R7694_2_REM_VE_COV_2_2_01.qxd 31/10/06 12:08 Page 2

VOLUME 2 NUMBER 2 2006

EDITORS-IN-CHIEF Graeme Moyle, Chelsea and Westminster Hospital, London, UK Jacob Lalezari, UCSF, San Francisco, CA, USA

Parenteral Inhibitors of HIV Entry in Current Development Jacob Lalezari, Stanley Lewis, G Mani Subramanian, Richard Krawiec, and Mary K Delmedico

The Mechanism and Inhibition of Influenza Entry Zhu-Nan Li and David A Steinhauer

What is the Role for HIV-Entry Inhibitors in Microbicides? Per Johan Klasse

Meeting Report from AIDS 2006

www.viralentry.com

This activity has been planned and implemented in accordance This journal is supported by an with the Essential Areas and Policies of ACCME through the joint educational grant from Pfizer. sponsorship of the University of Kentucky College of Medicine and Remedica. The University of Kentucky College of Medicine is accredited by the ACCME to provide continuing medical education for physicians. The University of Kentucky is an equal opportunity university. R7694_2_REM_VE_COV_2_2_01.qxd 31/10/06 12:08 Page 3

The Journal of Viral Entry is supported by an educational grant from Pfizer

Editorial Policy Publisher’s Statement The Journal of Viral Entry is an independent journal published by Remedica Medical Education © Remedica Medical Education and Publishing 2006. and Publishing. Editorial control is the sole responsibility of the Editors-in-Chief, Editorial Advisory Board, and the Editors. Before publication, all material submitted to the journal is subjected to All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, rigorous review by the Editors-in-Chief, Editorial Advisory Board, Editors, and/or independent or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or reviewers for suitability of scientific content, scientific accuracy, scientific quality, and conflict otherwise without the prior permission of the copyright owners. While every effort is made by the publisher and editorial board to see that no inaccurate or misleading data, opinions, or statements of interest. appear in this journal, they wish to make it clear that the material contained in the publication Aims and Scope represents a summary of the independent evaluations and opinions of the authors and contributors. The Journal of Viral Entry is designed to bring a critical analysis of the world antiviral medicine As a consequence, the board, publishers, and any supporting company accept no responsibility for literature, written by clinicians, for clinicians, to an international, multidisciplinary audience. Our the consequences of any such inaccurate or misleading data or statements. Neither do they endorse mission is to promote better understanding of antiviral medicine across the global healthcare the content of the publication or the use of any drug or device in a way that lies outside its current system by providing an active forum for the discussion of clinical and healthcare issues. licensed application in any territory. Leading Articles The Journal of Viral Entry (ISSN 1748-0329) is published twice a year. Subscriptions are available These major review articles are chosen to reflect topical clinical and healthcare issues at the following rates: Europe Eur65, USA and Canada and all other territories US$85. Additional in antiviral medicine. All contributions undergo a strict editorial review process. subscription information is available from the publisher. Clinical Reviews Remedica Medical Education and Publishing Ltd., Commonwealth House, 1 New Oxford Street, The most important papers from the best of the international literature on antiviral medicine London WC1A 1NU, UK. are systematically selected by an internationally recognized panel of experts. The Editors then Telephone: +44 (0)20 7759 2999 Fax: +44 (0)20 7759 2951 prepare concise and critical analyses of each paper, and, most importantly, place the findings Editorial Team: Cecilia Waller, Scott Millar into clinical context. Editorial Director: Reghu Venkatesan Meeting Reports Design and Artwork: AS&K Skylight Creative Services The Journal of Viral Entry also provides incisive reportage from the most important Publishers: Ian Ackland-Snow, Simon Kirsch international congresses. ISSN 1748-0329 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 2

Editors-in-Chief Contents Graeme Moyle Medical Director, St Stephen’s HIV Research, Editorials Chelsea and Westminster Hospital, London, UK Prevention and Therapeutics Jacob Lalezari Graeme Moyle and Jacob Lalezari 46 Assistant Clinical Professor of Medicine, Mount Zion Recent Results with CCR5 Antagonists Hospital of UCSF, and Director, Quest Clinical Research, Graeme Moyle and Jacob Lalezari 47 San Francisco, CA, USA Editorial Advisory Board Leading Articles Parenteral Inhibitors of HIV Entry David Cooper in Current Development Director and Professor of Medicine, National Centre in HIV Jacob Lalezari, Stanley Lewis, Epidemiology and Clinical Research, Sydney, NSW, Australia G Mani Subramanian, Richard Krawiec, and Mary K Delmedico 48 Robert Doms Chair, Department of Microbiology, University of Pennsylvania The Mechanism and Inhibition of Influenza School of Medicine, Philadelphia, PA, USA Virus Entry J Michael Kilby Zhu-Nan Li and David A Steinhauer 57 Associate Professor of Medicine, Medical Director, UAB 1917 (HIV) Clinic, Associate Director, UAB Pittman Commentary What is the Role for Entry Inhibitors in General Clinical Research Center, University of Alabama Microbicides for Blocking Transmission at Birmingham, Birmingham, AL, USA of HIV-1 via the Cervico–Vaginal or Peter Reiss Rectal Mucosa? Deputy Director of the Dutch National AIDS Therapy Per Johan Klasse 67 Evaluation Center (NATEC), Academic Medical Center, Clinical Reviews University of Amsterdam, Amsterdam, The Netherlands Coreceptors 73 Jonathan Schapiro Drug Interactions 73 Adjunct Clinical Professor, Center for AIDS Research, Stanford University School of Medicine, Stanford, CA, USA, Drug Resistance and Susceptibility 74 and Director, AIDS Service, National Hemophilia Center, Viral Variants 75 Tel Hashomer, Israel Antiretroviral Therapy 75 Vincent Soriano Service of Infectious Diseases, Instituto de Salud Carlos III, Meeting Report Madrid, Spain The 16th International AIDS Conference (AIDS 2006) Editors Toronto, ON, Canada, 13–18 August, 2006 77 James Hellinger Assistant Professor, Division of Infectious Diseases, Department of Medicine, Tufts University School of Medicine, Boston, MA, USA Andrew Luber Consultant, Division of Infectious Diseases, University of Pennsylvania, Philadelphia, PA, USA Mike Youle Director, Royal Free Centre for HIV Medicine, Royal Free Hospital, London, UK R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 46

Prevention and Therapeutics

Graeme Moyle1 and Jacob Lalezari2 1St Stephen’s HIV Research, Chelsea and Westminster Hospital, London, UK 2Mount Zion Hospital of University of California and Quest Clinical Research, San Francisco, CA, USA

J Viral Entry 2006:2(2):46. EDITORIAL This year’s 16th International AIDS conference in Toronto, fusion inhibitors. As reviewed by Jacob Lalezari (Mount Zion ON, Canada, and ICAAC in San Francisco, CA, USA, Hospital of University of California and Quest Clinical brought two themes to the fore: prevention and new Research, San Francisco, CA, USA) and colleagues, a treatments. The rising burden of HIV infection is number of these compounds are parenteral formulations disproportionately impacting lives, life expectancy, and that offer the potential of improved pharmacokinetic economies in many of the world’s poorest nations, creating properties and/or sustained release profiles, which might a vicious cycle. The limited availability and relatively high allow once-weekly, or even once-monthly, dosing. Although cost (relative to gross domestic product or health spend) of parenterally administered agents are unlikely to compete as antiretroviral drugs has not been sufficient to act as a public first- or second-line agents in most settings, their emergence health measure, although they do provide benefits to many could result in greater treatment choices for patients in need individuals. Similarly, the disappointingly slow progress on of salvage therapy. vaccine development means that this option remains As a journal of viral entry, the scope of in this issue distant, at best. As the majority of HIV transmissions occur expands to include a comprehensive discussion by Zhu-Nan through sexual intercourse, easy access to an inexpensive Li and David A Steinhauer (Emory University School of topical product, such as a microbicide preparation that is Medicine, Rollins Research Center, Atlanta, GA, USA) of the protective against HIV transmission and not an irritant to mechanism of entry of the influenza virus. The interaction the vaginal or rectal mucosa, would be highly attractive. between influenza viral hemagglutinin (HA) and host cell Furthermore, microbicides offer the advantage of a means surface glycoproteins containing sialic acid results in of protection that the passive party in a relationship can conformational changes within HA that produce a fusion control and potentially use without their sexual partner’s peptide – a helical coiled coil structure – that, like gp41 of knowledge. Such an agent would complement other HIV, is propelled to the target cell surface to mediate entry preventative measures including condoms, circumcision, into the cell. Interestingly, as noted by Li and Steinhauer, this and one day, hopefully, vaccinations. Importantly, as paradigm for viral entry is not only seen with influenza and described in the Commentary by Per Johan Klasse (Cornell HIV, but is shared with other viruses, including simian University, Weill Medical College, New York, NY, USA), immunodefiency virus and ebola virus. Given the ongoing HIV-specific microbicides such as those based around HIV- threat of an avian flu pandemic, it is hoped that a more entry inhibitors do not necessarily need to act as complete understanding of the mechanism of influenza spermicides and thus may allow safer conception [1]. entry will yield additional antiviral agents to protect Despite the number of obstacles that Dr Klasse describes, humanity from this scourge. the use of small-molecule entry inhibitors in topical microbicides represents an important potential avenue of References research, as several of these agents, such as HIV 1. Klasse PJ. What is the Role for Entry Inhibitors in Microbicides for Blocking Transmission of HIV-1 via the Cervico–Vaginal or Rectal Mucosa? J Viral Entry 2006:2(2):67–72. attachment inhibitors and CCR5 and CXCR4 antagonists, 2. Lalezari J, Lewis S, Subramanian GM et al. Parenteral Inhibitors of HIV Entry in Current possess attractive properties suitable for further evaluation. Development. J Viral Entry 2006:2(2):48–56. 3. Li Z-N, Steinhauer DA. The Mechanism and Inhibition of Influenza Virus Entry. In addition to effective HIV prevention strategies, there is J Viral Entry 2006:2(2):57–66. a pressing need for new agents in the management of patients with multi-class resistant HIV. Studies of the mechanism of HIV entry into CD4+ cells has yielded multiple targets for the development of new entry inhibitors, including HIV attachment inhibitors, CCR5 antagonists, and

46 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 EDITORIAL , + 10 47 increases . Patients + recovery recovery in cell numbers. + + was also seen in copies/mL in the cell count of 146 3 + 10 in the respective cohorts. 3 were were noted with maraviroc once-daily 3 in the placebo group. Emergence of X4 virus, copies/mL in the vicriviroc 5, 10, 15 mg-once- (Gulick R, et al. Abstract THLB0217) 3 ndon, UK copies/mL, and a median CD4 10 . The vicriviroc arms each included 30 individuals; 3 cal Research, San Francisco, CA, USA cal Research, 10 Vol 2 No 2 2006 Vol Results from the ACTG assessing 5211 the study, activity Taken Taken together, the data from these two small studies NTRY daily, daily, and placebo arms, respectively. changes were –1.51, –1.86 At and –1.68 log week 24, the and twice-daily dosing, respectively, suggesting that the X4 CD4 did not unfavorably affect virus emergence vicriviroc vicriviroc treatment arms, respectively, and –0.29 log persons with X4/R5-tropic virus. shortly. expected studies with CCR5 antagonists are Data from large scale of vicriviroc in combination with exclusively an harboring OBR R5 virus in were also individuals discussed at the conference copies/mL in the placebo recovery of +84, +142, and +142 cells/mm group. An impressive CD4 of 48 and 33 cells/mm the vicriviroc arms, compared respectively, with a decline of 9 cells/mm Emergence Emergence of X4 virus was observed in two of 58 patients, 12 of placebo 57 maraviroc once-daily patients, and 12 of 52 maraviroc twice-daily recipients. Interestingly, CD4 cells/mm either as a mixture or as the dominant quasi-species, was detected in 27%, 10%, and 7% of the vicriviroc recipients, respectively, and 4% (one individual) in the placebo group. The 5 mg-vicriviroc arm development of will vicriviroc use the higher doses examined was discontinued and further in this study. indicate that CCR5 antagonists demonstrate good activity in individuals who screen positive for R5 virus only, and may contribute to both antiviral efficacy and CD4 10, or 15 mg once daily, or 10, placebo, or after 15 which mg background once daily, therapy was optimized based on resistance mean change testing. in The viral load was –0.87, –1.15, +0.06 –0.92, log and 28 individuals were randomized to During the first 14 days, patients received either vicriviroc the 5 placebo group. were were triple-class experienced, had a median viral load 4.56 of log favored favored the maraviroc treatment groups, with mean increases of 35.7, 59.6, and 62.4 cells/mm E IRAL 10 V cell the also + 2 . Patients studies of cell counts + the activity of OURNAL OF J HE in vitro T copies/mL, respectively 10 copies/mL, respectively. These 10 and Jacob Lalezari and Jacob and a mean viral load of >5 log 1 3 (2):47. 2 2006: (Mayer H, et al. Abstract THLB0215) cell count. Baseline characteristics indicated that the + With With regards to secondary endpoints, the percentage of Although the expectation is that CCR5 antagonists will be St Stephen’s HIV Research, Chelsea and Westminster Hospital, Lo Chelsea and Westminster HIV Research, St Stephen’s Quest Clini of University of California and Mount Zion Hospital count of <100/mm combinations of entry inhibitors. patients achieving a viral load placebo, maraviroc of 150 mg and once-daily, maraviroc 150 mg <400 copies/mL in the twice-daily cohorts were respectively. For 24.1%, the 24.6%, ‘<50-copies’ assay, 15.5%, and the 21.1%, results 30.8%, and were reductions 26.9%, were relatively similar, changes in respectively. CD4 Although the data suggest antiviral enfuvirtide, as synergy has between been inferred maraviroc from and copies/mL. The primary analysis presented was 24-week data, which showed a change in viral load in the placebo, maraviroc 150 mg and once-daily, maraviroc 150 mg twice-daily groups of –0.97, –0.91, and –1.2 log Progress Progress in the development of small-molecule, oral CCR5 antagonists as therapeutic agents was reported during the 16th International AIDS Conference. This new class of drugs includes maraviroc, currently in Phase III clinical trials and vicriviroc, in Phase IIB trials. A third member of aplaviroc, the was discontinued due class, to hepatic toxicity; however, this problem has not been observed agents in ongoing development. with the other two of most use in individuals carrying only R5 HIV, Graeme Moyle Graeme 1 2 Entry J Viral Recent Results with CCR5 Antagonists CCR5 with Results Recent received received the fusion inhibitor enfuvirtide, the changes –0.89, were –1.26, and –1.44 log (differences (differences not statistically significant). In individuals who patients had very advanced infection with a mean CD4 maraviroc maraviroc in individuals with mixed- or dual-tropic (X4/R5) virus was assessed in a conference Phase IIB study reported at the were were randomized to receive placebo, maraviroc 150 mg once or daily, maraviroc 150 mg twice with daily, ritonavir assumed as a component of the optimized background regimen (OBR). The OBR consisted of three to six approved drugs, and was selected on the basis of resistance testing. Patients entering study were triple-class experienced, had a viral load of >1000 HIV RNA copies/mL to facilitate tropism evaluation, and any CD4 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 47 Page 16:39 2/11/06 R7694_2_REM_VE_2_2_07.qxd R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 48

Parenteral Inhibitors of HIV Entry in Current Development

Jacob Lalezari1, Stanley Lewis2, G Mani Subramanian3, Richard Krawiec4, and Mary K Delmedico5 1Quest Clinical Research, San Francisco, CA, 2Tanox, Inc., Houston, TX, 3Human Genome Sciences, Inc., Rockville, MD, 4Progenics Pharmaceuticals, Inc. Tarrytown, NY,and 5Trimeris, Inc., Morrisville, NC, USA LEADING ARTICLE Studies of the mechanism of HIV-1 entry into CD4+ cells have yielded multiple targets for new drug development. Processes of specific interest include virus attachment to CD4, binding of the virus to cellular coreceptors (CCR5 and CXCR4), and viral–cell fusion. Drugs that block these stages of the viral life cycle, the entry inhibitors, have demonstrated potent antiretroviral activity in preclinical and early stage clinical trials. Owing to their novel mechanisms of action, these new agents also remain active against viruses resistant to both reverse transcriptase and protease inhibitors. Following the example of enfuvirtide, the first entry inhibitor to gain approval for clinical use, a rich pipeline of parenterally delivered entry inhibitors is now in development and offer the potential of further diversifying treatment options as they initiate an era of once-weekly to once-monthly dosing regimens. These investigational agents include TNX-355, a monoclonal antibody that binds to CD4 to prevent post-attachment conformational changes required for entry; PRO-140 and HGS004, two monoclonal antibodies that block HIV binding to the CCR5 coreceptor; and the second-generation fusion inhibitors, TRI-999 and TRI-1144, the latter of which was recently selected for further development. J Viral Entry 2006;2(2):48–56.

The availability of highly active antiretroviral therapy Studies of the mechanisms of HIV-1 entry into CD4+ cells (HAART) has, since its introduction in the mid-1990s have advanced the understanding of the cellular and resulted in a marked improvement in outcomes for patients molecular basis for HIV-1 infection, and identified multiple with HIV type 1 (HIV-1) infection. However, the success of new targets within the extracellular stage of the viral life currently available antiretroviral therapy is limited by the cycle that precedes penetration of the target cell membrane emergence of drug-resistant viruses, the necessity of (reviewed in [3]). Viral entry involves a series of interactions sustained adherence to complex drug regimens, and the between viral envelope molecules and receptors on the cell potential for toxicity. surface, including: Multi-class resistance has rapidly become a problem in the treatment of HIV; 1–2% of newly infected patients in • Virus attachment to CD4. the US and EU harbor virus that is resistant to all three major • Virus binding to a cellular coreceptor (CCR5 or CXCR4). therapeutic classes: the protease inhibitors (PIs), non- • Virus–cell fusion. nucleoside reverse transcriptase inhibitors (NNRTIs), and nucleoside reverse transcriptase inhibitors (NRTIs) [1]. Each of these processes has become a target for the Resistance among the treatment-experienced population is development of drugs belonging to a novel class of also widespread, with an estimated 20% of these patients antiretrovirals, collectively referred to as entry inhibitors. having virus that has lost susceptibility to two of these Owing to their novel mechanisms of action, entry inhibitors three drug classes [2]. This, in combination with the need for remain active against viruses resistant to both RTIs and PIs. simple regimens and drugs with fewer adverse effects, Furthermore, for the same reason, cross-resistance to other clearly illustrates the urgency for novel classes of safe and classes of antiretrovirals is likely to be limited (reviewed in effective antiretroviral agents. [4], and there is also evidence of significant synergistic interactions between different entry inhibitors [5]. Address for correspondence: Jacob Lalezari, Quest Clinical Research, Enfuvirtide (also known as T-20), a synthetic peptide that 2300 Sutter Street, #202, San Francisco, CA 94115, USA. targets gp41-mediated fusion of the viral envelope with the Email: [email protected] host cell membrane [6], was the first entry inhibitor to gain

48 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 49

PARENTERAL INHIBITORS OF HIV ENTRY IN CURRENT DEVELOPMENT

Figure 1. TNX-355 mechanism of action. Attachment conformational epitope on domain 2 of the extracellular to domain 2 of CD4 allows TNX-355 to coat the cell and portion of the CD4 receptor (Fig. 1). The binding site is prevent HIV from entering. distinct from the site required for interaction of the receptor with major histocompatibility complex proteins (domain 1) [10]. Unlike anti-CD4 receptor antibodies that target domain GP120 1, TNX-355 does not interfere with immunological functions variable involving antigen presentation [9,10]. Binding of the MAb loops prevents entry of the virus by blocking post-attachment HIV-1 conformational changes between gp120 and CD4 prior to chemokine coreceptor interactions [9,10]. CD4 Preclinical data demonstrate potent antiviral activity TNX-355 against CCR5- and CXCR4-tropic viruses CCR5 CXCR4 TNX-355 has been shown to inhibit the infectivity of a diverse panel of HIV-1 variants in vitro, including CCR5- and CXCR4-tropic (R5 and X4, respectively) primary isolates [11]. In preclinical studies, TNX-355 reduced viral loads in rhesus monkeys infected with the simian immunodeficiency Cell virus of macaques (SIVmac) [12,13]. In a further study, virus

load was decreased by up to 1.7 log10 RNA copies/mL in

SIVmac-infected rhesus and cynomolgus monkeys after approval by the US Food and Drug Administration (FDA) for approximately 10 days of treatment with 3 mg/kg TNX-355 the treatment of HIV infection [7]. The success of this agent twice weekly [14]. Sustained binding of TNX-355 to has recently prompted a more aggressive goal of achieving CD4+ cells was observed without absolute decreases in undetectable viral load in triple-class experienced patients the CD4+ cell counts, and without immunosuppression, [8], and underscores the potential benefits of drugs that act in uninfected rhesus monkeys and in monkeys infected

on different steps in the viral life cycle. with SIVmac [9,14]. Following the example set by enfuvirtide, a number of entry inhibitors are now in the pipeline, offering the potential Phase I clinical trials show good tolerability and of further diversifying treatment options. In addition, several significant viral load reductions of these drugs have improved pharmacokinetic properties In the Phase IA protocol, Hu5A8.01, five sequential cohorts and/or have been designed as sustained-release formulations of six treatment-experienced HIV-positive patients received that potentially allow once-weekly, or even once-monthly, single-dose intravenous infusions of 0.3, 1, 3, 10, and dosing. Current investigational agents in development that 25 mg/kg TNX-355 [15]. Safety, pharmacokinetics, and require parenteral administration include TNX-355, which is a antiviral activity were assessed. When administered as a monoclonal antibody (MAb) that binds to CD4 to prevent single dose, TNX-355 was well tolerated. No serious adverse post-attachment conformational changes required for entry; events were reported in the study. In this monotherapy PRO-140 and HGS004, both MAbs that inhibit HIV setting, TNX-355 demonstrated antiviral activity as binding to the CCR5 coreceptor; and TRI-999 and TRI-1144, measured by reductions in viral load: median nadir

two second-generation fusion inhibitors that, like enfuvirtide, reductions of ≥1 log10 RNA copies/mL were seen following interfere with virus–cell fusion. This article discusses the treatment with the 10 mg/kg and 25 mg/kg doses. Clinically

present status of these novel compounds as they move from significant viral load reductions (>0.5 log10 RNA copies/mL) proof of concept to pivotal Phase III clinical trials. were maintained for 2–3 weeks following single infusions of the 10 mg/kg and 25 mg/kg doses. Mean CD4+ cell counts TNX-355 initially increased, then returned towards baseline over the TNX-355 is a humanized MAb in Phase II clinical trials. It was course of the follow-up. reconstructed from a murine progenitor antibody, 5A8, into a In protocol TNX-355.02, a Phase IB trial, multiple doses humanized immunoglobulin G subtype 4 (IgG4) molecule of TNX-355 were studied [16]. Nineteen patients were (Hu5A8 or TNX-355) that retains identical functional randomized to receive a 10 mg/kg initial dose, followed by specificity and activity of the predecessor antibody [9,10]. nine doses of 10 mg/kg weekly or five doses of 6 mg/kg TNX-355 blocks entry of HIV into target cells by binding to a twice weekly. Additionally, three non-randomized patients

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 49 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 50

J LALEZARI, S LEWIS, GM SUBRAMANIAN, R KRAWIEC, AND MK DELMEDICO

Figure 2. Mean reduction in viral load from baseline to 24 weeks (primary endpoint) in patients treated with the TNX-355 anti-CD4 antibody at 10 or 15 mg/kg plus optimized background regimen.

15 mg/kg Placebo 10 mg/kg Placebo 0 0

–0.20 –0.20

–0.5 0.75 –0.5 copies/mL) copies/mL)

10 10 0.96

–1.0 –0.95 –1.0

–1.16 at week 24 (log at week 24 (log

Mean HIV RNA reduction from baseline from Mean HIV RNA reduction –1.5 p<0.003 baseline from Mean HIV RNA reduction –1.5 p<0.001

Last observation carried forward analysis

were enrolled and received five doses of 25 mg/kg twice • TNX-355 15 mg/kg: –0.95 log10 (p=0.003 compared weekly. In this study, TNX-355 was administered to patients with placebo).

who were on a stable, failing regimen of HAART, or was • TNX-355 10 mg/kg: –1.16 log10 (p<0.001 compared given as a single agent. Again, TNX-355 was well tolerated. with placebo).

