Winnie Wing Yin Tong

BSc(Med) MBBS FRACP FRCPA

This thesis is submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

St Vincent’s Clinical School

Faculty of Medicine

February 2016

PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: TONG

First name: Winnie Other name/s: Wing Yin

Abbreviation for degree as given in the University calendar: PhD

School: St Vincent’s Clinical School Faculty: Faculty of Medicine

Title: T-cell responses to human papillomavirus in men with anal high-grade squamous intraepithelial lesions

Abstract 350 words maximum: (PLEASE TYPE)

Anal cancer incidence in men who have sex with men (MSM) infected with human immunodeficiency virus (HIV) is three times greater than cervical cancer incidence was before Papanicolaou screening. Analogous to cervical cancer, most anal cancers are caused by human papilloma virus type 16 (HPV16) and is preceded by a precancerous stage termed high-grade squamous intraepithelial lesion (HSIL). Unlike cervical cancer, the natural history of anal HSIL is poorly understood, and there is little evidence to inform prevention guidelines.

In a retrospective audit of 574 men (73% HIV-infected) attending our anal cancer screening clinic, annual progression rate to HSIL was 7.4%, and for HSIL to cancer was 1.2%. This was the first study to quantify a spontaneous regression rate from HSIL, which was common at 23.5% per year.

To further understand the role of the cellular immune response in anal HSIL spontaneous regression, two assays were developed to measure systemic HPV16-specific T-cell responses and applied to a cross-sectional substudy of the Study of the Prevention of ANal Cancer (SPANC), a prospective, natural history study of anal cancer precursors and HPV in MSM ≥35 years. CD4+ T-cell responses to the HPV16 oncogenic protein E6 were detected in 80 (N=134, 60%) men. These responses may be associated with recent HSIL regression – five of six (83%) regressors had these responses compared to seven of 20 (35%) non-regressors (Pexact=0.065).

T-cell immune responses in the anal mucosa were also studied in the above 26 men with HSIL at study entry. 24 (55%) biopsies had stromal lymphoid aggregates. Using whole slide imaging (600x) and two-colour immunofluorescent staining, CD4+ and CD8+ T-cell densities were measured. Lymphoid aggregates were CD4+ T-cell enriched (2.8-fold higher in density, P<0.01) compared to CD8+ T-cells (1.7-fold higher, P=0.08). Biopsies with HSIL diagnosis and having anal HPV16 detected were associated with higher total T-cell density.

In summary, anal HSIL spontaneous regression is common. A T-cell mediated mechanism for regression is plausible. This should be taken into account when developing screening and treatment strategies for anal HSIL.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…………………………………………………………… ……………………………………..……………… ……….……………………...…….…5th June 2016 Signature Witness Signature Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS

ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………......

Date ……………………………………………......

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed ……………………………………………......

Date ……………………………………………......

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed ……………………………………………......

Date ……………………………………………......

Originality Statement ...... 7

Acknowledgements ...... 8

Publications ...... 11

List of Figures ...... 12

List of Supplementary Figures ...... 13

List of Tables...... 14

List of Abbreviations...... 16

Chapter 1 - Literature review ...... 21

1.1 Introduction ...... 22 1.2 Human papillomavirus (HPV) ...... 22 1.2.1 Life cycle ...... 26 1.2.2 Classification ...... 27 1.3 Clinical manifestations of HPV infection ...... 29 1.3.1 With primary immunodeficiency ...... 29 1.3.2 With secondary immunodeficiencies ...... 30 1.3.2.1 Human immunodeficiency virus (HIV) ...... 30 1.3.2.2 Transplant recipients ...... 31 1.4 The immune response to HPV infection ...... 33 1.4.1 Innate ...... 33 1.4.2 Humoral ...... 35 1.4.3 Cellular ...... 36 1.5 Anal HPV, squamous intraepithelial lesions and cancer ...... 38

1.5.1 Epidemiology...... 38 1.5.1.1 Anal HPV and AIN ...... 38 1.5.1.2 Anal cancer ...... 39 1.5.2 Pathogenesis ...... 43 1.5.3 Diagnosis ...... 46 1.5.4 Treatment ...... 48 1.5.5 Prevention ...... 50 1.5.5.1 Screening ...... 50 1.5.5.2 Vaccination ...... 51 1.6 The anti-tumour immune response ...... 52 1.6.1 Cancer immunoediting ...... 56 1.6.2 Cancer immunotherapy ...... 57 1.7 Measuring HPV16-specific T-cell responses ...... 59 1.7.1 Functional T-cell assays ...... 59 1.7.2 The CD25/CD134 assay ...... 61 1.8 Conclusion ...... 61

Chapter 2 - Progression to and spontaneous regression of high-grade anal squamous intraepithelial lesions in HIV-infected and uninfected men ...... 63

2.1 Abstract ...... 64 2.2 Introduction ...... 66 2.3 Methods ...... 67 2.3.1 Study Population...... 67 2.3.2 Study Design ...... 68 2.3.3 Data Collection ...... 68 2.3.4 Endpoint Definitions...... 69 2.3.5 Statistical Analysis ...... 70 2.4 Results ...... 71 2.4.1 Patients ...... 71 2.4.2 Frequency of Testing and Follow-Up ...... 72 2.4.3 Progression ...... 74 2.4.4 Regression ...... 78 2.4.5 Cancer ...... 79 2.5 Discussion ...... 81

Chapter 3 - HPV16-specific T-cell responses and spontaneous regression of anal high- grade squamous intraepithelial lesions...... 89

3.1 Abstract ...... 91 3.2 Introduction ...... 92 3.3 Methods ...... 93 3.3.1 Study Population ...... 93 3.3.2 CD25/CD134 Assay ...... 94 3.3.3 Intracellular Cytokine Staining Assay ...... 95 3.3.4 Flow Cytometry Acquisition ...... 96 3.3.5 Assay Cut-offs ...... 96 3.3.6 Statistics ...... 98 3.4 Results ...... 98 3.4.1 Participant characteristics ...... 98 3.4.2 HPV16-E6 and E7-specific T-cell responses ...... 99 3.4.3 Clinical associations with HPV16-E6 and E7 CD4+ T-cell responses ...... 110 3.4.4 Clinical associations with HPV16-E6 and E7 CD8+ T-cell responses ...... 111 3.4.5 HPV16-E6 and E7 T-cell responses and spontaneous HSIL regression ...... 112 3.5 Discussion ...... 114

Chapter 4 - T-cells in the anal mucosa of men with anal high-grade squamous intraepithelial lesions ...... 125

4.1 Abstract ...... 126 4.2 Introduction ...... 128 4.3 Methods ...... 130 4.3.1 Study Population ...... 130 4.3.2 Haematoxylin & eosin stained diagnostic slides ...... 131 4.3.3 Immunofluorescent staining for CD4+ and CD8+ cells ...... 134 4.3.4 Microscopy ...... 136 4.3.5 CD4+ and CD8+ cell enumeration...... 138 4.3.6 Statistics ...... 139 4.4 Results ...... 142 4.4.1 Participant and biopsy characteristics ...... 142 4.4.2 Lymphoid aggregates ...... 145

4.4.3 Correlation of CD4:CD8 between systemic and mucosal compartments ...... 148 4.4.4 Factors associated with T-cell density ...... 149 4.5 Discussion ...... 152

Chapter 5 - Discussion ...... 155

5.1 Summary and context of findings ...... 156 5.2 Implications ...... 159 5.3 Future directions ...... 162 5.3.1 Addressing limitations of current work ...... 162 5.3.2 Future work ...... 163

References ...... 167

Appendix A ...... A

Appendix B ...... B

Appendix C ...... C

Appendix D ...... D

Thank you to the SPANC participants who are extraordinary in their commitment to this study, and without whom this work would not be possible.

Brian Acraman, Anu Aggarwal, Michelle Bailey, Leon Botes, Sam Breit, David Brown,

Alexander Carrera, Laura Cook, David Cooper, Alyssa Cornall, Leonie Crampton,

Philip Cunningham, Deborah Ekman, Kit Fairley, Annabelle Farnsworth, Eddie

Fraissard, Bertha Fsadni, Suzanne Garland, Kim Grassi, Andrew Grulich, Karl Hesse,

Richard Hillman, Denise Hsu, Susanna Ip, Jeff Jin, Graham Jones, Francois Lamoury,

Juliana Lamoury, Carmella Law, Yen Peng Lim, Melanie Lograsso, Dorothy Machalek,

Tara Maher, Kristin McBride, Patrick McGrath, Leo McHugh, Leon McNally, Alan

Meagher, Robert Mellor, Kate Merlin, Sharon Mitchell, Mee-Ling Munier, Richard

Norris, Matthew O’Dwyer, Robert Owe-Young, Chan Phetsouphanh, Maria Piperias,

Mary Poynten, Adele Richards, Robyn Richardson, Jenny Roberts, Maria Sarris, Daniel

Seeds, Bill Sewell, Ansari Shaik, Fei Shang, Kelsee Shepherd, Ivy Shih, Krista Siefried,

Brett Sinclair, Kate Sinn, Sepehr Tabrizi, David Templeton, Kate Thompson, Julia

Thurloe, Stuart Turville, Rick Varma, Susan Wan, Yin Xu and John Zaunders.

My heartfelt thanks to A/Professor Raymond Garrick for proofreading this thesis.

The St Vincent’s Clinic and Curran Foundations supported this work with Annual

Grants from 2011-2013, the Di Boyd Cancer Grant in 2013 and the K&A Collins

Cancer Grant in 2014.

Andrew Carr

Tony Kelleher

Your guidance and mentorship have made it possible for me to sail this “uncharted sea”.

The knowledge and wisdom you both have patiently and generously shared, and the

confidence you have shown in me, are gifts I will endeavour to pass on.

Sara & Sum

This thesis is dedicated to my Mum and Dad.

One day, cancer will be no more threatening than infection is today.

Ken

Without whose love, support and perseverance this thesis would not be complete.

This page has been intentionally left blank

Tong WWY, Jin F, McHugh LC, Maher T, Sinclair B, Grulich AE, Hillman RJ, Carr A.

Progression to and spontaneous regression of high-grade anal squamous intraepithelial lesions in HIV-infected and uninfected men. AIDS. 2013;27(14):2233-43. (Appendix

Machalek DA, Grulich AE, Hillman RJ, Jin F, Templeton DJ, Tabrizi SN, Garland SM,

Prestage G, McCaffery K, Howard K, Tong W, Fairley CK, Roberts J, Farnsworth A,

Poynten IM, for the SPANC team. The Study of the Prevention of Anal Cancer

(SPANC): design and methods of a three-year prospective cohort study. BMC Public

Health. 2013;13(1):946.

Tong WWY, Hillman RJ, Kelleher AD, Grulich AE, Carr A. Anal intraepithelial neoplasia and squamous cell carcinoma in HIV-infected adults. HIV Med.

2014;15(2):65-76. [Review] (Appendix A)

Tong WWY, Shepherd K, Garland S, Meagher A, Templeton DJ, Fairley CK, Jin F,

Poynten IM, Zaunders J, Hillman RJ, Grulich AE, Kelleher AD, Carr A, for the SPANC team. Human papillomavirus 16-specific T-cell responses and spontaneous regression of anal high-grade squamous intraepithelial lesions. J Infect Dis. 2015;211(3):405-15.

(Appendix C)

Figure 1.1 – Human papillomavirus ...... 24

Figure 1.2 – Life cycle of HPV illustrated in the cervical epithelium ...... 26

Figure 1.3 – Phylogenetic tree of 170 HPV types...... 28

Figure 1.4 – Typical course of HIV infection ...... 31

Figure 1.5 – Immune responses to HPV16 infection ...... 34

Figure 1.6 – Natural history of high-risk HPV infection ...... 43

Figure 1.7 – Functions of the E6 and E7 oncoproteins, and their interaction with each other in steps that lead to cell immortalisation ...... 45

Figure 1.8 – High-resolution anoscopy ...... 47

Figure 1.9 – Cancer immunoediting ...... 58

Figure 2.1 – Kaplan-Meier curve for progression to grade 3 anal intraepithelial neoplasia...... 75

Figure 3.1 – Representative data sets demonstrating T-cell assay gating strategies...... 97

Figure 3.2 – Positive responses detected by T-cell assays...... 103

Figure 3.3 – Qualitative analysis of positive responses detected by ICS...... 104

Figure 3.4 – Clinical course of participants with baseline high-grade disease who experienced spontaneous regression...... 113

Figure 4.1 – Anal biopsies with lymphoid infiltrates ...... 129

Figure 4.2 – Diagram of an anal biopsy block processed as per SPANC protocols ..... 132

Figure 4.3 – Annotated digitised slides of a diagnostic haematoxylin & eosin stained anal biopsy ...... 133

Figure 4.4 – Two-colour immunofluorescent staining protocol ...... 135

Figure 4.5 – CD4+ and CD8+ cell enumeration ...... 140

Figure 4.6 – Anal biopsies with and without lymphoid aggregates ...... 146

Figure 4.7 – Anal mucosal lymphoid aggregates are CD4+ T-cell enriched ...... 147

Figure 4.8 – Correlation of CD4:CD8 between systemic and mucosal compartments 148

Figure 5.1 – Incidence and regression of anal high-grade squamous intraepithelial lesions, and progression to cancer in untreated men ...... 157

Supplementary Figure 2.1 – Flowchart of patients in endpoint analyses ...... 87

Supplementary Figure 3.1 – Correlation between positive responses on CD25/CD134 assay and cytokine production on ICS...... 118

Supplementary Figure 3.2 – Qualitative analysis of individual positive responses ranked by response size...... 122

Table 1.1 – Anal cytology and histology terminology...... 23

Table 1.2 – HPV16 proteins ...... 25

Table 1.3 – Stage-3 (AIDS)-defining opportunistic illnesses in HIV infection ...... 32

Table 1.4 – Prevalence, incidence and risk factors for anal canal HPV, anal intraepithelial neoplasia (AIN) and anal cancer in HIV-infected men ...... 40

Table 1.5 – Prevalence, incidence and risk factors for anal canal HPV, anal intraepithelial neoplasia (AIN) and anal cancer in HIV-infected women ...... 41

Table 1.6 – Incidence of cancer or precancer spontaneous regression in placebo arms of randomised controlled trials (RCT) of treatment ...... 54

Table 1.7 – MHC class II restricted immune epitopes of HPV16 oncoproteins ...... 60

Table 2.1 – Cohort description at baseline ...... 73

Table 2.2 – Progression to and spontaneous regression of HGAIN* ...... 76

Table 2.3 – Progression to and spontaneous regression of histologically-confirmed

AIN3...... 77

Table 2.4 – Incident anal SCC patients ...... 80

Table 3.1 – Participant characteristics ...... 100

Table 3.2 – Clinical associations with positive HPV16-E6-specific T-cell responses

(univariable logistic regression) ...... 106

Table 3.3 – Clinical associations with positive HPV16-E7-specific T-cell responses

(univariable logistic regression) ...... 108

Table 4.1 – Antibodies used for two-colour immunofluorescent staining ...... 136

Table 4.2 – Summary of fluorescence microscopy acquisition settings (N=76) ...... 137

Table 4.3 – Participant characteristics ...... 143

Table 4.4 – Biopsy characteristics ...... 144

Table 4.5 – Factors associated with T-cell density (univariable generalised linear model)

...... 150

Table 5.1 – Wilson and Jungner criteria (283) for diseases suitable for screening applied to cervical and anal cancer ...... 160

Table 5.2 – Potential directions for future research arising from this thesis...... 166

AIN1/2/3 anal intraepithelial neoplasia grade 1/2/3

AIDS acquired immunodeficiency syndrome

AJCC American Joint Committee on Cancer

ANOVA analysis of variance

(c)ART (combination) antiretroviral therapy

ASC-H atypical squamous cells – cannot exclude high-grade

ASC-US atypical squamous cells of undetermined significance

CEF CMV, Epstein-Barr virus and influenza

CI confidence interval

CIN1/2/3 cervical intraepithelial neoplasia grade 1/2/3

CMV cytomegalovirus

CTLA-4 cytotoxic T-lymphocyte associated protein 4

DAPI 4',6-diamidino-2-phenylindole

DNA deoxyribonucleic acid

DOCK8 dedicator of cytokinesis 8

EDTA ethylenediaminetetraacetic acid

FSC-A/H forward scatter (area/height)

HGAIN high-grade anal intraepithelial neoplasia

H&E haematoxylin & eosin

HIV human immunodeficiency virus

HLA human leucocyte antigen

HPV16 human papillomavirus type 16

HR hazard ratio

HRA high-resolution anoscopy hr-HPV high-risk type HPV

HSIL high-grade squamous intraepithelial lesion

IARC International Agency for Research on Cancer

ICS intracellular cytokine staining

IDU injecting drug user

IEL intraepithelial lymphocyte

IFNγ interferon gamma

IL2 interleukin-2

IQR interquartile range

LCM laser capture microdissection

LCR long control region

LGAIN low-grade anal intraepithelial neoplasia lr-HPV low-risk type HPV

LSIL low-grade squamous intraepithelial lesion

mRNA messenger RNA

MHC major histocompatibility complex

MSM men who have sex with men

Neg negative (for squamous intraepithelial lesion or malignancy)

NK natural killer

OR odds ratio

PBMC peripheral blood mononuclear cell

PCR polymerase chain reaction

PD-1 programmed cell death protein 1

PD-1L programmed cell death 1 ligand 1

PHSIL possible high-grade squamous intraepithelial lesion

PLSIL possible low-grade squamous intraepithelial lesion

PY person-years

RAI receptive anal intercourse

RCT randomised controlled trial

RNA ribonucleic acid

SCC squamous cell carcinoma

SCJ squamocolumnar junction

SEB staphylococcal enterotoxin B

SD standard deviation

(A)SIL (anal) squamous intraepithelial lesion

SISCCA superficially invasive squamous cell carcinoma

SPANC Study of the Prevention of ANal Cancer

SSC-A side scatter (area)

START Strategic Timing of AntiRetroviral Treatment study

STI sexually transmitted infection

TIL tumour-infiltrating lymphocyte

TLR Toll-like receptor

TBS Tris (2-amino-2-hydroxymethyl-propane-1,3-diol) buffered saline

TWAUC time-weighted area under the curve

VLP virus-like particle

WHIM warts, hypogammaglobulinaemia, infections and myelokathexis

This page has been intentionally left blank

Sections of this chapter were published as a review in HIV Medicine 2014;15(2):65-76.

A PDF of the published article is included in Appendix A.

Anal high-grade squamous intraepithelial lesions (SIL) are considered to be the precursor to anal squamous cell carcinoma (SCC), a cancer usually treated with combined chemoradiation at significant cost and morbidity. Human immunodeficiency virus (HIV) coupled with oncogenic human papilloma virus (HPV) infection are strong risk factors for anal SCC (1) and high-grade SIL, especially in men who have sex with men (MSM). Similar to cervical cancer, greater than 90% of anal cancers are attributable to persistent infection with HPV (2). Terminology used for diagnosis of anal

SILs parallels that for the cervix (Table 1.1) (3). In this first chapter, the biology and clinical manifestations of HPV will be reviewed, followed by a summary of the current literature on anal HPV, SILs and cancer. This chapter concludes with a brief outline of the anti-tumour response as it relates to anal HPV, SILs and cancer, and background on methods of measuring HPV16-specific T-cell responses.

Papillomaviruses are host-specific, small (40-55 nanometres), non-enveloped, double- stranded DNA viruses of approximately 8000 bases (4). An icosahedral capsid (5) is formed by their major (L1) and minor (L2) structural proteins (Figure 1.1). There are six non-structural proteins of which E6 and E7 are the major oncoproteins (Table 1.2)

(4).

Table 1.1 – Anal cytology and histology terminology

Cytology Histology Australian Lower The Bethesda Modified Anogenital System (TBS) Bethesda Squamous Equivalent Terms (8) 2001 (6) System (AMBS) Terminology 2004 (7) (LAST) (3) Unsatisfactory* Unsatisfactory* Negative for Negative# Negative# AIN ASC-US PLSIL Condyloma acuminatum Koilocytosis / wart virus LSIL effect LSIL LSIL AIN1 / mild dysplasia LGAIN ASC-H PHSIL AIN2 / moderate dysplasia HSIL AIN3 / severe dysplasia HSIL HSIL Carcinoma in situ HGAIN SISCCA SCC SCC SCC SCC *Cytology results may be reported as unsatisfactory for evaluation due to inadequate cellularity, lack of anal transformation zone, lack of cell preservation, contamination with bacterial or faecal material. #Negative for intraepithelial lesion or malignancy. Abbreviations: AIN1/2/3 = anal intraepithelial neoplasia grade 1/2/3 ASC-H = atypical squamous cells – cannot exclude high-grade ASC-US = atypical squamous cells of undetermined significance HGAIN = high-grade anal intraepithelial neoplasia HSIL = high-grade squamous intraepithelial lesion LGAIN = low-grade anal intraepithelial neoplasia LSIL = low-grade squamous intraepithelial lesion PHSIL = possible high-grade squamous intraepithelial lesion PLSIL = possible low-grade squamous intraepithelial lesion SCC = squamous cell carcinoma SISCCA = superficially invasive squamous cell carcinoma.

Figure has been removed due to copyright restrictions.

Figure 1.1 – Human papillomavirus

A. Drawing of the molecular surface of an atomic model of the papillomavirus capsid.

Colours represent distance from the centre of the capsid. Reprinted from The EMBO

Journal, volume 21; Yorgo Modis, Benes L. Trus, Stephen C. Harrison; Atomic model

Organization, with permission from John Wiley and Sons (5).

B. Genomic organisation of HPV16. The three concentric circles correspond to the

three reading frames in which the sense strand can be translated. Eight genes (L1, L2,

E1, E2, E4, E5, E6 and E7) encode more than eight gene products as a result of mRNA

splicing. Reprinted from Vaccine, volume 30S; John Doorbar, Wim Quint, Lawrence

Banks, Ignacio G. Bravo, Mark Stoler, Tom R. Broker, Margaret A. Stanley; The

from Elsevier (9).

Table 1.2 – HPV16 proteins Protein Function E1 DNA helicase (10) E2 Forms helicase complex with E1 Transcription factor E4 Destabilisation of cytokeratin network (11) E5 Modulates function of various transmembrane proteins (12) E6 Binds and degrades p53 (13-15) Activation of telomerase (16-18) Inhibits degradation of Src-family of protein tyrosine kinases (17, 19) E7 Inactivates retinoblastoma protein (pRb) (20) Stimulation of S-phase genes cyclins A and E (21) Functional inactivation of cyclin-dependent kinase inhibitors (17) L1 Major capsid protein L2 Minor capsid protein

1.2.1 Life cycle

Papillomaviruses are strictly epitheliotropic. After infecting basal epithelial cells via microabrasions, they complete their life cycle in synchronisation with the maturing epithelium (Figure 1.2) (9).

Figure has been removed due to copyright restrictions.

Figure 1.2 – Life cycle of HPV illustrated in the cervical epithelium

Reprinted from Vaccine, volume 30S; John Doorbar, Wim Quint, Lawrence Banks,

Ignacio G. Bravo, Mark Stoler, Tom R. Broker, Margaret A. Stanley; The Biology and

Elsevier (9).

1.2.2 Classification

There are now 170 HPV types identified (22) based on their L1 open-reading frame

DNA sequence, each differing by more than 10% in this conserved region (23).

Genitally-transmitted, disease-causing types belong to the genus Alphapapillomavirus

(Figure 1.3) (24).

HPVs are also classified according to their carcinogenicity as “high-risk” based on a strong epidemiological association with cervical cancer (25) and cancers at other epithelial sites, or “low-risk” (cause benign warts). In 2012, the International Agency for Research on Cancer (IARC) classified HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52,

56, 58 and 59 as Group 1 carcinogens i.e. there is sufficient evidence of their carcinogenicity in humans. Other HPV types are considered probably (Group 2A e.g.

HPV68) or possibly (Group 2B e.g. HPV26) carcinogenic to humans, but classification of HPV types that are weakly carcinogenic (e.g. HPV73, HPV66, HPV53) on an epidemiological basis is challenging (23, 26, 27).

Figure has been removed due to copyright restrictions.

Figure 1.3 – Phylogenetic tree of 170 HPV types

Based on the L1 open reading frame using the maximum likelihood method. HPV16 indicated by yellow arrow, HPV18 by green arrow. Reprinted with minimal adaptations from Virology, volume 445; Ethel-Michele de Villiers; Cross-roads in the classification of papillomaviruses; page 6 © 2013, with permission from Elsevier (22).

Regardless of epithelial site, the consequence of HPV infection can range from asymptomatic carriage, to exophytic warts, to different grades of intraepithelial neoplasia with varying risk of progression to cancer. Low-risk types of HPV (e.g. HPV6 and 11) generally cause exophytic warts or low-grade SILs that are benign. In recurrent respiratory papillomatosis, a rare condition characterised by multiple benign papillomas of the larynx and trachea, there is potential for malignant transformation in 3-5% of patients (28).

High-risk types of HPV can cause high-grade SILs at mucosal sites, particularly in the anogenital region including the cervix, anus, vulva, vagina and penis (26). Most HPV- positive anal cancers are attributed to high-risk HPV types 16 (85%) or 18 (7%) (29,

30); these account for 70% of all cervical cancers (31). HPV has also been shown to be an important risk factor (odds ratio 12.3) for oropharyngeal SCC independent of smoking and alcohol use (32).

1.3.1 With primary immunodeficiency

HPV infection, including warts and SCC, are features of some (but not all) primary immunodeficiency syndromes. Epidermodysplasia verruciformis is a rare condition characterised by increased susceptibility to cutaneous manifestations of

Betapapillomaviruses, which are usually asymptomatic in normal individuals. The underlying genetic defect is in EVER1 or EVER2, which encode for endoplasmic reticulum transmembrane proteins important in zinc homeostasis (33, 34). Warts are also a feature of WHIM syndrome, accompanied by hypogammaglobulinaemia, infections and myelokathexis (35, 36). The underlying genetic defect is a gain-of-

function mutation in the chemokine receptor CXCR4 (37). Abnormal manifestations of

HPV infection are also seen in other primary immunodeficiency syndromes with diverse underlying defects such as DOCK8 and GATA2 deficiency (38).

1.3.2 With secondary immunodeficiencies

1.3.2.1 Human immunodeficiency virus (HIV)

HIV is an enveloped, single-stranded RNA retrovirus of the genus Lentivirus which causes the acquired immunodeficiency syndrome (AIDS) (39, 40). It productively infects CD4+ cells (41), causing CD4+ T-cell depletion and a clinical syndrome characterised by opportunistic infections, virally-induced cancers and dementia (Table

1.3) (42).

In the absence of treatment, the typical time course of HIV infection and the subsequent

CD4+ T-cell depletion and increasing susceptibility to opportunistic diseases is illustrated in Figure 1.4 (43). This CD4+ T-cell decline became reversible and preventable with the advent of effective combination antiretroviral therapy (cART) in the late 1990s (44).

Unlike invasive cervical cancer, anal cancer is not regarded as an AIDS-defining malignancy (probably for historical reasons e.g. lead time and length bias), although in many ways it behaves like one (e.g. caused by infection and associated with higher incidence in immunodeficiency).

HIV-infected women have a 6-fold increase in the incidence of cervical cancer compared to the general population, while HIV-infected adults have a 29-fold increase in the risk of anal cancer (45). HIV-infected MSM have a very high incidence of anal cancer (77.8/100,000; 95% confidence interval [CI] 59.4-96.2) despite cART (46). This 30

rate is 2.9 times higher than the incidence of cervical cancer was before the introduction of Papanicolaou testing (47). The epidemiology of anal cancer will be discussed in detail in a subsequent section.

1.3.2.2 Transplant recipients

Solid organ transplant recipients have a 2-fold increase in the incidence of cervical cancer compared to the general population, and a 5-fold increase of anal cancer (45).

Figure has been removed due to copyright restrictions.

Figure 1.4 – Typical course of HIV infection

From Annals of Internal Medicine, Fauci AS, Immunopathogenic Mechanisms of HIV

Physicians, Inc. (48).

Table 1.3 – Stage-3 (AIDS)-defining opportunistic illnesses in HIV infection Bacterial infections, multiple or recurrent* Candidiasis of bronchi, trachea, or lungs Candidiasis of esophagus Cervical cancer, invasive† Coccidioidomycosis, disseminated or extrapulmonary Cryptococcosis, extrapulmonary Cryptosporidiosis, chronic intestinal (>1 month’s duration) Cytomegalovirus disease (other than liver, spleen, or nodes), onset at age >1 month Cytomegalovirus retinitis (with loss of vision) Encephalopathy attributed to HIV Herpes simplex: chronic ulcers (>1 month’s duration) or bronchitis, pneumonitis, or esophagitis (onset at age >1 month) Histoplasmosis, disseminated or extrapulmonary Isosporiasis, chronic intestinal (>1 month’s duration) Kaposi sarcoma Lymphoma, Burkitt (or equivalent term) Lymphoma, immunoblastic (or equivalent term) Lymphoma, primary, of brain Mycobacterium avium complex or Mycobacterium kansasii, disseminated or extrapulmonary Mycobacterium tuberculosis of any site, pulmonary†, disseminated, or extrapulmonary Mycobacterium, other species or unidentified species, disseminated or extrapulmonary Pneumocystis jirovecii (previously known as “Pneumocystis carinii”) pneumonia Pneumonia, recurrent† Progressive multifocal leukoencephalopathy Salmonella septicemia, recurrent Toxoplasmosis of brain, onset at age >1 month Wasting syndrome attributed to HIV * Only among children aged <6 years. † Only among adults, adolescents, and children aged ≥6 years.

Immune responses to HPV have been studied predominantly in the context of cervical disease. While the HPV life cycle may be similar across epithelium at different sites, the natural history of AIN may be different from that of CIN because the anatomy, physiology (e.g. hormonal differences) and immune responses of the anus and cervix are different (49).

1.4.1 Innate

Innate immunity refers to the components of the immune response that are non-specific and does not require priming. Innate immunity at the mucosa is the first line of host defence against HPV infection (Figure 1.5). Components of innate immunity in the anus include the physical epithelial barrier, antimicrobial peptides (e.g. β-defensins)

(50), Toll-like receptors (TLRs) and various cells (phagocytes, natural killer [NK] cells, intraepithelial lymphocytes).

Oncogenic HPVs are adept at evading the innate immune response, delaying the activation of adaptive immunity (51). Studies using transfected cervical cancer cell lines and ex vivo cervical cancer specimens have shown that HPV16 oncogenes function to evade tumour necrosis factor (TNF)-mediated pro-apoptotic signals, inhibit the major antiviral cytokine interferon-γ (IFNγ) production in peripheral blood mononuclear cells

(PBMCs) and NK cells, and inhibit TLR9 transcription, resulting in suppression of activation of signalling pathways that result in acute inflammation, stimulation of adaptive immunity and induction of an antiviral state (52-54). HPV16 and 18 have more recently been shown to also dampen the downstream signalling of a broad range of innate immune receptors, leading to downregulation of a network of genes (centred

around interleukin-1β and interleukin-6) involved in antigen presentation, the inflammasome, and the production of antiviral, pro-inflammatory and chemotactic cytokines (55). These immune evasion strategies of oncogenic HPVs steer the host epithelial cell towards dysregulated proliferation and malignant transformation.

Figure 1.5 – Immune responses to HPV16 infection

Electron micrograph of HPV from http://en.wikipedia.org/wiki/File:Papilloma_Virus_%28HPV%29_EM.jpg [accessed 2

July 2012]. Source: NIH-Visuals Online #AV-8610-3067, Laboratory of Tumor Virus

Biology.

1.4.2 Humoral

Systemic antibody responses to natural HPV infection have been studied mostly in the context of women and cervical disease. IgG antibodies to L1 are HPV-type-specific, but are low titre and it remains unclear whether they protect against reinfection (56).

Furthermore, seroconversion is incomplete, with only 59.5% of women with incident

HPV16 infection (defined by DNA positivity in the cervix or vagina) positive for serum

HPV16-IgG at 18 months follow-up (57). Seroconversion is more likely to be persistent with HPV16/18 compared to HPV6, although the stability of serum HPV16-IgG levels over time is not established (57). There are very little data published regarding the natural humoral immune response to anal HPV infection, but a recent study in men showed seroprevalence varied by site of HPV infection (anal vs. external genitalia only)

(58), supporting the idea that site of HPV infection influences the immune response and therefore natural history of any subsequent disease.

