There are >100 different types of human papillomaviruses (HPV) that can infect either the cutaneous or mucosal epithelium of humans. Of these, the mucosal type HPV16 has been causally associated with approximately 60% of cervical cancers,1 and 70% of anal cancers and anal precursor lesions, such as anal intraepithelial neoplasia.2 There is evidence that anal cancer rates are increasing globally3 and that homosexual men are particularly at high risk of developing this malignancy.4 Those men who are HIV infected are at even higher risk, and incidence rates have not declined since the advent of highly active antiretroviral therapy.5 Anal cancer is now the most common non–AIDS-defining cancer in HIV-infected people in Australia.6
In contrast to the extensively researched natural history of the progression from cervical HPV infection to cervical malignancy, there is limited knowledge of the epidemiology and natural history of anal HPV infection. Few studies have prospectively examined the frequency of acquisition and duration of anal HPV infection in homosexual men,7,8 and data on HPV infection in older men, in whom rates of anal cancer are highest, are extremely limited. Most of the published research on HPV-related disease in homosexual men has reported on the prevalence of anal HPV infection by DNA detection.9,10 There have been only a few studies of HPV seroprevalence in homosexual men.11 – 15 Serological studies allow an estimate of lifetime cumulative exposure to HPV. In one study, the sensitivity of serology was estimated to be approximately 60% when measured against cervical DNA detection.16
Here, we explore prevalence, incidence, and risk factors for HPV16 seropositivity in Australian homosexual men.
MATERIALS AND METHODS
Two prospective community-based cohort studies of homosexual men were conducted in Sydney, Australia. The Health in Men (HIM) study involved HIV-negative homosexually active men. The methodology for the HIM study has been published previously.17 Men who were 18 years or older, tested HIV negative, and had sexual contact with at least one man in the past 5 years were eligible to enroll. The study recruited 1427 participants from June 2001 to December 2004 and concluded interviews in June 2007. The Positive Health (pH) study recruited 778 HIV-positive men from 1998 to 2006, who reported sex with at least one man during the previous 5 years or who identified as homosexual, using similar community-based methods as the HIM study.18 Final interviews were conducted in 2007. The HIM and pH studies and the serological research within the 2 studies received ethical approval from the University of New South Wales Human Research Ethics committee.
Participants underwent annual structured face-to-face interviews on a wide range of topics, including sexual relationships and practices. Quantitative sexual behavior data were collected, with number of partners and number of episodes of unprotected anal intercourse (UAI) recorded. At approximately 6 months between annual face-to-face interviews, information on recent sexual relationships and practices and injecting drug use was collected from HIV-negative participants via a short-version telephone interview. In the HIM study, serological testing for HIV and storage of serum samples were undertaken annually among consenting participants. In the pH study, annual serological testing was only performed on consenting participants in the period 2005–2007.
Serum Sample Collection, Shipment, and Testing
Serum samples were stored at −80°C and shipped on dry ice to the German Cancer Research Center (DFKZ), Heidelberg, Germany, for serology analysis. Specimens were labeled with a unique identifier, and all serum analyses were performed blindly. The samples were analyzed for antibodies to the capsid (L1) proteins of the mucosal α HPV types 16, as previously described.19
All samples were analyzed for HPV antibodies by multiplex serology, an antibody detection method based on glutathione S-transferase (GST) capture ELISA20,21 in combination with fluorescent bead technology.19,22 Briefly, full-length viral proteins were expressed in Escherichia coli bacteria in fusion with an N-terminal GST domain. Glutathione cross-linked to casein was coupled to fluorescence-labeled polystyrene beads (SeroMap, Luminex Corp., Austin, TX,) and GST fusion proteins were affinity purified on the beads. Each fusion protein was bound to a spectrally distinct bead set, and fusion protein-loaded bead sets were mixed. Sera were incubated with the mixed bead sets at a final dilution of 1:100, and bound antibodies were detected with biotinylated goat anti-human IgG secondary antibody and streptavidin-R-phycoerythrin. A Luminex analyzer (Luminex Corp) was used to identify the internal color of the individual beads and to quantify their reporter fluorescence (expressed as median fluorescence intensity (MFI) of at least 100 beads per set per serum).