Four serious adverse events were reported in the study; • Placebo: –0.20 log10. however, none were related to the study drug. As essential monotherapy, TNX-355 demonstrated clinically significant Treatment with the 10 mg/kg and the 15 mg/kg

median viral load reductions (approximately 1.0 log10) at regimens in combination with OBR continued to produce nadir measurements. statistically significantly greater mean viral load reductions than the placebo-plus-OBR treatment at 48 weeks (Fig. 3). Phase II clinical trials confirm significant viral load CD4+ cell-count increases were significantly greater in the reductions and show increased CD4+ cell counts TNX-355-treated patients than in the placebo group (Fig. 4) A Phase II protocol, TNX-355.03 was consequently [18]. Further studies of TNX-355 in treatment-experienced conducted, to assess the safety and antiviral activity of TNX- patients are planned. 355 in the setting of combination therapy [17]. This 48-week randomized, double-blind, placebo-controlled, three-arm CCR5 antagonists study enrolled 82 treatment-experienced HIV-positive Rationale for development of anti-CCR5 antibodies patients with viral loads ≥10 000 RNA copies/mL and for the treatment of HIV-1 infection ≥50 CD4+ cells/mm3 at baseline. The investigators selected a CCR5 is the chemokine receptor for the ligands macrophage resistance test-guided, optimized background regimen (OBR) inflammatory protein-1 (MIP-1) α and β, RANTES (regulated for each patient. Patients were subsequently randomized to on activation normal T cell expressed and secreted), and receive 15 mg/kg TNX-355 twice weekly, 10 mg/kg monocyte chemotactic protein-2. CCR5 is also the primary TNX-355 weekly for 9 weeks followed by 10 mg/kg twice coreceptor for HIV-1 transmission and replication, from the weekly, or placebo. The primary endpoint of the study was early stages of disease through progression to AIDS. In all the mean change in viral load at week 24 between either of stages of HIV-1 infection, R5 and dual-tropic (R5/X4) viruses the TNX-355 treatment arms and placebo. comprise the majority of viral strains detected, while At 24 weeks, the mean changes in viral load for the X4-exclusive viruses constitute a minority of cases. Naturally intent-to-treat population using last observation carried occurring host defects in CCR5 expression have forward analysis for non-completers were (Fig. 2): demonstrated the importance of this receptor in HIV-1

50 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 51

PARENTERAL INHIBITORS OF HIV ENTRY IN CURRENT DEVELOPMENT

Figure 3. Mean reduction in viral load from baseline to 48 weeks in patients treated with TNX-355.

15 mg/kg Placebo 10 mg/kg Placebo 0 0

–0.14 –0.14

0.57 –0.5 –0.5

copies/mL) copies/mL) 0.82 10 10 –0.71

–1.0 –1.0 –0.96 at week 48 (log at week 48 (log

Mean HIV RNA reduction from baseline from Mean HIV RNA reduction –1.5 p<0.010 baseline from Mean HIV RNA reduction –1.5 p<0.001

Last observation carried forward analysis All arms included an optimzed background regimen

infection, as it has been observed that individuals with a candidate based on potency in HIV-1 infectivity assays, and homozygous deletion (CCR5∆32) show resistance to the was converted to an IgG4 backbone. virus [19]. In addition, CCR5 is a potentially safe target in The ability of HGS004 to block infection with R5 HIV-1 HIV-1 treatment, as people lacking CCR5 expression appear was evaluated in a series of in vitro assays. HGS004 was healthy. Because MIP-1β and RANTES are proinflammatory, observed to potently inhibit: another potentially interesting role for this class of agents is their use in altering of the immune activation that has been • CCR5-dependent entry of 30 HIV-1 viruses

shown to promote disease progression in HIV-1 [20]. representative of clades A–G (median IC50 [the concentration required for 50% inhibition HGS004 of viral replication] 6 nM). The development of HGS004 was initiated with the • Enfuvirtide-resistant HIV-1.

generation of fully human MAbs from Xenomice (animals • Ligand binding (IC50 0.41 nM). genetically engineered to produce completely human MAbs; • Cell–cell fusion (Fig. 5). Abgenix, Thousand Oaks, CA, USA) immunized with mouse • Viral replication. cells that expressed CCR5. Binding specificity, affinity, antagonism of ligand binding, signaling, and antibody In addition, HGS004 did not induce signaling or mediate effector functions were determined using human peripheral any cellular toxicity upon binding to CCR5, nor did it blood mononuclear cells (PBMCs) or CCR5-expressing induce antibody-dependent, cell-mediated cytotoxicity, or cell lines. complement-dependent cytotoxicity in human cells [21]. To assess the risk of resistance development, two primary Preclinical data show potent inhibition of R5 viruses R5 virus isolates (RM and JC) were passaged in vitro for and synergy with other antiretrovirals >24 weeks in the presence of HGS004. When compared The antiretroviral activity of HGS004 was evaluated in HIV-1 with control cultures (grown without the antibody) that had envelope-dependent cell–cell fusion assays and with the been through a similar number of passages, there was no

PhenoSense HIV entry assay (Monogram, Inc., South San significant difference in IC50 values between the viruses, Francisco, CA, USA). Novel CCR5 MAbs were identified that suggesting a low likelihood of causing development specifically bind to CCR5, fail to induce intracellular of resistance [22]. signaling, block HIV entry, and inhibit HIV envelope Furthermore, in vitro combination studies with current dependent cell-cell fusion. HGS004 was selected as the lead antiretroviral agents representing each of the drug classes

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 51 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 52

J LALEZARI, S LEWIS, GM SUBRAMANIAN, R KRAWIEC, AND MK DELMEDICO

Figure 4. Mean absolute CD4+ cell counts at baseline and Figure 5. The anti-CCR5 monoclonal antibody, HGS004, 48 weeks in patients treated with the TNX-355 anti-CD4 blocks HIV-1 envelope-mediated cell fusion. antibody at 10 or 15 mg/kg plus OBR.

120 Baseline Week 48 347 350 100

299 300 274 80 ) at week 48 3 246 247 250 223 60

200 Cell fusion (%) 40 150 cell count (cells/mm + 20 Enfuvirtide 100 HGS004 IgG control MAb 0 50 0 0.01 0.1 1 10 100 1000 10000 Concentration (nM) Mean absolute CD4 0 15 mg/kg 10 mg/kg Placebo (p=0.016) (p=0.031) (p=NS) All arms included an OBR placebo was administered as an intravenous infusion. Blood samples were obtained pretreatment, post-infusion, and on OBR: optimized background regimen. days 1, 2, 4, 7, 14, 21, 28, 42, and 56. Adverse events, laboratory assessments (clinical chemistry and hematology), NRTIs (zidovudine and lamivudine), NNRTIs (efavirenz), PIs drug levels, CD4+ cell counts, viral load, receptor occupancy,

(indinavir), and fusion inhibitors (enfuvirtide), were and change in tropism or isolate sensitivity (IC50) were performed with the RM and JC isolates [22]. The results assessed. Overall, HGS004 was well tolerated and no dose- demonstrated that HGS004 acts synergistically with all limiting toxicities were observed. No grade 3–4 adverse currently approved classes of antiretroviral agents. events were noted, although two treatment-related events of moderate severity, involving infusion-related Phase 1 clinical trial demonstrates safety allergic reactions at the 2 mg/kg dose, necessitated and efficacy of HGS004 single-dose a protocol change to include premedication with oral intravenous injections diphenhydramine. The study further demonstrated: HGS004 is currently being evaluated in a Phase 1 clinical trial. The study is a randomized, placebo-controlled, dose- • Short-term increases in CD4+ and CD8+ cell counts. escalation, multi-center investigation in patients who are • Non-linear pharmacokinetics at doses of 0.4–40 mg/kg

infected with HIV-1 and not receiving concurrent antiretroviral (Cmax increased proportionally with dose while area therapy. The primary objective is to evaluate the safety and under the curve increased more than proportionally tolerability of escalating doses of a single intravenous infusion with the dose). of HGS004. The secondary objectives are to determine the • High levels of receptor occupancy of >80% through pharmacokinetics of HGS004, and to assess its effect on 28 days in the 8, 20, and 40 mg/kg-dose cohorts. + + plasma viral load, CD4 , and CD8 T cell counts over time. • Day 14 HIV-1 RNA reductions of >1 log10 in 14 of A total of 63 HIV-1-positive patients (R5 virus, HIV-1 26 (54%) subjects in the 8, 20, and 40 mg/kg-dose RNA >5000 copies/mL, CD4+ cell counts >250/mm3) were cohorts (Fig 6).

randomized to one of five dose cohorts receiving 0.4, 2, 8, • Day 28 HIV-1 RNA reductions of >1 log10 in four of 20, or 40 mg/kg HGS004 [23]. A single dose of MAb or 10 subjects in the 40 mg/kg cohort (Fig 6).

52 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 53

PARENTERAL INHIBITORS OF HIV ENTRY IN CURRENT DEVELOPMENT

Figure 6. Antiviral activity of the anti-CCR5 monoclonal antibody, HGS004, as measured by reduction in HIV-1 RNA, in a Phase I study.

0.0 copies/mL) 10

–0.5

Placebo 0.4 mg/kg 2 mg/kg 8 mg/kg –1.0 20 mg/kg 40 mg/kg Mean HIV-1 RNA reduction (log RNA reduction Mean HIV-1 0 4 7 14 21 28 42 56 Days

Preliminary data on post-treatment susceptibility of viral settings. In an evaluation using a panel of six primary HIV-1 isolates showed a shift in tropism to dual-mixed populations isolates selected for their genotypic and geographic diversity, in five subjects. the antibody was compared with RANTES [24]. Like RANTES, murine PRO 140 demonstrated potent and HGS101 subtype-independent inhibition of each of the viruses tested, µ Further research has subsequently identified HGS101, an with IC90 values of 2–3 g/mL. Murine PRO 140 inhibited alternative anti-CCR5 MAb candidate. In vitro data suggest HIV-1 replication in both isolated PBMCs and macrophage that HGS101 is 5.5-fold more potent against the clinical cell cultures, whereas RANTES was a poor inhibitor of isolates from the Phase I trials. Other attributes of HGS101 replication in macrophages. Thus, unlike the natural ligand are similar to those of HGS004, including favorable for CCR5, murine PRO 140 protects both T cells and pharmacokinetics, strong in vitro evidence of anti-viral activity macrophages from HIV-1 infection in vitro. Tests to assess that is additive or synergistic in combination with approved the CCR5 antagonistic effect of PRO 140 in vitro therapeutic agents, and a low likelihood of causing demonstrated effective suppression of viral replication at development of resistance based on long-term in vitro culture. concentrations that have little or no effect on the natural activity of CCR5. PRO 140 The antiviral activity of both murine and humanized PRO PRO 140 is a humanized IgG4κ MAb against CCR5, 140 were subsequently examined in vivo, using hu-PBL- developed by Progenics Pharmaceuticals, Inc (Tarrytown, SCID mice (a human lymphocyte-engrafted, severe NY, USA) [24,25]. It recognises a multidomain epitope on combined immunodeficient model) of chronic HIV-1 CCR5 and binds human PBMCs in vitro. The antiviral infection and therapy [26]. Viral loads were reduced to activities of humanized and murine PRO 140 have been undetectable levels (<400 copies/mL) in all animals (n=5) shown to be essentially indistinguishable, and the two mAbs treated with 1 mg-daily doses of humanized PRO 140 for

are equally potent in blocking HIV-1 envelope-mediated 8 days. The mean viral load reduction exceeded 1.5 log10 cell–cell fusion [24]. and was statistically significant (p<0.01), while the virus levels remained high in control animals. Murine PRO 140 Preclinical studies show subtype-independent was also effective in controlling viral replication in this inhibition of R5 viruses, without CCR5 antagonism model. PRO 140-treated animals had, on average, higher PRO 140 has demonstrated broad-spectrum activity against levels of human CD4+ cells than untreated animals, possibly primary R5 strains of HIV-1 in a variety of preclinical reflecting diminished levels of virus replication and cell killing

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 53 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 54

J LALEZARI, S LEWIS, GM SUBRAMANIAN, R KRAWIEC, AND MK DELMEDICO

in the presence of PRO 140. These findings indicate that the explore the safety and activity of PRO 140 in patients with antiviral effects of PRO 140 were related to CCR5 blockade HIV-1 disease. rather than to depletion of target cells. To investigate a potential synergistic activity against HIV, Next-generation fusion inhibitors: the effect of combinations of PRO 140 and the small- TRI-999 and TRI-1144 molecule CCR5 antagonists, SHC-C, vicriviroc, and TAK-779, Clinical trials have verified the effectiveness and safety of the were assessed in a fluorescence resonance energy transfer fusion inhibitor enfuvirtide. It is administered by twice-daily assay (a well-established model of HIV entry) [27]. The subcutaneous injection, with injection-site reactions being a antiviral activity consistently increased by at least four-fold, as common, but seldom treatment-limiting, adverse effect [29]. measured by the reduced drug concentration necessary to With the aim to improve the convenience of peptide-based block 50% of HIV membrane fusion. Synergy was observed antiretroviral therapies, two next-generation fusion inhibitors only with the addition of PR0 140 and not with combinations (NGFIs), TRI-999 and TRI-1144, are therefore currently of two small-molecule CCR5 antagonists alone. being developed, in another collaboration between the In additional laboratory studies, humanized or murine manufacturers of enfuvirtide, Trimeris-Roche (Trimeris, Inc., forms of PRO 140 were tested for activity against viruses Morrisville, NC, and Roche, Nutley, NJ, USA). selected for resistance to small-molecule CCR5 inhibitors To improve the potency and durability of the new in vitro. Viruses were cultured in the presence of small- agents, two peptides were derived from an optimized gp41 molecule CCR5 inhibitors until drug-resistant isolates heptad repeat (HR) 2 region, partially overlapping the emerged. The viruses acquired mutations that enabled them enfuvirtide amino acid sequence (Fig. 7) [30]. The natural to use CCR5 in the presence of the small-molecule sequences were modified using independent strategies, inhibitors; however, these mutations did not affect viral resulting in two compounds from distinct chemical series susceptibility to PRO 140. with promising potency, genetic barrier to resistance, and pharmacokinetic properties. Phase I clinical trial suggests infrequent dosing requirements Structural modifications to improve To assess PRO 140 in humans, a single-dose, double-blind, pharmacokinetics placebo-controlled dose-escalation study was conducted in TRI-999 is a 36-amino acid peptide containing a healthy volunteers to evaluate the safety, tolerability, polyethylene glycol linker-fatty acid moiety attached to an pharmacokinetics, and pharmacodynamics following internal lysine side chain. The peptide sequence overlaps the intravenous injection [28]. Four dose cohorts of five subjects sequence of enfuvirtide by 24 residues and contains each (four receiving drug and one placebo) were administered 12 additional HR2 residues N-terminal to the sequence from a single dose of 0.1, 0.5, 2, or 5 mg/kg PRO 140. No dose- which enfuvirtide is derived. These N-terminal residues make limiting toxicities or patterns of drug-related toxicities were contact with the deep pocket on the HR1 viral target

observed. Preliminary pharmacokinetic analyses suggest a [30–33]. The C18-fatty acid increases non-covalent binding serum half-life of 2–3 weeks, which could support dosing as to serum proteins; this protects the peptide from renal infrequently as monthly or bimonthly. Preliminary elimination and thus confers improved pharmacokinetic pharmacodynamic analyses demonstrate that a single 2.0 properties. The precise fatty acid position on the peptide and or 5.0 mg/kg dose of PRO 140 significantly coated the length and type of linker between the peptide and fatty CCR5-positive blood lymphocytes for 2 and 8 weeks, acid were chosen to minimize loss of potency when serum respectively. This extended period of coating was observed proteins are present in the HIV assay. without depletion of CCR5-positive lymphocytes from the TRI-1144 is a 38-amino acid peptide engineered to adopt circulation. Since coating of CCR5 is the mechanism by which a helical conformation in solution. The starting peptide PRO 140 inhibits HIV replication in vitro, these data support sequence is similar to that of TRI-999, and the wild-type the view that PRO 140 has the potential to inhibit viral residues that contact the viral target have been either retained replication in HIV-infected patients. or conservatively changed. Residues that are not involved in PRO 140 has been given a “fast track” designation by HR1 binding have been replaced with helix-promoting motifs. the FDA for the treatment of HIV infection. The Fast Track Development Program facilitates development and Preclinical data demonstrate superiority expedites regulatory review of drugs intended to address an over enfuvirtide unmet medical need for serious or life-threatening To assess antiviral activity, the peptides were tested against conditions. An ongoing Phase IB study is the first to a panel of 12 fusion inhibitor-sensitive clinical isolates.

54 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 55

PARENTERAL INHIBITORS OF HIV ENTRY IN CURRENT DEVELOPMENT

Figure 7. Sequences of the new-generation fusion inhibitor peptides, TRI-999 and TRI-1144. Both peptides are derived from an optimized HR2 sequence that partially overlaps the sequence of enfuvirtide.

HR2 Domain from gp41:

…MTWMEWDREINNYTSLIHSLIEESQNQQEKNEQELLELDKWASLWNWF…

Enfuvirtide sequence

Optimized HR2 sequence

NGFI Peptides:

LTWQEWDREINNYTSLIHSLIEESQNQQEKNEQELL TRI-999 O O || || C H -C-NH-CH -CH -O-CH -CH -O-CH -C } 17 35 2 2 2 2 2

TTWEAWDRAIAEYAARIEALLRALQEQQEKNEAALREL TRI-1144

HR: heptad repeat.

TRI-999 had a geometric mean IC50 of 1 nM – seven times and TRI-999, respectively, compared with enfuvirtide better than enfuvirtide, while TRI-1144 was similar to (40 mL/kg/h), following intravenous administration [30]. enfuvirtide at 7 nM [30]. Furthermore, the modified These improvements in pharmacokinetic parameters peptides were found to have a markedly narrower range allow the possibility of once-weekly dosing when coupled

of IC50 values than enfuvirtide. TRI-999 and TRI-1144 with a sustained-release formulation. Early experiments in were subsequently tested against a panel of eight virus rodents with TRI-1144 demonstrated steady sustained isolates that were resistant to enfuvirtide and other release of peptide over 1 week following a single peptide fusion inhibitors. The geometric means of the subcutaneous injection [35].

IC50 values demonstrated >250-fold and >150-fold greater Recently, the TRI-1144 peptide was advanced as the lead effectiveness of TRI-999 and TRI-1144, respectively, preclinical candidate for continued development due to its relative to enfuvirtide [34]. overall suitability as a candidate for HIV-1 therapy. These improvements in potency suggest higher genetic barriers to generation of resistance than that for enfuvirtide. Conclusion To verify this, passaging experiments were performed [30]. As we enter the third decade of HIV therapy, the success of In vitro selections with three different viruses in the presence antiretroviral drug development has been remarkable. of increasing concentrations of the peptides for >70 days Refinements in formulations and co-formulations along with showed only one- to eight-fold reductions in drug ritonavir boosting of PIs have provided many patients with the susceptibility. Similar experiments with enfuvirtide generated opportunity to treat HIV as a chronic manageable illness with viruses with >10-fold losses in susceptibility, revealing a the convenience of once-daily oral dosing. Much-anticipated comparatively superior durability of TRI-999 and TRI-1144 data on the small molecule CCR5 antagonists and integrase in vitro. inhibitors will likely translate into further options to control Furthermore, subsequent pharmacokinetic experiments viral replication with convenient oral regimens. in cynomolgus monkeys demonstrated four- to six-fold Against this backdrop, it is fair to ask where the slower clearance rates (11 and 7 mL/kg/h) for TRI-1144 parenteral agents currently under investigation for treatment

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 55 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 56

J LALEZARI, S LEWIS, GM SUBRAMANIAN, R KRAWIEC, AND MK DELMEDICO

14. Non-immunosuppressive anti-CD4 monoclonal antibody (5A8) for treatment of HIV-1 of HIV might fit. For some (TNX-355), the advent of an infection. Tanox, Inc. Pre-IND Briefing Document. May 10 2000. agent with a completely novel mechanism of action could 15. Kuritzkes DR, Jacobson J, Powderly WG et al. Antiretroviral activity of the anti-CD4 monoclonal antibody TNX-355 in patients infected with HIV type 1. J Infect Dis have an important role in salvage therapy irrespective of the 2004;189:286-91. need for intravenous infusion. For others (Tri-1144, the 16. Jacobson JM, Kuritzkes DR, Godofsky E et al. Phase Ib study of the anti-CD4 monoclonal antibody TNX-355 in HIV-1-infected subjects: safety and antiretroviral activity of multiple CCR5 MAbs), the trade-off might be to eventually forego doses. The 11th Conference on Retroviruses and Opportunistic Infections, San Francisco, some, or even all, daily therapy in favor of weekly CA, USA, February 8–11, 2004. Abstract 536. 17. Norris D, Morales J, Gathe J et al. TNX-355 in combination with optimized background subcutaneous dosing and/or monthly intravenous infusions. regimen (OBR) exhibits greater antiviral activity than OBR alone in HIV-treatment One thing is for certain: the rapidly expanding antiretroviral experienced patients. The 45th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC, USA, December 16–29, 2005. Abstract LB2-26. arsenal, including the parenterally administered entry 18. Norris D, Morales J, Gathe J et al. TNX-355, in combination with optimized background inhibitors in development, will offer clinicians and patients regimen (OBR), achieves statistically significant viral load reduction and CD4 cell count increase when compared with OBR alone in phase 2 study at 48 weeks. The 16th further options in constructing successful regimens to International AIDS Conference, Toronto, ON, USA, August 13–18, 2006. control HIV. Abstract THLB0218. 19. Liu R, Paxton WA, Choe S et al. Homozygous defect in HIV-1 coreceptor accounts for resistance of some multiply-exposed individuals to HIV-1 infection. Cell 1996;86:367–77. 20. Deeks SG, Kitchen CM, Liu L et al. Immune activation set point during early HIV infection Disclosures + predicts subsequent CD4 T-cell changes independent of viral load. Blood Dr Lalezari is Co-Editor-in-Chief of The Journal of Viral Entry, 2004;104:942–7. and receives research support from Human Genome Sciences, 21. Roschke V, et al. Characterization of novel human monoclonal antibodies that specifically antagonize CCR5 and block HIV-1 entry. 44th Annual Interscience Conference on Inc., Roche, and Trimeris Inc. Dr Lewis is an employee of Tanox, Inc., Antimicrobial Agents and Chemotherapy, 2004. Abstract 2871. Dr Subramanian is an employee of Human Genome Sciences, Inc., 22. Giguel F, Beebe L, Migone T. The anti-CCR5 mAb004 inhibits replication synergistically in combination with other antiretroviral agents but does not select for resistance during in Dr Krawiec is an employee of Progenics Pharmaceuticals, Inc., vitro passage. The 13th Conference on Retroviruses and Opportunistic Infections. February, 2006. Abstract 505 . and Dr Delmedico is an employee of Trimeris, Inc. 23. Lalezari, et al. A Phase 1, randomized, dose-escalation, placebo-controlled study of a fully human monoclonal antibody against CCR5 (HGS004) in patients with CCR5 tropic HIV-1 infection. The 46th Interscience Conference on Antimicrobial Agents and Chemotherapy, References September 27–30, San Francisco, CA, USA. Abstract H-1668. 1. Simon V, Vanderhoeven J, Hurley A et al. Evolving patterns of HIV-1 resistance to 24. Trkola A, Ketas TJ, Nagashima KA et al. Potent, broad-spectrum inhibition of human antiretroviral agents in newly infected individuals. AIDS 2002;16:1511–9. immunodeficiency virus type 1 by the CCR5 monoclonal antibody PRO 140. J Virol 2. Holodniy M, Charlebois ED, Bangsberg DR et al. Prevalence of antiretroviral drug 2001;75:579–88. resistance in the HIV-1-infected urban indigent population in San Francisco: a 25. Olson WC, Tsurushita N, Hinton PR et al. Potent, broad-spectrum inhibition of HIV-1 entry representative study. Int J STD AIDS 2004;15:543–51. with the humanized monoclonal antibody PRO 140. The 14th International AIDS 3. Laakso MM, Doms RW. The molecular basis of HIV entry and its inhibition. Conference, Barcelona, Spain, July 7–12, 2002. Abstract TuPeA4363. J Viral Entry 2006:2:4–12. 26. Franti M, Ramos L, Maloveste S et al. In vivo control of HIV-1 replication with PRO 140, 4. Briz V, Poveda E, Soriano V. HIV entry inhibitors: mechanisms of action and resistance a humanized monoclonal antibody to CCR5. The 15th International AIDS Conference, pathways. J Antimicrob Chemother 2006;57:619–27. Bangkok, Thailand, July 11–16, 2004. Abstract WeOrA1230. 5. Tremblay C. Effects of HIV-1 entry inhibitors in combination. Curr Pharm Des 27. Murga J, Olson W, Pevear D et al. Antiviral synergy between the CCR5 mAb PRO 140 and 2004;10:1861–5. small-molecule CCR5 antagonists. The 3rd IAS Conference on HIV Pathogenesis and 6. Kilby JM, Hopkins S, Venetta TM et al. Potent suppression of HIV-1 replication in humans Treatment, Rio de Janeiro, Brazil, July 24–27, 2005. Abstract TuOa0206. by T-20, a peptide inhibitor of gp41-mediated virus entry. Nat Med 1998;4:1302–7. 28. Olson W, Doshan H, Zhan C et al. First-in-humans trial of PRO 140, a humanized CCR5 7. FDA approves first drug in new class of HIV treatments for HIV infected adults and monoclonal antibody for HIV-1 therapy. The 3rd IAS Conference on HIV Pathogenesis and children with advanced disease. US Food and Drug Administration FDA News, March 13, Treatment, Rio de Janeiro, Brazil, July 24–27, 2005. Abstract WePe6.2C04. 2003, www.fda.gov/bbs/topics/NEWS/2003/NEW00879.html 29. Rice C, Wilantewicz H. Fuzeon (enfuvirtide): efficacy, safety, patient acceptance, and 8. Hammer SM, Saag MS, Schechter MS et al. Treatment for adult HIV infection, 2006 strategies for managing injection-site reactions. AIDS Read 2006;16:470–4; 490. recommendations of the International AIDS Society – USA Panel. JAMA 2006;296: 30. Delmedico MK, Bray BL, Cammack N et al. Next generation HIV peptide fusion inhibitor 827–43. candidates achieve potent, durable suppression of virus replication in vitro and improved 9. Reimann KA, Lin W, Bixler S et al. A humanized form of a CD4-specific monoclonal pharmacokinetic properties. Conference on Retroviruses and Opportunistic Infections, antibody exhibits decreased antigenicity and prolonged plasma half-life in rhesus monkeys Denver, CO, USA, February 5–9, 2006. Abstract 48. while retaining its unique biological and antiviral properties. AIDS Res Hum Retroviruses 1997;13:933–43. 31. Tan K, Liu J-H, Wang J-H et al. Atomic structure of a thermostable subdomain of HIV-1 gp41. PNAS 1997;94:12303–8. 10. Burkly LC, Olson D, Shapiro R et al. Inhibition of HIV infection by a novel CD4 domain 2- specific monoclonal antibody. J Immunol 1992;149:1779–87. 32. Chan DC, Fass D, Berger JM et al. Core structure of gp41 from the HIV envelope glycoprotein. Cell 1997;89:263–73. 11. Godofsky E, Zhang X, Sorenson M et al. In vitro antiretroviral activity of the humanized anti-CD4 monoclonal antibody, TNX-355, against CCR5, CXCR4, and dual-tropic isolates 33. Weissenhorn W, Dessen A, Harrison SC et al. Atomic structure of the ectodomain from and synergy with enfuvirtide. The 45th Interscience Conference on Antimicrobial Agents HIV-1 gp41. Nature 1997;387:426–30. and Chemotherapy, Washington, DC, USA, December 16–29, 2005. Abstract LB26. 34. Davison DK, Medinas RJ, Mosier SM et al. New fusion inhibitor peptides, TRI-999 and 12. Reimann KA, Cate RL, Wu Y et al. In vivo administration of CD4-specific monoclonal TRI-1144, are potent inhibitors of enfuvirtide and T-1249 resistant isolates. The 16th antibody: effect on provirus load in rhesus monkeys chronically infected with simian International AIDS Conference, Toronto, ON, Canada, August 13–18, 2006. immunodeficiency virus of macaques. AIDS Res Hum Retroviruses 1995;11:517–25. Abstract THPE0021. 13. Reimann KA, Burkly LC, Burrus B et al. In vivo administration to rhesus monkeys of a CD4- 35. Delmedico MK, Bray BL, Cammack N et al. Next generation fusion inhibitor candidates specific monoclonal antibody capable of blocking AIDS virus replication. AIDS Res Hum TRI-1144 and TRI-999 have improved pharmacokinetics and demonstrate sustained- Retroviruses 1993;9:199–207. release. XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006.