HIV infection is associated with higher prevalence of HPV16-IgG seropositivity in large studies of HIV-infected compared to HIV-uninfected women (59-61). This most likely reflects differences in risk of acquisition and persistence of HPV16 infection between the two populations. There are no data to suggest HIV infection per se impairs systemic antibody responses to HPV.

Mucosal antibody responses to HPV16 have also been studied in the cervix. After incident HPV16 DNA detection, HPV16-specific IgA in cervical secretions is detected earlier than serum IgA (median 10.5 versus 19.1 months, P=0.01), but this IgA response is relatively short-lived in both the cervical and serum compartments (median 12.0 and

13.6 months, respectively) compared to serum IgG (62). HIV infection does not impair the serum HPV16-IgA response (63, 64). However, acute and chronic HIV infection impairs the cervical production of HPV16-IgA (65, 66). The differences between

humoral responses to HPV infection in the anus compared to the cervix, and the impact of HIV infection, remain to be defined.

1.4.3 Cellular

The cellular immune response is implicated in the control and outcome of established

HPV infection. In animal models, systemic papillomavirus-specific T-cell responses peak with spontaneous wart regression (67). CD4+ T-cell infiltration during wart regression has been described in both animal models and human genital warts (68, 69).

Our understanding of how HPV-specific T-cells control and may clear mucosal HPV infection and neoplasia is incomplete (70). Circulating cytotoxic CD4+ and CD8+ T- cell responses (71) to HPV16 E6 and E7 in HPV16-positive women correspond with

CIN absence (72), and lack of cytotoxic responses to E6 associates with persistent

HPV16 infection (73). T-cell responses to HPV16 E2, E6 or E7 are absent or impaired in cervical cancer compared with healthy controls, and associates with persistence or progression of low-grade CIN (74, 75). Several studies have demonstrated that IFN-γ- producing, HPV16/18-specific CD4+ T-cells are important in protecting against intraepithelial neoplasia (76-78). Furthermore, memory CD4+ T-cell responses specific for HPV16/18 E2, E6 and/or E7 are detectable in 20-50% of healthy blood donors (76,

79).

Immunohistochemistry studies of local T-cell responses in cervical SIL and cancer biopsies show that CD4+ T-cells predominate in regressing SIL, while CD8+ T-cells predominate in invasive cancer. In 7 out of 10 CIN3 biopsies, CD4+ and CD8+ T-cells were visible as organised lymphoid follicular aggregates (80). Intraepithelial cytotoxic

CD8+ T-cells were associated with regressing low-grade cervical SIL in a prospective

study of 125 women (81). There are very few data regarding local T-cell responses in anal SIL and cancer (see Chapter 4).

The outcome of the anti-tumour immune response in the mucosa depends on a complex interplay of regulatory versus effector cells, as well as the local microenvironment (e.g. presence of inflammation). To date, attempts to correlate systemic HPV16-specific T- cell responses with CIN regression have had limited success, perhaps partly because therapeutic resection of high-grade CIN is mandatory (82, 83). HPV-specific CD4+ regulatory T-cells induced by natural infection have been shown to suppress the

HPV16-specific effector response in CIN and cervical cancer (84-86).

HIV-induced CD4+ T-cell immunodeficiency is a risk factor for detection of anal HPV,

AIN and anal cancer, in keeping with a central role of CD4+ T-cells in the control and outcome of HPV infection. In HIV-infected adults, a CD4+ T-cell count <200 cells/μL is a risk factor for detection of anal HPV (87-89), AIN (90-92) and anal cancer (1, 93).

Lower nadir CD4+ T-cell count (<50 cells/μL) is independently associated with anal high-grade disease (94) and incident anal cancer (hazard ratio [HR] 0.85 per 50 cells)

(93) in HIV-infected men. However, other unknown factors must also be important as twenty years of cART has not reduced the incidence of anal cancer. For example, regardless of the peripheral CD4+ T-cell count, HIV infection affects the quality of the cervical mucosal cellular immune response e.g. CD8+ T-cell dominant aggregates, less likely to have classical germinal centres, lower immune cell densities and IFNγ and regulatory cytokine expression (95, 96).

1.5.1 Epidemiology

1.5.1.1 Anal HPV and AIN

Anal canal HPV is almost universal (≥90%) in HIV-infected adults (Tables 1.4 and 1.5)

(46, 97), most of whom test positive to at least one high-risk HPV type. The only published prospective study estimated anal HPV16 incidence in HIV-infected men who have sex with men (MSM) at 13.0% per year and clearance at 14.6% per year (98). Co- infection with multiple HPV types is common (45-86% of HIV-infected men), with a mean seven HPV types detected intra-anally in HIV-infected MSM (87, 99).

Given the >90% prevalence of anal canal HPV detected by PCR in HIV-infected adults, it is clear that not all detected HPV results in AIN. In a recent meta-analysis, there was no statistically significant difference in the prevalence of anal HSIL between HIV- infected and HIV-uninfected MSM (46). While the small number of studies available for meta-analysis may account for this finding, it is plausible that the immune deficits associated with HIV infection are not the most significant determinants of high-risk

HPV types causing anal HSIL. For example, some HPV16 variants (subtypes) are associated with an increased risk of anal HSIL independent of HIV status and CD4+ T- cell count in MSM (100).

A risk factor for acquiring anal HPV is receptive anal intercourse (87, 88), but anal

HPV infection can be acquired in the absence of anal intercourse, suggestive of “field infection” (90). Risk of anal HPV detection declines with age in women (88, 89), whereas this is not the case in men. It is possible that this difference may be attributed to older homosexual men having higher numbers of sexual partners than older women

(101). In women, abnormal anal cytology is associated with abnormal cervical cytology or HPV detection (91).

1.5.1.2 Anal cancer

The incidence of anal cancer in the general population in Australia and other developed countries has increased over the last 25 years by around 50% (102, 103), but it is still an uncommon cancer with current annual incidence rate estimates between 0.8 to 1.8 cases per 100 000 (2, 102). In most developed countries, the incidence of anal cancer is higher in women compared to men (2, 102, 103), although the rate of increase is faster in men

(103). While older age (e.g. >65 years) is still associated with the highest incidence rates of anal cancer, the age at presentation has become much younger (median age at diagnosis in Denmark was 63 years in 1978-87, compared to mean age at diagnosis in a

United States AIDS population of 41 years in 1978-1996) (103-105).

The incidence of anal cancer in HIV-infected adults is about 30-fold higher than in the general population (106). In MSM, incidence of anal cancer in the HIV-infected is approximately five-fold that of HIV-uninfected (107). Women with cervical, vulvar or vaginal cancer have a 13-fold increase in risk of anal cancer compared to unaffected women (108). Since anal cancer is an infection-driven malignancy, it is surprising that incidence rates have not declined in the cART era (1, 106, 109, 110). Possible reasons for this include increased life expectancy with cART enabling persistent HPV infection to develop into anal cancer, improved health outcomes permitting increased risk behaviour leading to higher HPV incidence, the lack of protective effect of cART itself in the mucosa against HPV infection, and perhaps the restoration of systemic CD4+ T- cell counts from a low nadir does not restore local, HPV-specific immune responses in the anal mucosa responsible for clearance of established HPV infection.

Table 1.4 – Prevalence, incidence and risk factors for anal canal HPV, anal intraepithelial neoplasia (AIN) and anal cancer in HIV-infected men Anal Canal HPV AIN Anal Cancer Population Population Population Any abnormality 57.2%# (cytology) Any HPV 63.1%# (histology) 92.6%# Low-grade disease Prevalence MSM (46) 27.5%# (cytology) MSM (46) HPV16 31.3%# (histology) 33.8%# High-grade disease 6.7%# (cytology) 29.1%# (histology) HPV16 MSM (46) High-grade disease 45.9 / 100,000 PY# Incidence 5.9 – 10.8 /1000 person-mths MSM (98) MSM (46) Men (non-MSM) 8.5 – 15.4% per year 18.2 – 99.9 / 100,000 PY 4.4 / 1000 person-mths Men (111) (1, 105, 112) Any HPV MSM(87) Men MSM† RAI last 12 months (87) Men High-grade disease Older age† hr-HPV Older age^ Baseline CD4<200-350† Adults Risk Current CD4<200 (87) Men Nadir CD4<50 ART^ MSM (94)^ Lower nadir CD4† (1, 93, 105, 109, Factors* Anal bleeding past year (99) MSM Current ART <4 years^ Prior AIDS diagnosis† 112)† HPV16/18 Anal HPV16/18^ HIV diagnosis before Current CD4<250 (90) Male IDU 1996† Previous AIDS diagnosis (90) Male IDU CD4 = absolute CD4+ T-cell count (cells/μL); hr-HPV = high-risk type HPV; IDU = injecting drug user; RAI = receptive anal intercourse. #Meta-analytic estimate. *Risk factors are from cross-sectional studies unless otherwise specified. ^Prospective study. †Cohort studies.

Table 1.5 – Prevalence, incidence and risk factors for anal canal HPV, anal intraepithelial neoplasia (AIN) and anal cancer in HIV-infected women Anal Canal HPV AIN Anal Cancer Population Population Population Any abnormality 10.5 – 31% Any HPV Women Low-grade disease Women Prevalence 80-90% (92, 97) 12% (92, 113) High-grade disease 4.1 – 9% Any cytological abnormality Women Incidence 3.9 – 30 / 100,000 PY Women (1, 105) 22% per year (114) Any cytological abnormality CD4<200^ Current smoker^ Any HPV History of other STI^ MSM† History of anal intercourse (88) Anal HPV+^ Women Older age† Risk Current CD4<200 (88, 89) Adults Women Low-grade disease (91, 92, Baseline CD4<200† Factors* Younger age (88, 89) (1, 93, 105, 109)† History of RAI 114)^ Lower nadir CD4† White (race/ethnicity) (89) Younger age Prior AIDS diagnosis† Cervical HPV+ by PCR (89) Cervical or anal HPV+ High-grade disease Anal HPV+ CD4 = absolute CD4+ T-cell count (cells/μL); RAI = receptive anal intercourse; STI = sexually transmitted infection.

*Risk factors are from cross-sectional studies unless otherwise specified. ^Prospective study. †Cohort studies

There are no direct estimates of the progression rate from anal HSIL to anal cancer in

MSM. In predominantly female cohorts, Watson et al found eight of 72 patients (11%) with high-grade AIN progressed to cancer over a median follow up of 60 months, and

Scholefield et al found three of 35 patients progressed followed over a median of 63 months. (115, 116) Machalek et al calculated a progression rate to be 1 in 377 per year in HIV-infected MSM (compared to 1 in 4196 per year in HIV-uninfected MSM) (46).

These rates are somewhat lower than estimates of grade 3 cervical intraepithelial neoplasia (CIN3) progression to cancer (around 1% per year in HIV-uninfected women)

(117).

Given that the prevalence of anal HSIL in HIV-infected MSM is up to 53% (94), and the annual incidence rate of anal cancer in this population is 46 to 131 cases per 100 000

(1), it is highly likely regression of anal HSIL occurs even in the setting of HIV infection, as with CIN. However, there are no published estimates of regression rates, or risk factors for regression compared to anal HSIL persistence or progression to cancer

(Figure 1.6).

Figure 1.6 – Natural history of high-risk HPV infection

In the cervix (A) and the anus (B). Incidence rates are in blue, prevalence data are purple (46, 117, 118).

1.5.2 Pathogenesis

The biology of HPV-related squamous disease is thought to be similar across lower anogenital sites (3). The HPV life cycle is dependent upon epithelial differentiation, and

HPV replicates exclusively within the epithelial compartment (24). HPV infects basal keratinocytes through discontinuations in the overlying epithelium (microabrasions).

Recently, glycosaminoglycans and laminin have been described as “receptors” for HPV in the basement membrane, which lead to a conformational change of the viral capsid, enabling HPV to enter keratinocytes by binding to cell membrane proteins such as α6β4 integrin (119, 120).

The possible outcomes of HPV infection are clearance, productive infection (e.g. exophytic warts), latency, and persistence. Ultimately, persistence can cause intraepithelial neoplasia and cancer. The key step resulting in high-grade disease is deregulated expression of the HPV oncogenes E6 and E7 (Figure 1.7), leading to

increased but disorganised proliferation of the lower epithelial cell layers characteristic of high-grade disease (24). The E6 oncoprotein binds to the p53 tumour suppressor protein, causing deregulation of DNA damage and apoptotic pathways (121). E6 also activates telomerase, which is an intrinsic step in cell transformation and immortalisation (16). The E7 oncoprotein inactivates retinoblastoma protein (pRb) tumour suppressor functions by releasing E2F transcription activators (122). E7 also targets DNA methyltransferase, disrupting maintenance of CpG methylation patterns following cell replication (123). A recent study using a HPV K14 transgenic mouse model for anal cancer found that E7 may be a more potent oncogene than E6 (121).

Figure has been removed due to copyright restrictions.

Figure 1.7 – Functions of the E6 and E7 oncoproteins, and their interaction with each other in steps that lead to cell immortalisation

Reprinted by permission from Macmillan Publishers Ltd: Nature Reviews Cancer. zur Hausen H. Papillomaviruses and cancer: from basic studies to

1.5.3 Diagnosis

Anal HPV infection per se is asymptomatic, but low-risk type HPVs cause exophytic warts that may be palpable. In addition, anal HSIL and early anal cancer can be incidental findings on histopathology after common surgical procedures such as haemorrhoidectomy or wart excision. Anal cancer often presents late (39% stage 3A or higher at diagnosis) (124) with a lump, bleeding, incontinence from sphincter infiltration, fissure or fistula and pain, or with lymph node or distant metastases, but also non-specific symptoms such as pruritus, discomfort, pelvic mass sensation or change in bowel habit (125).

Anal SILs are diagnosed by biopsy or cytology. During high-resolution anoscopy

(HRA, Figure 1.8), a procedure adapted from cervical colposcopy, biopsies are taken intra-anally from the specialised non-keratinising squamous epithelium just distal to the anal squamocolumnar junction (SCJ, also known as the transitional or transformation zone) as well as from the keratinised epithelium of the anal verge and perianus.

An anoscope coated with lignocaine hydrochloride 2% gel is inserted into the anal canal, and the anal mucosa is visually inspected under magnification with a colposcope.

Acetic acid 3% followed by Lugol’s iodine is applied topically with cotton swabs to stain the SCJ, and HPV-related high-grade abnormalities of the anal mucosa can be identified by vascular patterns, lesion features (such as surface contours, punctation and ulceration) and staining characteristics (126). Small biopsies (around 2-6 millimetres) of abnormal areas are taken with Tischler’s or endoscopy forceps.

Figure 1.8 – High-resolution anoscopy

Panel A shows a clinic room prepared for HRA. The person being examined lies on the bed in the left lateral position. Panel B shows a disposable proctoscope which is inserted into the anal canal during HRA. Panel C is a digital photograph of the anal canal as seen during HRA with the colposcope (10x magnification). 3% acetic acid stain has whitened the stratified squamous epithelium, revealing the squamocolumnar junction.

Exfoliative cytology is another method of diagnosing HPV-related abnormalities of the anal mucosa and, if performed in conjunction with HRA, should be sampled first. Anal epithelial cells are collected by scraping a water-moistened, synthetic-fibre swab circumferentially around the anal canal (127). The swab is then processed as for cervical

Papanicolaou smears (a liquid-based preparation is preferred as it increases cell yield and decreases faecal contamination and air-drying artefact) (128). Terminology for anal cytological diagnosis is based on the Bethesda system for cervical smears (6). While the term “squamous intraepithelial lesion” (SIL) implies a cytological abnormality whereas

“anal intraepithelial neoplasia” implies a histological abnormality, usage of the terms interchangeably is common in the literature (8). Compared to cervical cytology, anal cytology is significantly less sensitive and specific for detecting high-grade disease at every cut-off (area under receiver operating characteristic curve 0.70 for anal cytology compared to 0.86 for cervical cytology) (129).

Anal HPV is detected by extraction of DNA from anal epithelia collected for cytology in a liquid-based medium, followed by HPV DNA detection and typing (mainly by polymerase chain reaction [PCR]) (130). As this test is not currently approved for routine diagnostic use, many laboratories will not perform this testing or if they do so, they will require a separate sample to the cytology sample.

1.5.4 Treatment

There is controversy as to whether treatment of anal HSIL is of benefit. At present, treatment options include “watchful waiting” (with HRAs at 6-12 monthly intervals in

HIV-infected patients with anal HSIL), surgical excision (but 15% developed secondary anal stenosis or faecal incontinence and 63% recurrence within a year) (125), and a variety of topical (imiquimod, trichloroacetic acid, 5-fluorouracil) or ablative (infrared 48

coagulation (131), laser, electrocautery) (132) therapies. The American Society of

Colon and Rectal Surgeons make strong recommendations for treatment of both low and high-grade anal SIL with topical 5% imiquimod cream, topical 5% 5-fluorouracil cream or targeted destruction all with close long-term follow-up, but state that all of these recommendations are based on low-quality evidence (133). Weak recommendations are made for observation alone, photodynamic therapy, close clinical long-term follow up and vaccination against HPV16/18 in high-risk patients such as

HIV-infected patients and MSM. In comparison, excision or ablation is strongly recommended for treatment of women with CIN2/3 based on level I evidence from at least one randomised, controlled trial (134).

The aim of treating anal HSIL is to prevent progression to anal cancer. However, there is a lack of adequately powered, randomised treatment trials of anal HSIL in both HIV- infected and uninfected adults that establish the efficacy of treatment. A recent

Cochrane review of interventions for AIN found only one randomised control trial which met the inclusion criteria (135). This trial of 53 HIV-infected MSM compared intra-anal topical imiquimod cream with placebo, and showed imiquimod had an effect in downgrading anal HSIL to LSIL compared to placebo but did not demonstrate persistence of this benefit (136). Also, interpretation of study findings is difficult due to early termination of the trial (135). Robust and well-designed clinical trials of anal

HSIL treatment to prevent anal cancer are needed.

The treatment of anal cancer in HIV-infected adults is the same as that for HIV- uninfected – combined modality chemoradiation with 5-fluorouracil and mitomycin C, with surgical excision reserved for small tumours of the anal margin which can be adequately resected without affecting sphincter function (137). Complete response occurs in 84-100% (depending on stage at diagnosis), and overall survival for HIV-

infected patients is 71-92% at 3 to 5 years, similar to 77-84% in the HIV-uninfected with similar acute and late toxicities in both groups (138, 139).

1.5.5 Prevention

1.5.5.1 Screening

Digital anal-rectal examination, anal cytology, PCR for HPV DNA, and HRA have been proposed in various combinations to screen for anal cancer in high-risk populations such as HIV-infected adults. However, currently there is no evidence base to determine if anal cancer screening is effective. There are limited data on the natural history of AIN and anal cancer, and so cost-effectiveness analyses of screening have yielded conflicting conclusions (140, 141).

Current national guidelines do not recommend cytology or HRA-based anal cancer screening (142). The European AIDS Clinical Society recommends a digital anal-rectal examination with or without anal cytology every 1-3 years in HIV-infected MSM, and

HRA if cytology is abnormal (based on expert opinion with unknown benefit) (143).

Based on expert opinion, the New York State Department of Public Health AIDS

Institute recommends an annual evaluation of all HIV-infected adults including anal symptoms, inspection of the perianal area and digital anal-rectal examination (144). In addition, annual anal cytology (and referral for HRA if any abnormality) is recommended for HIV-infected MSM, for any patient with a history of anogenital warts, and for women with abnormal cervical or vulvar histology.

Anal cytology is the best studied screening test, using HRA-directed biopsy as the gold standard. In HIV-infected MSM, the sensitivity of atypical squamous cells of undetermined significance (ASC-US) or worse on anal cytology for high-grade disease

is 87% and specificity is 42% (127), similar to figures in a systematic review (145).

This low specificity limits the usefulness of anal cytology on its own as a screening test as it would lead to many “unnecessary” referrals to HRA, an uncomfortable and costly procedure. How the introduction of prophylactic HPV vaccination will affect the performance of anal cytology and other screening methods in HIV-infected adults remains to be determined.

1.5.5.2 Vaccination

Virus-like particles (VLPs) are the basis of the quadrivalent (HPV16/18/6/11) and bivalent (HPV16/18) prophylactic vaccines that are now licensed worldwide (119). A 9- valent vaccine has also demonstrated efficacy in covering an additional five high-risk type HPVs in women (146). VLPs are synthetic type-specific HPV L1 proteins that self- assemble into non-infectious particles which are structurally analogous to the HPV viral capsid (147). Serum antibody responses to vaccination with HPV VLPs are high titre, robust and protective. In a study of young HIV-uninfected MSM, the quadrivalent vaccine prevented AIN associated with HPV16/18/6/11 in 77.5% (95% CI 39.6-93.3) of participants who had no evidence of previous infection with the four vaccine HPV types

(148). The bivalent vaccine has also demonstrated efficacy against anal HPV16/18 infection in HPV-naive young women (149). Despite demonstration of clinical efficacy at preventing HPV type-specific infection and induction of high titres of neutralising type-specific and cross-type serum antibodies in a range of populations, an immune correlate (e.g. antibody concentration, affinity or avidity) of vaccine-derived or infection-derived protection has not been defined (51).

To date, no efficacy trial has been published for the prophylactic HPV vaccines in HIV- infected adults. The standard vaccine regimen (three doses over 24 weeks) is safe and highly immunogenic in HIV-infected men, with HPV16 seroconversion in 100% of

participants who were HPV16 DNA negative and seronegative at enrolment (150). This

HIV-infected study population was highly treated (baseline cART use was 84% with

83% having suppressed plasma HIV viral load) and relatively immunocompetent

(baseline median CD4+ T-cell count 517 cells/μL and 87% with no history of AIDS- defining illness). A study in HIV-infected women also demonstrated safety and immunogenicity, although women with HIV viral load >10,000 copies/mL and/or CD4+

T-cell count ≤200 cells/μL had lower seroconversion rates (151). The United States

Centres for Disease Control and Prevention (CDC) currently recommends HPV vaccination for all HIV-infected adults through age 26 regardless of CD4+ T-cell count and for all MSM through age 26 (152).

Despite the incomplete understanding of cell mediated immunity in the clearance of

HPV-related neoplasia, a recent clinical trial of therapeutic vaccination with HPV16 E6 and E7 peptides was successful in inducing regression of grade 3 vulvar intraepithelial neoplasia in HIV-uninfected women (153), and regression correlated with the kinetics and phenotype of HPV16-specific T-cell responses (154).

The idea that the immune system can control cancer was first predicted by Paul Ehrlich in the first decade of the 20th century (155), and later expanded on by Macfarlane

Burnet and Lewis Thomas in the 1950s (“cancer immunosurveillance”) (156, 157).

Observations that support this hypothesis include: spontaneous regression of cancers

(158), paraneoplastic syndromes (159-162), higher incidence of cancers in immunosuppressed populations (45, 163, 164), prognostic significance of tumour- infiltrating lymphocytes (165, 166), and the recent advent of increasingly effective

cancer immunotherapies such as the so-called “immune checkpoint inhibitors” (167-

169).

The concept and definition of spontaneous regression in the setting of malignancy is controversial, but Everson and Cole’s definition in 1956 is still relevant and elegant.

They defined spontaneous regression of cancer “as the partial or complete disappearance of a malignant tumour in the absence of all treatment, or in the presence of therapy which is considered inadequate to exert a significant influence on neoplastic disease” (170). They emphasised spontaneous regression need not be complete, and is not synonymous with cure – it may occur synchronously or metachronously with persistence or progression of the same disease process.

Certain human malignancies have been reported to spontaneously regress at least partially, usually in association with an immune response. The prototypical example is melanoma. A 2009 review by Kalialis et al found 76 case reports of spontaneous regression of metastatic melanoma since 1866 (171). From 1960s Australian data,

McGovern estimated 13% of primary cutaneous melanomas partially regressed (172), a figure not dissimilar to a more recent Italian cohort where 349 of 1693 (21%) patients with American Joint Committee on Cancer (AJCC) stage I-II melanoma showed histological features of regression (173). Moreover, histological regression was an independent favourable factor in predicting overall survival (HR 0.43, P = 0.008). This supports earlier work showing the prognostic value of tumour-infiltrating lymphocytes

(TILs) in predicting 5- and 10-year survival rates, particularly if the pattern of infiltration was dense and diffuse throughout the tumour (termed a “brisk” pattern)

(166).

Other malignancies that are known to be immunogenic and have been reported to spontaneously regress in adults include renal cell carcinoma (174), thymic carcinoma 53

(175, 176) and virally-induced cancers such as hepatocellular carcinoma (177), lymphoma (178, 179) and Merkel cell carcinoma (180, 181) (158). Data on spontaneous regression of cancer or precancer from randomised controlled trials of treatment that had placebo arms are summarised in Table 1.6.

Table 1.6 – Incidence of cancer or precancer spontaneous regression in placebo arms of randomised controlled trials (RCT) of treatment Reference Year Disease Total N Placebo Incidence of randomised arm (n) spontaneous regression Correa 2000 Gastric cancer 976 117 7% (182) precursors Elhilali 2000 Metastatic renal 197 99 7% (174) cell carcinoma Prince 2012 CD25+ mycosis 144 44 15.9% (179) fungoides / Sézary syndrome

Autoimmune paraneoplastic syndromes associated with spontaneous regression of malignancies in adults also point to the immune response as a key player. For example, anti-Hu antibodies with encephalomyelitis are associated with spontaneous regression of small cell lung cancer (183), and are thought to confer a better prognosis (159).

Another approach is to compare epidemiological incidences of cancers in immunosuppressed populations with the general population. Grulich et al published a meta-analysis in 2007 that found infection-related cancers occurred with increased incidence in HIV-infected patients and transplant recipients, which might be expected given immunosuppression increases susceptibility to infections (45). However, four other cancers (kidney, multiple myeloma, leukaemia and melanoma) were also

increased in both populations, implying the importance of the host immune response in the pathogenesis of these particular malignancies.

Even in solid tissue cancers not traditionally thought to be “immunogenic”, the presence of TILs has been found to confer a favourable prognosis. For example, high CD45-RO+

(memory T-cell marker) TIL density independently predicted overall survival in colorectal cancer (P = 0.015) in the Nurses’ Health Study and Health Professionals

Follow-up Study (165), supporting the earlier findings by Galon et al (184). Similarly,

TILs have been shown to predict a favourable prognosis in breast (185), ovarian (186) and hepatocellular (187) carcinoma. However, in other settings TILs were predictive of poorer prognosis e.g. shorter relapse-free survival in surgically-treated local prostate cancer (188), and several studies suggested the ratio of regulatory to effector T-cells was of prognostic significance e.g. in small cell lung (189) and ovarian (190) cancer.

Therefore, there is likely to be a complex interplay between the tumour and host immune response that ultimately determines the outcome of premalignant and malignant lesions.

Despite the clues in these clinical and epidemiological observations, the hypothesis of an anti-tumour immune response remained controversial and lost momentum over decades as experiments in mouse models failed to support it. However, it has regained currency in the last two decades with improved mouse models of immunodeficiency, leading to the current paradigm of cancer immunoediting.

1.6.1 Cancer immunoediting

The cancer immunoediting hypothesis emphasises the dual host-protective and tumour- promoting actions of the anti-tumour immune response. The immune system is not only important in preventing cancer formation; it also plays a key role in the pathogenesis of cancer by shaping tumour immunogenicity. This concept arose from initial experiments by Shankaran et al who compared the development of carcinogen-induced sarcomas in immunodeficient (Rag2-/-) and immunocompetent (wild-type) mice (191). While immunodeficient mice were more susceptible to developing carcinogen-induced sarcomas compared to immunocompetent mice, when tumour cell lines derived from the two groups were injected into new immunocompetent mice, they behaved differently implying the immunogenicity of the tumour differed depending on the immune status of the original host in which the tumour arose in response to the same carcinogen (192). Some tumours that arose from an originally immunodeficient mouse regressed when injected into a new immunocompetent mouse, implying it was

“unedited” as it had developed on an immunodeficient background, and presumably displayed epitopes which the now immunocompetent mouse’s immune system could target. Tumours that arose from an originally immunocompetent mouse always progressed when injected into the new mouse (“edited”). In humans, an example of this phenomenon is the transfer of melanoma in donor organs to recipients, where the original donor had been cured of primary melanoma, but recipients develop fatal metastatic melanoma even if years had lapsed between the cured primary melanoma and organ transplantation (193).

Three phases of cancer immunoediting are described – elimination (an extrinsic tumour suppressor mechanism), equilibrium (adaptive immune response that sculpts tumour immunogenicity to prevent progression), and escape (tumour evades adaptive immunity

to progress to clinical disease). These phases are illustrated in Figure 1.9. Avoiding immune destruction and tumour-promoting inflammation are now recognised as hallmarks of cancer, alongside more traditional characteristics such as enabling replicative mortality and inducing angiogenesis (194).

In the setting of anal HPV16 and HSIL, these basic paradigms may underpin the natural history and risk factors for HSIL regression or progression to cancer, and illustrate why

HPV-induced precancer behaves differently in immunodeficient compared to immunocompetent hosts.

1.6.2 Cancer immunotherapy

William Coley first reported successfully treating ten cases of malignant sarcoma with bacterial toxins in 1893 (195, 196). The recent “renaissance of cancer immunotherapy”

(197) has led to a wide variety of more specific approaches in successfully manipulating the immune system to treat human malignancies. For example, clinical trials of therapeutic monoclonal antibodies targeted at PD-1 and CTLA-4 (so called “immune checkpoint inhibitors”) have shown efficacy in metastatic melanoma (167, 168) although not without significant immune-related adverse effects (198, 199). Adoptive

T-cell immunotherapy is also proving possible for both virally-driven [e.g. EBV (200,

201)] and epithelial cancers (202). Therapeutic vaccination using defined tumour antigens has also been successful in phase I and II clinical trials. For example, a 47% complete response rate at 24 months was achieved in 20 women with HPV16+ high- grade vulvar intraepithelial neoplasia vaccinated with synthetic long peptides of HPV16

E6 and E7 (153).

Figure has been removed due to copyright restrictions.

Figure 1.9 – Cancer immunoediting

Reproduced from Schreiber RD, Old LJ, Smyth MJ. Cancer immunoediting: integrating immunity's roles in cancer suppression and promotion. Science 2011 volume 331 page

1567 (192). Reprinted with permission from AAAS. Adapted from Vesely MD, Kershaw

MH, Schreiber RD, Smyth MJ. Natural innate and adaptive immunity to cancer. Annu

Rev Immunol. 2011;29:235-71.

The ability of T-cells to have memory in the secondary immune response was first described in a series of elegant mice experiments by Miller and Sprent in 1971 (203).

Since then, the measurement of antigen-specific T-cell responses has become a pillar of immunobiology research.

1.7.1 Functional T-cell assays

In general, functional T-cell assays involve firstly separating lymphocytes from whole blood, then stimulating the cells in culture with the antigen(s) of interest, then finally phenotyping the cells for markers of function.

Methods of T-cell separation were first developed in the 1970s, and included fractionating cell suspensions through nylon wool columns (204). However, the density- gradient, centrifugation methods (205) to isolate peripheral blood mononuclear cells

(PBMCs) have stood the test of time and are still standard procedures in (mostly research) immunology laboratories today. After separation, cells are cultured with antigen(s) for variable periods of time (depending on the read out used). Some assays may only require a few hours of incubation, whereas others may be cultured for days and require supplementation with various cytokines to ensure cell survival and proliferation (206). After culture, antigen-specific cells may be quantified with a variety of methods. 3H-thymidine assays to quantify lymphocyte proliferation were used for many decades, until the advent of ELISPOT assays to quantify cytokine (often IFNγ) production, and monoclonal labelling antibodies and flow cytometry. Combined with peptide-MHC tetramer technology from the mid-1990s (207), flow cytometry enabled the detection of low-frequency antigen-specific T-cells without the need for cell culture,

although it has proven to be far more feasible to make MHC class I tetramers and therefore identify antigen-specific CD8+ T-cells than MHC class II tetramers for antigen-specific CD4+ T-cells (208).

The literature on systemic HPV16-specific T-cell responses reflects the evolution of these methods, with most human studies published to date using data from 3H- thymidine, ELISPOT or intracellular cytokine staining (ICS) assays. Table 1.7 illustrates this by summarising the known MHC class II restricted immune epitopes of

HPV16 E6 and E7 and the assays used to define them, from the immune epitope database (http://www.iedb.org) (209).

