Human papillomavirus (HPV) is a common sexually transmitted infection and the primary etiologic agent in the development of anogenital cancers and condylomas.1 Factors shown to enhance progression of infection to cancer in women are smoking, long-term oral contraceptive use, and high parity.2 Sexually transmitted infections also have been investigated as possible cofactors because they may directly interact with HPV or damage the epithelial barrier that helps protect against HPV infection.1–3 Studies in women indicate that Chlamydia trachomatis (CT), assessed by the presence of DNA,4,5 may increase the risk of HPV persistence and that CT (assessed by the presence of DNA6 and antibodies7–9) increase the risk of cervical cancer, although some uncertainty remains.10–12 The association between herpes simplex virus type 2 (HSV-2; detected by DNA13,14 or antibodies15,16) and cervical cancer remains equivocal. Little is known about the influence of STIs on HPV natural history in men. One prospective study found that current CT infection increases the risk of acquiring an additional HPV type in HPV-positive men.17 Four cross-sectional studies showed that STIs were associated with current HPV infection,18–21 but not all studies adjusted fully for sexual behavior.
In a study among men,22 we examined the association of CT infection and HSV-2 serostatus with 4 outcomes—any HPV, oncogenic HPV, nononcogenic HPV only, and multiple HPV infections—after adjustment for confounding factors. Furthermore, we assessed whether such associations remained after stratifying by sexual behavior.
MATERIALS AND METHODS
The HPV in Men study is an ongoing prospective study of the natural history of HPV infection in men. Participants were recruited from Sao Paulo (Brazil), Cuernavaca (Mexico), and Tampa (United States), from March 2005 through September 2009. Men were eligible for participation if they were 18 to 70 years of age, reported no history of anogenital cancer or anogenital warts, had not participated in an HPV vaccine study, reported no current penile discharge or dysuria, and were not being treated for an STI. A full description of the study methods and procedures has been published previously.22
Data and Sample Collection
Participants provided written informed consent and underwent a clinical examination 2 weeks (baseline) before the enrollment visit. Participants completed a computer-assisted self-interview questionnaire to determine sociodemographic characteristics, sexual history, condom use practices, tobacco use, self- and partner history of STI, and self- and partner recent diagnosis of an STI or warts. Using 3 saline prewetted Dacron swabs, study clinicians collected exfoliated cells from the coronal sulcus, penile shaft, and scrotum to test for HPV infection. Samples were combined into 1 genital external specimen and stored at −70°C until analysis. Blood samples were collected to test for HSV-2 serostatus, and men provided a first-void urine specimen for CT infection testing. Baseline (visit 1) samples and study characteristics were used for analyses.
Detection of HPV DNA was done by means of polymerase chain reaction (PCR) amplification and genotyping. DNA was extracted with the QIAamp Media MDx (QIAgen, Valencia, CA, USA) by a robotic system, according to the manufacturer’s instructions. DNA was stored at −20°C until use. Thirty nanograms of DNA was amplified with the PGMY09/11 L1 consensus primer system.23 HPV DNA was detected by running the sample on a 2% agarose gel to visualize a band at 450 base pairs. The PCR product was used for HPV genotyping and analyzed by using the Linear Array method (Roche Molecular Diagnostics, Pleasanton, CA, USA).24 Human papillomavirus genotyping was conducted on all samples regardless of HPV PCR results.
Serum for HSV-2 testing was stored at 4°C and analyzed within 24 hours. Detection of HSV-2 IgG antibodies was conducted with the HerpeSelect2 ELISA IgG kit (Focus Diagnostics, Cypress, CA, USA). Samples were considered positive if the optical density was greater than 1.1 times above the cutoff calibrator.
Participants’ first-void urine samples were collected into urine collection cups free of preservatives. Samples were transferred within 24 hours after collection into the GenProbe specimen transport tube. C. trachomatis was tested using the Chlamydia LCx assay (Abbott Laboratories, Abbott Park, Chicago, IL, USA). Samples were concentrated (target capture), amplified (Transcription-Mediated Amplification), and detected (Dual Kinetic Assay) according to the manufacturer’s procedures.
