Infection with human papillomavirus (HPV) is highly prevalent throughout the world, and is associated with several types of cancer [1,2][1,2]. Persistent HPV infection with a high-risk (hr)HPV type (mainly HPV-16 or -18) is the likely cause of around 80% of anal cancers and 40–50% of penile cancers [3–5][3–5][3–5]. Anal cancer is relatively rare in the general population, but MSM, and especially HIV-infected MSM, are at much higher risk . Moreover, the incidence of anal cancer has shown an increasing trend in recent decades [7,8][7,8].
Prevalence of anogenital HPV infection in men varies widely across studies, with estimates up to 93% in HIV-infected MSM [9–11][9–11][9–11]. However, there is little insight in the anal and penile HPV incidence and clearance among MSM. In addition, although it is known that HIV-infected individuals are at increased risk of HPV infection and HPV-related cancer [6,12][6,12], the effect of HIV infection on the natural history of HPV infection is not well understood. In particular, it is not clear whether the increased HPV infection risk among HIV-infected individuals can be mainly explained by biological (i.e. HIV-related immunosuppression) or behavioral factors (i.e. shared routes of transmission).
The present study aimed to compare the anal and penile high-risk HPV incidence and clearance between HIV-infected and HIV-negative MSM over two years of follow-up, and to assess the effect of HIV-related immunosuppression on HPV incidence and clearance.
Materials and methods
Study methods have been described in detail previously . In short, HIV-negative and HIV-infected MSM were invited to participate in the HIV and HPV in MSM (H2M) cohort study from July 2010 to July 2011 at three sites in Amsterdam, the Netherlands: the Amsterdam Cohort Study (ACS) among MSM (Public Health Service Amsterdam), a sexually transmitted infections clinic (Public Health Service Amsterdam), and an infectious disease outpatient clinic (Jan van Goyen Medical Center). Men were eligible for participation if they were aged 18 years or older and conversant in Dutch and/or English. The Medical Ethics Committee of the Academic Medical Center Amsterdam approved this study and all participants provided written informed consent prior to enrolment.
Data were collected at baseline and at 6-month intervals during an intended follow-up time of 24 months, aiming at five visits per participant. At each visit, participants completed detailed self-administered questionnaires and collected an anal and a penile self-swab (regular flocked swab with 1 ml Universal Transport Medium; Copan, Brescia, Italy). For the anal swab, participants inserted the swab 3 cm into the anal canal and turned it around for 5–10 s. For the penile swab, participants rubbed the swab firmly over the skin of the penile shaft, including the outside of the foreskin, for 20 s (i.e. 5 s per side). HIV-related data were obtained from the Dutch HIV Monitoring Foundation's national HIV patients database.
Human papillomavirus DNA detection and genotyping
The anal and penile samples were stored at –20°C and analyzed as previously described . DNA extraction was performed using the MagNA Pure LC Total Nucleic Acid Isolation Kit (Roche, Mannheim, Germany). DNA amplification, testing for HPV DNA, and HPV genotyping was performed using the highly sensitive SPF10-PCR DEIA/LiPA25 system (version 1) . LiPA25 allows simultaneous detection of 25 specific mucosal HPV genotypes, of which HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 were classified as high-risk .
Baseline characteristics were compared between groups by χ2-squared test for categorical data; for continuous variables, the Student's t-test was used for normally distributed data and the rank sum test when the data were not normally distributed. To deal with missing questionnaire data (Supplementary Text 1, https://links.lww.com/QAD/A792) we imputed missing values using the MICE (multivariate imputation by chained equations) technique . Before imputing missing data, the number of recent (i.e. in the last 6 months) anal sex partners, lifetime male sex partners, and current and nadir CD4+ cell count were transformed to the logarithmic scale given their skewed distribution. In addition, complete-case analyses were performed, using only the available data per participant.
The outcomes of interest were anal and penile hrHPV incidence and clearance. In our main analyses (assumption I), incidence was defined as one positive test result for a specific hrHPV type at a given visit preceded by two consecutive negative test results for that HPV type. Clearance was defined as one positive test result for a specific hrHPV type followed by two consecutive negative visits. All analyses were performed on an HPV type-specific level and restricted to the 12 hrHPV types. Persons were at risk for a specific HPV type until their first event of that type, but remained at risk for other types.