56 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 LEADING ARTICLE n 57 ded forms 0 , a process required for 2 precursor, the native precursor, cleaved 0 subunit is required for fusion and HA 2 1 . In the endoplasmic reticulum, HA 0 ool of Medicine, subunits in the membrane-distal head domains of 1 Vol 2 No 2 2006 Vol s an alternative means for controlling influenza – an d the possibility that they may spread in the human al, and social problems. Vaccines provide a degree of ured, and distributed. In such situations, antiviral drugs gic planning for future pandemics. Two classes of anti- young and the elderly. In addition, there is the ominous NTRY (Fig. 2). High-resolution X-ray crystal available for many structures of the are 16 antigenic subtypes now of HA, and several structures also exist for HA complexes with receptor analogs [1–6]. In addition, for the H3 subtype, the structure has been determined for the HA this way, NA plays dissemination a of critical newly role structural assembled in and functional information virus the available on HA, particles. release NA, and and The M2 make them ideal targets for antiviral intervention, review. and these topics form the focus of this Structure of HA HA is a type I glycoprotein that is synthesized as a polypeptide precursor of approximately referred 550 to as HA amino acid residues, and viral surface sialic acid to remove functional receptors. I receptors. functional and viral surface sialic acid to remove protein, protein, and the low-pH conformation assumes during that the process of the membrane fusion molecule [6–8]. Thus, this article will concentrate equally applicable to the other subtypes. on concepts involved are the H3 subtype, but the homotrimers that are stabilized by non-covalent interactions. Each monomer of the trimer is subsequently cleaved into the disulfide-linked subunits HA the molecule, while the HA virus infectivity. The receptor-binding activity The of receptor-binding HA virus is infectivity. provi by the HA matches with the they circulating are viruses; least however, have a significant role to play in modulating the severity of ale, M2 inhibitors and neuraminidase inhibitors. While both E ach also suffers from drawbacks. Hence, there is an ongoing IRAL al V e, USA. (2):57–66. 2 OURNAL OF J HE 2006: T J Entry Viral Address for correspondence: David A Steinhauer, Department of David A Steinhauer, for correspondence: Address Emory University School of Medicin and Immunology, Microbiology Email: [email protected] Rollins Research Center, 1510 Clifton Road, Atlanta, GA 30322, 1510 Clifton Road, Atlanta, GA 30322, Center, Rollins Research matrix (M1) protein from the nucleocapsids, a process that frees them to enter the nucleus, where viral transcription and replication take place. At the end cycle, of the sialidase activity the provided by replication NA cleaves host cell Influenza A viruses are enveloped viruses with segmented single-stranded RNA genomes of negative hemagglutinin (HA) polarity. and neuraminidase The (NA) glycoproteins, along with the M2 protein, are incorporated into the viral membrane, and each plays a significant role in the virus life cycle (Fig. 1). Influenza A viruses attach to host interactions cells between HA via and cell surface glycoproteins and glycolipids that contain receptor-mediated terminal sialic endocytosis, acid. endosomes Following by cellular proton the pumps leads to two events that acidification are critical for virus of entry. One irreversible conformational is changes in the HA that triggering are required of to mediate fusion of the viral and endosomal membranes. The other involves the proton channel activity of the viral M2 protein, which, in response to the proton influx endosomes, induces a concurrent acidification into of the interior of virus particles. This results in the dissociation of the vir Influenza A viruses continue to cause serious medical, commerci Zhu-Nan Li and David A Steinhauer Zhu-Nan Li and University Sch Emory and Immunology, Department of Microbiology The Mechanism and Inhibition of and Inhibition Mechanism The Entry Virus Influenza USA Atlanta, GA, Center, Rollins Research protection when the strains used for their production are good effective in the populations that need them the most – the very threat posed by the emergence of antigenically novel viruses an population before effective vaccines can be developed, manufact may be useful. For these reasons, antivirals against influenza disease during the normal flu seasons, as well as in the influenza drugs are strate currently employed on a relatively large sc clearly are beneficial when administered in a timely fashion, e effort to improve existing and design new antiviral compounds a omnipresent moving target. R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 57 Page 16:39 2/11/06 R7694_2_REM_VE_2_2_07.qxd R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 58

ZHU-NAN LI AND DAVID A STEINHAUER

Figure 1. The influenza virus life cycle, illustrating virus binding of sialic acid-containing receptors on the host cell, internalization, and the processes that occur inside the endosome. For simplicity, the details regarding the synthesis of the viral glycoproteins in the endoplasmic reticulum, trafficking through the Golgi, and transport to the cell surface are not represented. As endosomes are acidified, the virion interior is also acidified, owing to the proton channel function of the M2 protein. This releases the nucleocapsids, which transfer to the nucleus following fusion of the viral and endosomal membranes. The drug amantadine blocks M2 channel function, which inhibits the dissociation of ribonucleoproteins from the matrix (M1) protein and prevent nucleocapsids from entering the nucleus. The inset image portrays a model of the transmembrane portion of the M2 tetramer and how amantadine might block the influx of protons into the virus particle. It also shows the position of one of the four mutations (Serine 31) on the inside of the channel that are commonly associated with drug resistance. The figure also depicts assembly and budding of progeny virions. Inhibition of viral neuraminidase at this stage results in aggregation of virus particles at the infected cell surface and inhibition of virus dissemination.

Neuraminidase Hemagglutinin

Nucleoprotein and polymerases

Influenza virus

Viral envelope Homotetrameric M2 protein

Matrix Neuramindase cleaves sialic acid from receptors to allow viral release Cell membrane Budding virus

Receptors containing sialic acid Viral assembly

Envelope proteins Endosome H+ ions

Ribosome Amantadine blocks + the influx of H Translation of ions through the viral messenger M2-proton channel, RNA inhibiting uncoating and release of free Polymerase and ribonucleoproteins into nucleoprotein the cytoplasm.

Viral uncoating Amantadine Serine 31 Virion RNA packaging Cytoplasm Viral ribonucleoproteins Viral messenger RNA H+ H+ H+ Virion RNA H+ Viral envelope Transcription Homotetrameric M2 protein Nucleus Replication Complementary RNA

Reproduced with permission from [92].

58 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:39 Page 59

THE MECHANISM AND INHIBITION OF INFLUENZA VIRUS ENTRY

Receptor binding Figure 2. Alpha carbon tracing of an HA monomer (left)

The HA trimer extends approximately 135 Å from the and the HA trimer (right). The HA1 subunit is shown in

membrane to the tip of the trimer, and the receptor-binding blue and the HA2 subunit in grey. The locations of the site is a shallow pocket located at the distal end of each receptor binding site and the fusion peptide (light blue)

monomer. The site is composed of a small helix (the 190 at the N-terminal region of HA2 are indicated. helix) at the membrane-distal edge of the pocket, an extended chain (the 130 loop) at the front of the site, and the 220 loop, which constitutes the left side of the binding domain. Residues Y98, W153, H183, and Y195 form the base of the pocket. Several of the residues in the binding site are highly or relatively well conserved, suggesting their involvement in stabilizing the structural integrity of the region, receptor interaction, or both (Fig. 3). HA binds to sialic acid components at the ends of carbohydrate chains on glycoproteins or glycolipids of host cells [9,10]. Avian influenza A viruses mainly bind to receptors with terminal sialic acids attached to galactose via α2,3 linkages, whereas human influenza A viruses generally prefer receptors with α2,6-linked sialic acid [11,12]. The specificity of binding reflects, to a great degree, the preponderance of receptors at the site of replication in the host. The digestive tract of bird species are rich in α2,3- linked sialic acid receptors, while the human respiratory tract HA: hemagglutinin. contains receptors with α2,6-linked sialic acid [13]. Interestingly, most of the α2,6-linked sialic acid receptors in human airways reside on non-ciliated cells, while avian monovalent analogs such as sialyllactose with an α2,3 or receptors are predominantly expressed on the ciliated cells α2,6 linkage indicate only a very slight difference in binding, [14]. In addition, mucins in the human respiratory tract despite their specific HA preferences. By comparison, several primarily contain sialic acids with α2,3 linkages, thus viral attachment proteins have a much greater affinity for providing a natural barrier to avian viruses [15]. their specific host cell receptors. For example, the HIV type 1 Studies on viral and bacterial ‘receptor-destroying (HIV-1) envelope protein binds to CD4 with dissociation enzymes’ in the 1940s led Burnet and colleagues to constants in the micromolar range [20]. postulate that such enzymes might be used therapeutically As monovalent receptor–HA interactions are inherently to remove influenza receptors from the respiratory tract, or weak, approaches involving monovalent inhibitors include that competitive inhibitors of these receptors might provide the synthesis of sialic acid compounds with substituents a barrier to infection [16–18]. These are still good ideas. designed to interact more strongly than natural sialosides Several lines of reasoning suggest that inhibition of receptor [21]. For example, derivatives with substituents that interact binding offers an attractive option for the design of antiviral with hydrophobic regions and polar residues adjacent to the compounds. For example, an abundance of data exists on binding site result in a 10- to 100-fold increase in binding how HAs bind to various receptor analogs. Furthermore, the affinity relative to naturally occurring monovalent sialosides.

binding site is accessible on the viral surface, so drugs would A second sialic acid binding site at an HA1–HA2 interface not have permeate cellular or viral membranes. Ideally, such toward the bottom of the head domains on the HA structure inhibitors should have the following properties: has also been defined with crystallography using α2,3-linked sialyllactose and other sialic acid derivatives. Although these • Bind more tightly to HA than sialic acid. structures suggest that the site is not biologically relevant for • Be capable of forming multivalent HA–receptor contacts. influenza binding to receptors, sites such as this could be • Be resistant to the action of viral NAs. exploited to enhance the affinity of polyvalent compounds designed for antiviral activity [22]. However, the binding of HA to monovalent sialosides is Approaches with multivalent inhibitors have been not exceptionally strong, exhibiting dissociation constants explored extensively [23]. Numerous naturally occurring

(Kd) of around 2–4 mM [19]. Interestingly, the Kd values for proteins that are heavily glycosylated are known to inhibit

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 59 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 60

ZHU-NAN LI AND DAVID A STEINHAUER

Figure 3. Schematic diagram of the HA receptor binding site. The 190 helix, 130 loop, and 220 loop are indicated, as are the side chains of residues 98, 153, 183, and 195, which constitute the base of the pocket. Sialic acid is depicted in grey, and hydrogen bonds are shown as dashed lines. The hydrogen bonds involving sialic acid atoms are shown in black dashed lines and those between HA side chain atoms are shown in blue dashed lines.

Tyr 195 190 HELIX

Leu 194 Glu 190

Thr 155

His 183 HO N – O OH O N H H OH O N HO 9 Trp 153 8 7 CH3 Ser 228 O NH OH 6 5 OH H RO 3 4 Ser 227 O 2 0 220 LOOP O C Leu 226 – 135 O HO

Gly 225 Ser 136

Arg 224 Tyr 98 H N 130 LOOP 137

138

HA: hemagglutinin.

influenza viruses by interfering with receptor binding [21]. increase the affinity of binding and probably interact with For example, horse serum is rich in α2-macroglobulin, which more than one trimer on the viral surface [26,27]. functions as a potent inhibitor of many influenza strains that A number of sialyl glycopolymer compounds have also been preferentially recognize α2,6-linked receptors. This is due to generated and shown to inhibit binding [21], but these have its high content of α2,6-linked sialosides, a significant not been advanced to the stage of clinical availability. percentage of which contain the 4-O-acetyl derivative of sialic acid that is resistant to neuraminidase [24]. If grown in HA-mediated membrane fusion

the presence of horse serum, changes in the HA receptor The cleavage of HA0 into HA1 and HA2 is a critical binding region that alter receptor preference to those prerequisite for HA fusion activity and influenza virus containing an α2,3 linkage are selected [25]. Given that infectivity (Fig. 4) [28,29]. Highly pathogenic avian strains natural molecules such as α2-macroglobin have antiviral often have HA cleavage sites that contain the furin properties, it stands to reason that it might be possible recognition motif (R-X-K/R-R), whereas the non-pathogenic to synthesize low-molecular-weight compounds that avian strains and human strains usually have only a single appropriately present multiple α-sialosides for the purpose arginine residue at the cleavage site [30]. Because the of binding inhibition. Bivalent and tetravalent compounds polybasic furin sites are recognized and cleaved

60 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 61

THE MECHANISM AND INHIBITION OF INFLUENZA VIRUS ENTRY

Figure 4. A: Ribbon diagram of the conformational changes that accompany precursor cleavage. The HA1 residues shown

in grey and the HA2 residues shown in blue do not change. The residues of the HA0 loop structure (light blue, left panel)

relocate following cleavage at residue 329 (arrow) such that the N-terminus of HA2 is inserted into the trimer interior of

the native cleaved HA (right panel). B: Space-filling model of the region of the cleavage loop showing HA0 (left) and native HA (right). The cleavage loop structure is oriented toward the reader in the left panel and the residues that become the

C-terminus of HA1 (white circle) and N-terminus of HA2 (black circle) are shown. The cavity that becomes occupied by the

fusion peptide is located to the left of the loop in HA0. The right panel shows the locations of the HA1 C-terminus (white

circle) and HA2 N-terminus (black circle) after cleavage.

A B

22 22 C14–C137 323 N C 329 12 323 1 12 5

N N HA HA 0 C C uncleaved HA0 cleaved HA

Reproduced with permission from [7]. HA: hemagglutinin.

intracellularly by ubiquitously expressed proteases, viruses is extruded from its buried position in the interior of native with these motifs can cause systemic infections in the host, HA trimers, and two structural rearrangements take place in whereas replication of viruses that have HAs with single the HA stalk [8]. One is the refolding of the extended chain arginine residues is generally restricted to locations at which that links the long alpha helix with the shorter one in the appropriate extracellular proteases are present (e.g. the native HA hairpin loop structure. These residues become lumen of the respiratory tract in humans). helical and lead to an N-terminal extension of the coiled coil.

A conformational change involving HA1 residues As a result, the small helix is also recruited into the extended

323–328 and HA2 residues 1–12 accompanies proteolytic coiled coil and residues N-terminal to it are concurrently

cleavage of HA0 [7]. The liberated HA2 N-terminal domain is relocated. Therefore, the extruded fusion peptide is a highly conserved hydrophobic region, and due to its role in propelled towards the end of the newly formed coiled coil to membrane fusion, it has been designated as the fusion insert into the target membrane. The other conformational peptide. After cleavage, the fusion peptide relocates into an change involves residues 106–112 in the membrane- adjacent cavity and contacts several ionizable residues. The proximal segment of the native HA coiled coil. These unfold positioning of residues 1–3 of the relocated fusion peptide to form a loop structure that causes residues at the also blocks access to another internal cavity in the trimer C-terminal end of each α-helix to reorient by 180º. interior that contains additional ionizable residues [1,3]. It is The short helical domains pack antiparallel to the coiled coil, thought that the relocation of the fusion peptide and the and residues C-terminal to this extend towards the same end changes in the environment of ionizable residues in this of the structure as the fusion peptide. The resulting structure region after cleavage may set the trigger for subsequent is a rod-shaped trimeric molecule with the N- and C-termini conformational changes required for fusion. at the same end [32]. Similar structures have been observed Acidification of native cleaved HA in the endosome leads for many other viral fusion proteins, such as those of to three prominent structural refolding events (Fig. 5). The Moloney murine leukemia virus, HIV-1, simian immuno-

globular membrane-distal heads of HA1 de-trimerize, but deficiency virus-1, human T cell leukemia virus-1, Ebola remain tethered to the trimeric HA stalk through residues virus, the paramyxovirus SV5, parainfluenza virus 5, and for

28–43 of HA1 [8,31]. In the HA2 subunit, the fusion peptide the vesicle (v)- and target (t)-SNARE complexes involved in

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 61 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 62

ZHU-NAN LI AND DAVID A STEINHAUER

Figure 5. Ribbon diagram summarizing the conformational provide attractive phases of the entry process for potential

changes that take place during membrane fusion based on intervention by antiviral drugs. For example, HA0 cleavage is the structures of the native HA (left) and the low pH HA required for membrane fusion activity and, thus, virus

(right). The membrane-distal HA1 head domains (grey) infectivity [28,29], and there are examples in which become detrimerized, and their location during fusion is inhibitors such as peptidyl chloromethylketones that mimic essentially unknown, as there is no structural information on the cleavage site have been shown to block cleavage the residues shown in hatched lines. The residues in the activation of HA in cell culture [35]. The calcium-specific extended chain (light grey) between the top of the native ionophore A23187 has also been found to inhibit cleavage HA long helix (light blue) and the shorter helix (dark grey) of HAs processed intracellularly by the calcium-dependent also become helical, and these form an extended coiled coil protease furin, presumably by interfering with the calcium (light blue–light grey–dark grey) in the low pH HA. The α- concentration in organelles where HA cleavage takes place helical residues of native HA shown in dark blue form a loop [36]. However, most strategies involving protease inhibition structure to invert the helical residues shown in blue in the are likely to encounter problems with toxicity. opposite direction. This also inverts residues C-terminal to On the other hand, approaches designed to prevent this in the direction antiparallel to the central coiled coil, and post-cleavage folding or acid-induced conformational positions the transmembrane domain at the same end of changes, using low-molecular-weight compounds with

the molecule as the fusion peptide [91]. limited toxicity, might be feasible. The HA0 cleavage loop resides adjacent to a cavity that is lined by ionizable residues,

as shown in Fig. 4 [7]. Following HA0 cleavage, the residues that make up the lower (membrane-proximal) half of the loop are liberated and insert into the cavity, burying the

conserved residues D109 and D112 in HA2 and the partially

conserved H17 in HA1. In addition, this relocation of the HA2 N-terminal domains occludes another cavity in the interior of the trimer from access to bulk solvent. The second cavity also contains ionizable residues that may play a role in

fusion. The relocation of the HA2 N-terminal fusion peptide and the change in the ionization state of nearby residues is Fusion pH thought to set the trigger for the subsequent acid-induced conformational changes required for fusion, as low pH fails N2 C 1 to induce the equivalent conformational changes in

uncleaved HA0 [7]. It is possible that compounds designed to bind in these cavities could either prevent the cleaved fusion peptide from inserting appropriately to prime HA for fusion, or might C2 influence the structure or ionization state of nearby residues such that they are not induced to initiate conformational N 1 changes upon acidification. Compounds have been identified using structure-based drug-design strategies such as these, and some demonstrate antiviral activity against certain HA Reproduced with permission from [91]. HA: hemagglutinin. subtype viruses. Inhibition of H3 subtype viruses by benzo- and hydroquinones was found to be due to the inhibition of synaptic fusion [33,34]. The formation of the rod-like acid-mediated HA conformational changes [37]. Similar structures is suggested to provide both a mechanism for the results were seen with quinolizidine-linked benzamide close apposition of the prefusion membranes and the driving compounds [38,39], which were shown to prevent HA force for the fusion of the bilayers. conformational changes of H1 and H2, but not H3, HA. Inhibition of H1 and H2 conformational changes was also Potential targets for HA fusion inhibition reported using an N-substituted piperidine compound [40]. The steps involving HA cleavage, formation of the native HA Photo-affinity labeling experiments [41] and mutant selection structure, and the subsequent conformational changes that studies [40] suggest that, as predicted, these compounds lead to membrane fusion are all vital for viral replication, and probably act by binding to the HA stalk region in proximity to

62 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 63

THE MECHANISM AND INHIBITION OF INFLUENZA VIRUS ENTRY

the native pH fusion peptide. In another report, it was shown these compounds were recognized for their capacity to that diiodofluorescein can inhibit virus infectivity by inhibit influenza A viruses, and amantadine (1-amino-1- facilitating HA conformational changes, presumably by adamantane hydrochloride) and its counterpart, rimantadine triggering the irreversible HA structural rearrangements (1-methyl-1-adamantane methylamine hydrochloride), were before fusion function is called for [42]. being utilized clinically long before their mode of action was As stated above, a number of viral fusion proteins adopt understood. In fact, the recognition of M2 as a proton stable rod-shaped structures during the fusion process. Like channel derived in large part from studies on influenza HA, these contain a trimeric central coiled coil and an mutants selected for resistance to these drugs. antiparallel polypeptide chain that functions to draw the Amantadine has two antiviral effects on influenza A virus viral and cellular membranes together. For many of these [49]. At high concentrations (>0.1 mmol/L), the drug fusion proteins, the antiparallel chain is in the form of a increases the pH of endosomes such that HA conformational second group of three α-helices that pack against the changes are not triggered during virus entry. This is outside of the coiled coil to form what is referred to as a six- analogous to that of other weak bases, such as ammonium helix bundle. For these proteins, which include the fusion chloride and chloroquine, and is non-specific, as the antiviral proteins of paramyxoviruses and retroviruses, strategies have activity of high concentrations of amantadine and other been undertaken to utilize helical peptide inhibitors designed lysosomotropic compounds is observed with influenza B as to competitively prevent six-helix bundle formation and well as several other enveloped viruses. This results in the membrane fusion [43,44]. Such approaches are less selection of HA mutants with an elevated pH during straightforward for influenza HA, as the antiparallel membrane fusion [50]. polypeptide chain in the low pH structure does not form a Lower concentrations (0.1–5 µmol/L) of either true six-helix bundle. In HA, the hairpin loop is followed by amantadine or rimantadine exhibits strain-specific antiviral only approximately four turns of helix in the antiparallel effects on influenza A viruses. For most susceptible virus direction, the protein then adopts an extended chain strains, the effect is seen at early stages of infection, where structure that traces along a groove in the central coiled coil. the drugs block the proton-dependent proton channel While it remains possible that peptide-based compounds activity of M2 in endosomes (Fig. 1). This abrogates the might be designed to prevent proper formation of HA helical acidification of the virus interior, which is necessary for the rods, the approaches could be more problematic than for the dissociation of nucleocapsids from the M1 matrix protein six-helix bundle proteins. However, benzotriazole-based and the subsequent uptake of nucleocapsids into the compounds, which appear to block formation of the trimer nucleus [51–54]. An effect late in infection can be seen with of hairpins in a manner similar to peptide inhibitors, show viruses that contain HAs that are cleaved intracellularly, such promise in respiratory syncytial virus [45,46], and it may be as pathogenic H5 and H7 avian strains. In these, M2 possible to develop similar compounds for other viruses that ‘chaperones’ the cleaved HAs to the cell surface by form trimeric rod structures during fusion, such as influenza. preventing the acidification of trans-Golgi vesicles. This stops the cleaved HAs undergoing a conformational change The proton channel function of M2 and action prior to reaching the plasma membrane. In the presence of of amantadine and rimantadine amantadine, intracellularly cleaved HAs are expressed on the The viral M2 protein is synthesized as a 97-amino acid cell surface in the low pH conformation and virus release is polypeptide chain with a 19-residue transmembrane impaired [55,56]. sequence, an N-terminal extracellular domain of 24 residues, Analysis of amantadine-resistant mutants revealed and a relatively long cytoplasmic domain of 54 amino acids. mutations in the transmembrane domain of the M2 protein It associates as homotetramers and higher order oligomers, [49], and along with observations that the proton which are expressed abundantly at the infected cell surface ionophore, monensin, antagonizes the effects of M2 by but are sparingly incorporated into budding virus particles equilibrating proton gradients within the cell, suggested that [47,48]. M2 is a proton channel that functions during virus M2 might act to modulate the pH of intracellular entry to acidify the interior of virus particles when they are compartments [57]. The finding that M2 forms tetramers in the endosome, a critical step that frees ribonucleoproteins allowed for modeling of the transmembrane domains in the to enter the nucleus following membrane fusion (Fig. 1). It form of a helical tetrameric channel in which the positions also functions at late stages of infection to help regulate the implicated in drug resistance orient towards the interior, as pH of trans-Golgi transport vesicles. shown in Fig. 1 [58,59]. Direct evidence for the ion channel It is the proton channel activity of M2 that is the target activity of M2 was obtained by electrophysiology studies for inhibition by α-adamantane compounds. Interestingly, using M2-expressing Xenopus oocytes, which have no