We considered 4 HPV categories in our analyses: any HPV, oncogenic HPV, nononcogenic HPV only, and multiple HPV infections. A participant was considered HPV positive if he tested positive by PCR or tested positive for at least 1 genotype. The oncogenic HPV category included men who were positive for at least 1 of 13 oncogenic types (HVP types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) regardless of coinfection with nononcogenic types. Nononcogenic HPV only included single or multiple infections with nononcogenic HPV types (HPV types 6, 11, 26, 40, 42, 53, 54, 55, 61, 62, 64, 66, 67, 69, 70, 71, 72, 73, 81, 82, 83, 84, IS39, and CP6108) in the absence of oncogenic infection.25 A participant was considered positive for multiple HPV types if he tested HPV positive for more than 1 genotype, regardless of type. Samples were considered valid if they were HPV positive by PCR or by genotyping, or were β-globin positive, regardless of HPV result. In total, 98% (3971/4074) of samples were considered valid. In a sensitivity analysis, similar results were obtained for the full cohort (n = 4074) as for the group with valid HPV samples only (n = 3971).
Pearson χ2 test was used to compare HPV prevalence by CT prevalence and HSV-2 seroprevalence. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using bivariable and multivariable logistic regression analyses. We constructed 4 separate models, one for each HPV outcome: any HPV, oncogenic HPV, nononcogenic HPV only, and multiple HPV infections. Men who had no detectable HPV (HPV negative on both PCR and genotyping assays) served as the reference group for all analyses.
Multivariable logistic regression analyses were performed using 3 different approaches. In approach 1, we used the same set of variables for all models; these were variables that had previously been associated with genital HPV prevalence or acquisition.22,26 This resulted in a list of 6 possible confounders for the CT model: (1) age (3 categories), (2) country of residence, (3) smoking status (never, former, current), (4) education, (5) number of female partners during the past 3 to 6 months, and (6) number of male anal sex partners during the past 3 months. For the HSV-2 model, we also selected possible confounders 1 through 4 and additionally adjusted for confounders 5 (number of lifetime female sex partners) and 6 (number of lifetime male anal sex partners). Recent number of sex partners was used with CT, and lifetime number of sex partners was used with HSV-2, as CT diagnosis indicates current infection, whereas HSV-2 antibody status reflects cumulative lifetime exposure.
In approach 2, a list of potentially confounding variables that were associated (P < 0.25) with each of the 4 HPV outcomes in bivariable analyses was added to the above-named model (e.g., marital status, circumcision status, age of sexual debut, ever diagnosed as having an STI, condom use during vaginal or anal sex in the past 3–6 months, and partner diagnosed as having genital warts or an STI during the past 6 months).
In approach 3, all variables associated (P < 0.25) with each of the 4 HPV outcomes in bivariable analyses were added to the main model and dropped one by one based on the likelihood ratio test at P < 0.05, until a parsimonious model was obtained.
To explore whether sexual behavior modified the association of CT infection and HSV-2 serostatus with the HPV outcomes, we performed a stratified analysis in which we divided the study population into 3 equal-size groups by number of recent sexual partners (RSPs) during the past 3 to 6 months and by number of lifetime sex partners (LSPs), respectively. Stratification was performed by number of partners, regardless of the sex of the partner. Men who refused to answer the number of male or female partners were excluded from the stratified analyses. In addition, stratified analyses by country were performed.
All statistical tests were 2 sided, and all analyses were performed using Stata software version 11.2 (Stata Intercooled, College Station, TX, USA).
Participating men had a median age of 31 years (interquartile range, 23–40) and were recruited from Brazil (35%), Mexico (33%), and the United States (32%). Overall, 2650 (67%) men were positive for any HPV, 1183 (30%) were positive for oncogenic HPV, 898 (23%) were positive for nononcogenic HPV only, and 1146 (29%) were positive for multiple HPV infections.