Glutathione-casein coupled bead sets were loaded with their respective antigens in 1 batch. Study sera were analyzed once on 1 of 4 consecutive assay days. A quality control panel of 188 samples was included each day on each run, made possible by the Luminex high-throughput method. This resulted in 4 quality control data sets to determine interday variation. Autofluorescence of each bead set and background reactions resulting from binding of secondary reagents to the antigen-loaded beads were determined in 1-well per plate without human serum. Mean bead background values were subtracted from raw MFI values and antigen-specific reactivity was determined by subtraction of the MFI of GST-tag from the MFI of the specific antigen.
Cut-off points to determine antibody positivity were selected by visual inspection of the distribution of MFI values among all study participants.23,24 Seropositivity was defined as having an MFI value of 400 or more. Seroconversion was defined by the cooccurrence of 2 conditions: (1) at least 2-fold increase of the previous serum MFI value and (2) MFI value above the cut-off point.25
HPV16 seroprevalence and seroincidence per 100 person-years (PYs) in those initially HPV16-seronegative participants were calculated according to each participant's HIV status. Seroincidence was calculated by the number of type-specific seroconversions divided by the total PYs followed. The exact binomial method was used to calculate 95% confidence intervals (CIs).
In the pH study, limited data were collected on lifetime sexual risk behavioral factors; therefore, risk factors for HPV16 seropositivity were only examined in HIM participants. Predictors of seroprevalent HPV16 were analyzed by logistic regression. Sociodemographic, biologic, and behavioral characteristics and their relation to HPV16 seroprevalence were examined. These included age, occupation, education, circumcision status, level of gay community involvement, lifetime number of male sexual partners, lifetime number of female sexual partners, duration since first sexual intercourse, duration since first anal sexual intercourse, cigarette smoking, marijuana use, history of syphilis, serological evidence of past infection with hepatitis A, B, and C, herpes simplex virus type 1 (HSV-1) and HSV-2, history of anal warts, and history of genital warts.
Predictors of seroincident HPV16 were analyzed by Cox regression among initially HPV16-seronegative HIM participants. Similar to the analysis of seroprevalence, sociodemographic, biologic, and behavioral characteristics and their relation to HPV16 seroincidence were examined. These included age, occupation, education, circumcision status, level of gay community involvement, number of male sexual partners in the past 6 months, cigarette smoking, marijuana use, UAI by partner type (regular or casual), UAI by anal intercourse position (insertive or receptive), UAI by partner's HIV status (positive, negative, or unknown), receptive oral sex with casual partners, insertive oral sex with casual partners, receptive rimming, receptive fingering, receptive fisting, urethral gonorrhoea and chlamydia, pharyngeal gonorrhoea, anal gonorrhoea and chlamydia, past syphilis infection, incident hepatitis A, B, and C, incident HSV-1, incident HSV-2, prevalent and incident anal warts, and prevalent and incident genital warts.
Multivariate models were developed using a backwards selection method. A significance threshold of a P value of 0.10 or less in the univariate models was used to identify variables to be included in the final multivariate models. Odds ratios (ORs) and hazard ratios (HRs) with corresponding 95% CIs were calculated for seroprevalent and seroincident HPV16 risk factors, respectively.
We performed further analyses exploring the association between circumcision status and HPV seroprevalence and seroincidence in all men and in those men who reported a preference for the insertive role in anal intercourse, using methodology we had previously developed for analyses of the association between sexually transmitted infection (STI)/HIV incidence and circumcision.26
Data analyses were performed using Stata 10.0 (Stata Corporation, College Station, TX).