Table 1.7 – MHC class II restricted immune epitopes of HPV16 oncoproteins Date range of Total number of Number of epitopes defined by assay publications epitopes defined 3H-thymidine ELISPOT ICS Other* E6 1996 - 2015 61 41 27 13 8 E7 1992 - 2015 37 20 10 7 8 *Other assays include: 51-chromium release, cytometric bead array and ELISA. Data from http://www.iedb.org (accessed 31 March 2015) (209).

While IFNγ-ELISPOT has been a frequently used assay in this field, multiparameter intracellular cytokine staining (ICS) is an alternative that offers several advantages. ICS has a shorter turnaround time (only six hours in culture), is compatible with whole blood (rather than requiring PBMC isolation), and it is possible to simultaneously describe the phenotype of antigen-specific T-cells by the cytokine(s) they produce and their surface markers (210, 211).

1.7.2 The CD25/CD134 assay

This is a simple in vitro assay on whole blood that detects antigen-specific CD4+ T-cell responses to a range of antigens with high specificity and sensitivity (212-215).

Antigens that have already been studied include mycobacteria, cytomegalovirus (CMV), herpes simplex virus-1, influenza, vaccinia (smallpox), tetanus toxoid, Candida albicans, streptokinase and HIV-1 peptides. A significant advantage of this assay is its simplicity – whole blood is cultured with the antigen of interest for 40-48 hours. Then antigen-specific CD4+ T-cells are identified by co-expression of surface CD25 and

CD134. CD25 is the α chain of the interleukin-2 (IL2) receptor and is a marker of T-cell activation (216, 217). CD134 (also known as OX40) is a member of the tumour necrosis factor receptor superfamily (218). It is a costimulatory molecule that is only expressed after T-cell activation and is essential for the late phase of memory T-cell generation.

CD134 is much more highly expressed on CD4+ compared to CD8+ T-cells, and is expressed on all CD4+ T-cell functional subsets (Th1, Th2, Th17, follicular helper and regulatory T-cells). The assay can be adapted to detect antigen-specific regulatory T- cells with the addition of CD39 (219). OX40 is a current molecule of interest in cancer immunotherapy as OX40 agonists have been shown in preclinical and phase I studies to enhance the anti-tumour immune response (220, 221).

Key unanswered questions include the natural history of anal SIL, progression of anal

HSIL to cancer and regression of anal HSIL and risk factors for these. Only with robust natural history data can evidence be established for an effective screening strategy and form the basis for randomised trials of treatment. Anal HSIL is a precancerous lesion for which treatment is not mandated, providing a unique opportunity to examine the role 61

of the immune system in dealing with virus-induced neoplasia and the impact of HIV infection. While the association between anal neoplasia, HPV and HIV has been known for more than two decades (222), much still remains to be understood. Given anal cancer is among the most common cancers occurring in people with HIV, an urgent expansion of research in this field is required.

This chapter was published in AIDS 2013;27(14):2233-43.

A PDF of the published article is included in Appendix B.

Objective: To quantify incidence of, and risk factors for, progression to and spontaneous regression of high-grade anal squamous intraepithelial lesions (ASILs).

Design: Retrospective review of patients at St Vincent’s Hospital Anal Cancer

Screening Clinic during a period when high-grade ASILs were not routinely treated

(2004-2011).

Methods: All patients who had an anal Papanicolaou smear or high-resolution anoscopy were included, except for patients with previous anal cancer. High-grade anal intraepithelial neoplasia (HGAIN) was defined as a composite of histologically- confirmed grade 2 or 3 anal intraepithelial neoplasia (AIN2/3) and/or high-grade squamous intraepithelial lesion on anal cytology. Analyses were repeated restricting to histologically-confirmed AIN3.

Results: There were 574 patients: median age 45 years (IQR 36-51), 99.3% male and

73.0% HIV-infected (median HIV duration was 13.8 years [IQR 6.4-19.8], median

CD4+ T-lymphocyte count was 500 cells/μL [IQR 357-662], 83.5% had undetectable plasma HIV viral load). Median follow-up was 1.1 years (IQR 0.26-2.76). Progression rate to HGAIN was 7.4/100 person-years (PY) (95%CI 4.73-11.63). No risk factor for progression to HGAIN was identified; progression to AIN3 was more likely with increasing age (Ptrend=0.004) and in those who were HIV-infected (hazard ratio 2.8

[95%CI 1.18-6.68] vs HIV-uninfected; P=0.019), particularly in those whose nadir

CD4+ T-lymphocyte count was <200 cells/μL (Ptrend=0.003). In 101 patients with

HGAIN, 24 (23.8%) patients had spontaneous regression (rate 23.5/100PY [95%CI

15.73-35.02]), mostly to AIN1. Regression was less likely in older patients

(Ptrend=0.048). Two patients with HGAIN developed anal cancer.

Conclusion: High-grade ASILs frequently spontaneously regress. Longer-term, prospective studies are required to determine whether these regressions are sustained.

In the era of combination antiretroviral therapy (cART), anal cancer has become the most common non-AIDS-defining cancer in HIV-infected adults in some developed countries (223). HIV-infected men who have sex with men (MSM) are at highest risk of anal cancer, with incidence rates approximately 15 times higher than HIV-uninfected

MSM (46). As with cervical cancer, anal squamous cell carcinoma (SCC) is caused by persistent infection with high-risk type human papillomavirus (HPV).

Anal SCC is preceded by a pre-cancerous phase of intraepithelial neoplasia termed high-grade squamous intraepithelial lesion (3), which is further sub-classified into grade

2 or 3 anal intraepithelial neoplasia (AIN2, AIN3). Similar to grade 3 cervical intraepithelial neoplasia (CIN3), AIN3 has the higher risk of progression to anal cancer.

The few natural history studies of anal HPV infection and anal squamous intraepithelial lesions (ASILs) report only progression rates to high-grade ASIL (94, 224). There are no published estimates of regression rates of high-grade ASILs, although an estimated

40% of CIN2 and 33% of CIN3 spontaneously regress (118).

CD4+ T-lymphocytic-predominant infiltrations are associated with regressing human genital warts (69). HPV type 16 (HPV16)-positive cervical cancers are associated with impaired CD4+ T-lymphocyte responses to HPV16 proteins (75). Therefore, it is biologically plausible that CD4+ T-lymphocyte immunodeficiency caused by HIV infection may impair host defences against HPV infection, leading to higher rates of progression to and persistence of intraepithelial neoplasia and lower rates of regression from intraepithelial neoplasia.

Anal cancer screening is not recommended within widely accepted national guidelines for the management of people with HIV (142, 143). Despite this, some centres with

appropriate expertise offer high-risk populations screening for ASILs with anal

Papanicolaou smears and/or high-resolution anoscopy (HRA) with biopsy of abnormal lesions (127). An anal cancer screening clinic was established at St Vincent’s Hospital in Sydney, Australia with one anoscopist (RJH). High-grade ASILs diagnosed at this clinic were not routinely treated because there is no high-quality evidence to indicate current treatments prevent anal cancer (135, 225). The practice at St Vincent’s is to discuss risk of progression with individual patients and, taking into account patient preferences, to co-manage those who prefer intervention with a colorectal surgeon.

We studied our cohort of men with untreated high-grade ASILs with the aim of describing, quantifying and examining risk factors for progression to and regression of high-grade ASILs. Our hypotheses were that a) high-grade ASILs can regress spontaneously, and b) the likelihood of progression to or regression of high-grade

ASILs is related to HIV status and degree of CD4+ T-lymphocyte immunodeficiency.

2.3.1 Study Population

All 574 patients who had an anal Papanicolaou smear or HRA performed at the St

Vincent’s Hospital anal cancer screening clinic from its inception in February 2004 to

21st January 2011 were included, except for seven patients with previous anal cancer.

The majority of men attending this clinic were MSM (>95%). Almost half (N=76) of our HIV-uninfected patients were participants in a clinical trial of the prophylactic quadrivalent HPV vaccine in preventing ASILs (148), and inclusion criteria for this trial included being aged 16 to 26 years with less than five lifetime sexual partners. St

Vincent’s Hospital Human Research Ethics Committee approval for the present analyses was granted on 6th December 2010 (File Number 10/210).

2.3.2 Study Design

This was a retrospective clinical audit of a cohort of patients with at least one visit for anal Papanicolaou smear or HRA. Patients were referred from within the St Vincent’s

Hospital HIV service and from local general practitioners specialising in HIV care.

Each patient’s enrolment date (hereafter referred to as baseline) was set as the date of their first anal Papanicolaou smear or HRA, whichever was first. Papanicolaou smear and HRA results within 90 days of each other were considered to be from the same visit. HRAs were performed as previously described (226), including staining of the intra-anal mucosa with acetic acid and Lugol’s iodine to identify areas of abnormality suspicious for ASILs. Biopsies were taken with Tischler’s forceps of these abnormal areas, particularly areas with vascular patterns or lesion features (such as surface contours, punctation and ulceration) and staining characteristics suggestive of high- grade ASILs. Liquid-based anal cytology and histology were reported by St Vincent’s

Hospital’s diagnostic laboratory (SydPath) using standard laboratory protocols developed for cervical reporting.

2.3.3 Data Collection

Medical records for all patients who attended the clinic were reviewed to confirm the dates of anal Papanicolaou smears and HRAs, Papanicolaou smear and HRA findings, previous anal surgery, HIV status and date of HIV diagnosis. Any purported results that could not be verified against the medical record were considered missing. All anal

cytology and intra-anal histology diagnoses (from HRA biopsies and anal surgery) were reviewed by one physician (WT) against the electronic medical record for accuracy and consistency with current terminology (the 2001 Bethesda System for cytology (6) and the Lower Anogenital Squamous Terminology Standardization Project for histology

(3)). CD4+ T-lymphocyte counts, HIV viral loads and date of HIV diagnosis were extracted from clinical and laboratory databases. Data were entered into a spreadsheet by an independent person. 100 (17%) patient records were audited to confirm accurate data entry.

2.3.4 Endpoint Definitions

The most abnormal result was used if histology results were available from more than one biopsy collected at any one HRA.

A composite endpoint of histologically-confirmed AIN2/3 and/or high-grade squamous intraepithelial lesion (HSIL) on anal cytology was defined as high-grade anal intraepithelial neoplasia (HGAIN) (227).

Progression to HGAIN was defined as:

 HGAIN with prior lower-grade histology.

If no prior histology was available, then by prior:

o anal cytology negative for intraepithelial lesion or malignancy; or

o visually normal HRA.

Regression of HGAIN was defined as:

 HGAIN with subsequent lower-grade histology.

If no subsequent histology was available, then by subsequent:

o anal cytology negative for intraepithelial lesion or malignancy; or

o visually normal HRA.

Analyses were also performed using an alternative endpoint based on histologically- confirmed AIN3.

Progression to AIN3 was defined as:

 Histologically-confirmed AIN3 with prior lower-grade histology.

If no prior histology was available, then by prior:

o anal cytology negative for intraepithelial lesion or malignancy; or

o visually normal HRA.

Regression of AIN3 was defined as:

 Histologically-confirmed AIN3 with subsequent lower-grade histology.

If no subsequent histology was available, then by subsequent:

o anal cytology negative for intraepithelial lesion or malignancy; or

o visually normal HRA.

2.3.5 Statistical Analysis

Logistic regression was used for cross-sectional analyses and Cox regression for longitudinal analyses. For analyses of progression, person-years were measured from baseline until an endpoint was reached (or if no endpoint was reached until the last visit for each patient). For analyses of regression, person-years were measured from the diagnosis of HGAIN or AIN3 until a subsequent lower-grade diagnosis as defined above (or if HGAIN or AIN3 persisted until the last visit for each patient). P-values of

0.05 or less were considered significant. Analyses were performed using Stata 12.1

(StataCorp, College Station, United States of America). We used four measures of

CD4+ T-lymphocyte immunodeficiency: current absolute count, nadir absolute count, time-weighted area under the curve (TWAUC, an estimate of cumulative degree of

CD4+ T-lymphocyte immunodeficiency) (228) and percentage of time (of available results) spent under thresholds of 200, 350 and 500 cells/μL. One patient was HIV- uninfected at enrolment but seroconverted 4 months later and was considered HIV- infected for the purposes of analysis.

2.4.1 Patients

Baseline characteristics of the 574 included patients (99.3% male, 73.0% HIV-infected) are shown in Table 2.1. The HIV-uninfected patients had a median age of 26 years

(interquartile range [IQR] 22-42). Baseline cytology results were available on 155

(100%) HIV-uninfected patients and of these 2.6% had a diagnosis of HSIL. Baseline histology results were available on 77 (49.7%) HIV-uninfected patients, an already selected population with abnormal Papanicolaou smears, and of these 40.3% had a diagnosis of AIN2/3. The HIV-infected patients had a median age of 47 years (IQR 42-

53). Baseline cytology results were available on 417 (99.5%) HIV-infected patients and of these 6.2% had a diagnosis of HSIL. Baseline histology results were available on 183

(43.7%) HIV-infected patients and of these 48.1% had a diagnosis of AIN2/3.

Considering only the HIV-infected patients, the median duration of HIV infection was

13.8 years (IQR 6.4-19.8) with a median absolute CD4+ T-lymphocyte count of 500 cells/μL (IQR 357-662); 83.5% had undetectable HIV viral load.

2.4.2 Frequency of Testing and Follow-Up

The mean number of clinic visits was 2.7 (SD 2.60, range 1-19). At least one anal

Papanicolaou smear was performed on 573 patients (1381 smears, mean 2.4 smears per patient, standard deviation [SD] 2.59). The mean interval between anal Papanicolaou smears was 7.4 months (SD 7.85).

At least one HRA was performed on 323 patients (534 HRAs, mean 1.7 HRAs per patient, SD 1.36). The mean duration between HRAs for those with at least two HRAs

(N=211) was 11.6 months (SD 8.57). Four-hundred fifty-five (85.0%) HRAs were visually abnormal, of which 432 (94.9%) proceeded to biopsy. Reasons for not performing intra-anal biopsies on visually abnormal HRAs included scarring, radiation proctitis, anticoagulation, haemorrhoids and having peri-anal biopsies performed. When intra-anal biopsies were taken, a mean 1.9 biopsies (SD 0.99, range 1–6) were taken per

HRA. Of 1009 histology results, 82% were from HRA-directed biopsies and the remainder were from anal surgery specimens.

Three-hundred twenty-two (56%) patients (211 HIV-infected, 111 HIV-uninfected) were seen on more than one occasion and were included in the longitudinal analysis.

For these patients, the median duration of follow-up was 1.1 years (IQR 0.26 – 2.76) and the mean time between visits was 0.5 years (SD 0.61). There were 16637 CD4+ T- lymphocyte counts on 411 HIV-infected patients, and 15182 HIV viral loads on 405

HIV-infected patients.

Table 2.1 – Cohort description at baseline

HIV-uninfected HIV-infected All N [%] 155 [27.0] 419 [73.0] 574 [100] Age (years, median [IQR]) 26 [22-42] 47 [42-53] 45 [36 – 51] Male (N, [%]) 152 [98.1] 418 [99.8] 570 [99.3] History of anal surgery (N, [%]) 10 [6.5] 84 [20.0] 94 [16.4] Cytology results (N) 155 417 572 Negative for intraepithelial lesion or malignancy 68 [43.9] 78 [18.7] 146 [25.5] Atypical squamous cells of undetermined significance (ASC-US) 31 [20.0] 88 [21.1] 119 [20.8] Low-grade squamous intraepithelial lesion (LSIL) 20 [12.9] 160 [38.4] 180 [31.5] Atypical squamous cells – cannot exclude high-grade (ASC-H) 8 [5.2] 39 [9.4] 47 [8.2] High-grade squamous intraepithelial lesion (HSIL) 4 [2.6] 26 [6.2] 30 [5.2] Unsatisfactory 22 [14.2] 26 [6.2] 48 [8.4] Missing 2 [1.3] 0 [0.0] 2 [0.3] Histology results (N) 77 183 260 Negative 14 [18.2] 17 [9.3] 31 [11.9] Anal intraepithelial neoplasia grade 1 (AIN1) or wart virus effect 32 [41.6] 76 [41.5] 108 [41.5] Anal intraepithelial neoplasia grade 2 (AIN2) 20 [26.0] 43 [23.5] 63 [24.2] Anal intraepithelial neoplasia grade 3 (AIN3) 11 [14.3] 45 [24.6] 56 [21.5] Squamous cell carcinoma (SCC) 0 [0.0] 2 [1.1] 2 [0.8] HIV duration (years, median [IQR])* - 13.8 [6.4 – 19.8] - CD4+ T-lymphocyte count (cells/μL, median [IQR])† - 500 [357 – 662] - Nadir CD4+ T-lymphocyte count (cells/μL, median [IQR])† - 216 [98 – 346] - TWAUC CD4+ T-lymphocyte count (cells/μL, median [IQR])† - 468 [344-586] - Undetectable HIV viral load (N [%])‡ - 338 [83.5] -

*Data available on 415 patients. †Data available on 411 patients. ‡Data available on 405 patients.

2.4.3 Progression

Of 152 patients included in the analysis of progression to HGAIN, 19 (12.5%) developed incident HGAIN over 256.1 person-years (PY) of follow-up, an incidence of

7.4 per 100PY (95% CI 4.73-11.63) (Table 2.2 and Supplementary Figure 2.1). The

19 patients who progressed were followed for 37.1 PY versus 219.0 PY for the 132 who did not progress (mean 2.0 and 1.7 PY follow-up respectively). Having an abnormal baseline diagnosis (i.e. AIN1 on histology) was not significantly associated with increased risk of progression to HGAIN (Ptrend=0.120).

Of 199 patients included in the analysis of progression to AIN3, 25 (12.6%) developed incident AIN3 over 308.4 PY of follow-up, an incidence of 8.1 per 100PY (95% CI

5.48-12. 00) (Table 2.3). The 25 patients who progressed were followed for 45.3 PY versus 263.1 PY for the 174 who did not progress (mean 1.8 and 1.5 PY follow-up respectively). Having an abnormal baseline diagnosis (i.e. AIN1 or AIN2 on histology) was significantly associated with increased risk of progression to AIN3 (HR 3.9,

Ptrend<0.001).

No risk factor was identified for progression to HGAIN. Increasing age, being HIV- infected and having a nadir CD4+ T-lymphocyte count <200 cells/μL (Figure 2.1) were significantly associated with increased risk of progression to AIN3. Restricting analyses to the HIV-infected group, increasing age was not significantly associated with increased risk of progression to AIN3 (Ptrend=0.163). Current CD4+ T-lymphocyte count, TWAUC CD4+ T-lymphocyte count, and percentage of time spent with CD4+ T- lymphocyte counts <200, 350 and 500 cells/μL (data not shown) were not significantly associated with differences in risk of progression to HGAIN or AIN3.

Figure 2.1 – Kaplan-Meier curve for progression to grade 3 anal intraepithelial neoplasia

Table 2.2 – Progression to and spontaneous regression of HGAIN*

Progression to HGAIN* Spontaneous Regression of HGAIN*

PY N Incidence HR 95% CI P value PY N Incidence HR 95% CI P value (per 100PY) (per 100PY) Overall 256.1 19 7.4 --- 4.73-11.63 --- 102.2 24 23.5 --- 15.73-35.02 --- Age (years) 0.471† 0.048† <35 160.1 12 7.5 1 --- 14.8 6 40.4 1 --- 35 – 45 20.8 3 14.4 1.63 0.41-6.44 25.5 8 31.4 0.81 0.27-2.41 45 – 55 52.5 2 3.8 0.42 0.09-1.98 41.1 9 21.9 0.59 0.20-1.71 >55 22.6 2 8.8 0.92 0.18-4.72 20.8 1 4.8 0.14 0.02-1.18 HIV status 0.506 0.254 Uninfected 164.8 13 7.9 1 --- 24.3 9 37.1 1 --- Infected 91.2 6 6.6 0.70 0.24-2.00 78.0 15 19.2 0.61 0.26-1.43 HIV duration (years) 1.07 0.95-1.22 0.273† 0.94 0.87-1.01 0.114† CD4+ T-lymphocyte count (cells/μL) Current 1.00 1.00-1.01 0.578 1.00 1.00-1.00 0.220 TWAUC 1.00 1.00-1.01 0.767 1.00 1.00-1.00 0.727 Nadir CD4+ T-lymphocyte count (cells/μL) 0.648† 0.212† HIV-uninfected 164.8 13 7.9 1 --- 24.3 9 37.1 1 --- CD4 nadir ≥200 44.5 3 6.7 0.68 0.18-2.57 41.1 9 21.9 0.69 0.27-1.79 CD4 nadir <200 43.2 3 6.9 0.79 0.21-2.92 36.8 6 16.3 0.51 0.18-1.48 HIV viral load (copies/mL) 0.752 0.934 Undetectable 75.4 5 6.6 1 --- 61.7 11 17.8 1 --- Detectable 12.1 1 8.3 1.43 0.16-13.19 14.2 3 21.1 1.06 0.29-3.83 * † HGAIN = composite of histologically-confirmed AIN2/3 and/or high-grade squamous intraepithelial lesion (HSIL) on anal cytology. Ptrend.

Table 2.3 – Progression to and spontaneous regression of histologically-confirmed AIN3

Progression to AIN3 Spontaneous Regression of AIN3

PY N Incidence HR 95% CI P value PY N Incidence HR 95% CI P value (per 100PY) (per 100PY) Overall 308.4 25 8.1 --- 5.48-12.00 --- 37.8 26 68.9 --- 46.88-101.14 --- Age (years) 0.004* 0.702* <35 167.9 6 3.6 1 --- 5.3 6 112.2 1 --- 35 – 45 33.1 3 9.1 2.48 0.61-10.11 9.2 6 65.1 0.49 0.14-1.66 45 – 55 72.5 10 13.8 3.62 1.28-10.23 17.0 13 76.7 0.91 0.31-2.70 >55 34.9 6 17.2 4.69 1.47-14.93 6.2 1 160.2 0.28 0.03-2.61 HIV status 0.019 0.149 Uninfected 178.6 8 4.5 1 --- 7.1 9 127.5 1 --- Infected 129.8 17 13.1 2.81 1.18-6.68 30.7 17 55.4 0.52 0.22-1.26 HIV duration (years) 1.04 0.97-1.12 0.281* 1.03 0.95-1.11 0.512* CD4+ T-lymphocyte count (cells/μL) Current 1.00 1.00-1.00 0.466 1.00 1.00-1.00 0.984 TWAUC 1.00 1.00-1.00 0.532 1.00 1.00-1.00 0.563 Nadir CD4+T-lymphocyte count (cells/μL) 0.003* 0.125* HIV-uninfected 178.6 8 4.5 1 --- 7.1 9 127.5 1 --- CD4 nadir ≥200 61.0 5 8.2 2.43 0.71-8.25 18.0 11 61.3 0.61 0.23-1.59 CD4 nadir <200 65.3 12 18.4 4.66 1.65-13.11 12.7 6 47.4 0.42 0.14-1.26 HIV viral load (copies/mL) 0.312 0.705 Undetectable 105.0 16 15.2 1 --- 25.0 14 56.0 1 --- Detectable 19.7 1 5.1 0.35 0.05-2.67 5.6 3 53.8 0.77 0.20-2.93 *

Ptrend.

2.4.4 Regression

Of 101 patients included in the analysis of regression of HGAIN, 31 regressed. Of these, six regressed after surgical excision and one after topical imiquimod treatment at an overseas clinic. Therefore, 24 (23.8%) patients experienced spontaneous regression from HGAIN over 102.2 PY of follow-up, an incidence of 23.5 per 100PY (95%CI

15.73-35.02). Seventeen (71%) patients regressed to AIN1 and seven (29%) regressed to negative biopsies. The 24 patients who spontaneously regressed were followed for

33.6 PY versus 68.6 PY for the 70 who did not spontaneously regress (mean 1.4 and 1.0

PY follow-up respectively).

Of 55 patients included in the analysis of regression of AIN3, 35 regressed. Of these, eight regressed after surgical excision and one after imiquimod treatment. Therefore, 26

(47%) patients experienced spontaneous regression from AIN3 over 37.8 PY of follow- up, an incidence of 68.9 per 100PY (95%CI 46.88-101.14). Eleven (42%) regressed to

AIN2, 11 (42%) regressed to AIN1 and four (15%) regressed to negative biopsies. The

26 patients who spontaneously regressed were followed for 24.0 PY versus 13.8 PY for the 21 who did not spontaneously regress (mean 0.9 and 0.7 PY follow-up respectively).

Increasing age was significantly associated with decreased risk of spontaneous regression of HGAIN (Ptrend=0.048). Restricting analyses to the HIV-infected group, increasing age was not significantly associated with decreased risk of spontaneous regression of HGAIN (Ptrend=0.151). HIV-infected patients had similar spontaneous regression rates of HGAIN (HR 0.6, P=0.254) and of AIN3 (HR 0.5, P=0.149) compared to HIV-uninfected patients. Current and nadir CD4+ T-lymphocyte count,

TWAUC CD4+ T-lymphocyte count, and percentage of time spent with CD4+ T- lymphocyte counts <200, 350 and 500 cells/μL (data not shown) were not significantly associated with differences in risk of spontaneous regression of HGAIN or AIN3.

2.4.5 Cancer

There were six patients with histologically-confirmed anal SCC in this cohort, of which two were baseline diagnoses. Therefore, four incident anal cancers occurred of which two had documented HGAIN before cancer diagnosis (Table 2.4). With two cases over

161.5 PY of follow-up, the incidence of anal cancer in patients with HGAIN was 1.2 per 100PY (95%CI 0.31-4.95).

Table 2.4 – Incident anal SCC patients

Age HIV Duration Nadir CD4+ Baseline diagnosis Documented Time of first Time of first Time of SCC (years) status of HIV T-lymphocyte HGAIN* HGAIN* AIN3 diagnosis /Gender infection count before SCC diagnosis diagnosis (years from (years) (cells/μL) diagnosis (years from (years from baseline) baseline) baseline) 1 52/M + 21.0 16 LSIL (cytology) No ------1.1 2 56/M + 21.7 154 AIN2 (histology) Yes 0.0 3.1 4.9 3 50/M + 5.1 171 ASC-US (cytology) Yes 0.8 1.3† 4.2 4‡ 48/M + 12.4 60 LSIL (cytology) No ------0.9§ *HGAIN = composite of histologically-confirmed AIN2/3 and/or high-grade squamous intraepithelial lesion (HSIL) on anal cytology. †Referred for surgical excision of AIN3 at 3.6 years. ‡Anal primary not confirmed. §SCC in surgically-excised enlarged inguinal lymph node. This patient had a pre-baseline history of intra-anal high-grade disease treated by surgery and peri-anal high-grade disease treated with topical imiquimod and 5-fluorouracil. Died at 1.6 years.

This is the first study to quantify spontaneous regression of high-grade ASILs in a cohort of HIV-infected and HIV-uninfected MSM untreated for their anal HPV- associated pathology. We found that 23.5% per year (95%CI 15.73-35.02) of men with

HGAIN (AIN2/3 on histology and/or HSIL on cytology) spontaneously regressed, mostly to AIN1. In contrast, the rate of progression from HGAIN to anal cancer was

1.2% per year (95%CI 0.31-4.95).

According to a recent systematic review, the incidence of high-grade ASILs is 8.5-

15.4% per year in HIV-infected MSM and 3.3-6.0% in HIV-uninfected MSM (46).

These rates are comparable to the rates found in our study: HIV-infected men had an incidence of progression to HGAIN of 6.6% per year and to AIN3 of 13.1% per year;

HIV-uninfected men had an incidence of progression to HGAIN of 7.9% per year and to

AIN3 of 4.5% per year.

Spontaneous regression of high-grade ASILs has not been previously described, in part because many centres that diagnose high-grade ASILs immediately treat them. We found 37.1% per year of HGAIN spontaneously regressed in HIV-uninfected men and this is broadly comparable to the estimates of CIN2/3 regression (overall 33-40%, annual rate not reported) in HIV-uninfected women (118). In HIV-infected men, we found 19.2% per year of HGAIN and 55.4% per year of AIN3 spontaneously regressed.

Spontaneous regression rates of AIN3 are higher than of HGAIN in our analyses because a large proportion (42%) of AIN3 regressed only to AIN2.

HIV-infected men were more likely to progress to AIN3 (HR 2.81) and tended to be less likely to spontaneously regress from AIN3 (HR 0.52) than HIV-uninfected men.

This trend was not seen in the analysis of progression to and spontaneous regression of

HGAIN, probably because of a lack of power but also because a higher proportion of

HIV-infected men already met the definition of HGAIN at baseline compared to HIV- uninfected men. If this finding is replicated in future larger prospective studies, it provides a possible explanation for the increased risk of anal cancer in HIV-infected adults: increased risk of progression to AIN3 combined with a decreased risk of spontaneous regression from AIN3 results in AIN3 persistence which, over time, increases the risk of progression to cancer.

The rate of progression from HGAIN to anal cancer was 1.2% per year (95%CI 0.31-

4.95), which is similar to high-grade cervical squamous intraepithelial lesions (117) but much higher than the theoretical rate of 1 in 377 per year in HIV-infected MSM calculated in the meta-analysis by Machalek et al (46). Both incident anal SCCs occurred in HIV-infected patients with nadir CD4+ T-lymphocyte counts <200 cells/μL.

Our data supports current guidelines that recommend any abnormality on anal cytology warrants further investigation with HRA (143, 144), as two of four incident anal cancers had only low-grade cytology a short time before the diagnosis of cancer.

The only measure of CD4+ T-lymphocyte immunodeficiency that was associated with increased risk of progression to AIN3 and tended towards decreased risk of spontaneous regression of AIN3 was being HIV-infected with a nadir CD4+ T-lymphocyte count

<200 cells/μL. Other measures of CD4+ T-lymphocyte immunodeficiency were not associated with changes in risk of progression to or regression of HGAIN or AIN3. The association of low nadir CD4+ T-lymphocyte count and increased risk of incident high- grade ASILs has previously been described (94), and it is also a known risk factor for anal cancer in HIV-infected adults (93, 229). It may be that the repertoire of lymphocytes required to successfully patrol the anal mucosa and effect clearance of

HPV infection and intraepithelial neoplasia is irreversibly damaged only if there is

severe CD4+ T-lymphocyte immunodeficiency from HIV infection which never recovers despite reconstitution of systemic CD4+ T-lymphocyte counts with cART. The central role of low nadir CD4+ T-lymphocyte counts in the natural history of ASILs in

HIV-infected adults may provide an explanation as to why rising anal cancer incidence rates have plateaued (1), as the CD4+ T-lymphocyte count threshold for starting cART has increased (from 200 cells/μL in the early cART era to currently ≥350 cells/μL

(230)).

Using undetectable HIV viral load as a surrogate marker for effective cART, we did not find any association between being on effective cART and changes in risk of progression to or regression of HGAIN or AIN3. This is consistent with Piketty et al’s prospective study that found cART initiation did not appear to have any effect on the natural history of ASILs in HIV-infected MSM (231). However, 83.5% of our cohort had undetectable HIV viral load at baseline, with only 16 patients (4.0%) who were initially detectable becoming undetectable during follow-up. Therefore, our cohort is not powered to assess the effect of cART initiation on ASILs.

Definitions of progression and regression of ASILs are problematic due to the multifocal nature of the disease and the limitations of anal cytology and HRA.

Compared to cervical cytology, anal cytology is significantly less sensitive and specific for detecting high-grade disease at every cut-off (area under receiver operating characteristic curve 0.70 for anal cytology compared to 0.86 for cervical cytology)

(129). HRA itself is an imperfect reference standard (232), because the highly involuted surface of the anal epithelium means that high-grade lesions can be missed (233) on visual inspection or at biopsy. It is generally accepted that a finding of HSIL on anal cytology has high specificity (93%) and good positive predictive value (89%) for high- grade ASILs particularly in high prevalence populations such as HIV-positive MSM

(127) and may signify high-grade ASILs that were missed at HRA. Given there is no consensus definition of progression and regression of ASILs in the literature, we adopted a composite definition of high-grade ASILs that included histologically- confirmed AIN2/3 and/or HSIL on anal cytology (which we labelled HGAIN). This is in keeping with the current recommended two-tiered nomenclature (3) and similar to the classification used by Wentzensen et al (227).

Our study has limitations. First, it was retrospective with selection bias particularly towards those patients that were more likely to return for follow-up (e.g. symptomatic patients who may have more severe disease). Also, our HIV-uninfected patient group was significantly younger and not directly comparable to the HIV-infected group. While we found that increasing age was strongly associated with increased risk of progression to AIN3 and decreased risk of spontaneous regression of HGAIN, younger age is confounded by being HIV-uninfected in our cohort. Approximately 25% of our HIV- uninfected group would have received the active prophylactic quadrivalent HPV vaccine, therefore our estimates of HGAIN and AIN3 incidence in the HIV-uninfected group may be underestimates. Our study was underpowered, particularly for the analyses of regression and we could not assess independent predictors of our endpoints.