Population characteristics are presented by CT infection and HSV-2 serostatus in Table 1. In total, 64 (1.6%) men were infected with CT at baseline. C. trachomatis infection was most common (2.4% CT infected) in the youngest age group (18–30 years; P < 0.001). C. trachomatis prevalence was highest in men recruited from Brazil (2.3%), followed by Mexico (1.3%) and the United States (1.2%) (P = 0.038). A higher number of recent female sexual partners during the past 3 to 6 months (P = 0.026) and having a partner diagnosed as having genital warts during the past 6 months (P = 0.028) were significantly associated with CT infection.
In total, 811 (20.4%) men had antibodies against HSV-2. Herpes simplex virus type 2 seroprevalence increased with age (P < 0.001), with men from Brazil having the highest HSV-2 seroprevalence (38.4%). Herpes simplex virus type 2 seroprevalence was associated with marital status (P < 0.001), education (P < 0.001), smoking (P = 0.045), and circumcision (P < 0.001). Herpes simplex virus type 2 seroprevalence was highest in men who reported 20 or more lifetime female sexual partners (P < 0.001) and increased with lifetime number of male anal sexual partners (P < 0.001). All other sexual behavior variables were associated with HSV-2 serostatus, except condom use during vaginal or anal sex during the past 3 to 6 months and number of recent female sexual partners.
Table 2 shows the prevalence of HPV infection by CT infection status and HSV-2 serostatus. Human papillomavirus prevalence was higher in CT-infected men compared with CT-negative men for the following HPV outcomes: any HPV, oncogenic HPV, 1 or more HPV genotypes, multiple HPV genotypes, HPV vaccine types, HPV 11, and HPV 18 (P ≤ 0.05). Human papillomavirus prevalence was higher in HSV-2–seropositive men than in HSV-2–seronegative men for the following HPV outcomes: any HPV, oncogenic HPV, nononcogenic HPV only, 1 or more HPV genotypes, and multiple HPV genotypes (P ≤ 0.05).
Bivariable and multivariable analyses (approach 1) examining the associations between the 4 HPV outcomes and the 2 main exposures (CT infection and HSV-2 serostatus) are presented in Table 3. In bivariable and multivariable analyses, CT infection was independently associated with any HPV (adjusted OR [aOR], 2.19; 95% CI, 1.13–4.24), oncogenic HPV (aOR, 3.10; 95% CI, 1.53–6.28), and multiple HPV infections (aOR, 3.43; 95% CI, 1.69–6.95). In the case of nononcogenic HPV only, the association was marginally significant in the bivariable analysis (crude OR, 2.16; 95% CI, 1.00–4.68; P = 0.050) and was not significant in the multivariable analysis (aOR, 1.95; 95% CI, 0.88–4.32; P = 0.101). Herpes simplex virus type 2 seropositivity was significantly associated with any HPV (aOR, 1.25; 95% CI, 1.02–1.52), nononcogenic HPV only (aOR, 1.38; 95% CI, 1.08–1.75), and multiple HPV infections (aOR, 1.33; 95% CI, 1.06–1.68). However, in the case of oncogenic HPV, HSV-2 serostatus was significantly associated in the bivariable analysis (crude OR, 1.52; 95% CI, 1.24–1.85) and marginally significant in multivariable analyses (aOR, 1.26; 95% CI, 1.00–1.59; P = 0.051).
Multivariable logistic regression analyses were performed in 2 additional ways (see “Materials and Methods”). In approach 2, the addition of the covariates mentioned earlier did not improve the model based on the likelihood ratio test and did not significantly alter the association between CT infection and HSV-2 serostatus with the 4 HPV outcomes by more than 5%. Using approach 3, the association between multiple HPV infections and HSV-2 seropositivity was not significant (aOR, 1.25; 95% CI, 0.99–1.57; P = 0.056); all other results were comparable.