A total of 1427 initially HIV-negative men were enrolled in the HIM study, and 5959 serological samples were available for analysis. The median age at enrollment was 35 years (range, 18–75 years), and men were followed for a median of 3.9 years, with a total follow-up time of 4538 PYs. In the pH study, the median age at enrollment (45 years; range, 22–71 years) was a decade more than in the HIM study. From the total of 245 men who had sera collected at least once, 394 serological samples were stored.
A quarter (25.4%, 95% CI: 23.1–27.9) of the HIV-negative participants were HPV16 seropositive at their baseline visit compared with nearly half of the HIV-positive participants (44.3%, 95% CI: 37.4–51.5; Table 1). The seroprevalence in HIV-positive men was significantly higher than in HIV-negative men (OR = 2.18, 95% CI: 1.73–2.75), which remained significant after controlling for age (adjusted OR = 1.88, 95% CI: 1.49–2.38). HPV16 seroincidence among HIV-negative men was 3.1/100 PYs and 95% CI was 2.5 to 3.8/100 PYs. HPV16 seroincidence among HIV-positive men was 1.3/100 PYs and 95% CI was 0.3 to 5.2/100 PYs. This finding was limited by the small number of HIV-positive men with a serological sample at more than 1 time point.
Predictors of HPV16 Seroprevalence
In univariate analyses of data among HIV-negative men, HPV16 seroprevalence was associated with older age (P < 0.001), a longer history of any sexual activity (P < 0.001) including anal sex (P < 0.001), and higher numbers of sexual partners (P < 0.001) (Table 2). Men with a history of anal warts or serological evidence of a number of STIs were also much more likely to be HPV16 seropositive at baseline, as were men who reported a preference for and predominantly practiced the receptive position in anal sex.26 Circumcision was not associated with seroprevalent HPV16 (P = 0.26). In multivariate analyses, seroprevalent HPV16 was associated with a higher lifetime number of male partners (P = 0.002), longer duration since first anal sex (P < 0.001), preference for the receptive role during anal sex (P = 0.032), a history of anal warts (P = 0.022), and serological evidence of past infection with hepatitis B virus (P = 0.035), HSV-1 (P = 0.034), and HSV-2 (P = 0.042).
Predictors of HPV16 Seroincidence
In contrast to HPV16 seroprevalence, HPV16 seroincidence declined with age from 4.1/100 PYs in those aged less than 25 years to less than 2/100 PYs in those aged more than 45 (P = 0.002) (Table 3). Behavioral predictors of HPV16 seroconversion included reporting higher numbers of male partners (P < 0.001), UAI with both casual and regular partners (P = 0.001) and with partners of unknown or positive HIV status (P = 0.002). Receptive anal sexual practices were also associated with seroincident HPV16, including receptive rimming (P = 0.001), fingering (P = 0.053), and fisting (P = 0.023). Several bacterial STIs were significantly associated with HPV16 seroincidence in the univariate model, although there was no association with any viral STIs. Overall, circumcision was not significantly associated with HPV seroincidence (HR = 0.70, 95% CI: 0.46–1.06, P = 0.09).
In multivariate analyses, seroincident HPV16 remained significantly associated with younger age (P = 0.049) and UAI with HIV-positive partners or partners of unknown HIV status (P = 0.046). Although not significant, there was a trend toward participants with higher numbers of recent male sexual partners (P = 0.089) being more likely to seroconvert (Table 3).
Circumcision and HPV16 Seroprevalence and Seroincidence
Further analysis was undertaken on the impact of circumcision on seroprevalent and seroincident HPV16 among men who predominantly practiced the insertive position in anal intercourse. Among these men (1109 PYs, 25% of total PYs), there was no relationship between being circumcised and HPV16 seropositivity at baseline (OR = 0.98, 95% CI: 0.60–1.61, P = 0.94), but there was a 57% reduction in seroincident HPV16 (HR = 0.43, 95% CI: 0.21–0.88, P = 0.021). When adjusted for potential confounders (number of sexual partners in the past 6 months, UAI by partner's HIV status, and age), the effect remained significant (HR = 0.47, 95% CI: 0.28–0.98, P = 0.043).
