What Drives the Number of High-Risk Human Papillomavirus Types in the Anal Canal in HIV-Positive Men Who Have Sex With Men?

MAJOR ARTICLE What Drives the Number of High-Risk Human Papillomavirus Types in the Anal Canal in HIVPositive Men Who Have Sex With Men? Julia Del Amo,1 Cristina González,1 Ronald B. Geskus,8 Montse Torres,2 Jorge Del Romero,3 Pompeyo Viciana,5 Mar Masiá,6 Jose R. Blanco,7 Beatriz Hernández-Novoa,4 and Marta Ortiz,2 on behalf of CoRIS-HPV Study Groupa 1 National Center of Epidemiology and 2National Center of Microbiology, Institute of Health Carlos III, 3Sanitary Center Sandoval, and 4Ramón y Cajal Hospital, Madrid, 5Virgen del Rocío Hospital, Sevilla, 6Elche Universitary Hospital, Alicante, and 7San Pedro–CIBIR Hospital, Logroño, Spain; and 8 Amsterdam Health Service and Academic Medical Center, Netherlands We estimated the effect of sexual behavior, age, and immunodeficiency on the number of high-risk human papillomavirus (HR-HPV) types in the anal canal among human immunodeficiency virus–positive men who have sex with men (MSM). Anal samples were genotyped with the Linear Array HPV Genotyping Test, and risk factors were investigated with Poisson regression. Of 586 MSM, 69% were Spanish, and 25.6% were Latin American; the median age was 34.9 years (interquartile range [IQR], 30.1–40.8). The median number of recent sex partners was 6 (IQR, 2–24 sex partners), and the median CD4+ T-cell count was 531.5 cells/mm3 (IQR, 403–701 cells/mm3). The prevalence of any and multiple HR-HPV infections was 83.4% and 60.5%, respectively. The most common types were HPV-16 (42%), HPV-51 (24%), HPV-39 (23.7%), and HPV-59 (23.5%). Age had a statistically significant, nonlinear association with the number of types, with the highest number detected around 35 years of age (P < .001). The number of recent sex partners had a statistically significant, fairly linear association on the log scale (P = .033). The high prevalence of HR-HPV types is associated with recent sexual behavior and age. Keywords. HR-HPV; HIV; MSMS; sexual behavior; anal cancer. Infection with high-risk human papillomavirus (HRHPV) among human immunodeficiency virus (HIV)– positive men who have sex with men (MSM) is extremely common [1–5]. Around 90% of HIV-positive MSM are coinfected with HR-HPV [1–5], a prevalence that is considerably higher than that among HIV-negative MSM [2, 6, 7]. Furthermore, the prevalence of multiple HR-HPV types among HIV-positive MSM has also been reported to be very high, at nearly 60% [2, 7], which is Received 6 July 2012; accepted 20 November 2012; electronically published 16 January 2013. a Members of the study group are listed at the end of the text. Presented in part: XIXthe International AIDS Conference, Washington, D. C., 20 July 2012. Poster number: TUPE 135. Correspondence: Julia del Amo Valero, MD, PhD, National Center of Epidemiology, Institute of Health Carlos III, Av. Monforte de Lemos no. 5, CP, 28029 Madrid, Spain ( ). The Journal of Infectious Diseases 2013;207:1235–41 © The Author 2013. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: . DOI: 10.1093/infdis/jit028 important because infection with multiple HR-HPV types is associated with a higher prevalence of anal intraepithelial neoplasia in this population [8–11]. Infection with HR-HPV types is a necessary cause of anal carcinoma, an emerging tumor whose incidence is increasing among MSM, particularly those infected with HIV [12–19]. Anal carcinoma is considered an opportunistic tumor, although it is not part of the AIDS case definition, in contrast with cervical carcinoma, which became an AIDS-defining condition in 1993 [20]. The very high prevalence of infection with multiple HR-HPV types among HIV-positive MSM could be driven by an increased persistence of HR-HPV infections due to compromised immunity and/or to a high incidence of new infections driven by sexual behavior, but to date, this has not been established. To address this question, we investigated the effect of self-reported past and current sexual behavior, age, and immunodeficiency on the number of baseline HR-HPV types in the anal canal in HIV-positive MSM. HR-HPV in HIV-Positive MSM • JID 2013:207 (15 April) • 1235 METHODS Statistical Analysis We present descriptive data on overall HR-HPV prevalence and number of HR-HPV types by covariate categories. For every covariate, we present the P value for its effect on HRHPV prevalence and number of HR-HPV types, based on univariable logistic regression models for overall HR-HPV prevalence and univariable Poisson regression models for the number of HR-HPV types. We used multivariable Poisson regression to establish the effect of exposure variables on the number of HR-HPV infections. We considered past and current sexual behavior, current age, and immunodeficiency as the main potential causes for the increased number of HR-HPV and treated the previously described variables as confounders. The number of sex partners had a very skewed distribution, so these values were transformed to the logarithmic scale. Similarly, CD4+ T-cell count was transformed to a square root scale, and HIV load was transformed to a logarithmic scale. We also extended the model with an interaction between CD4+ T-cell count and 3 variables: (1) the number of sex partners in the past 12 months, (2) unsafe anal intercourse, and (3) age at baseline. First, missing values were imputed using the MICE (multiple imputation by chained equations) technique [23]. This was based on a model that, besides the variables in Table 1, also included information on number of cigarettes smoked per day, history of vaginal sex, and results of anal cytology (either conventional or liquid). The effect of all continuous variables was allowed to vary smoothly by using natural cubic splines [24]. We also fit a negative binomial regression model, and results of the parameter estimates did not change, although the overall fit of the model was better. The R statistical computing environment was used for the analyses [25]. HPV DNA Detection and Genotyping Samples were collected with a citobrush and placed in 1 mL of specimen transport medium (Qiagen, Gaithersburg, MD), sent to the Retroviruses and Papillomavirus Unit of the National Center for Microbiology in Madrid, and stored at −20°C until required for testing. Anal HR-HPV infection was genotyped using polymerase chain reaction (PCR) amplification, followed by reverse line blot hybridization, using the Linear Array HPV 1236 • JID 2013:207 (15 April) • Amo et al RESULTS The study included 586 HIV-positive MSM, of whom 405 (69%) were Spanish. The median age was 34.9 years (interquartile range [IQR], 30.1–40.8 years), the median age at first sexual intercourse was 17 years (IQR, 15–18 years), the median lifetime number of sex partners was 100 (IQR, 40–300 CoRIS-HPV is a cohort study within CoRIS, the cohort of the Spanish Network of Excellence on HIV/AIDS Research. CoRIS is an open and multicenter cohort established (...truncated)