CI: 1.21-1.87), prior sexual experience (OR=1.54, conducted on a cross section of health care providers in 95% CI: 1.04–1.72) and involvement in high-risk sexual Jaipur to assess the knowledge, attitude and perception of behaviour (OR=1.34, 95% CI: 1.12–1.67). Decreased WTP health care providers to the ongoing HIV vaccine clinical tri- was associated with perceptions of HIV-related stigma als in the country. (OR=0.54, 95% CI: 0.32–0.64), superstitious beliefs about Objectives: To assess the knowledge, attitude and perception HIV (OR=0.64, 95% CI: 0.21–0.76), increased number of of health care providers on the ongoing Phase I vaccine trials doses of hypothetical vaccine (OR=0.41, 95% CI: 0.53–0.72) in the country in Jaipur and use of parenteral route for vaccine administration Methods: Two hundred health care providers including a (OR=0.61, 95% CI: 0.42–0.81). cross section of doctors, nurses and paramedics were admin- Conclusion: The high level of WTP among the artisans sug- istered a semi structured questionnaire between January to gests that HIV vaccines may be widely accepted if eventual- March 2006. The health care providers represented the mili- ly introduced among them. This is important because many tary hospital; the state government run public health hospi- trainee artisans fall into the most vulnerable age bracket for tals and a state level teaching hospital. HIV/AIDS. The general level of awareness about HIV/AIDS Results: The target group was assessed on socio demograph- and sexually transmitted diseases among this group of peo- ic data along with knowledge of vaccine development and the ple should be increased and efforts made to decrease HIV- current trials in the country. This cohort was 60% male, 40% related stigma, dispel rumors and correct misconceptions female with median age 42.6 years. Of the 200 participants about the origin the disease. These will assist in increasing in the study only 48.5% of the health care providers were the level of WTP. Larger studies involving focus group dis- aware of the currently ongoing vaccine clinical trials in India. cussions and in-depth interviews will be required to further Most respondents (86%) were not sure of the safety of the investigate perceived merits and demerits. vaccine and were apprehensive of volunteers becoming HIV positive by participating in the trials. Conclusions: It is pertinent that efforts on inculcating aware- ness regarding the HIV vaccine trials be strengthened special- ly among health care providers as complete and correct information on the subject is critical to the acceptance of the AIDS Vaccine 2006 227 Poster sessions_revGJ 14/8/06 16:34 Page 228 Amsterdam, Netherlands 29 August–1 September 2006 P13E-09 P13E-10 Towards a new GCP: “Good Community Practice” Increasing community involvement: model training in prevention research programs MJ Warren E Lee1, J Katzman2 and P Wilson3 AIDS Vaccine Advocacy Coalition, New York, NY, USA 1 AIDS Vaccine Advocacy Coalition (AVAC), New York, NY, USA; 2 Vaccine & Prevention Research Program, Washington DC, USA; Objectives/aim: Since 2004, controversy over pre-exposure 3 DAIDS/NIAID/NIH, Black AIDS Institute, Los Angles, CA, USA prophylaxis (PrEP) research has highlighted the challenge of communication between various stakeholders, and the need Background: HIV vaccine advocates have long stated the for mutual understanding of each other’s values. Left importance of community involvement in research efforts. unchecked, these differences undermine the progress of HIV However, community involvement can be defined broadly prevention trials and ultimately affect the ultimate beneficia- and may differ in meaning to various audiences around the ries of such research. world. Current resources primarily focus on educating and Methods: We participated in several global consultations mobilizing communities for enrollment into clinical trials – a over the past 18 months that sought to address the issues very narrow and specific aspect of community involvement. related to PrEP research and the implications for other pre- In order for communities to support long term HIV vaccine vention research, including HIV vaccine and microbicide research efforts, they must understand the nature of research development. Each consultation developed recommendations and its connection to their experiences with HIV beyond for new models of prevention research. becoming trial volunteers. To accomplish this, our under- Results: While stakeholders at each meeting recognized the standing and practice of community education and involve- value and importance of community involvement, there were ment must expand beyond trial-based efforts. different perspectives on how and when to engage communi- Objectives/Aims: Two models of HIV vaccine community ty. There was an expressed interest in working towards con- education with goals that are not linked to clinical trial sensus on best practices in community engagement that could enrollment have been developed and implemented over the be assessed by various stakeholders. While Community last three years. The Community Education and Outreach Advisory Boards (CABs) are an important and necessary Partnership Program (CEOPP) and the African American component of addressing some issues, they are not sufficient HIV University (AAHU) each aim to increase the capacity of to ensure meaningful community support and engagement. key constituents to address HIV vaccine research in their own Very often, individuals and CABs have limited understanding community settings without engaging directly in trial recruit- of clinical research processes and basic research concepts. In ment activities. addition, while trial investigators focus on the importance of Methods: Collaborating with the National Institute of the study at hand, community members may bring a complex Allergy and Infectious Diseases of NIH, Ogilvy Public set of different experiences, concerns and issues to the Relations and the Black AIDS Institute, the AIDS Vaccine research effort. These issues are not unique to any one trial or Advocacy Coalition has helped to develop training curricula one investigational prevention tool; they transcend and influ- to build a basic foundation of knowledge about HIV vaccine ence how prevention research happens everywhere. They also research and identify strategies to address community ques- will influence how trial results are disseminated and under- tions and concerns. These training curricula integrate vaccine stood and how new prevention technologies will be intro- advocacy into an overall HIV/AIDS advocacy agenda. duced and accepted. Results: These two programs have thus far trained more than Conclusion: There is an urgent need to develop, test, validate three dozen individuals and organizations, not directly asso- and implement guidelines for “Good Community Practice”, ciated with specific clinical trials, that are aware of and able modelled on the Good Clinical, Laboratory and to address community questions and concerns about HIV Manufacturing Processes (GCP, GLP, GCLP and GMP) that vaccine research. These experiences provide key insights into are the standard guides for research. Although community the development of community education efforts aimed at engagement activities will differ in each community, develop- long term support of and advocacy for vaccine and other ment of guidelines and minimum standards for engagement research into new prevention technologies. efforts, as well as approaches to measuring and monitoring Conclusions/Discussion: While community education is a engagement, would be invaluable in facilitating timely, com- critical aspect of clinical trial recruitment, the potential of prehensive and valid research. community education and involvement strategies is much more far reaching. Expansion of the types of community education activities resourced by the HIV vaccine field movement will be instrumental to ensuring sustained support of research efforts. 228 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 229 Amsterdam, Netherlands 29 August–1 September 2006 P13E-11 P13E-12 Research advocacy: maximizing preparedness for Successful enrollment of adolescents in pre-HIV HIV research vaccine clinical trials E Lee1, M Chatani2, M Ukpong3 and M Warren3 L Peralta1, R Gorle1, L Flores1, E Colocho1, C Boyer2 and B Griffin-Deeds1 1 AIDS Vaccine Advocacy Coalition (AVAC), New York, NY, USA; 2 African Microbicide Advocacy Group (AMAG), Accra, 1 University of Maryland School of Medicine, Baltimore, Ghana; 3 Nigerian HIV Vaccine and Microbicide Advocacy Maryland, USA; 2 University of California, San Francisco, San Group (NHVMAG), Lagos, Nigeria Francisco, CA, USA Objectives/Aims: Advocacy has played an important role in Background: Globally, young people under 25 years of age HIV research, especially in the development of AIDS treat- carry a large burden of HIV infection. It is therefore criti- ments. However, prevention research advocacy has general- cal to engage and enroll adolescents and young adults in ly occurred independently for vaccines and microbicides, HIV vaccine clinical trials. This study was designed to eval- while other prevention research (like male circumcision, cer- uate successful methods of recruiting adolescents 12–17 vical barriers, and pre-exposure prophylaxis) has had much years of age in a Hepatitis B vaccine clinical trial as a more limited advocacy. Yet the common issues for vaccines, model to enroll this population successfully in future HIV microbicides and other new prevention technology research vaccine clinical trials. in both developing and developed countries present clear Methods: A multifaceted study was designed to recruit ado- opportunities for strategic collaboration. This approach is lescents in a Hepatitis B vaccine clinical trial. The study con- framed by an understanding that the best response to sisted of educational and screening phases to reach HIV/AIDS is a comprehensive and integrated response that adolescents and their parents/guardians through community includes primary prevention, testing, treatment and care, mobilization activities. The educational component consisted research and social justice. of a presentation on Hepatitis B risk and consequences with Methods: AVAC, AMAG, NHVMAG and other partners pre- and post- presentation surveys to assess participants’ sought to establish links between community-based advoca- knowledge of Hepatitis B and attitudes and behaviors cy efforts for vaccines, microbicides and PrEP. Strategies towards a vaccine trial. The screening phase involved blood include message development, development of training mod- screening to detect Hepatitis B antigen-antibody reaction to ules, and joint educational activities to highlight connections confirm the adolescents’ immunization status. All non-immu- between each area. We reviewed strategies to build connec- nized youth were then invited to enroll in an ongoing tions between research advocacy for vaccine, microbicide Hepatitis B clinical trial to assess immune response to differ- and other new prevention technologies over the last four ent vaccination strategies. years, developed a timeline of major milestones in develop- Results: After 1 year, a total of 772 pre and post-presentation ing collaborative efforts and documented case studies of surveys were collected; 404 were completed by such efforts. parents/guardians and 368 by adolescent participants. Of Results: Efforts to link vaccine, microbicide and other pre- these, 209 adolescents aged 12–17 years agreed to be vention research advocacy have highlighted priority needs for screened for Hepatitis B serology. Approximately, 76 (36%) communities involved in HIV research across all prevention of adolescents were eligible for the Hepatitis B vaccine clini- research areas. Development of mechanisms for community cal trial; of these, 43 were successfully enrolled, 11 were eli- involvement in research (beyond CABs), increasing commu- gible and interested, 19 were lost to follow up, and 3 refused nity capacity in the areas of research ethics, HIV scientific lit- participation. In addition, 9 adolescents were enrolled as a eracy and planning for access and distribution are all issues result of Hepatitis B presentations and referrals increasing the that are shared between each research advocacy area. total enrollment to 52 teens. Successful recruitment strategies Conclusions/Discussion: Resources to educate and mobilize included presentations at after school programs and church communities around research for vaccines, microbicides and presentations for parents and adolescents. Reasons for refusal other new prevention technologies is limited, especially when included parental beliefs and mistrust of clinical research, it is not associated with recruitment into a specific clinical and preference to get the Hepatitis B vaccine from a primary trial. Establishing coordinated advocacy, education and train- care provider. ing efforts can maximize the impact of such efforts, build sus- Conclusion: A multifaceted educational program utilizing a tainable support for research overall and better prepare for community mobilization approach and one that includes edu- eventual new product introduction. Efforts to improve com- cating parents appears to be a successful strategy for identi- munity understanding of and support for prevention research fying and recruiting adolescents for a Hepatitis B vaccine are long-term investments with potentially high payoff for clinical trial. It is possible that these strategies hold promise developing new technologies and ensuring they get used in a for recruiting adolescents in future HIV