In these community-based cohorts of Australian homosexual men, we examined the seroepidemiology of HPV16. HPV16 seropositivity was common and was substantially higher among HIV-positive (44%) than HIV-negative (25%) men. Overall, about 3% of HIV-negative men seroconverted each year. Among HIV-negative men, both HPV16 seroprevalence and seroincidence correlated well with markers of sexual activity, and particularly with receptive anal sexual practices. An association between circumcision and decreased HPV16 seroconversion in those HIV-negative men who predominantly practiced the insertive role in anal sex was observed.
The high rates of HPV16 seroprevalence are comparable with findings from other serological studies involving homosexual men. In a Dutch population-based cross-sectional survey, men who reported sexual attraction to men had an HPV16 seroprevalence of 31.4% compared with 16.8% among heterosexual men.13 Similarly, in US cross-sectional national surveys, HPV16 seroprevalence was 37.7% among men who reported ever having had sex with a man,15 and it was 30% among 169 Austrian HIV-positive men, 59 of whom were homosexual.11 Unlike our findings, other studies have not found a difference in seroprevalence between HIV-positive and HIV-negative men. For instance, Hagensee et al reported a 48% HPV16 seroprevalence among 101 homosexual or bisexual HIV-negative men compared with 42% among 154 HIV-positive men in a serosurvey conducted in Seattle, WA.14
The only other Australian study to describe HPV16 seroprevalence among men found an overall seroprevalence rate of 7.9% in the male general population older than 15 years.27 Similar findings have been reported in a number of US studies involving heterosexual men. HPV16 seropositivity rates of 14.8% of 285 HIV-negative men recruited from mainly through community sources in Arizona in the HPV Infection in Men study,8 14.1% of 340 men in a community- and clinic-based cross-sectional study in Baltimore,28 and 18.7% of 786 HIV-negative STD clinic attendees7 provide consistent evidence that heterosexual men have lower HPV16 seroprevalence than homosexual men. In contrast, homosexual men have similar rates of HPV16 seropositivity to women (29.1%,13 30.2%,7 and 42.3%28) in a number of the studies discussed earlier. This higher seroprevalence among women and homosexual men is thought to be due to HPV infection via mucosal epithelium (vagina and anus) rather than keratinized epithelium (penis). It has been hypothesized that HPV infection of mucosal epithelium may be more likely to induce a humoral immune response than HPV infection of keratinized epithelium.7,12,15,28
Importantly, men who predominantly practiced the receptive position in anal sex were also more likely to have seroprevalent HPV16 (P = 0.014), further strengthening the theory that mucosal exposure leads to greater induction of a systemic antibody response.7,12 Similar risk factors of older age,7,8,12,14 unprotected sex with casual partners,7 lifetime number of male anal sex partners,12,14 and a history of other STIs,12 in particular HSV-2,14,28 have been previously associated with higher HPV16 seroprevalence.
Recently published studies have demonstrated a reduction of incident genital HPV detection among circumcised heterosexual HIV-negative and HIV-positive men.29,30 An important finding of our study was an observed decrease in HPV16 seroincidence among circumcised men who reported predominant practice of the insertive rather than the receptive position in anal sex. Circumcised men were approximately 60% less likely to seroconvert than uncircumcised men. Anal intercourse position preference was reported by 1211 HIM participants at least twice at consecutive annual interviews. Only 2.5% of participants reported a change from their initial preference for the insertive position at one or more subsequent interviews to either receptive or no preference. This indicates that sexual position preferences were constant and reflected actual behavior. Other studies have not found an association between circumcision and HPV16 seroprevalence7,12 or seroincidence.7
The HIM and pH studies have the strength of being large-scale prospective cohort studies. The samples were primarily community-based, and we believe that they can be considered as broadly representative of homosexual men in Sydney. This study measures HPV16 seroincidence in homosexual men and examines the effect of circumcision by anal sexual practices. Both are areas with limited published data available.