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 63 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 64

ZHU-NAN LI AND DAVID A STEINHAUER

channel activity of their own [60]. Importantly, subsequent The first crystal structures of NA were reported in 1983 studies showed that the channel activity was selective for [71], and the active site interactions with sialic acid were H+ ions [61]. shown by the structure of complexes in 1992 [72]. This led Amantadine has been in clinical use since the 1960s and to the development of the NA inhibitor zanamivir, which rimantadine was widely used in Russia long before it was was designed to bind to this active site of the molecule [73], approved in the US, in 1993 [62]. These two drugs have and several such inhibitors have now been characterized. Of been shown to be effective against many influenza A these, zanamivir – which is administered as an inhalant – viruses, both prophylactically and therapeutically. They are and its orally administered counterpart, oseltamivir, have reasonably well tolerated at low doses, but at higher been in clinical use since 1999. concentrations, amantadine exhibits a greater frequency of The NA inhibitors lead to reduced virus dissemination in neurological and gastrointestinal side effects compared with a fashion that is reminiscent of NA-defective viruses, and rimantadine [63]. can be effective against viruses of all subtypes. They do not Amantadine and rimantadine could prove useful during block virus infection, but are effective at reducing the the initial stage of a pandemic provided that the virus strains severity of disease if administered within the first few days involved are susceptible to the drugs. This would be after contracting the infection. Several studies have especially important before vaccines could be developed and described NA inhibitor-resistant viruses, and the mutations manufactured, particularly for segments of the population leading to resistance map not only to the vicinity of the NA such as healthcare providers. These agents are also less active site, as would be expected, but also to HA regions expensive than the neuraminidase inhibitors discussed below, predicted to affect virus attachment [74]. Thus, the virus can and have a long shelf-life so are suitable for stockpiling. circumvent the inhibitor-mediated block to virus shedding by However, both drugs suffer from the drawback that they are mutating HA to bind less tightly, or possibly by mutating to strain-specific and that resistant mutants arise at a relatively alter glycosylation. This provides further evidence for the high frequency in virus-infected patients receiving treatment existence of a functional balance between HA and NA [64,65]. A worrying feature of H3N2 viruses is the recent proteins in influenza virus infections [75]. Unlike the dramatic increase in resistance to M2 inhibitors among strains situation with amantadine and rimantadine, NA inhibitor- circulating in the US, from <2% of isolates examined to resistant mutants are not readily isolated from patients >90% over the past 2 years [66]. Among the avian H5N1 receiving these drugs [76,77]. To date, zanamivir-resistant strains that have spread from Southeast Asia in recent years, viruses have not been isolated from immunocompetent a lineage that is currently prevalent in Indonesia and through patients, and oseltamivir-resistant mutants arise at levels of China, Mongolia, Russia, and Turkey seems to be sensitive, approximately 1% for adults and 5% for children. In while others are resistant [67]. Thus, it will be necessary to addition, the resistant mutants that have been analyzed are continue monitoring emerging strains to develop plans to often found to be debilitated [78–81]. However, the control them. situation could change and vigilance is crucial in continuing to monitor clinical isolates and emerging strains for NA and its inhibitors susceptibility to NA inhibitors. In the event of a pandemic, Although this review focuses on virus entry, it is appropriate to these inhibitors may provide a critical line of defense, briefly discuss the critical balance between the HA receptor particularly if the strains are resistant to M2 inhibitors. binding and NA receptor-destroying functions, together with currently available NA inhibitors. NA may have some functions Conclusion in virus entry, e.g. to counteract the inhibitory effects of mucins The processes involved in the entry of influenza viruses into that block access to epithelial cells of the respiratory tract [68]. host cells and NA function at the end of the virus life cycle However, the primary function of the viral NA is to cleave sialic provide many steps that might be exploited for the design of acid from host cell and viral surfaces at the end of the virus life antiviral drugs. Among existing agents, the M2 inhibitors, cycle (Fig. 1). Studies on temperature-sensitive mutants [69], amantadine and rimantadine, and the NA inhibitors, and viruses with truncated NA proteins [70] show that zanamivir and oseltamivir, have been most widely used infection, RNA replication, and protein synthesis are not clinically; several review articles detail their modes of action, affected in the absence of NA function, but that virus particles effectiveness when administered therapeutically or accumulate at the surface of infected cells. This is due to the prophylactically, and their advantages and disadvantages presence of an abundance of functional receptors at these [62,77,82–84]. However, it is vital to persevere with sites, and their interactions with sialic acid binding sites of the research into novel inhibitors of influenza, such as those viral HA glycoproteins. discussed here, and others that modulate pathways such as

64 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 65

THE MECHANISM AND INHIBITION OF INFLUENZA VIRUS ENTRY

cell signaling [85], endosomal trafficking [86], and those 23. Roy R, Zanini D, Meunier SJ et al. Solid-phase synthesis of dentritic sialoside inhibitors of influenza A virus hemagglutinin. J Chem Soc Chem Commun 1993;1869–72. involving various stages of the RNA “life cycle” [87–90]. 24. Pritchett TJ, Paulson JC. Basis for the potent inhibition of influenza virus infection by The continued expansion of knowledge of the structure, equine and guinea pig alpha 2-macroglobulin. J Biol Chem 1989;264:9850–8. 25. Rogers GN, Paulson JC, Daniels RS et al. Single amino acid substitutions in influenza function, and biology of influenza viruses and their haemagglutinin change receptor binding specificity. Nature 1983;304:76–8. components should lead to the further development of 26. Glick GD, Toogood PL, Wiley DC et al. Ligand recognition by influenza virus. The binding improved methods for the control of these viruses. of bivalent sialosides. J Biol Chem 1991;266:23660–9. 27. Chinarev AA, Tuzikov AB, Gambarian AS. Tetravalent blockers for influenza virus hemagglutinin. In: Sialobiology and Other Novel Forms of Glycosylation. Inoue Y, Lee LC, Acknowledgment Troy FA, editors. Osaka, Japan: Gakushin Publishing Co., 1999:135–43. 28. Lazarowitz SG, Choppin PW. Enhancement of the infectivity of influenza A and B viruses The authors would like to thank William Langley and by proteolytic cleavage of the hemagglutinin polypeptide. Virology 1975;68:440–54. Byeong-Jae Lee for critical comments on the manuscript. 29. Klenk HD, Rott R, Orlich M et al. Activation of influenza A viruses by trypsin treatment. Virology 1975;68:426–39. 30. Bosch FX, Garten W, Klenk HD et al. Proteolytic cleavage of influenza virus Disclosures hemagglutinins: primary structure of the connecting peptide between HA1 and HA2 determines proteolytic cleavability and pathogenicity of avian influenza viruses. Virology Dr Steinhauer receives research support from The US National Institutes 1981;113:725–35. of Health (grant AI66870). Neither author has any relevant financial 31. Bizebard T, Gigant B, Rigolet P et al. Structure of influenza virus haemagglutinin complexed with a neutralizing antibody. Nature 1995;376;92–4. relationships to disclose. 32. Chen J, Skehel JJ, Wiley DC. N- and C-terminal residues combine in the fusion-pH

influenza hemagglutinin HA2 subunit to form an N cap that terminates the triple-stranded coiled coil. Proc Natl Acad Sci USA 1999;96:8967–72. References 33. Harrison SC. Mechanism of membrane fusion by viral envelope proteins. Adv Virus Res 1. Ha Y, Stevens DJ, Skehel JJ et al. H5 avian and H9 swine influenza virus haemagglutinin 2005;64:8967–72. EMBO J 21 structures: possible origin of influenza subtypes. 2002; :865–75. 34. Skehel JJ, Wiley DC. Receptor binding and membrane fusion in virus entry: the influenza 2. Gamblin SJ, Haire LF, Russell RJ et al. The structure and receptor binding properties of the hemagglutinin. Annu Rev Biochem 2000;69:531–69. Science 303 1918 influenza hemagglutinin. 2004; :1838–42. 35. Stieneke-Grober A, Vey M, Angliker H et al. Influenza virus hemagglutinin with multibasic 3. Russell RJ, Gamblin SJ, Haire LF et al. H1 and H7 influenza haemagglutinin structures cleavage site is activated by furin, a subtilisin-like endoprotease. EMBO J extend a structural classification of haemagglutinin subtypes. Virology 2004;325:287–96. 1992;11:2407–14. 4. Stevens J, Corper AL, Basler CF et al. Structure of the uncleaved human H1 hemagglutinin 36. Klenk HD, Garten W, Rott R. Inhibition of proteolytic cleavage of the hemagglutinin of from the extinct 1918 influenza virus. Science 2004;303:1866–70. influenza virus by the calcium-specific ionophore A23187. EMBO J 1984;3:2911–5. 5. Stevens J, Blixt O, Tumpey TM et al. Structure and receptor specificity of the 37. Bodian DL, Yamasaki RB, Buswell RL et al. Inhibition of the fusion-inducing conformational hemagglutinin from an H5N1 influenza virus. Science 2006;312:404–10. change of influenza hemagglutinin by benzoquinones and hydroquinones. Biochemistry 1993;32:2967–78. 6. Wilson IA, Skehel JJ, Wiley DC. Structure of the haemagglutinin membrane glycoprotein of influenza virus at 3 Å resolution. Nature 1981;289:366–73. 38. Luo G, Colonno R, Krystal M. Characterization of a hemagglutinin-specific inhibitor of influenza A virus. Virology 1996;226:66–76. 7. Chen J, Lee KH, Steinhauer DA et al. Structure of the hemagglutinin precursor cleavage site, a determinant of influenza pathogenicity and the origin of the labile conformation. 39. Luo G, Torri A, Harte WE et al. Molecular mechanism underlying the action of a novel Cell 1998;95:409–17. fusion inhibitor of influenza A virus. J Virol 1997;71:4062–70. 8. Bullough PA, Hughson FM, Skehel JJ et al. Structure of influenza haemagglutinin at the pH 40. Plotch SJ, O’Hara B, Morin J et al. Inhibition of influenza A virus replication by compounds of membrane fusion. Nature 1994;371:37–43. interfering with the fusogenic function of the viral hemagglutinin. J Virol 1999;73:140–51. 9. Gottschalk A, Lind PE. Product of interaction between influenza virus enzyme and 41. Cianci C, Yu KL, Dischino DD et al. pH-dependent changes in photoaffinity labeling ovomucin. Nature 1949;164:232. patterns of the H1 influenza virus hemagglutinin by using an inhibitor of viral fusion. J Virol 1999;73:1785–94. 10. Klenk E, Faillard H, Lempfrid H. Enzymatic effect of the influenza virus. Hoppe Seylers Z Physiol Chem 1955;301:235–46. 42. Hoffman LR, Kuntz ID, White JM. Structure-based identification of an inducer of the low- pH conformational change in the influenza virus hemagglutinin: irreversible inhibition of 11. Rogers GN, D’Souza BL. Receptor binding properties of human and animal H1 influenza infectivity. J Virol 1997;71:8808–20. virus isolates. Virology 1989;173;317–22. 43. Kilby JM, Eron JJ. Novel therapies based on mechanisms of HIV-1 cell entry. N Engl J Med 12. Connor RJ, Kawaoka Y, Webster RG et al. Receptor specificity in human, avian, and equine 2003;348:2228–38. H2 and H3 influenza virus isolates. Virology 1994;205:17–23. 44. Lambert DM, Barney S, Lambert AL et al. Peptides from conserved regions of 13. Ito T, Couceiro JN, Kelm S et al. Molecular basis for the generation in pigs of influenza A paramyxovirus fusion (F) proteins are potent inhibitors of viral fusion. Proc Natl Acad Sci viruses with pandemic potential. J Virol 1998;72:7367–73. USA 1996:93;2186–91. 14. Matrosovich MN, Matrosovich TY, Gray T et al. Human and avian influenza viruses target 45. Cianci C, Langley DR, Dischino DD et al. Targeting a binding pocket within the trimer-of- different cell types in cultures of human airway epithelium. Proc Natl Acad Sci USA hairpins: small-molecule inhibition of viral fusion. Proc Natl Acad Sci USA 2004;101:4620–4. 2004;101:15046–51. 15. Couceiro JN, Paulson JC, Baum LG. Influenza virus strains selectively recognize 46. Cianci C, Meanwell N, Krystal M. Antiviral activity and molecular mechanism of an orally sialyloligosaccharides on human respiratory epithelium; the role of the host cell in selection active respiratory syncytial virus fusion inhibitor. J Antimicrob Chemother of hemagglutinin receptor specificity. Virus Res 1993;29:155–65. 2005;55:289–92. 16. Burnet FM. Mucins and mucoids in relation to influenza virus action, III: inhibition of virus 47. Lamb RA, Zebedee SL, Richardson CD. Influenza virus M2 protein is an integral membrane haemagglutination by glandular mucins. Aust J Exp Biol Med Sci 1948;26:371–9. protein expressed on the infected-cell surface. Cell 1985;40:627–33. 17. Burnet FM, Stone JD. The receptor-destroying enzyme of V. cholerae. 48. Zebedee SL, Richardson CD, Lamb RA. Characterization of the influenza virus M2 integral Aust J Exp Biol Med Sci 1947;25:227–33. membrane protein and expression at the infected-cell surface from cloned cDNA. J Virol 18. Anderson SG, Burnet FM, de St. Groths F et al. Mucins and mucoids in relation to 1985;56:502–11. influenza virus action, VI. General discussion. Aust J Exp Biol Med Sci 1948;26:403–11. 49. Hay AJ, Wolstenholme AJ, Skehel JJ et al. The molecular basis of the specific anti-influenza 19. Sauter NK, Bednarski MD, Wurzburg BA et al. Hemagglutinins from two influenza virus action of amantadine. EMBO J 1985;4:3021–4. variants bind to sialic acid derivatives with millimolar dissociation constants: a 500-MHz 50. Daniels RS, Downie JC, Hay AJ et al. Fusion mutants of the influenza virus hemagglutinin proton nuclear magnetic resonance study. Biochemistry 1989;28;8388–96. glycoprotein. Cell 1985;40:431–9. 20. Lasky LA, Nakamura G, Smith DH et al. Delineation of a region of the human 51. Bukrinskaya AG, Vorkunova NK, Kornilayeva GV et al. Influenza virus uncoating in immunodeficiency virus type 1 gp120 glycoprotein critical for interaction with the CD4 infected cells and effect of rimantadine. J Gen Virol 1982;60:49–59. receptor. Cell 1987;50:975–85. 52. Bukrinskaya AG, Vorkunova NK, Pushkarskaya NL et al. Uncoating of a rimantadine- 21. Matrosovich M, Klenk HD. Natural and synthetic sialic acid-containing inhibitors of resistant variant of influenza virus in the presence of rimantadine. J Gen Virol influenza virus receptor binding. Rev Med Virol 2003;13:85–97. 1982;60:61–6. 22. Sauter NK, Glick GD, Crowther RL et al. Crystallographic detection of a second ligand 53. Martin K, Helenius A. Nuclear transport of influenza virus ribonucleoproteins: the viral binding site in influenza virus hemagglutinin. Proc Natl Acad Sci USA 1992;89:324–8. matrix protein (M1) promotes export and inhibits import. Cell 1991;67:117–30.

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 65 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 66

ZHU-NAN LI AND DAVID A STEINHAUER

54. Bui M, Whittaker G, Helenius A. Effect of M1 protein and low pH on nuclear transport of 73. von Itzstein M, Wu WY, Kok GB et al. Rational design of potent sialidase-based inhibitors influenza virus ribonucleoproteins. J Virol 1996;70:8391–401. of influenza virus replication. Nature 1993;363:418–23. 55. Steinhauer DA, Wharton SA, Skehel JJ et al. Amantadine selection of a mutant influenza 74. McKimm-Breschkin JL. Resistance of influenza viruses to neuraminidase inhibitors – a virus containing an acid-stable hemagglutinin glycoprotein: evidence for virus-specific review. Antiviral Res 2000;47:1–17. regulation of the pH of glycoprotein transport vesicles. Proc Natl Acad Sci USA 75. Wagner R, Matrosovich M, Klenk HD. Functional balance between haemagglutinin and 1991;88:11525–9. neuraminidase in influenza virus infections. Rev Med Virol 2002;12:159–66. 56. Sugrue RJ, Bahadur G, Zambon MC et al. Specific structural alteration of the influenza 76. Monto AS, McKimm-Breschkin JL, Macken C et al. Detection of influenza viruses resistant haemagglutinin by amantadine. EMBO J 1990;9:3469–76. to neuraminidase inhibitors in global surveillance during the first 3 years of their use. 57. Hay AJ. The mechanism of action of amantadine and rimantadine against influenza Antimicrob Agents Chemother 2006;50:2395–402. viruses. In: Concepts of Viral Pathogenesis III. Notkins AL, Oldstone MBA, editors. New York, USA: Springer Verlag, 1989:561–7. 77. McKimm-Breschkin JL. Management of influenza virus infections with neuraminidase inhibitors: detection, incidence, and implications of drug resistance. Treat Respir Med 58. Holsinger LJ, Lamb RA. Influenza virus M2 integral membrane protein is a homotetramer 2005;4:107–16. stabilized by formation of disulfide bonds. Virology 1991;183:32–43. 78. Carr J, Ives J, Kelly L et al. Influenza virus carrying neuraminidase with reduced sensitivity 59. Sugrue RJ, Hay AJ. Structural characteristics of the M2 protein of influenza A viruses: to oseltamivir carboxylate has altered properties in vitro and is compromised for infectivity evidence that it forms a tetrameric channel. Virology 1991;180;617–24. and replicative ability in vivo. Antiviral Res 2002;54:79–88. 60. Pinto LH, Holsinger LJ, Lamb RA. Influenza virus M2 protein has ion channel activity. Cell 79. Le QM, Kiso M, Someya K et al. Avian flu: isolation of drug-resistant H5N1 virus. Nature 1992;69:517–28. 2005;437:1108. 61. Chizhmakov IV, Geraghty FM, Ogden DC et al. Selective proton permeability and pH 80. Yen HL, Herlocher LM, Hoffmann E et al. Neuraminidase inhib itor-resistant influenza regulation of the influenza virus M2 channel expressed in mouse erythroleukaemia cells. viruses may differ substantially in fitness and transmissibility. Antimicrob Agents J Physiol 1996;494:329–36. Chemother 2005;49:4075–84. 62. Hayden FG. Antivirals for influenza: historical perspectives and lessons learned. Antiviral Res 2006;71:372–8. 81. Zurcher T, Yates PJ, Daly J et al. Mutations conferring zanamivir resistance in human influenza virus N2 neuraminidases compromise virus fitness and are not stably maintained 63. Hayden FG, Hoffman HE, Spyker DA. Differences in side effects of amantadine in vitro. J Antimicrob Chemother 2006;58:723–32. hydrochloride and rimantadine hydrochloride relate to differences in pharmacokinetics. Antimicrob Agents Chemother 1983;23:458–64. 82. Hayden FG, Hay AJ. Emergence and transmission of influenza A viruses resistant to amantadine and rimantadine. Curr Top Microbiol Immunol 1992;176:119–30. 64. Hayden FG, Sperber SJ, Belshe RB et al. Recovery of drug-resistant influenza A virus during therapeutic use of rimantadine. Antimicrob Agents Chemother 1991;35:1741–7. 83. McKimm-Breschkin JL. Neuraminidase inhibitors for the treatment and prevention of influenza. Expert Opin Pharmacother 2002;3:103–12. 65. Hall CB, Dolin R, Gala CL et al. Children with influenza A infection: treatment with rimantadine. Pediatrics 1987;80:275–82. 84. Gubareva LV. Molecular mechanisms of influenza virus resistance to neuraminidase inhibitors. Virus Res 2004;103:199–203. 66. Bright RA, Shay DK, Shu B et al. Adamantane resistance among influenza A viruses isolated early during the 2005-2006 influenza season in the United States. JAMA 85. Garcia-Sastre A and Biron CA, Type 1 interferons and the virus-host relationship: a lesson 2006;295:891–4. in detente. Science 2006;312:879–82. 67. Hayden F, Klimov A, Tashiro M et al. Neuraminidase inhibitor susceptibility network 86. Marsh M and Helenius A. Virus entry: open sesame. Cell 2006;124:729–40. Antivir Ther position statement: antiviral resistance in influenza A/H5N1 viruses. 87. Ge Q, Filip L, Bai A et al. Inhibition of influenza virus production in virus-infected mice by 10 2005; :873–7. RNA interference. Proc Natl Acad Sci USA 2004;101:8676–81. 68. Matrosovich MN, Matrosovich TY, Gray T et al. Neuraminidase is important for the 88. Tompkins SM, Lo CY, Tumpey TM et al. Protection against lethal influenza virus challenge initiation of influenza virus infection in human airway epithelium. by RNA interference in vivo. Proc Natl Acad Sci USA 2004;101:8682–6. JVirol2004;78:12665–7. 89. Twu KY, Noah DL, Rao P et al. The CPSF30 binding site on the NS1A protein of influenza 69. Palese P, Tobita K, Ueda M et al. Characterization of temperature sensitive influenza virus A virus is a potential antiviral target. J Virol 2006;80:3957–65. mutants defective in neuraminidase. Virology 1974;61:397–410. 70. Liu C, Air GM. Selection and characterization of a neuraminidase-minus mutant of 90. Fechter P and Brownlee GG. Recognition of mRNA cap structures by viral and cellular influenza virus and its rescue by cloned neuraminidase genes. Virology 1993;194:403–7. proteins. J Gen Virol 2005;86:1239–49. 71. Colman PM, Varghese JN, Laver WG. Structure of the catalytic and antigenic sites in 91. Skehel JJ, Steinhauer DA, Wharton SA et al. Receptor binding and membrane fusion by influenza virus neuraminidase. Nature 1983;303:41–4. influenza hemagglutinin. In: Cellular receptors for Animal Viruses. Cold Spring Harbor Laboratory Press, 1994:187–93. 72. Varghese JN, McKimm-Breschkin JL, Caldwell JB et al. The structure of the complex between influenza virus neuraminidase and sialic acid, the viral receptor. Proteins 92. Hayden FG. Antiviral resistance in influenza viruses – implications for management and 1992;14:327–32. pandemic response. N Engl J Med 2006;354:785–8.

