Of 406 men who have sex with men, 36% with prevalent high-risk anal human papillomavirus had type-specific persistence for ≥24 months. Men with ≥1 recent male anal sex partner had double the risk for ≥12 months’ persistence.
Keywords: natural history, anal cancer, anal condyloma, HPV DNA, prospective
Abstract
Background. Given high rates of anal disease, we investigated the natural history of high-risk anal human papillomavirus (HPV) among a multinational group of men who have sex with men (MSM) aged 18–64 years.
Methods. Anal specimens from human immunodeficiency virus–negative men from Brazil, Mexico, and the United States were genotyped. Over 2 years, 406 MSM provided evaluable specimens every 6 months for ≥2 visits. These men were stratified into men who have sex only with men (MSOM, n = 70) and men who have sex with women and men (MSWM, n = 336). Persistence was defined as ≥12 months’ type-specific duration and could begin with either a prevalent or incident infection. Prevalence ratios and 95% confidence intervals were calculated by Poisson regression.
Results. Median follow-up time was 2.1 years. Retention was 82%. Annual cumulative incidence of 9-valent vaccine types was 19% and 8% among MSOM and MSWM, respectively (log-rank P = .02). Duration of anal HPV did not differ for MSOM and MSWM and was a median of 6.9 months for HPV-16 after combining men from the 2 groups. Among men with prevalent high-risk infection (n = 106), a total of 36.8%, retained the infection for at least 24 months. For those with prevalent HPV-16 (n = 27), 29.6% were persistent for at least 24 months. Persistence of high-risk HPV was associated with number of male anal sex partners and inversely associated with number of female sex partners.
Conclusions. MSM with prevalent high-risk HPV infection should be considered at increased risk for nontransient infection.
Human papillomavirus (HPV) vaccination is effective against high-risk HPV infection at the anal canal, which otherwise can result in anal intraepithelial neoplasia and potentially invasive cancer; however, HPV vaccination of boys is still debated in some countries and vaccines are not licensed for persons older than age 26 years in the United States [1, 2]. The issue is pressing given that anal cancer incidence is increasing among women and men, but is particularly high among men who have sex with men (MSM), who may have an anal cancer incidence that is up to 25 times higher than that of heterosexual men [3–5]. Unlike heterosexual men who may benefit through herd immunity from female-only vaccination strategies, HPV-associated disease among MSM is unlikely to decline in female-only vaccination settings [6].
For MSM not benefiting from vaccination, developing the most appropriate anal cancer screening protocol may depend on anal HPV natural history data among persons across the lifespan to help identify those at highest risk of malignancy. Anal HPV natural history may also help support appropriate screening recommendations if those recommendations involve HPV DNA testing as they now do with cervical cancer screening [7, 8].
Our objective was to estimate cumulative incidence, median duration, and persistence of anal HPV among a multinational sample of MSM. We stratified the sample to investigate differences between men who have sex only with men (MSOM) and men who have sex with both women and men (MSWM). In addition, we assessed factors associated with persistence of high-risk anal HPV.
METHODS
Men were recruited in São Paulo, Brazil; Cuernavaca, Mexico; and Tampa, Florida, United States, beginning in June 2005 for the prospective HPV Infection in Men (HIM) Study—a study of genital HPV with optional anal canal sampling. Inclusion criteria included an age of 18–70 years, no prior genital warts or anal cancer, and no current sexually transmitted infection (STI) including human immunodeficiency virus (HIV). Men returned to clinic every 6 months for up to 4 years. We analyzed biological specimens and behavioral data collected up until the first 30 months of anal canal sampling, or a total of 5 visits. The study design has been previously described [9, 10].
MSM were not targeted for recruitment. Most recruitment in São Paulo targeting the general population but also included recruitment from a genitourinary clinic that tests for HIV and STIs. Men attending the clinic for STI symptoms or treatment were excluded. In Cuernavaca, men were recruited through a health plan, factories, and the military. Men in Tampa were recruited from a university campus and the general public. All participants consented to the study and received a nominal incentive for participation. The study was approved by human subjects committees at each study site.
Between June 2005 and February 2009, 4123 men enrolled in the HIM Study. At each visit, men completed an 88-item computer-assisted self-interview (CASI) written in the region's primary language (Portuguese, Spanish, or English). The CASI elicited demographic, substance use, and sexual behavior data. After the CASI, a clinician examined the participant for STIs before using 3 saline-wetted swabs to collect exfoliated skin cells from the penis and scrotum. Then, using a fourth swab, cells were collected from between the anal os and dentate line, after which the swab was placed into standard transport medium and stored at −80°C.
HPV Analyses
Specimens were analyzed for HPV DNA as described previously [9]. In brief, DNA was extracted using the QIAamp Media MDx Kit (Qiagen, Inc). The polymerase chain reaction consensus primer system (PGMY 09/11) was used to amplify a fragment of the HPV L1 gene [11]. HPV genotyping was conducted on all specimens using DNA probes labeled with biotin to detect 36 HPV types: 6, 11, 16, 18, 26, 31, 33, 34, 35, 39, 40, 42, 44, 45, 51–54, 56, 58, 59, 61, 62, 66–73, 81–84, and 89 [12]. Accuracy and potential contamination were assessed using nontemplate negative controls and CaSki DNA-positive controls. Anal specimen β-globin positivity averaged 90.9% over all visits.