We do not have data on HPV DNA, size of high-grade lesions, smoking history, details of cART regimens and relevant behaviours. As is the case in many longitudinal studies of ASILs, many visits had an anal Papanicolaou smear that was not accompanied by

HRA and biopsies. There is likely to be variability in the reporting of anal cytology and histology results due to more than one reporting pathologist (234, 235). The technical skill of the anoscopist in diagnosing high-grade ASILs may have improved over the study period, although this effect would not be expected to affect our analyses of progression and regression of disease within individual patients as all HRAs were

performed by the one anoscopist. Finally, this type of analysis does not take into account the likely dynamic nature of ASILs. Our overall follow-up duration was short and we do not know if the spontaneous regressions we describe are complete or sustained. The only way to address these limitations is with a detailed, prospective natural history study of ASILs with several years of follow-up. An example of this is the Study of the Prevention of Anal Cancer which is currently being conducted at St

Vincent’s Hospital (236).

Strengths of our study include our robust endpoint definitions, despite the lack of agreement in the literature and the inherent limitations of anal cytology and HRA. The similarity of our progression to high-grade ASIL rates with those in the literature increases our confidence that our definitions are appropriate, and that our estimation of spontaneous regression rates of high-grade ASIL are meaningful. We repeated our analyses using histologically-confirmed AIN3 as the most robust indication of precancer, and have presented both sets of results for comparison. We had very comprehensive CD4+ T-lymphocyte count and HIV viral load data on our HIV-infected patients.

The implications for clinical care from our data will vary according to local practice.

For those who currently adopt a “watchful waiting” approach in the management of high-grade ASILs, our data provide reassurance that high-grade ASILs diagnosed at any one time-point are much more likely to spontaneously regress than progress to cancer.

For those who currently treat high-grade ASILs, it may be that the risk of treatment in some patients would outweigh the benefits.

In summary, this is the first study to quantify spontaneous regression of high-grade

ASILs. Spontaneous regression occurs far more commonly than progression to cancer, suggesting that not all patients with high-grade ASILs warrant treatment. Future studies 85

in this field should report regression as well as progression rates. Prospective studies to delineate risk factors and biomarkers that predict those at highest risk of progression to cancer are needed so that intervention can be targeted, and avoided in those for whom it is unlikely to be of value.

Supplementary Figure 2.1 – Flowchart of patients in endpoint analyses

This page has been intentionally left blank

This chapter was published in The Journal of Infectious Diseases 2015;211(3):405-15.

A PDF of the published article is included in Appendix C.

This page has been intentionally left blank

Background: Most anal cancers are attributable to persistent HPV16 infection. The anal cancer precursor, high-grade squamous intraepithelial lesion (HSIL), frequently regresses spontaneously. We hypothesized that T-cell responses are associated with

HSIL regression.

Methods: In men who have sex with men (MSM) undergoing anal cytology and high- resolution anoscopy (clinicaltrials.gov NCT02007421), we measured responses to

HPV16 oncogenic proteins E6 and E7 using the CD25/CD134 assay for CD4+ antigen- specific T-cells, and intracellular cytokine staining for CD4+ and CD8+ antigen-specific

T-cells.

Results: Of 134 participants (mean age 51 years [SD 9.3], 31 [23.1%] HIV-infected),

51 (38.1%) had HSIL. E6 and E7-specific CD4+ T-cell responses were detected in 80

(59.7%) and 40 (29.9%) participants, respectively. E6 and E7-specific CD8+ T-cell responses were each detected in 25 (18.7%) participants. HSIL was significantly associated with E7-specific CD8+ T-cell responses (OR 4.09 [95%CI 1.55-10.77],

P=0.004), but not with any CD4+ T-cell response (P≥0.092). Twenty-six participants had HSIL a mean one year prior to measuring T-cell responses, of which six (23%) regressed. Five (83%) regressors had E6-specific CD4+ T-cell responses versus seven of 20 (35%) non-regressors (Pexact=0.065).

Conclusions: Systemic HPV16-E6 and E7-specific T-cell responses were common in

MSM. E6-specific CD4+ T-cell responses may be associated with recent HSIL regression.

Anal high-grade squamous intraepithelial lesions (HSILs) are precursors to squamous cell carcinoma (SCC) (3). Approximately 85% of anal cancer is caused by high-risk types of human papillomavirus (HPV), of which type 16 (HPV16) is the most common cause (90% of HPV-related anal cancers) (26, 237, 238). Men who have sex with men

(MSM) and HIV-infected adults are at particularly high risk (1).

Although anal HSIL is detected in one-third of MSM (46), the low incidence of anal cancer suggests that few progress to cancer each year (239, 240). Therefore, most anal

HSIL in MSM spontaneously regresses or persists without progression. A retrospective study found 23.5% of anal HSIL in men spontaneously regressed per year, mostly to low-grade squamous intraepithelial lesions (LSILs), although long term follow-up is required to determine if these regressions are sustained (240). Cellular immune responses are a central mechanism underlying regression of HPV-related lesions (67,

81, 241). In keeping with this hypothesis, HIV-related CD4+ T-cell immunodeficiency is a risk factor for anal HPV, SILs and cancer (242).

HPV16-specific T-cell responses have been measured in women with cervical SIL and healthy volunteers by delayed-type hypersensitivity skin tests, interferon- (IFN)

ELISPOT and 3H-thymidine proliferation assays (74, 79, 241). These traditional T-cell assays have variable sensitivity, especially for antigens that elicit small systemic immune responses, and do not predict HSIL regression in the cervix (82, 243).

Screening strategies for prevention of anal SCC lack evidence of effectiveness and safety (242). Furthermore, there is no proven treatment for anal HSIL (135) and complete excision (the standard of care for cervical HSIL) is not possible without unacceptable morbidity (134, 244). Therefore, an immune response-based assay that

predicts which patients with anal HSIL are more likely to regress spontaneously would enable screening and treatment resources to be targeted to those at higher risk of persistence or progression.

We measured T-cell responses to the HPV16 oncogenic proteins, E6 and E7, in a prospective natural history study of anal HPV and SILs in MSM, the Study of the

Prevention of Anal Cancer (SPANC). Our aim was to describe in cross-sectional describe HPV16-specific T-cell responses in a subset of SPANC participants and correlate these with anal HSILs. We hypothesised HPV16-specific T-cell responses would be associated with recent anal HSIL regression.

3.3.1 Study Population

SPANC is a community-recruited study of HIV-uninfected and infected MSM aged ≥35 years in Sydney, Australia (clinicaltrials.gov identifier NCT02007421) (245). In brief, participants undergo liquid-based Papanicolaou tests for anal cytology and high- resolution anoscopy (HRA) with biopsies of visually abnormal areas at each of five study visits over three years. Diagnosed anal SILs were not routinely treated. HPV

DNA detection and typing were performed on liquid-based anal cytology specimens using the Roche Linear Array® assay (Alameda, USA).

We defined anal high-grade disease as a composite of cytological HSIL (Bethesda

System 2001) (6) and/or histological grade 2 or 3 anal intraepithelial neoplasia

(AIN2/3) (3, 227). Atypical Squamous Cells cannot exclude HSIL (ASC-H) was not included. A participant was considered to have experienced spontaneous regression if

high-grade disease was diagnosed at study entry, but not at the Immunology Substudy visit in the absence of intervention.

The present Immunology Substudy protocol was approved by St Vincent’s Hospital

Human Research Ethics Committee (File Number 09/203 amendment 15th December,

2011). After written, informed consent was obtained, 48mL of peripheral blood was collected once, and processed within six hours. Absolute CD4+ and CD8+ T-cell counts were determined using TrucountTM tubes (Becton Dickinson [BD], Franklin Lakes,

USA) (246, 247).

3.3.2 CD25/CD134 Assay

This simple assay detects antigen-specific CD4+ T-cell responses (by CD25 and CD134 co-expression) to a range of antigens with high sensitivity in various clinical settings

(212, 213, 215, 248). Heparinised whole blood was mixed with Iscove’s Modified

Dulbecco’s Medium with GlutaMAXTM (Gibco, Carlsbad, USA) 1:1 (total volume

500μL) in 24-well plates. The following stimulants were added per well: medium only

(negative control), Staphylococcus enterotoxin B 1μg/mL (SEB, Sigma-Aldrich, St

Louis, USA; positive mitogen control), cytomegalovirus (CMV) lysate 1μL (Meridian

Life Science, Cincinnati, USA; positive antigen control), duplicate wells with a pool of

30 peptides (15mers overlapping by 10) spanning E6 10μg/mL, and duplicate wells with a pool of 18 15mer peptides spanning E7 10μg/mL (JPT Peptide Technologies, Berlin,

Germany). Stock solutions of each pool, covering the full published sequences (NCBI

RefSeq Accession NP_041325.1 for HPV16-E6; NP_041326.1 for HPV16-E7), were initially prepared in dimethyl sulfoxide (DMSO; MP Biomedicals, Solon, USA), and then diluted in culture medium for the assays at equivalent final concentrations of

0.04% by volume, to control for any effects on T-cell responses. Cultures were incubated in a humidified atmosphere at 37°C, 5% CO2 for 44 ± 2 hours.

Cultures were then centrifuged (500g, 7 minutes), and the supernatant aspirated and discarded. Cells were stained with CD3-PerCP-Cy5.5, CD4-PE-Cy7, CD25-APC and

CD134-PE (BD) for 15 minutes at room temperature. Red cells were lysed with

Optilyse C (Beckman-Coulter, Marseilles, France) and washed twice with Dulbecco’s phosphate-buffered saline (PBS, Gibco). Cells were resuspended in 0.5% paraformaldehyde in PBS until acquisition within 24 hours.

3.3.3 Intracellular Cytokine Staining Assay

This assay measures antigen-specific IFN and interleukin-2 (IL2) production from

CD4+ and CD8+ T-cells (210, 249). Heparinised whole blood (500μL) was mixed with the following in 15mL Falcon® conical tubes (Corning, Tewksbury, USA): blood only,

SEB 1μg/mL (Sigma-Aldrich), CMV lysate 1μL (Meridian) and CMV, Epstein-Barr virus and influenza (CEF) peptides 2μg/mL (Auspep, Tullamarine, Australia) (250), E6 peptides 10μg/mL and E7 peptides 10μg/mL (JPT) as detailed above in DMSO 0.04% by volume. CD28 and CD49d 1μg/mL (BD) were added to each tube and cultures were incubated in a humidified atmosphere at 37°C, 5% CO2 for 6 hours. Brefeldin A

10μg/mL (Sigma-Aldrich) was added after 2 hours.

After incubation, EDTA 2mM was added to each culture. Cells were treated with FACS

Lysing Solution (BD) for 10 minutes at room temperature, then washed with 0.5% bovine serum albumin and 0.1% sodium azide (both Sigma-Aldrich) in PBS (PBA) by centrifugation (500g, 7 minutes). FACS Permeabilizing Solution 2 (BD) was added to resuspended cells for 10 minutes at room temperature, washed by centrifugation (644g,

7 minutes) with PBA, and stained with CD3-PerCP-Cy5.5, CD4-PE-Cy7, CD8-APC-

Cy7, CD69-APC, IFN-FITC and IL2-PE (BD) for 30 minutes in the dark at 4°C. Cells were washed (644g, 7 minutes) with PBA twice and resuspended in Stabilizing Fixative

(BD) until acquisition within 24 hours.

3.3.4 Flow Cytometry Acquisition

Samples were acquired on a four-laser LSRII flow cytometer using FACSDiva version

6.1.2 (BD). Photo multiplier tube voltages were set to optimise separation between

CD4-negative lymphocytes, CD4-dim monocytes and CD4+ T-cells for each channel

(251). Compensation was determined using single fluorochrome-stained cells before sample acquisition.

For the CD25/CD134 assay, 500,000 total events were collected per sample. For E6 and

E7, .fcs files from duplicate wells were concatenated and analysed in FlowJo (version

7.6.5, TreeStar, Ashland, USA) to total 1,000,000 events. For the intracellular cytokine staining (ICS) assay, 1,000,000 total events were collected per sample. The gating strategy is shown in Figure 3.1.

3.3.5 Assay Cut-offs

For the CD25/CD134 assay, the two mandatory criteria defining positive reactivity were: a) percentage of CD4+CD25+CD134+ T-cells was at least double the negative control (stimulation index [SI] ≥2); and b) absolute count of CD4+CD25+CD134+ T- cells was >20 cells after subtracting double the number of cells in the corresponding region of the negative control culture (to correct for total 1,000,000 events).

For the ICS assay, the two mandatory criteria defining positive reactivity were: a) percentage of CD4+ or CD8+ T-cells that were CD69+ and cytokine-positive was at least double the negative control (SI ≥ 2); and b) absolute count of CD69+ and cytokine-positive cells was >15 cells after subtracting the number of cells in the corresponding region of the negative control culture. For statistical analysis, the ICS assay was considered positive to an antigen if IFN and/or IL2 reactivity met the above criteria for each T-cell subset.

Figure 3.1 – Representative data sets demonstrating T-cell assay gating strategies.

Panel A shows the gating strategy for the CD25/134 assay. Panel B shows the gating strategy for the intracellular cytokine staining (ICS) assay. Antigen-specific CD8+ T- cell responses are shown; the same strategy was applied to CD4+ T-cells. For both assays, the primary readout of interest is the top right quadrant (indicated by thick lines).

3.3.6 Statistics

There was no pre-planned sample size as this was a descriptive study. Cross-sectional associations were tested for using Pearson χ2 or Fisher’s exact test for dichotomous variables, and Student’s t-test to compare means for continuous variables. Logistic regression was used to determine factors associated with HPV16-specific T-cell responses in univariable analyses. Variables with P≤0.1 were included in multivariable analyses using backward, stepwise, logistic regression. Due to biological plausibility, anal HPV16 status, CD4+/CD8+ T-cell ratio, HIV status and age group were forced into the multivariable model in sensitivity analyses. Assay correlations were assessed with

Spearman’s correlation coefficients. Two-sided testing with a significance level of

P≤0.05 was used for all analyses. Analyses were performed using Stata 12.1 (StataCorp,

College Station, USA) and Prism 6.04 (GraphPad Software, La Jolla, USA).

3.4.1 Participant characteristics

Immunology Substudy participant characteristics (N=134) are shown in Table 3.1.

Mean age was 50.8 years (SD 9.31) and 31 (23.1%) were HIV-infected. Fifty-six

(41.8%) were assessed at their baseline visit in the main study, 34 (25.4%) at 6 months,

29 (21.6%) at 12 months, and 15 (11.2%) at 24 months (mean interval 13.4 months).

HIV-uninfected men had more visually normal HRAs than HIV-infected men (P=0.05).

Overall, cytology diagnoses were similar between the two groups. Histological HSIL was found in 54% of HIV-infected men vs. 42% of HIV-uninfected men (P=0.30), and

HPV16 in 39% vs. 26%, respectively (P=0.18). There was fair agreement between histological and cytological HSIL (kappa=0.38, 95% CI 0.22-0.54, n=126, P<0.0001).

3.4.2 HPV16-E6 and E7-specific T-cell responses

Across both assays, CD4+ T-cell responses were detected in 80 (59.7%) men to E6 and in 40 (29.9%) men to E7. Using the CD25/CD134 assay, 68 (50.8%) men had CD4+ T- cell responses to E6 and 32 (23.9%) men to E7 (Figure 3.2). Using the ICS assay, 24

(17.9%) men had CD4+ T-cell responses to E6 and 16 (11.9%) men to E7; 25 (18.7%) men had CD8+ T-cell responses to each of E6 and E7. Comparing assays, 12 (17.7%) men had CD4+ T-cell responses to E6 on both the CD25/CD134 and ICS assays; 8

(25.0%) men had these responses to E7 on both assays. Further data on assay correlations and performance are presented in Supplementary Figure 3.1. Combining results from both assays, thirty-five (26.1%) men had CD4+ T-cell responses to both E6 and E7; 15 (11.2%) men had CD8+ T-cell responses to both.

Median positive CD4+ and CD8+ T-cell responses to E6 and E7 were a mean 9.8-times

(SD 3.0) smaller than positive responses to CMV. Most positive responses were dominated by cells producing a single cytokine, generally IFN, with the exception of

E7-specific CD4+ T-cell responses, which were predominantly IL2+IFN- (Figure 3.3).

100

Table 3.1 – Participant characteristics

HIV-uninfected HIV-infected All N (%) 103 (76.9) 31 (23.1) 134 (100) Age [years, mean (SD)] 51.2 (9.58) 49.5 (8.33) 50.8 (9.31) Months between baseline and immunology substudy bleeda [mean (SD)] 13.6 (6.59) 12.9 (7.39) 13.4 (6.73) Anal cytology [n (%)] 103 (100) 31 (100) 134 (100) Unsatisfactory 6 (5.8) 2 (6.5) 8 (6.0) Negative for intraepithelial lesion or malignancy 41 (39.8) 12 (38.7) 53 (39.6) Atypical squamous cells of undetermined significance (ASC-US) 14 (13.6) 4 (12.9) 18 (13.4) Low-grade squamous intraepithelial lesion (LSIL) 8 (7.8) 4 (12.9) 12 (9.0) Atypical squamous cells cannot exclude HSIL (ASC-H) 18 (17.5) 3 (9.68) 21 (15.7) High-grade squamous intraepithelial lesion (HSIL) 16 (15.5) 6 (16.1) 22 (16.4) High resolution anoscopy (HRA) findings [n (%)] 103 (100) 31 (100) 134 (100) Visually normal (no biopsy taken) 27 (26.2) 3 (9.7) 30 (22.4) Visually abnormal (at least one biopsy taken) 76 (73.8) 28 (90.3) 104 (77.6) Anal histologyb [n (%)] 76 (100) 28 (100) 104 (100) Negative for squamous intraepithelial lesion 28 (36.8) 7 (25.0) 35 (33.7) Low-grade squamous intraepithelial lesion (LSIL) 16 (21.1) 6 (21.4) 22 (21.1) High-grade squamous intraepithelial lesion (HSIL) 32 (42.1) 15 (53.6) 47 (45.2)

HIV-uninfected HIV-infected All Anal human papillomavirus (HPV) detected [n (%)] 99 (100)c 31 (100) 130 (100)c HPV type 16 26 (26.3) 12 (38.7) 38 (29.2) Any high-risk HPV type 55 (55.6) 22 (71.0) 77 (59.2) Number of HPV types [median, IQR] 2 [1-4] 3 [1-6] - Absolute T-cell count [cells/μL, mean (SD)] CD4+ 822 (253.8)d 584 (249.8) 766 (271.5) CD8+ 465 (198.3) 859 (496.4) 556 (337.1) CD4+/CD8+ T-cell ratio [mean (SD)] 2.0 (0.87) 0.8 (0.42) - Currently on antiretroviral therapye [n (%)] - 29 (94) - Undetectable HIV viral loade [n (%)] - 28 (90) - Ever had AIDS defining illnesse [n (%)] - 11 (35) - aNot including those immunology substudy participants bled at main study baseline. bMost serious diagnosis found if more than one biopsy. cFour participants were unassessable on the HPV typing assay. dThree participants’ absolute CD4+ T-cell count results were excluded due to laboratory quality control measures.

eSelf-reported on audio computer-assisted self-interview (ACASI).

101

6 7 E "' C:>~ ~ -E ,u 134 SEB \ (100) ~ {< CDS ~C:> 131 CMV (97.8) ~ <;>" ~() :X . ~ c,9 . C:>" 32 E7 ,u (23.9) -17 '\ CD25+CD134+ CD4 ~ {< 0 (JC:> " 20 60 40 . ~ ~ cl) cl) 0 c a. a. .... > (/) ' - = E6 68 s:: 0 - (/) cu; a.~ !::! t:: ..... ! ~ • · (50.8) ~ Assay " ... 132 126 SEB SEB (94.0) (98.5) ~ 80 110 (82.1) (59.7) + CMV+CEF CMV+CEF 16 CD69+1L2+ (4.5) (11.9) + 5 6 . . 9 E6 E7 E6 E7 (3.7) (6.7) + -+- .. 131 133 SEB SEB (99.3) (97.8) 117 121 (90.3) (87.3) CMV+CEF CMV+CEF * 0 20 (0) CD69+1FNy+ (14.9) ~ I E6 E7 E6 E7 17 24 (12.7) (17.9) n n 1 1 (%) (%) 10 10 0.1 0.1 100 100 0.01 0.01 I I Q) (/) (.) (.) 0 0 0 0) 0) + + ~ 0 0 co 0 -a; 0 ...... !e. .2 ~ ~ Q Q (.) (.) ...... :::. - - 1- 1-

102

Figure 3.2 – Positive responses detected by T-cell assays.

Top panel displays CD4+ T-cell responses, bottom panel CD8+ T-cell responses. Size of positive responses is expressed as percentage of CD4+ or

CD8+ T-cells. Number of participants with a positive response to each antigen is shown along the x-axes, N=134. E6 and E7 refer to HPV16 E6 and

E7 overlapping peptide pools; CMV = cytomegalovirus lysate; CEF = CMV, Epstein-Barr virus and influenza peptides; SEB = staphylococcal

enterotoxin B. Horizontal bars indicate median, whiskers indicate interquartile range. Boxed inset shows percentage of participants with positive

responses to E6 and E7 by assay. ICS refers to intracellular cytokine staining assay for interferon gamma (IFN) and interleukin-2 (IL2).

CD25/CD134 is an assay for antigen-specific CD4+ T-cells (212). Asterisk (*) indicates no IFN E7-specific CD4+ T-cell responses were detected on

ICS.

103

Figure 3.3 – Qualitative analysis of positive responses detected by ICS.

Panel A describes CD4+ T-cell responses; panel B CD8+ T-cell responses.

Participants were included if they had an overall positive response to E6 (for CD4

responses n=24, CD8 n=25) or E7 (CD4 n=16, CD8 n=25). Some responses were

excluded from this analysis if background reactivity on the alternative cytokine was

larger than the positive response measured, final n included in each group are as indicated. Horizontal bars indicate median; whiskers indicate interquartile range. Each

pie chart represents the average distribution of the cytokine response for each group,

with each included participant weighted equally. For details of individual responses

ranked by response size, see Supplementary Figure 3.2. The area of each pie chart is

proportional to the mean response size measured as the absolute count of

CD69+cytokine+ cells.

104

105

106

Table 3.2 – Clinical associations with positive HPV16-E6-specific T-cell responses (univariable logistic regression) T-cell subset CD4+ CD8+ Assay CD25/CD134 ICS ICS Prevalence OR P Prevalence OR P Prevalence OR P n (%) (95% CI) n (%) (95% CI) n (%) (95% CI) Age (years) 0.064a 0.125a 0.212a 35-44 22 (68.8) 1 9 (28.1) 1 7 (21.9) 1 45-55 26 (44.8) 0.37 9 (15.5) 0.47 13 (22.4) 1.03 (0.15-0.92) (0.16-1.34) (0.36-2.92) ≥55 20 (45.5) 0.38 6 (13.6) 0.40 5 (11.4) 0.46 (0.15-0.98) (0.13-1.28) (0.13-1.60) HIV status 0.764 0.811 0.031 Uninfected 53 (51.5) 1 18 (17.5) 1 15 (14.6) 1 Infected 15 (48.4) 0.88 6 (19.4) 1.13 10 (32.3) 2.79 (0.40-1.98) (0.41-3.16) (1.10-7.09) CD4+/CD8+ T-cell 0.236 0.464 0.657 ratio - 0.80 - 1.19 - 0.90 (0.55-1.16) (0.75-1.87) (0.55-1.46) Anal HPV status HPV16 0.822 0.321 0.523 Not detected 48 (52.2) 1 19 (20.7) 1 19 (20.7) 1 Detected 19 (50.0) 0.92 5 (13.2) 0.58 6 (15.79) 0.72 (0.43-1.95) (0.20-1.69) (0.26-1.97) Any high-risk HPV type 0.059 0.921 0.590 Not detected 22 (41.5) 1 10 (18.9) 1 9 (17.0) 1 Detected 45 (58.4) 1.98 14 (18.2) 0.96 16 (20.8) 1.28 (0.97-4.03) (0.39-2.35) (0.52-3.17)

T-cell subset CD4+ CD8+ Assay CD25/CD134 ICS ICS Prevalence OR P Prevalence OR P Prevalence OR P n (%) (95% CI) n (%) (95% CI) n (%) (95% CI) Number of HPV types detected 0.042a 0.789a 0.292a 0 8 (30.8) 1 5 (19.2) 1 3 (11.5) 1 1 to 2 25 (55.6) 2.8 7 (15.6) 0.77 9 (20.0) 1.92 (1.01-7.80) (0.22-2.74) (0.47-7.83) ≥3 34 (57.6) 3.1 12 (20.3) 1.07 13 (22.0) 2.17 (1.15-8-15) (0.34-3.43) (0.56-8.37) Anal SIL screening diagnosis Cytology 0.092 0.922 0.426 ≤ASC-H 50 (48.1) 1 18 (17.3) 1 22 (21.2) 1 HSIL 15 (68.2) 2.31 4 (18.2) 1.06 3 (13.6) 0.59 (0.87-6.14) (0.32-3.51) (0.16-2.17) Worst histology 0.623 0.687 0.965 ≤AIN2 50 (52.1) 1 18 (18.8) 1 18 (18.8) 1 AIN3 18 (47.4) 0.83 6 (15.8) 0.81 7 (18.4) 0.98 (0.39-1.76) (0.30-2.23) (0.37-2.57) High-grade disease 0.966 0.950 0.825 (HSIL cytology or AIN2/3 histology) Not detected 42 (50.6) 1 15 (18.1) 1 15 (18.1) 1 Detected 26 (51.0) 1.02 9 (17.7) 0.97 10 (19.6) 1.11 (0.51-2.04) (0.39-2.42) (0.45-2.69) a

Ptrend

107

108

Table 3.3 – Clinical associations with positive HPV16-E7-specific T-cell responses (univariable logistic regression) T-cell subset CD4+ CD8+ Assay CD25/CD134 ICS ICS Prevalence OR P Prevalence OR P Prevalence OR P n (%) (95% CI) n (%) (95% CI) n (%) (95% CI) Age (years) 0.438a 0.610a 0.821a 35-44 12 (37.5) 1 6 (18.8) 1 5 (15.6) 1 45-55 8 (13.8) 0.27 4 (6.9) 0.32 12 (20.7) 1.41 (0.09-0.75) (0.08-1.24) (0.45-4.43) ≥55 12 (27.3) 0.63 6 (13.6) 0.68 8 (18.2) 1.20 (0.24-1.66) (0.20-2.36) (0.35-4.08) HIV status 0.764 0.415 0.909 Uninfected 25 (24.3) 1 11 (10.7) 1 19 (18.5) 1 Infected 7 (22.6) 0.88 5 (16.1) 1.61 6 (19.4) 1.06 (0.40-1.98) (0.51-5.05) (0.38-2.94) CD4+/CD8+ T-cell 0.824 0.474 0.08 ratio - 0.95 - 0.80 - 1.48 (0.62-1.47) (0.43-1.48) (0.95-2.31) Anal HPV status HPV16 0.171 0.179 0.409 Not detected 25 (27.2) 1 9 (9.8) 1 16 (17.4) 1 Detected 6 (15.8) 0.50 7 (18.4) 2.08 9 (23.7) 1.47 (0.19-1.35) (0.71-6.07) (0.59-3.71) Any high-risk HPV type 0.493 0.411 0.153 Not detected 11 (20.8) 1 5 (9.4) 1 7 (13.2) 1 Detected 20 (26.0) 1.34 11 (14.3) 1.60 18 (23.4) 2.00 (0.58-3.09) (0.52-4.91) (0.77-5.21)

T-cell subset CD4+ CD8+ Assay CD25/CD134 ICS ICS Prevalence OR P Prevalence OR P Prevalence OR P n (%) (95% CI) n (%) (95% CI) n (%) (95% CI) Number of HPV types detected 0.568a 0.048a 0.181a 0 6 (23.1) 1 0 (0) - 4 (15.4) 1 1 to 2 9 (20.0) 0.83 6 (13.3) 1 6 (13.3) 0.85 (0.26-2.68) (0.22-3.33) ≥3 16 (17.2) 1.24 10 (17.0) 1.33 15 (25.4) 1.88 (0.42-3.64) (0.44-3.97) (0.56-6.32) Anal SIL screening diagnosis Cytology 0.134 0.323 0.127 ≤ASC-H 22 (21.2) 1 11 (10.6) 1 18 (17.3) 1 HSIL 8 (36.4) 2.13 4 (18.2) 1.88 7 (31.8) 2.23 (0.79-5.72) (0.54-6.56) (0.80-6.25) Worst histology 0.629 0.151 0.019 ≤AIN2 24 (25.0) 1 14 (14.6) 1 13 (13.5) 1 AIN3 8 (21.1) 0.80 2 (5.3) 0.33 12 (31.6) 2.95 (0.32-1.98) (0.07-1.51) (1.20-7.25) High-grade disease 0.732 0.551 0.015 (HSIL cytology or AIN2/3 histology) Not detected 19 (22.9) 1 11 (13.3) 1 10 (12.1) 1 Detected 13 (25.5) 1.15 5 (9.8) 0.71 15 (29.4) 3.04 (0.51-2.59) (0.23-2.18) (1.24-7.44) a

Ptrend

109

3.4.3 Clinical associations with HPV16-E6 and E7 CD4+ T-cell responses

CD4+ T-cell response frequency using the CD25/CD134 assay was not associated with detection of concurrent high-grade disease (E6: 51% no high-grade disease vs 51% high-grade disease, P=0.966 and E7: 23% no high-grade disease vs 26% high-grade disease, P=0.732), with similar findings using the ICS assay (Tables 3.2 and 3.3). CD4+

T-cell response frequency was not associated with age, HIV or anal HPV16 status.

Increasing number of HPV types detected was significantly associated with an increased chance of detecting HPV16-specific CD4+ T-cell responses (to E6 on the CD25/CD134 assay, and to E7 on the ICS assay; Ptrend=0.042 and 0.048, respectively). There were fewer E6-specific CD4+ T-cell responses on both CD25/CD134 and ICS assays with increasing age (P=0.064 and 0.125, respectively). E6-specific CD4+ T-cell responses on the CD25/CD134 assay were more common among those participants with any high- risk type HPV detected (P=0.059) and with anal HSIL diagnosed on cytology

(P=0.092).

There was no difference in mean CD4+ T-cell counts between HIV-uninfected men with or without E6-specific CD4+ T-cell responses on the CD25/CD134 assay (818 vs

825 cells/μL respectively, P=0.889). Results were similar for the ICS assay and for E7- specific CD4+ T-cell responses on both assays (data not shown). For HIV-infected men, those with E6-specific CD4+ T-cell responses had higher mean CD4+ T-cell counts

(662 cells/μL) than men without (510 cells/μL, P=0.091). Similar trends were seen for the ICS assay and E7-specific CD4+ T-cell responses on the CD25/CD134 assay, although the differences were not significant (data not shown).

For HIV-infected men, those reporting a prior AIDS-defining illness were less likely to have E6-specific CD4+ T-cell responses on the CD25/CD134 assay (27.3% vs. 60.0%;

OR 0.25 [95% CI 0.05-1.24], P=0.090). As most (≥90%) of the 31 HIV-infected

110

participants were on antiretroviral therapy (ART) and virally suppressed, meaningful analysis of the relationships of ART and HIV viral load to T-cell responses was not possible.

The only variable independently associated with HPV16-specific CD4+ T-cell responses was the detection of more anal HPV types: E6-specific CD4+ T-cell responses on the CD25/CD134 assay (OR 1.65 [95% CI 1.02-2.68], Ptrend=0.042) and

E7-specific CD4+ T-cell responses on the ICS assay (OR 2.38 [95% CI 1.01-5.59],

Ptrend=0.048). These ORs and significance levels did not change significantly after anal

HPV16 status, CD4+/CD8+ T-cell ratio, HIV status and age were forced into the multivariable model (data not shown). In the forced model, having HIV infection and higher CD4+/CD8+ T-cell ratio both were independent predictors for lower odds of having E6-specific CD4+ T-cell responses on the CD25/CD134 assay (OR 0.33 [95%

CI 0.11-0.99], P=0.048; OR 0.54 [95% CI 0.31-0.95], P=0.031; respectively).

