Background: To develop strategies for prevention and early treatment of human papillomavirus (HPV) anal and penile cancer, a better understanding of related sexual behavior risk factors is needed.
Goal: The goal of this study was to establish the prevalence of anal and coronal sulcus HPV in a group of men who have sex with men participating in a Dutch gay-cohort study, to identify risk factors associated with HPV infection in this group, and to investigate the presence of identical HPV types in couples with stable relationships.
Study Design: A cross-sectional study of 241 HIV-negative and 17 HIV-positive men who have sex with men visiting the sexually transmitted disease clinic of the Erasmus MC for a regular and scheduled examination. Participants underwent a routine venereological examination including HIV serologic analysis, and swabs were taken from the coronal sulcus and anus for HPV DNA testing. All subjects were asked to complete a questionnaire on sexual risk behavior.
Results: HPV DNA was detected at the coronal sulcus in 23.5% of the HIV-positive men and in 15.8% of the HIV-negative men (P = 0.492). In anal specimens, HPV DNA was detected in 64.7% of the HIV-positive men and 32.8% of the HIV-negative men (P = 0.015). High-risk HPV types (P = 0.007) and 2 or more different HPV genotypes (P = 0.006) were seen more often in anal specimens of HIV-positive persons than in specimens of HIV-negative persons. A factor possibly associated with the presence of anal HPV infection was a concomitant anal infection with Chlamydia trachomatis, gonococci, or herpes simplex virus (P = 0.059). In only 16.7% of HPV-positive steady couples, both companions showed the presence of one or more identical HPV genotypes.
Conclusion: In this study, anal HPV DNA was detected more often than HPV DNA at the coronal sulcus. HIV positivity was associated with a higher prevalence of high-risk, but not with low-risk HPV types, at the anus. No association was found between HIV positivity and presence of high-risk HPV at the coronal sulcus. No sexual behavioral determinants for the presence of HPV could be identified. Concomitant anal infection with C trachomatis, gonococci, or herpes simplex virus may be associated with HPV infection. In the majority of steady couples, partners were infected with different HPV types.
This study of men who have sex with men participating in a Dutch gay cohort showed that concomitant anal infection with Chlamydia trachomatis, gonococci, or herpes simplex virus may be associated with anorectal human papillomavirus infection.
From the *Department of Dermatology and Venereology, †Department of Virology, and ‡Institute of Epidemiology and Biostatistics, Erasmus MC, Rotterdam, The Netherlands
Reprint requests: Eric M. van der Snoek, MD, Erasmus MC, Department of Dermatology and Venereology, Dr. Molewaterplein 40, 3015 GD Rotterdam, The Netherlands. E-mail: email@example.com
Received December 20, 2002,
revised February 19, 2003, and accepted February 21, 2003.
Human papillomavirus (HPV) infections are the most common sexually transmitted viral infections and have a steadily increasing prevalence. 1 Genital warts or condylomata acuminata, one of the clinical manifestations of anogenital HPV infection, are predominantly associated with HPV types 6 and 11 and are estimated to affect about 1% of the sexually active population in the United States and Western Europe. 2,3 The estimated prevalence of subclinical and latent HPV infection in sexually active women and men ranges from 10% to 46%. Only 1% of all persons infected with HPV will have visible genital warts. 4
HPV is a double-stranded DNA virus. To date, more than 100 genotypes of HPV have been characterized. At least 35 of these genotypes have a predilection for the anogenital tract. 5 The virus infects the basal layer of the epithelium and increases cell proliferation and viral replication in fully differentiated keratinocytes. The arising papillomas are usually benign, but can progress to dysplasia or neoplasia in a small percentage of cases. The latter primarily occurs in case of infection with so-called high-risk types of HPV, particularly types 16 and 18. 6 The viral regulatory genes E6 and E7 inactivate the tumor-suppressor protein p53 and retinoblastoma (pRb) protein, which renders cellular DNA susceptible to carcinogenic effects of mutagens and increases the risk of malignant transformation. 7
In almost all cervical cancers, HPV DNA (mostly HPV-16) has been identified. 5,8 The yearly incidence of anal cancer in the United States among both men and women in the general population is approximately one tenth that of cervical cancer. In men who have sex with men (MSM) with a history of receptive anal intercourse, the incidence of anal cancer was estimated to be at least 44 times higher, namely 35 per 100,000 per year. 5 The incidence of anal cancer among HIV-positive men may be about twice that of HIV-negative men. In a population-based study linking AIDS and cancer registries, the risk of anal cancer among persons with AIDS was 84 times greater than that in the general population. 9,10 Daling et al. 11 stated that a history of condylomata acuminata carried a relative risk of 27 for the development of anal cancer and anal sexual intercourse carried a relative risk of 50. According to Carter et al., 12 HPV-16 seropositivity is associated with a sixfold increased prevalence of anal cancer in men.