Stratified analyses for the association between CT infection and the 4 HPV outcomes are presented in Table 4. A significant association for all outcomes was found in the group that reported 2 or more RSPs, except for the association between CT infection and nononcogenic HPV only (aOR, 3.22; 95% CI, 0.86–12.06). Stratified analyses by LSPs for the association between HSV-2 serostatus and the 4 HPV outcomeffs are shown in Table 5. Herpes simplex virus type 2 serostatus was significantly associated with the 4 HPV outcomes among men reporting 0 to 5 LSPs, but this association was weaker among men reporting 6 to 14 LSPs and absent among men reporting 15 or more LSPs. The same trend was observed for the 3 other HPV outcomes.
In additional stratified analyses by country, CT was positively associated with HPV in each country; however, the strength of these associations was highest in the United States, followed by Mexico, and weakest in Brazil. Herpes simplex virus type 2 was positively associated with HPV in each country (data not shown).
This is one of the first studies investigating the association of genital HPV infection with CT infection and HSV-2 serostatus in men. After adjusting for confounding factors, CT infection was positively associated with any HPV, oncogenic HPV, and multiple HPV infections; HSV-2 was positively associated with any HPV, nononcogenic HPV, and multiple HPV infections. In expanded models adjusting for additional variables generally focused on sexual behavior, ORs for CT and HSV-2 were unchanged and remained significant. Thus, additional correction for sexual behavior did not alter our conclusion, suggesting that in this cross-sectional analysis, confounding caused by sexual behavior was well controlled.
We measured prevalent genital HPV infection, whereas HSV-2 seropositivity represents cumulative sexual exposure. The association between 2 markers reflecting different stages of infection may be questioned. However, because HSV-2 after a primary infection leads to clinical recurrences, HSV-2 seropositive individuals may have frequent reactivations of genital HSV-2 eruptions (in another study 54% of the men with genital HSV-2 infection or HSV-2 antibodies had 6 recurrences per year).27
The association we observed between CT infection and HPV infection may raise some questions because these infections affect different anatomical sites: CT infection is located in the urethra, whereas HPV infection is mostly located at the coronal sulcus, penile shaft, and scrotum. The biological mechanism responsible for an association between these 2 STIs therefore remains to be elucidated.
Men who reported a current penile discharge or dysuria and men who were being treated for an STI were excluded from participation in this study. Therefore, we are only measuring the associations between asymptomatic CT and HPV infections. Although this is a limitation, greater than 50% of CT infections in men are asymptomatic.28 This explains why, despite the exclusion criteria, we observed 1.6% of men had a CT infection.
A limitation of the current study is the inability to address temporality. It is unclear whether the STI was transmitted concurrently with HPV, or preceded or followed HPV. In the case of HSV-2, temporality may not be an issue because HSV-2 seroprevalence indicates past infections that may have influenced the current HPV infection. Although HSV-2 assessed with antibodies does not distinguish between oral, anal, and genital HSV-2 infection, HSV-2 seropositivity indicates genital herpes in almost all patients.29 Because both CT and HPV infections are current infections, it is challenging to draw conclusions on temporality of these 2 infections. Another limitation is the relatively low prevalence of each HPV type; therefore, it was not possible to investigate CT/HSV-2 associations with individual HPV types.
Our study concerns a well-characterized international cohort of men with a wide age range, a large sample size, extensive data regarding demographic and sexual behavior characteristics, high rate of specimen adequacy (98%), and comprehensive laboratory ascertainment. As such, the study provides a good starting point to investigate the association of CT infection and HSV-2 serostatus with HPV in more detail.