66 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 COMMENTARY n 67 activity against experimentation, in vitro in vitro (2):67–72. 2 2006: J Viral Entry J Viral Weill Medical College, New York, NY, USA NY, College, New York, Medical Weill Vol 2 No 2 2006 Vol In the future, it may also be possible to combine these Different classes of active agents in microbicides have ew are already in clinical trials. Entry inhibitors interfere , recombinant receptor fragments, peptides, chemically with HIV-1. and small organic molecules. This article considers some ntry inhibitors act either on the viral or the cellular side, NTRY agents with vaccines. So far, the agents most with intractable vaccines. i difficulty So far, HIV-1 vaccine development has been to elicit high enough levels of neutralizing antibodies. HIV-1-specific cytotoxic lymphocytes can readily considered be sufficient induced, for sterile but immunity cellular immunity this in [3]. combination with a However, microbicidal is barrier not that substitutes for [4]. adequate protection mucosal antibodies might provide been tested in cell culture transmission, and some and have progressed animal to clinical studies. models of One viral candidate, Nonoxinol-9 ethanol [N-9]), a non-ionic surfactant that disrupts (nonylpenoxypolyethoxy- the viral lipid envelope, has already been assessed and rejected in a clinical trial, as it was found to damage the vaginal mucosa and thus enhance the risk notwithstanding, of infection other [5]. This surfactants failure evaluated are in currently human trials, being as polyanionic are gels, and inhibitors various of HIV pH reverse transcriptase modifiers, [6–9]. This review, however, will focus on another diverse l or rectal mucosa in order to block the transmission of HIV nclude entry inhibitors. Is inhibition of infectious entry into E nsidered for such use, based on their al envelope with the cytoplasmic membrane of the target cell, ral attachment to susceptible cells, docking onto receptors, o receptors and coreceptors. These agents are very diverse and organism? bypass Can a HIV-1 mucosal microbicidal barrier by re discussed against the background of

IRAL V gy and k Avenue, OURNAL OF J HE T Transmission Transmission to women is currently more common than to men; among 15- to 24-year-olds in sub-Saharan Africa, females are three times more likely to be than males [2]. Microbicides HIV-1-infected offer the advantage of a means of protection that women can control; they may be used without their sexual partners’ exclude knowledge the and possibility do conception. not that Hence, there intercourse are may as a valuable complement to condoms. microbicides many lead reasons to to regard With With an incidence of approximately 4 million HIV-1 infections per year [1], the world urgently needs effective measures preventive to curb the spread of this transmissions virus. The majority of occur through satisfactory vaccine is unlikely to become sexual available within the intercourse. next few years, As one preventive a microbicides, which strategy are agents now designed to focuses prevent infection on mucosa. when applied topically to the cervico-vaginal or rectal macaque transmission models, and the biology of human infection macaque transmission models, and the modified chemokines, carbohydrate-binding proteins, polyanions, important questions in developing effective microbicides that i target cells in the mucosa sufficient to prevent entry into the dendritic cells? These getting issues captured a by trafficking NY 10021, USA. Email: [email protected] Box 62, New York, Address for correspondence: PJ Klasse, Department of Microbiolo for correspondence: Address belong to a range of different classes of molecules: antibodies Medical College, 1300 Yor Weill Cornell University, Immunology, HIV-1 HIV-1 infection; some have been tested in animal models and a f with the earliest events in the viral replication cycle: the vi subsequent interactions with coreceptors, and fusion of the vir all of which are mediated by the viral envelope glycoprotein. E binding to different subunits of the envelope glycoprotein or t Per Johan Klasse Cornell University, and Immunology, Department of Microbiology Microbicides are designed for topical application on the vagina type 1 virus. (HIV-1) Diverse substances are currently being co What is the Role for Entry Inhibitors in Inhibitors for Entry Role is the What Transmission for Blocking Microbicides or via the Cervico–Vaginal of HIV-1 Mucosa? Rectal R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 67 Page 16:40 2/11/06 R7694_2_REM_VE_2_2_07.qxd R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 68

PER JOHAN KLASSE

group of substances, namely molecules that specifically interfere with viral entry into susceptible cells.

Molecular interactions of viral entry into the host cell of the antibodies

HIV-1 particles, or virions, are studded with spikes of n activation,

the envelope glycoprotein, Env, which mediate virus receptor entry [10]. Each protomer of Env consists of a surface

moiety (SU) and a transmembrane protein (TM) that sylated proteins

anchors the complex in the phospholipid bilayer of the teins viral membrane. The SU and TM are linked to each other through non-covalent bonds. Trimers of these SU–TM heterodimers stud the surface of the virion. An infectious cycle begins when a virion approaches

the target-cell surface and a trimer binds to the main R4 DC-SIGN Mannose G5 AZN-D1 Clone 19, -defensins Cell surface cellular receptor for the virus, CD4. The binding β induces conformational changes in the Env oligomer that promote interactions with a cellular coreceptor – CCR5 for R5 viral strains, or CXCR4 for X4 strains (for reviews, see [10–12]). The coreceptor interactions in omplex but binds Fabs well [21]. Such antibody fragments or any al glycosylation sites must be mutated for viral escape [16]. maraviroc vicriviroc, AMD3465

turn trigger a refolding of TM into a coiled coil, which upo adhesion molecule; RANTES: regulated t; ICAM: intercellular promotes the fusion of the viral and cellular membranes [13,14]. Each individual step in the entry process, from attachment to fusion, can be blocked by -defensins PSC-RANTES cell- Stromal ICAM-3 Mannosylated specific inhibitors that bind to Env, the receptors, or the θ coreceptors (Table 1). - and α Viral ligands Sulfated polyanions interact with positively charged side chains in Env, particularly the hypervariable V3 region in X4 viral strains [15]. Other compounds, such as Cyanovirin-N (CV-N), coat virus particles through interactions with the abundant and mannose-rich carbohydrate moieties on Env [16]. This type of binding may inhibit infection by increasing the Escape from this agent may have a high genetic barrier as sever Escape from included to illustrate the principles of entry inhibition. repulsion of virions from target cells and by interfering sibility for whole immunoglobulin G antibodies in the CD4-Env c -3 grabbing non-integrin; Fab: antigen-binding antibody fragmen with viral binding to ancillary cell-surface attachment molecules. It may also block binding to specific receptors through steric hindrance [17]. The use of these molecules in microbicides has been reviewed gp120 gp41 CD4 Coreceptor Mannose type-C lectin elsewhere [18–20]. Several specific ligands for Env are also being coreceptor bsFabs*peptide PRO2000, Carraguard, molecules and others 1 and 2 and others receptor considered for microbicide use. However, there are only a few potent neutralizing antibodies (NAbs) that are active against multiple strains of HIV-1, partly because of the extensive sequence variation of the env gene molecules CCR5 and lectin-like BMS-378806 5MO30 NSC 13778 CMPD167, AMD3100, Mannan Mannan CD4 bs V3 region/ Mannose shield Ecto-domain Mainly domains CCR5 CXC [21]. Among these broadly neutralizing antibodies, some appear quite promising. The NAb b12, directed

against the CD4-binding site (CD4bs) on gp120, has inhibitors. . Categories of HIV type 1 entry and capture been shown to provide protection against vaginal peptides soluble CD4 N-terminal other lectins defensins factor-1, Pro proteins/ proteins/ other sulfated defensins, and C52L, T-1249, derived Manno Monoclonal antibody b12Other 447-52D, PRO 542,Polyanions 2G12 Tyrosine- anti-CD4i Small molecules Cyanovirin-N**, Enfuvirtide, 2F5, 4E10, SPL7013, Q4120, TNX- 355, PRO 140, Z13 12 and others 2D7, and others Clone 15-2 Table 1 Table Inhibitor Virus bs: binding site; DC: dendritic cell; DC-SIGN: DC-specific ICAM **CV-N blocks both infection of lymphocytes and capture by DC. blocks both infection of lymphocytes and capture **CV-N mentioned may not be feasible as microbicide components but are mentioned may not be feasible as microbicide challenge with a simian–human immunodeficiency and secreted. expressed normal T-cell * CD4i = CD4-induced epitope on gp120: this site has poor acces

68 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 69

WHAT IS THE ROLE FOR HIV ENTRY INHIBITORS IN MICROBICIDES?

virus, SHIV-SF162P4 [22]. A tetra-valent recombinant form Cell surface-receptor ligands of CD4 could act similarly [7]. Another anti-gp120 antibody, HIV-1 infection depends on the binding of gp120 to CD4. This 2G12, binds an epitope that comprises distal mannose interaction tethers the viral envelope to the cell membrane and residues of the glycan shield on the outer domain of the triggers the subsequent steps in the entry process. The blocking molecule [21]. Three cross-reactive NAbs (2F5, 4E10, and of CD4 with monoclonal antibodies (MAbs) such as TNX-355 – Z13) directed against gp41, have also been identified; these a humanized MAb currently in Phase II clinical trials for systemic recognize adjacent epitopes close to the TM segment [23]. use – has therefore been considered for microbicide application The anti-gp120 and anti-gp41 NAbs may neutralize the [7]. The defensins, mentioned above, bind to CD4 in addition virus by different mechanisms: the former hinder attachment to binding to Env, and this may contribute to their inhibition of to certain target cells; the latter block subsequent steps of infection [26]. A small-molecule ligand for CD4, NSC 13778, the entry process [17]. that prevents gp120 binding has also been described [34]. Direct interference with the fusion between the virus and Inhibitory occupancy on cell surface receptors can be achieved the host cell can also inhibit infection. Peptides derived from before exposure to the virus, and, if dissociation is slow, the gp41 sequence can block viral fusion by interfering with protection will be maintained for hours after application [9]. the formation of a fusogenic six-helix bundle [24]. One such The next step in host cell entry is the interaction with the peptide, C52L, has shown partial protection from vaginal coreceptor. PSC-RANTES, a modified version of the challenge with SHIV in the macaque model [25]. chemokine RANTES (regulated upon activation, normal Interestingly, the innate immune peptides, α- and θ- T-cell expressed and secreted), and the small molecule defensins, bind to gp120 with dissociation constants in the CMPD167, both of which bind to CCR5, have shown good nanomolar range and also block HIV-1 infection [26]. protection in the macaque model of vaginal R5 SHIV In addition, θ-defensins have been shown to bind to gp41 transmission [25,35]. This is encouraging, as almost and to block the formation of the fusogenic six-helix exclusively CCR5-using strains are transmitted [11,12]. bundle [27]. However, X4 virus can be present in the semen of an infected It remains to be seen whether the production of NAbs or individual. It would be of grave concern if the blocking of R5 other proteins for use in microbicides can be made transmission gave a relative preference to the transmission of economically feasible [9]. One strategy may be to colonize X4 variants, as presence of X4 virus is associated with a more the vagina or rectum with recombinant bacteria in order to rapid progression to AIDS. Hence, there is a rationale for also produce the inhibitors in situ [28–31]. However, whether this trying to block X4 infection with microbicides. Some approach can yield sufficient local concentrations for inhibitors, such as the sulfated polyanions that interact more inhibition is debatable [9]. Aside from their high cost, strongly with the X4 gp120 than the R5 gp120, would inhibit proteins also have other disadvantages: they may degrade, X4 at least as efficiently as R5 virus. Others, including the denature, and elicit immune responses. Inflammation in the cross-neutralizing antibodies, small-molecule gp120 ligands, mucosa would be likely to augment the risk of infection soluble CD4-constructs, fusion-blocking peptides, and CD4 through recruitment of susceptible cells (Fig. 1). Therefore, ligands would block most strains of either tropism. And, in small-molecule ligands of Env are attractive microbicide analogy with blocking CCR5, the coreceptor for X4 virus, candidates. One such molecule, BMS-378806, which binds CXCR4, could also be blocked. This can be achieved with the to the CD4bs on gp120 and blocks viral docking onto the physiological ligand for CXCR4, stromal cell-derived factor 1, receptor [32], has shown protection in the macaque model or possibly by certain defensins [36]. The small-molecule [25]. Small-molecule gp41 ligands that mimic the fusion- CXCR4 ligands, AMD3100 and AMD3465, inhibit X4 blocking peptides are now also being identified [33], and infection in vitro, and could be combined with CCR5 ligands would provide valuable additions to the microbicide arsenal. and general blocking agents in a microbicide [7,9]. Viral ligands can only bind once the infectious inoculum is deposited on the mucosa. They must then attach rapidly, Combinations of inhibitors when the semen and micribicide solution mix by diffusion, in In addition to the aforementioned worry that blockade of the order to prevent infectious encounters with the first CCR5 receptor might somehow promote transmission of X4 potential target cells. Those ligands that depend on receptor virus, there is also a potential risk that resistant strains could be interactions with Env for the induction of their binding sites preferentially transmitted, or amplified in tissues with must be present at sufficiently high concentrations in the suboptimal concentration of inhibitor. This is a further reason relevant tissue compartment at that late step in entry. The to include several inhibitors in a formulation, a strategy that situation is principally different for the counterpart inhibitors would also generally enhance the effectiveness at obtainable on the cellular side. concentrations. Indeed, combining the CCR5 ligand,

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 69 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 70

PER JOHAN KLASSE

Figure 1. The female genital mucosa: the locale where microbicides may prevent transmission. The squamous stratified epithelium of the vagina forms an efficient barrier to viral penetration when intact. The columnar epithelium of the cervix uteri provides a weaker barrier against infection, but cervical mucus may trap virions [45]. However the plug of mucus would not protect ectopic cervical epithelium. Ex-plant experiments suggest that HIV cannot cross the epithelial barrier in the absence of a rift. The figure depicts virus in infectious semen that has been deposited in the vagina. Both CCR5- and CXCR4-using virions (blue and grey, respectively) are present. With few exceptions, R5 virus comes to dominate in a newly infected host; it may be preferentially transmitted or disproportionately amplified early in infection [11,12]. The dendritic cell (DC) to the left has captured a virion from the lumen. DCs express CD4 and CCR5 and can thus be infected by HIV; deeper in the tissue, an infected DC has progeny virions budding from the cell surface (sections studded with black Env spikes). DCs also express DC-SIGN (DC-specific ICAM-3 grabbing non-integrin) and other mannose C-type lectins (MCLR) on their surface, which are used for binding to the virus and subsequent internalization. Whether DCs contribute to natural infection by conveying virus bound to MCLRs across the epithelial barrier is not known [37]. To establish an infection, the virus must reach a number of CD4+ CCR5+ lymphocytes as shown in the propagation step in the middle of the figure. Infected macrophages (shown on the right) are also considered to be an important source of progeny virus [46]. The arrows show different routes of traversing the mucosa, infection of the first target cells, and migration of progeny virus or DCs with captured virus to the lymph nodes, via the afferent lymphatics. The lymph nodes are rich in CD4+ lymphocytes. From these target cells and infected macrophages, new generations of progeny virus cascade to the next level of lymphatic tissue. Viremia peaks when the virus reaches the gut-associated lymphoid tissue (GALT); its subsequent decline may be due to a depletion of the reservoir of susceptible cells in the GALT. Eventually, the virus disseminates via efferent lymphatics and blood to the spleen, brain, liver, and lungs.

Crossing X4 the barrier R5

Mucus Vagina Cervix CLR CCR5

CD4

Infected DC CD4/CCR5 infection CD4 Propagation of infection CCR5

Infected Infected CD4+ T cell macrophage CD4/CCR5 infection

DC with captured virus CLR - capture (R5 or X4) Dissemination Afferent lymphatics

T cell Spread of captured and/or de novo virus

Enhanced virus spread to CD4+ T cells

Lymph node

Further dissemination via central lymphatics and blood to spleen, Efferent lymphatics brain, liver and lungs, etc.

Reproduced with permission from [4].

70 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 71

WHAT IS THE ROLE FOR HIV ENTRY INHIBITORS IN MICROBICIDES?

CMPD167, or the gp120 ligand, BMS-378806, with the inhibitors that bind to the cell as opposed to the virus [8,9]. fusion-blocking peptide, C52L, has shown improved protection How can this be explained? The viral doses used in the two in the macaque model compared with using single drugs. This situations are usually comparable; furthermore, if some 90% might be explained by the synergy observed in vitro between of the infectivity of the inoculum were blocked in the lumen C52L and either CMPD167 or BMS-378806 [25]. or on the mucosa, this should be reflected in a substantial degree of protection, since when 10-fold lower challenge Is entry block enough? doses than the usual are used in the macaque model, Studies have shown that HIV-1 can bind to and infect dendritic significantly lower frequencies of infection ensue. Hence, the cells (DCs) [37]. If dendrites of these cells become exposed in explanation does not seem to be that a greater extent the lumen of the vagina, for example through coitally-induced of inhibition is required in vivo than in vitro microabrasions, virus particles may be picked up, transported [9,18,22,35,39–44]. Nor does capture by DCs seem to be a to regional lymph nodes with the trafficking DCs, and there major explanatory factor, since an agent such as CV-N, infect lymphocytes (Fig. 1). The half-life of DC-trapped virus is which blocks both infection and interactions with MCLRs, prolonged and DCs may also enhance the infectivity of the also evinces high in vivo/in vitro ratios [9,16,39,40]. virus it presents to lymphocytes compared with the What does appear relevant is the concentration gradient corresponding amounts of virions in cell-free suspension. of inhibitor that is created by diffusion from the lumen of Furthermore, it has been observed that HIV-1 can cross the vagina or rectum to interstitial spaces deep in the some epithelia through transcytosis and infect underlying mucosal tissue. This explanation is compatible with a greater lymphocytes in the lamina propria. This would be a more in vivo/in vitro ratio for cell surface receptors than for viral plausible mechanism for transmission across the rectal or ligands. The receptor ligands would have to reach colonic epithelium, than via the female genital tract [7,38]. If sufficiently high concentrations in the deeper tissues to yield this route of infection is relevant, it would be insufficient to blocking occupancies there; the viral ligands, on the other maintain sterilizing concentrations of inhibitors in the lumina hand, would achieve their blocking occupancies in the of the vagina or the rectum. The odds for protection by cell- lumen, then slowly dissociate as the coated virions diffuse or surface-receptor ligands would seem even worse than for traffic with DCs into deeper tissues. Other considerations viral ligands in such a scenario. The latter might stay bound that apply equally to cell surface-receptor and viral ligands to the DC-captured virus, or even prevent the capture. involve the diffusion of inhibitor into the luminal fluids, both However, if infection of DCs – as opposed to their mere before and after deposition of the inoculum. Substantial capture of virus – is an obligatory step in crossing the losses and dilution would be expected, as would variations mucosal barrier, at least agents preventing R5 infection in local distribution due to incomplete mixing [9]. would be effective. In any case, viral ligands that bind only after receptor docking, such as the fusion-blocking peptides, Conclusion do protect in the vaginal macaque model. This would suggest It may seem that the earlier in the viral replication cycle an that cells accessible by viral diffusion from the lumen are inhibitor interferes, the better. But interference with any step important targets in infection. Whether protection could be obligatory to infection is sufficient. Potency, extent, and further enhanced by also blocking the interactions with Env- breadth of inhibition, rather than the mechanism, are of the binding mannose C-type lectin receptors (MCLRs) on DCs is essence. Hence, many entry blockers are attractive microbicide another question. The interaction of virus with MCLRs, such candidates but not prima facie more promising than, for as DC-SIGN (dendritic-cell-specific intercellular adhesion example, reverse transcriptase inhibitors. Due to the extreme molecule 3-grabbing non-integrin), can be blocked by genetic variation of HIV-1, and the palpable risk of selecting mannan and CV-N [16] and perhaps by defensins, which for viral escape, combinations of inhibitors may be the wisest have lectin activity [26] (Table 1). Such inhibitors should strategy. Regardless, it must be shown that the drug therefore be tested in combination with others. formulations do not induce inflammation or in other ways compromise mucosal integrity, which would increase the risk of The high ratio of inhibitory concentrations infection. Candidates that out-compete others based on in vivo/in vitro breadth of inhibition in vitro and degree of protection against It has been observed that inhibitor concentrations several prototype strains in the macaque model may qualify for orders of magnitude higher are required to protect a human clinical trials. But they must also be easy to use and proportion of macaques from vaginal or rectal challenge, relatively cheap [9]. The potency and molecular disparity of the than to block a similar fraction of infectivity in vitro. entry inhibitors mean that some may well fit the bill for Tentatively, this difference is particularly pronounced for inclusion in future microbicides.

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 71 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 72

PER JOHAN KLASSE

22. Veazey RS, Shattock RJ, Pope M et al. Prevention of virus transmission to macaque Acknowledgment monkeys by a vaginally applied monoclonal antibody to HIV-1 gp120. Nat Med The author is grateful to John P Moore for a critical reading 2003;9:343–6. 23. Zwick MB, Saphire EO, Burton DR. gp41: HIV’s shy protein. Nat Med 2004;10:133–4. of the manuscript. 24. Matthews T, Salgo M, Greenberg M et al. Enfuvirtide: the first therapy to inhibit the entry of HIV-1 into host CD4 lymphocytes. Nat Rev Drug Discov 2004;3:215–25. 25. Veazey RS, Klasse PJ, Schader SM et al. Protection of macaques from vaginal SHIV Disclosures challenge by vaginally delivered inhibitors of virus-cell fusion. Nature 2005;438:99–102. Dr Klasse receives research support from the US National Institutes of 26. Wang W, Owen SM, Rudolph DL et al. Activity of alpha- and theta-defensins against primary isolates of HIV-1. J Immunol 2004;173;515–20. health (grant U19 AI 65413). 27. Gallo SA, Wang W, Rawat SS et al. Theta-defensins prevent HIV-1 Env-mediated fusion by binding gp41 and blocking 6-helix bundle formation. J Biol Chem 2006;281:18787–92. 28. Rao S, Hu S, McHugh L et al. Toward a live microbial microbicide for HIV: commensal bacteria References secreting an HIV fusion inhibitor peptide. Proc Natl Acad Sci USA 2005;102:11993–8. 1. Centers for disease control and prevention (CDC). The Global HIV/AIDS pandemic, 2006. 29. Pusch O, Boden D, Hannify S et al. Bioengineering lactic acid bacteria to secrete the HIV-1 MMWR Morb Mortal Wkly Rep 2006;55:841–4. virucide cyanovirin. J Acquir Immune Defic Syndr 2005;40:512–20. 2. Quinn TC, Overbaugh J. HIV/AIDS in women: an expanding epidemic. Science 30. Chang TL, Chang CH, Simpson DA et al. Inhibition of HIV infectivity by a natural human 2005;308:1582–3. isolate of Lactobacillus jensenii engineered to express functional two-domain CD4. Proc Natl Acad Sci USA 2003;100:11672–7. 3. McMichael AJ. HIV vaccines. Ann Rev Immunol 2006;24:227–55. 31. Liu X, Lagenaur LA, Simpson DA et al. Engineered vaginal lactobacillus strain for mucosal 4. Gupta K, Klasse PJ. How do viral and host factors modulate the sexual transmission of delivery of the human immunodeficiency virus inhibitor cyanovirin-N. Antimicrob Agents HIV? Can transmission be blocked? PLoS Med 2006;3:e79. Chemother 2006;50:3250–9. 5. Hillier SL, Moench T, Shattock R et al. In vitro and in vivo: the story of nonoxynol 9. 32. Guo Q, Ho HT, Dicker I et al. Biochemical and genetic characterizations of a novel human J Acquir Immune Defic Syndr 2005;39:1–8. immunodeficiency virus type 1 inhibitor that blocks gp120-CD4 interactions. J Virol 6. Weber JN, Lacey CJ. The development of novel vaginal microbicides: from the bench to 2003;77:10528–36. the clinic. AIDS 2001;15:S35–7. 33. Frey G, Rits-Volloch S, Zhang XQ et al. Small molecules that bind the inner core of gp41 and inhibit HIV envelope-mediated fusion. Proc Natl Acad Sci USA 2006;103:13938–43. 7. Shattock RJ, Moore JP. Inhibiting sexual transmission of HIV-1 infection. Nat Rev Microbiol 2003;1:25–34. 34. Yang QE, Stephen AG, Adelsberger JW et al. Discovery of small-molecule human immunodeficiency virus type 1 entry inhibitors that target the gp120-binding domain of 8. Lederman MM, Offord RE, Hartley O. Microbicides and other topical strategies to prevent CD4. J Virol 2005;79:6122–33. vaginal transmission of HIV. Nat Rev Immunol 2006;6:371–82. 35. Lederman MM, Veazey RS, Offord R et al. Prevention of vaginal SHIV transmission in 9. Klasse PJ, Shattock RJ, Moore JP. Which topical microbicides for blocking HIV-1 rhesus macaques through inhibition of CCR5. Science 2004;306:485–7. transmission will work in the real world? PLoS Med 2006;3:e351. 36. Feng Z, Dubyak GR, Lederman MM et al. Cutting edge: human beta defensin 3—a novel 10. Poignard P, Saphire EO, Parren PW et al. gp120: biologic aspects of structural features. antagonist of the HIV-1 coreceptor CXCR4. J Immunol 2006;177:782–6. Annu Rev Immunol 2001;19:253–74. 37. Pope M, Haase AT. Transmission, acute HIV-1 infection and the quest for strategies to 11. Moore JP, Kitchen SG, Pugach P et al. The CCR5 and CXCR4 coreceptors—central to prevent infection. Nat Med 2003;9:847–52. understanding the transmission and pathogenesis of human immunodeficiency virus type 38. Miller CJ, Shattock RJ. Target cells in vaginal HIV transmission. Microbes Infect 1 infection. AIDS Res Hum Retroviruses 2004;20:111–26. 2003;5:59–67. 12. Margolis L, Shattock R. Selective transmission of CCR5-utilizing HIV-1: the ‘gatekeeper’ 39. Tsai CC, Emau P, Jiang Y et al. Cyanovirin-N gel as a topical microbicide prevents rectal problem resolved? Nat Rev Microbiol 2006;4:312–7. transmission of SHIV89.6P in macaques. AIDS Res Hum Retroviruses 2003;19:535–41. 13. Weissenhorn W, Dessen A, Harrison SC et al. Atomic structure of the ectodomain from 40. Tsai CC, Emau P, Jiang Y et al. Cyanovirin-N inhibits AIDS virus infections in vaginal HIV-1 gp41. Nature 1997;387:426–30. transmission models. AIDS Res Hum Retroviruses 2004;20:11–8. 14. Chan DC, Fass D, Berger JM et al. Core structure of gp41 from the HIV envelope 41. Otten RA, Adams DR, Kim CN et al. Multiple vaginal exposures to low doses of R5 simian- glycoprotein. Cell 1997;89:263–73. human immunodeficiency virus: strategy to study HIV preclinical interventions in 15. Hartley O, Klasse PJ, Sattentau QJ et al. V3: HIV’s switch-hitter. AIDS Res Hum nonhuman primates. J Infect Dis 2005;191:164–73. Retroviruses 2005;21:171–89. 42. McDermott AB, Mitchen J, Piaskowski S et al. Repeated low-dose mucosal simian 16. Balzarini J. Inhibition of HIV entry by carbohydrate-binding proteins. Antiviral Res immunodeficiency virus SIVmac239 challenge results in the same viral and immunological 2006;71:237–47. kinetics as high-dose challenge: a model for the evaluation of vaccine efficacy in nonhuman primates. J Virol 2004;78:3140–4. 17. Klasse PJ, Sattentau QJ. Occupancy and mechanism in antibody-mediated neutralization 43. Marx PA, Spira AI, Gettie A et al. Progesterone implants enhance SIV vaginal transmission of animal viruses. J Gen Virol 2002;83:2091–108. and early virus load. Nat Med 1996;2:1084–9. 18. Weber J, Nunn A, O’Connor T et al. ‘Chemical condoms’ for the prevention of HIV 44. Boadi T, Schneider E, Chung S et al. Cellulose acetate 1,2-benzenedicarboxylate protects infection: evaluation of novel agents against SHIV(89.6PD) in vitro and in vivo. AIDS against challenge with pathogenic X4 and R5 simian/human immunodeficiency virus. 2001;15:1563–8. AIDS 2005;19:1587–94. 19. Stone A. Microbicides: a new approach to preventing HIV and other sexually transmitted 45. Maher D, Wu X, Schacker T et al. HIV binding, penetration, and primary infection in infections. Nat Rev Drug Discov 2002;1:977–85. human cervicovaginal tissue. Proc Natl Acad Sci USA 2005;102:11504–9. 20. Dhawan D, Mayer KH. Microbicides to prevent HIV transmission: overcoming obstacles to 46. Igarashi T, Brown CR, Endo Y et al. Macrophage are the principal reservoir and sustain + chemical barrier protection. J Infect Dis 2006;193:36–44. high virus loads in rhesus macaques after the depletion of CD4 T cells by a highly 21. Pantophlet R, Burton DR. gp120: target for neutralizing HIV-1 antibodies. Annu Rev pathogenic simian immunodeficiency virus/HIV type 1 chimera (SHIV): Implications for Immunol 2006;24:739–69. HIV-1 infections of humans. Proc Natl Acad Sci USA 2001;98:658–63.