Because anal condylomata acuminata are more common than penile warts in MSM, most studies on HPV in this group have focused only on anal warts and HPV infection. 9 However, 75% to 100% of patients with penile intraepithelial neoplasia have high-risk HPV DNA types, mostly HPV-16, whereas up to 50% of persons with invasive penile cancers test positive for HPV DNA. 3,13,14 Penile cancer is most often of the squamous cell type and is less common in the United States and Western Europe than anal cancer, with a yearly incidence of 0.3 to 1.0 per 100,000.
To the best of our best knowledge, there are no epidemiologic data on the possible association between HPV infection and penile cancer in (HIV-positive) homosexual and bisexual men. We consider information on the prevalence of HPV infection at the coronal sulcus important, because of the assumed relation with penile cancer. 13,14 However, there are no longitudinal studies on HPV-related penile disease in gay men. Therefore, we performed a cross-sectional study in a Dutch cohort of MSM to establish the prevalence of both anal and coronal sulcus HPV infection and to identify sexual behavior risk factors related to the presence of HPV infection. The presence of identical HPV genotypes in steady couples was also investigated. To develop strategies for prevention and early treatment of HPV-associated high-grade anal squamous intraepithelial lesions (HSIL), a better understanding of sexual behavior risk factors is needed. 15
Study Population and Study Design
The study was performed at the sexually transmitted disease (STD) clinic of the Department of Dermatology and Venereology, Erasmus MC, Rotterdam, The Netherlands. Beginning in February of 1999, we recruited MSM to participate in the Rotterdam Gay-Cohort study. Homosexual and bisexual STD clinic attendees were informed about the opportunity to take part in the study. In addition, men visiting gay bars and saunas in Rotterdam were asked to participate in the study. To enroll a diverse group of men, trained volunteers visited so-called meeting places, where gay men have (anonymous) sexual encounters. Advertisements were published in newspapers and gay periodicals. The last participant was enrolled in February of 2000. Cohort participants were tested for STDs and HIV every 6 months for a period of 3 years. HPV specimens were taken only during the third visit. Therefore, the data reported in this article only concern the third visit of the cohort participants.
Data Collection and Questionnaires
Demographic and sexual behavioral data were collected. These included ethnic background, age, educational qualification, sexual orientation, number of sexual partners during the last 6 months, practice of anal intercourse, intravenous drug use, participation in prostitution, and earlier diagnoses of STD, including HIV infection. Additional data regarding age of first sexual experience, estimated number of lifetime sexual partners, and condom use were collected with self-administered questionnaires. Participants were asked if they had a steady sex partner for longer than 6 months; partners who answered this question positively were recorded as steady couples.
At enrollment and at each semiannual visit, all participants underwent a standardized venereological examination as described previously. 16 In HIV-positive men, blood samples were taken for a CD4+ lymphocyte count.
HPV DNA Sample Collection
Between January of 2000 and August of 2001, specimens for assessment of HPV DNA were collected from all participants using a dry swab (Medical Wire & Equipment Co. (Bath) Ltd., Corsham, Wiltshire, United Kingdom) with sampling the coronal sulcus and the anal area. 17 In this study, we chose to sample the coronal sulcus for HPV. According to Holmes et al., 18 condylomata acuminata in men first appear on the frenulum and coronal sulcus because the area is liable to trauma during intercourse, allowing entry of an infecting agent. The swabs were immediately placed into standard collecting tubes without transport medium and sent to the Department of Virology for further processing.
Detection and Typing of HPV DNA
In 119 of the 258 men (46.1%), HPV DNA testing was performed by using a specific HPV-type detection polymerase chain reaction (PCR) for HPV types 6, 11, 16, 18, 31, and 33, as described elsewhere. 19 Later on, the SPF reverse-hybridization line probe assay (LiPA) HPV PCR test, which detects 25 different HPV genotypes, was used in 139 of 258 males (53.9%). Kleter et al. showed that LiPA results were highly concordant with those of genotype-specific PCR tests. 20,21 The total nucleic acid DNA was extracted by using the total nucleic acid isolation kit on a MagnaPure LC system (Roche Applied Science, Penzberg, Germany).