Type-specific anal and penile hrHPV incidence and clearance rates were calculated separately by HIV status, with corresponding incidence rate ratios (IRR) and clearance rate ratios (CRR). Kaplan–Meier curves were constructed to explore the cumulative incidence and clearance of HPV-16 and HPV-18 by HIV status. To further assess the effect of HIV infection, crude and adjusted IRR and CRR for HIV-infected versus HIV-negative MSM were calculated for HPV-16 and -18 separately using Poisson regression. We also assessed the effect for the 12 hrHPV types combined via Poisson regression, using generalized estimating equations with an exchangeable correlation structure to account for multiple HPV types per person .
Variables that were a priori included in the multivariable models for anal HPV incidence and clearance – based on literature  and our previous analyses  – were as follows: age, number of lifetime male sex partners (both collected at baseline), number of recent anal sex partners, current smoking, recent cannabis and/or poppers use, anal sex position, having been rimmed (i.e., anal–oral contact), receptive fisting (all time-updated and asked over the previous 6 months), anal sexually transmitted infections (STI) diagnosis at time of visit (chlamydia and/or gonorrhea), and HPV type. The same variables were a priori included in the multivariable models for penile HPV incidence and clearance, with the exception of having been rimmed, receptive fisting and anal STI, and the addition of circumcision status at baseline and urethral STI diagnosis at time of visit (chlamydia and/or gonorrhea). All continuous variables (i.e., age, number of lifetime and recent sex partners) were allowed to vary smoothly using natural cubic splines .
We also assessed crude and adjusted anal and penile IRR and CRR according to immune status (categorized into HIV-negative; HIV-infected with nadir CD4+ cell count >350 cells/μl; HIV-infected with nadir CD4+ cell count between 200 and 350 cells/μl; and HIV-infected with nadir CD4+ cell count <200 cells/μl) using the same models and outcomes as described above. Furthermore, we assessed anal and penile CRR for incident (i.e., infections that were not detected at baseline) versus prevalent (i.e., infections that were detected at baseline) hrHPV infections using the same models and outcomes as described above.
Sensitivity analyses were performed using alternative definitions of incidence and clearance. In sensitivity analyses with assumption II, incidence was defined as one positive test result preceded by one negative visit, whereas clearance was defined as one positive test result followed by one negative visit. In sensitivity analyses with assumption III, incidence was defined as two consecutive positive test results preceded by one negative visit, whereas clearance was defined as two consecutive positive test results followed by one negative visit.
Statistical analyses were performed using Stata software package version 13.1 (Stata Intercooled, College Station, Texas, USA) and the R statistical computing environment version 3.0.2. .
Role of the funding source
The funders had no role in study design; the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
Characteristics of the study population
Seven hundred and ninety-five MSM were enrolled in the H2M study, of which 750 provided sufficient baseline data and attended at least one follow-up visit. The 45 excluded MSM did not significantly differ with respect to HIV status or age compared with the 750 included MSM.
Of the 750 MSM, 302 (40%) were HIV-infected at baseline. Median follow-up time was 25 months (interquartile, IQR 24–26) in HIV-negative and 24 months (IQR 23–25) in HIV-infected MSM; 83% had five study visits. HIV-infected MSM were significantly older, with a median age of 46 years (IQR 40–53) compared with 38 years (IQR 34–42) in HIV-negative MSM (P < 0.001). The median number of lifetime male sex partners was 300 (IQR 100–1000) in HIV-infected versus 100 (50–400) in HIV-negative MSM (P < 0.001). In HIV-infected MSM, the median nadir CD4+ cell count was 230 cells/μl (IQR 170–320), 87% was using combination antiretroviral therapy and 79% had undetectable HIV viral load at baseline.
At baseline, anal hrHPV infection was detected in 46% [95% (confidence interval, CI 41–50%) of HIV-negative and 63% (95% CI 58–69%) of HIV-infected MSM (P < 0.001)]. Penile hrHPV infection was detected in 17% (95% CI 13–20%) and 31% (95% CI 26–37%) of MSM, respectively (P < 0.001). A detailed description of the study population with anal and penile HPV prevalence at baseline has been published previously .
