Objectives Anal human papillomavirus (HPV) is highly prevalent in men who have sex with men (MSM) of all ages, whereas cervical HPV declines with age. We explore the hypothesis that different sexual behavioural patterns are the basis of this difference in age distribution.
Methods Published data on age-specific HPV prevalence for women (cervical HPV) were extracted from a large meta-analysis and for MSM (anal HPV) from the EXPLORE study of HIV-negative MSM. Age-specific data on recent sexual activity were extracted from two behavioural surveys: the second Australian Study of Health and Relationships survey and the 2013 Gay Community Periodic Survey.
Results At least 50% of MSM at all ages reported more than one sexual partner in the past 6 months. In comparison, 33% of women aged 16–19 years reported more than one partner over the past year. This decreased to 19% and 6% in women aged 20–29 and 30–39 years, respectively, and to fewer than 5% of women in older age groups. Prevalent anal HPV was detected in over 50% of MSM in each age group. Prevalence did not decline with age. In contrast, there was a steady decrease in cervical HPV prevalence with age. Cervical HPV prevalence fell from 23% among North American women aged <25 years to 3% in women aged ≥65 years.
Conclusions In contrast to the decreasing prevalence with age among heterosexual women, the high prevalence and lack of decline in prevalent anal HPV among older MSM are likely to be related to continuing high rates of newly acquired HPV infection from ongoing sexual exposure through new partners.
- ANOGENITAL CANCER
- BEHAVIOURAL SCIENCE
- GAY MEN
Statistics from Altmetric.com
Infection with high risk (HR) types of human papillomavirus (HPV) causes almost all cervical and over 90% of anal cancers.1 Although rare in the general population,w1 squamous cell cancer of the anal canal (ASSC) is more common in some specific population groups. Incidence is highest among men who have sex with men (MSM). HIV-positive MSM are at the highest risk, with rate ratios of almost 80 compared with rates among HIV-negative men (rate ratio 78.8, 95% CI 40.8 to 152.1).2 w2 High-grade squamous intraepithelial lesions (HSIL), the presumed anal squamous cell cancer (ASCC) precursor lesions, are highly prevalent among MSM.3 Screening programmes to detect these anal precursor lesions, based on analogy to screening programmes for cervical cancer, have been proposed.w3
Not surprisingly, anal HPV infection, the necessary cause of ASSC, is also highly prevalent among MSM. The HPV in men (HIM) study found that MSM had a markedly higher prevalence of any anal HPV detection compared with heterosexual men (47.2%, 95% CI 39.6 to 54.8 vs 12.2%, 95% CI 10.5 to 14.1).w4 A systematic review and meta-analysis of anal HPV among MSM demonstrated that the detection of any HPV type and any HRHPV type were very common among MSM, with pooled estimates of prevalence of any HPV type for HIV-positive men of 92.6% (95% CI 90.8 to 94.5) and any HRHPV 73.5% (95% CI 63.9 to 83.0) and of any HPV type for HIV-negative men of 63.9% (95% CI 55.2 to 72.6) and any HRHPV 37.2% (95% CI 27.4 to 47.0).3
With cervical HPV, prevalence peaks in young women and then declines dramatically with age (figure 1).w5 A meta-analysis of 194 studies in 59 countries included over a million women with normal cytological findings who had been tested for cervical HPV. The overall prevalence of cervical HPV fell threefold from 24% in women aged <25 years to 7.5% in women aged >55 years.4 Similar patterns in age distribution have been seen with cervical HPV incidence. Among 24 105 women participating in screening trials, the incidence of HRHPV detection decreased from 5.7% per year in women aged 29–33 years to 1.0% per year in women aged 54–56 years in the Netherlands and from 8.7% per year in women aged 24–28 years to 0.29% per year in women aged 57–60 years in Italy.w6
Substantial age-related declines in HPV prevalence are not seen in the most of the available epidemiological studies among MSM. In the EXPLORE study of 1218 HIV-negative MSM recruited from four US cities, there was no change in anal HPV prevalence by age.5 No significant association between age group and anal HPV prevalence was found in 258 HIV-negative MSM recruited in Italy between August 2009 and November 2010w7 or in a 2005 Australian study of 316 MSM.w8 However, the HIM study discussed above found a decrease in prevalence with increase in age for both any HPV and oncogenic types among MSM (p=0.001 for both).w4
Various explanations have been proposed for the lack of a decline in prevalent anal HPV with age in MSM. These include biological reasons such as impaired clearance of HPV in the anal canal compared to the cervix, age-related decreases in immune responses, the impact of hormones on cervical tissue and other organ-specific differences.5 Higher rates of newly acquired infections among older MSM, related to behavioural differences, have also been proposed.5 A considerable proportion of MSM continue to be sexually active into their later years.6 In this report, we use published data on HPV prevalence and data from two large behavioural surveys to explore the hypothesis that different sexual behavioural patterns among older heterosexual women and MSM are the basis of the difference between the age-related patterns of cervical and anal HPV detection.