Data were compared to assess statistically significant differences in the prevalence of HPV related to sexual behavior parameters. Prevalence was calculated as the number of positive tests per 100 tested individuals. Logistic regression analyses were used to test differences between the HPV groups. In these analyses, HIV status was included as a covariable along with the sexual behavior parameters. We also tested whether HIV modified the effect of these sexual behavior parameters. The result of a test was considered significant if the P value was less than 0.05.
A total of 258 men were enrolled in this study, including 17 HIV-positive men (group I) and 241 HIV-negative men (group II). Of the 17 HIV-positive men, 3 were detected during the third semiannual visit and had recently (less than 6 months) seroconverted. Fourteen subjects had tested positive for HIV before the study, within a median of 2 years (range, 1–9 years). Five HIV-positive men were on antiretroviral therapy. The median age of participants was 42.0 years (range, 29–59 years) in group I and 41.0 years (range, 19–76 years) in group II (P = 0.772). In group I, 94.1% of participants were of Dutch descent, compared with 93.8% in group II, a difference that was not significant. In group I, 88.2% of men were homosexual and 11.8% were bisexual, whereas 89.1% of men in group II were homosexual and 10.9% were bisexual; this difference was not significant. There were no significant differences between the groups in terms of educational qualification. No exact data were available regarding the circumcision status of the participants, but an estimated minority of less than 5% of these men were circumcised.
Sexual Behavior Characteristics
Participants in group I had had more sexual partners during the last 6 months, with a median number of 15 (range, 2–78) compared with 8 partners (range, 0–140) in group II (P = 0.031). Not unexpectedly, the men in group I also had had more lifetime sexual partners, with a median number of 400 compared with 100 in group II (P = 0.001). In group I, no man had “never” had anal sex, whereas 15.5% of group II participants stated to have never practiced anal sex; however, this difference was not statistically significant. Data concerning condom use, which were collected using self-administered questionnaires, did not show significant differences between groups. Four of 16 men (25.0%) in group I and 48 of 204 men (23.5%) in group II said that they “always” used condoms during anal sex. No participants in group I and 13 of 204 men (6.4%) in group II participants said that they “never” used condoms.
No significant differences were found between groups regarding age at first sexual experience with a male partner and participation in prostitution or intravenous drug use (data not shown). Twenty-one steady couples participated in the current study, and none of these steady couples was mutually monogamous. Partners also had other sexual partners during the last 6 months, with a median number of 15 (range, 1–140).
Results of Venereological Examination
No differences regarding gonococcal, chlamydial, and herpes simplex virus infections, in the past or at the present visit, were found between the 2 groups. Reports of formerly diagnosed perianal and penile warts were similar in both groups. The median CD4+ lymphocyte count in HIV-positive men was 600 (range, 340–1,020). The median CD4+ count in the 5 HPV-negative men was 690 versus 490 in the 12 HPV-positive men, though this difference was not significant. CD4+ lymphocyte counts were not available for 3 of the 17 (17.6%) HIV-positive subjects.
Table 1 summarizes the detection of HPV in anal specimens in this cohort. HPV DNA was detected in coronal sulcus specimens of 4 (23.5%) out of 17 HIV-positive men and in 38 (15.8%) out of 241 HIV-negative men (P = not significant [NS]). In anal specimens, HPV DNA was detected in 11 (64.7%) out of 17 HIV-positive men and in 79 (32.8%) out of 241 HIV-negative men (P = 0.015). In HIV-positive men, HPV-16 was the most frequently HPV subtype detected in both anal and coronal sulcus specimens, and was found in 5 (29.4%) and 2 (11.8%) of these patients in the respective samples. In HIV-negative men, HPV-6 was the most frequently detected HPV subtype, and was detected 30 times (12.4%) in anal specimens and 12 times (5.0%) in coronal sulcus specimens.