Anal and penile high-risk human papillomavirus incidence
In total, 705 MSM attended at least two follow-up visits, and were included in our main analyses. Among these 705 MSM, 652 incident anal and 419 incident penile hrHPV infections were observed.
Type-specific anal and penile hrHPV incidence rates, stratified by HIV status, are shown in Table 1. Anal incidence rates were higher in HIV-infected compared with HIV-negative MSM (except for HPV-51), with statistically significant differences for HPV types 16, 31, 35, 52, and 56. Anal HPV-16 incidence rate was 4.7/1000 person-months of observation (PMO) in HIV-negative and 9.1/1000 PMO in HIV-infected MSM (IRR 2.0; 95% CI 1.2–3.2). Penile incidence rates were also higher in HIV-infected compared with HIV-negative MSM (except for HPV-39), with statistically significant differences for HPV types 31 and 35. Penile HPV-16 incidence rate was 4.2/1000 PMO in HIV-negative and 5.7/1000 PMO in HIV-infected MSM (IRR 1.4; 95% CI 0.8–2.3). Figure 1 shows Kaplan–Meier curves of the cumulative incidence of anal and penile HPV-16 and -18 infections, respectively, stratified by HIV status.
Similar patterns were observed in sensitivity analyses with assumptions II and III (Supplementary Tables 1 and 2, respectively, https://links.lww.com/QAD/A792), with most IRRs above 1.0 for both anal and penile HPV infection.
Effect of HIV infection on anal and penile high-risk human papillomavirus incidence
In univariable Poisson analyses, the anal incidence rates among HIV-infected MSM were significantly higher than among HIV-negative MSM for HPV-16 and for the 12 hrHPV types combined, although not for HPV-18 (Table 2). In multivariable analyses, the associations for HPV-16 and the hrHPV types combined remained significant, with an adjusted IRR (aIRR) of 1.9 (95% CI 1.1–3.2) and 1.6 (95%CI 1.3–2.1), respectively. In univariable analyses for incident penile HPV infection, a significantly increased IRR was observed for the 12 hrHPV types combined, which remained borderline significant in multivariable analyses with an aIRR of 1.4 (95% CI 1.0–2.1). In sensitivity analyses with assumptions II and III similar results were obtained (Table 2). Complete-case analyses showed largely similar results (data not shown).
Anal and penile high-risk human papillomavirus clearance
In total, 744 events of cleared anal hrHPV infection were observed in 519 MSM; 397 cleared penile infections were observed in 262 MSM.
Type-specific anal and penile hrHPV clearance rates, stratified by HIV status, are shown in Table 3. Anal clearance rates were lower in HIV-infected compared with HIV-negative MSM (except for HPV-58 and HPV-59), with statistically significant differences for HPV types 31, 35, 45, and 52. Anal HPV-16 clearance rate was 46.9/1000 PMO in HIV-negative and 43.7/1000 PMO in HIV-infected MSM (CRR 0.9; 95%CI 0.6–1.4). In contrast, penile clearance rates were comparable to or higher among HIV-infected compared with HIV-negative MSM. Penile HPV-16 clearance rate was 71.4/1000 PMO in HIV-negative and 132.8/1000 PMO in HIV-infected MSM (CRR 1.9; 95%CI 1.0–3.3). Figure 2 shows Kaplan–Meier curves of the cumulative clearance of anal and penile HPV-16 and -18 infections stratified by HIV status.
Comparable patterns were observed in sensitivity analyses with assumptions II and III (Supplementary Tables 3 and 4, respectively, https://links.lww.com/QAD/A792), with most CRR below 1.0 for anal infection and above 1.0 for penile infection.
Effect of HIV infection on anal and penile high-risk human papillomavirus clearance
In univariable Poisson analyses, the anal CRR for HIV-infected versus HIV-negative MSM was 0.7 (95% CI 0.6–0.8) for the 12 hrHPV types combined; the CRR remained significantly below 1.0 in multivariable analyses [adjusted CRR (aCRR) 0.7; 95%CI 0.6–0.9] (Table 4). The penile CRR in univariable analyses was 1.3 (95% CI 1.0–1.7) for the hrHPV types combined, and was nonsignificantly increased in multivariable analyses (aCRR 1.3; 95% CI 1.0–1.7). In sensitivity analyses with assumptions II and III comparable results were obtained (Table 4). Complete-case analyses showed similar results (data not shown).