Statistical Analyses
A specimen was considered positive for high-risk HPV if it was positive for ≥1 of 13 genotypes (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68 [13]) regardless of the presence of other genotypes. Similarly, specimens were labeled as low risk if any of the remaining 23 types were detected regardless of the presence of high-risk types.
Based on sexual behavior from all visits, 623 men were identified as MSM, of whom 596 (96%) returned for at least one 6-month follow-up visit. Of these, 559 (94%) contributed the optional anal canal specimen at ≥2 visits. A total of 434 men had ≥2 β-globin–positive specimens (78%) and 28 men acknowledged HIV during the study, leaving 406 men available for analysis. A total of 75 men were lost to follow-up over the 5 visits; thus, the retention rate was 82%.
MSM were classified as MSOM (n = 70) if they acknowledged sex only with men after enrollment in the study and MSWM (n = 336) if they acknowledged sex with women and men after enrollment. Classification of the men's sexual behavior was based on their answers to 17 questions about vaginal, anal, and oral sex in either the prior 3 or 6 months.
MSOM and MSWM were analyzed separately to compare demographics, behavioral characteristics, cumulative incidence, and duration of HPV. Because anal HPV infection duration was not different between the groups, MSOM and MSWM were combined for estimates of duration for prevalent and incident infection and for assessment of factors associated with persistence.
A 2-sided Fisher exact or a χ2 test was used to assess differences in characteristics between MSOM and MSWM. An incident infection was defined as the presence of a type-specific infection at any follow-up visit when that type was absent at enrollment. The cumulative risk for incident infection was estimated using the Kaplan–Meier method with right censoring at time of event or last negative visit. Duration of type-specific infection was calculated as the time from prevalent or incident detection with observations right-censored at the last positive visit. Two consecutive negative visits after detection (or a single negative result at the participant's last visit) was defined as clearance.
We estimated the proportion of men at risk for 6-, 12-, 18-, and 24-month persistence with high-risk HPV. The numerator was the sum of men with persistence that began with either prevalent or incident infection. For the denominator, we estimated the number at risk as men with prevalent or incident infection that put men at risk for either 6-, 12-, 18-, or 24-month persistence. For example, the number of men at risk for 24-month persistence (ie, men who retained type-specific infection for all 5 visits of the analysis) were all men with high-risk HPV at the baseline visit, whereas men at risk for 18-month persistence was the sum of all men with infection at baseline plus incident infection at the first 6-month visit.
For assessment of risk factors, persistence was defined as having a specific HPV type for ≥12 months at any time during the 24 months under analysis; thus, we assessed persistence events at the individual level that began with either prevalent or incident infection. Men with a persistent infection for ≥1 of 13 high-risk types were defined as cases. Men with no persistence for ≥12 months for high-risk genotypes were defined as noncases. Prevalence ratios (PRs) and 95% confidence intervals (CIs) were calculated by Poisson regression using a robust sandwich estimator of the variance [14]. Univariable associations with persistence were assessed for age, residence, race, ethnicity, marital status, duration of primary sexual relationship, years of school, circumcision, cigarette smoking, alcohol consumption, clinician-diagnosed anogenital warts, number of male anal sex partners in the last 3 months, number of female sex partners in the last 6 months, lifetime number of male anal sex partners, lifetime number of female sex partners, and frequency of condom use for anal sex in the last 6 months.
Age, residence, and variables with a log-rank P value ≤.25 were included in multivariable modeling with the exception of marital status, which was strongly correlated with duration of relationship and was therefore not included in multivariable modeling. Independent risk factors for anal HPV persistence were identified using a backward-elimination method. Age and residence were considered confounders and retained in all modeling, whereas other variables with a P value >.05 on a likelihood ratio test were individually removed until a final set of risk factors remained. Data were analyzed using SAS software version 9.3 (SAS Institute, Cary, North Carolina).
RESULTS
Median follow-up time was 2.1 years. Compared with MSWM, MSOM tended to be slightly younger in age, have a current primary sexual relationship of shorter duration, and were less likely to be a current cigarette smoker (all P < .001). The majority of both MSOM (63%) and MSWM (66%) were recruited in São Paulo (Table 1). MSOM had a higher number of recent and lifetime number of male anal sex partners compared with MSWM (P < .001 and P = .005, respectively).
Table 1.