There were several limitations in this research. Multiplex serology was used here, a serological method that has been used in numerous other studies. Some assays use virus-like particles as antigens, whereas in this research, we used GST-L1-based assays. Cut-off definitions for positivity vary between laboratories. These differences in serological methods may limit direct comparisons with other studies. HPV seroconversion can occur up to 18 months after HPV DNA detection,22 and thus it is possible that some seroincident cases may represent infection before study entry. Although HPV serology is believed to be specific, it is only a moderately sensitive measure of infection. Therefore, with estimated seroconversion rates of 60%, not all those infected with HPV will develop antibodies.16 The true HPV infection rate is potentially underestimated, which may lead to misclassification bias.13,16 Exposure misclassification may also occur due to waning of antibodies over time8 and due to varying immune response based on exposure via mucosal rather than squamous epithelium.
These results clearly demonstrate that HPV16 seropositivity is common among both HIV-negative and HIV-positive homosexual men. Both HPV16 seroprevalence and seroincidence correlated well with other markers of high-risk sexual activity, particularly receptive anal sexual practices. Circumcision was associated with a reduction in HPV seroincidence in men who were mainly exposed via the penis rather than the anus. Because HPV16 seroincidence declined only gradually with age, our data suggest that there may be a role for prophylactic vaccination of young gay men.
1. de Sanjose S, Quint W, Alemany L, et al.. Human papillomavirus genotype attribution in invasive cervical cancer: A retrospective cross-sectional worldwide study. Lancet Oncol 2010; 11:1048–1056.
2. De Vuyst H, Clifford G, Nascimento M, et al.. 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; 124:1626–1636.
3. Johnson L, Madeleine M, Newcomer L, et al.. Anal cancer incidence and survival: The surveillance, epidemiology, and end results experience, 1973–2000. Cancer 2004; 101:281–288.
4. Daling J, Weiss N, Hislop T, et al.. Sexual practices, sexually transmitted diseases, and the incidence of anal cancer. N Engl J Med 1987; 317:973–977.
5. Van Leeuwen M, Vajdic C, Middleton M, et al.. Continuing declines in some but not all HIV-associated cancers in Australia after widespread use of antiretroviral therapy. AIDS 2009; 23:2183–2190.
6. Australian Institute of Health and Welfare & Australian Association of Cancer Registries 2010; Vol 60: CAN56. Cancer in Australia: an overview 2010. Canberra.
7. Thompson D, Douglas JM, Foster M, et al.. Seroepidemiology of infection with human papillomavirus 16, in men and women attending sexually transmitted disease clinics in the United States. J Infect Dis 2004; 190:1563–1574.
8. Lu B, Hagensee M, Lee J, et al.. Epidemiologic factors associated with seropositivity to human papillomavirus type 16 and 18 virus–like particles and risk of subsequent infection in men. Cancer Epidemiol Biomarkers Prev 2010; 19:511–516.
9. Dunne E, Nielson C, Stone K, et al.. Prevalence of HPV infection among men: A systematic review of the literature. J Infect Dis 2006; 194:1044–1057.
10. van der Snoek E, Niesters H, Mulder P, et al.. Human papillomavirus infection in men who have sex with men participating in a Dutch gay-cohort study. Sex Transm Dis 2003; 30:639–644.
11. Höpfl R, Petter A, Thaler P, et al.. High prevalence of high risk human papillomavirus-capsid antibodies in human immunodeficiency virus-seropositive men: A serological study. BMC Infect Dis 2003; 3.
12. Lu B, Viscidi R, Lee J, et al.. Human papillomavirus (HPV) 6, 11, 16 and 18 seroprevalence is associated with sexual practice and age: Results from the Multi-National HPV Infection in Men Study (HIM Study). Cancer Epidemiol Biomarkers Prev 2011; 20:990–1002.