72 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 73

CLINICAL REVIEWS Commentary and Analysis on Recent Key Papers

Clinical reviews were prepared by James Hellinger, Andrew Luber, and Mike Youle

Coreceptors disease progression occurred. This was in the absence of any change in viral phenotype, suggesting that the protection afforded by these anti-CCR5 antibodies is independent of Long-lasting CCR5 internalization by antibodies in viral strain and may be an attractive target for potential a subset of long-term nonprogressors: a possible vaccine development. protective effect against disease progression Address for reprints: Lopalco, Infectious Diseases Clinic, San Raffaele Pastori C, Weiser B, Barassi C et al. Scientific Institute, via S. D’Ancona 20, 20127 Milan, Italy. Blood 2006;107:4825–33. Email: [email protected]

The authors of this report discuss the possibility that antibodies against CCR5 may be preventing HIV infection, Drug Interactions or slowing down or abrogating disease progression. In a cohort of patients with and without highly active antiretroviral therapy, anti-CCR5 antibodies were only An anti-CCR5 monoclonal antibody and small seen in long-term non-progressors, in whom they molecule CCR5 antagonists synergize by inhibiting occurred as a fifth of the selected sample. Loss of these different stages of human immunodeficiency virus antibodies was associated with progression to disease. type 1 entry Safarian D, Carnec X, Tsamis F et al. Blockade of the coreceptor CCR5 is now possible with Virology 2006;352:477–84. several antiretroviral agents and appears to reduce the viral burden. However, naturally occurring antibodies that In this study, Safarian and colleagues document that recognize an epitope in the first extracellular loop of the sensitivity to two small-molecule inhibitors (SMIs) of receptor have been identified, and found to down-regulate CCR5-mediated viral entry, SCH-C and TAK-779, is CCR5 on the target cell surface as well as render CD4+ cells regulated by the hypervariable V3 crown of gp120, differ resistant to infection by CCR5-tropic (R5) virus strains. in the timing of inhibition from the monoclonal antibody A selection of 290 subjects from cohorts in Italy and the PA14, and that both SMIs synergize with PA14 to inhibit US were assessed for the presence of anti-CCR5 antibodies CCR5-tropic strains of HIV type 1. and their benefit in preventing progression of HIV disease. The only subjects in whom anti-CCR5 antibodies were Following binding of HIV gp120 to the cell surface CD4 found were long-term non-progressors (23.5% of this receptor, the former undergoes a conformational change to group), while they were not seen in age-, sex-, and risk expose the chemokine receptor binding site. At least two factor-matched controls with AIDS or equally matched contact points occur between gp120 and the CCR5 coreceptor: controls with chronic HIV-infection who were receiving highly active antiretroviral therapy (p<0.001). • The second extracellular loop of CCR5 (ECL2) abuts the Since almost all viruses transmitted are R5 variants, these crown of the gp120 hypervariable V3 loop. strains may have a role in preventing infection. In this study, • The CCR5 N-terminal domain binds the V3 stem and the presence of antibodies targeted to CCR5 was associated bridging sheet of gp120. with a lack of progression, and in nine of 20 long-term non- progressors who lost these antibodies over the follow-up In previous work, this group has shown that both SCH-C

period, viral burden increased (>1 log10 rise in HIV RNA) and and TAK-779 bind the transmembrane domain of CCR5, in a

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 73 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 74

CLINICAL REVIEWS

pocket distinct from the gp120 binding site, and non- are now beginning to reach the clinic. It has previously been competitively inhibit gp120 binding to CCR5 [1,2]. Monoclonal speculated that they might act in tandem to reduce the antibodies (MAbs) to ECL2, such as PA14, are potent inhibitors ability of HIV to enter target cells effectively. One of the of viral entry without blocking gp120 binding. agents currently in development is TNX-355, a humanized In this study, the investigators documented small- monoclonal antibody, which targets a unique epitope in CD4

molecule inhibitor (SMI) and MAb IC50 values to HIV type 1 that is involved in a conformational change allowing virus (HIV-1) reporter viruses pseudotyped with envelope entry after CD4 binding. glycoprotein from 13 different primary CCR5-tropic (R5) In this report, a series of experiments were conducted using viruses. In contrast to a narrow six-fold range in sensitivity to activated peripheral blood mononuclear cells from HIV- PA14, sensitivity to SCH-C and TAK-779 spanned four orders negative individuals, exposed to six strains of HIV and one of of magnitude. Safarian and colleagues then showed that human T-cell leukemia virus-III, to examine the effect of single modifications in the V3 crown determine sensitivity to SMIs, or continuous exposure to TNX-355 at varying concentrations that the greater sensitivity to both SMIs is due to reduced of enfuvirtide. Synergistic antiretroviral activity was seen across binding of the gp120 V3 crown with the coreceptor, and that all experiments with TNX-355 and enfuvirtide, and appeared this is linked to features of the V3 crown. greatest when both agents were present throughout the Furthermore, synergy studies demonstrated that culture period. The fact that there are now three tightly linked combinations of either SMI with PA14 synergize to inhibit targets around viral entry, all with candidate drugs, might allow envelope-mediated entry. Finally, an assay was established a three-way antiviral synergy against this staged process by to determine the time course of viral inhibition of HIV-1 which the virus gains a foothold in the cell. entry. The time required to achieve half-maximal inhibition Address for reprints: RT Schooley, University of California, San Diego, with SCH-C was 9 min and with PA14 32 min, compared San Diego, CA 92093, USA. Email: [email protected] with 46 min for the gp41-membrane fusion-inhibitor enfuvirtide. Together, these data indicate significant differences in the mechanism of action for SMIs and PA14, Drug Resistance and Susceptibility and thus, excellent potential as combination HIV chemotherapy warranting further study. Characterization of envelope glycoprotein gp41 1. Dragic T, Trkola A, Thompson DA et al. A binding pocket for a small molecule inhibitor of genotype and phenotypic susceptibility to HIV-1 entry within the transmembrane helices of CCR5. Proc Natl Acad Sci USA 2000;97:5639–44. enfuvirtide at baseline and on treatment in the 2. Tsamis F, Gavrilov S, Kajumo F et al. Analysis of the mechanism by which the small- Phase III clinical trials TORO-1 and TORO-2 molecule CCR5 antagonists SCH-351125 and SCH-350581 inhibit human immunodeficiency virus type 1 entry. J Virol 2003;77:5201–8. Melby T, Sista P, DeMasi R et al.

Address for reprints: T Dragic, Department of Microbiology and AIDS Res Hum Retroviruses 2006;22:375–85. Immunology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461, USA. Email: [email protected] This article presents further data on the genotype and phenotype of viruses isolated at baseline and after 48 in vitro Synergistic antiretroviral activity of a weeks of enfuvirtide therapy within the TORO (T-20 humanized monoclonal anti-CD4 antibody versus Optimized Regimen Only) studies. Since (TNX-355) and enfuvirtide (T-20) enfuvirtide resistance has been seen both in vitro and Zhang XQ, Sorensen M, Fung M et al. in vivo, where it develops rapidly in the face of ongoing Antimicrob Agents Chemother 2006;50:2231–3. viral replication, the identification of amino acid changes resulting in loss of susceptibility is clinically important. A number of new agents are currently being investigated for blocking entry of HIV type 1 into the host cell. In this Enfuvirtide works by competitively inhibiting interactions study, synergy was demonstrated in vitro between a between the heptad repeat (HR) 1 and HR-2 domains of humanized monoclonal antibody against an epitope of gp41; thus preventing the reconfiguration that allows the CD4 (TNX-355) and the fusion inhibitor enfuvirtide. virus and cell surfaces to meet and fusion to occur. Changes The fact that this dual attack on the entry cascade in the conserved sequence of gp41 at positions 36–45 may gives increased benefit may help to design successful result in loss of enfuvirtide activity. combinations of agents that act on the entry process. The TORO trials demonstrated a wide range of baseline

susceptibilities with a 3 log10 difference across the patients Drugs with activity against the HIV-receptor, CD4, and included. Few polymorphisms in this region were identified, coreceptors, CCR5 and CXCR4, as well as fusion inhibitors, the most common, seen in 16% of subjects, was at

74 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 75

ANTIRETROVIRAL THERAPY

position 42. However, changes in the 36–45 stretch of At day 11, two subjects receiving maraviroc 100 mg once daily amino acid residues were observed in 92.7% of the patients had CXCR4-using (X4) virus. The baseline CD4+ cell counts who lost virological control during the studies, and in and viral load reductions in these subjects were 98.8% of those with a greater than four-fold reduction in indistinguishable from the rest of the study group. Clonal susceptibility to enfuvirtide. analysis by sequencing of gp160 revealed that a minority While coreceptor tropism and viral subtype had no effect population of dual-tropic virus existed at baseline in one on viral response – although there were limited numbers of subject. Phylogenetic analysis of each subject showed that the non-B subtypes in these studies – the 3.5% of viruses using X4 virus was genetically distinct from the R5 clones, and thus

the CXCR4 receptor showed higher IC50 values (the emerged from pre-existing viral reservoirs rather than resulted concentration required for 50% inhibition of viral replication) from mutations in the R5 isolates. A third subject was than CCR5-using or dual-tropic viruses. The presence of the incorrectly enrolled into the trial with a mixed-tropic virus on N42S polymorphism had an independent association with day 1 and experienced no drop in viral load following 10 days’

lower baseline IC50; however, it was not possible to follow treatment with maraviroc 100 mg twice daily. This subject had changes in susceptibility over time in this cross-sectional study. persistent X4 viral clones throughout the study period that Taken together with the results of previous studies, those were genetically distinct from the CCR5-using isolates. In all shown here attest to the relative fragility of enfuvirtide when three subjects, circulating virus reverted to predominantly given alone and argue for its use with an effective backbone CCR5-using variants following discontinuation of maraviroc. regimen of drugs to which the patient remains sensitive. These results suggest that the development of CXCR4- With the impending availability of CCR5 receptor blockers using virus is uncommon following short-term maraviroc and integrase inhibitors, the future seems rosy indeed. monotherapy, and occurs as a result of pre-existing viral reservoirs. Whether or not pre-existing CXC4-using variants Address for reprints: ML Greenberg, Trimeris Inc., 3500 Paramount Parkway, Morrisville, NC 27560, USA. Email: [email protected] will evolve under pressure from highly active antiretroviral therapy containing multiple agents in combination with a CCR5 antagonist has yet to be determined. However, based Viral Variants on the findings of this study, the screening for viral tropism prior to use of CCR5 inhibitors is warranted. Emergence of CXCR4-using human Address for reprints: M Westby, Pfizer Ltd., PGRD, Sandwich Labs, Kent immunodeficiency virus type 1 (HIV-1) variants in CT13 9NJ, UK. Email: [email protected] a minority of HIV-1-infected patients following treatment with the CCR5 antagonist maraviroc is Antiretroviral Therapy from a pretreatment CXCR4- using virus reservoir Westby M, Lewis M, Whitcomb J et al. J Virol 2006;80:4909–20. A pharmacokinetic–pharmacodynamic model to optimize the Phase IIa development program The current study describes the coreceptor tropisms of maraviroc observed before, during, and after 10 days’ treatment Rosario MC, Poland B, Sullivan J et al. with the CCR5 antagonist maraviroc. The emergence J Acquir Immune Defic Syndr 2006;42:183–91. of CXCR4-using virus was rare and was the result of pre-existing variants that reverted to predominantly CCR5- The results of this modeling study indicate that it is using virus following discontinuation of maraviroc therapy. possible to accelerate drug development by replacing some arms or trials using simulations based on data from The development of CCR5 receptor antagonists has been existing studies. In this report, the development plan for complicated by concern over the emergence of CXCR4- the CCR5 inhibitor maraviroc was assisted by results from using (X4) viral variants with the potential for more the first monotherapy study in HIV patients to plan the accelerated disease progression, which could arise from doses to be studied thereafter. selective pressure during treatment with these agents. The current publication describes 10-day monotherapy, The authors of this article examine the use of a modeling dose-escalation studies with the CCR5 antagonist maraviroc, approach to the pharmacokinetics and pharmacodynamics and the viral tropism changes before, during, and after of the CCR5 receptor blocker maraviroc (formerly UK- therapy. Following treatment, the virus remained CCR5-tropic 427,857). Clinical data were used from a monotherapy (R5 virus) in 97% (62 of 64) of the assessed patients. study (A4001007), in which 44 HIV type 1 (HIV-1)-infected

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 75 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 76

CLINICAL REVIEWS

subjects received maraviroc, given without food, at a dose of chemotaxis. The sole ligand for this receptor is stromal cell- 25 mg/day, or 50, 100, or 300 mg twice daily, or placebo, derived factor-1 (SDF-1, also called SDF-1α, or CXCL12), a for 10 days. chemotactic cytokine that is constitutively produced in Using the day-10 data, HIV-1 RNA changes and viral numerous tissues. This receptor–cytokine axis has important susceptibilities of the individuals over the study period were roles in diverse aspects of the body including T lymphocyte incorporated in a mixed effects-modeling approach to functions (lymphopoiesis, maturation, trafficking to bone develop a pharmacokinetic–pharmacodynamic model. The marrow throughout the body), B cell growth and parameters from this model were then used to calculate a development, vascularization of the gastrointestinal tract, variety of outcomes, including average viral inhibition and normal cerebellar development. AMD3100 is a specific fraction and decay rate of actively infected cells. These data inhibitor of CXCR4 that has been widely studied, not only to were entered into a Monte Carlo simulation, which validate CXCR4 as a target for HIV therapy, but also for predicted the viral load changes over a number of doses to mobilization of hematopoietic stem cells in patients with be examined in a future study, A4001015. multiple myeloma and non-Hodgkin’s lymphoma. Animal The advantage of a model is that it can be updated using studies indicate that CXCR4 blockade with AMD3100 may new clinical data when available as well as data from the have beneficial effects in inflammatory diseases, including literature. The model predicted that inhibition of the virus rheumatoid arthritis and asthma. would range from 0.15–0.38 for the 25-mg once-daily dose, Fricker and colleagues provide detailed data character- to 0.88–0.96 for the 300-mg twice-daily dose. The mean izing the specificity of action of AMD3100 receptor binding, decline rate of HIV-1 RNA was similar or greater than and the lack of agonist effects following receptor binding, published data for monotherapies of other antiretrovirals, and of cross-reactive binding with cells expressing CXCR3, but occurred more slowly than in patients receiving CCR1, CCR2b, CCR4, CCR5, and CCR7. The article combination therapy. This has been reported as a good demonstrates that AMD3100 binding to CXCR4 is tight and predictor of subsequent clinical outcomes. slowly reversible, without allowing: There was good agreement between the predicted and observed data, although the effect of food on maraviroc levels • SDF-1 binding. (approximately 50% reduction) was overestimated; the model • SDF-1-mediated calcium flux. predicted a 14% reduction in viral load benefit, whereas • SDF-1-stimulated chemotaxis. only an 8% reduction was recorded in the subsequent study. It may be that this discrepancy is due to a lack of The use of AMD3100 for inhibition of HIV has been pharmacokinetic data on this fed dose for use in the model. limited by the requirement for continuous intravenous Overall, however, this simulation approach helped to reduce infusion, dose-related side effects including gastrointestinal unnecessary evaluation of particular doses of maraviroc. symptoms, paresthesias, and premature ventricular

Address for reprints: MC Rosario, Department of Clinical Pharmacology, contractions. The authors note that AMD3100 does not Pfizer Clinical Research and Development, Eastern Point Road, Groton, result in significant declines in HIV-viral load, despite specific CT 06340, USA. Email: [email protected] activity against CXCR4-tropic virus. The report cites recent clinical trials in multiple myeloma Characterization of the molecular pharmacology of and non-Hodgkin’s lymphoma, which indicate that AMD3100: a specific antagonist of the G-protein AMD3100 expands myeloid cells in the bone marrow, coupled chemokine receptor, CXCR4 resulting in release of stem cells into the circulation. These Fricker SP, Anastassov V, Cox J et al. studies suggest that AMD3100 may be safe and effective in Biochem Pharm 2006;72:588–96. improving stem cell mobilization when combined with granulocyte colony-stimulating factor (G-CSF), as compared This article confirms that AMD3100 binds to CXCR4, with the current use of G-CSF alone. It remains unclear inhibiting binding of the receptor’s natural ligand, stromal whether the clinical issues raised with AMD3100 in HIV cell-derived factor-1 (SDF-1; CXCL12), chemotaxis, and patients were unique to the agent used, or a consequence of SDF-1-mediated calcium flux. No inhibition of calcium successful CXCR4 blockade. The presence of the CXCR4 flux was observed in cells expressing CXCR3, CCR1, receptor system in diverse tissues warrants particularly CCR2b, CCR4, CCR5, or CCR7. careful assessment of short- and long-term safety of AMD3100, and any other agent that interferes with CXCR4.

The chemokine receptor CXCR4 is present on many different Address for reprints: SP Fricker, AnorMED Inc., #200 20353 64th cell types and plays an important role in leukocyte Avenue, Langley, BC V2Y 1N5, Canada. Email: [email protected]

76 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 77

The 16th International AIDS Conference MEETING REPORT (AIDS 2006)

Toronto, ON, Canada, 13–18 August, 2006

Mike Youle Royal Free Centre for HIV Medicine, Royal Free Hospital, London, UK

Earlier this summer I gave a series of lectures in Canada, one the government’s AIDS policy, more infections occur daily of which was in the same hotel in Vancouver where almost than patients enter into HAART programs. exactly 10 years earlier we heard the first seriously good Although this meeting was not primarily scientific, there results from protease inhibitor (PI)-based highly active was a selection of data for many of the new compounds in antiretroviral therapy (HAART). The treatment of HIV/AIDS development (Table 1). has changed dramatically since then and now includes better- tolerated drugs and formulations, fewer pills per regimen, and Entry inhibitors more once-daily dosing. In my cohort, the last death from TNX-355 AIDS in a patient with no therapy options left happened in Thomas Duensing (Tanox, Inc., Houston, TX, USA) presented May 2000; the major cause of HIV disease and death in the new 48-week data from a randomized, double-blind, UK is late presentation. This, of course, is not the case for the placebo-controlled study of triple-class-experienced patients majority of people in the world, and the 16th International treated with the novel entry inhibitor TNX-355 [1]. TNX-355, AIDS Conference Conference reflected this in the prominence a monoclonal antibody that prevents the conformational rightly given to access and empowerment issues, such as the change in gp120 needed to expose the coreceptor binding plight of the South African population, where, as a result of site, is given as a weekly intravenous infusion. Eighty-two

Table 1. Antiretroviral agents discussed at AIDS 2006 (some of these drugs are still under development).

Drug Company Class Abstract number TNX-355 Tanox, Inc. Entry I TUPE0058, THLB0218 Maraviroc Pfizer CCR5 I THLB0215 Vicriviroc Schering-Plough CCR5 I TUPE0074, THLB0217 TRI-999, TRI-1144 Trimeris, Inc./Roche Fusion I THPE0021, THAA0303 Fosalvudine Heidelberg Pharma NRTI THPE0025, THLB0216 Etravirine (formerly TMC125) Tibotec, Inc. NNRTI TUPE0061, TUPE0082, TUPE0084, TUPE0086, THPE0136 Rilpivirine (formerly TMC278) Tibotec, Inc. NNRTI TUPE0087 GS9137 Gilead Sciences Integrase I TUPE0080, TUPE0088 MK-0518 Merck & Co. Integrase I THPE0020, THPE0027, THLB0214, THAA0302 PL-100 Merck Pharmaceuticals PI THAA0304 Brecanavir GlaxoSmithKline PI TUAB0105, THPE0023 Darunavir (formerly TMC114) Tibotec PI TUAB0104, TUPE0060, TUPE0062, TUPE0063, TUPE0069, TUPE0078, TUPE0083, TUPE0086, THPE0136 I: inhibitor; NNRTI: non-nucleoside reverse transcriptase inhibitor; NRTI: nucleoside reverse transcriptase inhibitor; PI: protease inhibitor

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 77 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 78

MIKE YOULE

Figure 1. The viral load declines in a study assessing TNX-355: 48-week efficacy results.