Enrollment Characteristics of Men Who Returned for the First 6-Month Follow-up Visit
| No. (%) of Participants |
|||
|---|---|---|---|
| Characteristic | MSOM (n = 70) | MSWM (n = 336) | Total MSM |
| Age, y | |||
| 18–24 | 28 (40.0) | 66 (19.6) | 94 (23.2) |
| 25–34 | 26 (37.1) | 124 (36.9) | 150 (37.0) |
| 35–44 | 14 (20.0) | 107 (31.9) | 121 (29.8) |
| 45–64 | 2 (2.9) | 39 (11.6) | 41 (10.1) |
| Median (range) | 28 (18–47) | 33 (18–64) | 32 (18–64) |
| Residence | |||
| São Paulo, Brazil | 44 (62.9) | 222 (66.1) | 266 (65.5) |
| Cuernavaca, Mexico | 11 (15.7) | 66 (19.6) | 77 (19.0) |
| Tampa, Florida | 15 (21.4) | 48 (14.3) | 63 (15.5) |
| Race | |||
| White | 52 (74.3) | 182 (55.0) | 234 (58.4) |
| Black | 6 (8.6) | 70 (21.2) | 76 (19.0) |
| Mixed/other | 12 (17.1) | 79 (23.9) | 91 (22.7) |
| Ethnicity | |||
| Non-Hispanic | 39 (58.2) | 198 (59.8) | 237 (59.6) |
| Hispanic | 28 (41.8) | 133 (40.2) | 161 (40.5) |
| Duration of relationship with primary sex partner | |||
| No primary sex partner | 44 (63.8) | 108 (33.5) | 152 (38.9) |
| <2 y | 19 (27.5) | 77 (23.9) | 96 (24.6) |
| ≥2 y | 6 (8.7) | 137 (42.6) | 143 (36.6) |
| Years of school | |||
| <12 | 6 (8.7) | 72 (21.4) | 78 (19.2) |
| 12 | 17 (24.6) | 84 (25.0) | 101 (24.9) |
| 13–15 | 22 (31.9) | 75 (22.3) | 97 (23.9) |
| 16 | 20 (29.0) | 79 (23.5) | 99 (24.4) |
| ≥17 | 4 (5.8) | 26 (7.8) | 30 (7.4) |
| Circumcision status | |||
| Prepuce present | 50 (71.4) | 245 (72.9) | 295 (72.7) |
| No prepuce present | 20 (28.6) | 91 (27.1) | 111 (27.3) |
| Cigarette smoking status | |||
| Never smoker | 53 (75.7) | 169 (50.3) | 222 (54.7) |
| Former smoker | 3 (4.3) | 67 (19.9) | 70 (17.2) |
| Current smoker | 14 (20.0) | 100 (29.8) | 114 (28.1) |
| Alcoholic drinks in past month | |||
| 0–30 | 45 (69.2) | 224 (70.4) | 269 (70.2) |
| 31–60 | 8 (12.3) | 47 (14.8) | 55 (14.4) |
| >60 | 12 (18.5) | 47 (14.8) | 59 (15.4) |
| Anogenital warts (clinician report) | |||
| Yes | 3 (4.3) | 14 (4.2) | 17 (4.2) |
| No | 67 (95.7) | 322 (95.8) | 389 (95.8) |
| No. of male anal sex partners in last 3 mo | |||
| 0 | 12 (18.5) | 180 (56.1) | 192 (49.7) |
| 1 | 23 (35.4) | 62 (19.3) | 85 (22.0) |
| ≥2 | 30 (46.1) | 79 (24.6) | 109 (28.2) |
| Median (range) | 1 (0–10) | 0 (0–150) | 1 (0–150) |
| No. of female sex partners in last 6 mo | |||
| 0 | 67 (100.0) | 157 (49.2) | 224 (58.0) |
| 1 | 0 (0.0) | 73 (22.9) | 73 (18.9) |
| ≥2 | 0 (0.0) | 89 (27.9) | 89 (23.1) |
| Median (range) | 0 (0–0) | 1 (0–30) | 0 (0–30) |
| Lifetime No. of male anal sex partners | |||
| 0–2 | 15 (23.8) | 142 (45.0) | 157 (41.4) |
| 3–9 | 16 (25.4) | 81 (25.6) | 97 (25.6) |
| 10–19 | 12 (19.1) | 32 (10.1) | 44 (11.6) |
| ≥20 | 20 (31.7) | 61 (19.3) | 81 (21.4) |
| Median (range) | 10 (0–150) | 3 (0–2000) | 4 (0–2000) |
| Lifetime No. of female sex partners | |||
| 0–2 | 65 (100.0) | 96 (30.6) | 161 (42.4) |
| 3–9 | 0 (0.0) | 98 (31.2) | 98 (25.9) |
| 10–19 | 0 (0.0) | 59 (18.8) | 59 (15.6) |
| ≥20 | 0 (0.0) | 61 (19.4) | 61 (16.1) |
| Median (range) | 0 (0–1) | 6 (0–600) | 4 (0–600) |
| Frequency of condom use for anal sex in the last 6 mo for men who report anal sex (n = 295) | |||
| Always | 24 (39.3) | 93 (39.7) | 117 (39.7) |
| Sometimes | 28 (45.9) | 88 (37.6) | 116 (39.3) |
| Never | 9 (14.8) | 53 (22.7) | 62 (21.0) |
P values using 2-sided Fisher exact or χ2 tests and only nonmissing observations for comparing MSOM and MSWM were ≤0.001 for age, duration of primary sexual relationship, cigarette smoking status, number of male anal sex partners in the last 3 months, number of female sex partners in the last 6 mo, and lifetime number of female sex partners; lifetime number of male anal sex partners, P = .005; frequency of condom use for anal sex, P = .007; and race, P = .02. All other P values >.05.