vaccine trials. comprehensive and integrated response to the epidemic. AIDS Vaccine 2006 229 Poster sessions_revGJ 14/8/06 16:34 Page 230 Amsterdam, Netherlands 29 August–1 September 2006 P13E-13 P13E-14 Selectively willing and conditionally able: motiva- African American participation in HIV preventive tions, deterrents, and barriers to participation in clinical trial – a preliminary examination of will- HIV vaccine trials among women at “high risk” of ingness to participate in HIV vaccine clinical trials HIV infection in Philadelphia and the impact of HIV vaccine and other clinical trial knowledge on willingness C Voytek, KT Jones, T Brown, R White, K Mackey, A Fleck, A Rodriguez, D Dunbar, I Frank and K Shakira Washington DS Metzger Community Education Group, Inc., Washington, DC, USA University of Pennsylvania, Philadelphia, USA Background: In the District of Columbia (DC) 1 out of 50 is Background: Women at “high risk” of HIV infection have living with AIDS and has rates of 179.2 per 100,000. Further, been under-represented in prior clinical trials of HIV vac- African Americans account for 75% of all reported AIDS cines. To address this limitation, current multi-site trials have cases in the District. Given these high rates, DC is an ideal stipulated inclusion of a significant percentage of women. location to assess willingness to participate and knowledge of These women are often marginalized due to drug use, sexual HIV preventive vaccine clinical trials. behaviors, poverty, racism, and sexism, and may be leery of Objectives: To assess willingness to participate in HIV pre- involvement with research or the medical establishment. ventive vaccine clinical trials, the level of knowledge of clini- Objectives: To examine motivations, deterrents, and barriers cal trials in this community, and its impact on individual to participating in clinical HIV vaccine trials among women willingness to participate in HIV preventive vaccine trials. at high risk of HIV infection in Philadelphia. Methods: African Americans in DC were recruited to com- Methods: During recruitment for a Phase II HIV vaccine trial, plete an ACASI survey on HIV vaccine clinical trials. in-depth interviews were conducted among 16 African Participants were presented a knowledge assessment (general American women residing in neighborhoods with high over- clinical trials, HIV trials and Tuskegee Syphilis study) and a all HIV prevalence, who use crack cocaine and/or exchange “step-wise” mock consent process of an HIV vaccine clinical sex for money or drugs. The social contexts of drug use, HIV, trial. Willingness to participate was assessed on a four point and research were explored, including reasons for or against scale. An ordinal logistic regression was conducted to deter- involvement in HIV vaccine research. mine the impact of demographics and knowledge on willing- Results: Respondents expressed varying degrees of desire ness to participate. to participate, influenced by the type of research, proce- Results: Of the 395 enrolled, 69% responded to 80% of the dures involved, perceived risks and benefits, and the signif- general clinical trial instrument correctly; 25% responded to icance of the study in their lives. Major motivators 80% of the HIV clinical trial instrument correctly; and 46% included: compensation, the opportunity to gain informa- had heard of the Tuskegee Syphilis study. Only 29% of this tion, and to potentially help others. Several respondents group answered 6 or more of the remaining 7 items correct- reported positive impressions of study staff and recruit- ly about Tuskegee. Between 33–47% of all respondents indi- ment procedures, and some perceived that they experience cated that they were “not at all” willing to participate in a benefits from their participation in research. Major deter- HIV vaccine trial and 3–5% indicating that they were “defi- rents included: wariness about being taken advantage of or nitely” willing to participate. An ordinal regression analysis used (as a “guinea pig”), concern about unknown potential found that an individuals HIV risk status (high risk; 95% CI: future consequences of participation (such as side effects), 1.30–5.65) and knowing someone with AIDS (95% CI: and aversions to needles and injections. Respondents also 1.04–3.22) increasing likelihood of participation in such discussed how personal commitments/priorities, contingen- studies. (χ2=27.6, P<0.0001) . cies, and convenience issues could impede their ability to Conclusions: Knowledge is relatively low and in need of attend study appointments. improvement in this community if decisions regarding partic- Conclusions: Findings highlight the importance of under- ipation are to come from an informed framework. standing factors which influence vaccine research participa- Preliminary data appear to indicate that high risk individuals tion among women at “high risk”, and using such knowledge and persons knowing someone with AIDS are more willing to to tailor study procedures to local contexts. The goal is not participate in these types of clinical trials and that recruiting to convince unwilling women to take part in research, rather efforts should include such individuals in their outreach to address concerns and barriers through education and efforts or recruitment plans. efforts to make participation more convenient to drug-using women. By understanding the structural nature of certain barriers, screening:enrollment ratios can be more precisely estimated and recruitment efforts can be more effectively designed and efficiently implemented. 230 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 231 Amsterdam, Netherlands 29 August–1 September 2006 P13E-15 P13E-16 Willingness to participate in HIV vaccine trials: Willingness to participate in future HIV vaccine the impact of trial characteristics (Project VIBE) trials among persons at risk in South India PA Newman1, S-J Lee2, ET Rudy3, DS Seiden2, M Suhadev1, AM Nyamathi2, S Swaminathan1, L Kakinami2, WE Cunningham2 and N Duan2 P Venkatesan1, R Sakthivel1, Shenbagavalli1, A Suresh1 and JL Fahey3 1 University of Toronto, Toronto, ON, Canada; 2 University of California, Los Angeles, Los Angeles, CA, USA; 3 Los Angeles 1 Tuberculosis Research Center, Indian Council of Medical County Health Department, Los Angeles, CA, USA Research, Chennai, India; 2 School of Nursing, University of California at Los Angeles (UCLA), USA; 3 David Geffan School of Background: The recruitment of tens of thousands of high- Medicine, University of California at Los Angeles (UCLA), USA risk volunteers needed to participate in Phase III HIV vaccine trials for the foreseeable future raises formidable sociobehav- Aim: to explore knowledge and attitude of the persons at risk ioral challenges. of HIV infection on HIV vaccines and their willingness to Objectives: To assess willingness to participate (WTP) in participate in future HIVVTs among the high-risk popula- hypothetical Phase III HIV vaccine trials, and to quantify the tions in Tamilnadu, India. relative impact of an array of trial attributes on WTP using Methods: This cross-sectional study was conducted among conjoint analysis, among ethnically diverse adults from com- 501 respondents during 2005 using semi-structured interview munities at increased HIV risk. schedules among transport workers, people attending sexual- Methods: Participants (n=123; median age=38 y; 69% male; ly transmitted infection clinics, injection drug users, men hav- 37% Latino; 14% African-American) were recruited using ing sex with men, women in sex industry and a representative multi-site, venue-based sampling among vulnerable commu- sample of monogamous married women. Since there was a nities in Los Angeles, California. Conjoint analysis, a tech- general lack of knowledge about HIVVTs, the participants nique for systematically assessing consumer preferences for were educated on HIVVTs after evaluating their baseline discrete choice alternatives, was used to map preferences knowledge. among different HIV vaccine trials. WTP was assessed for 8 Results: Out of 501 respondents, 55% were males, 58% hypothetical HIV vaccine trials that varied across 7 dichoto- under the age group of 31–50 years, 64% were married and 7-4 mous attributes, according to a 2 fractional factorial exper- 79% were school educated. Overall willingness to volunteer imental design. The impact of each trial attribute on WTP for HIVVTs was 82% and the desire to be protected from was estimated using within-subject analysis of variance, then HIV/AIDS was the main reason for being willing to partici- meta-analysed across participants. Statistical significance of pate in a hypothetical vaccine trial. The most common rea- the mean impact of each trial attribute was tested using a sons for refusing vaccinations were not being at risk of HIV two-sided one-sample t test. infection, fear about the stigma, uncertainity about the Results: WTP ranged from 18.5 (SD=30.5) to 70.3 (SD=37.9) receipt of vaccine or placebo and the concern about the safe- on a 0-100 scale across 8 hypothetical HIV vaccine trials ty of the vaccine.The knowledge scale reported a significant (mean WTP across trials=40.5, SD=22.1). Trial attributes increase in their knowledge (P<0.05). 76% revealed the hope associated with greater WTP were risk of vaccine-induced that there would be an effective vaccine in a few years and HIV infection (none versus. small risk, impact=22.1; 95% CI: 71% hoped that the HIV vaccine would protect them from 16.2–28.0; P<0.0001), false-induced HIV positives (no versus HIV infection. Main concern was the unknown efficacy of yes, 13.4–95% CI: 8.8–17.9, P<0.0001), provision of free the vaccine (50%) and the effects of a HIV vaccine on par- HIV medications (yes versus no, 13.1; 95% CI: 8.6–17.6; ticipants’ lives (51%). Overall, 76% agreed that sex without P<0.0001), and duration (3 versus 5 years, 6.8; 95% CI: a condom would not be safe whether or not there was a HIV 3.3–10.2; P=0.0002). Minor physical side effects, number of vaccine. doses, and reimbursement per visit ($75 versus $25) had no Conclusions: It is likely that high-risk volunteers will be will- significant impact on WTP. ing to enroll in HIVVTs. Addressing barriers and concerns by Conclusions: HIV vaccine trial attributes may strongly influ- providing information through appropriate agencies will ence WTP among communities at increased risk for HIV spell out success for HIVVTs in India infection. While existing candidate vaccines cannot cause HIV infection, perceptions of risk may nonetheless impede WTP. Eliciting trial preferences and concerns prior to trial design and implementation may enable investigators to accommodate certain preferences and to tailor educational and social marketing interventions to address concerns, which may facilitate informed and ethical trial recruitment and participation. AIDS Vaccine 2006 231 Poster sessions_revGJ 14/8/06 16:34 Page 232 Amsterdam, Netherlands 29 August–1 September 2006 P13E-17 P13E-18 How do adolescents compare with adults in atti- Factors influencing participation and retention of tudes and outcomes in Vaccine Preparatory Cohort healthy volunteers in a Phase I trial of a preven- Studies tative candidate HIV vaccine in New York City L Bekker1, K Middelkoop1 , H Jaspan1 , L Myer1,2, S Vasan1,2, A Hurley1,2, D Dugin1, E Londoño1,2, and P Mthimunye1 and R Wood1 S Schlesinger1,2 1 Desmond Tutu HIV Centre, IIDMM, UCT; 2 School of Public 1 Aaron Diamond AIDS Research Center, New York, NY USA; 2 Health, UCT Rockefeller University, New York, NY USA Introduction: South Africa is considered a key country in Background: Recruitment of healthy low-risk seronegative which to conduct HIV Vaccine Trials. In light of the high volunteers into HIV preventative vaccine trials has historical- HIV incidence in youth, it is also a country where adoles- ly been difficult. We recently completed a Phase I trial of a cents may be important participants in future vaccine trials. HIV-1 vaccine candidate in New York City, where the epi- However, incidence is not the only determinant of the feasi- demic has hit hardest in the United States. bility of vaccine trials - willingness to participate (WTP) in Objectives: To evaluate factors influencing recruitment and vaccine trials will impact recruitment and retention of par- retention of volunteers in a Phase I preventative HIV-1 vac- ticipants is also critical. We compare reported WTP and cine trial in New York City. retention between adolescent and adult participants in an Methods: 45 healthy low-risk HIV seronegative volunteers HIV-negative cohort study. between the ages of 18–60 (<10% over 45) were pre-screened Methods: We recruited 200 sexually active HIV-negative indi- via phone interview, and then screened by history, physical viduals aged 16–40 years into the prospective cohort in a and laboratory examination. Volunteers received 3 intramus- periurban township near Cape Town, South Africa. Three cular doses of a DNA-based vaccine (ADVAX) and were fol- monthly HIV testing was performed and follow-up was over lowed for 18 months. Compensation was US$100 per visit a twelve-month period. Seroconvertors were followed up on for 12 visits. The study was conducted at the Aaron Diamond a separate protocol. Questionnaires on WTP in HIV vaccine AIDS Research Center (ADARC) and the University of trials were administered at enrollment and 12 months. Rochester. After study completion, motivation for participa- Information on HIV vaccines was provided in a series of tion and retention at ADARC were evaluated with an anony- structured education sessions, held over the study period. mous survey. Retention was calculated as numbers of people still on study Results: 30 volunteers were enrolled at ADARC, in 3 dose- at 12 months excluding seroconvertors. escalation groups of 10 volunteers each. Time from screen- Results: 98 adolescents (16–21 y) and 102 adults (22–39 y) ing to full enrollment improved with each group, at 9, 5, were enrolled. Females made up 87% and 72% respectively. and 4 weeks respectively. Trial volunteers were 60% female 7 adolescents (7%) and 6 adults (6%) sero converted during (18/30) 40% male (12/30), 10% (3/30) Hispanic, and 17% the study. 