In anal specimens, two or more different types of HPV were more often found in HIV-positive men than in HIV-negative men; namely, six times (54.5%) versus 23 times (29.1%; P = 0.006). With regard to coronal sulcus specimens, multiple HPV types were found once (25.0%) in the HIV-positive men versus eight times (21.1%) in HIV-negative men (P = NS). Only high-risk HPV types were found in the anal specimens of seven HIV-positive men (41.2%) and in 32 (13.3%) of the HIV-negative men (P = 0.007). In coronal sulcus specimens, only high-risk types were found in three HIV-positive men (17.6%) and in 19 (7.9%) of the HIV-negative men (P = NS). HPV-16 was detected more often in both coronal and anal specimens in HIV-positive men (P = 0.023). The simultaneous presence of only high-risk HPV genotypes in coronal sulcus and anal specimens was found three times (17.6%) in HIV-positive men versus five times (2.1%) in HIV-negative men (P = 0.011).
In those specimens tested with the SPF LiPA HPV PCR test, HPV-52 was most frequently detected in anal specimens, and was detected in 16 of 139 (11.5%) patients. Anal HPV-52 was more often found in the nine HIV-positive men than in the 130 HIV-negative men; namely, 6 times (66.7%) versus 10 times (7.7%; P < 0.0005). HPV-68 was seen in 13 (9.4%) of all anal specimens: three times in HIV-positive men (33.3%) and 10 times in HIV-negative men (7.7%; P = 0.039). In coronal sulcus specimens, both HPV-68 and HPV-70 were most often found by using the LiPA test; this method detected these HPV subtypes in 6.5% of all men. HPV-68 was seen more often (three times) in HIV-positive men, i.e. (33.3%, P = 0.013).
In 18 of the 21 steady couples (85.7%), at least one partner was HPV positive. In 3 of these 18 couples (16.7%), both partners shared the same type of HPV infection. Seventeen of the 18 couples (94.4%) were discordant in terms of at least one HPV type.
The effect of risk factors, adjusted for HIV-infection status, on HPV types 6 to 33 in coronal sulcus and anal specimens is shown in Tables 2 and 3. Having a concomitant chlamydial, gonococcal, or herpetic infection may be a risk factor for having a positive anal HPV test result (P = 0.059). No other risk factors could be identified.
No relation was found between HPV risk and the variables Dutch descent, educational qualification, number of sex partners during the last 6 months, history of chlamydial, gonococcal, or herpetic anorectal infection, penile warts at present visit, or having had no anal sex during the last 6 months (data not shown in Tables 2 and 3). After including an interaction term between the risk factors and HIV infection in the model, no significant effect-modifying role of HIV could be found for all the explanatory variables listed in Tables 2 and 3, possibly because the number of HIV-positive participants in this study was small.
In this study, we examined risk factors for the detection of coronal sulcus and anal HPV infections a Dutch cohort of MSM. Unlike most other studies, we also focused on coronal sulcus HPV infection, because of the assumed relation between HPV infection and penile cancer. In our study, HIV positivity was associated with a higher prevalence of anal high-risk HPV types, but not low-risk types. No association was found between the HIV serostatus and the prevalence of high-risk or low-risk coronal sulcus HPV infection. A limitation of this study is the small number of HIV-positive participants.
Factors associated with the presence of coronal sulcus HPV infection were not found. Possible risk factors associated with the presence of anal HPV infection could be a concomitant anal infection with Chlamydia trachomatis, gonococci, or herpes simplex virus. No sexual behavior determinants for the presence of HPV could be found.
Other studies also found an association between HIV-positivity and a higher prevalence of anal high-risk HPV types. 9,22,23 In some studies, a relation between CD4+ counts lower than 200 × 106/l and high-risk HPV was detected as well. 9,22,23 However, other studies did not confirm this finding. 24,25 It is unclear why we did not find an association between HIV serostatus and prevalence of high-risk coronal sulcus HPV. Van Doornum et al. 26 found more HPV-infections in swabs taken from the urethra than from the coronal sulcus in heterosexual men. In that study, which enrolled 65 Dutch heterosexual men with at least five heterosexual partners in the last 6 months, the presence of HPV DNA in coronal sulcus specimens was only 3%, whereas 17% of the urethral samples tested positive for HPV DNA. 26 Possibly swabs taken from the coronal sulcus often fail to detect prevalent HPV infections, perhaps because of penile hygienic measures. No exact data were available about the circumcision status of our study participants. In uncircumcised men, the maceration from epithelial debris and glandular secretions could possibly give rise to a more prevalent HPV infection or HPV persistence. An estimated number of less than 5% of men in this study were circumcised. A higher “natural” susceptibility of the anal skin and more frequently occurring trauma caused by receptive anal sex could explain the higher prevalence of HPV types at the anus. Mucoid anal discharge and moisture are easily retained within the natal cleft, and might be a reason for HPV persistence at the anal epithelium.