Abbreviations: MSM, men who have sex with men; MSOM, men who have sex only with men; MSWM, men who have sex with women and men.
There were few statistically significant differences in rate of HPV acquisition, although there was a pattern of higher rates of acquisition among MSOM compared to MSWM, especially for low-risk types and 9-valent vaccine types (Table 2 and Figure 1). There was no statistically significant difference in age-specific acquisition for MSOM, MSWM, or after the groups were combined (Supplementary Appendix Figure 2).
Table 2.
Number of Prevalent Infections, 12-Month Cumulative Incidence, and Median Duration of Anal Human Papillomavirus
| HPV Type | MSOM (n = 70) |
MSWM (n = 336) |
MSM (n = 406) |
|||||
|---|---|---|---|---|---|---|---|---|
| Prevalent Infections | Incident Infections | Cumulative Incidence (95% CI) | Prevalent Infections | Incident Infections | Cumulative Incidence (95% CI) | Prevalent Duration, mo (95% CI) | Incident Duration, mo (95% CI) | |
| Any HPV type | 37 | 19 | 0.33 (.19–.53) | 132 | 84 | 0.15 (.11–.21) | 24.9 (22.2–26.5) | 10.8 (10.5–12.9) |
| 9-valent types | 20 | 19 | 0.19 (.10–.33) | 69 | 68 | 0.08 (.05–.12) | 23.8 (20.0–24.8) | 9.5 (8.0–12.7) |
| 4-valent types | 17 | 14 | 0.16 (.08–.29) | 53 | 53 | 0.06 (.04–.10) | 23.9 (21.2–26.5) | 7.1 (8.9–12.0) |
| Any high-risk type | 24 | 17 | 0.19 (.10–.34) | 82 | 71 | 0.10 (.07–.14) | 23.9 (20.4–24.9) | 9.5 (8.0–12.4) |
| 16 | 6 | 8 | 0.07 (.03–.17) | 21 | 22 | 0.02 (.01–.05) | 22.2 (20.3–27.1) | 6.9 (6.8–10.5) |
| 18 | 2 | 2 | 0.00 (.0–.0) | 13 | 20 | 0.02 (.01–.04) | 24.1 (19.0–30.7) | 8.1 (7.8–11.5) |
| 31 | 0 | 2 | 0.00 (.0–.0) | 4 | 9 | 0.01 (.01–.03) | 23.7 (11.3–32.5) | 6.0 (4.2–8.4) |
| 33 | 1 | 2 | 0.00 (.0–.0) | 4 | 5 | 0.00 (.0–.0) | 27.1 (14.4–41.1) | 8.0 (2.5–23.1) |
| 45 | 3 | 3 | 0.00 (.0–.0) | 11 | 21 | 0.02 (.01–.04) | 21.4 (13.5–35.1) | 7.8 (7.5–12.0) |
| 51 | 2 | 8 | 0.03 (.01–.12) | 17 | 25 | 0.03 (.02–.06) | 21.7 (17.1–25.2) | 6.9 (6.8–10.5) |
| 52 | 1 | 3 | 0.03 (.01–.12) | 7 | 15 | 0.02 (.01–.04) | 13.6 (8.3–23.4) | 6.2 (6.2–11.6) |
| 58 | 0 | 4 | 0.05 (.02–.14) | 6 | 12 | 0.01 (.00–.03) | 19.4 (5.3–33.9) | 6.7 (6.4–13.7) |
| 59 | 3 | 7 | 0.02 (.0–.11) | 9 | 21 | 0.03 (.01–.05) | 21.3 (19.9–27.6) | 7.3 (6.7–11.8) |
| Any low-risk type | 30 | 21 | 0.24 (.13–.42) | 103 | 86 | 0.14 (.10–.20) | 25.0 (22.6–27.8) | 9.8 (10.1–12.4) |
| 6 | 10 | 6 | 0.05 (.02–.16) | 16 | 26 | 0.03 (.01–.05) | 24.1 (20.0–25.5) | 11.0 (8.7–14.0) |
| 11 | 0 | 4 | 0.03 (.01–.12) | 7 | 7 | 0.01 (.01–.03) | 21.1 (3.5–50.7) | 6.3 (4.7–13.0) |
| 44 | 4 | 1 | 0.00 (.0–.0) | 3 | 15 | 0.03 (.01–.05) | 20.5 (9.2–30.0) | 7.0 (7.0–12.9) |
| 53 | 6 | 5 | 0.03 (.01–.13) | 17 | 25 | 0.02 (.01–.04) | 26.5 (18.3–46.9) | 11.6 (9.4–13.1) |
| 61 | 7 | 6 | 0.05 (.02–.15) | 7 | 17 | 0.01 (.00–.03) | 24.0 (18.6–25.4) | 6.2 (6.7–10.5) |
| 70 | 4 | 4 | 0.05 (.02–.14) | 8 | 14 | 0.01 (.00–.03) | 23.8 (13.8–38.4) | 7.3 (7.2–12.3) |
| 72 | 0 | 7 | 0.05 (.02–.14) | 9 | 18 | 0.02 (.01–.04) | 23.7 (13.1–31.6) | 6.7 (7.3–10.8) |
| 73 | 3 | 2 | 0.02 (.0–.11) | 9 | 15 | 0.03 (.01–.05) | 27.1 (16.9–33.6) | 6.3 (6.1–8.5) |
| 83 | 3 | 2 | 0.03 (.01–.13) | 7 | 13 | 0.01 (.00–.03) | 27.0 (9.5–94.3) | 6.7 (6.7–11.8) |
| 84 | 4 | 5 | 0.00 (.0–.0) | 24 | 31 | 0.03 (.01–.05) | 24.5 (16.6–29.8) | 6.3 (6.3–8.1) |
| 89 | 4 | 7 | 0.03 (.01–.12) | 12 | 35 | 0.04 (.02–.06) | 26.4 (19.7–36.1) | 8.3 (6.7–12.5) |
Data are shown as No. unless otherwise specified. Genotypes are not shown if the 12-month incidence for MSOM and MSWM is ≤0.02 (excepting types included in HPV vaccines). Derived from Cox regression and assessing the difference between MSOM and MSWM incidence, all P values for cumulative incidence were ≥.05 except for any HPV type, P = .004; any low-risk HPV, P = .01; and any 9-valent vaccine type, P = .01.