Effect of immune status on anal and penile human papillomavirus incidence and clearance
No effect of immune status on anal or penile HPV incidence or clearance rates was observed; 95% confidence intervals for different categories of nadir CD4+ cell count in HIV-infected MSM overlapped in all analyses, without any clear trends (Table 5).
Clearance among incident versus prevalent human papillomavirus infections
The clearance rates of both anal and penile hrHPV infections were significantly higher for incident compared with prevalent hrHPV infection [aCRR for anal HPV 1.4 (95% CI 1.2–1.6); aCRR for penile HPV 1.3 (95% CI 1.1–1.6)] (Supplementary Table 5, https://links.lww.com/QAD/A792).
HIV-infected MSM had an increased incidence of anal and penile hrHPV infection, including anal HPV-16, the main causative agent of anal cancer. In addition, HIV-infected MSM had a decreased clearance of anal hrHPV. No significant independent effect of HIV infection was observed for penile hrHPV clearance. Anal and penile HPV incidence or clearance did not significantly differ by level of nadir CD4+ cell count among HIV-infected MSM.
Anal HPV incidence rates were largely comparable to those found in other studies among HIV-negative [20,21][20,21] and HIV-infected [21–24][21–24][21–24][21–24] MSM, although comparisons across studies are hampered by differences in study populations, laboratory methods, and statistical methods used (e.g. definitions of incidence and clearance). Anal HPV clearance rates observed in our study were mostly higher than reported for HIV-infected MSM so far [22,23][22,23]. Interestingly, the clearance rate of HPV-16 was low compared with other HPV types, which is in line with previous studies [22,23][22,23], and may partly explain the high oncogenic potential of HPV-16.
Penile HPV incidence and clearance rates were considerably higher than previously reported in HIV-infected MSM . Our observed incidence rates among HIV-negative MSM were comparable with rates of genital HPV infection reported in a large multinational sample of HIV-negative, mostly heterosexual men . The high penile HPV clearance rates observed in our study are striking. Although we have no clear explanation for the difference in clearance between anal and penile HPV infection, it may be conceivable that infections in keratinized epithelium (such as at the penile shaft, where penile samples were taken) clear faster than mucosal infections (such as at the anal canal, where anal samples were taken), or that HPV viral latency differs between both anatomical sites.
HIV infection showed a significant effect on anal and penile HPV incidence, and anal HPV clearance, independent of reported sexual behavior or other potential confounders. These findings are in line with the few data available regarding anogenital HPV infection in men [21,26][21,26] and cervical infection in women . However, even in our study where we adjusted for multiple sexual behavior variables, residual confounding cannot be ruled out. For example, assortative mixing (whereby HIV-infected men preferentially have sex with other HIV-infected men) may lead to increased HPV incidence among HIV-infected MSM. Nevertheless, our data suggest a biological effect of HIV infection on HPV acquisition and clearance, implying more persistent hrHPV infections independent of sexual behavior among HIV-infected individuals, which presumably contributes to their increased risk of HPV-related cancer. The fact that we could not demonstrate an effect of immune status as measured by nadir CD4+ cell count may be due to a relatively low number of participants with severe immunosuppression, or due to HIV-related immunological effects not captured by CD4 values.
As persistent hrHPV infection poses a risk of becoming malignant, understanding which infections are transient versus persistent is crucial. Our analyses confirm previous data  that incident HPV infections clear faster than prevalent infections. This seems logical, as prevalent infections are a combination of recently acquired infections and long-standing infections that are less likely to clear.
Strengths of this study include the large and well-characterized cohort consisting of HIV-negative and HIV-infected MSM (enabling us to assess the effect of HIV infection whereas adjusting for multiple potential confounders), the longitudinal analyses of both anal and penile hrHPV infection, and the sensitive laboratory methods used. In addition, by conducting various sensitivity analyses, we were able to show the major impact of different methodological definitions on reported incidence and clearance rates, whereas the independent effect of HIV infection on incidence and clearance was not affected by assumptions.