87% and 82% respectively were still present on the (5/30) African American. 10/30 volunteers identified as study at 12 months. 16% of adolescents were WTP in future homosexual (5 male, 5 female). Two volunteers did not vaccine trials at baseline, increasing to 61% at 12 months complete the study for vaccine unrelated reasons. Twenty- and the remainder were uncertain. In the adult group, one six surveys were returned. Half of the volunteers (13/26) person reported unwillingness to participate in future vaccine were motivated to participate by a friend or family member trials at baseline, 41% reported WTP, increasing to 61% with HIV/AIDS. 15/26 would have participated with no WTP at 12 months. Following education, 47% of adolescents monetary compensation, 11 of whom knew someone with who had reported uncertainty at baseline, subsequently HIV. 25/26 would participate in a second vaccine study, of reported WTP, compared to 28% of adults. whom 12/25 attributed this to a high quality of care from Conclusions: Seroincidence in this study is high amongst both the clinic staff, and 12/25 cited a desire to contribute to youth and adult volunteers. Retention is good in both groups. developing a vaccine against HIV. Whilst young people tended to be uncertain at the start, more Conclusion: We were able to rapidly recruit highly motivated reported WTP after 12 months. Education was more likely to and committed volunteers, over half of whom knew someone increase adolescents’ WTP as compared to adults. with HIV/AIDS. Financial compensation was a much greater factor in participation for those who did not have personal involvement with infected individuals. Future recruitment efforts for clinical trials of HIV vaccine candidates in devel- oped countries may be best targeted to groups and/or indi- viduals directly affected by this epidemic. The greatest factors influencing retention and repeat participation were a positive clinical experience, as well as altruism. 232 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 233 Amsterdam, Netherlands 29 August–1 September 2006 P13E-19 Circumcision as a component of standard of pre- vention for HIV vaccine trials: potential impact on public sector services in Soweto G de Bruyn1, G Gray1, J McIntyre1, M Smith2, R Wesson2, G Dos Passos2 and N Martinson1,3 1 Perinatal HIV Research Unit, University of the Witwatersrand, Johannesburg, South Africa; 2 Department of Surgery, Chris Hani Baragwanath Hospital, and University of the Witwatersrand, Johannesburg, South Africa; 3 School of Medicine, Johns Hopkins University, Baltimore, MD, USA Objectives/Aims: A recently completed randomized con- trolled trial of male circumcision for prevention of HIV acquisition, together with many observational studies, demonstrated that male circumcision is associated with a reduction in the risk of infection. These findings add to the options for personal protection against HIV among partici- pants in HIV vaccine trials, and would likely be considered a component of the standard of prevention that should be in place for all trial participants. Plans are underway to conduct test of concept trials in South Africa soon. It is currently unclear what capacity exists to supply such an intervention to trial participants. Methods: The numbers of circumcision procedures per- formed by general surgical services in the only public sector hospital in Soweto for July 1998 – March 2006 were retrieved from the procedure logs of the services. These num- bers reflect requests for elective procedures among adults (>9 years), and do not include cases performed by pediatric sur- geons or urologists. Results: 2786 procedures were performed in the 93 months of procedure logs reviewed (mean: 30/month). The median age of patients was 21 years. Demand varied by season, being approximately 50% greater in the second half of the year, and the median age of persons requesting procedures also varied by season. The age of persons undergoing procedures has shown significant decline over the time period by a non- parametric test (Cuzick test, P<0.001). Discussion/Conclusions: Uncircumcised men being screened for HIV vaccine trials would potentially benefit from being circumcised prior to participation. Public sector capability exists to perform procedures for this indication. However that capacity would be significantly stretched by the addi- tional burden from trial participants, which may number in the hundreds. This may pose a hindrance to delivering the standard of prevention for clinical trial participants. AIDS Vaccine 2006 233 Poster sessions_revGJ 14/8/06 16:34 Page 234 Amsterdam, Netherlands 29 August–1 September 2006 TOPIC 14: SOCIAL/ETHICAL/ACCESS/REGULATORY ISSUES P14-01 Conclusion: The new guidelines for HIV/AIDS vaccine research will greatly facilitate the process of conducting Development of national guidelines for research vaccine trials in Kenya. The approval process for such tri- and development of HIV/AIDS vaccines: the als is now expected to take a quarter of the time it would previously take. Kenyan Experience WG Jaoko1, KM Bhatt1, F Manguyu2, T Mboya-Okeyo3, JJ Bwayo1, AO Anzala1, J Ndinya-Achola1 C Kambili2 P14-02 and S Kalibala2 Ethical challenges in the recruitment and informed 1 Kenya AIDS Vaccine Initiative (KAVI), University of Nairobi, consent process for an HIV vaccine efficacy trial Nairobi, Kenya; 2 International AIDS Vaccine Initiative (IAVI), in Haiti New York, USA; 3 Ministry of Health, Nairobi, Kenya P Joseph1, S Nerette1, M Bernard1, J Geffrard1, RI Verdier1, M Ascensio1, K Henrys1, N Coicou1, SS Jean1, Background: Clinical trials in most African countries are 1 3 2 2 relatively new phenomena. Approval process for conduct- MM Deschamps , PF Wright , DW Fitzgerald , W Johnson , 1, 2 ing these trials is often not well defined, and may take quite and JW Pape long. In Kenya, these approvals have on average taken 1 year. Given the urgency of responding to the HIV/AIDS 1 Groupe Haïtien d’Etude du Sarcome de Kaposi et des epidemic, it is critical that governments expedite review Infections Opportunistes (GHESKIO) Port au Prince, Haïti; and approval of HIV/AIDS vaccine trial protocols. In 2 Division of International Medicine and Infectious Diseases, 2005, the Kenya government reviewed its approval process Weill Medical College of Cornell University, New York, NY, USA; to make it more efficient, culminating in development of 3 The Division of Pediatric Infectious Disease, Vanderbilt the ‘Kenya National Guidelines for research and University, Nashville, TN, USA Development of HIV Vaccines’. Objectives: To streamline the approval process required for conducting HIV vaccine trials in Kenya, in order to facili- Objectives: The GHESKIO centers offer primary care, HIV tate and support research in this field. voluntary counseling and testing (VCT), and services for Methods: Development of the guidelines was facilitated by HIV associated diseases. Since March 2001, GHESKIO has the Kenya Government. This involved participation of been part of 3 HIV vaccine trials. Volunteers recruitment stakeholders from universities, research institutions, non- and ethical considerations can be, potential obstacles in the governmental organizations, community leaders, and peo- conduct of HIV vaccine trials, especially in low education- ple living with AIDS, among others. The process entailed al/resources settings. Our informed consent process holding of discussions in plenary sessions followed by addresses concerns about volunteers’ understanding and workshops in small thematic groups dealing with different real desire to participate in the study. This abstract aspects of the guidelines, using a generic model developed describes the ethical challenges for a Phase II-b HIV vaccine by WHO through Africa AIDS vaccine programme trial (STEP) in a low illiteracy population. (AAVP). Finally, a national consensus workshop was held Methods: In 2005 the trial was introduced to about 15,000 to complete the document. adults by social workers and by posters, pamphlets and a Results: The Kenya government has developed a national generic video played in waiting rooms. High-risk persons guideline for research and development of HIV/AIDS vac- willing to participate in the trial have three educational ses- cines with a clear protocol approval process and timelines. sions focusing on key aspects of the study and the informed The approval process has 2 well defined stages. In stage consent process. These combine one-on-one meetings with one, a concept paper is submitted to the HIV vaccine sub- counselors and social workers with video and audio tapes committee of the Ministry of Health for review, paying reviewing important aspects of the trial and the consent particular attention to preclinical safety and immunogenic- process. A formal evaluation of understanding and real desire ity data. The subcommittee is expected to respond to the to participate is conducted by a psychologist not involved in investigators within 4 weeks. In stage two, investigators the training process. Volunteers must score higher than 80% submit the full proposal simultaneously to two government to proceed. Contact with participants is facilitated using pre- boards mandated to review the science and the investiga- paid phone cards given to participants, home visits by field tional product, respectively. Response from these boards workers as a reminder of upcoming appointments, and at should be given to the investigators within 8 weeks. The each visit a calendar scheduler highlighting the next visit and approval process is therefore expected to take approxi- documentation of any change of residence. mately 3 months. Results: Of the 2002 high-risk persons referred to the vaccine unit over 40% (825) indicated willingness to participate in 234 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 235 Amsterdam, Netherlands 29 August–1 September 2006 the trial. All three educational sessions were attended by 49% research in South Africa (n=10). Confidentiality issues raised (408/825) of those interested in participating. Over 90% of included the right of parents to children’s HIV or STD infor- those evaluated passed the test (304/335). Of the 304 who mation (n=2) and the obligation to report under-age sexual passed the evaluation, 203 were eligible, 190 signed the con- activity (n=3). Cultural concerns (n=4) such as recruitment sent form and 116 were enrolled. from conservative religious groupings were raised. Conclusions: VCT centers can serve as site recruitment for HIV Stakeholders recommended enrollment challenges could be vaccine trials. Only 25% of volunteers willing to take part were resolved by education (n=3), upfront discussions (n=2) and eligible for enrollment. A comprehensive educational process empirical research and philosophical debate (n=2). for this low literacy population helped ensure informed study Conclusion: This data suggests that stakeholders require participation. Recruiting volunteers for large scale vaccine trials urgent clarification of the legal context for child participa- necessitate intensive and continuous effort. tion. Guidance in meeting complex legal and operational demands for consenting child participants is also required. P14-03 P14-04 What do stakeholders in South Africa view as major ethical challenges in adolescent HIV vaccine Payment for HIV vaccine research participation: trials? South African stakeholders’ perceptions C Milford1, C Slack1 and A Strode2 C Milford, N Barsdorf and G Lindegger 1 HIV AIDS Vaccines Ethics Group (HAVEG), School of HIV AIDS Vaccines Ethics Group (HAVEG), School of Psychology, Psychology, University of KwaZulu-Natal, Pietermaritzburg, University of KwaZulu-Natal, Pietermaritzburg, South Africa South Africa; 2 School of Law, University of KwaZulu-Natal, Pietermaritzburg, South Africa Background: Paying clinical trial participants is increasingly a topic of ethical concern, particularly with more clinical tri- Background: South Africa’s ethical-legal framework is in a als being conducted in resource-poor settings. There is lack of state of flux. New legislation has been passed but not fully consensus on whether or not to pay participants and on cri- implemented, that will change the way that health research teria to determine level of payment. Although payment as with children is conducted. Children under 21 years will compensation is recognized as fair, there are some concerns not be able to consent independently to health research. about payment as an incentive and the implications of paying Parents/legal guardians must participate in all decisions “too much”. relating to involvement of children. The application of Objectives/Aims: This study aimed to elucidate stakeholders’ these principles to trials will lead to many complexities as perceptions of ethical challenges in HIV vaccine trials. One a careful balance is required between child participation issue spontaneously raised by participants was payment. and child autonomy. which is the focus of this paper. Objectives: This research aimed to explore stakeholder per- Methods: Thirty-one intensive interviews were conducted