Abbreviations: CI, confidence interval; HPV, human papillomavirus; MSM, men who have sex with men; MSOM, men who have sex only with men; MSWM, men who have sex with women and men.
Figure 1.
Kaplan–Meier estimates of the cumulative incidence (with 95% confidence interval [CI]) of human papillomavirus (HPV) type 16, HPV-6, 4-valent vaccine types, 9-valent vaccine types, high-risk types, and low-risk types among men who have sex only with men (MSOM) and men who have sex with women and men (MSWM).
There was no difference between MSOM and MSWM in HPV duration in the anal canal with the exception of HPV-89 incident infections (MSOM, 6.0 months [95% CI, 5.6–6.4 months] and MSWM, 8.8 months [95% CI, 8.2–13.7 months]) (data not shown); therefore, we combined the 2 groups. The median duration of prevalent infection, depending on genotype, ranged from 13.6 to 27.1 months; however, incident infections were of much shorter duration. For example, the median duration of incident HPV-16 was 6.9 months (95% CI, 6.8–10.5 months) (Table 2).
Of 106 men with high-risk prevalent infection at baseline, 36.7% retained the high-risk HPV type for at least 24 months, whereas 8 of 27 men (29.6%) with prevalent HPV-16 retained the infection at 24 months (Figure 2).
Figure 2.
Number and proportion of men who have sex with men with persisting human papillomavirus (HPV) among men at risk for persistence. Both prevalent and incident persistent events are included.
A total of 88 men had ≥12-month high-risk persistence. In univariable analysis, number of male sex partners, both recent and lifetime, were associated with high-risk persistence among the combined MSOM and MSWM group (ie, MSM), whereas age, duration of relationship, and number of female sex partners, both recent and lifetime, were inversely associated with high-risk persistence. For example, compared to single men with no primary relationship, MSM with a primary sexual relationship of ≥2 years were half as likely to have high-risk HPV persistence for ≥12 months (Table 3).
Table 3.
Factors Associated With ≥12-Month Persistence of High-Risk Anal Canal Human Papillomavirus Among Men Who Have Sex With Men (n = 406)
| Factor | Univariable |
Multivariable | |
|---|---|---|---|
| No. (%) | PR (95% CI) | aPR (95% CI) | |
| Age, y | |||
| 18–24 | 26 (27.7) | Reference | Reference |
| 25–34 | 33 (22.0) | 0.80 (.52–1.21) | 0.75 (.49–1.14) |
| 35–44 | 25 (20.7) | 0.75 (.48–1.17) | 0.89 (.55–1.43) |
| 45–64 | 4 (9.8) | 0.35 (.15–.83) | 0.54 (.23–1.25) |
| Residence | |||
| São Paulo, Brazil | 59 (22.2) | 1.07 (.66–1.76) | 1.05 (.64–1.73) |
| Cuernavaca, Mexico | 16 (20.8) | 1.01 (.55–1.83) | 1.23 (.68–2.23) |
| Tampa, Florida | 13 (20.6) | Reference | Reference |
| Race | |||
| White | 52 (22.2) | Reference | … |
| Black | 15 (19.7) | 0.89 (.56–1.42) | … |
| Mixed/other | 18 (19.8) | 0.89 (.58–1.38) | … |
| Ethnicity | |||
| Non-Hispanic | 49 (20.7) | Reference | … |
| Hispanic | 36 (22.4) | 1.08 (.76–1.54) | … |
| Duration of relationship with primary sex partner | |||
| No primary sex partner | 40 (26.3) | Reference | … |
| <2 years | 29 (30.2) | 1.15 (.78–1.69) | … |
| ≥2 years | 18 (12.6) | 0.48 (.31–.75) | … |
| No. of male anal sex partners in last 3 mo | |||
| 0 | 21 (10.9) | Reference | Reference |
| 1 | 24 (28.2) | 2.58 (1.63–4.08) | 2.01 (1.19–3.37) |
| ≥ 2 | 41 (37.6) | 3.44 (2.28–5.19) | 1.88 (1.06–3.32) |
| Lifetime No. of male anal sex partners | |||
| 0–2 | 17 (10.8) | Reference | Reference |
| 3–9 | 20 (20.6) | 1.90 (1.15–3.16) | 1.41 (.80–2.50) |
| 10–19 | 13 (29.6) | 2.73 (1.55–4.80) | 1.67 (.88–3.16) |
| ≥20 | 30 (37.0) | 3.42 (2.15–5.45) | 2.06 (1.07–3.97) |
| No. of female sex partners in last 6 mo | |||
| 0 | 63 (28.1) | Reference | … |
| 1 | 10 (13.7) | 0.49 (.28–.83) | … |
| ≥ 2 | 13 (14.6) | 0.52 (.32–.84) | … |
| Lifetime No. of female sex partners | |||
| 0–2 | 48 (29.8) | Reference | Reference |
| 3–9 | 27 (27.6) | 0.92 (.64–1.34) | 1.20 (.81–1.76) |
| 10–19 | 7 (11.9) | 0.40 (.21–.74) | 0.54 (.28–1.06) |
| ≥20 | 3 (4.9) | 0.17 (.07–.41) | 0.29 (.12–.74) |
In univariable analysis, χ2 P values for a difference among categories of a factor were all ≥.05 except lifetime number of male anal sex partners, lifetime number of female sex partners, and number of male anal sex partners in past 3 months, P < .001; duration of relationship, P = .002; and number of female sex partners in past 6 months, P = .005. In multivariable analysis, all factors in the table were included in multivariable analysis with age and residence forced into all modeling. Estimates and 95% CIs are given in boldface if the CI excludes 1.00.