The concomitant presence of HIV and other STDs, some of which cause skin and/or mucosal defects (e.g., syphilis and genital herpes), has been described previously. 9,15,27,28 Mucosal defects may explain the higher prevalence of HPV in MSM with concomitant rectal STDs, as found in this study.
A number of studies have examined factors associated with the detection of anal HPV infection in MSM. These studies showed high rates of HPV infection among those who had practiced receptive anal intercourse and had a large number of (lifetime) sexual partners. There also is an association between a high prevalence of anal HPV infection and trauma of the anal epithelium, younger age, rectal drug use, history of laxative use, hepatitis B infection, positive C trachomatis serologic result, history of gonorrhea, history of rectal discharge, anal fissures or fistulas, and smoking. 9,15,29
In this study, the factors type of anal sex practiced, age at first sexual experience with a male partner, and estimated number of (lifetime) sexual partners were not associated with the detection of anal HPV, contrary to results of other studies. 10,16 A recent study on the prevalence and determinants of genital HPV in heterosexual men could not detect an association of penile HPV infection and the factor age at first sexual intercourse. Also, no association was found between penile HPV infection and number of sexual partners. 29
Risks factors for penile cancer are penile tear, being uncircumcised, difficulty in retracting foreskin, more than 30 lifetime sexual partners, condylomata acuminata, lichen sclerosus of the glans penis, smoking, ultraviolet A photograph chemotherapy, and smegma. 14
The simultaneous presence of identical HPV types in steady couples was rarely seen. The majority (94.7%) of prevalent HPV types were not shared by both partners. The simultaneous presence of identical HPV types in steady couples is remarkably low. Rapid clearance, high numbers of sexual partners other then their steady partner, inadequate collection of specimens and HPV-induced type-specific immunity could perhaps explain this finding. To get a better idea of clearance and persistence rates, all individuals participating in this study will be retested for the presence of HPV DNA after 1.5 years, during their sixth semiannual visit.
Concomitant anal infection with C trachomatis, gonococci or herpes simplex virus might have been associated with anorectal HPV infection in this group.
1. O'Mahony C, Law, C, Gollnick HPM, et al. New patient-applied therapy for anogenital warts is rated favourably by patients. Int J STD AIDS 2001; 12: 565–570.
2. Syrjänen SM, Krogh Gvon, Syrjänen KJ. Anal condyloma in men. 1. Histopathological and virological assessment. Genitourin Med 1989; 65: 216–224.
3. Dillner J, Meijer CJLM, Krogh G von et al. Epidemiology of human papillomavirus infection. Scand J Urol Nephrol Suppl 2000; 205: 194–200.
4. Bauer HM, Ting Y, Greer CE, et al. Genital human papillomavirus infection in female university students as determined by a PCR-based method. JAMA 1999; 265: 472–477.
5. Palefsky JM. Anal squamous intraepithelial lesions in human immunodeficiency virus-positive men and women. Semin Oncol 2000; 4: 471–479.
6. Höpfl R, Guger M, Widschwendter A. Humane Papilloviren und ihre Rolle bei der Karzinogenese. Hautarzt 2001; 52: 834–48.
7. Howley PM. Papillomavirinae and their replication. In: Fields BN, Knipe DM and Howley PM, eds. Fundamental Virology. Philadelphia: Lippincott-Raven, 1996.
8. Lie AK, Skarsvag S, Haugen OA, et al. Association between the HLA DQB1*0301 gene and human papillomavirus infection in high-grade cervical intraepithelial neoplasia. Int J Gynecol Pathol 1999; 18: 206–210.
9. Palefsky JM, Holly EA, Ralston ML, et al. Prevalence and risk factors for human papillomavirus infection of the anal canal in human deficiency virus (HIV)-positive and HIV-negative homosexual men. J Infect Dis 1998; 177: 361–367.