ceptions of critical ethical issues in HIV vaccine trials, includ- with 7 groups: community advisory boards (CABs), site staff, ing adolescent trials. sponsors, ethics committees (RECs), government, civil society Methods: 31 semi-structured interviews were conducted with and media. Participants were asked to identify and describe 7 groups, namely: community advisory boards, site staff, ethical challenges related to HIV vaccine trials. Interviews sponsors, ethics committees, government, civil society and were transcribed. Key themes were identified and coded. media. Interviews were transcribed. Key themes were identi- Data was organised using NVivo, a qualitative data analysis fied and coded. Data was organised with NVivo, a qualitative package. data analysis package. Results: Sixteen out of 31 participants (52%) identified con- Results: Only 5 participants (16%) spontaneously listed child cerns about participant payment. The majority of issues were participation as a concern while others referred to it when identified by site staff (4 issues). The fewest issues were iden- discussing issues such as consent and vulnerable groups. tified by CABs and sponsors (2 issues each). Stakeholders When asked specifically about child participation concerns, were concerned about the influence of payment on choices site staff and civil society representatives raised the most about participation (n=12). Some felt that payment would issues (12 each), followed by sponsors (11). Government rep- unduly influence participation (n=3). Others worried about resentatives raised the fewest issues (5). Almost half of all the impact payment would have on choices of individuals participants asserted that there was a scientific imperative to from vulnerable communities (n=3). The possibility that enrol children in HIV vaccine trials (n=17). The most fre- money was perceived as a benefit and that people participat- quently discussed concern was obtaining consent and/or ed in return for this benefit was also raised (n=4). Three par- assent for trial participation (n=20). Sponsors did not discuss ticipants felt that payment as compensation would be fair. consent/assent. The second largest issue identified by all The tension between reasonable and unreasonable payment groups, was difficulty interpreting the law in relation to child was also described (n=3). Suggested solutions to this ethical issue included negotiations with CABs as to what would be AIDS Vaccine 2006 235 Poster sessions_revGJ 14/8/06 16:34 Page 236 Amsterdam, Netherlands 29 August–1 September 2006 appropriate for the specific community, discussions with Conclusions: Gender may influence not only an individual’s REC members, and transparency with participants. Decisions willingness to enroll and continue participation, but also how around payment should be consultative and may vary others respond in supporting participation. Applying a gen- according to research question and context. der-lens to understanding the motivations, fears, enabling Conclusion/Discussion: The issue of payment to participants factors and constraints of women and men around participa- in HIV vaccine trial research is of concern to many stake- tion, and using that information to shape recruitment, reten- holders. Incorporating these views with sound ethical princi- tion and support strategies will strengthen the potential for a ples, although challenging, might inform substantive and successful outcome. procedural decision making for payment in trials. P14-06 P14-05 Building a new framework to explore global Gender concerns in HIV vaccine research: reflec- demand scenarios for preventive AIDS vaccines tions from key stakeholders in East Africa G Gandhi1, R Hecht1 E Roca1, C Kaufmann2, M Yeh2 and L Nyblade1, J Becker2, S Kalibala3, F Manguyu3 and W Woods2 B Bender3 1 International AIDS Vaccine Initiative, New York, NY, USA; 2 1 International Center for Research on Women, Washington, DC, Boston Consulting Group, Boston, MA, USA USA; 2 International AIDS Vaccine Initiative, New York, NY, USA; 3 International AIDS Vaccine Initiative, Nairobi, Kenya Background: Demand scenario building is important for AIDS vaccines to inform donors about likely future funding Background: Gender issues are likely to impact recruitment requirements, provide developers with credible estimates of and retention of women, and potentially men, in AIDS vac- market potential, and help design R&D incentives that cine trials, and may also influence the social impact of par- include realistic market scenarios. Previous estimates of glob- ticipation on volunteers. In East Africa, the ratio of female al AIDS vaccine demand can be strengthened to address sev- to male participants in early trials declined from one-to-one eral important factors: adoption and utilization patterns at initial contact to one woman for every eight men upon across countries, variations in delivery capacity, price effects, enrollment. While current trials are recruiting more women, and impacts of vaccine characteristics. and clinical research in preparation for efficacy trials has Objective: Develop a framework to explore scenarios of glob- not encountered such disparities, it is unknown whether al demand for future AIDS vaccines. women will readily put themselves forward for an injection Methods: A framework for forecasting demand was devel- in future trials. Participation of women is key because there oped, drawing on previous academic and industrial is potential for a vaccine to behave differently in men and approaches. Estimations of need were based on UNAIDS epi- women and because of the need to ensure that women demiological projections. Countries were grouped by antici- access any future vaccine. pated speed/breadth of AIDS vaccine adoption and coverage, Objectives: Given the paucity of data on women’s participa- using multivariate cluster analyses of time series data (includ- tion, an initial assessment was conducted in Kenya and ing GNI/capita, HIV/AIDS prevalence, Hepatitis B vaccine Uganda to inform development of a social research study and uptake, API score and ARV uptake). Over 50 experts from begin delineation of research questions. the public and private sectors, international organizations, Methods: Individual interviews and group discussions were and the HIV/AIDS community provided input via interviews conducted with vaccine trial staff, community advisory board on the likely impact of vaccine characteristics and price on members, peer leaders and other stakeholders. expected demand. Results: Emergent themes include: a) requirements to avoid Results: The modelling generated a wide range of demand pregnancy during a trial in the face of social pressure to bear forecasts in developed and developing countries, depending children; fears about impact of a vaccine on future fertility on differences in regional and national reactions to expected and pregnancies; contraception side-effects; and vaccine-con- vaccine characteristics and views about broad versus targeted traception interactions; b) women’s concerns about possible vaccination campaigns. This information was combined with negative impact on health and ramifications for families due analyses of secondary data on adoption rates and achievable to their role as caregivers; and concerns about blood draws coverage levels to produce demand scenarios, of which a key given blood loss through menstruation; c) social costs includ- component was high uptake in the initial “catch up” phase, ing sanctions women may face if they do not behave in the when countries may vaccinate a wide age spectrum of sexu- expected way (e.g. visiting a research site); women’s lack of ally active adults. power to make decisions; influence of communities, families, Discussion: AIDS vaccine demand scenario development providers and partners on decision-making; childcare and poses challenges not faced by demand forecasters for exist- mobility constraints; d) economic costs including time away ing/late stage or EPI-administered vaccines. Characteristics of from income generating activities. AIDS vaccines are currently undetermined, the demographics 236 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 237 Amsterdam, Netherlands 29 August–1 September 2006 and uptake patterns for adolescent/adult vaccines will be dif- important prevention tools given the scientific questions ferent, and AIDS vaccines may target hard-to-reach groups posed by HIV vaccine research. Collection and dissemination whose numbers may be unclear. Innovative model-building, of annual data on R&D investments in HIV vaccine R&D framework testing and additional regional and national have proven critical to monitoring current levels of effort and research are necessary to better understand the relationships may be used to: identify trends in investment, spending, and between demand drivers and AIDS vaccine forecasts and to research focus; identify areas needing more resources and create robust tools to help advocates, policymakers, and vac- effort; assess the impact of public policies aimed at increasing cine developers accelerate the advent and uptake of a vaccine investment in new prevention technologies; and provide a that can help end the epidemic. fact base for policy advocacy on R&D investment and allo- cations. P14-07 P14-08 Expanding resources, expanding options: funding for HIV vaccine research and development (R&D) Toward a collaborative model of HIV vaccine between 2000 and 2005 research K Fisher1, G Lamourelle3, M Warren1, P Harrison2, G K Fisher and M Warren Gandhi3, and J Rowley AIDS Vaccine Advocacy Coalition, New York, NY, USA 1 AIDS Vaccine Advocacy Coalition, New York, NY, USA; 2 Alliance for Microbicide Development, USA; 3 International AIDS Objectives: HIV is a disease which poses scientific questions Vaccine Initiative, USA cutting across an unprecedented number of fields including immunology, virology and genetics. Researchers developing Objectives: Information on financial investments in research and testing potential models for an HIV vaccine are required and development (R&D) for new prevention technologies – to obtain rights from numerous parties prior to engaging in vaccines and microbicides – is critical in understanding polit- preclinical research. ical and social commitment to these technologies and the Methods: Product development for the HIV pandemic has impact of public policies aimed at accelerating scientific been hindered by the secrecy and unilateral research model progress. The Alliance for Microbicide Development, AIDS that have limited sharing of intellectual property. Such shar- Vaccine Advocacy Coalition, International AIDS Vaccine ing is vitally important in the HIV vaccine field where vaccine Initiative and UNAIDS jointly developed a comprehensive candidates often include intellectual property (IP) from sever- methodology for tracking resource trends in microbicide and al holders. We analysed over 1300 U.S. patents to private, preventive HIV vaccine R&D. This presentation will provide government and academic assignees covering the most signif- an update on total funding levels and sources for 2005, com- icant AIDS vaccine product matter compositions of HIV mitments to date for 2006, and additional information on genes or proteins, vectors, sequences. how funds for HIV vaccine R&D have been spent. Results: Review of HIV vaccine related patents highlight that Methods: Data were collected on product development, clin- complex vaccines/biologics have IP thickets much greater ical trial preparation, community education, and policy advo- than drugs. National systems vary in their granting of rights cacy efforts to estimate annual investment levels for over samples and roles of trial participants in ways that affect preventive HIV vaccine R&D related activity. All primary research progress. HIV research requires an approach to IP funders were asked to provide data on annual disbursements, management to track risks and uncertainties as they evolve as this gives a more accurate picture of annual investments throughout the changes in the R&D and development cycle, than commitments or pledges made. The field is urgently in need of contractual agreements such Results: In 2004, public, philanthropic and commercial as the “covenant not to sue,” which suspends limits on an investment in HIV vaccines R&D was an estimated US$680 investigator’s freedom to experiment with third party IP. The million. The public sector was the main source of these funds, covenant can accelerate the process of IP sharing by creating accounting for US$602 million (88%) of total investments in a safe haven for research. Research efforts as part of the HIV vaccines. The commercial sector (pharmaceutical and Global HIV Vaccine Enterprise are also moving toward a sys- biotechnology companies) spent about US$68 million (10%) tem of IP sharing at the preclinical stage. and the remaining 2% (US$12 million) came from philan- Conclusions: Development of an HIV vaccine, and new ther- thropic sources. apeutics, require innovative synergies and crosscutting Funding for HIV vaccine research has increased significantly research. We propose a research and action agenda to ensure over the last five years, more than doubling between 2000 that the HIV/AIDS vaccine field learns from the lessons of IP and 2004. In 2005, public sector funding for HIV vaccines collaboration and sharing mechanisms. IP issues cannot be further increased to an estimated US $654 million. resolved completely. Legislative proposals may be useful but Discussion: Despite increased resources, current funding lev- are slow to take effect. A few modest steps can be imple- els are significantly