3.4.4 Clinical associations with HPV16-E6 and E7 CD8+ T-cell responses

Being HIV-infected was the only independent variable associated with increased odds of having E6-specific CD8+ T-cell responses (OR 2.57 [95% CI 1.01-6.54], P=0.047), even after forced adjustment for anal HPV16 status, CD4+/CD8+ T-cell ratio, HIV status and age (Tables 3.2 and 3.3). There was no significant difference in mean CD8+

T-cell counts between men with or without these responses (451 vs 467 cells/μL respectively for HIV-uninfected men, P=0.782; 923 vs 829 cells/μL for HIV-infected men, P=0.634).

High-grade disease was independently associated with E7-specific CD8+ T-cell responses (OR 4.09 [95% CI 1.55-10.77], P=0.004). Higher CD4+/CD8+ T-cell ratio

111

was also independently associated with this response (OR 1.77 [95% CI 1.10-2.82],

P=0.018). These ORs and significance levels were similar after anal HPV16 status, HIV status and age were forced into the multivariable model (data not shown).

3.4.5 HPV16-E6 and E7 T-cell responses and spontaneous HSIL regression

Using longitudinal cytology and histology data from the main study, 26 men had high- grade disease diagnosed at their baseline study visit, which was a mean one year (SD

6.4 months) prior to their Immunology Substudy visit. Of these men, six (23%) experienced spontaneous regression by the time T-cell assays were performed (Figure

3.4). Five (83%) of these six regressors had E6-specific CD4+ T-cell responses on the

CD25/CD134 assay, compared to seven of 20 (35%) non-regressors (P=0.065, Fisher’s exact test). ICS and E7-specific T-cell responses were not associated with regression.

112

Figure 3.4 – Clinical course of participants with baseline high-grade disease who experienced spontaneous regression.

Solid line = completed visits; dashed line = future visits; double vertical mark = visit at which T-cell assays were performed; asterisk (*) = incident HIV infection.

113

This is the first study to describe systemic HPV16-E6 and E7-specific T-cell responses in the setting of anal HSILs. These responses were detected in 60% of MSM aged ≥35 years, but were not associated with concurrent anal HSIL. However, E6-specific CD4+

T-cell responses on the CD25/CD134 assay showed a trend towards association with recent spontaneous regression of anal HSIL.

Systemic HPV16-specific T-cell responses were small, about ten-fold lower than responses to CMV, and E6 elicited more T-cell responses than E7. These findings are in keeping with those in women with cervical HSIL and in healthy adults (72, 82, 252,

253). Given the E6 oncoprotein has functions distinct from E7, it is not surprising that they interact with the immune response differently, although the exact mechanisms remain unknown (13, 20). E7 appears less immunogenic (or more immunosuppressive) than E6. E7-specific CD4+ T-cell responses were very small and predominantly with

IL2, not IFN. In contrast, E7-specific CD8+ T-cells were generally IFN-dominant; therefore, it seems unlikely that E7 interferes with IFN production in lymphocytes. IL2 production from E7-specific CD4+ T-cells has been previously described in women with cervical HSIL (254).

The CD25/CD134 assay was more sensitive than ICS at detecting systemic CD4+ T-cell responses to E6 and E7. This may be due to minimal background reactivity, and is in keeping with the performance of the CD25/CD134 assay with other antigens (212, 213,

215, 248). This may account for our ability to show a trend between recent anal HSIL regression and systemic E6-specific CD4+ T-cell responses, in contrast to the conflicting findings in the cervical field using less sensitive assays (74, 82, 83, 255).

114

Another possible explanation is the CD25/CD134 assay can detect antigen-specific

CD4+ T-cells that do not produce the Th1 cytokines IFNγ and IL2, for example, regulatory T-cells. CD4+ regulatory T-cells have been described in both cervical SIL and cancer (85, 86), and they were found to predict therapeutic vaccine failure in vulvar

HSIL (154). The CD25/CD134 assay has been shown to consistently detect antigen- specific regulatory T-cells (219).

CD8+ T-cell responses to E6 were more likely in HIV-infected men, independent of

CD8+ T-cell counts and CD4+/CD8+ T-cell ratios. This has not been previously described in peripheral blood, although Kobayashi et al described similar findings in the cervical local immune response (96). We did not find HIV status or CD4+ T-cell count to be a significant factor for detecting E6 or E7-specific CD4+ T-cell responses. This may be due to successful ART among most of our HIV-infected participants, as evidenced by mean CD4+ T-cell count of 584 cells/μL and 90% reporting an undetectable HIV plasma viral load. Low nadir, rather than current, CD4+ T-cell counts are associated with increased incidence of anal cancer in HIV-infected adults (93, 109).

No systematically collected, long-term, historical CD4+ T-cell count data are available in our cohort. However, the possible trend we found between self-reported AIDS and decreased prevalence of E6-specific CD4+T-cell responses is in keeping with the hypothesis that cumulative rather than current CD4+ T-cell immunodeficiency is the more important factor in determining systemic HPV16-specific T-cell immunity.

HPV16-specific CD4+ T-cell responses were associated with increasing number of anal

HPV types detected. Possible explanations for this include behavioural confounders which may increase the risk of repeated exposures to anal HPV, and potential cross- reactivity of the T-cell assays to non-HPV16 types.

115

Men with anal high-grade disease were more likely to have E7-specific CD8+ T-cell responses, implying that persistent anal high-grade disease may be an ongoing source of

E7 antigen that stimulates an effector, cytotoxic response. However, these responses were not associated with regression of high-grade disease, calling into question the efficacy of these IFNγ-predominant E7-specific CD8+ T-cell responses. Similar findings have been described in other anogenital SIL settings (74, 256).

Strengths of this study include the large, well-characterised population that includes both HIV-infected and uninfected MSM. The CD25/CD134 assay we have optimised for measuring systemic E6/E7-specific CD4+ T-cells is sensitive, and may correlate with clinically significant outcomes such as spontaneous regression of HSIL. Therefore, it has the potential to be incorporated into screening and treatment algorithms for the prevention of anal cancer, perhaps by contributing to an individual patient’s risk category for progression to cancer and thereby determining a safe interval before repeat screening. Given the scarcity of HRA and anal cytology expertise, the use of an immune response-based assay may result in significant cost savings if use of the assay safely increased the period between HRAs. Further, the CD25/CD134 assay is relatively simple to perform because it can be done on whole blood (no need for cell separation), and relies on surface rather than intracellular staining. Therefore, the assay is feasibly translatable to a diagnostic laboratory. Finally, we have used a composite endpoint

(cytology and histology) for detection of HSIL in our analyses, which improves the sensitivity of anal HSIL diagnosis (227).

Our study has limitations. First, the T-cell assays could not be prospectively validated in an independent, well-characterised population against known standards, therefore they remain investigative. Second, there was no pre-planned sample size due to the descriptive and exploratory nature of this study, but these results will form the basis of

116

sample size calculations for future work. Third, we only measured two Th1 cytokines, and therefore cannot describe the true cytokine profile of the HPV-specific T-cell responses. However, the CD25/CD134 assay readout can be adapted to achieve this

(257). Additionally, our analyses were mostly cross-sectional and, although our sample size is one of the largest to date, numbers for some analyses were small. Therefore, the associations we have described require careful interpretation, and many of the associations though showing trends, did not reach statistical significance. Lastly, the definition of HSIL regression remains problematic across the field. Limitations with the diagnostic performance of anal cytology, HRA and histology, and the multifocal nature of anal HPV disease all contribute to difficulties in defining HSIL regression.

In summary, a simple assay for HPV16-specific CD4+ T-cells can detect systemic E6- specific responses in more than half of MSM who undergo screening for anal HSIL.

Positive responses trended towards a significant association with recent anal HSIL spontaneous regression; the value of this assay in the screening and treatment of HPV- induced precancerous lesions should be explored. Future work should focus on validating these early findings prospectively and longitudinally, how these systemic responses reflect responses in the anal mucosa, the roles of HIV infection, ART and

CD4+ T-cell recovery, and on more detailed immunological studies of the detected responses, including epitope mapping, further phenotypic characterisation including identification of regulatory subsets, assay comparison with ELISPOT, and correlation with E6/E7 serology.

117

Supplementary Figure 3.1 – Correlation between positive responses on

CD25/CD134 assay and cytokine production on ICS.

Assay correlations were assessed with Spearman’s correlation coefficients (r).

118

Positive responses to CMV on the CD25/CD134 assay positively correlated with production of IFN (r=0.64, 95% CI 0.54-0.74, P<0.0001) and IL2 (r=0.41, 95% CI

0.26-0.55, P<0.0001) by CD4+ T-cells. The same correlation between the CD25/CD134

119

and ICS assays was not seen for E6 and E7. For dichotomised results of CMV response, there was slight to fair agreement between the CD25/CD134 assay and production of both cytokines from CD4+ and CD8+ T-cells (kappa range 0.07-0.27, all P≤0.02). For dichotomised E6 and E7 responses, slight agreement was found between CD25/CD134 positive responses and IL2 production from E7-specific CD4+ T-cells (kappa=0.21,

95% CI 0.02-0.39, P=0.0045).

T-cell assay performance

Median background reactivity of the CD25/CD134 assay was 0.003% (interquartile range [IQR] 0 – 0.006%) of CD4+ T-cells. The intra-assay coefficient of variation (CV) over 10 replicates was 6.1% for SEB and 6.2% for CMV. The inter-operator CV was

3.5% for SEB and 9.4% for CMV. For the ICS assay, median background reactivity of

CD4+ T-cells for IFN was 0.013% (IQR 0.004 – 0.024%) and 0.016% (IQR 0.007 –

0.058%) for IL2. Median background reactivity of CD8+ T-cells for IFN was 0.026%

(IQR 0.008 – 0.051%) and 0.023% (IQR 0.010 – 0.081%) for IL2. The intra-assay CV for CD4+CD69+IFN+ over 6 replicates was 7.9% for SEB and 17.5% for CMV/CEF; for CD8+CD69+IFN+ it was 3.4% for SEB and 18.8% for CMV/CEF.

120

This page has been intentionally left blank

121

Supplementary Figure 3.2 – Qualitative analysis of individual positive responses ranked by response size.

122

NB. Four digit numbers starting with 1 are individual participant identifiers.

123

This page has been intentionally left blank

124

Data presented in part at the World STI & HIV Congress (poster 10.10) and

Australasian HIV & AIDS Conference (poster 9), Brisbane, 13-18th September 2015.

125

Introduction: The host cellular immune response plays an integral role in controlling

HPV-induced precancerous lesions. The local cellular immune response in the anal mucosa is poorly described. We aimed to describe and quantify T-cells in anal biopsies of men with prior anal high-grade (grade 2 or 3) squamous intraepithelial lesions

(HSIL), and correlate this with systemic HPV16-specific T-cell responses.

Methods: The Study of the Prevention of Anal Cancer (SPANC) is a longitudinal natural history study of anal HPV infection in men who have sex with men aged ≥35 years. Anal HPV DNA and systemic HPV16-E6 and E7-specific T-cell responses were detected as described in Chapter 3. Participants with prior anal HSIL were selected and their subsequent high-resolution anoscopy (HRA) guided intra-anal biopsies taken at one visit were studied. Presence of lymphoid aggregates in these biopsies were assessed by visual inspection of haematoxylin & eosin-stained sections. Additional sections were immunofluorescently stained to enumerate stromal and intra-epithelial CD4+ and CD8+

T-cells. Whole slide imaging to reveal full tissue architecture at high resolution (600x) was used. T-cell density for each biopsy was estimated by dividing absolute T-cell

2 counts by biopsy area (mm ). Student’s t-test of log10-transformed T-cell density was used to compare means; a generalized, linear model was used to determine factors associated with total (CD4 plus CD8) T-cell density (biopsy-based analysis with intra- subject adjustment).

Results: Of 26 men (mean age 53 years, SD 10.5), 7 (27%) were HIV-infected.

Seventeen (68%) had concurrent anal HPV16 in anal swabs. A total of 44 biopsies were taken a mean of 12.8 months (SD 6.7) after initial HSIL diagnosis. Twenty-four biopsies (55%) had lymphoid aggregates localised in the stroma adjacent to the epithelium, and 26 (59%) biopsies had persistent HSIL. Presence of lymphoid 126

aggregates was associated with higher CD4+ T-cell density (mean 192 vs. 69 cells/mm2,

P<0.001), but not higher CD8+ T-cell density (106 vs. 62 cells/mm2, P=0.077). A biopsy diagnosed with HSIL was significantly associated with higher total T-cell density [odds ratio (OR) 11.80, 95% CI 1.51 – 92.08, P=0.02], as was having anal

HPV16 detected (OR 14.08, 95%CI 1.15 – 172.71, P=0.04). Presence of low-risk HPV genotypes was not associated (OR 1.37, 95%CI 0.12 – 15.14, P=0.80). There was no association found between biopsy T-cell density and systemic immune responses to

HPV16 oncogenic proteins. Participants classified as recent regressors tended to have lower total T-cell density in their biopsies compared to non-regressors (OR 0.21, 95%

CI 0.03 – 1.31, P=0.10).

Conclusion: CD4+ T-cell enriched lymphoid aggregates in the anal mucosa were associated with anal HSIL and HPV 16.

127

The host cellular immune response plays an integral role in controlling HPV-induced precancerous lesions. While we have detected systemic HPV16-specific T-cell responses in up to half of MSM aged ≥35 years, it is unknown if these circulating cells reflect HPV-related events in the anal mucosa. During regular multidisciplinary reviews of SPANC histology, roughly 20% of all intra-anal biopsies were reported to show

“follicular inflammation” i.e. lymphoid aggregates usually situated in the stroma

(Figure 4.1). These local immune responses have not been previously described in the anus.

Several studies have described the local immune response in CIN and correlated this with risk of progression, persistence or regression. In a longitudinal study of 125 women with low-grade CIN followed prospectively for a year, Woo et al performed immunohistochemistry for lymphocyte subtypes on cervical biopsies (81). They found that women with low-grade CIN that subsequently regressed had significantly higher numbers of cytotoxic (granzyme B+) cells in their initial biopsies, and that regressors had higher CD8/CD4 ratios compared with progressors. They concluded that early infiltration of low-grade CIN lesions by cytotoxic T-cells may protect against progression.

Trimble et al found that CD8+ T-cell infiltration of high-grade CIN lesions correlated with subsequent regression, and that T-cell access to cervical tissue was mediated by adhesion molecules on vascular endothelium (258). Ovestad et al also found that regressing high-grade CIN lesions had higher numbers of stromal CD8+ T-cells by quantitative immunohistochemistry, and that stromal CD8+ T-cell numbers independently predicted regression (although none of the regressing high-grade CIN

128

Figure 4.1 – Anal biopsies with lymphoid infiltrates

Four examples of intra-anal biopsies from SPANC reported by histopathologists to show “follicular” or “intense inflammation” (arrows). Haematoxylin & eosin stained.

Each scale bar represents 500 μm.

129

lesions in this study contained HPV16) (259). Kojima et al, using cervical cytobrush samples and flow cytometry analysis, found other local immune cell populations

(αEβ7+ intraepithelial lymphocytes, CD4+ regulatory T-cells) correlate with spontaneous regression of CIN (260, 261). Cross-sectional studies have described T-cell subsets using immunohistochemistry in cervical biopsies with and without CIN, (80) and the effect of HIV infection (95, 96).

Other than Gervaz et al who were the first to describe immune cells in the normal anal mucosa using immunohistochemistry (262), there are only two published studies which assess local immune responses in the anal mucosa also from the same group. They found that HIV infection was an independent risk factor for having lower numbers of

CD1a+ Langerhans’ cells (OR 6.0, P<0.001) (263), and that lower CD1a+ counts were predictive of AIN3 and anal cancer (OR 9.4, P<0.0001) (264).

This pilot study aimed to describe and quantify T-cell infiltration in anal biopsies of men diagnosed with anal HSIL, and correlate this with systemic HPV16-specific T-cell responses. We explored CD4+ and CD8+ T-cell counts, their distribution in stromal vs. intraepithelial compartments, and investigated for potential factors that associated with quantity of T-cell infiltration.

4.3.1 Study Population

We selected 26 SPANC Immunology Substudy participants who had prior anal HSIL diagnosed. This population was chosen as we reasoned that it was most likely to have abnormal tissue related to persistent HPV infection and ongoing HPV16-E6 and/or E7 expression, and they had systemic HPV16-E6/E7-specific responses measured in

130

Chapter 3. Of these 26, six were defined as recent regressors as detailed in Chapter 3.

Participants had a median of 2 intra-anal biopsies (range 0-4) taken under HRA- guidance at their Immunology Substudy visit, yielding a total of 44 biopsies that were included in this pilot study.

4.3.2 Haematoxylin & eosin stained diagnostic slides

After formalin fixation and paraffin embedding, SPANC biopsies have nine levels prepared, of which levels one to three and seven to nine are stained with haematoxylin and eosin (H&E) for diagnostic reporting (Figure 4.2) (245). For each diagnostic level two consecutive 3μm sections were taken (whereas only one section is taken at each level for research sections B1-9). The two consecutive sections corresponding to level

A9 were scanned and digitised using an Aperio ScanScope XT (Leica Biosystems,

Nussloch, Germany) slide scanner with an Olympus UPlanSApo 20x/0.75 (with 2x auxiliary magnification to produce 40x) objective at the Histology & Microscopy Unit,

School of Medical Sciences, University of New South Wales (UNSW).

Digitised slides were visualised and analysed using Aperio ImageScope software version 12.0.1.5027 (Leica Biosystems). Biopsy area (μm2) was measured by hand annotating the digital slide (Figure 4.3). For each biopsy, these measures were taken on the two digitised consecutive sections at the same level (A9) and averaged.

Each biopsy was assessed visually for the presence or absence of lymphoid aggregates on H&E at low power (20-40x). Examples of these aggregates are show in Figures 4.1 and 4.6. Of the 24 biopsies with lymphoid aggregates, 13 (54%) had one aggregate, 10

(42%) had two aggregates and one (4%) biopsy had three aggregates. The median

131

Figure 4.2 – Diagram of an anal

biopsy block processed as per

SPANC protocols

Levels A1-3 and A7-9 are H&E

stained. Levels A4-6 are left unstained

in case diagnostic histochemistry

stains (e.g. p16INK4a) are required.

Levels B1 and B9 are H&E stained.

Levels B6-8 (each 3μm thick) are

reserved for immunostaining and two

sections were used for this pilot

substudy.

132

Figure 4.3 – Annotated digitised slides of a diagnostic haematoxylin & eosin stained anal biopsy

(a) and (b) are two consecutive sections from the same level, corresponding to A9 in

Figure 4.2. Green annotations in (a) and (b) measure biopsy area in μm2. Final biopsy area was averaged between the two measurements.

(c) is the same image as (a). Blue annotations in (c) measure maximal diameter of lymphoid aggregates present. Measurements in (c) were taken once on the 24 biopsies with lymphoid aggregates present.

Annotations were drawn using Aperio ImageScope version 12 (Leica Biosystems,

Nussloch, Germany). The freeform annotations in (a) and (b) were hand drawn using a

Jot Pro stylus (Adonit; Austin, USA & Taipei, Taiwan) on a 15.6-inch touch screen

(resolution 3200 x 1800 pixels). Each scale bar represents 700 µm.

133

maximal diameter of these aggregates was 514 µm (IQR 371 – 916). The smallest aggregate measured 224 µm in maximal diameter (see Figure 4.3c).

4.3.3 Immunofluorescent staining for CD4+ and CD8+ cells

Paraffin embedded anal tissue blocks were retrieved and further sectioned as described in Figure 4.2. For each biopsy, sections B6 and B7 were used for this pilot substudy and dual-stained for CD4 and CD8 using primary and secondary antibodies as detailed below. The primary antibodies were omitted for one section as a negative (“no primary”) control. Biopsies were stained over six runs; each run included human normal tonsil sections as controls.

The staining protocol is illustrated in Figure 4.4. In detail, sections were rehydrated in xylene (2 x 5 minutes), 100% ethanol (3 x 1 minute), 70% ethanol (1 minute) and then running tap water for 1 minute using a Gemini AS automated slide stainer (Thermo

Scientific, Waltham, USA). Antigen retrieval was heat-induced (95°C for 15 minutes) in a Dako PT Link pre-treatment module (Agilent Technologies, Glostrup, Denmark) using a high pH (9.0) EDTA-based retrieval solution (Dako K8000, Agilent

Technologies). After rinsing with distilled water, slides were blocked in 2% skim milk solution for 20 minutes at room temperature, rinsed in Tris-buffered saline x3 (TBS,

Dako K8007, Agilent Technologies), then blocked with 10% goat serum (Sigma-

Aldrich G9023, St Louis, USA) in TBS for 30 minutes at room temperature to minimise non-specific staining.

Tissue was then incubated with two primary antibodies concurrently (as specified in

Table 4.1) for 90 minutes at room temperature, followed by 3 x 5-minute rinses in TBS.

134

Then tissue was incubated with fluorescently-labelled secondary antibodies for 45 minutes in the dark at room temperature, followed by 3 x 5-minute rinses in TBS.

Figure 4.4 – Two-colour immunofluorescent staining protocol

AF = Alexa Fluor®, DAPI = 4',6-diamidino-2-phenylindole; EDTA = ethylenediaminetetraacetic acid.

135

Table 4.1 – Antibodies used for two-colour immunofluorescent staining Manufacturer Target Species Isotype Dilution* (Catalogue no.) Primary (monoclonal) Clone Cell Marque Human CD4 Rabbit IgG 1:20 SP35 (104R-15) Dako Human CD8 Mouse IgG , κ 1:150 C8/144B 1 (M7103) Secondary (polyclonal) Conjugate Life Technologies Rabbit IgG Goat IgG 1:200 AF594 (A-11012) Life Technologies Mouse IgG Goat IgG 1:200 AF488 (A-11001) *All antibodies diluted using antibody diluent containing TBS, sodium azide and protein (Dako K8006, Agilent Technologies).

To reduce autofluorescence, tissue was incubated with 0.1% Sudan Black B in 70% ethanol for 1 minute and rinsed in distilled water (3 x 5 minutes) after antibody staining.

Finally, specimens were mounted using ProLong® Gold Antifade Mountant with DAPI

(4',6-diamidino-2-phenylindole) nuclear stain (Life Technologies P36935, Thermo

Fisher Scientific, Waltham, USA). Slides were then allowed to cure overnight at room temperature and then subsequently stored in the dark at 4°C until microscopy acquisition within one week of staining (median = 2 days, IQR 2-6 days).

4.3.4 Microscopy

Images were acquired on a fluorescent microscope (IX71; Olympus) controlled by a

DeltaVision Elite System (Applied Precision, GE Healthcare) with softWoRx version

6.1.1 release 5 software (Applied Precision) and fitted with an electron multiplying charge coupled device (EMCCD) high-sensitivity camera (Evolve 512, Photometrics) that acquires images at 512 x 512-pixel resolution. We used an Olympus 60x high numerical aperture (1.42) oil immersion objective (with Applied Precision immersion oil N=1.514) and images were acquired with a pixel size of 0.26570 μm. 136

Each image was acquired using a seven-colour Insight solid-state illumination light source with accompanying filter sets supporting excitation and emission of DAPI,

AF488 and AF594. Image capture on the Evolve 512 EMCCD camera was set at 50X gain. Exposure and transmission settings for each channel were set on a human tonsil stained positive control to establish the dynamic range of the sample in addition to avoiding any fluorescence bleaching. This enabled repeated image acquisition to establish hard-focusing across large distances, while preserving fluorophore integrity.

Settings were optimised on each sample but always kept constant between “no primary” control and stained slides (Table 4.2).

Table 4.2 – Summary of fluorescence microscopy acquisition settings (N=76) Channel* DAPI AF488 AF594 Transmission (%) mean 11.7 9.9 27.7 (range) (10 – 32) (5 – 10) (10 – 32) Exposure time (seconds) mean 0.014 0.017 0.040 (SD) (0.0049) (0.0109) (0.0355) *Evolve 512 EMCCD camera gain set at 50X. SD = standard deviation

Limits of each tissue sample were defined by visual inspection and a rectangular grid set up for panel collection in a snake-by-rows manner. High content image acquisition in this context of whole slide fluorescence imaging provided two main challenges. Firstly, over large distances, the influence of coverslip glass imperfections and/or mounting variation (i.e. slope) resulted in microscope focal plane drift. Secondly, samples often varied in focal plane across the sample (i.e. sample topology). To address the issue of objective to coverslip variation, focus was maintained using a hardware autofocus module that uses a system of lasers that detects and corrects for focal distance to the coverslip (DeltaVision UltimateFocus). To address tissue topology variation, we further

137

applied contrast-based autofocusing using the nuclear stain DAPI. Whole slide imaging was achieved by panel stitching using softWoRx (Applied Precision). The smallest biopsy had 270 images (= 1.7 mm2) and the largest biopsy had 3774 images (= 23.3 mm2).

4.3.5 CD4+ and CD8+ cell enumeration

CD4+ and CD8+ cells in each biopsy were manually counted using Fiji (265) software installed with Cell Counter (266) and Grid (267) plugins. Brightness and contrast settings were optimised for visual inspection of each image (γ=1), but kept constant between matching stained and “no primary” control slides.

A cell was considered CD4 (red) or CD8 (green) positive if continuous, linear cell membrane fluorescence staining could be discerned encircling a cell nucleus (blue) at full (600x) magnification. Cell location (stromal vs intraepithelial) was noted (Figure

4.5c).

“No primary” control slides were counted in the same manner. Of 33 “no primary” control slides, 30 slides had no discernible positive cells. Three slides had one or two cells considered positive, and these counts were subtracted from those obtained from the matching stained slide. Nine biopsies only had one section available for this work, these were stained and counted without a matching “no primary” control.

Three biopsies did not have CD8 data (human normal tonsil in-run positive control was negative), one biopsy was tangentially cut and did not have stroma, and one biopsy was ulcerated and did not have epithelium.

138

4.3.6 Statistics

There was no pre-planned sample size as this was a pilot study. Student’s t-test was used to compare means. T-cell density was calculated by dividing absolute T-cell counts by biopsy area in mm2. CD4 and CD8 T-cell counts were summed for total T-cell density. Density data were not normally distributed, therefore log10 transformed for analysis. A generalized, linear model was used to determine univariate factors associated with total T-cell density. The analysis was per-biopsy-based, with adjustment for clustering and variability per-participant. Multivariate analysis was not performed due to small sample size. Correlations were assessed with Pearson’s correlation coefficients. Analyses were performed using Stata 13.1 (StataCorp, College Station,

USA) and Prism 6.05 (GraphPad Software, La Jolla, USA).

139

Figure 4.5 – CD4+ and CD8+ cell enumeration

140

All images (paired “no primary” and stained) with counting grid and markers are included with this thesis on a Universal Serial Bus (USB) drive in Appendix D.

141

4.4.1 Participant and biopsy characteristics

Twenty-six participants had 44 HRA-guided, intra-anal biopsies taken a mean of 12.8 months (SD 6.7) after initial HSIL diagnosis. Participant characteristics are shown in

Table 4.3 and biopsy characteristics in Table 4.4. Participant mean age was similar to the larger Immunology Substudy group (53 vs. 51 years), as was proportion infected with HIV (27 vs. 23%). Twenty-five (96%) participants had visually abnormal HRAs and therefore at least one biopsy taken, and 17 (68%) had anal HPV16 detected. Twelve

(46%) participants had one biopsy taken, eight (31%) had two and the remaining five

(19%) participants had three or four biopsies taken. Sixteen (62%) participants had

HPV16-E6-specific T-cell responses detected systemically, while eight (31%) participants had T-cell responses to HPV16-E7 detected. Eight (31%) participants had

T-cell responses to both E6 and E7 detected.

Of 44 biopsies, 26 (59%) had a histological diagnosis of HSIL, the majority (n=19,

73%) of which were AIN3. There were no anal SCC or SISCCA diagnoses in this substudy. Mean biopsy area was 3.2 mm2 (standard deviation [SD] 2.13). Twenty-four

(55%) biopsies were judged to have lymphoid aggregates on examination of the H&E diagnostic slide, and these biopsies had significantly higher T-cell densities than those biopsies judged not to have lymphoid aggregates (mean 294 vs 141 cells/mm2, t-test,

P=0.01).

142

Table 4.3 – Participant characteristics All Anal HSIL diagnosed at SPANC study entry (%) 26 (100) Time from HSIL diagnosis to biopsy (months, SD) 12.8 (6.7)

Recent regressor at current visit (%) 6 (23) HIV-infected (%) 7 (27) Age [years, mean (SD)] 53 (10.5) Visually abnormal HRA (%) 25 (96) Number of biopsies per participant [median (range)] 2 (0-4) Anal HPV16 detected at immunology substudy visit (%)* 17 (68) Any hr-HPV (%)* 24 (96) Any lr-HPV (%)* 20 (80) Number of HPV detected [median (IQR)] * 4 (2-5) HIV- HIV+ All Peripheral T-cell counts (cells/μL) Absolute CD4 count [mean (SD)]** 938 (275.4) 585 (175.8) 839 (296.0) Absolute CD8 count [mean (SD)] 602 (220.7) 1232 (595.5) 771 (448.9) CD4:CD8 ratio [mean (SD)] 1.7 (0.62) 0.6 (0.35) 1.4 (0.74) E6 E7 E6 or E7 Systemic HPV-specific T-cell assay positive (%) 16 (62) 8 (31) 21 (81) CD25/CD134 assay (%) 12 (46) 5 (19) 13 (50) Intracellular cytokine stain assay CD4 (%) 6 (23) 5 (19) 8 (31) Intracellular cytokine stain assay CD8 (%) 4 (15) 7 (27) 9 (35) *Data missing on one participant as not assessable on HPV typing assay **Data missing on one HIV-uninfected participant due to laboratory quality control measures 143

Table 4.4 – Biopsy characteristics Lymphocytic infiltration N Presence of lymphoid aggregates on H&E (%) 44 24 (55) Total T-cell density [cells/mm2, mean (SD)] 39 312 (297.0) CD4+ T-cell density [cells/mm2, mean (SD)] 42 189 (181.5) CD8+ T-cell density [cells/mm2, mean (SD)] 39 130 (147.4) Stromal CD4:CD8 ratio [mean (SD)] 40 2.0 (1.06) Intraepithelial CD4:CD8 ratio [mean (SD)] 40 0.5 (0.69) Intraepithelial:stromal T-cell ratio [mean(SD)] 39 0.1 (0.34)

Lymphoid aggregates No (n=20) Yes (n=24) Total (N=44) Histology diagnosis Negative for anal SIL (%) 3 (15) 2 (8) 5 (11) Low-grade SIL (%) 7 (35) 5 (21) 12 (27) High-grade SIL (%) 10 (50) 16 (67) 26 (59) AIN2 (%) 4 (20) 3 (13) 7 (16) AIN3 (%) 6 (30) 13 (54) 19 (43) Non-diagnostic – ulcerated (%) 0 1 (4) 1 (2)

144

4.4.2 Lymphoid aggregates

Figure 4.6 shows biopsies with and without lymphoid aggregates. When lymphoid aggregates were present, mean CD4+ T-cell density was 2.8 fold greater – 192 (95% CI

129-285) vs 69 (95% CI 47-102) cells/mm2 (t-test, P<0.01). Mean CD8+ T-cell density was 1.7 fold greater at 106 (95% CI 71-159) vs. 62 (95% CI 39-99) cells/mm2 (t-test,

P=0.08) (Figure 4.7). The greater mean CD4+ T-cell density associated with presence of lymphoid aggregates remained statistically significant regardless of HIV status (HIV- uninfected, n=30, 83 vs. 203 cells/mm2, P=0.006; HIV-infected, n=12, 47 vs. 164 cells/mm2, P=0.05).

Of lymphocytes present in a biopsy, a mean of 10% (SD 34%) were in the intraepithelial compartment. CD8+ cells were predominant in the intraepithelial compartment (mean CD4:CD8 = 0.5), whereas CD4+ cells were predominant in the stroma (mean CD4:CD8 = 2.0).

Histology diagnosis was not associated with the presence or absence of lymphoid aggregates (χ2 test for trend, P=0.22).

145

Figure 4.6 – Anal biopsies with and without lymphoid aggregates

Representative biopsies without (top two panels) and with (bottom two panels) lymphoid aggregates (arrows). H&E staining (left) with matching CD4/CD8 dual immunofluorescence staining on the right. All scale bars = 300 μm.