Abbreviations: aPR, adjusted prevalence ratio; CI, confidence interval; PR, prevalence ratio.
In multivariable analysis, MSM with ≥20 lifetime male anal sex partners and MSM with 1 or more male anal sex partners in the prior 3 months were approximately twice as likely to have persistent high-risk HPV (eg, compared to MSM with 0–2 partners during the lifetime, the adjusted PR was 2.06 [95% CI, 1.07–3.97] for men with ≥20 sex partners); however, MSM with the highest number of female sex partners (ie, ≥20 lifetime partners) were associated with a 71% reduced likelihood of persistence compared with MSM with only 0–2 female partners in their lifetime.
DISCUSSION
In this multinational study of MSM, we observed a pattern of a higher rate of anal HPV acquisition among MSOM compared with MSWM, although only the group of 9-valent vaccine types and low-risk types achieved statistical significance. No difference between MSOM and MSWM was detected with regard to anal HPV duration. After combining MSOM and MSWM, 36% of men with prevalent high-risk HPV retained the infection for at least 2 years.
Bearing in mind differences in assays, populations, and choice of reported estimates, 2 published articles are most comparable to the current study with regard to HIV-negative MSM across the lifespan [15, 16]. In a study where most men had only 1 follow-up visit, 123 HIV-negative Thai MSM had HPV-16 twelve-month cumulative incidence of 7.2% [15] (comparable to the current study). A second study of 287 MSM in the United States reported a 24-month incidence of 54% for any HPV type (including unclassified types) [16], which is somewhat higher than our observation of 41% (Supplementary Appendix Figure 1). In the US study, the median number of sex partners in the prior year was 4.5, which is comparable to the number of sex partners reported by men in the current study (Table 1). However, the inclusion of unclassified types in the US study may have resulted in the higher estimate. In both the US and Thai studies, men were recruited from HIV prevention programs that may have enrolled MSM with different risk profiles compared to the current study.
In contrast to the older population in the current study, other studies of anal HPV incidence have focused on young adult MSM [17–19]. These studies reported incidence rates for high-risk HPV ranging from 3.1 per 100 person-years for 94 young MSM (aged 16–30 years) recruited in Seattle [17] to 76.8 per 100 person-years among 200 teenage HIV-negative MSM recruited in Melbourne, Australia [18]. Among women, a study of 650 Hawaiian women with a median age of 40 years observed a 12-month incidence of high-risk anal HPV of 24% [20], which is similar to the estimate observed in the current study among MSOM (19%).
The median duration of incident infection, almost without regard to HPV group or type, was similar among MSOM and MSWM, likely reflecting an underlying common biological process. We observed a somewhat longer duration of incident infection in MSM compared to what has been reported among women [21]. For example, the median duration in months in the current study of MSM and the Hawaiian study of women was 9.5 vs 5.0 for high-risk types, 6.9 vs 4.4 for HPV-16, 9.8 vs 7.4 for low-risk types, and 11.0 vs 6.0 for HPV-6. Nevertheless, regardless of sex, approximately 50% of persons with anal HPV cleared their infection within 1 year in both studies. Given that the median duration of high-risk types was 9.5 months in these men, a 12-month persistence definition, compared with 6 months, likely better captures men who are at risk for nontransient high-risk infections.
Our finding of a much longer duration for prevalent infection vs incident infection likely reflects survivorship bias, a feature of populations as they enter a study—that is, persons with longer durations of disease are more likely to be enrolled at study start than persons who have shorter durations of the disease. A similar relationship between prevalent and incident HPV infection has been previously reported for anal and cervical HPV in women [22, 23] and genital HPV in men [10], and warrants investigation of different clinical procedures for prevalent vs incident anal HPV.