short of those needed to develop these mented now: consortia and patent pools that must address players of unequal IP position; preclinical covenant not to sue AIDS Vaccine 2006 237 Poster sessions_revGJ 14/8/06 16:34 Page 238 Amsterdam, Netherlands 29 August–1 September 2006 agreements, use of trial exemptions, later licensing and pur- chasing maximize freedom to operate and invent; and har- monization and clarity for research exemptions. P14-10 P14-09 HIV vaccine research & development: modelling the path to speedier success Speeding vaccine R&D – what does history tell us? H Wong1, C Gingerich2, K White1 and J Michaud2 H Wong2, S Srinivas1 and S Post2 1 International AIDS Vaccine Initiative, New York, NY, USA; 2 Bill and Melinda Gates Foundation, Seattle, WA, USA 1 John F Kennedy School of Government, Harvard University, Cambridge, MA, USA; 2 International AIDS Vaccine Initiative, Background: As researchers work to overcome the daunting New York, NY, USA scientific hurdles in developing an HIV vaccine, they must also gauge the likelihood and timeline of success. Resolving Background: A major premise of product development pub- this latter challenge is essential for maintaining support for lic-private partnerships, including the International AIDS the HIV vaccine effort, and requires that policy-makers, Vaccine Initiative, is that R&D must be tightly integrated donors and the public have realistic expectations about the and conducted on a large scale to be most efficient. Most time and resources needed to develop an effective vaccine. current AIDS vaccine R&D is conducted in a small-scale, Objectives: To estimate the time and money needed to devel- investigator-driven manner. Many studies assume these two op a successful HIV vaccine, and to identify opportunities for approaches correspond in a one-to-one match to R&D in accelerating the development timeframe. “industrial” and “academic” settings. Yet there is consider- Methods: Given the many parameters and uncertainties able blurring between public and private R&D in bio-med- affecting success, we developed a computer simulation model icine, and “industrial” and “academic” R&D are terms that of the HIV vaccine development pipeline. Model components hide as much as they reveal. A greater understanding of include the number of current candidates in the pipeline, the these terms is needed to gauge optimal arrangements for number of candidates entering clinical trials annually, the speeding vaccine R&D. probability that a candidate successfully moves to the next Objectives: To analyse organizational and institutional con- development phase, and cost and duration of each phase. texts in which vaccine development has occurred to deter- Values for each component were estimated from data on HIV mine elements that contribute to an efficient and rapid R&D vaccine candidates (or other vaccines), and based on guid- process. In particular, to investigate assumptions of what ance from experts, such as the Global HIV Vaccine Enterprise constitutes so-called “industrial” and ”academic” approach- Working Groups. The model was run with “status quo” es: Do these terms represent history accurately? What other assumptions (current spending and scientific knowledge) and, organizational and institutional models have been used? alternatively, with assumptions based on increased invest- What lessons can be applied to today’s AIDS vaccine field? ment in key areas. The difference in outcomes represents the Methods: Secondary data on vaccine R&D history are potential savings in time that targeted investments can yield. analysed to identify institutional contexts in which vaccines Results: Under the status quo scenario, the model predicted were developed in public and private domains. Structured a 50% chance of developing an effective HIV vaccine with- interviews with researchers and others involved in vaccine in 15 years. The cumulative probability of success reaches R&D from academia, industry, and the public sector, and a 90% over 35–40 years. Strategic investments focused on literature review, are used to identify organizational models key pipeline attributes can potentially reduce the timeline and institutional environments for vaccine R&D, types of significantly. Under these latter assumptions, the cumulative markers used for progress across models, and scientific and probability of success within 15 years increases to 80%, technological challenges faced. These factors are incorporat- and to 96% after 20 years (improving the outer timeline by ed into a preliminary framework for their relative contribu- 15–20 years). tion to the success of vaccine development and introduction. Discussion: Computer simulation models, although far Results: While every vaccine has shown unique scientific from perfect, can estimate how improvements in specific characteristics and challenges, vaccine R&D processes, espe- attributes of the vaccine development pipeline can affect cially in recent years, share some similar characteristics that success. Additional investments in the range of $140 mil- led to success. In some cases, these included coordinated lion per year, focused on vaccine discovery as well as labo- efforts with thoughtful project management and oversight ratory optimization and standardization, could throughout the stages of development. In other cases, break- substantially accelerate the timeline. throughs occurred under different organizational parameters or institutional contexts. The implications of differences across models, and a discussion of the blurred boundaries of “industrial” and “academic” research today, are elaborated with specific suggestions for future research into the issues and recommendations for accelerating AIDS vaccine R&D. 238 Antiviral Therapy 11, Supplement 2 Poster sessions_revGJ 14/8/06 16:34 Page 239 Amsterdam, Netherlands 29 August–1 September 2006 P14-11 P14-12 Community activism, good science and the poli- Be part of something Big! Promoting volunteerism tics of HIV vaccine research in Canada: a case in HIV Vaccine clinical trials study of Phase IIb industry-sponsored trials with marginalized women MdR Leon1, P Goicochea2, M Bernal3, H Sanchez2, and J Sanchez2 D Elliott 1 Asociacion Civil Impacta Salud y Educacion, Community Simon Fraser University, Vancouver, BC, Canada Education, Lima, Peru; 2 Asociacion Civil Impacta Salud y Educacion, Grimaldo del Solar , Miraflores - Lima, Peru; 3 Freelance Background: HIV preventive vaccine research continues to designer, Calle Don Aurelio, dpto. Surco, Peru be plagued by both social and scientific barriers in drug research. Vancouver’s Downtown Eastside, an inner city Aim: Large scale clinical research trials in healthy popula- community internationally known for its illicit drug econo- tions of men who have sex with other men are a new con- my, sex industry, and epidemic rates of HIV and HCV infec- cept in Peru. They need to know that efforts are being tion has been identified as a site for industry-sponsored made to deal with the AIDS epidemic and that require of Phase IIb preventive vaccine trials targeting at-risk women, clinical trials and community involvement. We developed a who in this community are politically and economically dis- communication campaign (“we love you/we want you. Be advantaged (primarily IDUs, impoverished, Aboriginal, and part of something Big!”) targeted at promoting the partici- young). pation of MSM in HIV prevention studies that include HIV Objectives: This paper examines on-going controversies vaccines, pre-exposure prophylaxis and the use of acyclovir and socio-cultural barriers surrounding the implementa- for genital herpes. tion of Phase IIb industry-sponsored HIV preventive vac- Methods: Using strategic communication planning we cine trials in Vancouver where in April 2006 they began assessed community expectations on HIV prevention clini- enrolling trial subjects. cal research using qualitative approaches. We performed Methods: This paper is based on qualitative, ethnographic nine focus groups in two large cities of Peru, a discussion research in Vancouver – using observation at a community with a MSM community and consultations with the forum; open-ended interviews with vulnerable women, com- Community Advisory Board. Formative research revealed munity activists, and the trials clinical team; and a content concerns of community members regarding social harm and media analysis between December 2005 to August 2006. stigmatization, and the “Guinea pig” affair. Motivations for Results: The analysis reveals that community members and trial participation included altruism and be part of the solu- activists have concerns regarding meaningful informed con- tion to the AIDS epidemic, having access to comprehensive sent, the practice of honoraria, stigmatization, increased vul- sexual health care. nerabilities of trial subjects, the safety of new drugs, and the Results: “We love you/we want you. Be part of something possibility of a global bioethics. Researchers attempted to Big!” was launched in June of 2005 and has contributed to highlight the positive impact an AIDS vaccine could have on successful increase in enrollment figures in a HIV Vaccine global suffering, stressing altruistic principles and the exi- Trial, and in an acyclovir intervention for the prevention gency of pharmaceutical research for all. Three divergent HIV among HSV-2(+) HIV(-) MSM in three different cities positions emerged from the analysis characterized by con- in the country. trasting views of 1). who “community” was and who had the Conclusions: Strategic communication planning may provide right to speak for the community, 2). the ability of impover- innovative approaches to promote participation in HIV pre- ished populations to make informed decisions when cash vention studies. It must address altruism (to include the com- payment was involved, and 3) the possibility of positive, munity of interest as part of the solution of an international immediate and direct impact on the local community (as effort) with an emphatic tone and an edutainment approach. opposed to impacting the “global AIDS pandemic”). While Also provide key messages to educate community members some degree of resolution was achieved, the community con- and potential study participants and to train peer educators tinues to be split regarding its support for the trials. and counselors. Conclusions: This paper reveals that community concerns surrounding the inclusion of historically vulnerable popula- tions in clinical trials continue to be a significant barrier for researchers carrying out preventive vaccine research in urban Canadian settings - forcing us to reconsider how vac- cine trials can be conducted in a manner which produces good science and yet are socially responsible. AIDS Vaccine 2006 239 Poster sessions_revGJ 14/8/06 16:34 Page 240 Author index 14/8/06 16:47 Page 241 Amsterdam, Netherlands 29 August–1 September 2006 AUTHOR INDEX Aasa-Chapman M OA05-05 Altfeld M S04-03 Ball JK P09a-41 Bergmann S P05-34 Abbink P OA08-02 OA07-05 Balla-Jaghjoorsingh S P09b-12 Berkhout B P09d05 S02-02 P05-32 P09d-08 P09d06 OA08-01 P03-05 Balzarini J OA07-02 OA02-05 Abdool Karim SS OA01-04 P03-07 Bandawe G P08-09 Bermudez C P06-05 P08-09 P03-08 Bansal A OA04-07 Bernal M P14-12 Abdul Hussein S P09b-18 Altman JD OA01-07 Barin F P01-03 Bernard M P14-02 Abel K OA09-01 Amara R P09c-07 Barnett S P09a-49 Bernasconi D P05-08 S06-05 Ambrozak D P05-09 P09c-03 Berneman ZN P05-01 Aboud S P11-05 Americo J P09b-22 P09a-65 Berry N OA09-03 P11-07 An I P09b-23 P09a-74 P10-10 P13e-02 P09c-08 P09a-61 Bess Jr. J P09a-48 OA03-02 Andersen H P05-31 P09a-66 Bessette B OA04-01 Acharya P S05-03 Anderson D P09a-67 P09d-10 Betts MR OA08-03 Adam RI OA07-01 Anderson DE P09a-44 S05-06 OA01-07 Adams WC P05-13 P09a-50 P09a-38 Bewley CA P09a-22 P05-09 Andersson S P09a-63 P09a-45 Beyrer C P08-03 Adeyemi AO P13e-01 Andrews C OA02-07 P09a-29 P13c-08 Adjé-Touré C OA01-05 Andrews S P09b-32 P12-01 PL02-01 Adon C P13c-08 Anklesaria P OA03-01 P09b-05 Bhatt KM P14-01 Aerts JL P09a-77 Annappa S P09a-75 Baroncelli S P10-05 P11-01 Aga E S10-02 Anton P P10-10 Barouch DH S02-02 Bhattacharya T PL02-02 Aguilar-Sino RO P09a-24 Anzala AO P11-01 OA08-01 OA02-03 Ahmad S P09b-06 P14-01 OA08-02 Biberfeld G P11-05 Ahmed R OA04-02 Anzala O P01-02 Barreto-de-Souza V P09a-08 P11-07 Ahues P OA01-02 P13e-03 P03-02 P13e-02 Ait-Mohand H P05-18 P13a-08 Barsdorf N P14-04 OA03-02 Alaeus A P05-08 OA06-01 Barsov E P05-31 Bibollet-Rouch F P06-08 Alam M S05-01 P13b-04 Bart PA S10-01 OA01-03 Alam SM P09a68 Arancio A P09a-39 Bartholow BN OA06-03 OA05-07 P06-12 Aricò E P09b-19 Barzee S P09c-10 OA07-06 OA05-06 Arntzen C OA02-04 Bass E OA06-07 S01-03 Albert D P09a-50 Arreola SG P13a-10 Basu S P09a-72 Bichare S P05-14 Albert J P11-06 Arroyo M PL02-01 P09a-73 P05-15 Albert P P12-08 Arthos J P01-05 P09a-69 Bieler K P09a-35 Alberti C P09a-71 Asbach B P09a-33 Bates AT OA08-01 P09b-13 Albuquerque A P05-12 Ascensio M P14-02 Batten J C P07-01 Binder A P09b-11 Alcântara LCJ P09a-46 Asher TE P05-22 Bauer A P09a32 Binley J OA05-04 Alfsen A OA02-04 Astronomo R P09a-25 Bäuerle M P05-34 Bins A P09b-15 Alicea C P12-08 Ataman-Onal Y P09b-09 Baumgartner T P05-26 Birungi J OA06-01 P05-28 Athanasopoulos P P09b-03 Bautista C P06-05 P13c-06 P09a56 Atkinson J P11-10 Bear J P12-08 Birx D P05-07 Allard SD P09a77 Aubertin AM OA05-02 Beard C P09b-07 P13a-03 Allen M OA06-02 P06-03 Beaton A P09c-01 P08-03 P11-09 Aucoin S P09a-44 Becker J P14-05 P06-05 P11-11 August TJ P09a-56 Beddows S P09a-14 P09a-20 Allen S S08-04 Auma B P05-06 Beels D P05-33 P11-02 P09a-41 P05-18 Beenhakekr N P09d-08 PL02-01 P13b-04 S10-01 P09b-12 P13a-04 OA01-02 P05-19 Beer B