Study recruitment began in July, 2005, and ended in September, 2009, and 4299 men provided consent— 1443 from Brazil, 1429 from Mexico, and 1427 from the USA. The first 1159 men (subcohort) recruited from June, 2005, to December, 2006, who completed a minimum of 2 weeks of follow-up (median 27·5 months, range 0·5 to 40·5; mean 23·6 months, IQR 18·0–31·2) were followed up until March, 2009. 132 (11%) of 1159 men completed 2 weeks of follow-up, 94 (8%) completed 6 months, 82 (7%) completed 12 months, 118 (10%) completed 18 months, 247 (21%) completed 24 months, 373 (32%) completed 30 months, and 113 (10%) completed 36 months. Participants and non-participants, and men in the subcohort and full cohort were similar for most demographic and behavioural characteristics (). Mean age of participants was 32·1 years (SD 10·8). In the subcohort, the most common self-reported race was white, most men were uncircumcised, and most were non-smokers at enrolment (). 46 (15%) of 306 men who reported no female sexual partners in the past 3–6 months reported sexual intercourse with men.
Baseline characteristics of men enrolled and not enrolled
Oncogenic HPV types with the highest incidences per 1000 person months were 16, 51, 52, and 59 (). Non-oncogenic HPV types with the highest incidences per 1000 person months were 6, 62, 84, and CP6108 (). The incidences of any, non-oncogenic, or oncogenic HPV infections did not vary with age (figure). Differences in cumulative incidences were also noted by country (webappendix pp 2–4). At 12 months, Brazil had the highest incidence of oncogenic infection (p=0·0109) and non-oncogenic (p<0·0001) infection; however, the risk of HPV 16 infection was highest in Mexico (p=0·0195; webappendix p 2).
Prevalence at enrolment, and incidence and clearance of human papillomavirus (HPV) infections in men
Median time to clearance of infection of any HPV type was significantly longer in men aged 18–30 years than in the other age groups (figure). For the oncogenic HPV types, median time to clearance was longest for HPV types 16, 35, and 51 (). For the non-oncogenic HPV types, median time was longest for HPV types IS39, 40, and 62 (). Median time to clearance of non-oncogenic and oncogenic infections was longest in men aged 18–30 years (figure). By contrast, median time to clearance of HPV 16 was not age dependent (figure). No differences in HPV clearance by country were noted (webappendix pp 5–8). At 12 months and 24 months, 885 (66%) of 1342 and 1023 (90%) of 1134 men, respectively, cleared newly acquired HPV infections. 31 (44%) of 71 newly acquired HPV 16 infections persisted at 12 months and seven (13%) of 56 persisted at 24 months. Five (22%) of 23 HPV 6 infections, neither of two HPV 11 infections, one (7%) of 14 HPV 16 infections, and none of six HPV 18 infections that initially cleared were detectable at a later study visit. In every case, reappearance of these HPV types was after a minimum of 12 months of non-detectable infection (data not shown).
In the multivariate analysis, acquisition of oncogenic HPV infection was significantly higher with ten to 49 or at least 50 lifetime female sexual partners compared with not more than one, and with at least three male anal-sex partners versus none in the previous 3 months (). Acquisition of non-oncogenic HPV infection was significantly increased in men reporting two to nine and ten to 49 lifetime female sexual partners versus not more than one, current smokers, those with 13–15 years and more than 17 years of education compared with those with less than 12 years of education, and in men residing in Brazil versus those living in the USA ().
Factors independently associated with infection with human papillomavirus (HPV)
Men residing in Brazil and Mexico versus those living in the USA, and those reporting two to nine, ten to 49, and at least 50 lifetime female partners compared with no more than one were less likely to clear an oncogenic infection (). The probability of clearing an oncogenic HPV infection increased with increasing age (). Factors that reduced the likelihood of clearing a non-oncogenic infection were testing positive for HPV at enrolment, reporting a high number of female partners in the past 3 months (three partners compared with none), and residing in Brazil versus USA ().
Factors independently associated with clearance of human papillomavirus (HPV) infections