146

Figure 4.7 – Anal mucosal lymphoid aggregates are CD4+ T-cell enriched

Bars indicate mean T-cell density; whiskers indicate 95% CI. N=44.

147

4.4.3 Correlation of CD4:CD8 between systemic and mucosal compartments

The systemic CD4:CD8 ratio correlated with the same ratio in both the stromal and intraepithelial compartments of anal biopsies, although the correlation was stronger between the systemic and stromal compartments (r = 0.61) than between the systemic and intraepithelial compartments (r = 0.41) (Figure 4.8).

Figure 4.8 – Correlation of CD4:CD8 between systemic and mucosal compartments

(n = 39)

148

4.4.4 Factors associated with T-cell density

A biopsy diagnosed with HSIL vs. no HSIL was significantly associated with higher total T-cell density (OR 11.80, 95% CI 1.51 – 92.08, P=0.02), and this association was significant for both CD4+ and CD8+ T-cell density (Table 4.5). Having anal HPV16 detected was also significantly associated with higher total T-cell density (OR 14.08,

95% CI 1.15 – 172.71, P=0.04), in particular CD8+ T-cell density (OR 29.44, 95% CI

1.66 – 521.42, P=0.02). Presence of low-risk HPV genotype(s) was not associated with

T-cell density (OR 1.37, 95% CI 0.12 – 15.14, P=0.80).

There was no association found between biopsy T-cell density and systemic immune responses to HPV16 oncogenic proteins. Participants classified as recent regressors showed a trend towards lower total T-cell density in their biopsies compared to non- regressors (OR 0.21, 95% CI 0.03 – 1.31, P=0.10).

149

150

Table 4.5 – Factors associated with T-cell density (univariable generalised linear model)

Total T-cell density CD4+ T-cell density CD8+ T-cell density mean mean mean OR OR OR n (cells/ P n (cells/ P n (cells/ P (95% CI) (95% CI) (95% CI) mm2) mm2) mm2) Histology 39 0.018 42 0.004 39 0.036 No HSIL 16 134 1 16 71 1 16 58 1 11.80 10.94 6.00 HSIL 23 286 26 168 23 106 (1.51 – 92.08) (2.11 – 56.72) (1.13 – 32.00) HIV status 39 0.775 42 0.303 39 0.636 Uninfected 27 217 1 30 138 1 27 78 1 0.71 0.33 1.77 Infected 12 192 12 88 12 96 (0.07 – 7.15) (0.04 – 2.71) (0.17 – 18.8) Anal HPV16 38 0.039 40 0.114 38 0.021 Not detected 10 114 1 12 79 1 10 41 1 14.08 3.93 29.44 Detected 28 249 28 134 28 102 (1.15 – 172.71) (0.72 – 21.44) (1.66 – 521.42) Any hr-HPV* 38 –* 40 –* 38 –* Not detected 1 520 1 1 430 1 1 90 1

Detected 37 198 –* 39 111 –* 37 80 –* Any lr-HPV 38 0.799 40 0.434 38 0.937 Not detected 10 187 1 10 91 1 10 83 1 1.37 2.11 0.90 Detected 28 209 30 123 28 80 (0.12 – 15.14) (0.32 – 13.77) (0.06 – 13.16)

Total T-cell density CD4+ T-cell density CD8+ T-cell density mean mean mean OR OR OR n (cells/ P n (cells/ P n (cells/ P (95% CI) (95% CI) (95% CI) mm2) mm2) mm2) Systemic T-cell immune assays Any CD4+ 39 0.542 42 0.271 39 0.593 No response 14 184 1 14 96 1 14 72 1 1.78 2.34 1.83 Response 25 226 28 136 25 90 (0.28 – 11.28) (0.52 – 10.64) (0.20 – 16.96) Any CD8+ 39 0.433 42 0.731 39 0.453 No response 23 189 1 26 116 1 23 75 1 2.02 1.32 1.96 Response 16 243 16 130 16 96 (0.35 – 11.77) (0.26 – 6.67) (0.34 – 11.33) To HPV16-E6 39 0.868 42 0.959 39 0.966 No response 16 217 1 17 122 1 16 84 1 0.85 0.96 0.96 Response 23 205 25 120 23 82 (0.13 – 5.56) (0.18 – 4.97) (0.14 – 6.58) To HPV16-E7 39 0.372 42 0.223 39 0.423 No response 32 196 1 34 109 1 32 78 1 2.90 3.78 2.36 Response 7 285 8 186 7 108 (0.28 – 30.02) (0.44 – 32.19) (0.29 – 19.24) HSIL regression 39 0.095 42 0.120 39 0.071 Non-regressor 31 234 1 34 132 1 31 94 1 0.21 0.31 0.15 Regressor 8 137 8 83 8 50 (0.03 – 1.31) (0.07 – 1.35) (0.02 – 1.18)

*Only one HIV-uninfected participant did not have any hr-HPV detected, analysis unreliable.

151

We have described CD4+ T-cell enriched lymphoid aggregates in more than half of anal biopsies taken from men with previous HSIL. HSIL histological diagnosis and the presence of anal HPV16 were associated with higher mucosal T-cell density. We did not detect any association between mucosal T-cell density and systemic HPV16-specific T- cell responses on the CD25/CD134 or intracellular cytokine staining assays in this pilot study.

These descriptive findings of tissue lymphocytic infiltration in association with SILs in the anus are very similar to those previously described in the cervix. For example,

Monnier-Benoit et al found CD4+ and CD8+ cells visible as organised lymphoid structures in 7 out of 10 CIN3 biopsies (80). More recently in the setting of HPV16 therapeutic vaccination of high-grade cervical SILs, Maldonado et al demonstrated the formation of lymphoid aggregates with germinal centre features (central CD20+ and associated CD3+ cells) in post-vaccination participants, whereas in the unvaccinated the lymphoid aggregates localised to the stroma without these features (268). Also in keeping with cervical descriptions is the inverted CD4:CD8 ratio in the intraepithelial compartment, with anal intraepithelial lymphocytes being predominantly CD8+ T-cells.

Woo et al found the predominant intraepithelial cells in both low and high-grade CIN were CD8+ and CD56+ cytotoxic T-cells, whereas CD4+ and FOXP3+ T-cells predominated in the stromal compartment (81). This predominance of intraepithelial

CD8+ T-cells was described by Gervaz et al in the normal anal mucosa (262).

We could not demonstrate any association between mucosal T-cell density and systemic

HPV16-specific T-cell responses in this pilot study. However, the strong associations between HSIL diagnosis and higher mucosal T-cell density, and positive relationship between anal HPV16 and higher CD8+ T-cell density are suggestive of an interaction 152

between these factors. In a comparable size group (N=33) of women with cytological cervical abnormalities, Passmore et al were able to demonstrate correlations between cervical and systemic HPV16-specific T-cell responses using cervical cytobrush samples for functional assays (252). Although the pilot work presented in this chapter does not establish the antigen specificity of the lymphocytes described in the anal mucosa, HPV-specificity might be demonstrable if these data were combined with laser capture microdissection (LCM) studies of the causative HPV type in a high-grade lesion with a subadjacent lymphoid aggregate in a particular anal biopsy (245, 269).

Many cervical studies have been able to demonstrate associations between mucosal T- cell immune responses and SIL regression (259, 261), and the recent Maldonado et al therapeutic vaccination paper (268) supports the hypothesis that these localised immune responses may induce regression. Our pilot study is small and regressor status was assigned based on recent, retrospective regression of anal HSIL. However, our finding that recent regressors were likely to have lower mucosal T-cell density, although not statistically significant, suggests that the lymphocytic infiltration may have receded after mediating successful HSIL regression. The only way to resolve these questions is to extend this work to longitudinal anal biopsy samples and a prospective analysis of regression status, all of which is possible and feasible in SPANC.

Strengths of this study include the establishment of a whole-slide imaging technique that reveals the phenotype and location of the mucosal T-cell immune response in the context of local tissue architecture and in high-resolution. A successful two-colour immunofluorescent staining technique maximises the amount of information obtained from precious study tissue samples. This pilot work is the first to quantify and systematically describe T-cell lymphoid aggregates in anal HSIL, and sets the stage for future work to understand their role, if any, in spontaneous regression.

153

Study design was limited by a biased population (previous anal HSIL), lack of controls and relatively small numbers. Ideally this pilot work should be extended to include all

Immunology Substudy participants (regardless of diagnosis at study entry), or longitudinally in a smaller group. An alternative would be a nested case-controlled design. In the absence of a formal definition of “lymphoid aggregate”, independent review by two or more readers should be planned for future work. Likewise, for manual cell counting of large images, independent replicate counts would increase precision. To make reproducible analysis of a larger sample size feasible, an automated cell counting method using image analysis software would be ideal. This approach was attempted but could not be optimised in the time available. Our staining technique only looked at

CD4+ and CD8+ cells, although other tissue immune cells (e.g. B, NK and dendritic cells, macrophages) may be important. Indeed, intraepithelial CD4+ cells in the basal layer are likely to be dendritic rather than T-cells, and future work should look at adapting the current staining protocol for other markers (e.g. CD1a for dendritic cells,

CD20 for B-cells, FOXP3 for regulatory T-cells, perforin/granzyme for cytotoxic cells, etc.)

In summary, T-cell rich lymphoid aggregates are common in anal biopsies and higher total T-cell density is associated with anal HSIL and HPV16. The ability to correlate lesion-specific and detailed histological, viral and immune data in a large, well- characterised, longitudinal clinical cohort (with both HIV-infected and uninfected men) will be a powerful means to further our understanding of the basic mechanisms of immune control of HPV-related disease, and the implications for screening, diagnosis and treatment of anal HSIL and cancer.

154

Using robust, composite cytology-histology endpoint definitions for anal HSIL,

Chapter 2 presents the first published estimate of spontaneous regression incidence in

HIV-infected and uninfected men (240, 270, 271). Regression was more common than progression regardless of the definition employed (Figure 5.1). Others have published comparable progression rates to HSIL (e.g. 10.5% per year in HIV-infected MSM in

Spain (272), 29% and 8% at 12 months of HIV-infected and uninfected Thai MSM respectively (273)). Confirmatory estimates of anal HSIL spontaneous regression in other cohorts are yet to be published.

Rates of progression from anal HSIL to cancer have been published from treated cohorts around the world. In a treated cohort of 138 HIV-infected MSM with anal cancer diagnosis in San Francisco from 1997-2011, progression from anal HSIL to cancer could be documented in 27 men with previous biopsy-proven HSIL at the cancer site, although an incidence estimate was not available due to retrospective study design and loss to follow-up (274). Cachay et al estimated a progression rate of 1.7% over 5 years in HIV-infected patients with baseline HSIL cytology followed for a median of 4 years with access to infrared photocoagulation ablation therapy (275). Similarly, Dalla

Pria et al found a cumulative risk of anal cancer from first AIN3 diagnosis of 3.2% over

5 years in 368 HIV-infected MSM followed for a median of 4.2 years (239).

In keeping with the published literature, the only measure of CD4+ T-cell immunodeficiency that associated with increased risk of progression to AIN3 was nadir absolute CD4+ T-cell count <200 cells/μL [HR 4.66 (95% CI 1.65 – 13.11), P=0.003].

In our data, this factor also showed a trend for reduced risk of AIN3 regression [HR

0.42 (95% CI 0.14 – 1.26), P=0.125], although this was not significant.

156

Figure 5.1 – Incidence and regression of anal high-grade squamous intraepithelial lesions, and progression to cancer in untreated men

Compare Figure 1.6B. Data in blue are from Chapter 2 (240). Data in red are preliminary data from SPANC (276, 277).

To further understand how the T-cell response is important in HPV disease control and the impact of HIV, two assays were optimised to measure HPV16-specific T-cell responses in the circulation. Similar to other studies of these responses in the cervix, the

HPV16 oncogenic protein E6 was more immunogenic than E7, and there is early data to support the hypothesis that HPV16-E6-specific CD4+ T-cell responses are associated with recent anal HSIL regression (Chapter 3). While these responses were common, they were very small in magnitude, previously presenting a technical challenge to measure and quantify ex vivo without in vitro manipulation. These results require validation against other viral and host-related biomarkers, in particular anal HPV DNA detection and typing, in predicting anal HSIL risk of progression, persistence and spontaneous regression.

157

There are no other published data to date on systemic HPV-specific T-cell responses in the context of anal HPV and SIL, although Comi et al presented a poster on HPV16-L1- specific Th1 responses detected on intracellular cytokine staining in 40 HIV-infected men who were screened for anal HPV and SIL with cytology only (278). They described increased local and systemic T-cell activation in men with cytology evidence of (mostly low-grade) SIL regression compared to non-regressors. In other settings,

HPV16-specific T-cell responses have been described. For example, nine of 10 children had detectable HPV16-E2-specific T-cell responses in the Finnish Family HPV cohort, and children of mothers with incident CIN3 had higher IFNγ and TNFα levels compared to other children (279). HPV16-E2-specific T-cell responses were also detected in six of eight women with vulvar intraepithelial neoplasia and their asymptomatic male partners, but were absent when lesions were persistent despite treatment (280).

In Chapter 4, CD4+ T-cell enriched lymphoid aggregates were described in 55% of

HRA-directed anal biopsies taken from MSM previously diagnosed with anal HSIL.

Biopsies with histological HSIL diagnosis were significantly associated with higher total T-cell density in the biopsy (OR 11.80, P=0.02), as was having anal HPV16 detected (OR 14.08, P=0.04). The distribution of lymphoid aggregates in the stroma subadjacent to SILs, and inversion of CD4:CD8 ratio in the intraepithelial compartment, is analogous to descriptions of these lymphoid aggregates in cervical HSIL and cancer

(258, 259). Although a relationship was not demonstrated between lymphoid aggregates and systemic HPV16-specific T-cell responses, recent HSIL regressors showed a trend towards lower total T-cell density in their biopsies compared to non-regressors (OR

0.21, P=0.10), which could be explained by a resolution of local mucosal inflammation after HSIL clearance.

158

The description and quantification of spontaneous regression of anal HSIL (Chapter 2), even if temporary and not to normal, has implications regarding who might be screened and treated for anal HSIL to prevent cancer.

Compared to cervical HSIL, perhaps anal HSIL carries less clinical significance once detected, because regression is much more common than progression to cancer.

Therefore, additional data (e.g. clinical factors, viral or host immune response biomarkers) are required to help distinguish those most at risk of progression for increased surveillance or intervention.

Although anal cytology, HPV DNA testing, HRA and digital anorectal examination exist as screening tests for anal HSIL and cancer, and share with cervical HSIL the aetiology of being HPV-induced, significant differences in the epidemiology and natural history between anal and cervical HSIL impacts on the suitability and effectiveness of screening in preventing anal compared to cervical cancer (Table 5.1). For example,

HSIL prevalence and the at-risk populations are different: >30% prevalence of anal

HSIL for HIV-infected MSM compared to generally ≤2% for cervical HSIL in women

(281), and MSM have a higher risk of having anal HSIL and HIV infection compared to women. Furthermore, anal cytology and HRA are not widely available (127) or government-funded, and there remains no consensus on how to manage anal HSIL once it is identified (282).

Current anal HSIL therapeutic strategies are limited to topical or ablative approaches.

Chapters 3 and 4 demonstrated a potentially central role of HPV-specific T-cell immunity, both systemic and mucosal, in controlling anal HSIL. One major implication of being able to measure systemic HPV16-specific T-cell responses is the potential to

159

develop a host immune response-based biomarker to help predict anal HSIL prognosis, and triage those patients at highest risk of progression for treatment. While traditional functional T-cell assays are usually too cumbersome for the diagnostic laboratory, the

CD25/CD134 assay described in Chapter 3 is sensitive enough to detect these small responses, and simple enough to be feasibly implemented in a routine diagnostic setting.

Table 5.1 – Wilson and Jungner criteria (283) for diseases suitable for screening applied to cervical and anal cancer

Cervical cancer Anal cancer

An important health problem Yes Yes

Accepted treatment exists Yes Yes

Facilities for diagnosis and treatment available Yes Yes

Recognisable latent or early stage Yes Yes

Suitable test and examination Yes Unknown

Test is acceptable to the population Yes Unknown

Natural history adequately understood Yes No

Agreed policy on whom to treat Yes No

Cost-benefit Depends Unknown

Continuing process Yes Unknown

Our data in Chapter 4 is suggestive that anal HPV16 and HSIL may be responsible for the presence of lymphoid aggregates in the anal mucosa that are CD4+ T-cell enriched.

If these findings can be confirmed, and HPV16-specificity demonstrated in these mucosal immune cells, it would strongly imply that CD4+ T-cells (including regulatory and other subsets) play an important role in controlling HPV16-induced anal HSIL. This 160

would provide a plausible biological mechanism such that consideration should be given to trialling treatment approaches in patients that harness this anti-tumour response e.g. therapeutic vaccination (284-287), adoptive T-cell therapies (288) including chimeric antigen receptor T-cells (289, 290), and systemic immune checkpoint inhibitors such as anti-PD-1 monoclonal antibodies (168, 169) and other T-cell modulators. At present, the use of immune checkpoint inhibitors in anal HSIL, an asymptomatic condition, may be limited by significant toxicity (291).

The description of CD4+ enriched lymphoid aggregates in the anal mucosa of MSM with HSIL also has implications for HIV transmission, and may provide a mechanism to explain the epidemiological association noted between HPV infection and increased risk of HIV acquisition (292). Further clinical and biological studies to elucidate this relationship are warranted and of both clinical and public health significance

In HIV infection, data regarding the effect of cART on anal cancer risk are conflicting.

Protease inhibitors have been associated with higher anal cancer risk in two large cohorts (293, 294). Other studies have found cART to be protective against anal HPV and SIL (295, 296). However, given the Strategic Timing of AntiRetroviral Treatment

(START) study published in August 2015 showed net benefit in initiating cART in

HIV-positive adults with CD4+ T-cell counts >500 cells/mm3 compared to deferring until CD4+ T-cell counts fell to <350 cells/mm3 (297), the majority of the next generation of HIV patients with access to cART are unlikely to reach the low nadir

CD4+ T-cell counts that are known to be risk factors for anal cancer. How this impacts on the incidence of anal cancer remains to be seen, but it seems reasonable to hypothesise that the early initiation of cART will decrease anal cancer incidence (over years to decades) in effectively treated HIV-infected populations. START did not show an effect on HPV-related cancers but the data are not definitive due to small numbers

161

and limited follow up (there were two cases of incident anal cancer in the deferred cART arm compared to one case in the immediate arm).

In summary, not all anal HSIL are the same. Some are more likely to persist or progress, and host immune factors are likely to be an important determinant. Treatment of anal

HSIL should only be advocated for those at highest risk of persistence or progression, and these patients should be selected based on a rational understanding of the underlying biology. A robust evidence base needs to be established to determine the preferred therapeutic approach.

5.3.1 Addressing limitations of current work

In order to confirm the findings presented in this thesis and standardise clinical research in the field of anal cancer prevention, consensus endpoints are needed including how to combine anal cytology and histology in defining HSIL (270). Another controversial issue that needs to be addressed is defining HSIL spontaneous regression. Longer and larger studies are needed to differentiate partial or temporary regression that may be related to limitations of current screening tests, from clinically meaningful regression that is associated with a reduced risk of developing anal cancer. The natural history of anal SILs requires confirmation from prospective studies that are currently in progress such as SPANC in Sydney, Australia (245) and the observation arm of studies commencing recruitment such as ANCHOR in the United States (ClinicalTrials.gov identifier NCT02135419).

The CD25/CD134 and ICS assays which measure systemic HPV16-specific T-cells should be applied to prospective cohorts, and analysed for ability to predict HSIL

162

spontaneous regression. Perhaps they could also have a role in predicting SCC recurrence. In particular, relatively simple assays such as the CD25/CD134 assay could be implemented into routine diagnostic laboratories as immune-based biomarkers as part of a wider anal cancer screening strategy or algorithm. In keeping with the literature, further optimisation of these assays using HPV16 L1 and E2 peptides may increase test sensitivity for HSIL and/or spontaneous regression.

The immunofluorescence microscopy methods described in Chapter 4 need to be applied to normal and LSIL anal biopsies, perhaps in a case-controlled manner with the work already done on HSIL biopsies. This work should also be extended to prospectively acquired, longitudinal anal biopsies to compare the mucosal immunobiology of HSIL that regresses to those that persist or progress. Other possible comparisons include anal HPV16 DNA presence compared with absence, or high-risk

HPV presence compared with absence. Further studies in this field should aim to define and characterise the observed lymphoid aggregates, including more detailed phenotyping of the T-cell infiltrates and staining for other immune cell types. The development of an automated, reproducible cell enumeration method (e.g. using software such as CellProfiler) will go a long way towards making larger and prospective studies of this kind feasible (298).

5.3.2 Future work

Risk factors for those at highest risk of anal HSIL persistence or progression to cancer need to be established. For example, lesion size in the cervix can predict the risk of progression to cancer (299, 300), and whether this factor is important in anal HSIL is unknown. Likewise, genome-wide association studies have shown the influence of genetic determinants of the immune response in determining susceptibility to cervical 163

cancer (301) and HPV persistence (302) – these findings require study and confirmation in anal HPV, HSIL and cancer. The epidemiology and natural history of anal HSIL and cancer in other at-risk populations (e.g. transplant and other immunosuppressed patients, women with cervical HSIL) should also be studied and compared with HIV- infected and uninfected MSM. Systemic and mucosal HPV-specific T-cell responses can also be studied in HPV-associated cancers outside the urogenital tract e.g. head and neck squamous cell carcinomas.

For those at highest risk of developing anal cancer, randomised controlled trials of anal

HSIL treatment are mandatory. While standard approaches such as the various forms of ablation do require comparison with each other and with watchful waiting, the combination of ablation and immune-based treatments such as imiquimod may provide additional benefit, for example, by reducing recurrence after treatment. Another hypothesis is that the prophylactic HPV-L1 VLP vaccine may prevent new precancerous lesions in this population. Novel immune-based therapeutic approaches such as therapeutic vaccination (284, 303) and adoptive T-cell therapy (288) have also yet to be applied to anal HSIL.

In terms of underlying biology, the methods to detect and describe systemic and mucosal cellular immune responses in this thesis could be applied to larger and longitudinal natural history studies to elucidate in more detail the role of the immune response in controlling HPV-induced precancer. These same techniques can then be applied to HPV-specific treatment studies (prophylactic and therapeutic vaccines, immunomodulatory and adoptive T-cell therapies etc.). Much more detailed studies of the systemic immune response are possible with already established methods e.g. sub- phenotyping E6/E7-specific CD4+ T-cell responses to look for regulatory T-cells, non-

Th1 cytokines etc. A study correlating systemic E6/E7-specific T-cell responses with

164

E6/E7-specific serology i.e. B-cell responses would also currently be novel in the field of HPV immunology. Establishing HPV-specificity of anal mucosal T-cells is a technical challenge. One approach is to correlate this work with laser capture microdissection of adjacent HSIL lesions to detect the culprit HPV type, or perhaps a method for single-cell PCR on fixed tissue could be developed. These and other potential directions for future research arising from this thesis are summarised in Table

5.2.

165

Table 5.2 – Potential directions for future research arising from this thesis Research question Comments (e.g. potential study design, feasibility etc.)

Clinical Who is most at risk of progression from anal HSIL to cancer? Longitudinal, prospective cohorts. In progress. What factors predict anal HSIL progression/persistence vs. regression? Host (immune, genetic) as well as viral factors. Lesion size. Who will benefit most from treatment of anal HSIL? Randomised controlled trials. Novel immunomodulatory What is the best way to treat anal HSIL safely and effectively? approaches (therapeutic vaccination, immune-checkpoint inhibitors, adoptive T-cell therapy). Combined treatments. Assay Is the CD25/CD134 assay for HPV16-E6 CD4+ T-cells predictive of Partially in progress with stored frozen samples from SPANC. development anal HSIL regression? Can it be used in a routine diagnostic laboratory Requires more samples at later time points to be taken. as part of a cost-effective anal cancer prevention programme? Would extending the assays to other HPV antigens (e.g. HPV16-L1, Cost may be a constraint. HPV16-E2, other high-risk HPV antigens) be clinically meaningful? How do the assays perform on frozen PBMCs? Important for application in clinical trials. Can the assays be extended to measure CD4+ T-cell subsets (e.g. Important for better understanding of underlying biology. regulatory T-cells) and non-Th1 cytokines? Can these assays predict response to treatment? Apply to randomised controlled trials of anal HSIL treatment. Biology Are the lymphoid aggregates described in Chapter 4 HPV-specific? Descriptive cross-sectional and longitudinal studies. Technical What are they doing there? How did they get there? Which T-cell issues to be addressed include demonstrating antigen- subsets are involved? What other (non-T, non-lymphocyte) immune specificity of lymphocytes in fixed tissue. cells are important in this mucosal response? Can measuring systemic HPV-specific T-cell responses predict an Will need larger numbers of anal biopsies imaged and effective anal mucosal immune response against HPV-induced lesions? analysed and related to outcome data. Could having anal HPV and HSIL increase the risk of acquiring HIV Descriptive in vitro work using ex vivo anal biopsy samples. due to CD4+ T-cell enriched lymphoid aggregates in the anal stroma acting as target cells? 165

1. Silverberg MJ, Lau B, Justice AC, Engels E, Gill MJ, Goedert JJ, et al. Risk of anal cancer in HIV-infected and HIV-uninfected individuals in North America. Clin Infect Dis. 2012;54(7):1026-34. 2. Wu X, Watson M, Wilson R, Saraiya M, Cleveland JL, Markowitz L. Human Papillomavirus-Associated Cancers - United States, 2004-2008. MMWR. 2012;61(15):258-61. 3. Darragh TM, Colgan TJ, Cox JT, Heller DS, Henry MR, Luff RD, et al. The Lower Anogenital Squamous Terminology Standardization Project for HPV-Associated Lesions: background and consensus recommendations from the College of American Pathologists and the American Society for Colposcopy and Cervical Pathology. J Low Genit Tract Dis. 2012;16(3):205-42. 4. Frazer IH. Prevention of cervical cancer through papillomavirus vaccination. Nat Rev Immunol. 2004;4(1):46-54. 5. Modis Y, Trus BL, Harrison SC. Atomic model of the papillomavirus capsid. EMBO J. 2002;21(18):4754-62. 6. Solomon D, Davey D, Kurman R, Moriarty A, O'Connor D, Prey M, et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. JAMA. 2002;287(16):2114-9. 7. National Health and Medical Research Council. Screening to prevent cervical cancer: guidelines for the management of asymptomatic women with screen-detected abnormalities. Australia 2006. Available from: http://www.nhmrc.gov.au/guidelines/publications/wh39. Accessed 11th June 2012. 8. Roberts JM, Ekman D. The reporting of anal cytology and histology samples: establishing terminology and criteria. Sex Health. 2012;9(6):562-7. 9. Doorbar J, Quint W, Banks L, Bravo IG, Stoler M, Broker TR, et al. The biology and life-cycle of human papillomaviruses. Vaccine. 2012;30 Suppl 5:F55-70. 10. Wilson VG, West M, Woytek K, Rangasamy D. Papillomavirus E1 proteins: form, function, and features. Virus Genes. 2002;24(3):275-90. 11. Roberts S, Ashmole I, Johnson GD, Kreider JW, Gallimore PH. Cutaneous and mucosal human papillomavirus E4 proteins form intermediate filament-like structures in epithelial cells. Virology. 1993;197(1):176-87. 12. Hwang ES, Nottoli T, Dimaio D. The HPV16 E5 protein: expression, detection, and stable complex formation with transmembrane proteins in COS cells. Virology. 1995;211(1):227-33. 13. Scheffner M, Werness BA, Huibregtse JM, Levine AJ, Howley PM. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell. 1990;63(6):1129-36. 14. Scheffner M, Huibregtse JM, Vierstra RD, Howley PM. The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of p53. Cell. 1993;75(3):495-505. 15. Crook T, Vousden KH, Tidy JA. Degradation of p53 can be targeted by HPV E6 sequences distinct from those required for p53 binding and trans-activation. Cell. 1991;67(3):547-56. 16. Klingelhutz AJ, Foster SA, McDougall JK. Telomerase activation by the E6 gene product of human papillomavirus type 16. Nature. 1996;380(6569):79-82. 17. zur Hausen H. Papillomaviruses and cancer: from basic studies to clinical application. Nat Rev Cancer. 2002;2(5):342-50.

167

18. Veldman T, Horikawa I, Barrett JC, Schlegel R. Transcriptional activation of the telomerase hTERT gene by human papillomavirus type 16 E6 oncoprotein. J Virol. 2001;75(9):4467-72. 19. Oda H, Kumar S, Howley PM. Regulation of the Src family tyrosine kinase Blk through E6AP-mediated ubiquitination. Proc Natl Acad Sci U S A. 1999;96(17):9557- 62. 20. Dyson N, Howley PM, Munger K, Harlow E. The human papilloma virus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science. 1989;243(4893):934-7. 21. Zerfass K, Schulze A, Spitkovsky D, Friedman V, Henglein B, Jansen-Durr P. Sequential activation of cyclin E and cyclin A gene expression by human papillomavirus type 16 E7 through sequences necessary for transformation. J Virol. 1995;69(10):6389-99. 22. de Villiers EM. Cross-roads in the classification of papillomaviruses. Virology. 2013;445(1-2):2-10. 23. IARC. Human Papillomaviruses. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. 2007;90. 24. Doorbar J. The papillomavirus life cycle. J Clin Virol. 2005;32 Suppl 1:S7-15. 25. Munoz N, Bosch FX, de Sanjose S, Herrero R, Castellsague X, Shah KV, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med. 2003;348(6):518-27. 26. IARC. Part B: Biological Agents. A Review of Human Carcinogens. 2012;100:255-313. 27. Schiffman M, Clifford G, Buonaguro FM. Classification of weakly carcinogenic human papillomavirus types: addressing the limits of epidemiology at the borderline. Infectious agents and cancer. 2009;4:8. 28. Chadha NK, James A. Adjuvant antiviral therapy for recurrent respiratory papillomatosis. Cochrane Database Syst Rev. 2012;12:CD005053. 29. Abramowitz L, Jacquard AC, Jaroud F, Haesebaert J, Siproudhis L, Pradat P, et al. Human papillomavirus genotype distribution in anal cancer in France: The EDiTH V study. Int J Cancer. 2010. 30. Hoots BE, Palefsky JM, Pimenta JM, Smith JS. Human papillomavirus type distribution in anal cancer and anal intraepithelial lesions. Int J Cancer. 2009;124(10):2375-83. 31. Grulich AE, Jin F, Conway EL, Stein AN, Hocking J. Cancers attributable to human papillomavirus infection. Sex Health. 2010;7(3):244-52. 32. D'Souza G, Kreimer AR, Viscidi R, Pawlita M, Fakhry C, Koch WM, et al. Case-control study of human papillomavirus and oropharyngeal cancer. N Engl J Med. 2007;356(19):1944-56. 33. Ramoz N, Rueda LA, Bouadjar B, Montoya LS, Orth G, Favre M. Mutations in two adjacent novel genes are associated with epidermodysplasia verruciformis. Nat Genet. 2002;32(4):579-81. 34. Lazarczyk M, Pons C, Mendoza JA, Cassonnet P, Jacob Y, Favre M. Regulation of cellular zinc balance as a potential mechanism of EVER-mediated protection against pathogenesis by cutaneous oncogenic human papillomaviruses. J Exp Med. 2008;205(1):35-42. 35. Krill CE, Jr., Smith HD, Mauer AM. Chronic Idiopathic Granulocytopenia. N Engl J Med. 1964;270:973-9. 36. Zuelzer WW. "Myelokathexis"--a New Form of Chronic Granulocytopenia. Report of a Case. N Engl J Med. 1964;270:699-704.