Slightly more than one-third of the MSM exhibited persistence of prevalent high-risk types for ≥24 months, indicating that prevalent high-risk anal HPV infection may be a clinically important event; however, the ratio of all high-risk persistent infections compared with HPV-16 persistent infections decreased from 1.52 for 6-month persistence to 1.24 for 24-month persistence, which is consistent with an increased role for HPV-16 in squamous cell carcinomas of the anal canal [24].
Men who reported any vs no recent receptive anal sex partners were approximately twice as likely to exhibit ≥12-month persistence of a high-risk HPV type. Our prior study found no association between partners and 6-month persistence of any HPV type among MSM [25]. Differences in persistence definition and the outcome in the current article (ie, prevalent and incident cases of ≥12-month persistence of high-risk types) may better demonstrate the risk of an increased number of partners.
The MSM were 406 largely urban men (two-thirds were from São Paulo), which may limit the study's generalizability. Also, due to self–reporting, it is possible that some MSM were misclassified as men who have sex only with women. While all analyzed men reported being HIV-negative, we did not test for HIV. We cannot rule out that incident detection of HPV is actually a reappearance of an existing infection [26], nor can we rule out clearance and reinfection with the same types between visits. Also, our observed anal HPV persistence may result not from the repeated sampling of a single focal infection but rather from the repeated sampling, over multiple visits, of metachronous type-specific infections with overlapping durations. Finally, the duration of prevalent infection, while longer than incident infection, is ultimately even longer than our estimates as we cannot determine the length of infection before study entry. In addition, right censoring after 30 months may mean we underestimated duration.
A study strength was inclusion of a multinational group of men. This provided a diverse group of men and a rich data source. Virtually all agreed to anal canal sampling, and β-globin positivity was high.
With some exceptions, we observed a similar incidence of anal HPV among MSOM and MSWM. But once HPV was detected, its duration among MSOM and MSWM was comparable, with prevalent HPV-16 being particularly likely to persist compared with incident HPV-16.
These results may help inform future anal cancer screening that uses HPV DNA testing—for example, men with baseline HPV-16 results may be counseled to seek anal Pap testing or other clinical follow-up. However, in a repeated screening setting, our data indicate that an incident high-risk infection, even HPV-16, is more likely than not to clear within an annual screening window. These data support not only development of anal precancer screening policy, but also appropriate counseling of MSM who have prevalent vs incident infection at the anal canal. These results may also inform debates in countries considering vaccination for males by informing anal HPV natural history modeling [1]. Finally, number of anal sex partners is a modifiable behavior, and, if repeated in other studies, may be included in interventions designed to reduce anal disease among MSM, particularly older MSM who do not have access to HPV prophylactic or therapeutic vaccination [2].
Supplementary Data
Supplementary materials are available at http://cid.oxfordjournals.org. Consisting of data provided by the author to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the author, so questions or comments should be addressed to the author.
Notes
Acknowledgments. Special thanks to the men who provided personal information and biological specimens for the study. Thanks to the HPV Infection in Men (HIM) Study Team in São Paulo, Cuernavaca, and Tampa: Lenice Galan, Elimar Gomes, Elisa Brito, Filomena Cernicchiaro, Rubens Matsuo, Vera Souza, Ricardo Cintra, Ricardo Cunha, Birgit Fietzek, Raquel Hessel, Viviane Relvas, Fernanda Silva, Juliana Antunes, Graças Ribeiro, Roberta Bocalon, Rosária Otero, Rossana Terreri, Sandra Araujo, Meire Ishibashi, the Centro de Referência e Treinamento em Doenças Sexualmente Transmissíveis/AIDS nursing team, Aurelio Cruz, Pilar Hernandez, Griselda Diaz Garcia, Oscar Rojas Juarez, Rossane del Carmen Gonzales Sosa, Rene de Jesus Alvear Vazquez, Christine Gage, Kathy Cabrera, Nadia Lambermont, Kayoko Kennedy, Kim Isaacs-Soriano, Andrea Bobanic, Michael O'Keefe, Bradley Sirak, and Ray Viscidi (HIM Study Coinvestigator, Johns Hopkins).
Disclaimer. The contents of this report are solely the responsibility of the authors and do not necessarily represent the official views of the National Institute of Allergy and Infectious Diseases (NIAID) or the National Cancer Institute (NCI).
Financial support. This work was supported by the NIAID (grant number 5R21AI101417-3 to A. G. N.) and the NCI (grant number RO1CA09880301–A1 to A. R. G.), National Institutes of Health.