15 mg/kg Placebo 10 mg/kg Placebo 0 0

–0.14 –0.14

0.57 –0.5 –0.5

copies/mL) copies/mL) 0.82 10 10 –0.71

–1.0 –1.0 –0.96 at week 48 (log at week 48 (log

Mean HIV RNA reduction from baseline from Mean HIV RNA reduction –1.5 p<0.010 baseline from Mean HIV RNA reduction –1.5 p<0.001

Last observation carried forward analysis All arms included an optimzed background regimen

patients with CD4+ T cell counts >50 cells/mm3 and viral loads TNX-355. This suggests that cross-resistance between these >10 000 copies/mL were randomized to one of the following two drugs may not represent a major problem. three groups: Vicriviroc • TNX-355 10 mg weekly for nine doses, given every On a poster by Angela Sanone (Schering-Plough Research 2 weeks thereafter. Organization, Kenilworth, NJ, USA) the effect of the CCR5 • TNX-355 15 mg twice every 2 weeks. receptor antagonist vicriviroc in combination with a range • An optimized background regimen (OBR). of ritonavir-boosted PIs (atazanavir, saquinavir, indinavir, fosamprenavir, and tipranavir) and unboosted nelfinavir was Virological failures were rolled over onto TNX-355 plus evaluated in a pharmacokinetic study in healthy volunteers a new OBR. The viral load declines were approximately [4]. It appears that the addition of any of these agents has

0.2 log10 less than the equivalent 24-week data shown a negligible effect on the key vicriviroc pharmacokinetic previously and data on a withdrawn poster (Fig. 1) [2] parameters, and no significant adverse events were (D Norris, Comprehensive Research Center, Tampa, FL, USA). reported. Clearly, additional data with boosted lopinavir The mean increase in CD4+ T cell count was 51 cells/mm3 for and darunavir (formerly TMC114) are required to complete the 15-mg group and 48 cells/mm3 for the 10-mg group, the evaluation. compared with only 1 cell/mm3 for the placebo arm (p<0.05) After the disappointment of the earlier vicriviroc studies, The time to loss of virological response was similar in the it was reassuring to hear the positive news given by Roy two treatment arms at 8 months (Fig. 2), but no data were Gulick (Cornell University, New York, NY, USA) in a late- presented on the outcomes of patients who subsequently breaker presentation on the ACTG 5211 study [5]. Three received open-label drug. No serious adverse events (SAEs) doses of vicriviroc were assessed against placebo in a highly were attributed to TNX-355, and the mild side effects treatment-experienced patient population with CCR5-tropic reported were similar to those in the OBR group. (R5) virus. One hundred and eighteen subjects (median viral A poster by S Weinheimer (Tanox, Inc.) provided further load 36 380 HIV RNA copies/mL and CD4+ T cell count data on the activity of this agent, in this case against virus 146 cells/mm3) were randomized to receive vicriviroc (5, 10, with enfurvitide-resistance substitutions (G36D, V38A, and or 15 mg/day) or placebo (Fig. 3). Viral load declines were N43D) generated through site-directed mutagenesis [3]. significantly greater in all three vicriviroc arms compared Compared with wild-type virus, HIV envelopes with these with placebo at day 14 and week 24 (p<0.01, intention-to- mutations showed an 11–32-fold reduced sensitivity to treat analysis) (Table 2; Fig. 4). Although not statistically enfuvirtide but less than a two-fold reduced susceptibility to significant, the lower dose of 5 mg/day resulted in a greater

78 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 79

THE 16TH INTERNATIONAL AIDS CONFERENCE

Figure 2. Time to loss of virological response in a study Figure 3. Design of study ACTG 5211 assessing therapy assessing TNX-355. with vicriviroc in treatment-experienced patients with CCR5-tropic virus.

300 p=0.003 48 weeks 253 p=0.003 24 weeks 250 230 Primary analysis Placebo + OBR 200 Vicriviroc 5 mg once daily + OBR 150 Screening Days Vicriviroc 10 mg once daily + OBR 100 Vicriviroc 15 mg once daily + OBR

50 First 14 days vicriviroc or placebo only. OBR added on day 14. 0 G Triple-class experienced 0 G 15 mg/kg 10 mg/kg Placebo VL >1000 copies/mL (median 4.56 log10 copies/mL) G Any CD4+ cell count (median 146/mm3) Last observation carried forward analysis G NB. Some samples screened R5 were X4/R5 at baseline) All arms included an optimzed background regimen

OBR: optimized background regimen.

Table 2. Change in viral load and CD4+ cell counts at day 14 and week 24 in study ACTG 5211 assessing treatment with vicriviroc in patients with CCR5-using virus.

Vicriviroc 5 mg Vicriviroc 10 mg Vicriviroc 15 mg OBR once daily+OBR once daily+OBR once daily+OBR (n=30) (n=30) (n=30) (n=28)

Change in HIV-1 RNA (log10 copies/mL) –0.87 –1.15 –0.92 +0.06 at day 14 (mean)

Change in HIV-1 RNA (log10 copies/mL) –1.51 –1.86 –1.68 –0.29 at week 24 (mean) Change in CD4+ cell count (mm3) at week 24 +84 +142 +142 –9 % change in coreceptor use to X4/R5 only 27% 10% 7% 4%

OBR: optimized background regimen.

number of patients with virological failure and emergence of When the response was analyzed according to baseline CXCR4-using (X4) virus. Thirty-three percent of participants tropism, good suppression was seen even in the X4/R5- had previous enfuvirtide experience, 92% were male, 66% tropic populations. Caucasian, 20% black, and 12% Hispanic. At study entry, 86% of patients harbored R5 virus exclusively and 10% had Maraviroc dual- or mixed-tropic variants. Considering that only a few Howard Mayer (Pfizer Global Research and Development, weeks earlier, tropism screening had shown all patients as New London, CT, USA) discussed the 24-week results from having R5 virus, this reveals some of the limitations of the the A4001029 study, in the only talk to present new data on current technology to exactly determine the tropism of a the CCR5 inhibitor maraviroc [6]. A4001029 is an ongoing, patient’s virus. However, emerging data that mixed- or dual- double-blind, placebo-controlled, exploratory Phase IIB trial tropic virus responds to CCR5 receptor blockers may make in subjects with triple drug-class failure, >5000 HIV RNA the use of screening assays less vital. copies/mL, and non-R5 virus – a patient population in which

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 79 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 80

MIKE YOULE

Figure 4. Change in viral load at day 14 and week 24 in study ACTG 5211 assessing therapy with vicriviroc in patients with CCR5-using virus.

0.5 Week 2 Week 24 5 mg 10 mg 15 mg Placebo copies/mL)

10 0

–0.5

–1

–1.5

–2 Mean HIV RNA change from baseline (log Mean HIV RNA change from Viral load response at week 24 by baseline tropism (10 and 15 mg doses) G R5 tropic (n=71) –1.83 log10 copies/mL G X4/R5 tropic (n=10) –0.77 log10 copies/mL; (p=0.01)

one may expect the drug to have only limited benefit. A (Morrisville, NC, USA) and Roche (Nutley, NJ, USA), a total of 190 subjects were randomized to maraviroc (150 mg collaboration that has already produced enfuvirtide. In a once or twice daily) or placebo, all with OBRs. Patient poster presentation by DK Davidson (Trimeris, Inc.), the ability characteristics at baseline are shown in Table 3; tropism of these new compounds to inhibit virus isolates resistant to testing revealed that most subjects had dual- or mixed-tropic enfuvirtide and the previous lead compound, T-1249, was virus. Approximately 50% of the patients received concurrent examined [7]. Through in vitro selection, a series of mutations enfuvirtide therapy, 60% in each group had between two developed in the heptad repeat 1 (HR1) domain, and, less and four active drugs in their OBR, baseline CD4+ T cell frequently, in HR2 of gp41. While all enfuvirtide-resistant count was approximately 95 cells/mm3, and viral load was isolates demonstrated a <10-fold loss of susceptibility to

5.0 log10 copies/mL. As would be expected, mean viral load T-1249, mutations arising through selection with T-1249 were changes did not differ greatly between the groups (Table 4). highly resistant to enfuvirtide. When these isolates were However, what was surprising was that CD4+ T cell increases tested against either TRI-999 or TRI-1144, they retained were greater in the maraviroc-treated patients than in those sensitivity to these new agents. Both required a greater who received placebo; this reached the lowest level of number of mutations to acquire resistance. Furthermore, virus statistical significance when all randomized subjects were breakthrough was seen at concentrations 30-fold lower than included; Table 5). In addition, the use of enfuvirtide in the those required of enfuvirtide or T-1249, suggesting a much- OBR pointed towards a trend of a greater than additive enhanced genetic barrier against resistance development. benefit. This raises the question of whether maraviroc may Mary Delmedico (Trimeris, Inc.) then showed exciting actually be useful in a broader range of patients than was data on advances towards a once-weekly formulation of the first thought. No drug-specific toxicity was noted; in fact, two lead compounds, administered by subcutaneous injection a greater number of grade 3 and 4 abnormalities in liver [8]. The researchers’ aim was to develop formulations that function were recorded in the placebo group. require a minimum dose with maximum drug exposure. Using a rat model, they first tried a peptide–organic salt New-generation fusion inhibitors complex. For TRI-1144, this resulted in 13% bioavailability, Two new fusion inhibitors, TRI-999 and TRI-1144, are being but a long half-life and an improved area under the curve developed through a joint venture between Trimeris, Inc. (AUC) when the release rate of the compound was increased

80 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 81

THE 16TH INTERNATIONAL AIDS CONFERENCE

Table 3. Baseline patient characteristics of the study population in the maraviroc Phase IIB trial, A4001029.

Randomized (190) Treated (186)

Placebo+OBR Maraviroc once Maraviroc twice (n=62) daily+OBR (n=63) daily+OBR (n=61)

Change in HIV-1 RNA (log10 copies/mL) –0.87 –1.15 –0.92+0.06 Mean age (years), (range) 44.6 (23–65) 42.7 (16–59) 42.5 (16–62) Female, n (%) 9 (14.5) 10 (15.9) 6 (9.8) Race, n (%) – White 40 (64.5) 46 (73.0) 44 (72.1) – Black 18 (29.0) 17 (27.0) 13 (21.3) – Other 4 (6.5) 0 (0) 4 (6.6) Tropism, n – X4 2 2 4 – R5 0 1 0 – Not phenotyped, or not reported 2 3 5 – Dual/mixed-tropic 58 57 52 OBR: optimized background regimen (three to six antiretroviral drugs).

Figure 5. Rat pharmacokinetic model assessing area under Figure 6. Rabbit pharmacokinetic model assessing area under curve for an organic salt complex of TRI-1144. curve for a modified gel formulation of TRI-999.

100 10 Immediate-release formulation

10 Immediate-release formulation

1 1 Release rate increased Peptide gel formulation 0.1 0.1 Plasma concentration (µg/mL) Plasma concentration (µg/mL) 0.01 Rat subcutaneous injection Rat subcutaneous injection normalized to 3 mg/kg normalized to 3 mg/kg 0.001 0.01 01234567 0 1234567 Time (days) Time (days)

(Fig. 5). However, when a modified gel formulation of the formulation, Dr Delmedico said that its pharmacokinetic other lead compound, TRI-999, was assessed in a rabbit profile did not lend itself to this and no work in this direction model, bioavailability was increased to near 100% and the is planned. This makes the rapid development of these two half-life greatly extended (Fig. 6). These data are exciting, new compounds all the more vital. as they demonstrate that a once-weekly injectable fusion inhibitor is achievable. This is a big leap towards improved Nucleoside reverse transcriptase inhibitors clinical use of parenteral compounds that appear to overcome Fosalvudine tidoxil many of the current problems with enfuvirtide. When asked Pedro Cahn (Fundacion Huesped, Buenos Aires, Argentina) if enfuvirtide was a candidate for an extended-release showed data from a Phase I study conducted in Argentina

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 81 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 82

MIKE YOULE

Table 4. Efficacy results in the maraviroc Phase IIB trial, A4001029.

All treated patients with Placebo+OBR Maraviroc once daily Maraviroc twice daily dual/mixed-tropic HIV-1 (n= 58) +OBR (n=57) +OBR (n=52)

Mean decrease in HIV-1 RNA (log10 copies/mL)* –0.97 –0.91 –1.20 Treatment difference (maraviroc–OBR) in HIV-1 +0.06 –0.23

RNA decrease (log10 copies/mL) (97.5% CI) (–0.53,+0.64) (–0.83,+0.36)

HIV RNA <400 copies/mL (%) 24.1 24.6 30.8 HIV RNA <50 copies/mL (%) 15.5 21.1 26.9 Mean decrease in HIV-1 RNA in patients using –0.89 –1.26 –1.44

enfuvirtide**(log10 copies/mL)

*Primary end point. **Last observation carried forward analysis. OBR: optimized background regimen.

Table 5. Changes in CD4+ cell count in patients with dual/mixed-tropic HIV type 1 at screening in the maraviroc Phase IIB trial, A4001029.

CD4+ cell change from baseline Placebo+OBR Maraviroc once daily Maraviroc twice daily baseline (cells/mm3, mean) +OBR +OBR All treated patients with dual/mixed-tropic HIV-1 +36* +60 +62 (at 24 weeks, last observation carried forward) (n=58) (n=57) (n=52) Patients with only X4-tropic HIV-1 detectable -104 +48 +33 at time of virologic failure (n=2) (n=12) (n=12) *Data for patient 4 is missing.

on the safety, tolerability, and pharmacokinetics of Non-NRTIs fosalvudine tidoxil in HIV patients [9]. This agent is a Etravirine prodrug of the nucleoside reverse transcriptase inhibitor Data on non-NRTIs (NNRTIs) at the conference exclusively (NRTI) alovudine, and has a narrow therapeutic window. In concerned etravirine (formerly TMC125) and rilpivirine a previous study, 4 weeks’ treatment with alovudine (7.5 mg (formerly TMC278) from Tibotec, Inc. (Yardley, PA, USA).

once daily) resulted in a median reduction of 1.88 log10 The only new information on the latter agent was a healthy in HIV RNA levels in patients with multiple thymidine- volunteer study (n=16) that examined the interaction with associated mutations [10]. Another early trial of alovudine ketoconazole, an inhibitor of the liver enzyme cytochrome was stopped after a case of liver failure. At the end of the P3A4 (CYP3A4) [12] (Rolf van Heeswijk, Tibotec BVBA). session, the investigator of that particular study, Charles Rilpivirine is a substrate for CYP3A4, and concomitant Flexner (John Hopkins University School of Medicine, administration with ketoconazole increased exposure to

Baltimore, MD, USA), commented that this was an rilpivirine by 49% (AUC24h), probably by inhibiting its idiosyncratic hepatic necrosis, and went on to advise metabolism in the liver. The results suggest that the investigators of this compound to be very careful. After the combination of rilpivirine and inhibitors of cellular experience with lodenosine, which caused high rates of liver cytochromes may require dose modifications. changes, the present author agrees. Nevertheless, the Two pharmacokinetic studies of etravirine, which is present study in 24 healthy volunteers showed that further along in development than rilpivirine, were presented fosalvudine tidoxil has a half life of 5–7 h at single doses of by M Schöller-Gyüre, (Tibotec BVBA). In the first of these, 5, 10, 20 (molar equivalent of 7.5 mg alovudine), and 40 mg. there seemed to be no significant interaction between this The prodrug is thought to be safer than the parent compound, drug and methadone in healthy, HIV-negative volunteers as animal and in vitro studies have shown that it has who were stabilized on methadone therapy [13]. In a 14-day minimal bone marrow distribution and high protein-binding study of 16 individuals, no clinically relevant changes in drug properties [11] (Frank Reuss, Heidelberg Pharma GmbH, handling of either agent were seen and no significant Ladenburg, Germany), and, apart from headaches, no symptoms of methadone withdrawal (which can be a adverse events were reported in the present study. problem with the currently licensed NNRTIs) were observed.

82 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 83

THE 16TH INTERNATIONAL AIDS CONFERENCE

Figure 7. Change in viral load at 48 weeks in the TMC125-C223 trial assessing therapy with etravirine.

0.5 Active control (n=40) 400 mg twice daily (n=80) 800 mg twice daily (n=79) 0 –0.14

–0.5

–0.88, p=0.018 –1.0 –1.01, p=0.002 viral load (± standard error) viral load (± standard 10

–1.5

–2.0

Mean change in log –2.5 024 8 12 16 20 24 32 40 48 Time (weeks) No statistical difference between the TMC125 groups was observed. p values versus active control.

A second bioavailability study of etravirine in 18 healthy, Following pharmacokinetic data presented by Marta HIV-negative volunteers assessed co-administration of the Boffito (Chelsea and Westminster Hospital, London, UK)

agent with the H2 antagonist ranitidine and the proton- [16] in London last year, which defined the optimal doses pump inhibitor omeprazole [14]. Drug–drug interactions of the combination of darunavir/ritonavir and etravirine as with medications used for ulcer pain and gastritis have been 600 mg/100 mg + 200 mg, respectively, twice daily, this problematic for some other HIV drugs, notably, unboosted combination has now entered into two large Phase III atazanavir. However, neither ranitidine nor omeprazole had studies (DUET 1 and 2) for safety and efficacy assessment. any clinically relevant effect on etravirine in this study, Thomas Kakuda (Tibotec, Inc.) and co-workers reported although etravirine exposure was increased by 41% by a formal two-way crossover trial to further assess concomitant omeprazole, possibly as a result of the pharmacokinetic interactions in 32 HIV-negative volunteers, inhibition of CYP2C19. randomized to two groups [17]. All volunteers received Clinical data on the TMC125-C223 trial were discussed by etravirine 100 mg twice daily for 8 days. After a washout Cal Cohen (Community Research Initiative of New England, period of 14 days, they were given darunavir/ritonavir Boston, MA, USA). A total of 199 subjects with documented 600 mg/100 mg twice daily for 16 days. From day 9 to 16, NNRTI resistance were randomized to etravirine (400 mg or group A received etravirine 100 mg twice daily, while group 800 mg, twice daily), or a standard-of-care control regimen B received 200 mg twice daily. Concomitant administration [15]. Baseline resistance analysis showed a mean of two NNRTI of etravirine and darunavir/ritonavir was found to have no mutations and a phenotypic mean fold change of only 1.7 for clinically significant effect on the pharmacokinetics of darunavir etravirine, compared with 41 for efavirenz and 61 for or ritonavir. However, the etravirine 100 mg twice-daily

nevirapine. At week 48, the viral load change was –0.88 log10 exposure was decreased by 37%, with similar decreases in

and –1.04 log10 for the two etravirine arms, compared with Cmax and Cmin (32% and 49%, respectively). Pharmacokinetic –0.14 log for the control arm, in which 78% of patients had parameters for etravirine observed after co-administration of withdrawn due to virological failure (Fig. 7). When the etravirine 200 mg twice daily with darunavir/ritonavir was response was assessed against baseline resistance to NNRTIs, slightly lower than the historical control for healthy

the number of mutations predicted the viral response (Fig. 8). volunteers given the same dose and formulation (AUC12h

However, even with three mutations, the response to etravirine 7638±2254; Cmax 876±2336). Overall, the safety profiles for remained greater than the active control. etravirine, darunavir/ritonavir, and their co-administration

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 83 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 84

MIKE YOULE

Figure 8. Assessment of the effect of baseline resistance to NNRTIs on virological response in the TMC125-C223 trial at week 48.

Etravirine Active Baseline NNRTI mutations in etravirine 800 mg twice daily 800 mg twice daily control 0* 1 2 ≥3 0 n=79 n=40 n=14 n=19 n=16 n=30 –0.14

–0.5 –0.54

–1.0 –0.9

viral load (± standard error) viral load (± standard –1.01 10

–1.5 –1.38

–1.67

–2.0 Mean change in log Patients discontinuing the trial for any reason had their viral load response imputed as no change from baseline *All patients had NNRTI mutations from prior genotyping

NNRTI: non-nucleoside reverse transcriptase inhibitor.

were similar. Apart from a rash, which was seen more Figure 9. Design of the Protocol 004 study assessing therapy frequently during co-administration, use of etravirine with with MK-0518 in antiretroviral-naïve subjects. darunavir/ritonavir in healthy volunteers was generally safe and well tolerated. 48 weeks A number of groups around the world have early data 24 weeks on compassionate-use groups receiving the two drugs Interim analysis simultaneously. One of the more clinically relevant studies for Efavirenz 600 mg once daily + TDF/FTC salvage therapy was a poster by Julio Montaner (British Columbia Centre for Excellence in HIV/AIDS, Vancouver, BC, MK 518 100 mg twice daily + TDF/FTC Canada) on the outcomes of combining darunavir/ritonavir with etravirine [18]. In four heavily pre-treated subjects Screening MK 518 200 mg twice daily + TDF/FTC administered this combination, three of whom also received

enfuvirtide as a new drug, a drop of 2 log10 in HIV RNA was MK 518 300 mg twice daily + TDF/FTC recorded. No adverse events were reported. These results are similar to data obtained by the present author from patients MK 518 400 mg twice daily + TDF/FTC in London, UK (unpublished data). This triple regimen offers First 10 days MK518 or placebo only a reasonable new HAART for patients who cannot wait for in eight of 38 patients/group novel classes such as integrase inhibitors to become available. G Therapy blinded throughout G Viral load 5000 HIV RNA copies/mL

Integrase inhibitors (mean 4.6 –4.8 log10 copies/mL) MK-0518 G CD4+ cell count ≥100 cells/mm3 MK-0518, the first integrase inhibitor developed by Merck TDF/FTC: tenofovir/emtricitabine. & Co., Inc. (Whitehouse Station, NJ, USA), is an agent that prevents HIV from incorporating its genetic information into that of the host cell, thereby preventing the establishment of Because of its novel mode of action within the nucleus of the infection in T cells. It has shown good efficacy in vitro and cell, cross-resistance is not likely to become a problem. acts synergistically with many available antiretroviral drugs. Therefore, integrase inhibitors represent a vital new option for

84 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 85

THE 16TH INTERNATIONAL AIDS CONFERENCE

Figure 10. Virological efficacy in the Protocol 004 study assessing therapy with MK-0518 in antiretroviral-naïve subjects.

120 <400 copies/mL <50 copies/mL 100 100 100 100 94 94 94 93 89 88 86 80

60

Percent of patients Percent 40

20

0 100 mg 200 mg 300 mg 400 mg Efavirenz MK-518

3 Efficacy: All doses showed >2.2 log10 decline in HIV RNA with CD4 increases of 75–135 cells/mm . Tolerability: Similar adverse events in all groups; nausea, headache, and dizziness most common. Infrequent lab adverse events. One discontinuation due to raised LFTs in 600 mg group.

patients with multi-drug resistance [19] (Vincenzo Summa, Exciting new data on MK-0518 were presented by IRBM-Merck Research Laboratories, Pomezia, Rome, Italy). In Martin Markowitz (The Rockefeller University, Aaron addition, favorable pharmacokinetic and metabolic properties Diamond AIDS Research Center, New York, NY, USA) from a suggest that MK-0518 may have good safety and efficacy pivotal study in antiretroviral-naïve subjects [22]. The study profiles in HIV-positive patients [20] (Ralph Laufer, design is shown in Figure 9. Considering that efavirenz MSD/IRBM, Merck, Pomezia, Rome, Italy). Furthermore, and the combination tenofovir/emtricitabine (Truvada, Michael Miller (Merck Research Laboratories, West Point, PA, Gilead Sciences, Inc., Foster City, CA, USA) have proved to USA) provided an update of the biochemical and antiviral be both safe and highly effective in treatment-naïve activity of MK-0518, which charted the history of the patients, it is quite a gamble to compare a new drug with development of this drug over the past 12 years, thanks to these. Thus, it was reassuring to see that MK-0518 the tenacity of Daria Hazuda (Merck Research Laboratories) performed well, especially since the 10-day monotherapy and her team [21]. MK-0518 has been shown to: lead-in could jeopardize the outcome if resistance were to emerge (Fig. 10). The enrolled subjects had a mean age of • Be highly specific for HIV integrase. 36 years, 80% were male, 69% were non-white, and 34% • Be active against HIV-1 and HIV-2. had AIDS. Not only did MK-0518 outperform efavirenz at • Have either additive or synergistic effects with all tested each dose tested, but it also appeared to result in a greater antiretroviral agents. proportion of patients with undetectable virus at an earlier time point (Fig. 11). No data are available on the use of either tipranavir or The tolerability profile of MK-0518 was at least as good darunavir in combination with MK-0518; however, no as that for efavirenz (Table 6), if not better, and so the negative effects/interactions are expected. A member of the longer-term follow-up of this study will be very interesting audience asked whether the drug crosses the blood–brain and could potentially lead to a new option for patients who barrier, but this has not yet been assessed. are naïve to antiviral therapy.

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 85 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 86

MIKE YOULE

Figure 11. Percent of patients with HIV RNA of <50 copies/mL at week 24 in the Protocol 004 study assessing therapy with MK-0518 in antiretroviral-naïve subjects.

100

80

60

40 *

* 20

p<0.001 for MK-0518 at each dose vs. efavirenz Patients with HIV RNA <50 copies/mL (%) 0 024 8 12 16 24 Time (weeks) MK-0518 100 mg 39 39 39 39 39 39 MK-0518 200 mg 40 40 40 40 40 40 MK-0518 400 mg 41 41 41 41 41 41 MK-0518 600 mg 40 40 40 40 40 40 Efavirenz 38 38 83 38 38 37

GS-9137 Table 6. The tolerability profile of MK-0518 in Protocol 004: GS-9137 is the potent HIV integrase inhibitor from Gilead, common (>5%) drug related adverse experiences. which, at a dose of GS-9137/ritonavir 50 mg/100 mg, MK-0518* (all doses) Efavirenz* resulted in a >2.0 log decrease in HIV RNA in a 10-day 10 n=160 (%) n=38 (%) monotherapy study recently published by Edwin De Jesús (Orlando Immunology Center, Orlando, FL, USA) and Nausea 11 13 co-workers [23]. At this conference, two pharmacokinetic Headache 9 24 interaction studies conducted by Brian Kearney (Gilead Dizziness 8 26 Sciences Inc.) and his group assessed whether zidovudine Diarrhea 7 11 or tenofovir/emtricitabine demonstrated meaningful Insomnia 7 11 interactions with this drug [24,25]. Although GS-9137 is Abnormal dreams 6 18 primarily cleared by the cytochromes in the liver, it is also, like zidovudine, excreted renally through glucuronidation, Flatulence 6 – indicating a potential for drug–drug interactions. In a *with tenofovir/lamivudine. Additional adverse events seen at >5% in the efavirenz group: nightmares standard crossover design, 28 subjects were given the (11%), vomiting (8%), malaise (8%), fatigue (5%), disturbance in attention highest planned dose of GS-9137/ritonavir (200 mg/100 mg (5%), lethargy (5%), anxiety (5%). once daily) concomitantly with zidovudine 250 mg (Fig. 12). No clinically significant alteration of either drug level was 27-day dosing period, suggesting that these agents can be apparent, co-administration was safe and well tolerated, and given together with impunity. the half-life of this dose of GS-9137 was 9.1 h, supporting once-daily dosing. In a similarly designed study, a further Protease inhibitors 26 subjects were enrolled to establish whether GS-9137 at PL-100 the same dose can be administered concurrently with Those who went to the late-breaker presentations on new tenofovir/emtricitabine (Fig. 13). No grade 3 or 4 events, drug development were rewarded with news of an exciting and no meaningful interactions were noted during the new PI – PL-100 [26] (Serge Dandache, Ambrilia Biopharma,

86 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 87

THE 16TH INTERNATIONAL AIDS CONFERENCE

Figure 12. Pharmacokinetic interaction studies assessing concomitant administration of the highest planned dose of GS-9137/ritonavir (200 mg/100 mg once daily) and zidovudine. Plasma concentration–time profiles are shown.