The present study has also some limitations. First, we may have had limited power to observe an effect of immune status on HPV incidence or clearance. Second, in this kind of epidemiologic studies it is impossible to distinguish true HPV clearance from viral latency, and to distinguish true incident infection from re-activation of latent infection. As viral latency may be more common in HIV-infected individuals , this may partly explain the observed increased HPV incidence rates. Third, no clinical examinations were performed, and therefore we could not link virological endpoints to clinical lesions. Fourth, our cohort consists of highly sexually active, adult MSM. Therefore, our results may not be generalizable to all MSM.
In conclusion, anal and penile hrHPV incidence rates were higher among HIV-infected compared with HIV-negative MSM, and anal hrHPV clearance rates were lower, independent of sexual behavior. This suggests a biological effect of HIV infection on anogenital hrHPV infection. Further studies are needed into the mechanisms by which HIV exerts its effect on HPV infection and sequelae.
The authors would like to thank the members of the H2M steering committee for their advice; the personnel of the Amsterdam Cohort Studies, M.C. Jan van Goyen, and the Amsterdam STI clinic for their contributions to the implementation of the study and data collection; Martijn van Rooijen and Linda May for assisting in data management; Wilma Vermeulen (Public Health Service of Amsterdam) for HPV genotyping; Jan Sonsma and Marina Burger (RIVM) for DNA isolation, and Stichting HIV Monitoring (SHM) for their contribution in data collection. Above all, we gratefully acknowledge all study participants for their co-operation.
Sources of funding: This work was supported by Aids Fonds [grant number 2009029], and additional funding was provided by the Public Health Service of Amsterdam and RIVM. The Amsterdam Cohort Studies on HIV infection and AIDS, a collaboration between the Public Health Service of Amsterdam, the Academic Medical Center of the University of Amsterdam, Sanquin Blood Supply Foundation, and the University Medical Center Utrecht, are part of the Netherlands HIV Monitoring Foundation and financially supported by the Netherlands National Institute for Public Health and the Environment.
Author contributions: M.S.vdL. designed and leads the H2M study as principal investigator. M.vdS., R.C., I.S., H.dV., A.S., P.S., C.M., and S.M. contributed to the study design. A.S., H.dV., A.vE., and S.M. contributed to the implementation of the study and data collection. A.S., A.K., and S.M. were responsible for laboratory analyses. D.vS. performed the statistical analyses and interpreted the data together with S.M., R.G., and M.S.vdL. S.M. wrote a draft manuscript, which was reviewed by all authors for important intellectual content. All authors have seen and approved the final version of the manuscript.
Requests for reprints: Maarten Schim van der Loeff, MD, PhD, Health Service of Amsterdam (GGD), Infectious Diseases, Department of Research, Nieuwe Achtergracht 100, 1018 WT Amsterdam, The Netherlands. Phone: +31-20-555 5083; Fax: +31-20-555 5533; E-mail: [email protected]
Conflicts of interest
Conflicts of interest: M.S.vdL. received funding for a substudy on HPV from Sanofi Pasteur MSD and is a co-investigator in a Merck-funded investigator-initiated study on Gardasil. M.S.vdL., H.dV., and C.M. are investigators on a Sanofi-Pasteur-sponsored trial of a nonavalent HPV vaccine in heterosexual men. For the remaining authors no conflicts of interest were declared.