10. Friedman HB, Saah AJ, Sherman ME, et al. Human papillomavirus, anal squamous intraepithelial lesions and human immunodeficiency virus in a cohort of gay men. J Infect Dis 1998; 178: 45–52.
11. Daling JR, Weiss NS, Hislop TG, et al. Sexual practices, sexually transmitted diseases, and the incidence of anal cancer. N Engl J Med 1987; 317: 973–977.
12. Carter JJ, Madeleine MM, Shera K, et al. Human papillomavirus 16 and 18 L1 serology compared across anogenital cancer sites. Cancer Res 2001; 61: 1934–1940.
13. Bezerra ALR, Lopes A, Landman G, et al. Clinicopathologic features and human papillomavirus DNA prevalence of warty and squamous cell carcinoma of the penis. Am J Surg Pathol 2001; 25: 673–678.
14. Dillner J, Krogh Gvon, Horenblas S, et al. Etiology of squamous cell carcinoma of the penis. Scand J Urol Nephrol Suppl 2000; 205: 189–193.
15. Critchlow CW, Hawes SE, Kuypers JM, et al. Effect of HIV infection on the natural history of anal human papillomavirus infection. AIDS 1998; 12: 1177–1184.
16. Snoek van der EM, Götz HM, Mulder PG, et al. Prevalence of STD and HIV infections among attenders of the Erasmus MC STD clinic, Rotterdam, The Netherlands, during the years 1996 to 2000. Int J STD AIDS 2003; 14: 119–124.
17. Lazcano-Ponce E, Herrero R, Munoz N, et al. High prevalence of human papillomavirus infection in Mexican males: comparative study of penile-urethral swabs and urine samples. Sex Transm Dis 2001; 28: 277–280.
18. Oriel D. Genital human papillomavirus. In: Holmes KK, Mårdh PA, Sparling PF, et al, eds. Sexually Transmitted Diseases, 2nd ed. New York: McGraw-Hill, 1990: 435–436.
19. Brule van den AJC, Claas ECJ, Maine du M, et al. The use of anti-contamination primers in the polymerase chain reaction for the detection of human papilloma virus genotypes in cervical scrapes and biopsies. J Med Virol 1989; 29: 20–27.
20. Kleter B, Doorn van LJ, Schegget ter J, et al. Novel short-fragment PCR assay for highly sensitive broad-spectrum detection of anogenital human papilloma viruses. Am J Path 1998; 153: 1731–1739.
21. Doorn van LJ, Quint w, Kleter B, et al. Genotyping of human papillomavirus in liquid cytology cervical specimens by the PGMY line blot assay and the SPF10 line probe assay. J Clin Microbiol 2002; 40: 979–983.
22. Aynaud O, Piron D, Barrasso R, et al. Comparison of clinical, histological and virological symptoms of HPV in HIV-1 infected men and immunocompetent subjects. Sex Transm Infect 1998; 74: 32–34.
23. Palefsky JM, Holly EA, Ralston ML, et al. High incidence of anal high-grade squamous intra-epithelial lesions among HIV-positive and HIV-negative homosexual and bisexual men. AIDS 1998; 12: 495–503.
24. Hagensee ME, Kiviat N, Critchlow CW, et al. Seroprevalence of human papillomavirus types 6 and 16 capsid antibodies in homosexual men. J Infect Dis 1997; 176: 625–631.
25. van Doornum GJJ, Hoek van den JAR, Ameijden van EJC. Cervical human papillomavirus infection among HIV-infected prostitutes addicted to hard drugs. J Med Virol 1993; 41: 185–191.
26. van Doornum GJJ, Hooykaas C, Juffermans LHJ, et al. Prevalence of human papillomavirus infections among heterosexual men and women with multiple sexual partners. J Med Virol 1992; 37: 13–21.
27. Greenblatt RM, Lukehart SA, Plummer FA. Genital ulceration as a risk factor for human immunodeficiency virus infection. AIDS 1998; 2: 46–50.
28. Chen CY, Ballard RC, Beck-Sague CM, Dangor T. Human immunodeficiency virus infection and genital ulcer disease in South Africa: the herpetic connection. Sex Transm Dis 2000; 27: 21–29.
© Copyright 2003 American Sexually Transmitted Diseases Association
29. Franceschi S, Castellsagué X, Dal Maso L, et al. Prevalence and determinants of human papillomavirus genital infection in men. Br J Cancer 2000; 86: 705–711.