168

37. Hernandez PA, Gorlin RJ, Lukens JN, Taniuchi S, Bohinjec J, Francois F, et al. Mutations in the chemokine receptor gene CXCR4 are associated with WHIM syndrome, a combined immunodeficiency disease. Nat Genet. 2003;34(1):70-4. 38. Leiding JW, Holland SM. Warts and all: human papillomavirus in primary immunodeficiencies. J Allergy Clin Immunol. 2012;130(5):1030-48. 39. Barre-Sinoussi F, Chermann J, Rey F, Nugeyre M, Chamaret S, Gruest J, et al. Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS). Science. 1983;220(4599):868-71. 40. Gallo R, Salahuddin S, Popovic M, Shearer G, Kaplan M, Haynes B, et al. Frequent detection and isolation of cytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS. Science. 1984;224(4648):500-3. 41. Dalgleish AG, Beverley PCL, Clapham PR, Crawford DH, Greaves MF, Weiss RA. The CD4 (T4) antigen is an essential component of the receptor for the AIDS retrovirus. Nature. 1984;312(5996):763-7. 42. Centers for Disease Control and Prevention (CDC). Revised surveillance case definition for HIV infection--United States, 2014. MMWR Recommendations and reports : Morbidity and mortality weekly report Recommendations and reports / Centers for Disease Control. 2014;63(RR-03):1-10. 43. Pantaleo G, Graziosi C, Fauci AS. New concepts in the immunopathogenesis of human immunodeficiency virus infection. N Engl J Med. 1993;328(5):327-35. 44. Autran B, Carcelain G, Li TS, Blanc C, Mathez D, Tubiana R, et al. Positive Effects of Combined Antiretroviral Therapy on CD4+ T Cell Homeostasis and Function in Advanced HIV Disease. Science. 1997;277(5322):112-6. 45. Grulich AE, van Leeuwen MT, Falster MO, Vajdic CM. Incidence of cancers in people with HIV/AIDS compared with immunosuppressed transplant recipients: a meta- analysis. Lancet. 2007;370(9581):59-67. 46. Machalek DA, Poynten M, Jin F, Fairley CK, Farnsworth A, Garland SM, et al. Anal human papillomavirus infection and associated neoplastic lesions in men who have sex with men: a systematic review and meta-analysis. Lancet Oncol. 2012;13(5):487-500. 47. Gustafsson L, Ponten J, Bergstrom R, Adami HO. International incidence rates of invasive cervical cancer before cytological screening. Int J Cancer. 1997;71(2):159- 65. 48. Fauci AS. Immunopathogenic Mechanisms of HIV Infection. Annals of Internal Medicine. 1996;124(7):654. 49. Iwasaki A. Antiviral immune responses in the genital tract: clues for vaccines. Nat Rev Immunol. 2010;10(10):699-711. 50. Kreuter A, Skrygan M, Gambichler T, Brockmeyer NH, Stucker M, Herzler C, et al. Human papillomavirus-associated induction of human beta-defensins in anal intraepithelial neoplasia. Br J Dermatol. 2009;160(6):1197-205. 51. Stanley M. HPV - immune response to infection and vaccination. Infectious agents and cancer. 2010;5:19. 52. Filippova M, Song H, Connolly JL, Dermody TS, Duerksen-Hughes PJ. The human papillomavirus 16 E6 protein binds to tumor necrosis factor (TNF) R1 and protects cells from TNF-induced apoptosis. J Biol Chem. 2002;277(24):21730-9. 53. Lee SJ, Cho YS, Cho MC, Shim JH, Lee KA, Ko KK, et al. Both E6 and E7 oncoproteins of human papillomavirus 16 inhibit IL-18-induced IFN-gamma production in human peripheral blood mononuclear and NK cells. J Immunol. 2001;167(1):497- 504.

169

54. Hasan UA, Bates E, Takeshita F, Biliato A, Accardi R, Bouvard V, et al. TLR9 expression and function is abolished by the cervical cancer-associated human papillomavirus type 16. J Immunol. 2007;178(5):3186-97. 55. Karim R, Meyers C, Backendorf C, Ludigs K, Offringa R, van Ommen GJ, et al. Human papillomavirus deregulates the response of a cellular network comprising of chemotactic and proinflammatory genes. PLoS One. 2011;6(3):e17848. 56. Gravitt PE. The known unknowns of HPV natural history. J Clin Invest. 2011;121(12):4593-9. 57. Carter JJ, Koutsky LA, Hughes JP, Lee SK, Kuypers J, Kiviat N, et al. Comparison of human papillomavirus types 16, 18, and 6 capsid antibody responses following incident infection. J Infect Dis. 2000;181(6):1911-9. 58. Lu B, Viscidi RP, Wu Y, Nyitray AG, Villa LL, Lazcano-Ponce E, et al. Seroprevalence of human papillomavirus (HPV) type 6 and 16 vary by anatomic site of HPV infection in men. Cancer Epidemiol Biomarkers Prev. 2012;21(9):1542-6. 59. Viscidi RP, Ahdieh-Grant L, Clayman B, Fox K, Massad LS, Cu-Uvin S, et al. Serum immunoglobulin G response to human papillomavirus type 16 virus-like particles in human immunodeficiency virus (HIV)-positive and risk-matched HIV- negative women. J Infect Dis. 2003;187(2):194-205. 60. Namujju PB, Waterboer T, Banura C, Muwonge R, Mbidde EK, Byaruhanga R, et al. Risk of seropositivity to multiple oncogenic human papillomavirus types among human immunodeficiency virus-positive and -negative Ugandan women. J Gen Virol. 2011;92(Pt 12):2776-83. 61. Silverberg MJ, Schneider MF, Silver B, Anastos KM, Burk RD, Minkoff H, et al. Serological detection of human papillomavirus type 16 infection in human immunodeficiency virus (HIV)-positive and high-risk HIV-negative women. Clin Vaccine Immunol. 2006;13(4):511-9. 62. Onda T, Carter JJ, Koutsky LA, Hughes JP, Lee SK, Kuypers J, et al. Characterization of IgA response among women with incident HPV 16 infection. Virology. 2003;312(1):213-21. 63. Petter A, Heim K, Guger M, Ciresa-Ko Nig A, Christensen N, Sarcletti M, et al. Specific serum IgG, IgM and IgA antibodies to human papillomavirus types 6, 11, 16, 18 and 31 virus-like particles in human immunodeficiency virus-seropositive women. J Gen Virol. 2000;81(Pt 3):701-8. 64. Viscidi RP, Ahdieh-Grant L, Schneider MF, Clayman B, Massad LS, Anastos KM, et al. Serum immunoglobulin A response to human papillomavirus type 16 virus- like particles in human immunodeficiency virus (HIV)-positive and high-risk HIV- negative women. J Infect Dis. 2003;188(12):1834-44. 65. Marais DJ, Vardas E, Ramjee G, Allan B, Kay P, Rose RC, et al. The impact of human immunodeficiency virus type 1 status on human papillomavirus (HPV) prevalence and HPV antibodies in serum and cervical secretions. J Infect Dis. 2000;182(4):1239-42. 66. Marais DJ, Carrara H, Ramjee G, Kay P, Williamson AL. HIV-1 seroconversion promotes rapid changes in cervical human papillomavirus (HPV) prevalence and HPV- 16 antibodies in female sex workers. J Med Virol. 2009;81(2):203-10. 67. Jain S, Moore RA, Anderson DM, Gough GW, Stanley MA. Cell-mediated immune responses to COPV early proteins. Virology. 2006;356(1-2):23-34. 68. Nicholls PK, Moore PF, Anderson DM, Moore RA, Parry NR, Gough GW, et al. Regression of canine oral papillomas is associated with infiltration of CD4+ and CD8+ lymphocytes. Virology. 2001;283(1):31-9.

170

69. Coleman N, Birley HD, Renton AM, Hanna NF, Ryait BK, Byrne M, et al. Immunological events in regressing genital warts. Am J Clin Pathol. 1994;102(6):768- 74. 70. Frazer IH. Interaction of human papillomaviruses with the host immune system: a well evolved relationship. Virology. 2009;384(2):410-4. 71. Nakagawa M, Stites DP, Palefsky JM, Kneass Z, Moscicki AB. CD4-positive and CD8-positive cytotoxic T lymphocytes contribute to human papillomavirus type 16 E6 and E7 responses. Clin Diagn Lab Immunol. 1999;6(4):494-8. 72. Nakagawa M, Stites DP, Farhat S, Sisler JR, Moss B, Kong F, et al. Cytotoxic T lymphocyte responses to E6 and E7 proteins of human papillomavirus type 16: relationship to cervical intraepithelial neoplasia. J Infect Dis. 1997;175(4):927-31. 73. Nakagawa M, Stites DP, Patel S, Farhat S, Scott M, Hills NK, et al. Persistence of human papillomavirus type 16 infection is associated with lack of cytotoxic T lymphocyte response to the E6 antigens. J Infect Dis. 2000;182(2):595-8. 74. Woo YL, van den Hende M, Sterling JC, Coleman N, Crawford RA, Kwappenberg KM, et al. A prospective study on the natural course of low-grade squamous intraepithelial lesions and the presence of HPV16 E2-, E6- and E7-specific T- cell responses. Int J Cancer. 2010;126(1):133-41. 75. de Jong A, van Poelgeest MI, van der Hulst JM, Drijfhout JW, Fleuren GJ, Melief CJ, et al. Human papillomavirus type 16-positive cervical cancer is associated with impaired CD4+ T-cell immunity against early antigens E2 and E6. Cancer Res. 2004;64(15):5449-55. 76. Welters MJ, van der Logt P, van den Eeden SJ, Kwappenberg KM, Drijfhout JW, Fleuren GJ, et al. Detection of human papillomavirus type 18 E6 and E7-specific CD4+ T-helper 1 immunity in relation to health versus disease. Int J Cancer. 2006;118(4):950-6. 77. van Poelgeest MI, van Seters M, van Beurden M, Kwappenberg KM, Heijmans- Antonissen C, Drijfhout JW, et al. Detection of human papillomavirus (HPV) 16- specific CD4+ T-cell immunity in patients with persistent HPV16-induced vulvar intraepithelial neoplasia in relation to clinical impact of imiquimod treatment. Clin Cancer Res. 2005;11(14):5273-80. 78. Seresini S, Origoni M, Lillo F, Caputo L, Paganoni AM, Vantini S, et al. IFN- gamma produced by human papilloma virus-18 E6-specific CD4+ T cells predicts the clinical outcome after surgery in patients with high-grade cervical lesions. J Immunol. 2007;179(10):7176-83. 79. de Jong A, van der Burg SH, Kwappenberg KM, van der Hulst JM, Franken KL, Geluk A, et al. Frequent detection of human papillomavirus 16 E2-specific T-helper immunity in healthy subjects. Cancer Res. 2002;62(2):472-9. 80. Monnier-Benoit S, Mauny F, Riethmuller D, Guerrini JS, Capilna M, Felix S, et al. Immunohistochemical analysis of CD4+ and CD8+ T-cell subsets in high risk human papillomavirus-associated pre-malignant and malignant lesions of the uterine cervix. Gynecol Oncol. 2006;102(1):22-31. 81. Woo YL, Sterling J, Damay I, Coleman N, Crawford R, van der Burg SH, et al. Characterising the local immune responses in cervical intraepithelial neoplasia: a cross- sectional and longitudinal analysis. BJOG. 2008;115(13):1616-21; discussion 21-2. 82. Trimble CL, Peng S, Thoburn C, Kos F, Wu TC. Naturally occurring systemic immune responses to HPV antigens do not predict regression of CIN2/3. Cancer Immunol Immunother. 2010;59(5):799-803. 83. Peng S, Trimble C, Wu L, Pardoll D, Roden R, Hung CF, et al. HLA-DQB1*02- restricted HPV-16 E7 peptide-specific CD4+ T-cell immune responses correlate with

171

regression of HPV-16-associated high-grade squamous intraepithelial lesions. Clin Cancer Res. 2007;13(8):2479-87. 84. Visser J, Nijman HW, Hoogenboom BN, Jager P, van Baarle D, Schuuring E, et al. Frequencies and role of regulatory T cells in patients with (pre)malignant cervical neoplasia. Clin Exp Immunol. 2007;150(2):199-209. 85. Molling JW, de Gruijl TD, Glim J, Moreno M, Rozendaal L, Meijer CJ, et al. CD4(+)CD25hi regulatory T-cell frequency correlates with persistence of human papillomavirus type 16 and T helper cell responses in patients with cervical intraepithelial neoplasia. Int J Cancer. 2007;121(8):1749-55. 86. van der Burg SH, Piersma SJ, de Jong A, van der Hulst JM, Kwappenberg KM, van den Hende M, et al. Association of cervical cancer with the presence of CD4+ regulatory T cells specific for human papillomavirus antigens. Proc Natl Acad Sci U S A. 2007;104(29):12087-92. 87. Guimaraes MD, Grinsztejn B, Melo VH, Rocha GM, Campos LN, Pilotto JH, et al. Anal HPV prevalence and associated factors among HIV-seropositive men under antiretroviral treatment in Brazil. J Acquir Immune Defic Syndr. 2011;57 Suppl 3:S217- 24. 88. Goncalves MA, Randi G, Arslan A, Villa LL, Burattini MN, Franceschi S, et al. HPV type infection in different anogenital sites among HIV-positive Brazilian women. Infectious agents and cancer. 2008;3:5. 89. Palefsky JM, Holly EA, Ralston ML, Da Costa M, Greenblatt RM. Prevalence and risk factors for anal human papillomavirus infection in human immunodeficiency virus (HIV)-positive and high-risk HIV-negative women. J Infect Dis. 2001;183(3):383- 91. 90. Piketty C, Darragh TM, Da Costa M, Bruneval P, Heard I, Kazatchkine MD, et al. High prevalence of anal human papillomavirus infection and anal cancer precursors among HIV-infected persons in the absence of anal intercourse. Ann Intern Med. 2003;138(6):453-9. 91. Baranoski AS, Tandon R, Weinberg J, Huang FF, Stier EA. Risk factors for abnormal anal cytology over time in HIV-infected women. Am J Obstet Gynecol. 2012;207(2):107 e1-8. 92. Hessol NA, Holly EA, Efird JT, Minkoff H, Schowalter K, Darragh TM, et al. Anal intraepithelial neoplasia in a multisite study of HIV-infected and high-risk HIV- uninfected women. AIDS. 2009;23(1):59-70. 93. Crum-Cianflone NF, Hullsiek KH, Marconi VC, Ganesan A, Weintrob A, Barthel RV, et al. Anal cancers among HIV-infected persons: HAART is not slowing rising incidence. AIDS. 2010;24(4):535-43. 94. de Pokomandy A, Rouleau D, Ghattas G, Trottier H, Vezina S, Cote P, et al. HAART and Progression to High-Grade Anal Intraepithelial Neoplasia in Men Who Have Sex with Men and Are Infected with HIV. Clin Infect Dis. 2011;52(9):1174-81. 95. Kobayashi A, Greenblatt RM, Anastos K, Minkoff H, Massad LS, Young M, et al. Functional attributes of mucosal immunity in cervical intraepithelial neoplasia and effects of HIV infection. Cancer Res. 2004;64(18):6766-74. 96. Kobayashi A, Darragh T, Herndier B, Anastos K, Minkoff H, Cohen M, et al. Lymphoid follicles are generated in high-grade cervical dysplasia and have differing characteristics depending on HIV status. Am J Pathol. 2002;160(1):151-64. 97. Kojic EM, Cu-Uvin S, Conley L, Bush T, Onyekwuluje J, Swan DC, et al. Human papillomavirus infection and cytologic abnormalities of the anus and cervix among HIV-infected women in the study to understand the natural history of HIV/AIDS in the era of effective therapy (the SUN study). Sex Transm Dis. 2011;38(4):253-9.

172

98. de Pokomandy A, Rouleau D, Ghattas G, Vezina S, Cote P, Macleod J, et al. Prevalence, clearance, and incidence of anal human papillomavirus infection in HIV- infected men: the HIPVIRG cohort study. J Infect Dis. 2009;199(7):965-73. 99. Vajdic CM, van Leeuwen MT, Jin F, Prestage G, Medley G, Hillman RJ, et al. Anal human papillomavirus genotype diversity and co-infection in a community-based sample of homosexual men. Sex Transm Infect. 2009;85(5):330-5. 100. Xi LF, Critchlow CW, Wheeler CM, Koutsky LA, Galloway DA, Kuypers J, et al. Risk of anal carcinoma in situ in relation to human papillomavirus type 16 variants. Cancer Res. 1998;58(17):3839-44. 101. Poynten IM, Grulich AE, Templeton DJ. Sexually transmitted infections in older populations. Curr Opin Infect Dis. 2013;26(1):80-5. 102. Nielsen A, Munk C, Kjaer SK. Trends in incidence of anal cancer and high- grade anal intraepithelial neoplasia in Denmark, 1978-2008. Int J Cancer. 2012;130(5):1168-73. 103. Jin F, Stein AN, Conway EL, Regan DG, Law M, Brotherton JM, et al. Trends in anal cancer in Australia, 1982-2005. Vaccine. 2011;29(12):2322-7. 104. Frisch M, Melbye M, Moller H. Trends in incidence of anal cancer in Denmark. BMJ. 1993;306(6875):419-22. 105. Frisch M, Biggar RJ, Goedert JJ. Human papillomavirus-associated cancers in patients with human immunodeficiency virus infection and acquired immunodeficiency syndrome. J Natl Cancer Inst. 2000;92(18):1500-10. 106. van Leeuwen MT, Vajdic CM, Middleton MG, McDonald AM, Law M, Kaldor JM, et al. Continuing declines in some but not all HIV-associated cancers in Australia after widespread use of antiretroviral therapy. AIDS. 2009;23(16):2183-90. 107. D'Souza G, Wiley DJ, Li X, Chmiel JS, Margolick JB, Cranston RD, et al. Incidence and epidemiology of anal cancer in the multicenter AIDS cohort study. J Acquir Immune Defic Syndr. 2008;48(4):491-9. 108. Saleem AM, Paulus JK, Shapter AP, Baxter NN, Roberts PL, Ricciardi R. Risk of anal cancer in a cohort with human papillomavirus-related gynecologic neoplasm. Obstet Gynecol. 2011;117(3):643-9. 109. Piketty C, Selinger-Leneman H, Grabar S, Duvivier C, Bonmarchand M, Abramowitz L, et al. Marked increase in the incidence of invasive anal cancer among HIV-infected patients despite treatment with combination antiretroviral therapy. AIDS. 2008;22(10):1203-11. 110. Seaberg EC, Wiley D, Martinez-Maza O, Chmiel JS, Kingsley L, Tang Y, et al. Cancer incidence in the multicenter AIDS Cohort Study before and during the HAART era: 1984 to 2007. Cancer. 2010;116(23):5507-16. 111. Darwich L, Canadas MP, Videla S, Coll J, Molina-Lopez RA, Sirera G, et al. Prevalence, clearance, and incidence of human papillomavirus type-specific infection at the anal and penile site of HIV-infected men. Sex Transm Dis. 2013;40(8):611-8. 112. Legarth R, Helleberg M, Kronborg G, Larsen CS, Pedersen G, Pedersen C, et al. Anal carcinoma in HIV-infected patients in the period 1995-2009: a Danish nationwide cohort study. Scandinavian journal of infectious diseases. 2013;45(6):453-9. 113. Hou JY, Smotkin D, Grossberg R, Suhrland M, Levine R, Smith HO, et al. High prevalence of high grade anal intraepithelial neoplasia in HIV-infected women screened for anal cancer. J Acquir Immune Defic Syndr. 2012;60(2):169-72. 114. Durante AJ, Williams AB, Da Costa M, Darragh TM, Khoshnood K, Palefsky JM. Incidence of anal cytological abnormalities in a cohort of human immunodeficiency virus-infected women. Cancer Epidemiol Biomarkers Prev. 2003;12(7):638-42.

173

115. Watson AJ, Smith BB, Whitehead MR, Sykes PH, Frizelle FA. Malignant progression of anal intra-epithelial neoplasia. ANZ journal of surgery. 2006;76(8):715- 7. 116. Scholefield JH, Castle MT, Watson NF. Malignant transformation of high-grade anal intraepithelial neoplasia. The British journal of surgery. 2005;92(9):1133-6. 117. McCredie MR, Sharples KJ, Paul C, Baranyai J, Medley G, Jones RW, et al. Natural history of cervical neoplasia and risk of invasive cancer in women with cervical intraepithelial neoplasia 3: a retrospective cohort study. Lancet Oncol. 2008;9(5):425- 34. 118. Ostor AG. Natural history of cervical intraepithelial neoplasia: a critical review. Int J Gynecol Pathol. 1993;12(2):186-92. 119. Frazer IH, Leggatt GR, Mattarollo SR. Prevention and treatment of papillomavirus-related cancers through immunization. Annu Rev Immunol. 2011;29:111-38. 120. Raff AB, Woodham AW, Raff LM, Skeate JG, Yan L, Da Silva DM, et al. The evolving field of human papillomavirus receptor research: a review of binding and entry. J Virol. 2013;87(11):6062-72. 121. Thomas MK, Pitot HC, Liem A, Lambert PF. Dominant role of HPV16 E7 in anal carcinogenesis. Virology. 2011;421(2):114-8. 122. Fera D, Schultz DC, Hodawadekar S, Reichman M, Donover PS, Melvin J, et al. Identification and Characterization of Small Molecule Antagonists of pRb Inactivation by Viral Oncoproteins. Chemistry & biology. 2012;19(4):518-28. 123. Burgers WA, Blanchon L, Pradhan S, de Launoit Y, Kouzarides T, Fuks F. Viral oncoproteins target the DNA methyltransferases. Oncogene. 2007;26(11):1650-5. 124. Simpson S, Turner R. Four decades of anal cancer in Tasmania, Australia: what do the case data tell us? Sex Health. 2012;9(3):213-9. 125. Simpson JA, Scholefield JH. Diagnosis and management of anal intraepithelial neoplasia and anal cancer. BMJ. 2011;343:d6818. 126. Jay N, Berry JM, Hogeboom CJ, Holly EA, Darragh TM, Palefsky JM. Colposcopic appearance of anal squamous intraepithelial lesions: relationship to histopathology. Dis Colon Rectum. 1997;40(8):919-28. 127. Darragh TM, Winkler B. Anal cancer and cervical cancer screening: Key differences. Cancer Cytopathology. 2011;119(1):5-19. 128. Darragh TM, Jay N, Tupkelewicz BA, Hogeboom CJ, Holly EA, Palefsky JM. Comparison of conventional cytologic smears and ThinPrep preparations from the anal canal. Acta Cytol. 1997;41(4):1167-70. 129. Cachay ER, Agmas W, Mathews WC. Relative accuracy of cervical and anal cytology for detection of high grade lesions by colposcope guided biopsy: a cut-point meta-analytic comparison. PLoS One. 2012;7(7):e38956. 130. WHO. Human papillomavirus laboratory manual. First ed2009. 131. Goldstone RN, Goldstone AB, Russ J, Goldstone SE. Long-term follow-up of infrared coagulator ablation of anal high-grade dysplasia in men who have sex with men. Dis Colon Rectum. 2011;54(10):1284-92. 132. Marks DK, Goldstone SE. Electrocautery ablation of high-grade anal squamous intraepithelial lesions in HIV-negative and HIV-positive men who have sex with men. J Acquir Immune Defic Syndr. 2012;59(3):259-65. 133. Steele SR, Varma MG, Melton GB, Ross HM, Rafferty JF, Buie WD, et al. Practice parameters for anal squamous neoplasms. Dis Colon Rectum. 2012;55(7):735- 49.

174

134. Wright TC, Jr., Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D, et al. 2006 consensus guidelines for the management of women with cervical intraepithelial neoplasia or adenocarcinoma in situ. Am J Obstet Gynecol. 2007;197(4):340-5. 135. Macaya A, Munoz-Santos C, Balaguer A, Barbera MJ. Interventions for anal canal intraepithelial neoplasia. Cochrane Database Syst Rev. 2012;12:CD009244. 136. Fox PA, Nathan M, Francis N, Singh N, Weir J, Dixon G, et al. A double-blind, randomized controlled trial of the use of imiquimod cream for the treatment of anal canal high-grade anal intraepithelial neoplasia in HIV-positive MSM on HAART, with long-term follow-up data including the use of open-label imiquimod. AIDS. 2010;24(15):2331-5. 137. Glynne-Jones R, Northover JM, Cervantes A. Anal cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2010;21 Suppl 5:v87-92. 138. Seo Y, Kinsella MT, Reynolds HL, Chipman G, Remick SC, Kinsella TJ. Outcomes of chemoradiotherapy with 5-Fluorouracil and mitomycin C for anal cancer in immunocompetent versus immunodeficient patients. Int J Radiat Oncol Biol Phys. 2009;75(1):143-9. 139. Fraunholz I, Rabeneck D, Gerstein J, Jack K, Haberl A, Weiss C, et al. Concurrent chemoradiotherapy with 5-fluorouracil and mitomycin C for anal carcinoma: are there differences between HIV-positive and HIV-negative patients in the era of highly active antiretroviral therapy? Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology. 2011;98(1):99-104. 140. Czoski-Murray C, Karnon J, Jones R, Smith K, Kinghorn G. Cost-effectiveness of screening high-risk HIV-positive men who have sex with men (MSM) and HIV- positive women for anal cancer. Health Technol Assess. 2010;14(53):iii-iv, ix-x, 1-101. 141. Goldie SJ, Kuntz KM, Weinstein MC, Freedberg KA, Welton ML, Palefsky JM. The clinical effectiveness and cost-effectiveness of screening for anal squamous intraepithelial lesions in homosexual and bisexual HIV-positive men. JAMA. 1999;281(19):1822-9. 142. Centers for Disease Control and Prevention. Guidelines for Prevention and Treatment of Opportunistic Infections in HIV-Infected Adults and Adolescents. MMWR. 2009;58(No. RR-4):68-75. 143. European AIDS Clinical Society. EACS Guidelines Version 6. November 2012. Available from: http://www.europeanaidsclinicalsociety.org/images/stories/EACS- Pdf/EacsGuidelines-v6.1-2edition.pdf. Accessed 28th January 2013. 144. New York State Department of Health AIDS Institute. Anal dysplasia and cancer. New York 2007. Available from: http://www.hivguidelines.org/wp- content/uploads/2009/05/NEOPLACOMPLIC.pdf. Accessed 16th May 2012. 145. Chiao EY, Giordano TP, Palefsky JM, Tyring S, El Serag H. Screening HIV- infected individuals for anal cancer precursor lesions: a systematic review. Clin Infect Dis. 2006;43(2):223-33. 146. Joura EA, Giuliano AR, Iversen OE, Bouchard C, Mao C, Mehlsen J, et al. A 9- valent HPV vaccine against infection and intraepithelial neoplasia in women. N Engl J Med. 2015;372(8):711-23. 147. Zhou J, Sun XY, Stenzel DJ, Frazer IH. Expression of vaccinia recombinant HPV 16 L1 and L2 ORF proteins in epithelial cells is sufficient for assembly of HPV virion-like particles. Virology. 1991;185(1):251-7.

175

148. Palefsky JM, Giuliano AR, Goldstone S, Moreira ED, Aranda C, Jessen H, et al. HPV Vaccine against Anal HPV Infection and Anal Intraepithelial Neoplasia. N Engl J Med. 2011;365(17):1576-85. 149. Kreimer AR, Gonzalez P, Katki HA, Porras C, Schiffman M, Rodriguez AC, et al. Efficacy of a bivalent HPV 16/18 vaccine against anal HPV 16/18 infection among young women: a nested analysis within the Costa Rica Vaccine Trial. Lancet Oncol. 2011;12(9):862-70. 150. Wilkin T, Lee JY, Lensing SY, Stier EA, Goldstone SE, Berry JM, et al. Safety and immunogenicity of the quadrivalent human papillomavirus vaccine in HIV-1- infected men. J Infect Dis. 2010;202(8):1246-53. 151. Kojic EM, Kang M, Cespedes MS, Umbleja T, Godfrey C, Allen RT, et al. Immunogenicity and safety of the quadrivalent human papillomavirus vaccine in HIV- 1-infected women. Clin Infect Dis. 2014;59(1):127-35. 152. CDC. Recommended adult immunization schedule—United States, 2012. MMWR. 2012;61(4). 153. Kenter GG, Welters MJ, Valentijn AR, Lowik MJ, Berends-van der Meer DM, Vloon AP, et al. Vaccination against HPV-16 oncoproteins for vulvar intraepithelial neoplasia. N Engl J Med. 2009;361(19):1838-47. 154. Welters MJ, Kenter GG, de Vos van Steenwijk PJ, Lowik MJ, Berends-van der Meer DM, Essahsah F, et al. Success or failure of vaccination for HPV16-positive vulvar lesions correlates with kinetics and phenotype of induced T-cell responses. Proc Natl Acad Sci U S A. 2010;107(26):11895-9. 155. Strebhardt K, Ullrich A. Paul Ehrlich's magic bullet concept: 100 years of progress. Nat Rev Cancer. 2008;8(6):473-80. 156. Burnet M. Cancer; a biological approach. I. The processes of control. Br Med J. 1957;1(5022):779-86. 157. Thomas L. On immunosurveillance in human cancer. The Yale journal of biology and medicine. 1982;55(3-4):329-33. 158. Challis GB, Stam HJ. The spontaneous regression of cancer. A review of cases from 1900 to 1987. Acta oncologica. 1990;29(5):545-50. 159. Darnell RB, Posner JB. Paraneoplastic syndromes involving the nervous system. N Engl J Med. 2003;349(16):1543-54. 160. Dalmau J, Rosenfeld MR. Paraneoplastic syndromes of the CNS. The Lancet Neurology. 2008;7(4):327-40. 161. Lennon VA, Kryzer TJ, Griesmann GE, O'Suilleabhain PE, Windebank AJ, Woppmann A, et al. Calcium-Channel Antibodies in the Lambert–Eaton Syndrome and Other Paraneoplastic Syndromes. New England Journal of Medicine. 1995;332(22):1467-75. 162. Anhalt GJ, Kim S, Stanley JR, Korman NJ, Jabs DA, Kory M, et al. Paraneoplastic pemphigus: an autoimmune mucocutaneous disease associated with neoplasia. New England Journal of Medicine. 1990;323(25):1729-35. 163. Vajdic CM, McDonald SP, McCredie MR, van Leeuwen MT, Stewart JH, Law M, et al. Cancer incidence before and after kidney transplantation. Jama. 2006;296(23):2823-31. 164. Serraino D, Piselli P, Busnach G, Burra P, Citterio F, Arbustini E, et al. Risk of cancer following immunosuppression in organ transplant recipients and in HIV-positive individuals in southern Europe. Eur J Cancer. 2007;43(14):2117-23. 165. Nosho K, Baba Y, Tanaka N, Shima K, Hayashi M, Meyerhardt JA, et al. Tumour-infiltrating T-cell subsets, molecular changes in colorectal cancer, and prognosis: cohort study and literature review. J Pathol. 2010;222(4):350-66.

176

166. Clemente CG, Mihm MC, Jr., Bufalino R, Zurrida S, Collini P, Cascinelli N. Prognostic value of tumor infiltrating lymphocytes in the vertical growth phase of primary cutaneous melanoma. Cancer. 1996;77(7):1303-10. 167. Hodi FS, O'Day SJ, McDermott DF, Weber RW, Sosman JA, Haanen JB, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363(8):711-23. 168. Wolchok JD, Kluger H, Callahan MK, Postow MA, Rizvi NA, Lesokhin AM, et al. Nivolumab plus ipilimumab in advanced melanoma. N Engl J Med. 2013;369(2):122-33. 169. Hamid O, Robert C, Daud A, Hodi FS, Hwu WJ, Kefford R, et al. Safety and tumor responses with lambrolizumab (anti-PD-1) in melanoma. N Engl J Med. 2013;369(2):134-44. 170. Everson TC, Cole WH. Spontaneous Regression of Cancer: Preliminary Report. Annals of surgery. 1956;144(3):366-80. 171. Kalialis LV, Drzewiecki KT, Klyver H. Spontaneous regression of metastases from melanoma: review of the literature. Melanoma research. 2009;19(5):275-82. 172. McGovern VJ. Spontaneous Regression of Melanoma. Pathology. 1975;7(2):91- 9. 173. Ribero S, Osella-Abate S, Sanlorenzo M, Savoia P, Astrua C, Cavaliere G, et al. Favourable prognostic role of regression of primary melanoma in AJCC stage I-II patients. Br J Dermatol. 2013;169(6):1240-5. 174. Elhilali MM, Gleave M, Fradet Y, Davis I, Venner P, Saad F, et al. Placebo- associated remissions in a multicentre, randomized, double-blind trial of interferon gamma-1b for the treatment of metastatic renal cell carcinoma. The Canadian Urologic Oncology Group. BJU international. 2000;86(6):613-8. 175. Yutaka Y, Omasa M, Shikuma K, Okuda M, Taki T. Spontaneous regression of an invasive thymoma. General thoracic and cardiovascular surgery. 2009;57(5):272-4. 176. Fukui T, Taniguchi T, Kawaguchi K, Yokoi K. Spontaneous regression of thymic epithelial tumours. Interactive cardiovascular and thoracic surgery. 2014;18(3):399-401. 177. Oquinena S, Guillen-Grima F, Inarrairaegui M, Zozaya JM, Sangro B. Spontaneous regression of hepatocellular carcinoma: a systematic review. Eur J Gastroenterol Hepatol. 2009;21(3):254-7. 178. Gattiker HH, Wiltshaw E, Galton DAG. Spontaneous regression in non- Hodgkin's lymphoma. Cancer. 1980;45(10):2627-32. 179. Prince HM, Duvic M, Martin A, Sterry W, Assaf C, Straus DJ. Incidence of spontaneous remission in patients with CD25-positive mycosis fungoides/Sezary syndrome receiving placebo. J Am Acad Dermatol. 2012;67(5):867-75. 180. O'Rourke MGE, Bell JR. Merkel Cell Tumor with Spontaneous Regression. The Journal of Dermatologic Surgery and Oncology. 1986;12(9):994-7. 181. Connelly TJ, Cribier B, Brown TJ, Yanguas I. Complete Spontaneous Regression of Merkel Cell Carcinoma: A Review of the 10 Reported Cases. Dermatologic Surgery. 2000;26(9):853-6. 182. Correa P, Fontham ET, Bravo JC, Bravo LE, Ruiz B, Zarama G, et al. Chemoprevention of gastric dysplasia: randomized trial of antioxidant supplements and anti-helicobacter pylori therapy. J Natl Cancer Inst. 2000;92(23):1881-8. 183. Byrne T, Mason WP, Posner JB, Dalmau J. Spontaneous neurological improvement in anti-Hu associated encephalomyelitis. Journal of neurology, neurosurgery, and psychiatry. 1997;62(3):276-8.