1. Bouvard V, Baan R, Straif K, Grosse Y, Secretan B, El Ghissassi F, et al. A review of human carcinogens–Part B: biological agents
. Lancet Oncol
2. Giuliano AR, Tortolero-Luna G, Ferrer E, Burchell AN, de Sanjosé S, Kjaer SK, et al. Epidemiology of human papillomavirus infection in men, cancers other than cervical and benign conditions
3. 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
4. Dillner J, von Krogh G, Horenblas S, Meijer CJ. Etiology of squamous cell carcinoma of the penis
. Scand J Urol Nephrol Suppl
5. Hoots BE, Palefsky JM, Pimenta JM, Smith JS. Human papillomavirus type distribution in anal cancer and anal intraepithelial lesions
. Int J Cancer
6. 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
7. Johnson LG, Madeleine MM, Newcomer LM, Schwartz SM, Daling JR. Anal cancer incidence and survival: the surveillance, epidemiology, and end results experience, 1973–2000
8. Silverberg MJ, Lau B, Justice AC, Engels E, John Gill M, Goedert JJ, et al. Risk of anal cancer in HIV-infected and HIV-uninfected individuals in North America
. Clin Infect Dis
9. Dunne EF, Nielson CM, Stone KM, Markowitz LE, Giuliano AR. Prevalence of HPV infection among men: a systematic review of the literature
. J Infect Dis
10. Smith JS, Gilbert PA, Melendy A, Rana RK, Pimenta JM. Age-specific prevalence of human papillomavirus infection in males: a global review
. J Adolesc Health
11. Del Amo J, González C, Geskus RB, Torres M, Del Romero J, Vinciana P, et al. What drives the number of high risk human papillomavirus types in the anal canal in HIV-positive men who have sex with men?
. J Infect Dis
12. Silverberg MJ, Chao C, Leyden WA, Xu L, Tang B, Horberg MA, et al. HIV infection and the risk of cancers with and without a known infectious cause
13. Van Aar F, Mooij SH, van der Sande MA, Speksnijder AG, Stolte IG, Meijer CJ, et al. Anal and penile high-risk human papillomavirus prevalence in HIV-negative and HIV-infected MSM
14. Molijn A, Kleter B, Quint W, Van Doorn LJ. Molecular diagnosis of human papillomavirus (HPV) infections
. J Clin Virol
15. Van Buuren S, Groothuis-Oudshoorn K. MICE: Multivariate imputation by chained equations in R
. J Stat Softw
16. Xue X, Gange SJ, Zhong Y, Burk RD, Minkoff H, Massad LS, et al. Marginal and mixed-effects models in the analysis of human papillomavirus natural history data
. Cancer Epidemiol Biomarkers Prev
17. Schim van der Loeff MF, Mooij SH, Richel O, de Vries HJ, Prins JM. HPV and anal cancer in HIV-infected individuals
. Curr HIV/AIDS Rep
18. Hastie TJ (1992) Generalized additive models. In: Chambers JM, Hastie TJ editors. Statistical models in S
. Pacific Grove, CA: Wadsworth & Brooks/Cole.
19. R Development Core Team. R: A language and environment for statistical computing
. Vienna, Austria: R foundation for statistical computing, 2011.
20. Glick SN, Feng Q, Popov V, Koutsky LA, Golden MR. High rates of incident and prevalent anal human papillomavirus infection among young men who have sex with men
. J Infect Dis
21. Phanuphak N, Teeratakulpisarn N, Pankam T, Kerr SJ, Barisri J, Deesua A, et al. Anal human papillomavirus infection among Thai men who have sex with men with and without HIV infection: prevalence, incidence, and persistence
. J Acquir Immune Defic Syndr
22. 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
23. de Pokomandy A, Rouleau D, Ghattas G, Vezina S, Coté 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
24. Hernandez AL, Efird JT, Holly EA, Berry JM, Jay N, Palefsky JM. Incidence of and risk factors for type-specific anal human papillomavirus infection among HIV-positive MSM
25. Giuliano AR, Lee JH, Fulp W, Villa LL, Lazcano E, Papenfuss MR, et al. Incidence and clearance of genital human papillomavirus infection in men (HIM): a cohort study
26. Mbulawa ZZ, Marais DJ, Johnson LF, Coetzee D, Williamson AL. Impact of human immunodeficiency virus on the natural history of human papillomavirus genital infection in South African men and women
. J Infect Dis
27. De Vuyst H, Lillo F, Broutet N, Smith JS. HIV, human papillomavirus, and cervical neoplasia and cancer in the era of highly active antiretroviral therapy
. Eur J Cancer Prev
28. Beachler DC, D'Souza G, Sugar EA, Xiao W, Gillison ML. Natural history of anal versus oral HPV infection in HIV-infected men and women
. J Infect Dis
29. Gravitt PE. The known unknowns of HPV natural history
. J Clin Invest