Data on sexual behaviour for heterosexual women were taken from the second Australian Study of Health and Relationships survey (ASHR2), undertaken in 2012–2013. A nationally representative sample of over 20 000 respondents underwent computer-assisted telephone interviews, with a participation rate of 66.2%.7 Data from MSM were extracted from the 2013 Gay Community Periodic Survey (GCPS). The GCPS are cross-sectional surveys conducted annually in most Australian states and territories. Men are approached at gay community venues and events and invited to self-complete an anonymous questionnaire. Approximately 9000 men participate each year and the response rate is typically 70%.8 The ASHR2 protocol was approved by the Human Research Ethics Committees (HREC) of La Trobe University (HEC 11–040) and the GCPS protocol was approved by the University of New South Wales HREC (HC13366).
As it has been shown that HPV may be acquired through non-intercourse behaviours,w9 the measure of sexual behaviour used was the proportion of respondents reporting more than one male sexual partner, rather than any specific anal or vaginal sex practices. This was measured numbers of male partners in the past year (ASHR2) or 6 months (GCPS), as determined by question design in each survey. This was examined by age group (16–19, 20–29, 30–39, 40–49, 50–59 and 60–69 years).w10 A new analysis of the GCPS responses was undertaken, whereas published ASHR2 data for women who reported heterosexual activity in the last year were used.7 HPV prevalence by age for women (cervical HPV) was extracted from the large meta-analysis described above4 and for men (anal HPV) from the EXPLORE study.5 It is important to note that the median age in the EXPLORE study was 37 years (IQR 31–43) with 80% >30 years, and there were limited data on younger MSM.5
Fifty per cent of MSM aged 16–19 years reported more than one sexual partner in the past 6 months. This increased to over 60% in all the other age groups, remaining at 62% in the oldest group, aged 60–69 years. High prevalence of receptive anal sex continued to be reported into older age in MSM.w11 In comparison, 33% of women aged 16–19 years reported more than one partner over the past year. This decreased to 19% and 6% in women aged 20–29 and 30–39 years, respectively. In older age groups, fewer than 5% of women reported more than one sexual partner in the last 12 months. This was as low as 1.4% in women aged 60–69 years (figure 1).
In MSM in the EXPLORE study, prevalent anal HPV was detected in over 50% of HIV-negative MSM in each age group. Prevalence did not decline with age. In contrast, results of the meta-analysis of women showed there was a steady decrease in cervical HPV prevalence with age.4 This was particularly evident among women from North America, where cervical HPV prevalence fell from 23% among women aged <25 years to 2.7% in women aged ≥65 years (figure 1).
We believe that the clear association between anal and cervical HPV detection and reported numbers of recent sexual partners provides strong evidence that sexual behaviour plays a major role in the rapid decline in cervical HPV prevalence with age, and in the persistently high anal HPV prevalence in MSM, regardless of age.