P12-01 OA03-02 OA01-03 Avila MM P13b-06 Beirnaert E OA05-05 P08-14 OA06-06 Azizi A P09a-50 Bekan B P11-10 Bishop K P05-16 P13e-06 P09a-44 P13a-05 Bizimana J P13a-05 OA07-06 Ba L P08-12 P13d-01 P13d-01 P13a-07 Baak I P09b-12 Bekker L P13e-17 P11-10 S01-03 P09d-08 PL03-01 P13a-07 P13a-05 Baan E P01-04 P13e-04 P13e-06 P11-10 Baden LR P11-12 Bell J OA04-04 Blaak H P05-09 Allen SA P13d-01 P09b-21 Belli R P10-04 Blackwell JL OA01-03 Allen T S04-03 Baedeker N P12-05 P10-05 OA07-06 P03-08 Bagaya B P13b-03 Bellino S P01-09 Blackwood E S06-05 Allen TM OA07-01 Baglyos L P09b-21 Bello G P08-01 Blanc C P05-19 P07-05 Bahati N P13a-08 Bender B P14-05 Blattner W P09-b21 OA07-05 Bailer R S07-03 P13a-08 Blay W OA07-07 Allgaier R P03-08 OA08-03 Benenson M P11-03 Blazevic V P12-02 Allison M P08-04 Bajcz M P09a-57 OA03-04 Blish C OA05-01 Almeida MR P13e-05 Bakari M P13e-02 P11-15 Blomberg P P09b-01 P13b01 OA03-02 P13b-05 Blumenthal R P09a-20 P05-19 Baker B P05-29 P11-08 Boaz MJ P13a-05 Almond N OA09-03 Baker D P09a-23 OA03-03 OA06-01 Alter G P03-07 S05-04 Benesi AJ P09a20 Boberg A P09a-02 P05-32 Baksaas I P12-03 Berg L P03-01 P09a-03 P03-05 Baldwin C OA02-05 Bergamaschi C P09a-56 Boeckl K P09b-14 AIDS Vaccine 2006 241 Author index 14/8/06 16:47 Page 242 Amsterdam, Netherlands 29 August–1 September 2006 Boeras D OA01-02 Buonaguro FM P10-06 Charifi F P09a62 Coovadia HM OA04-02 Boeser-Nunnink B P06-04 P09b19 P06-02 Cope A P10-10 Bogers W P09a29 P08-10 Charles MH P09b09 Coplan PM P13d02 Bogoshi M OA06-05 Buonaguro L P09b19 Chatani M P13e11 Coppens S P09a18 Bollinger L S09-05 P10-06 Chatora B OA01-02 Corey L PL01-02 Bollinger RC P06-11 P08-10 Chavan R P09c08 P13a-09 Bomsel M OA02-04 S06-05 Chen B OA08-02 P11-11 Bonnet B P11-16 Buonocore-Buzzelli L S06-05 Chen H P09a58 P09b21 Bontjer I OA02-05 Burastero S P09a71 Chen L S05-03 Corpus E P01-07 Boren D P09a68 Burda S P07-04 Chen Y OA08-04 Cortesão C P05-12 Borges AR S01-02 P08-06 P09b30 Cosma A P07-08 Borghans JAM P05-02 Burke B P09c03 Chen Z P08-12 Costagliola D P09a09 P09a40 P09a49 P09b26 P05-19 Borowski L P05-25 P09b05 OA09-05 Cotter C P09b22 Borrow P P12-07 P09a54 P08-12 Coutinho-Silva A P05-11 Borsetti A P10-05 Burke D P11-09 Cheng C OA02-07 Couto-Fernandez JC P08-01 P09a38 Burton DR S05-02 Chennareddi L P09a60 Cox J P05-10 Bosch RJ S10-02 P09a24 Chequer-Fernandez SL P08-01 P09b01 Bosire K P11-01 P09a26 Cherni I P09a55 OA03-02 Bou-Habib DC P03-02 P09a78 Chernyaeva E P08-05 P05-07 P09a08 OA05-04 Chertova E P09a48 P11-02 Boulassel R OA04-01 OA02-06 Cheruiyot J P13a03 Cranage M P10-10 P05-26 P09a25 Chetty S OA04-02 Crawford H OA07-04 Bourgault Villada I P11-16 OA02-01 Chevalier A P11-13 P05-20 Bourn W P09b11 P09a22 Chikhilar PR P09a56 Crimi C P09a52 Bowers D P09b11 P06-06 Chin AC P05-24 Cristillo A P09b17 Bowlsbey A P12-01 Busari AA Chirmule N P11-16 Critchfield JW OA01-06 Boyer JD OA09-07 Chiu JN PL02-01 Crooks E OA05-04 P12-08 Busari OA P13e01 Choge I P02-01 Crostarosa F P10-04 Bozza MT P03-02 P13e01 OA01-04 Cruz C P01-05 Bozza PT P09a08 P13e01 P07-02 Cunha-Neto E P05-11 Braibant M P01-03 Busch MP S01-04 Chohan B P01-01 Cunningham WE P13d04 Brander C PL02-02 Butman B OA02-07 Chomatas ERV P13b01 P13e15 P05-16 Butto S P05-08 Chomatas M P13e05 Currier J P05-07 P05-21 P09a39 Chomba E OA06-06 Cuzin L P11-16 P05-22 Bwayo JJ P13a01 P13a07 D’Agnolo R P05-11 OA07-01 P14-01 S01-03 Da Silva B P05-19 OA07-04 P11-01 Chomont N OA04-01 Dai A P09c06 Branson BM S08-02 Byland R S01-05 Choudhary A P09a76 Dai JJ P09b02 Bratt G OA03-02 Cafaro A P03-04 S01-02 Dally L OA06-01 Bråve A OA03-02 P09a38 Choudhry V S01-02 Dang V OA02-07 P09b01 P10-05 Chulay J P11-09 Darden J P13c04 P09c02 Cahn P P05-10 Chultemsuren B P01-07 Das AT P09d06 P09c11 Caire-Faria-Neto HC P09a08 Chuma D P13a04 P09d05 P11-06 Calarese D OA02-01 Chunsuttiwat S OA03-03 Das B P09a04 P09a02 Calarota SA P09c06 Cicala C P01-05 P09a01 Breda D P09a71 Calori G P09a70 Ciccozzi M P08-10 Davenport M P07-01 Bredl S OA08-07 Cameron M P05-30 Cisto J P09a65 Davies A P13c09 Brenchley JM P05-22 Campbell C P13a10 P09a66 Davies S P06-07 P09a57 Campbell GR P09a75 Clerici M P09a74 Davis D P09a29 Briggs J P09a17 Campbell-Gardener L P09a14 Cleveland B P09a67 Davis KL OA05-07 Brigido LFM P08-11 Capecchi B P09a49 Cleveland M P09c01 Dawson P OA02-01 P13b01 Caputo A P09a74 Climent N P09d02 Dawson PE P09a22 P13e05 Cardoso R OA02-01 Clivio A P10-09 Day CL OA04-02 Brockman M P07-05 Cardozo T P09a43 Clore GM P09a22 OA04-06 OA07-01 Carlin N P09a02 Clumeck N OA03-01 Day S P09c04 P05-27 Carta M P01-09 Cohen C P13b02 De Araujo J P13c03 Broder C P09a76 Caruso L P09b30 Cohen J S03-02 De Boer RJ P09a10 S01-02 Casimiro D OA03-07 Coicou RN P14-02 P05-02 Broder G P11-18 P11-16 Coldiron M P13d01 De Bruyn G P13a11 Bronson R P09b10 Casoli C P09a71 Coley S OA03-05 P13e19 Brown A OA03-04 Castejon MJ P08-11 Collacchi B P09a39 de Gleria K P12-07 Brown B P06-05 Castro P P09d02 P01-09 De Groot A P13b02 Brown BK P09a20 Catone S P10-05 Colleton BA P05-17 P09a27 Brown D P12-07 Celentano C P13c08 Collins K P11-16 P09a28 Brown JA OA04-02 Celentano D P08-03 Colocho E P13e12 de Haard H OA05-05 Brown T P13e13 Cellerai C OA04-05 Colombo G P10-09 De Mareuil J P09a75 Brunel FM P09a22 S04-02 Colón R S06-05 De Mori G P10-09 OA02-01 Centlivre M P09d05 Combadière B P05-18 de Oliveira T P09a46 Bu X P06-01 Chahroudi A P09b25 Combes O P09a45 De Rose R P07-01 Buchbinder S S07-02 Chaiwong V P13b05 Condra J OA03-07 de Souza M P11-02 P13a10 Chakrabarti B P03-06 Conlon C P12-07 Debard N P13c04 P13a09 Chalwe V P08-13 Connolly M P09a65 Debre P P09a09 P11-14 Cham F P09a76 Connolly NC P05-25 Decker J P06-08 Bucy RP S10-02 S01-02 Connors M P06-09 P06-06 Bull L P11-18 Chang G P08-12 Consonni R P10-09 OA01-03 Bunnik E P06-10 Chang H P09b06 Conway J OA04-07 OA07-06 Chaplin P P09a52 Cooper D OA07-05 S01-03 Coovadia H OA04-06 242 Antiviral Therapy 11, Supplement 2 Author index 14/8/06 16:47 Page 243 Amsterdam, Netherlands 29 August–1 September 2006 Decoville T OA05-02 Eamsila C OA03-03 Fast P OA06-06 Franti M P09a14 P06-03 P08-14 P13e06 P06-02 Deeds B P13e12 Earl P P09b22 P13a07 OA05-04 Deers M P09b21 P09b01 P13c06 P09b31 Delair T P09b-09 S06-05 P11-10 P10-03 Delgado K P06-02 OA03-02 P13e03 S06-01 Delmarcelle Y P09-c01 Easley KA P09b25 OA03-01 Frasca V P09a66 Delwart EL S01-04 Easlick J OA09-01 P13b04 Freeman GJ OA04-02 Demanet C OA01-05 S06-05 P11-01 Frey S P09b21 Deml L OA08-07 Eda Y P07-07 PL03-02 P11-11 Deneka M S01-05 Effros RB P05-24 Fauce SR P05-24 Frost R P09a50 Denner J P09a-53 Egan M OA08-06 Fauci AS P01-05 P09a44 P10-07 Eibl MM P01-06 Fazio A P09a39 Fuchs J P13c04 Dennis M P10-10 Eichbaum Q OA04-02 Fedanov A P09b32 P11-11 Derby N P09a54 Eismann K P05-34 Feinberg MB P09b24 Fujiwara M P05-05 Derdeyn CA OA07-06 El Habib R S10-02 P05-23 P07-03 OA01-03 Eldridge J OA08-06 P09c08 Fuller S P09a17 S01-03 P11-11 P09b23 Fumero E P09d02 Desaint C P11-16 Elizaga M P11-11 OA01-07 Funkhouser R OA02-03 Deschamps MM P13a-09 P09b21 P09b25 Gachie J P01-02 P11-14 Eller L P13b03 Felber B P09b10 Gadgil T P13a02 P14-02 P11-02 Felber BK P05-28 Gagarina S P08-05 DeSouza M PL02-01 Eller M P13b03 P09a56 Gagnon B P05-26 P08-14 P11-02 P12-08 Gaines H P11-07 P13b03 Elliott D P14-11 Feng S OA08-05 Gakhal AK P06-06 Desrosiers R P09b10 Elsley W OA09-03 Feng Y P09a76 Galant S P09b11 Dey AK P09a14 Enama ME S07-03 Fenyö EM P10-08 Galatchyants YuP P09b04 Dey B P09a64 Engfeld S P12-05 Ferarri G P11-09 Gall J OA02-07 P09a78 Engram JC OA01-07 Ferguson D OA09-03 Gallart T P09d02 Di Carlo A P01-09 Engström G P11-06 Ferko B P09a42 Galley L P11-02 Diab BY OA09-07 Ensoli B P01-09 P09a47 Gallo RC P12-04 Diaz-Mitoma F P09a44 P09a74 Ferlenghi I P09a49 Gallotta G P09a70 P09a50 P10-04 Fernandez C P07-01 Galvão-Castro B P09a46 Dickson G P09b03 P03-04 Fernandez M P13b06 Gandhi G P14-06 Dieltjens T P09a18 P09a38 Fernandez V P11-16 P14-07 Dillard Smith C OA06-03 P10-05 Ferrantelli F P09a38 Gandhi R P03-08 Dilraj A OA08-01 P09a39 P03-04 Ganesh L P03-06 Dimitrov D P09a78 Ensoli F P09a39 Ferrari G P11-12 Gao F PL02-02 S01-02 P01-09 Ferreira L P13e05 S05-01 Dittmar MT P09a42 Erfle V P07-08 Fiebig EW S01-04 P08-13 Djomand G P11-18 P09a74 Fiebig U P09a53 Gao H P06-11 D’Offizi G P01-09 Ertl HC P09b29 Finnefrock A OA03-07 Garber DA P09b24 Doms RW P06-06 Ertl P P10-02 Fiorelli V P01-09 P09b25 Donastorg Y P11-14 Esparza J PL03-03 P09a39 P09c08 Dong M P09a76 Esquieu D P09a75 P03-04 P09b23 Dong T OA04-04 Essex M P02-04 Fisch DN P09a14 P05-23 P12-07 P13c01 Fischer W OA02-03 OA01-07 Donnelly J P09a-61 Esteban M P09d03 PL02-02 Garcia JC OA01-06 P09c-03 P04-01 Fischl M S10-02 Garcia M P05-12 P09b-05 P09d08 Fisher K OA06-07 P06-05 P09a-65 Evangelista ACA P13e05 P14-07 P09d02 P09a-66 P13b01 P14-08 García F P09d02 P12-01 Ewald BA OA08-02 Fitzgerald D S07-02 Garcia- Martinez JV S06-04 Donners H P06-06 Excler JL P11-04 P14-02 Gardiner D P09a72 Doria-Rose N OA07-07 Excler J-L OA03-01 P13a09 P09a73 Dorrell L P12-07 Eyer-Silva WA P08-01 Fleck A P13e13 P09a69 Dos Passos G P13e-19 Fagrouch Z P09a29 Flisher A P13e04 Garg S P01-08 Douek D P05-20 Fahey J P13e16 Flood D P13c04 Garry D P09b10 P09a57 Faircloth KL P05-22 Flores L P13e12 Gatell JM P09d02 P05-22 P05-21 Florese RH S05-06 Gaunitz S P09a02 Douglass N P09b-11 Fajola A P13e08 Foley B OA07-07 Gaunt C P11-16 Draghia-Akli R P09b-17 Famoyin C P13e08 Fonseca LAM P05-11 Gavioli R P09a74 Du XS P09a58 Fan Z P05-25 Fonseca SG P05-11 Gazarian K P09a21 Duan N P13e15 Fanales-Belasio E P03-04 Forsman A OA05-05 P09a12 P13d03 Fang Q OA09-05 Forthal D OA02-06 Gazarian T P09a12 P13d04 P09b26 S05-06 Geffrard J P14-02 Dubey S OA03-07 Farah B OA06-01 Fox L S10-02 Geijtenbeek TBH OA07-03 P11-16 P11-01 Foxall R P05-12 Geldmacher C P05-07 Dudek T S03-05 Farcomeni S P09a38 Frachette M P09b12 Geller L P05-30 Duerr A S07-02 P10-04 Frahm N P05-21 Genescà M OA04-03 Dugin D P13e18 P10-05 P05-22 OA09-06 Dukhovlinov IV P09b04 Farid R P08-10 PL02-02 Geng Y P09d07 Dukhovlinova E P08-05 Francavilla V P01-09 Gerhardt M P05-07 P09b04 P09a39 Geyer B OA02-04 Dunbar D P13e13 Franchette MJ P09d03 P09a55 Durham M OA06-03 Franchini G P12-08 Ghanakaran S S01-03 Durier C P11-16 Frank I P13e13 Ghate M P05-15 Duvall M P05-09 Frankel F P09b28 Gherardi MM P05-10 Dwek R P09a25 P04-01 AIDS Vaccine 2006 243 Author index 14/8/06 16:47 Page 244 Amsterdam, Netherlands 29 August–1 September 2006 Ghezzi S P09a70 Gray G OA06-04 Harron S OA09-03 Holl V OA05-02 Ghorbani M P09a44 P13e19 Hartigan-O’Connor D OA09-01 P06-03 P09a50 P13a11 Hartog K P09a61 Holmes H P09a74 Ghys P P08-07 OA06-05 P09a45 Holt R OA06-02 Giavedoni L P09b27 Greer C P09b05 Harwig A P09d06 Holterman L OA08-01 Gil C P09d02 Grieser H P09a07 Haule A P05-07 P05-13 Gilbert P P11-09 Grise H P09a48 Havenga M P05-13 S02-02 Gilbert P P09b21 Grobler J P08-02 S02-02 OA08-02 Gilks C P05-06 Grosskurth H P05-06 OA08-02 Honda A P07-07 Gill D OA06-01 Grünewald K P09a17 OA08-01 Honeyborne I OA07-04 Gill H P09b32 Gruters RA P09a77 Hayes P OA06-01 P05-20 Gillespie G OA04-04 Guan L P11-16 Haynes B S05-01 Hong KX P08-08 Gillis A P06-05 Gudmundsdotter L P05-08 OA08-05 P09b02 Gilmour J P13a05 P11-06 P09a68 Honnen W P06-01 OA03-01 P09a02 OA02-03 Hoshiro Y S03-05 P11-01 Guenaga FJ P09a23 PL02-02 Hotrawarikarn S P01-03 OA06-01 P06-09 P06-08 Hoxie JA OA05-07 P11-14 Guérin C P11-16 OA05-06 Hu N P09b06 Gingerich C P14-10 Guha-Niyogi A P03-06 P08-13 Hu S-L P09a67 Gissman L S03-03 Guillet JG P11-16 P06-12 Huang C P09a25 Giuchuki C P01-02 P09b21 He L P09b17 Huang X P09a15 Giulianelli M P01-09 Guillon C P09a13 Heald A OA03-01 P05-17 Giuliani M P01-09 Guimarães ML P08-01 Healey D P05-26 P05-25 Gladden AD OA07-05 Guimaraes-Walker A P12-07 Hebblewaite D P12-01 Huang Y P09a69 Glashoff R P09a74 Gupta P OA08-04 Hecht R P14-06 P09a72 Glaunsinger T OA03-01 P09b30 S09-05 P09a73 Gnanakaran G OA07-06 P05-25 Heckerman D P05-16 Hübner J P10-07 Gnanakaran S OA07-07 Gupta S P13c08 P05-21 Hudacik L P09b17 Godoy-Ramirez K P11-07 Gurunathan S OA03-04 Heeney J P09b15 Hughes P P13b04 Goebel FD P07-08 Gurwith M OA03-04 P10-02 Hull R OA09-03 Goepfert P OA03-05 Gutierrez E P05-11 P09d08 Hunsmann G P03-03 P06-08 Guzman CA P09a74 P09b12 OA09-02 P09b21 Haanen J P09b15 P09a29 P10-03 OA04-07 Haddad EK OA04-01 Heidari S P09b12 S06-01 Goicochea P P14-12 P05-30 Hejdeman B OA03-02 Hunter E S01-03 Goldberg E P09b27 Hahn B PL02-02 P05-08 P13c07 Goldberg M P09a27 S05-01 P11-05 P09b23 Gomez P OA02-07 OA02-03 P13e02 OA01-02 S07-03 OA05-07 Helm M P12-05 OA01-03 Gómez C P09d08 S01-03 Helmus RA P09b30 P13a05 P09d03 OA01-03 Hemelaar J P08-07 OA07-06 Gomez-Roman VR P09c03 OA07-06 Henrys K P14-02 Hunter M OA02-06 P09b05 Haigwood N S05-06 Herrera C P10-10 Hural J P09b21 S05-06 OA07-07 Herrmann E P05-07 Hurley A P13e18 Gonzalez-Canali G P11-16 Hain J P12-05 Hessell AJ OA02-06 Hurtado C P09d02 Goode D P01-05 P09a52 Hewitt HS P05-21 Huskens D OA07-02 Goonetilleke N P12-07 Hallermalm K P09a02 P05-22 Idiart R P09a58 Goonetilleke N OA06-01 Halsey RJ P09a19 Heyd B P09a45 Im E-J P09a06 Gorle R P13e12 Halwani R OA09-07 P09d10 Indangasi J OA06-01 Gorny M P06-01 Ham C P10-10 Heyndrickx L P05-01 Indangasi S P11-01 OA05-03 OA09-03 P05-33 India IAVI Team P13a02 Gorriahn D P12-05 Hambira R P13c01 P07-04 Isaacs R OA03-07 Gorse GJ P11-12 Hamer R OA04-04 Hicks C P06-08 Isaacs Z P09b11 Gostick E OA08-03 Hammer D P09a30 Hilt S P09a66 Isaguliants M P09a02 Gotch F P09b03 Hammer J P05-11 P09b05 P09a03 P05-06 Hammer S P13a09 Hinkula J P09c11 Israel H OA06-02 P09a74 P09b21 P03-01 Ito TS P13b01 Goudima G P11-13 P11-14 P09c02 Iwai LK P05-11 Goudsmit J P05-13 S08-05 P10-06 Iyer SPN P09a14 S02-02 Hangartner L OA02-06 P09a02 Jackson A P11-16 OA08-02 Hanke T P09a06 P09b01 Jackson B P13a10 OA08-01 P09a05 Hinz A P09a42 Jaeger H P12-05 Goulder P P05-20 P12-07 P09a47 Jamieson B P05-24 OA07-04 Hannah S P13a08 Ho D S07-05 Janabi M P13e02 