The present analysis included 379 men enrolled in the cohort. The mean age of participants was 29 years. The majority of men were white, single (never married), and had at least some post-secondary education. Seventy-seven percent of men were heterosexual. Fifty-three percent reported a total of 6 or more female sex partners during their lifetimes. A history of soliciting female commercial sex workers was reported by 13% of men. HIV infection was reported by 7% of men. HIV-positive individuals were excluded from all analyses of HPV status, resulting in 351 men for subsequent analyses.
In univariate analysis, cigarette smoking, history of sex with men, lifetime number of female sex partners, history of genital warts, and circumcision status were associated with HPV status at 1 or more anatomic sites. Age, race/ethnicity, birthplace, education level, marital status, age at initial sexual intercourse with a female, and condom use were not associated with HPV status.
In multivariate analysis (adjusting for age, birthplace, race/ethnicity, education level, lifetime number of female sex partners, history of sex with men, age at initial sex, condom use, history of genital warts, cigarette smoking, and circumcision status), only sex with men, lifetime number of female sex partners, and circumcision status were associated with HPV status. History of sex with men was inversely associated with HPV infection of the shaft (adjusted OR, 0.44 [95% CI, 0.21-0.94]). Lifetime number of female sex partners was positively associated with HPV infection of the shaft and scrotum (adjusted ORs of 6.93 [95% CI, 2.34-20.48] [Ptrend < .0001] and 3.54 [95% CI, 1.14-10.96] [Ptrend = .003], respectively, for >15 partners relative to <2 partners).
Circumcision status was based on clinical examination; 79% of men were circumcised. Four men who reported being circumcised were observed to have a foreskin on examination and were classified as uncircumcised. Visible genital warts were observed in 4% of study participants.
Compared with circumcised men, uncircumcised men had an older age distribution (P = .01), were more likely to be of Asian or other racial/ethnic background (P = .003), were more likely to be born outside of the United States (P < .001), had a broader distribution by education level (P = .003), and were less likely to have a history of genital warts (P = .007) (). There were no differences between circumcised and uncircumcised men with respect to sexual history, condom use, cigarette smoking history, and history of sexually transmitted infections other than genital warts.
Characteristics of the male cohort, by circumcision status
Overall, 90% (2001/2226) of specimens were sufficient on the basis of β-globin detection. There was no difference in specimen sufficiency by circumcision status for any anatomic site (data not shown).
The overall HPV prevalence was highest in the penile shaft (52%) followed by the scrotum (40%), glans/corona (32%), urine (10%), and semen (6%) (). Among uncircumcised men, HPV prevalence in the foreskin (44%) was comparable to that in the glans/corona (46%).
Association between circumcision and human papillomavirus (HPV) prevalence, by anatomic site
In comparisons by circumcision status, uncircumcised men had a higher prevalence of any HPV infection of the glans/corona (46% vs. 29%; adjusted OR, 1.96 [95% CI, 1.02-3.75]). Uncircumcised men also had an increased risk of oncogenic infection (31% vs. 16%; adjusted OR, 2.51 [95% CI, 1.11-5.69]) and infection with multiple HPV types in the glans/corona (31% vs. 12%; adjusted OR, 3.56 [95% CI, 1.50-8.50]).
The highest prevalence of HPV was observed in the penile shaft for both groups. HPV prevalence for circumcised men (50%) was lower than for uncircumcised men (60%), although differences between the 2 groups were not statistically significant. No association with circumcision status was observed for sites other than the glans/corona.
Overall, 33 different HPV genotypes were detected across all sites (data not shown). The indeterminate risk type HPV-84 was the most common type found overall and in the shaft (14%), foreskin (14%), scrotum (10%), and urine (7%).
In the glans/corona, the distribution of oncogenic HPV types varied by circumcision status (). HPV-84 was the most common HPV type detected in the glans/corona of both uncircumcised and circumcised men. The most common oncogenic HPV types detected in the glans/corona were HPV-16 and HPV-39 among circumcised men and HPV-66 followed by HPV-52, -53, and -73 among uncircumcised men. The most common nononcogenic type detected in the glans/corona in both circumcised and uncircumcised men was CP6108.
Distribution of human papillomavirus (HPV) genotypes detected in the glans penis and coronal sulcus, by circumcision status
Correlation in type-specific HPV detection was evaluated between paired anatomic sites from the same individuals. Among uncircumcised men, type-specific agreement was observed between the foreskin and the glans/corona (κ = 0.52 [95% CI, 0.30-0.74]) and the foreskin and the shaft (κ = 0.43 [95% CI, 0.22-0.65]). Among circumcised men, type-specific agreement in HPV infection was observed between the shaft and scrotum (κ = 0.56 [95% CI, 0.45-0.67]), but this association was not among uncircumcised men (κ = 0.39 [95% CI, 0.19-0.59]).