GS-9137/ritonavir alone GS-9137/ritonavir + zidovudine 10000

1000 GS-9137 plasma concentration (ng/mL) 100 0 6 12 18 24 Time (h) Zidovudine alone Zidovudine + GS-9137/ritonavir 1000

100

10 Zidovudine plasma concentration (ng/mL) 1 04812 Time (h) Zidovudine alone Zidovudine + GS-9137/ritonavir 10000

1000

100

G-ZDV plasma 10 concentration (ng/mL) 1 0 2 4 6 8 10 12 Time (h) G Data are mean ± standard deviation (n=24) G No clinically relevant drug–drug interaction

Inc., Verdun, QC, Canada). Developed by Ambrilia stages. The precursor, PPL-100, is converted into PL-100, Biopharma, Inc., in conjunction with Mark Wainberg (McGill which has a half-life of approximately 35 h and does not University, Montreal, QC, Canada), this agent was recently require ritonavir boosting. In vitro selection studies produced acquired by Merck Pharmaceuticals. PL-100 is a potent orally viruses with reduced sensitivity to PL-100 (mutations selected bioavailable PI with a simple synthesis process of only five were novel and included K45R, M46I, T80I, and P81S), and

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 87 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 88

MIKE YOULE

Figure 13. Pharmacokinetic interaction studies assessing concomitant administration of the highest planned dose of GS-9137/ritonavir (200 mg/100 mg once daily) and tenofovir/emtricitabine. Plasma concentration–time profiles are shown.

GS-9137/ritonavir alone GS-9137/ritonavir + tenofovir/emtricitabine 1000

100 GS-9137 plasma concentration (ng/mL) 10 0 6 12 18 24 Time (h) Tenofovir/emtricitabine alone Tenofovir/emtricitabine + GS-9137/ritonavir 1000

100 Tenofovir plasma Tenofovir concentration (ng/mL) 10 0 6121824 Time (h) Tenofovir/emtricitabine alone Tenofovir/emtricitabine + GS-9137/ritonavir 10000

1000

100 Emtricitabine plasma concentration (ng/mL) 10 0 6121824 Time (h) G Data are mean ± standard deviation (n=24) G No clinically relevant drug–drug interaction

at least four were required to generate a 10.8-fold reduction 600, 1200, and 2400 mg once daily, either fasted or with a

in drug sensitivity. There appears to be no cross-resistance light meal. Although food reduced the Cmax, it had no effect with currently licensed PIs, and the T80I mutation results in on the AUC. No significant toxicity was seen (no grade 3 or hypersensitivity to saquinavir and nelfinavir. A Phase I study 4 events) and the compound is now moving forward to was undertaken in healthy volunteers, using doses of 300, clinical development.

88 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 89

THE 16TH INTERNATIONAL AIDS CONFERENCE

Figure 14. Pharmacokinetic studies assessing darunavir bilirubin changes were noted with atazanavir that probably exposure under fasted and fed conditions, and by meal type. resulted from the increased drug levels. The final study, with tipranavir/ritonavir 500 mg/200 mg Standard breakfast (n=23) and brecanavir 600 mg, was discontinued prematurely as Faster conditions (n=12) seven out of 12 healthy volunteer subjects developed High-fat breakfast (n=12) Nutritional protein-rich drink (n=10) elevated alanine aminotransferase (ALT) levels (grade 1 Croissant with coffee (n=11) [n=3], grade 2 [n=2], grade 3 [n=2]); this was believed 6000 to indicate a high risk of poor drug–drug interactions. All changes in the volunteers returned to normal following drug cessation. In conclusion, from a pharmacokinetic perspective, 5000 brecanavir seems best partnered with lopinavir/ritonavir (Kaletra, Abbott Laboratories, Abbott Park, IL, USA) and 4000 should not be used with tipranavir.

3000 Darunavir Among the PIs, the largest body of new data was for darunavir, the next anti-HIV drug to be licensed in many 2000 countries. Fitting that most of the results came from the pivotal POWER 1 and 2 trials, augmented by POWER 3; all 1000 but one of the presentations were posters and several were

Darunavir plasma concentration (ng/mL) combined analyses. 0 Starting with clinical pharmacology, Vanitha Sekar 0481216 20 24 (Tibotec, Inc.) and co-workers presented an overview of all Time (h) studies that have led to the recommended dose of darunavir/ritonavir of 600 mg/100 mg twice daily [29]. In healthy volunteers, ritonavir boosting increased the Brecanavir absolute oral bioavailability of darunavir from 37% to 82%, The second PI heading towards clinical use is brecanavir. and systemic darunavir exposure by approximately 14-fold. Data presented by Charles Craig (GlaxoSmithKline, Administering a higher dose of 200 mg of ritonavir did not Stevenage, UK) discussed the susceptibility in vitro of a increase darunavir levels to any greater extent. Darunavir is panel of 105 “worst case scenario” viruses from Monogram 95% plasma protein-bound, mostly with alpha-1-acid Biosciences, Inc. (South San Francisco, CA, USA), along with glycoprotein, and is metabolized almost exclusively by other viruses with single, double, or triple PI resistance- CYP3A4. A mass balance study showed that darunavir was associated mutations at positions 32, 33, 46, 47, 50, 54, 82, excreted mainly in feces (80%) and to a lesser extent in urine 84, and 90 [27]. Throughout this panel, brecanavir retained (12%), with a terminal half-life of 15 h. Although darunavir

activity at sub-nanomolar levels (median IC50 300 pM), exposure was approximately 30% lower under fasted than suggesting that increased numbers of resistance mutations fed conditions, it was unaffected by meal type (Fig. 14). will be required for the drug to fail. Amino acid substitutions Tolerability data for the three studies combined at residues 32, 47, and 50 were associated with the greatest were compared with data of those who received the reduction in sensitivity (six-fold). comparator PIs (CPIs) [30] (José Valdez Madruga, Centro de Data from three multiple-dose studies on the interactions Referência e Treinamento DST/AIDS, Mariana-São Paulo, of brecanavir (doses of 300 mg or 600 mg with 100 mg Brazil). The number of patient-years of exposure was greater ritonavir boosting) with a variety of PIs were shown by in the darunavir/ritonavir group than the CPI group (310 vs. Mark Shelton (GlaxoSmithKline) [28]. In the first study with 75 person-years), due to the longer mean duration of brecanavir 300 mg and lopinavir/ritonavir, no significant treatment and the larger number of patients treated in this changes in the levels of either agent were seen, and no group. Eleven percent of the darunavir/ritonavir group and unexpected adverse events were reported. In the second, 81% of the CPI group discontinued the drug because of with brecanavir 300 mg and atazanavir/ritonavir adverse events, and 3% and 67%, respectively, discontinued

300 mg/100 mg, both agents showed increases in Cmax the drug following virological failure. Darunavir/ritonavir (38% for brecanavir and 48% for atazanavir) and AUC was generally well tolerated and no specific toxicity was (41% for brecanavir and 21% for atazanavir). Some associated with the treatment. The incidence of diarrhea was

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 89 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 90

MIKE YOULE

Figure 15. Changes in lipids following treatment with darunavir/ritonavir 600 mg/100 mg after the first 12 weeks of therapy.

Grade 2 Grade 3 Grade 4 10 Triglycerides Cholesterol

5 Patients (%)

0 All CPI(s) All CPI(s) All CPI(s) All CPI(s) de novo de novo de novo de novo Week 4 Week 12 Week 4 Week 12

CPI: comparator protease inhibitor.

lower in the darunavir/ritonavir arm than in the CPI arm • Increased amylase (7% vs. 5%). (16% vs. 28%), as was headache (11% vs. 20%). • Increased total cholesterol (5% vs. 2%). Approximately 29% of the patients in each treatment group reported grade 3/4 adverse events. Fifteen percent of Changes in lipids over the first 12 weeks are shown in darunavir/ritonavir-treated patients and 14% of CPI-treated Fig. 15 and illustrate low rates of meaningful abnormalities patients reported at least one SAE, the most common over this short follow-up. Longer periods of therapy will being pneumonia and metabolic acidosis; individual SAEs be needed to clarify the comparative status of darunavir in occurred in <1% of patients. Eleven patients died in the this area. darunavir/ritonavir arms. Causes of death were illicit drug Finally, a poster abstract by D Dubois (Johnson & Johnson overdose, acute and chronic pulmonary emboli, anal cancer, Pharmaceutical Services, Beerse, Belgium) evaluated quality- multiple organ failure due to sepsis, bacterial endocarditis, of-life changes from POWER 1 and 2 with the validated acute respiratory failure, meningoencephalitis, septic shock, Functional Assessment of HIV Infection (FAHI) score, and chronic cryptosporidial diarrhea, brain edema, and unknown demonstrated that clinically significant improvements were cause; all deaths were considered by the investigator to be overall more frequent in the darunavir arms (20–40% for unrelated to the study medication. One patient died in the various scores) compared with the CPI group (15–25%) [32]. CPI group 2 weeks after follow-up. Two posters showed data for POWER 3 alone, a follow- An integrated analysis of laboratory abnormalities in the up of POWER 1 and 2, in which all subjects received same population was performed by Tony Vangeneugden darunavir/ritonavir 600 mg/100 mg from the outset. (Tibotec BVBA) [31]. The incidences of graded and non- POWER 3 was a non-randomized open-label trial that enrolled graded laboratory abnormalities were generally low and 327 patients; of these 75% were Caucasian, 87% male, the

similar in darunavir/ritonavir- and CPI-treated patients. Most mean entry viral load was 4.62 log10 HIV RNA copies/mL, were grade 1–2, and discontinuations because of these and the mean CD4+ T cell count was 115 cells/mm3. To assess changes were uncommon (1% of patients in both groups). whether there was a relationship between drug exposure The commonest grade 3/4 changes in the darunavir/ and efficacy or safety, trough and peak pharmacokinetic ritonavir group were: samples were taken at week 4. In addition, the baseline resistance measurement was used to calculate the inhibitory • Decreased white blood cell count (6% vs. 7% quotient (IQ) of the drug, i.e. how much drug is required to in the CPI group). overcome resistance to it [33]. Not surprisingly, it appears • Increased triglycerides (9% vs. 7%). from these data that the darunavir IQ is a strong predictor of

90 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 91

THE 16TH INTERNATIONAL AIDS CONFERENCE

Figure 16. Percent of patients with HIV RNA of <50 copies/mL at week 48 in the POWER 1 and 2 studies assessing therapy with darunavir/ritonavir. The response in baseline subgroups is shown.

Darunavir/ritonavir 600/100 mg twice daily 70 CPI(s)

60 21/36 44/82 50 27/61 40

30 Patients (%) 5/25 20 2/13 4/35 7/70 11/100 10 1/15 0/18 0 Enfurvitide used Enfurvitide used Enfurvitide 0 sensitive 1 sensitive (naïve) (non-naïve) not used ARV in OBR ARV in OBR

ARV: antiretroviral drug; OBR: optimized background regimen. Use of enfuvirtide was not randomized in POWER 1 and 2.

virological outcome, and that this is mainly driven by the from 244 patients who received CPI regimens [35]. In the baseline resistance fold change to the drug and not darunavir darunavir group, 91 patients had <50 HIV RNA copies/mL, exposure levels. With regard to safety, there seemed to be no compared with 22 in the CPI group (p<0.001). The association of darunavir drug levels with any adverse events. breakdown by baseline subgroups clearly demonstrated that It is comforting that signature toxicity does not seem to be a use of enfurvitide in the back group is important for patients hallmark of this compound. with virus that is still sensitive to the drug (Fig. 16). The second POWER 3 analysis [34], presented by CD4+ T cell responses were good in the darunavir group Jean-Michel Molina (Hospital Saint-Louis, Paris, France), compared with the CPI group in both POWER 1 (mean showed results at week 24 to be broadly similar to the first increase of 92 vs. 17 cells/mm3; p<0.001) and POWER 2 two studies: (mean increase of 102 vs. 19 cells/mm3; p<0.05). Adverse events were similar to those seen earlier, with approximately

• A >1 log10 reduction in HIV RNA in 65% of patients. 20% having some diarrhea and 15% headaches; the • Less than 50 HIV RNA copies/mL reached by 40% majority of adverse events were grades 1 and 2. It should be of patients. noted, however, that it is always difficult to assess side

• A mean reduction in HIV RNA of 1.65 log10. effects in this group of late-stage patients since immune reconstitution can produce many constitutional symptoms. The most common adverse events were diarrhea (14%), nasopharyngitis (11%), and nausea (10%). Grade 3 or 4 Conclusion triglyceride, cholesterol, and ALT and aspartate Overall, a hopeful set of results were presented at AIDS 2006, aminotransferase elevations occurred in 6%, 4%, 2%, and putting into perspective the widening gap between those who 2% of patients, respectively; these rates are similar to those have access to efficacious drugs and those that do not. seen in previous studies with darunavir/ritonavir. However, despite the great advances that have produced Sharon Walmsley (Toronto General Hospital, Toronto, new therapies, it remains vital that doctors and healthcare ON, Canada) gave an oral presentation on the week-48 professionals know how to use these treatments wisely to combined analysis of POWER 1 and 2, and compared the avoid the emergence of resistance, which has been the results from 241 subjects treated with darunavir with data hallmark of poor clinical practice in the past.

THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 91 R7694_2_REM_VE_2_2_07.qxd 2/11/06 16:40 Page 92

MIKE YOULE

Acknowledgement 18. Montaner J, Harris M, Larsen G et al. Combination of TMC114/ritonavir and TMC125 in patients with multidrug resistant HIV. XVI International AIDS Conference, Toronto, The author would like to thank the investigators who shared ON, Canada, 13–18 August, 2006 (Abstr. THPE0136). slides from their presentations. 19. Summa V, Pace P, Petrocchi A et al. Discovery of MK-0518 a novel, potent and selective HIV integrase inhibitor in phase III clinical trials. XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. THPE0020). References 20. Laufer R, Gonzalez-Paz O, Taliani M et al. Preclinical pharmacokinetics and metabolism 1. Norris D, Morales J, Godofsky E et al. TNX-355, in combination with optimized of MK-0518, a potent HIV Integrase inhibitor in phase III clinical trials. XVI International background regimen (OBR), achieves statistically significant viral load reduction and CD4 AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. THPE0027). cell count increase when compared with OBR alone in Phase 2 study at 48 weeks. XVI 21. Miller M, Witmer M, Stillmock K et al. Biochemical and antiviral activity of MK-0518, International AIDS Conference, Toronto, Canada, 13–18 August, 2006 (Abstr. THLB0218). a potent HIV integrase inhibitor. XVI International AIDS Conference, Toronto, ON, 2. Norris D, Morales J, Gathe J et al. Phase 2 efficacy and safety of the novel entry inhibitor, Canada, 13–18 August, 2006 (Abstr. THAA0302). TNX-355, in combination with optimized background regimen (OBR). XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0058). 22. Markowitz M, Nguyen BY, Gotuzzo F et al. Potent antiretroviral effect of MK-0518, a novel HIV-1 integrase inhibitor, as part of combination ART in treatment-naïve 3. Weinheimer S, D’arigo K, Fung M et al. TNX-355 is active against enfuvirtide resistant HIV-1 infected patients. XVI International AIDS Conference, Toronto, ON, Canada, HIV. XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. THPE0024). 13–18 August, 2006 (Abstr. THLB0214). 4. Sansone A, Keung A, Tetteh E et al. Vicriviroc (VCV) pharmacokinetics (PK): lack of impact 23. DeJesus E, Berger D, Markowitz M et al. Antiviral activity, pharmacokinetics, and dose of ritonavir (RTV)-boosted protease inhibitors (PI). XVI International AIDS Conference, response of the HIV-1 integrase inhibitor GS-9137 (JTK-303) in treatment-naive and Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0074). treatment-experienced patients. J Acquir Immune Defic Syndr 2006;43:1–5. 5. Gulick R, Su Z, Flexner C et al. ACTG 5211: Phase II study of the safety and efficacy 24. Ramanathan S, Skillington J, Plummer A et al. Lack of clinically relevant drug-drug of vicriviroc in HIV-infected treatment-experienced subjects. XVI International AIDS interaction between ritonavir-boosted GS-9137 (GS-9137/r) and emtricitabine (FTC)/ Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. THLB0217). tenofovir disoproxil fumarate (TDF). XVI International AIDS Conference, Toronto, 6. Mayer H, van der Ryst E, Saag M et al. Safety and efficacy of maraviroc (MVC), a novel ON, Canada, 13–18 August, 2006 (Abstr. TUPE0080). CCR5 antagonist, when used in combination with optimized background therapy (OBT) 25. Ramanathan S, Lagan K, Plummer A et al. Lack of clinically relevant drug-drug for the treatment of antiretroviral-experienced subjects infected with dual/mixed-tropic interaction between the ritonavir-boosted HIV integrase inhibitor GS-9137/r and HIV-1: 24-week results of a Phase 2b exploratory trial. XVI International AIDS Conference, zidovudine (ZDV). XVI International AIDS Conference, Toronto, ON, Canada, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. THLB0215). 13–18 August, 2006 (Abstr. TUPE0088). 7. Davison DK, Medinas RJ, Mosier SM et al. New fusion inhibitor peptides, TRI-999 and TRI-1144, are potent inhibitors of enfuvirtide and T-1249 resistant isolates. 26. Dandache S, Wainberg MA, Panchal C et al. PL-100, a novel protease inhibitor with XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 a high genetic barrier. XVI International AIDS Conference, Toronto, ON, Canada, (Abstr. THPE0021). 13–18 August, 2006 (Abstr. THAA0304). 8. Delmedico M, Bray B, Cammack N et al. Next generation fusion inhibitor candidates 27. Craig C, Yates P, St Clair M et al. Survey of brecanavir (BCV) and other protease inhibitor TRI-1144 and TRI-999 have improved pharmacokinetics: progress towards once/week (PI) susceptibility to HIV-1 variants containing PI resistance-associated amino acid dosing. XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 substitutions (RAS). XVI International AIDS Conference, Toronto, ON, Canada, (Abstr. THAA0303). 13–18 August, 2006 (Abstr. THPE0023). 9. Cahn P, Reuss F, Rolon M et al. A Phase I study to explore the safety, tolerability, and 28. Shelton MJ, Ford S, Anderson MT et al. Overview of drug interactions between brecanavir pharmacokinetics of fosalvudine tidoxil in patients infected with HIV-1. XVI International (BCV) and other HIV protease inhibitors (PIs). XVI International AIDS Conference, AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. THLB0216). Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUAB0105). 10. Katlama C, Ghosn J, Tubiana R et al. MIV-310 reduces HIV viral load in patients 29. Sekar V, Spinosa-Guzman S, Lefebvre E et al. Clinical pharmacology of TMC114 – failing multiple antiretroviral therapy: results from a 4-week phase II study. a potent HIV protease inhibitor. XVI International AIDS Conference, Toronto, ON, AIDS 2004;18:1299–304. Canada, 13–18 August, 2006 (Abstr. TUPE0083). 10. Reuss F, Kulke M, Braspenning J et al. Fosalvudine tidoxil, a novel alovudine-derived prodrug is activated and inhibits the replication of HIV-1 in human PBMC. XVI 30. Valdez Madruga J, Lafeuillade A, Beatty G et al. TMC114/r is well tolerated by International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 treatment-experienced patients in power 1, 2 and 3: integrated clinical safety analysis. (Abstr. THPE0025). XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0062). 12. van Heeswijk R, Hoetelmans R, Kestens D et al. The pharmacokinetic interaction between ketoconazole and TMC278, an investigational non-nucleoside reverse transcriptase 31. Vangeneugden T, Van Baelen B, De Paepe E et al. TMC114/r in treatment-experienced inhibitor (NNRTI), in healthy HIV-negative subjects. XVI International AIDS Conference, patients in power 1, 2 and 3: integrated analysis of laboratory parameters. Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0087). XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 13. Schöller-Gyüre M, Woodfall B, Vanaken H et al. Lack of interaction between TMC125 and (Abstr. TUPE0063). methadone. XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 32. Dubois D, Smets E, Vangeneugden T et al. Improved quality of life in treatment- 2006 (Abstr. TUPE0084). experienced HIV patients treated with TMC114/r vs control protease inhibitors: results 14. Schöller-Gyüre M, De Smedt G, Vanaken H et al. TMC125 bioavailability is not affected of power 1 and 2 functional assessment of HIV infection (FAHI). XVI International by ranitidine and omeprazole. XVI International AIDS Conference, Toronto, ON, Canada, AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0069). 13–18 August, 2006 (Abstr. TUPE0082). 33. Sekar V, De Meyer S, Vangeneugden T et al. Absence of TMC114 exposure-efficacy 15. Cohen C, Steinhart C, Ward D et al. Efficacy and safety results at 48 weeks with the novel and exposure-safety relationships in power 3. XVI International AIDS Conference, NNRTI, TMC125, and impact of baseline resistance on the virologic response in study Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0078). TMC125-C223. XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0061). 34. Molina JM, Cohen C, Katlama C et al. TMC114/r in treatment-experienced HIV patients in power 3: 24-week efficacy and safety analysis. XVI International AIDS Conference, 16. Boffito M, Back D, Stainsby-Tron M et al. Pharmacokinetics of saquinavir hard gel/ritonavir (1000/100 mg twice daily) when administered with tenofovir diproxil fumarate in HIV-1- Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0060). infected subjects. Br J Clin Pharmacol 2005;59:38–42. 35. Lazzarin A, Queiroz-Telles F, Frank I et al. TMC114 provides durable viral load suppression 17. Kakuda T, Schöller-Gyüre M, Peeters M et al. Pharmacokinetic interaction study with in treatment-experienced patients: POWER 1 and 2 combined week 48 analysis. TMC125 and TMC114/rtv in HIV-negative volunteers. XVI International AIDS Conference, XVI International AIDS Conference, Toronto, ON, Canada, 13–18 August, 2006 Toronto, ON, Canada, 13–18 August, 2006 (Abstr. TUPE0086). (Abstr. TUAB0104).

92 THE JOURNAL OF VIRAL ENTRY Vol 2 No 2 2006 R7694_2_REM_VE_COV_2_2_01.qxd 31/10/06 12:08 Page 4

THE JOURNAL OF VIRAL ENTRY

Reader Survey – Let Us Know What You Think!

Please take a few moments to complete this survey. We would value your opinion.

Please photocopy this page, complete the survey below, and fax it back to Remedica on +44 (0)20 7759 2951. Or you can visit the VIRAL ENTRY website and complete the survey online (registration online is FREE): www.viralentry.com

1. We are aiming to provide practical information for healthcare professionals working in the field of antiviral medicine. How would you rate the information presented in this issue?

Strongly agree Strongly disagree

a) The technical quality of information included in THE JOURNAL OF VIRAL ENTRY was acceptable: 12345

b) The information was relevant to my practice: 12345

c) The information was presented clearly: 12345

d) The leading articles provide new information regarding the understanding and use of antiviral medicines 12345

e) The clinical review section was helpful and I would like to see analyses in future issues: 12345

2. Did you learn anything through the CME activity THE JOURNAL OF VIRAL ENTRY that will change the way you practice medicine? ■ Yes ■ No

If so, what?......

3. Is there anything you learned from the CME activity THE JOURNAL OF VIRAL ENTRY that prompts you to seek further information that may influence the way you practice medicine in the future? ■ Yes ■ No

If so, what?......

4. Would you like to recommend THE JOURNAL OF VIRAL ENTRY to a colleague? ■ Yes ■ No

My colleague’s email is: ......

5. What specific topics do you think should be covered in future issues?

......

Name ...... Job title ......

Institution ......

Address ......

Country ...... Post/zip code ......

Email ...... R7694_2_REM_VE_COV_2_2_01.qxd 31/10/06 12:08 Page 1

The Journal of Viral Entry is supported by an educational grant from Pfizer