177

184. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce-Pages C, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313(5795):1960-4. 185. Rathore AS, Kumar S, Konwar R, Makker A, Negi MP, Goel MM. CD3+, CD4+ & CD8+ tumour infiltrating lymphocytes (TILs) are predictors of favourable survival outcome in infiltrating ductal carcinoma of breast. Indian J Med Res. 2014;140(3):361-9. 186. Stumpf M, Hasenburg A, Riener MO, Jutting U, Wang C, Shen Y, et al. Intraepithelial CD8-positive T lymphocytes predict survival for patients with serous stage III ovarian carcinomas: relevance of clonal selection of T lymphocytes. Br J Cancer. 2009;101(9):1513-21. 187. Chew V, Chen J, Lee D, Loh E, Lee J, Lim KH, et al. Chemokine-driven lymphocyte infiltration: an early intratumoural event determining long-term survival in resectable hepatocellular carcinoma. Gut. 2012;61(3):427-38. 188. Karja V, Aaltomaa S, Lipponen P, Isotalo T, Talja M, Mokka R. Tumour- infiltrating lymphocytes: A prognostic factor of PSA-free survival in patients with local prostate carcinoma treated by radical prostatectomy. Anticancer Res. 2005;25(6C):4435-8. 189. Wang W, Hodkinson P, McLaren F, MacKinnon A, Wallace W, Howie S, et al. Small cell lung cancer tumour cells induce regulatory T lymphocytes, and patient survival correlates negatively with FOXP3+ cells in tumour infiltrate. Int J Cancer. 2012;131(6):E928-37. 190. Shah CA, Allison KH, Garcia RL, Gray HJ, Goff BA, Swisher EM. Intratumoral T cells, tumor-associated macrophages, and regulatory T cells: association with p53 mutations, circulating tumor DNA and survival in women with ovarian cancer. Gynecol Oncol. 2008;109(2):215-9. 191. Shankaran V, Ikeda H, Bruce AT, White JM, Swanson PE, Old LJ, et al. IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity. Nature. 2001;410(6832):1107-11. 192. Schreiber RD, Old LJ, Smyth MJ. Cancer immunoediting: integrating immunity's roles in cancer suppression and promotion. Science. 2011;331(6024):1565- 70. 193. MacKie RM, Reid R, Junor B. Fatal melanoma transferred in a donated kidney 16 years after melanoma surgery. N Engl J Med. 2003;348(6):567-8. 194. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646-74. 195. Coley WB. The treatment of malignant tumors by repeated inoculations of erysipelas: with a report of ten original cases. The American Journal of the Medical Sciences. 1893;105(5):487-510. 196. Coley WB. The Treatment of Inoperable Sarcoma by Bacterial Toxins (the Mixed Toxins of the Streptococcus erysipelas and the Bacillus prodigiosus). Proceedings of the Royal Society of Medicine. 1910;3(Surg Sect):1-48. 197. Finn OJ, Schuler G. Introduction to The Renaissance of Cancer Immunotherapy. Annals of the New York Academy of Sciences. 2013;1284(1):v-vii. 198. Camacho LH. CTLA-4 blockade with ipilimumab: biology, safety, efficacy, and future considerations. Cancer medicine. 2015. 199. Kong YC, Flynn JC. Opportunistic Autoimmune Disorders Potentiated by Immune-Checkpoint Inhibitors Anti-CTLA-4 and Anti-PD-1. Frontiers in immunology. 2014;5:206.

178

200. Bollard CM, Gottschalk S, Torrano V, Diouf O, Ku S, Hazrat Y, et al. Sustained complete responses in patients with lymphoma receiving autologous cytotoxic T lymphocytes targeting Epstein-Barr virus latent membrane proteins. J Clin Oncol. 2014;32(8):798-808. 201. Chia WK, Teo M, Wang WW, Lee B, Ang SF, Tai WM, et al. Adoptive T-cell transfer and chemotherapy in the first-line treatment of metastatic and/or locally recurrent nasopharyngeal carcinoma. Mol Ther. 2014;22(1):132-9. 202. Tran E, Turcotte S, Gros A, Robbins PF, Lu YC, Dudley ME, et al. Cancer immunotherapy based on mutation-specific CD4+ T cells in a patient with epithelial cancer. Science. 2014;344(6184):641-5. 203. Miller JFAP, Sprent J. Cell-to-cell interaction in the immune response: VI. Contribution of thymus-derived cells and antibody-forming cell precursors to immunological memory. The Journal of Experimental Medicine. 1971;134(1):66-82. 204. Julius MH, Simpson E, Herzenberg LA. A rapid method for the isolation of functional thymus-derived murine lymphocytes. Eur J Immunol. 1973;3(10):645-9. 205. English D, Andersen BR. Single-step separation of red blood cells. Granulocytes and mononuclear leukocytes on discontinuous density gradients of Ficoll-Hypaque. J Immunol Methods. 1974;5(3):249-52. 206. Munier CM, Zaunders JJ, Ip S, Cooper DA, Kelleher AD. A culture amplified multi-parametric intracellular cytokine assay (CAMP-ICC) for enhanced detection of antigen specific T-cell responses. J Immunol Methods. 2009;345(1-2):1-16. 207. Altman JD, Moss PAH, Goulder PJR, Barouch DH, McHeyzer-Williams MG, Bell JI, et al. Phenotypic Analysis of Antigen-Specific T Lymphocytes. Science. 1996;274(5284):94-6. 208. Wooldridge L, Lissina A, Cole DK, Van Den Berg HA, Price DA, Sewell AK. Tricks with tetramers: how to get the most from multimeric peptide–MHC. Immunology. 2009;126(2):147-64. 209. Vita R, Overton JA, Greenbaum JA, Ponomarenko J, Clark JD, Cantrell JR, et al. The immune epitope database (IEDB) 3.0. Nucleic Acids Res. 2015;43(Database issue):D405-12. 210. Maecker HT, Maino VC, Picker LJ. Immunofluorescence analysis of T-cell responses in health and disease. J Clin Immunol. 2000;20(6):391-9. 211. Letsch A, Scheibenbogen C. Quantification and characterization of specific T- cells by antigen-specific cytokine production using ELISPOT assay or intracellular cytokine staining. Methods. 2003;31(2):143-9. 212. Zaunders JJ, Munier ML, Seddiki N, Pett S, Ip S, Bailey M, et al. High levels of human antigen-specific CD4+ T cells in peripheral blood revealed by stimulated coexpression of CD25 and CD134 (OX40). J Immunol. 2009;183(4):2827-36. 213. Keoshkerian E, Helbig K, Beard M, Zaunders J, Seddiki N, Kelleher A, et al. A novel assay for detection of hepatitis C virus-specific effector CD4(+) T cells via co- expression of CD25 and CD134. J Immunol Methods. 2012;375(1-2):148-58. 214. Hsu DC, Zaunders JJ, Plit M, Leeman C, Ip S, Iampornsin T, et al. A novel assay detecting recall response to Mycobacterium tuberculosis: Comparison with existing assays. Tuberculosis. 2012;92(4):321-7. 215. Hsu DC, Kerr SJ, Iampornsin T, Pett SL, Avihingsanon A, Thongpaeng P, et al. Restoration of CMV-specific-CD4 T cells with ART occurs early and is greater in those with more advanced immunodeficiency. PLoS One. 2013;8(10):e77479. 216. Uchiyama T, Broder S, Waldmann TA. A monoclonal antibody (anti-Tac) reactive with activated and functionally mature human T cells. I. Production of anti-Tac monoclonal antibody and distribution of Tac (+) cells. J Immunol. 1981;126(4):1393-7.

179

217. Uchiyama T, Nelson DL, Fleisher TA, Waldmann TA. A monoclonal antibody (anti-Tac) reactive with activated and functionally mature human T cells. II. Expression of Tac antigen on activated cytotoxic killer T cells, suppressor cells, and on one of two types of helper T cells. J Immunol. 1981;126(4):1398-403. 218. Mallett S, Fossum S, Barclay AN. Characterization of the MRC OX40 antigen of activated CD4 positive T lymphocytes--a molecule related to nerve growth factor receptor. EMBO J. 1990;9(4):1063-8. 219. Seddiki N, Cook L, Hsu DC, Phetsouphanh C, Brown K, Xu Y, et al. Human antigen-specific CD4(+) CD25(+) CD134(+) CD39(+) T cells are enriched for regulatory T cells and comprise a substantial proportion of recall responses. Eur J Immunol. 2014;44(6):1644-61. 220. Curti BD, Kovacsovics-Bankowski M, Morris N, Walker E, Chisholm L, Floyd K, et al. OX40 is a potent immune-stimulating target in late-stage cancer patients. Cancer Res. 2013;73(24):7189-98. 221. Jensen SM, Maston LD, Gough MJ, Ruby CE, Redmond WL, Crittenden M, et al. Signaling through OX40 enhances antitumor immunity. Semin Oncol. 2010;37(5):524-32. 222. Frazer IH, Medley G, Crapper RM, Brown TC, Mackay IR. Association between anorectal dysplasia, human papillomavirus, and human immunodeficiency virus infection in homosexual men. Lancet. 1986;2(8508):657-60. 223. Deeken JF, Tjen ALA, Rudek MA, Okuliar C, Young M, Little RF, et al. The rising challenge of non-AIDS-defining cancers in HIV-infected patients. Clin Infect Dis. 2012;55(9):1228-35. 224. Palefsky JM, Holly EA, Ralston ML, Jay N, Berry JM, Darragh TM. High incidence of anal high-grade squamous intra-epithelial lesions among HIV-positive and HIV-negative homosexual and bisexual men. AIDS. 1998;12(5):495-503. 225. Fox PA. Treatment options for anal intraepithelial neoplasia and evidence for their effectiveness. Sex Health. 2012;9(6):587-92. 226. Palefsky JM. Practising high-resolution anoscopy. Sex Health. 2012;9(6):580-6. 227. Wentzensen N, Follansbee S, Borgonovo S, Tokugawa D, Schwartz L, Lorey TS, et al. Human papillomavirus genotyping, human papillomavirus mRNA expression, and p16/Ki-67 cytology to detect anal cancer precursors in HIV-infected MSM. AIDS. 2012;26(17):2185-92. 228. Byakwaga H, Zhou J, Petoumenos K, Law MG, Boyd MA, Emery S, et al. Effect of nucleoside reverse transcriptase inhibitors on CD4 T-cell recovery in HIV-1- infected individuals receiving long-term fully suppressive combination antiretroviral therapy. HIV Med. 2009;10(3):143-51. 229. Piketty C, Selinger-Leneman H, Bouvier AM, Belot A, Mary-Krause M, Duvivier C, et al. Incidence of HIV-Related Anal Cancer Remains Increased Despite Long-Term Combined Antiretroviral Treatment: Results From the French Hospital Database on HIV. J Clin Oncol. 2012;30(35):4360-6. 230. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services; 2012. Available from: http://aidsinfo.nih.gov/contentfiles/lvguidelines/AdultandAdolescentGL.pdf. Accessed 14 January 2013. 231. Piketty C, Cochand-Priollet B, Lanoy E, Si-Mohamed A, Trabelsi S, Tubiana R, et al. Lack of regression of anal squamous intraepithelial lesions despite immune restoration under cART. AIDS. 2013;27(3):401-6.

180

232. Mathews WC, Cachay ER, Caperna J, Sitapati A, Cosman B, Abramson I. Estimating the accuracy of anal cytology in the presence of an imperfect reference standard. PLoS One. 2010;5(8):e12284. 233. Berry JM, Palefsky JM, Jay N, Cheng SC, Darragh TM, Chin-Hong PV. Performance characteristics of anal cytology and human papillomavirus testing in patients with high-resolution anoscopy-guided biopsy of high-grade anal intraepithelial neoplasia. Dis Colon Rectum. 2009;52(2):239-47. 234. Lytwyn A, Salit IE, Raboud J, Chapman W, Darragh T, Winkler B, et al. Interobserver agreement in the interpretation of anal intraepithelial neoplasia. Cancer. 2005;103(7):1447-56. 235. Darragh TM, Tokugawa D, Castle PE, Follansbee S, Borgonovo S, LaMere BJ, et al. Interrater agreement of anal cytology. Cancer Cytopathol. 2013;121(2):72-8. 236. Grulich AE, Hillman RJ. Study of the Prevention of Anal Cancer. Available from: http://www.nchecrsurveys.unsw.edu.au/spanc/index.html. Accessed 6th February 2012. 237. De Vuyst H, Clifford GM, Nascimento MC, Madeleine MM, Franceschi S. Prevalence and type distribution of human papillomavirus in carcinoma and intraepithelial neoplasia of the vulva, vagina and anus: a meta-analysis. Int J Cancer. 2009;124(7):1626-36. 238. Hillman RJ, Garland SM, Gunathilake MP, Stevens M, Kumaradevan N, Lemech C, et al. Human papillomavirus (HPV) genotypes in an Australian sample of anal cancers. Int J Cancer. 2014;135(4):996-1001. 239. Dalla Pria A, Alfa-Wali M, Fox P, Holmes P, Weir J, Francis N, et al. High- resolution anoscopy screening of HIV-positive MSM: longitudinal results from a pilot study. AIDS. 2014;28(6):861-7. 240. Tong WW, Jin F, McHugh LC, Maher T, Sinclair B, Grulich AE, et al. Progression to and spontaneous regression of high-grade anal squamous intraepithelial lesions in HIV-infected and uninfected men. AIDS. 2013;27(14):2233-43. 241. Hopfl R, Heim K, Christensen N, Zumbach K, Wieland U, Volgger B, et al. Spontaneous regression of CIN and delayed-type hypersensitivity to HPV-16 oncoprotein E7. Lancet. 2000;356(9246):1985-6. 242. Tong WW, Hillman RJ, Kelleher AD, Grulich AE, Carr A. Anal intraepithelial neoplasia and squamous cell carcinoma in HIV-infected adults. HIV Med. 2014;15(2):65-76. 243. Goodell V, dela Rosa C, Slota M, MacLeod B, Disis ML. Sensitivity and specificity of tritiated thymidine incorporation and ELISPOT assays in identifying antigen specific T cell immune responses. BMC immunology. 2007;8:21. 244. Brown SR, Skinner P, Tidy J, Smith JH, Sharp F, Hosie KB. Outcome after surgical resection for high-grade anal intraepithelial neoplasia (Bowen's disease). The British journal of surgery. 1999;86(8):1063-6. 245. Machalek DA, Grulich AE, Hillman RJ, Jin F, Templeton DJ, Tabrizi SN, et al. The Study of the Prevention of Anal Cancer (SPANC): design and methods of a three- year prospective cohort study. BMC public health. 2013;13(1):946. 246. CDC. Guidelines for performing single-platform absolute CD4+ T-cell determinations with CD45 gating for persons infected with human immunodeficiency virus. MMWR. 2003(52(No. RR-2)):1-11. 247. Schnizlein-Bick CT, Spritzler J, Wilkening CL, Nicholson JK, O'Gorman MR. Evaluation of TruCount absolute-count tubes for determining CD4 and CD8 cell numbers in human immunodeficiency virus-positive adults. Site Investigators and The

181

NIAID DAIDS New Technologies Evaluation Group. Clin Diagn Lab Immunol. 2000;7(3):336-43. 248. Hsu DC, Kerr SJ, Thongpaeng P, Iampornsin T, Pett SL, Zaunders JJ, et al. Incomplete restoration of Mycobacterium tuberculosis-specific-CD4 T cell responses despite antiretroviral therapy. J Infect. 2014;68(4):344-54. 249. Kelleher AD, Puls RL, Bebbington M, Boyle D, Ffrench R, Kent SJ, et al. A randomized, placebo-controlled phase I trial of DNA prime, recombinant fowlpox virus boost prophylactic vaccine for HIV-1. AIDS. 2006;20(2):294-7. 250. Currier JR, Kuta EG, Turk E, Earhart LB, Loomis-Price L, Janetzki S, et al. A panel of MHC class I restricted viral peptides for use as a quality control for vaccine trial ELISPOT assays. J Immunol Methods. 2002;260(1-2):157-72. 251. Maecker HT, Trotter J. Flow cytometry controls, instrument setup, and the determination of positivity. Cytometry A. 2006;69(9):1037-42. 252. Passmore JA, Milner M, Denny L, Sampson C, Marais DJ, Allan B, et al. Comparison of cervical and blood T-cell responses to human papillomavirus-16 in women with human papillomavirus-associated cervical intraepithelial neoplasia. Immunology. 2006;119(4):507-14. 253. Welters MJ, de Jong A, van den Eeden SJ, van der Hulst JM, Kwappenberg KM, Hassane S, et al. Frequent display of human papillomavirus type 16 E6-specific memory t-Helper cells in the healthy population as witness of previous viral encounter. Cancer Res. 2003;63(3):636-41. 254. de Gruijl TD, Bontkes HJ, Walboomers JM, Stukart MJ, Doekhie FS, Remmink AJ, et al. Differential T helper cell responses to human papillomavirus type 16 E7 related to viral clearance or persistence in patients with cervical neoplasia: a longitudinal study. Cancer Res. 1998;58(8):1700-6. 255. Kim KH, Greenfield WW, Cannon MJ, Coleman HN, Spencer HJ, Nakagawa M. CD4+ T-cell response against human papillomavirus type 16 E6 protein is associated with a favorable clinical trend. Cancer Immunol Immunother. 2012;61(1):63- 70. 256. Todd RW, Roberts S, Mann CH, Luesley DM, Gallimore PH, Steele JC. Human papillomavirus (HPV) type 16-specific CD8+ T cell responses in women with high grade vulvar intraepithelial neoplasia. Int J Cancer. 2004;108(6):857-62. 257. Phetsouphanh C, Xu Y, Amin J, Seddiki N, Procopio F, Sekaly RP, et al. Characterization of transcription factor phenotypes within antigen-specific CD4+ T cells using qualitative multiplex single-cell RT-PCR. PLoS One. 2013;8(10):e74946. 258. Trimble CL, Clark RA, Thoburn C, Hanson NC, Tassello J, Frosina D, et al. Human papillomavirus 16-associated cervical intraepithelial neoplasia in humans excludes CD8 T cells from dysplastic epithelium. J Immunol. 2010;185(11):7107-14. 259. Ovestad IT, Gudlaugsson E, Skaland I, Malpica A, Kruse AJ, Janssen EA, et al. Local immune response in the microenvironment of CIN2-3 with and without spontaneous regression. Mod Pathol. 2010;23(9):1231-40. 260. Kojima S, Kawana K, Fujii T, Yokoyama T, Miura S, Tomio K, et al. Characterization of gut-derived intraepithelial lymphocyte (IEL) residing in human papillomavirus (HPV)-infected intraepithelial neoplastic lesions. Am J Reprod Immunol. 2011;66(5):435-43. 261. Kojima S, Kawana K, Tomio K, Yamashita A, Taguchi A, Miura S, et al. The prevalence of cervical regulatory T cells in HPV-related cervical intraepithelial neoplasia (CIN) correlates inversely with spontaneous regression of CIN. Am J Reprod Immunol. 2013;69(2):134-41.

182

262. Gervaz E, Dauge-Geffroy MD, Sobhani I, Vissuzaine C, Mignon M, Benhamou G, et al. Quantitative analysis of the immune cells in the anal mucosa. Pathology, research and practice. 1995;191(11):1067-71. 263. Sobhani I, Walker F, Aparicio T, Abramowitz L, Henin D, Cremieux AC, et al. Effect of anal epidermoid cancer-related viruses on the dendritic (Langerhans') cells of the human anal mucosa. Clin Cancer Res. 2002;8(9):2862-9. 264. Yaghoobi M, Le Gouvello S, Aloulou N, Duprez-Dutreuil C, Walker F, Sobhani I. FoxP3 overexpression and CD1a+ and CD3+ depletion in anal tissue as possible mechanisms for increased risk of human papillomavirus-related anal carcinoma in HIV infection. Colorectal disease : the official journal of the Association of Coloproctology of Great Britain and Ireland. 2011;13(7):768-73. 265. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nature methods. 2012;9(7):676-82. 266. De Vos K. Available from: http://rsb.info.nih.gov/ij/plugins/cell-counter.html. Accessed December 2014. 267. Rasband WS. Available from: http://rsb.info.nih.gov/ij/plugins/grid.html. Accessed December 2014. 268. Maldonado L, Teague JE, Morrow MP, Jotova I, Wu TC, Wang C, et al. Intramuscular therapeutic vaccination targeting HPV16 induces T cell responses that localize in mucosal lesions. Science translational medicine. 2014;6(221):221ra13. 269. Cornall AM, Roberts JM, Garland SM, Hillman RJ, Grulich AE, Tabrizi SN. Anal and perianal squamous carcinomas and high-grade intraepithelial lesions exclusively associated with "low-risk" HPV genotypes 6 and 11. Int J Cancer. 2013;133(9):2253-8. 270. Wentzensen N. Screening for anal cancer: endpoints needed. Lancet Oncol. 2012;13(5):438-40. 271. Poynten IM, Machalek DA, Jin F, Hillman R, Templeton DJ, Roberts J, et al. Use of a composite cytology-histology endpoint in the diagnosis of anal high grade squamous intra-epithelial lesions (HSIL). 29th International Papillomavirus Conference and Public Health & Clinical Workshops; 21-25 August 2014; Seattle, USA. 2014. p. 307. 272. Burgos J, Curran A, Tallada N, Guelar A, Navarro J, Landolfi S, et al. Risk of progression to high-grade anal intraepithelial neoplasia in HIV-infected MSM. AIDS. 2015;29(6):695-702. 273. Phanuphak N, Teeratakulpisarn N, Triratanachat S, Keelawat S, Pankam T, Kerr SJ, et al. High prevalence and incidence of high-grade anal intraepithelial neoplasia among young Thai men who have sex with men with and without HIV. AIDS (London, England). 2013;27(11):1753-62. 274. Berry JM, Jay N, Cranston RD, Darragh TM, Holly EA, Welton ML, et al. Progression of anal high-grade squamous intraepithelial lesions to invasive anal cancer among HIV-infected men who have sex with men. Int J Cancer. 2014;134(5):1147-55. 275. Cachay E, Agmas W, Mathews C. Five-year cumulative incidence of invasive anal cancer among HIV-infected patients according to baseline anal cytology results: an inception cohort analysis. HIV Med. 2014. 276. Grulich A, Jin F, Poynten IM, Roberts J, Farnsworth A, Templeton DJ, et al. Incidence and clearance of anal high-grade squamous intraepithelial lesions (HSIL) in HIV positive and HIV negative homosexual men. 29th International Papillomavirus Conference and Public Health & Clinical Workshops; 24 August 2014; Seattle, USA. 2014. p. 217.

183

277. Grulich AE, Jin F, Poynten IM, Roberts J, Farnsworth A, Templeton DJ, et al. Incidence and clearance of anal high-grade squamous intraepithelial lesions (HSIL) in HIV-positive and HIV-negative homosexual men. 16th International Workshop on Co- morbidities and Adverse Drug Reactions in HIV; Philadelphia, USA.: Antiviral Therapy. Vol 19; 2014. p. A15. 278. Comi L, Tincati C, Bellistrì GM, Pandolfo A, Martinez JS, Rainone V, et al. Mucosal and Systemic Immune Responses in HIV/HPV Co-Infected Males [CROI abstract 719]. In Special Issue: Abstracts From the 2014 Conference on Retroviruses and Opportunistic Infections; Boston, USA: Top Antivir Med; 2014. Vol 22(e-1). p. 365. 279. Koskimaa HM, Paaso AE, Welters MJ, Grenman SE, Syrjanen KJ, van der Burg SH, et al. Human papillomavirus 16 E2-, E6- and E7-specific T-cell responses in children and their mothers who developed incident cervical intraepithelial neoplasia during a 14-year follow-up of the Finnish Family HPV cohort. J Transl Med. 2014;12:44. 280. Jacobelli S, Sanaa F, Moyal-Barracco M, Pelisse M, Berville S, Villefroy P, et al. Anti-HPV16 E2 protein T-cell responses and viral control in women with usual vulvar intraepithelial neoplasia and their healthy partners. PLoS One. 2012;7(5):e36651. 281. Ting J, Kruzikas DT, Smith JS. A global review of age-specific and overall prevalence of cervical lesions. Int J Gynecol Cancer. 2010;20(7):1244-9. 282. Alam NN, White DA, Narang SK, Daniels IR, Smart NJ. Systematic Review of guidelines for the assessment and management of High Grade Anal Intraepithelial Neoplasia (AIN II/III). Colorectal disease : the official journal of the Association of Coloproctology of Great Britain and Ireland. 2015. 283. Wilson JMG, Jungner G. Principles and practices of screening for disease. Geneva, Switzerland. World Health Organization 1968. Available from: http://whqlibdoc.who.int/php/WHO_PHP_34.pdf. Accessed November 2015. 284. Trimble CL, Morrow MP, Kraynyak KA, Shen X, Dallas M, Yan J, et al. Safety, efficacy, and immunogenicity of VGX-3100, a therapeutic synthetic DNA vaccine targeting human papillomavirus 16 and 18 E6 and E7 proteins for cervical intraepithelial neoplasia 2/3: a randomised, double-blind, placebo-controlled phase 2b trial. Lancet. 2015;386(10008):2078-88. 285. van Poelgeest MI, Welters MJ, Vermeij R, Stynenbosch LF, Loof NM, Berends- van der Meer DM, et al. Vaccination against oncoproteins of HPV16 for non-invasive vulvar/vaginal lesions: lesion clearance is related to the strength of the T-cell response. Clin Cancer Res. 2016. 286. Alvarez RD, Huh WK, Bae S, Lamb LS, Jr., Conner MG, Boyer J, et al. A pilot study of pNGVL4a-CRT/E7(detox) for the treatment of patients with HPV16+ cervical intraepithelial neoplasia 2/3 (CIN2/3). Gynecol Oncol. 2016;140(2):245-52. 287. Kim TJ, Jin HT, Hur SY, Yang HG, Seo YB, Hong SR, et al. Clearance of persistent HPV infection and cervical lesion by therapeutic DNA vaccine in CIN3 patients. Nat Commun. 2014;5:5317. 288. Stevanovic S, Draper LM, Langhan MM, Campbell TE, Kwong ML, Wunderlich JR, et al. Complete regression of metastatic cervical cancer after treatment with human papillomavirus-targeted tumor-infiltrating T cells. J Clin Oncol. 2015;33(14):1543-50. 289. Eskander RN, Tewari KS. Immunotherapy: an evolving paradigm in the treatment of advanced cervical cancer. Clin Ther. 2015;37(1):20-38. 290. Barrett DM, Singh N, Porter DL, Grupp SA, June CH. Chimeric antigen receptor therapy for cancer. Annual review of medicine. 2014;65:333-47.

184

291. Villadolid J, Amin A. Immune checkpoint inhibitors in clinical practice: update on management of immune-related toxicities. Transl Lung Cancer Res. 2015;4(5):560- 75. 292. Houlihan CF, Larke NL, Watson-Jones D, Smith-McCune KK, Shiboski S, Gravitt PE, et al. Human papillomavirus infection and increased risk of HIV acquisition. A systematic review and meta-analysis. AIDS. 2012;26(17):2211-22. 293. Chao C, Leyden WA, Xu L, Horberg MA, Klein D, Towner WJ, et al. Exposure to antiretroviral therapy and risk of cancer in HIV-infected persons. AIDS. 2012;26(17):2223-31. 294. Bruyand M, Ryom L, Shepherd L, Fatkenheuer G, Grulich A, Reiss P, et al. Cancer risk and use of protease inhibitor or nonnucleoside reverse transcriptase inhibitor-based combination antiretroviral therapy: the D: A: D study. J Acquir Immune Defic Syndr. 2015;68(5):568-77. 295. van der Snoek EM, van der Ende ME, den Hollander JC, Schutten M, Neumann HA, van Doornum GJ. Use of Highly Active Antiretroviral Therapy Is Associated With Lower Prevalence of Anal Intraepithelial Neoplastic Lesions and Lower Prevalence of Human Papillomavirus in HIV-Infected Men Who Have Sex With Men. Sex Transm Dis. 2012;39(7):495-500. 296. Hidalgo-Tenorio C, Rivero-Rodriguez M, Gil-Anguita C, Lopez De Hierro M, Palma P, Ramirez-Taboada J, et al. Antiretroviral Therapy as a Factor Protective against Anal Dysplasia in HIV-Infected Males Who Have Sex with Males. PLoS One. 2014;9(3):e92376. 297. INSIGHT START Study Group, Lundgren JD, Babiker AG, Gordin F, Emery S, Grund B, et al. Initiation of Antiretroviral Therapy in Early Asymptomatic HIV Infection. N Engl J Med. 2015;373(9):795-807. 298. Carpenter AE, Jones TR, Lamprecht MR, Clarke C, Kang IH, Friman O, et al. CellProfiler: image analysis software for identifying and quantifying cell phenotypes. Genome Biol. 2006;7(10):R100. 299. Tidbury P, Singer A, Jenkins D. CIN 3: the role of lesion size in invasion. British journal of obstetrics and gynaecology. 1992;99(7):583-6. 300. Yang HP, Zuna RE, Schiffman M, Walker JL, Sherman ME, Landrum LM, et al. Clinical and pathological heterogeneity of cervical intraepithelial neoplasia grade 3. PLoS One. 2012;7(1):e29051. 301. Chen D, Juko-Pecirep I, Hammer J, Ivansson E, Enroth S, Gustavsson I, et al. Genome-wide association study of susceptibility loci for cervical cancer. J Natl Cancer Inst. 2013;105(9):624-33. 302. Madeleine M, Leo P, Wang S, Schwartz S, Hemminki K, Hallmans G, et al. Genomewide association study identifies MHC associations with and high heritability of cervical pre-cancer and cancer as markers of HPV persistence. 29th International Papillomavirus Conference and Public Health & Clinical Workshops; 24 August 2014; Seattle, USA. 2014. p. 92. 303. Bourgault Villada I. Vaccination against HPV-16 for vulvar intraepithelial neoplasia. N Engl J Med. 2010;362(7):655-6; author reply 6.

185

Tong WWY, Hillman RJ, Kelleher AD, Grulich AE, Carr A. Anal intraepithelial neoplasia and squamous cell carcinoma in HIV-infected adults. HIV Medicine.

2014;15(2):65-76. [Review]

Tong WWY, Jin F, McHugh LC, Maher T, Sinclair B, Grulich AE, Hillman RJ, Carr A.

Progression to and spontaneous regression of high-grade anal squamous intraepithelial lesions in HIV-infected and uninfected men. AIDS. 2013;27(14):2233-43.

Tong WWY, Shepherd K, Garland S, Meagher A, Templeton DJ, Fairley CK, Jin F,

2015;211(3):405-15.

This USB drive contains:

1. All the anal biopsy microscopy images acquired for Chapter 4 with counting

grid and markers. Images are in TIFF format.

2. A digital copy of this thesis.

He who studies medicine without books sails an uncharted sea, but he who studies

medicine without patients does not go to sea at all.

– W. Osler