OA04-06 Haque VK OA02-07 P09a73 Janbazian L OA04-01 P05-16 Harari A OA04-05 P09a69 Janssens W P07-04 OA04-02 S04-02 P09a72 P09a18 Gouws E P08-07 S10-01 OA09-05 Jaoko W P13e03 Graf M P09a32 P09b12 P08-12 P11-01 P09a35 Harbort J P03-03 Hocker L P09b17 P14-01 P09d03 Harley CB P05-24 Hodara V P09b27 Jarrett O P10-07 Graham B OA03-06 Harrad D P13e05 Hoelscher M P08-02 Jaspan H P13e04 P11-02 P13b01 P05-07 P13e17 S07-03 Harrer E P12-05 Hoffman I P13a06 Jaye A P06-08 OA08-03 P05-34 Hoffmann D P07-08 Jean S P11-14 Graham S P13c09 Harrer T P12-05 Hofman S P10-02 P13a09 Gray E P07-02 P05-34 Hofmann H P09b14 P14-02 OA01-04 Harrison P P14-07 Hohn O P07-06 Jeena PM OA08-01 P02-01 Harro C P09b21 Holdsworth IM OA06-05 244 Antiviral Therapy 11, Supplement 2 Author index 14/8/06 16:47 Page 245 Amsterdam, Netherlands 29 August–1 September 2006 Jeffs S P09a35 Kan E P09a61 Kestens L P05-33 Kornbluth R P09c10 P09b18 P09a66 OA01-05 Kornilaeva G P11-13 P06-07 P09a29 Ketter N P13b04 Korobova S P11-13 Jennes W OA01-05 P09a49 Ketunuti M P02-03 Kosovac D P09a34 Jensen R P09a22 P09a65 P02-04 Köstler J P09b14 Jia M P08-12 P09c03 Khamboonroeng C OA03-04 Koup R P05-09 P08-08 P09a45 P11-08 S07-03 Jiang S S01-02 P09d10 P13b05 P05-13 Jiménez M P09d08 P09b05 Khan A P09b17 OA08-03 Jiménez V P09d03 Kan-Mitchell J P09a57 Khawaja S P13c08 Koutsoukos M P10-02 Jin X P05-04 Kanungi J P13c09 Kibler J OA06-03 Kozlov A P08-05 Jittiwutikarn J P08-03 Kappes J OA04-07 Kibuuka H P13b03 P09b04 Johansen K P09c11 Karamov E P11-13 Kiepela P OA04-02 Kozlov AP P13a06 Johansson S P03-01 Karim S P11-09 P05-16 Kozlov AP P09b04 P09a02 P02-01 OA04-06 Krachmarov C P06-01 Johnson B P13c06 Karita E P13e06 P05-20 Kraehenbuhl J-P P13c04 Johnson LS P13b04 P11-10 OA07-04 Kran A-MB P12-03 Johnson P OA09-04 P13a05 Kierstead L P11-16 Krebs FC P09a20 OA03-01 P13a07 OA03-07 Krebs M OA06-06 Johnson RW OA01-03 S01-03 Kijak G P08-03 Krogsad P P09d07 Johnson W P06-06 P13d01 P08-14 Krohn K P12-02 Johnson WD P13a09 P13e03 PL02-01 Kuate S P09b31 Johnson Jr W P14-02 P13b04 Kilby JM S10-02 Kuiken C OA02-03 Johnston N P09b11 Kärkkäinen T P12-02 Kim J P03-06 Kulkarni SS P06-11 Johnston R P09b20 Karlén K P11-05 OA03-04 Kunasol P P11-08 OA09-04 P11-07 P08-14 OA03-04 P09b07 Karpov VL P09a03 P11-15 Kurth R P09a53 Jojic N P05-21 Karras M P13c05 P13b05 P10-07 Joly P P09d10 Karre K P03-01 P11-08 P07-06 Jones B P09b06 Kaslow RA OA04-07 OA03-03 Kutsch O OA05-07 Jones J P09a14 Kasongo W P08-13 P11-03 Kutzler M P09a28 Jones KT P13e13 Kaspers J OA08-01 Kim KS OA09-02 P09a27 Jones Y OA04-04 Kaspryzyk T OA07-07 S06-01 Kvale D P12-03 Joseph P P14-02 Kataaha P P13b03 Kimura T P09a43 Kwon Yd S05-03 Jourdain G P01-03 Katinger D P09b18 OA05-03 Kwong P P09a23 Jurgens C OA09-04 Katinger H P09a42 King CR OA02-07 P09a64 P09b20 OA02-01 Kiplangat S P13a03 S05-03 Kabamba K OA03-01 P01-06 P13a04 OA05-07 Kadie C P05-21 P09a47 Kirchherr JL P08-13 P09a78 Kaewkungwal J P11-15 Katinger HWD S01-02 Kjeken R OA08-06 Lacor P P09a77 P11-08 Katlama C P05-18 Klasse P J P06-02 LaFaunce A P13b02 OA03-03 Kato PK OA06-01 P09a62 Lahaie M P07-05 OA03-04 Katzman J P13e10 Klaver B P09d05 Lai L P09a60 Kafaar Z P13c02 Kaufmann C P14-06 P09d06 P09c07 Kagee A P13c02 Kaufmann D P05-27 Klenerman P OA04-02 Lai Z P06-01 Kakade M P11-04 P05-29 Klimov N P08-05 Lakhi S OA06-06 Kakinami L P13e15 Kaur A P09b10 P09b04 Lallemant M P01-03 Kalams S P11-11 S03-05 Klots I P09a67 Lally M P09a28 Kalbina I P09a63 Kaushik S P09a16 Klumpp SA P09b10 P13b02 Kaldma K P10-01 Kavanagh D P03-08 Knipe D P09b10 P09a27 Kaldor J OA07-05 P05-27 S03-05 Lam J P13c04 Kaleebu P P05-06 OA04-06 Koenig E P13c08 Lambert A P11-18 P13e03 Kawashima Y P05-05 Koevoets C OA01-01 Lamourelle G P14-07 P13b04 Kayda D P12-01 Koh W OA05-05 Lancelot S P09a75 P13c06 Kayitenkore K P13e03 Kohli R P09a01 Landucci G OA02-06 Kalibala S P14-05 P13b04 Koita OA P09a27 S05-06 P14-01 P13e06 Koito A P07-07 Lang S P09b10 P13a08 P11-10 Koizumi H P05-05 Langat L P13a03 Kalil J P05-11 Kayitesi K P13a07 Kolber M P04-02 P13a04 Kallas E P05-11 Kazenova EV P09b04 Komaroff W P11-10 Lange J S10-01 Kalman D P09b25 Keefer M OA06-02 P11-01 P01-04 Kalyanaraman VS S05-06 P09b21 Konde C P13c06 Langhammer S P10-07 Kamali A P13b04 P11-11 Kone M P09a27 Larke N P09a06 P13a08 Keita N P13b02 P13b02 Lasaro MO P09b29 P13e06 Kelleher A OA07-01 Koning F OA01-01 Lau C OA06-02 P13e03 OA07-05 P08-06 Laufer D P11-11 Kambili C P13e06 P05-27 Koopman G P10-02 Laufer N P05-10 P14-01 Kelvin D P05-30 P09b15 Launay O P11-16 P11-10 Kennedy J OA03-05 P09a29 Laurén A P10-08 P13b04 Kensinger RD P09a20 Koornstra W P09a29 Laurikainen K P12-02 P13e03 Kent S P05-03 Kootstra N OA01-01 Lavereys L P01-01 P11-01 P07-01 Korber B PL02-02 Lawrence D S07-02 Kepler T OA08-05 OA07-07 P11-12 Kesmir C P05-02 S05-01 Lawrence J S10-02 P09a40 OA02-03 Lawson B P01-08 Kessler P P09a45 S01-03 Lazzarin A P09a71 P09d10 OA07-06 P01-09 Kestelyn E P13a07 P05-21 Le Gall S OA02-02 Korn K P05-34 Le Goff F P10-01 AIDS Vaccine 2006 245 Author index 14/8/06 16:47 Page 246 Amsterdam, Netherlands 29 August–1 September 2006 Le Grand R P09a74 Liu H P09b05 Manguyu F P14-01 McElrath M P11-11 P10-01 Liu J P09a48 P14-05 P08-04 Leavitt R OA03-07 P09b06 Manigart O OA01-02 McGowan I P10-10 Lee E P13e11 Liu L P09b26 P13c07 McIntyre J P13a11 P13e10 OA09-05 OA06-06 P13e19 OA06-07 Liu Q P09b02 Manyonyi GO P11-01 OA06-04 Lee F P06-06 Liu Y P09c05 Marcon L P09a28 OA06-05 Lee H P09a25 P09b02 P09a27 McKinney-Baker D P09a52 Lee S-J P13d04 P09a15 Marfatia K P09c08 McKnight A OA05-05 P13e15 P09b02 Marincola F P09b19 McLellan-Lemal E P13a10 Leechanachai P P01-03 P09a15 Mark D P13e04 Mcleod K P01-05 Legg H P09b05 Livingston B P11-12 Markham P OA03-05 McMichael A S07-01 Leiherer A P02-02 Ljungberg K P09b07 P12-08 OA04-04 Leikam D P09a31 P09b01 P09b17 P09a06 Leite O P05-11 Lockett G OA06-03 Markowitz M P01-06 P12-07 Leitner T PL02-02 Londono E P13e18 Marone R P13b06 P09a05 Lemckert A S02-02 Long R P11-16 Marovich M OA03-02 McMurry JA P09a27 Lemckert AAC OA08-01 Longhi R P10-09 Marques Jr. ET P09a56 P09a28 OA08-02 P09a70 Marsch M S01-05 McNally LM OA08-01 Lemongello D OA01-06 Longo O P01-09 OA09-01 McQuoid M P09b24 Lennox J S10-02 P09a39 S06-05 OA01-07 Leon MdR P14-12 Lopalco L P09a70 S05-06 P09b23 Leone P P09a38 P09a71 Martin G P09a45 Medeiros TX P09a08 Leroy O P13c05 P10-09 P09d10 Mehandru S P01-06 Lesch A P13c02 Lore K P05-09 P09a61 P11-04 Leseka N P02-01 P05-13 Martin JE S07-03 OA03-01 OA01-04 Loret EP P09a75 Martin L P09d10 P13a02 Leslie A P05-20 Lori F S10-05 P09a65 Mehrotra D S07-02 P05-16 P09c06 P09a61 P11-16 OA07-04 Louis DH P13a09 P09a45 OA03-07 OA04-02 Louis JM P09a22 Martin S P13a04 Meier A P03-05 Letourneau S P09a05 Lu B OA09-05 Martin WD P09a28 P05-32 Letvin N OA02-07 Lu D OA09-06 P09a27 Meier Angela P03-07 OA02-03 Lu H S01-02 Martinelli E P01-05 Menez A P09a61 PL02-02 Lu L P08-12 Martinon F P10-01 P09a45 Leung A P09a69 Lu S OA03-05 Martinson N P13e19 P09d10 P09a72 Lu W P05-19 Marx PA OA02-06 Mesesan K OA06-05 P09a73 Lu XZ P08-08 Mascola J OA02-07 Metch B OA06-02 Lewis GK P09b19 Lubeck M P11-11 P05-09 P11-11 Li B OA01-03 Luchters SMF P01-04 P09a64 Metzger DS P13e13 OA07-05 Luckay A OA08-06 P06-09 Meyer B P11-16 Li DF P09c05 Ludvigsen A P09a52 S07-03 Meyerhans A P05-07 Li F P08-04 Ludwig C P02-02 Masharsky A P08-05 Meyers A P09a36 Li H OA05-07 P09a33 P09b04 P09a19 OA03-06 Lütschg V P09a34 Masse B P13a06 Meyers T P07-02 P09d01 Lyamuya E P13e02 Mathews C P13e04 Mhalu F P11-05 P11-12 Lynch DM OA08-02 Mathiesen I OA08-06 P13e02 Li HJ P09b02 Ma M OA06-03 Mathy N P09c01 OA03-02 Li HY P05-11 Ma T P09c05 P10-02 Mhlanga P OA06-01 Li J OA09-06 Maboko L P05-07 P09a76 Michael E OA02-05 Li Y P06-09 Macchia I P09a38 OA09-03 Michael N OA03-04 Li Y P09a67 P10-04 Matoba N OA02-04 P13a04 Li Z P09b28 P03-04 P09a55 PL02-01 Lian Y P09b05 Mackey E P05-27 Matsunaga R P08-11 P13b03 P09a29 Mackey K P13e13 Matsushita S P07-07 OA03-02 P09a66 Madden V OA09-04 Maulen S P13b06 Michael NL P06-05 P09c03 Maddon P P09a14 Maxi A P14-02 Michaud J P14-10 Liang H P09d01 Mage RG P09a22 Mayer K P13b02 Middelkoop K P13e17 Liao H-X P06-12 Maggiorella MT P10-04 P09a27 Miedema F P09a40 S05-01 P10-05 Mayol K P09a13 Mildvan D S10-02 Lichterfeld M P03-08 P09a38 Mboko L P08-02 Milford C P14-03 S04-03 Magnani M P09a74 Mboya-Okeyo T P14-01 P14-04 Lidman K P05-08 Mahalanabis M S05-06 McAdams M P09a68 Miller CJ S04-04 Lifson J P09a48 Mahmoud A PL03-03 P06-12 OA04-03 P09b10 OA06-01 McBurney S P09a11 OA09-06 Lifson JD P05-31 Mairena EC P05-11 P09a07 Miller J P09d07 S10-04 Majeed S S05-03 OA08-04 Miró JM P09d02 Liljestrom P P09b12 Majluf-Cruz A P09a12 McCaffrey R P09a54 Misher L OA07-07 P09d03 Makhema J P13c01 McChesney MB OA04-03 Mishra S OA06-01 Lind I P09a63 Malaza AL P08-02 S06-05 Mitsuyasu R S10-02 Linde CH P05-22 Malhotra U P08-04 McClelland RS P13c09 Miura T P03-08 P05-21 Malogo R P11-01 McCutchan F P08-02 Mkhize P P05-20 Lindegger G P14-04 Malone S PL03-03 P05-07 Mlisana K P08-09 Liniger M P09d09 Maloveste S P06-02 P08-03 OA01-04 P09b16 P09a62 P08-14 P02-01 P09d04 Mambo Muvunyi C P11-10 PL02-01 Mlungwana O OA06-04 Lisziewicz J S10-05 Mandaliya K OA05-01 P06-05 OA06-05 P09c06 Mandl J P01-08 McElrath J P09b21 Mogale M OA06-04 Liu D P05-04 P05-23 246 Antiviral Therapy 11, Supplement 2 Author index 14/8/06 16:47 Page 247 Amsterdam, Netherlands 29 August–1 September 2006 Mogg R P11-16 Müller S P05-34 Niccolai LM OA06-05 Overbaugh J OA05-01 OA03-07 Mullins J P08-04 Nickle DC P05-21 P01-01 Mohlala A P13c03 P08-09 Nicolette C P05-26 P09a14 Moise L P09a28 OA07-04 Nielsen L P13c06 Owzar K OA08-05 P09a27 P05-16 Nieuwenhuis I P09b12 Oxendine S P13a10 Mokili JL OA07-06 P05-21 P09d08 Oyaro M P11-01 Molina E P13a10 Mumba T P13c09 Nieves-Duran L P12-01 Pacheco GJ P09a08 Mónaco D P04-01 Munderi P P05-06 Nikolaeva I P11-13 Page M OA09-03 Monini P P03-04 Munier S P09b09 Nilsson C P11-05 Pahwa S P04-02 P09a38 P09a13 P11-07 Pal R P09b17 Montefiori D OA03-05 Muntean W P01-06 OA03-02 OA03-05 P09a67 Murakami T P07-07 Niphuis H P09a29 Palacios-Rodríguez Y P09a12 P06-11 Murashev BV P09b04 Nitayaphan S OA03-04 Palamara G P01-09 P09b05 Murasheva IV P09b04 OA03-03 Palasudhi K P11-15 P09a50 Murphy G P13c08 P08-14 P13e02 P09a60 Musengamana V P13a05 Nkengasong JN OA01-05 OA03-02 S01-02 Musonda RM P08-13 Nkosi PS P09a37 Pancera M P09a64 Moodie Z OA03-06 Mutea G P01-02 Nolin J P08-04 S05-03 Moody MA OA05-06 Mutimba S P13c09 Noonan E P11-12 Paniccia G P09a39 S05-01 Mutua G P13e03 Norley S P07-06 S04-02 Moog C OA05-02 Muturi M P01-02 P09b31 Pantaleo G OA04-05 P06-03 Muvunyi C P13e06 Norris PJ S01-04 P09b12 Mooij P P09b12 Mwami J P13e02 North D OA09-03 S10-01 P09d08 Mwananyanda L P13c01 Notka F P09a31 P09d03 P09b15 P08-13 P09a32 Pantophlet R P09a24 Mooketsi U P13c01 Mwangoma M P13c09 P09a30 P06-06 Moon T P06-08 Myer L P13e17 P09a35 P09a25 Moore J S06-05 Myszka D P09a51 Novitsky V P02-03 Paolucci C P09a71 PL01-03 Nabatov AA OA07-03 Nwaigwe C P09a07 Papagatsias T P09b03 OA02-05 Nabel G P03-06 Nyamathi A P13e16 Pape JW P11-14 P09a14 Nabel G OA02-07 Nyambi P P07-04 P13a09 Moore P OA05-04 S02-01 P08-06 P14-02 P02-01 P09a78 Nyange J P13a08 Pappalardo BL S01-04 P07-02 S07-03 P11-01 Paranjape R P05-15 OA01-04 Naderi HR P08-10 Nyblade L P14-05 P05-14 Moore S OA06-01 Naim H P09d09 Nyhus J P12-03 P06-11 Mor T OA02-04 Naim HY P09d04 Oballah P P11-02 OA03-01 P09a55 P09b16 P13b03 Paris RM P11-03 Moraes SL P05-11 Nájera J P09d08 O’Brien KL OA08-01 Park KS S06-01 Moran T P09b07 P09d03 Ofek G P09a23 P09b31 Moreau A P01-03 Nakwagala F P13c06 S05-03 Parker S P11-11 Moreau DR P09a46 Naluyima P P13b03 P09a48 Parks R P06-12 Moretti S P03-04 Namwat C P11-08 Ogola S OA06-01 Parodi L P09b27 P10-05 OA03-03 P11-01 Parren PWHI OA02-06 Morgado MG P08-01 P13b05 Ogrel A P09a44 Parrington M P09b06 Morgan C OA03-06 Nanda A OA08-02 P09a44 Parrino J OA08-03 Morgan PA P11-03 S02-02 Oguto H P13a08 Parry J P13c08 Morris L P09a37 OA08-01 P11-01 Pastori C P09a71 OA05-04 Nanfack A P08-06 Ohlen C P05-31 P10-09 P07-02 Nanvubya A P13b04 Oka S P07-03 P09a70 P06-11 P13c06 P05-05 Patel V P09a56 OA01-04 P13e03 Oksi J P12-02 Patil B P13a02 P02-01 Nappi F P03-04 Oligbu G P13e01 Patterson S P09b03 Morris MM OA01-06 Narciso P P01-09 Oliva H P09d02 Paul E P09a57 Morrow M P05-28 Nascimento AS P13e05 Oliveira CAF P08-11 Pavlakis G P09b10 Morrow-Hall G P13a10 Nason M S07-03 P13b01 P12-08 Mortier D P10-02 Natwijuka A P13b03 Oliveira CM P13b01 P05-28 Moss B P09b01 Naus C P13c05 Olivier T P06-08 Pavlakis GN P09a56 P09b22 Navis M OA01-01 Olivieri E P09a38 Pavlova MS P09b04 S06-05 Navratilova I P09a51 P03-04 Pavlova T P11-13 OA03-02 Ndambuki R P11-01 Olson WC P09a14 Pavone-Cossut MR P03-04 Motta MC P09a08 Ndinya-Achola J P14-01 O’Mara L P09b23 P10-04 Moussu H P10-01 P11-01 Ondondo B P12-07 Paxton WA OA07-03 Moysi E P09a54 Ndirangu R P13a04 Onigbogi O P13e08 P01-04 Mphatswe W OA07-04 Ndzamela N OA06-01 Onyango J P11-01 OA02-05 Mthimunye P P13e17 Negri DRM P10-05 Oostermeijer H P09a29 Peacock J OA08-05 Mugaba S P05-06 Nelson JD P09a22 Opi S P09a75 S05-01 Mugisha E P13c06 Nerette Fontain S P14-02 Opollo M P11-02 Peiperl L OA03-06 Mugusi F P13e02 Neuberger MJ P06-07 Osmanov S P08-07 P09b21 Mui S P03-08 Newman M P11-12 Osterhaus ADME P09a77 Pelchen-Matthews A S01-05 Muldoon M OA07-06 P09a52 Osterrieder N P09b14 Peralta L P13e12 P05-21 Newman PA P13d04 Ostrowski M P09b06 Perdue T P13a06 Mulenga J P13c07 P13d03 Ott D P05-31 Peressin M OA05-02 S01-03 P13e15 Otten G P09c03 P06-03 OA01-02 Ng H P09d07 P09b05 Pereyra F P05-29 OA01-03 P05-24 Otten GR P12-01 P05-27 P13e03 Ngo-Giang-Huong N P01-03 Ouma B P13b03 P03-08 OA07-06 Ngumbela KC OA04-06 P11-02 Perez J P11-14 P13b04 Nguyen M OA05-01 Perez M P11-14 AIDS Vaccine 2006 247 Author index 14/8/06 16:47 Page 248 Amsterdam, Netherlands 29 August–1 September 2006 Perez-Jimenez E P09d03 Rabussay