Potential conflicts of interest. A prior employer of A. G. N. (Moffitt Cancer Center and Research Institute) received research funding from Merck & Co. L. L. V. and A. R. G. are on the speakers' bureau of, consult for, and are advisory board members for Merck & Co. A. R. G. has grants from Merck & Co. The institution of J. S. received grant and travel support from GlaxoSmithKline, Qiagen, and Merck & Co., and travel support from GlaxoSmithKline and Qiagen. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References
- 1.King EM, Gilson R, Beddows S et al. Human papillomavirus DNA in men who have sex with men: type-specific prevalence, risk factors and implications for vaccination strategies. Br J Cancer 2015; 112:1585–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Centers for Disease Control and Prevention. Recommendations on the use of quadrivalent human papillomavirus vaccine in males—Advisory Committee on Immunization Practices (ACIP), 2011. MMWR Morb Mortal Wkly Rep 2011; 60:1705–8. [PubMed] [Google Scholar]
- 3.Daling JR, Weiss NS, Klopfenstein LL, Cochran LE, Chow WH, Daifuku R. Correlates of homosexual behavior and the incidence of anal cancer. JAMA 1982; 247:1988–90. [PubMed] [Google Scholar]
- 4.D'Souza G, Wiley DJ, Li X et al. Incidence and epidemiology of anal cancer in the Multicenter AIDS Cohort Study. J Acquir Immune Defic Syndr 2008; 48:491–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Machalek DA, Poynten M, Jin F 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; 13:487–500. [DOI] [PubMed] [Google Scholar]
- 6.Read TR, Hocking JS, Chen MY, Donovan B, Bradshaw CS, Fairley CK. The near disappearance of genital warts in young women 4 years after commencing a national human papillomavirus (HPV) vaccination programme. Sex Transm Infect 2011; 87:544–7. [DOI] [PubMed] [Google Scholar]
- 7.Moyer VA. Screening for cervical cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med 2012; 156:880–91. [DOI] [PubMed] [Google Scholar]
- 8.Wentzensen N, Follansbee S, Borgonovo S 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; 26:2185–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Giuliano AR, Lazcano-Ponce E, Villa LL et al. The Human Papillomavirus Infection in Men Study: human papillomavirus prevalence and type distribution among men residing in Brazil, Mexico, and the United States. Cancer Epidemiol Biomarkers Prev 2008; 17:2036–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Giuliano AR, Lee JH, Fulp W et al. Incidence and clearance of genital human papillomavirus infection in men (HIM): a cohort study. Lancet 2011; 377:932–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gravitt PE, Peyton CL, Alessi TQ et al. Improved amplification of genital human papillomaviruses. J Clin Microbiol 2000; 38:357–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gravitt PE, Peyton CL, Apple RJ, Wheeler CM. Genotyping of 27 human papillomavirus types by using L1 consensus PCR products by a single-hybridization, reverse line blot detection method. J Clin Microbiol 1998; 36:3020–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bouvard V, Baan R, Straif K et al. A review of human carcinogens—part B: biological agents. Lancet Oncol 2009; 10:321–2. [DOI] [PubMed] [Google Scholar]
- 14.Deddens JA, Petersen MR. Approaches for estimating prevalence ratios. Occup Environ Med 2008; 65:501–6. [DOI] [PubMed] [Google Scholar]
- 15.Phanuphak N, Teeratakulpisarn N, Pankam T 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 2013; 63:472–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Critchlow CW, Hawes SE, Kuypers JM et al. Effect of HIV infection on the natural history of anal human papillomavirus infection. AIDS 1998; 12:1177–84. [DOI] [PubMed] [Google Scholar]
- 17.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 2014; 209:369–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Zou H, Tabrizi SN, Grulich AE et al. Site-specific human papillomavirus infection in adolescent men who have sex with men (HYPER): an observational cohort study. Lancet Infect Dis 2015; 15:65–73. [DOI] [PubMed] [Google Scholar]
- 19.Palefsky JM, Giuliano AR, Goldstone S et al. HPV vaccine against anal HPV infection and anal intraepithelial neoplasia. N Engl J Med 2011; 365:1576–85. [DOI] [PubMed] [Google Scholar]
- 20.Goodman MT, Shvetsov YB, McDuffie K et al. Acquisition of anal human papillomavirus (HPV) infection in women: the Hawaii HPV Cohort Study. J Infect Dis 2008; 197:957–66. [DOI] [PubMed] [Google Scholar]
- 21.Shvetsov YB, Hernandez BY, McDuffie K et al. Duration and clearance of anal human papillomavirus (HPV) infection among women: the Hawaii HPV Cohort Study. Clin Infect Dis 2009; 48:536–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Moscicki AB, Ma Y, Farhat S et al. Natural history of anal human papillomavirus infection in heterosexual women and risks associated with persistence. Clin Infect Dis 2014; 58:804–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Rodriguez AC, Schiffman M, Herrero R et al. Longitudinal study of human papillomavirus persistence and cervical intraepithelial neoplasia grade 2/3: critical role of duration of infection. J Natl Cancer Inst 2010; 102:315–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Gillison ML, Chaturvedi AK, Lowy DR. HPV prophylactic vaccines and the potential prevention of noncervical cancers in both men and women. Cancer 2008; 113:3036–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Nyitray AG, Carvalho da Silva RJ, Baggio ML et al. Six-month incidence, persistence, and factors associated with persistence of anal human papillomavirus in men: the HPV in Men Study. J Infect Dis 2011; 204:1711–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Insinga RP, Perez G, Wheeler CM et al. Incidence, duration, and reappearance of type-specific cervical human papillomavirus infections in young women. Cancer Epidemiol Biomarkers Prev 2010; 19:1585–94. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.