Data for heterosexual women were derived from a probability sample survey.7 Probability sample surveys tend to yield small numbers of MSM and thus the MSM data are from a community survey (GCPS).8 Differences in sexual behaviour may be due to differences in the participants sampled. It is also important to note that the GCPS data are collected for sexual partners in the last 6 months, and thus the comparison may be an underestimate of the difference between heterosexual women and MSM.
Age-specific prevalence of anal HPV has been examined in a handful of studies of healthy immunocompetent women. In the Hawaii HPV Cohort Study, the prevalence of concurrent cervical and anal HPV detection decreased with increasing age from 25% among women aged 18–24 years, to <5% among those aged ≥60 years. The same pattern was observed in women with cervical HPV without accompanying anal HPV infection.w12 In contrast, among women with detectable anal HPV only (without concurrent cervical infection) the prevalence of anal HPV remained constant in all age groups.w13 Current practice of anal sex significantly reduced the chances of anal HPV clearance in this study.w14 Taken together, these data suggest that differences in HPV profiles between the cervix and the anus may be largely driven by behaviour, rather than biological differences between the two anatomical sites.
If recent sexual behaviour drives high anal HPV prevalence, a large proportion of detected HRHPV infections in adult MSM are likely to be recent. This is analogous to the situation in female adolescents, who often report multiple partners and acquire HPV shortly after initiating sexual intercourse, and where the vast majority of cervical HPV infections are transient. Half of those who clear HPV infection will have a repeat HPV infection within 3 years.w15 In addition, most cervical intraepithelial neoplasia (CIN) grade 1 lesions and up to 60% of CIN grade 2 lesions regress within several years.9w16 Many experts believe this epidemiological evidence strongly supports observation over intervention for adolescents with CIN 1 and 2.10 US guidelines recommend that CIN2 in adolescents be managed with close observation by regular colposcopy, rather than by ablative treatment.w16–w18
It may be the case that as in female adolescents, a high proportion of prevalent HRHPV (and HSIL) in older MSM represents an acute rather than a chronic infection. However, the incidence of cervical cancer among adolescents is very low, unlike the relatively high incidence of anal cancer among older MSM. Differentiating acute HRHPV infection from longstanding infection is a fundamental challenge for anal HSIL/HRHPV screening in adult MSM. Studies exploring the optimal screening and management of anal HPV and anal dysplasia among MSM are underway.w19 w20
In summary, high rates of ongoing sexual activity with new partners may contribute to the high prevalence and lack of decline in prevalent anal HPV among older MSM. Rather than solely persistent infection, older MSM may have continuing high rates of newly acquired, transient HPV infection. Prospective natural history studies comparing MSM with women across all age-ranges are required to confirm this.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
- Data supplement 1 - Online supplement
Handling editor Jackie A Cassell
Contributors IMP was responsible for the study concept and design. IMP performed the descriptive data analysis and drafted the manuscript. DM participated in the study design, created the manuscript figures and reviewed the manuscript for important intellectual content. DT participated in the study design and reviewed the manuscript for important intellectual content/FJ participated in the study design and reviewed the manuscript for important intellectual content. RH participated in the study design and reviewed the manuscript for important intellectual content. IZ participated in the study design and reviewed the manuscript for important intellectual content. MH participated in the study design, performed the descriptive data analysis and reviewed the manuscript for important intellectual content. AG was responsible, with MP, for the study concept and participated in its design, analysis and interpretation of data. AG reviewed the manuscript for important intellectual content. All authors read and approved the final manuscript.
Funding In 2013 the Gay Community Periodic Surveys were funded by the Australian Capital Territory Department of Health, the New South Wales Ministry of Health, the Queensland Department of Health and the Victorian Department of Health. The Kirby Institute, UNSW Australia is funded by the Australian Government of Health and Ageing.
Competing interests AG has received honoraria and research funding from CSL Biotherapies, honoraria and travel funding from Merck and sits on the Australian advisory board for the Gardasil HPV vaccine. RH has received support from CSL Biotherapies and MSD.
Ethics approval ASHR2 protocol was approved by La Trobe University HREC (HEC 11-040). The GCPS protocol was approved by the University of New South Wales HREC (HC13366).
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.