D P12-01 Roederer M PL02-03 Sandstrom E P13e02 Perkins S OA02-03 Racz P P10-04 S07-03 OA03-02 Perrin G P13a09 P09b31 OA08-03 P11-06 Peshu N P13c09 Ramjee G P08-09 P05-09 P11-07 Pevear DC P09a14 Ramshaw I P09c04 Roelofsen PML P01-04 P11-05 Phadke M P13a02 Ranasinghe C P09c04 Rollier C P09b15 Sanford H P09b10 Phogat S P09a51 Ranki A P12-02 Rollman E P05-03 Santana-Bagur J S10-02 P06-09 Rao S OA02-07 Rollman E P07-01 Santelli Antonille T P06-05 Pialoux Gi P11-16 Rappuoli R P09a49 Rong R OA07-06 Santiago LE P11-14 Piantadosi A P01-01 Rathod A P05-29 Rosa Borges A P09a20 Santos LA P09a46 Piazza P P05-25 Rathod A P05-20 Rosati M P09a56 Santos S P05-11 Picker LJ S02-04 P05-16 P05-28 Saraiva EM P09a08 Picq I P10-01 P05-27 P12-08 Sarche J P13a10 Piechocka A P05-29 Reddy S OA07-04 Rosati M P09b10 Sateren W P13b03 Piippo I P12-02 OA04-02 Rose J S06-05 Sattentau Q P09a17 Pilcher C S10-02 Reece J P07-01 Rosenberg E P03-08 P06-07 Pinter A P06-01 Rees H PL01-01 P05-27 Sauermann U P10-03 Pitisutthitham P P11-03 Reeves P P09b25 Ross T P09a07 P03-03 OA03-04 Regis EG P03-02 P09a11 OA09-02 P11-15 Regoes R P05-23 OA08-04 S06-01 Plana M P09d02 P01-08 Rossenkhan R P02-04 Sawe F P13a04 Plonk K P09a68 Reichman R P05-04 Roth P P12-08 Scanlan C P09a25 Pobsuk C P11-15 Remme H P10-01 P09a56 Scarselli M P09a49 Podack E P04-02 Ren L P09a15 Rourke T OA04-03 Scearce R P09a68 Poignard P P09a62 Rerk-N S P11-15 Rousseau C P08-09 P06-08 P06-02 Rerks-Ngarm S OA03-04 Routy J-P OA04-01 P06-12 Poizot-Martin I P11-16 OA03-03 P05-26 Schätz B P05-34 Polacino P P09a67 P08-14 Roux K P09a48 Schätzl HM P07-08 Poli G P09a39 P11-08 Rouzioux C P05-19 Schaubert KL P09a57 P09a74 P13b05 Rowland-Jones S OA04-04 Schellens IMM P09a40 Pollakis G P01-04 P11-03 P06-08 Schengrund CL P09a20 OA02-05 Restrepo S P09b17 P05-09 Scherbak N P09a63 OA07-03 Revesz K OA05-03 P12-07 Scherer EM P09a26 Pollard RB OA01-06 Rezza G P08-10 Rowley J P14-07 Schief W S05-04 S10-02 Richardson B P09a67 Rowley M P09a12 Schiff B P09a23 Polo J P09b05 Richardt JL P07-06 Rozis G P09b03 Schlesinger S P13e18 Polonis V P09a20 Richman D OA05-04 Ruan YH P08-08 Schmidt B P05-34 S01-02 P06-06 Rudy ET P13e15 Schmidt C P13c06 Polonis VR P06-05 Riddler SA P05-25 Ruff LE P05-22 OA03-01 Poonam P OA08-04 Ridolfi B P10-05 Ruiz M P01-09 Schmidt K OA09-01 P09b30 Rinaldo C P05-25 Ruiz-Alvarez MJ P09a39 S06-05 Post S P14-09 P05-17 Russell N P11-09 Schmidt S OA05-02 Powell R P08-06 Riou C P05-30 OA03-06 P06-03 Prado J OA07-04 Rischard J-F OS05 P09b21 Schmolke M P09a53 Prausnitz M P09b32 Ristola M P12-02 Russell ND P11-12 Schneider L P12-05 Precopio M P05-09 Rivera D P09a27 Ruzagira E P13e03 Schneidewind A OA07-01 OA08-03 P09a28 P13b04 P07-05 P05-13 Robb M P13a03 Rwanyonga M.R. P13a08 Schnuriger A P05-19 Premsri N OA03-03 OA03-02 Rybczyk K OA06-02 Schols D OA07-02 P11-08 P08-03 Rybicki EP P09a19 Schröder U P09c02 P13b05 P11-02 P09a36 P09c11 Pressley M P09b07 P13a04 Saa D P08-06 P10-06 Price D P05-20 PL02-01 Sadoff J S02-05 Schuck P P01-05 P05-22 P13b03 Sahasrabuddhe S P11-04 Schuitemaker H P06-10 OA08-03 P06-05 Sahay S P13a02 OA01-01 P09a57 Robert-Guroff M S05-06 P11-04 P06-04 Price M P13c09 Roberts D S02-02 OA03-01 Schulte R OA09-02 P13e06 Roberts DM OA08-02 Sahner D S10-02 S06-01 P13e03 Roberts J P12-07 Sakthivel R P13e16 Schumacher T P09b15 P13a08 Roberts KJ P13d03 Sakthivel S P09b32 Schürmann D P07-06 P13b04 P13d04 Salles J-P P09a75 Schwarz R OA06-04 Priddy F P09b32 Robertson M S07-02 Salmon D P11-16 Scoglio A P09a39 Princiotta M P05-31 OA03-07 Salomon H P05-10 P01-09 Puissant B P05-18 Robinson H P09a60 P04-01 Scott J P13b04 Pujari A OA06-01 P09c07 Saltzman M P09b27 Seamons L OA06-01 Pulendran B S04-05 Robinson J OA05-04 Salvi M. SLL01 Sebinang K P13c01 Pumarola T P09d02 P06-12 Samad F P06-04 Seder R PL02-03 Purandare B OA06-01 P06-08 Samleerat T P01-03 Seebach J P07-08 Puri A P09a20 Roca E P14-06 Sanchez H P14-12 Segura M P13b06 Py P P13c04 S09-05 Sanchez J P14-12 Segurado AC P05-11 Qiu C P09a15 Rockstroh J OA03-01 P13b06 Seiden DS P13d03 Quakkelaar E P06-04 Rodrigues H P05-11 Sánchez-Palomino S P09d02 P13d04 Queiróz ATL P09a46 Rodrigues R P13b01 Sanders E P13a08 P13e15 Quendler H P09a47 P13e05 P13c09 Sekaly RP OA09-07 P09a42 P08-11 Sanders R OA02-05 P05-30 Quinnan G P09a76 Rodriguez A P13e13 Sanders-Buell E PL02-01 OA04-01 QuinnanJr. GV S01-02 P04-01 P05-26 Quirk E OA03-07 Self S P08-04 Qureshi H P09a04 S08-03 248 Antiviral Therapy 11, Supplement 2 Author index 14/8/06 16:47 Page 249 Amsterdam, Netherlands 29 August–1 September 2006 Sempowski G OA08-05 Soare C P09a44 Strimmer K P07-08 Thakar M P05-15 Sernicola L P10-04 Soares R P05-12 Strode A P14-03 P11-04 P10-05 Sobek V P12-05 Strominger J P09a09 OA03-01 Serwanga J P05-06 Sodroski J P09a51 Stutz H P09b11 OA06-01 Seth P P09a01 P09a64 Suh YS S06-01 P05-14 P09a04 P09a78 OA09-02 P06-11 Sette A P05-11 Soka NF P13e04 P09b31 Than S P11-10 Sewell AK P09a57 Sommerfelt MA P12-03 P10-03 P11-01 Shaboltas AV P13a06 Somogyi R P05-30 Suhadev M P13e16 Thanh Thuy NT P01-07 Shacklett BL OA01-06 Song Y P09a73 Sulivan E P01-07 Theiler J OA02-03 Shaffer D P13a03 P09a69 Sumpter BS OA01-07 Theve-Palm R P05-08 P13a04 P09a72 Sun J P09c05 Thielemans K P09a77 Shao Y P09a15 Sopper S OA09-02 Sun MS P09b02 Thior I P13c01 P09a76 S06-01 Sun P P01-05 P11-12 P09d01 Sørensen B P12-03 Sun Q P09b02 Thobakgale C OA07-04 P09b02 Sosa Estani S P13b06 Sun Y P09a29 Thomaz M P13e05 P08-08 Sourial S P09a17 P09a61 P13b01 P09c05 Sousa AE P05-12 P09a45 Thompson J P09b07 Sharma V P09a61 Spearman P P11-11 P09a66 Thomsen L P10-02 P09a29 Spriewald B P05-34 P09a65 Thomson H P13c09 P09a45 Sreenivas V P09a16 P09c03 P11-01 P09a65 Srinivas S P14-09 P09d10 Thongcharoen P OA03-04 P09a66 Sripaipan T P08-03 Sung YC P10-03 Thorner A S02-02 Sharpe S P10-10 Srivastava I P09b05 P09b31 Thorner AR OA08-01 Shattock R P09b18 P09a38 OA09-02 OA08-02 P10-10 P09c03 S06-01 Thorstensson R P10-08 Shaw G P06-08 P09a61 Surasaerneewongse V P01-03 Thulin S P09a63 P06-06 P09d10 Suregaonkar S P05-14 Tichacek A OA06-06 OA05-07 P09a29 P05-15 P13a07 S01-03 P09a45 Surenaud M P11-16 P11-10 OA01-03 P09a49 Suresh A P13e16 Tien P P09b26 OA07-06 P09a65 Suriyanon V P08-03 OA09-05 Sheets RL OA02-07 P09a66 Suschak J P09b17 Timofeev AV P09a03 S09-01 Srivastava R P12-01 Suscovich T P03-07 Tischer K P09b14 Shen H P09b27 Ssemaganda A OA06-01 Sutherland L P09a68 Titti F P09a38 Shen R P09d01 Stahl-Hennig C S06-01 P06-08 P10-04 Shen X P09a52 P09b31 P06-12 P10-05 Shenbagavalli P13e16 P10-03 Suttill A P12-07 Tobery T OA03-07 Shephard E P09a36 OA09-02 Svehla K P06-09 Tomaras G P09b21 P09b11 Stamatatos L P09a54 P09a64 Toppin C P09c10 Sheppard NC P06-07 P09a66 Swaminathan S P13e16 Tornesello ML P09b19 Shevchenko K P12-06 Stamegna P OA02-02 Swanson PE OA08-02 P10-06 Shi Y P08-12 Stanescu I P12-02 OA08-01 P08-10 Shibata J P07-07 P09a74 Swartz L P13c02 Torres JV P09a44 Shiver J OA03-07 P10-01 Tabengwa O P13c01 Toussova OV P13a06 P11-16 Stanfield RL OA02-01 Tack V OA05-05 Tovanabutra S P08-14 Shrotri A P13a02 Staprans S P01-08 Tagliamonte M P08-10 P08-03 P11-04 P09b32 P10-06 PL02-01 Shu Y P06-09 P09b25 Tähtinen M P12-02 Tran B P09c10 Shutes E P11-10 Starodubova ES P09a03 Takiguchi M P05-05 Trautmann L OA04-01 P13b04 Stauber B P07-06 P07-03 Treurnicht F P08-09 P13e06 Stebbings R OA09-03 Tambussi G P01-09 P02-01 P13a07 Stefano-Cole K P09a07 Tang H P06-11 Tripathy SP P06-11 P13a05 Stein J OA02-07 Tang M P09a59 Tripiciano A P09a39 P13e03 Steinman R P09b31 P06-09 P03-04 Siddiqui RA P10-03 Stepanenko RN P09a03 Tang T P09b05 P01-09 Sidney J P05-11 Stevens G P13c07 Tanuma J P07-03 Trivett M P05-31 Sidorov IA S01-02 P13b04 Tanzer FL P09a19 Trocio JN P09c06 Sidorovich I P11-13 OA06-01 P09a36 Trubey CM P05-31 Sikut R P10-01 Stewart-Jones G OA04-04 Tapia G OA04-05 Tsuji M P09a72 Silbermann B P11-16 Sticht H P05-34 S10-01 Tuin K OA02-05 Silva AR P09a08 Stiegler G P09a47 Tarragona T OA06-01 Turk G P05-10 Silva IO P13e05 P09a42 P13a05 Turnbull E P12-07 Silvera P P12-01 P01-06 Tawar R P09a41 Überla K P09b31 Silvestri G OA01-07 S01-02 Taylor K P09a48 Udem S S02-03 Singh M P13e07 Stivali F P09a39 Tcherepanova I P05-26 Ueda M P08-11 Singharaj P P11-03 Stoiber H P09b31 Teigen N P03-05 Ueno T P05-05 Sirijongdee N P11-03 Stone G P09c10 P05-32 Ukpong M P13e11 Sirithadthamrong S P01-03 Stout J P13a05 Teigen Nickolas P03-07 Ulmer J P09c03 Sites D P12-01 Stout R P09b01 Teixeira SL P08-01 P09a61 Sivamurthy R P03-08 OA03-02 Tekirya E P13d01 P09d10 Sjödin A P05-08 Stover J S09-05 P13a05 P09a45 Skhosana N P13a11 Straus SE S03-05 Tener-Racz K P09b31 P09b05 Slack C P14-03 Straus W P13c08 Tenev V S01-02 P09a65 Smith M P05-03 Strbo N P04-02 Tenner-Racz K P10-04 P09a66 P13e19 Streeck H P03-05 Terrat C P09a13 P12-01 P07-01 S04-03 Teruya K P07-03 Urassa W P13e02 Snarsky V P09c10 Streeck Hendrik P03-07 Teyton L P09a26 P11-05 Snyder G P01-05 Strid Å P09a63 Urbanski M P08-06 AIDS Vaccine 2006 249 Author index 14/8/06 16:47 Page 250 Amsterdam, Netherlands 29 August–1 September 2006 Ustav M P10-01 von Gegerfelt A P12-08 Warren M P14-08 Wilson J P09c01 Vajpayee M P09a16 P05-28 P14-07 Wilson L OA03-01 Valentin A P05-28 von Lieven A P13c06 OA06-07 Wilson P P13e10 P09a56 P13a08 P13e11 Wincker N P05-18 P12-08 Vorauer-Uhl K P09a47 P13e09 Winstone N P12-07 Valley-Omar Z P09a36 Voss G P09c01 Washington KS P13e14 Wiriyakijja W P08-14 van Alphen F P06-04 P10-02 Washington S OA06-03 P13b05 van Baalen CA P09a77 OA09-03 Wasunna M P13a03 P11-08 van Baarle D P09a40 P09a76 P13a04 OA03-03 van der Elst E P13c09 Voytek C P13e13 Watana S P11-15 P11-15 van der Ende ME P09a77 Vu B S01-02 Watanabe D S03-05 Wit F S10-01 Van Grevenynghe J P05-30 Vu N P13b02 P09b10 Woldegabriel E P02-04 Van Gulck E P05-01 Vuylsteke B OA01-05 Watanna S P08-14 Wolf H P01-06 van Haaften P P09b12 Vwalika C OA06-06 Watkins DI S06-02 P09b13 P09d08 OA01-02 Watkins JD P09a75 P09b15 van Harmelen J P09b11 Vyankandondera J P01-04 Wattana S P13b05 P09b12 van Laethem K OA07-02 Wabire-Mangen F P11-02 P11-08 Wong C P09a25 Van Leeuwen S P06-12 Wabwire-Mangen F P13b03 OA03-03 Wong C-H OA02-01 Van Lunzen J OA03-01 Wagner A P09a47 Wecker M P11-12 Wong H P14-10 van Montfort T OA07-03 P09a42 P11-09 P14-09 Van Niekirk R OA06-04 Wagner R P02-02 Wei L P09d01 Wood R P13e04 OA06-05 P09a31 Wei X OA01-03 P13e17 van Nuenen A P06-10 P09a33 Weibel S P09b16 Woods W P14-06 P06-04 P09a06 P09d09 Woratanarat T P11-03 van Rompay K OA09-01 P09a30 P09d04 Wright P P09b21 S06-05 P09a34 Weiner D P12-08 P14-02 S05-06 P09b13 Weiner DB OA09-07 P13a09 Van Ryk D P01-05 OA08-07 P09c06 Wrin T P09a54 van Uriaan J OA02-05 P09a32 P09a28 Wu H P05-04 VanCott T OA03-02 P09b15 P09a27 Wyatt L P09b22 P09a14 P09d03 Weinhold K P11-11 Wyatt R P09a37 Vang L P09a52 P09a35 P11-12 P09a59 Vanham G P05-33 P09b14 Weiser B P09a70 P09a51 P05-01 P09d08 Weiss D P09b17 P09a23 P09a18 P09b12 Weiss RA OA05-05 P09a64 P07-04 Wahren B OA03-02 Weissenbacher M P13b06 P09a78 P06-06 P09a74 Weissenhorn W P09a47 P06-09 Vardas E OA06-04 P03-01 P09a42 Xia S-M P06-12 P13c03 P09c02 Welcher B P06-09 Xiang S P09a78 OA06-05 P09a03 Wesson R P13e19 Xiang W P09d01 P09a74 P09c11 West K OA03-05 Xiao W P09b22 Vasan S P13e18 P11-07 Westman M P11-06 Xing H P08-08 Vcelar B P01-06 P11-06 Whalen RG P09a58 Xu J P09a15 Venkatesan P P13e16 P05-08 White K P14-10 P09c05 Ventevogel M OA08-05 P09a02 White R P13e13 Xu K P06-03 Venturi V P07-01 P13e02 Whiteside TL P05-25 Xu L P09a58 Venzon D S05-06 Wakasiaka S P13a01 Whitmore A P09b07 P09a78 Verdier RI P14-02 P11-01 Wieczorek L P06-05 Yamashiro R P08-11 Verevochkin S P08-05 Wakefield S P11-18 P09a20 Yan W P08-12 P13a06 Wala E P01-02 Wiesel M OA08-07 Yang GB P09b02 Verheyden S OA01-05 Walker B P07-05 Wig N P09a16 Yang H P12-07 Verkade E OA02-05 P05-29 Wigdahl B P09a20 Yang L P08-12 Verkoczy L S05-01 P05-20 Wild J P09b13 Yang O P09d07 OA05-06 P05-27 P09a30 P05-24 Vermeire K OA07-02 OA02-03 P09a34 Yassine Diab B P05-26 Vermeulen J S10-01 OA07-04 OA08-07 P05-30 Verrier B P05-18 PL02-02 P09a32 Yegorov O P05-26 P09a13 P03-08 P09a35 Yeh M P14-06 P09b09 OA02-02 P09b14 Yewdell J P09b06 Verschoor E P09a29 OA07-05 Willems B P09a18 Yoshimura K P07-07 Verstrepen B P09b15 OA07-01 Wille-Reece U PL02-03 Young AT OA06-03 Vets E OA03-01 P05-16 Williams C OA05-03 Young DH OA01-06 Victorino RMM P05-12 P05-21 Williams I P09b32 Young K P09b20 Vieillard V P09a09 OA04-02 Williams K P03-08 P09a07 Vieira SM P06-07 S01-01 Williams P P12-07 Yu J P05-04 Villafana T P13c01 Walker BW OA04-06 Williams PM P09a41 Yu Q P09b06 Villinger F P09c10 Walter H P05-34 Williamson A-L P09a19 Yu W OA09-05 Vink M P09d06 Wang H P09b26 P09b11 P09b26 Visintini R P01-09 OA09-05 P09a36 Yu X S04-03 Vita C P09d10 Wang L P09a76 P09a06 P03-08 P09a45 Wang S OA03-05 Williamson C P09a06 Yuan F P05-31 Vogels R OA08-01 Wang X-H OA05-03 P08-02 Yue L OA04-07 S02-02 Wang Y P09d01 P09b11 Yusim K P05-21 Vogt J P09b22 Wanninger V P09a35 OA01-04 OA04-07 Vollmar J P09a52 Warachit B P01-03 P08-09 OA02-03 P12-05 P02-01 PL02-02 Volques C P11-14 Wilson CM OA04-07 Zadorin Y P12-06 Volsky B OA05-03 Wilson I P09a25 Zagury D P12-04 OA02-01 Zaharatos J P09a69 250 Antiviral Therapy 11, Supplement 2 Author index 14/8/06 16:47 Page 251 Amsterdam, Netherlands 29 August–1 September 2006 Zanetti G P09a17 Zhang X P09d01 Zhu P P09a48 Zuber B P05-03 Zaunders J P05-27 Zhang Y P08-12 Zhu W P09b26 Zuniga A P09b16 Zhang GM P09a56 Zhang YW P09c05 OA09-05 P09d09 Zhang JH P08-08 Zhang YZ P08-08 Zhuang K OA09-05 P09d04 Zhang L P09b26 Zhang Z P09a76 Zhuang K P09b26 zur Megede J P12-01 OA09-05 Zhao H P08-04 Zolla-Pazner S S05-05 Zwick M OA02-01 P08-12 Zhao J P09c07 P09a43 P09a22 P12-01 Zhao X P09b28 OA05-02 P09a26 Zhang M P09b28 Zhou C P09b28 P06-01 S01-02 Zhou KM P09c05 P07-04 PL02-02 Zhou T P09a78 OA05-03 P09a78 P09a64 Zolopa A S10-02 Zhang P P09a76 S05-03 Zorrilla CD P11-14 AIDS Vaccine 2006 251