Our results are in good agreement with previous studies in women.10,30 A prospective study among men from Denmark found CT infection to be a risk factor for acquiring an additional HPV type in HPV-positive men.17 A recent study investigating heterosexual couples in Rwanda found a moderate association between high-risk HPV infection and HSV-2 seropositivity (OR, 2.4; 95% CI, 1.2–5.0) in men.19 A Japanese study in men found CT not to be associated with HPV in a bivariable analysis, although in a multivariable analysis, STIs were significantly associated with HPV infection (OR, 2.41; 95% CI, 1.04–5.61).20 In a study among male attendees of the STI clinic in the Netherlands, an association between history of chlamydia and gonorrhea with HPV infection was observed.18 Among Kenyan men, CT infection was significantly independently associated with HPV presence in the glans (aOR, 1.87; 95% CI, 1.24–2.81) but not the shaft, and HSV-2 serostatus was only marginally independently associated with HPV presence in the glans (aOR, 1.25; 95% CI, 1.01–2.81).21
In stratified analyses regarding CT and HPV infection, the strength of the association increased with increasing number of RSPs. Men with 2 or more RSPs may be acquiring both STIs simultaneously from their high-risk partners. Because the association among men with 2 or more RSPs was stronger than in the whole study population, this may indicate residual confounding caused by sexual behavior. However, it is also possible that CT increases the risk of HPV acquisition or persistence through a biological interaction. In an analogy of the hypothesized interaction between CT and HPV at the cervix, CT induces an inflammatory reaction that indirectly may increase HPV persistence through decreases in cell-mediated immunity.3 When stratifying by LSPs, the strength of the HSV-2 and HPV association increased with decreasing number of LSPs. In the case of HSV-2, the biological interaction in men reporting 6 or more LSPs may be obscured by riskier sexual behavior that overwhelms the added risk conferred by HSV-2.
Further prospective studies in men are needed to clarify the role of STIs as a cofactor in HPV natural history, specifically HPV acquisition and persistence. There may be concern that STIs are surrogates for higher-risk behaviors that increase the exposure to HPV. Our stratified analyses for HSV-2 indicate the contrary. Prospective studies in men are needed in which multiple measures of HPV infections and STIs would help differentiate STIs as true HPV cofactors from STIs, or as indicators of sexual behavior. Overall, these results show that after correcting for sexual behavior, men infected with CT and men seropositive for HSV-2 are at higher risk for HPV infection.
1. WHO. IARC monographs on the evaluation of carcinogenic risks to humans: Human papillomaviruses, vol. 90. IARC: Lyon, 2007.
2. Castellsague X, Bosch FX, Munoz N. Environmental co-factors in HPV carcinogenesis. Virus Res 2002; 89: 191–199.
3. Castle PE, Giuliano AR. Chapter 4: Genital tract infections, cervical inflammation, and antioxidant nutrients—assessing their roles as human papillomavirus cofactors. J Natl Cancer Inst Monogr 2003; (31): 29–34.
4. Kim S, Arduino JM, Roberts CC, et al. Incidence and predictors of human papillomavirus-6, -11, -16, and -18 infection in young norwegian women. Sex Transm Dis 2011; 38: 587–597.
5. Samoff E, Koumans EH, Markowitz LE, et al. Association of Chlamydia trachomatis
with persistence of high-risk types of human papillomavirus in a cohort of female adolescents. Am J Epidemiol 2005; 162: 668–675.
6. Lehtinen M, Ault KA, Lyytikainen E, et al. Chlamydia trachomatis
infection and risk of cervical intraepithelial neoplasia. Sex Transm Infect 2011; 87: 372–376.
7. Arnheim DL, Andersson K, Luostarinen T, et al. Prospective seroepidemiologic study of human papillomavirus and other risk factors in cervical cancer. Cancer Epidemiol Biomarkers Prev 2011; 20: 2541–2550.
8. Smith JS, Bosetti C, Munoz N, et al. Chlamydia trachomatis
and invasive cervical cancer: A pooled analysis of the IARC multicentric case-control study. Int J Cancer 2004; 111: 431–439.
9. Anttila T, Saikku P, Koskela P, et al. Serotypes of Chlamydia trachomatis
and risk for development of cervical squamous cell carcinoma. JAMA 2001; 285: 47–51.
10. Oakeshott P, Aghaizu A, Reid F, et al. Frequency and risk factors for prevalent, incident, and persistent genital carcinogenic human papillomavirus infection in sexually active women: Community based cohort study. BMJ 2012; 344: e4168.