13. Heiligenberg M, Michael K, Kramer M, et al.. Seroprevalence and determinants of eight high-risk human papillomavirus types in homosexual men, heterosexual men, and women: A population-based study in Amsterdam. Sex Transm Dis 2010; 37:672–680.
14. Hagensee M, Kiviat N, Critchlow C, et al.. Seroprevalence of human papillomavirus types 6 and 16 capsid antibodies in homosexual men. J Infect Dis 1997; 176:625–631.
15. Stone K, Karem K, Sternberg M, et al.. Seroprevalence of human papillomavirus type 16 infection in the United States. J Infect Dis 2002; 186:1396–1402.
16. Carter J, Koutsky L, Hughes J, et al.. Comparison of human papillomavirus types 16, 18, and 6 capsid antibody responses following incident infection. J Infect Dis 2000; 181:1911–1919.
17. Jin F, Prestage G, Mao L, et al.. Transmission of herpes simplex virus types 1 and 2 in a prospective cohort of HIV-negative gay men: The Health in Men study. J Infect Dis 2006; 194:561–570.
18. Prestage G, Mao L, Fogarty A, et al.. How has the sexual behaviour of gay men changed since the onset of AIDS: 1986–2003. Aust N Z J Public Health 2005; 29:530–535.
19. Waterboer T, Sehr P, Michael K, et al.. Multiplex human papillomavirus serology based on in situ–purified glutathione S
-Transferase Fusion Proteins. Clin Chem 2005; 51:1845–1853.
20. Sehr P, Muller M, Hopfl R, et al.. HPV antibody detection by ELISA with capsid protein L1 fused to glutathione S
-transferase. J Virol Methods 2002; 106:61–70.
21. Sehr P, Zumbach K, Pawlita M. A generic capture ELISA for recombinant proteins fused to glutathione S
-transferase: Validation for HPV serology. J Immunol Methods 2001; 254:153–162.
22. Waterboer T, Sehr P, Pawlita M. Suppression of non-specific binding in serological Luminex assays. J Immunol Methods 2006; 309:200–204.
23. Carter J, Paulson K, Wipf G, et al.. Association of Merkel cell polyomavirus-specific antibodies with Merkel cell carcinoma. J Natl Cancer Inst 2009; 101:1510–1522.
24. Herrero R, Castellsagué X, Pawlita M, et al.. Human papillomavirus and oral cancer: The International Agency for Research on Cancer Multicenter Study. J Natl Cancer Inst 2003; 95:1772–1783.
25. Syrjanen S, Waterboer T, Sarkola M, et al.. Dynamics of human papillomavirus serology in women followed up for 36 months after pregnancy. J Gen Virol 2009; 90:1515–1526.
26. Templeton D, Jin F, Prestage G, et al.. Circumcision and risk of sexually transmissible infections in a community-based cohort of HIV negative homosexual men in Sydney, Australia. J Infect Dis 2009; 200:1813–1819.
27. Newall A, Brotherton J, Quinn H, et al.. Population seroprevalence of human papillomavirus types 6, 11, 16, and 18 in men, women and children in Australia. Clin Infect Dis 2008; 46:1647–1655.
28. Kreimer A, Alberg A, Viscidi R, et al.. Gender differences in sexual biomarkers and behaviors associated with human papillomavirus-16, -18 and -33 seroprevalence. Sex Transm Dis 2004; 31:247–256.
29. Tobian A, Serwadda D, Quinn T, et al.. Male circumcision for the prevention of HSV-2 and HPV infections and syphilis. N Engl J Med 2009; 360:1298–1309.
30. Gray R, Serwadda D, Kong X, et al.. Male circumcision decreases acquisition and increases clearance of high-risk human papillomavirus in HIV-negative men: A randomized trial in Rakai, Uganda. J Infect Dis 2010; 201:1455–1462.