11. Safaeian M, Quint K, Schiffman M, et al. Chlamydia trachomatis
and risk of prevalent and incident cervical premalignancy in a population-based cohort. J Natl Cancer Inst 2010; 102: 1794–1804.
12. Castle PE, Escoffery C, Schachter J, et al. Chlamydia trachomatis
, herpes simplex virus 2, and human T-cell lymphotrophic virus type 1 are not associated with grade of cervical neoplasia in Jamaican colposcopy patients. Sex Transm Dis 2003; 30: 575–580.
13. Koffa M, Koumantakis E, Ergazaki M, et al. Association of herpesvirus infection with the development of genital cancer. Int J Cancer 1995; 63: 58–62.
14. Tran-Thanh D, Provencher D, Koushik A, et al. Herpes simplex virus type II is not a cofactor to human papillomavirus in cancer of the uterine cervix. Am J Obstet Gynecol 2003; 188: 129–134.
15. Smith JS, Herrero R, Bosetti C, et al. Herpes simplex virus-2 as a human papillomavirus cofactor in the etiology of invasive cervical cancer. J Natl Cancer Inst 2002; 94: 1604–1613.
16. Lehtinen M, Koskela P, Jellum E, et al. Herpes simplex virus and risk of cervical cancer: A longitudinal, nested case-control study in the nordic countries. Am J Epidemiol 2002; 156: 687–692.
17. Kjaer SK, Munk C, Winther JF, et al. Acquisition and persistence of human papillomavirus infection in younger men: A prospective follow-up study among Danish soldiers. Cancer Epidemiol Biomarkers Prev 2005; 14: 1528–1533.
18. Vriend HJ, Boot HJ, van der Sande MA. Type-specific human papillomavirus infections among young heterosexual male and female STI clinic attendees. Sex Transm Dis 2012; 39: 72–78.
19. Veldhuijzen NJ, Dhont N, Vyankandondera J, et al. Prevalence and concordance of HPV, HIV, and HSV-2 in heterosexual couples in Kigali, Rwanda. Sex Transm Dis 2012; 39: 128–135.
20. Shigehara K, Kawaguchi S, Sasagawa T, et al. Prevalence of genital mycoplasma, ureaplasma, gardnerella, and human papillomavirus in Japanese men with urethritis, and risk factors for detection of urethral human papillomavirus infection. J Infect Chemother 2011; 17: 487–492.
21. Smith JS, Backes DM, Hudgens MG, et al. Prevalence and risk factors of human papillomavirus infection by penile site in uncircumcised Kenyan men. Int J Cancer 2010; 126: 572–577.
22. 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: 932–940.
23. Gravitt PE, Peyton CL, Alessi TQ, et al. Improved amplification of genital human papillomaviruses. J Clin Microbiol 2000; 38: 357–361.
24. Gravitt PE, Peyton CL, Apple RJ, et al. 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: 3020–3027.
25. Bouvard V, Baan R, Straif K, et al. A review of human carcinogens—part B: Biological agents. Lancet Oncol 2009; 10: 321–322.
26. Giuliano AR, Lazcano E, Villa LL, et al. Circumcision and sexual behavior: Factors independently associated with human papillomavirus detection among men in the HIM study. Int J Cancer 2009; 124: 1251–1257.
27. Wald A, Zeh J, Selke S, et al. Genital shedding of herpes simplex virus among men. J Infect Dis 2002; 186 (suppl 1): S34–S39.
28. Manavi K. A review on infection with Chlamydia trachomatis
. Best Pract Res Clin Obstet Gynaecol 2006; 20: 941–951.
29. Wald A, Ashley-Morrow R. Serological testing for herpes simplex virus (HSV)-1 and HSV-2 infection. Clin Infect Dis 2002; 35: S173–S182.
© Copyright 2013 American Sexually Transmitted Diseases Association
30. Howell-Jones R, de SN, Akpan M, et al. Prevalence of human papillomavirus (HPV) infections in sexually active adolescents and young women in England, prior to widespread HPV immunisation. Vaccine 2012; 30: 3867–3875.