Oral HPV Infection in HIV-Infected and HIV-Uninfected Adults
Oral HPV Infection in HIV-Infected and HIV-Uninfected Adults
The Persistent Oral Human Papillomavirus Study (POPS) is an ongoing prospective study nested within 2 longitudinal studies of HIV infection. We included participants from the Chicago, Illinois; Washington, DC/Baltimore, Maryland; and Pittsburgh, Pennsylvania sites of the Multicenter AIDS Cohort Study (MACS) and participants from the Chicago, Bronx (New York City), and Brooklyn (New York City) sites of the Women's Interagency HIV Study (WIHS). A convenience sample of HIV-infected persons and HIV-uninfected persons (who were at risk for HIV and were similar to HIV-infected persons in terms of demographic and behavioral characteristics) was enrolled in the POPS between October 2009 and March 2011, as previously reported. Enrollment was stratified by study site, by HIV status, and by ever use of combination antiretroviral therapy (cART), also known as highly active antiretroviral therapy (HAART). We restricted the current analysis to POPS data collected between 2010 and 2013. POPS participants had similar demographic, behavioral, and biological characteristics as the MACS and WIHS participants, except that more of them were cART-naive. This study's definition of cART was reported use of 3 or more antiretroviral medications. The POPS protocol was approved by the executive committees and institutional review boards of all study sites. All of the participants provided written informed consent.
Participants were followed semiannually for up to 3 years, through March 2013. At each visit (approximately 6 months apart), exfoliated epithelial cells were collected by means of a 30-second oral rinse and gargle sample using Scope mouthwash (Procter & Gamble, Cincinnati, Ohio), which has shown a strong reliability and a higher sensitivity than alternative methods. Participants who preferred not to use Scope (<5% of participants) used saline instead to collect oral exfoliated cells, since both solutions have been shown to have strong DNA yields, DNA quality, and stability at room temperature. Risk factor data were collected semiannually through computer-assisted self-interview in the MACS and through interviewer-administered questionnaires in the WIHS. The study's definition for the number of oral sex partners included all male or female partners that the participant had performed oral sex on (fellatio or cunnilingus). Recent behaviors were defined as those performed in the past 6 months.
Oral rinse samples were stored at 4°C for up to 2 weeks until processed. DNA was purified from the oral rinse using a magnetic bead-based automated platform (QIAsymphony SP; QIAGEN, Germantown, Maryland), as previously described. Purified DNA was evaluated for 37 different HPV DNA genotypes utilizing PGMY09/11 polymerase chain reaction primer pools and primers for β-globin, followed by reverse line blot hybridization to the Roche LINEAR ARRAY HPV Genotyping Test (Roche Molecular Systems, Pleasanton, California). HPV types were classified as either oncogenic (high-risk) or nononcogenic (low-risk) according to the criteria of the International Agency for Research on Cancer. Oncogenic HPV types included types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, and 73, while nononcogenic types included types 6, 11, 26, 40, 42, 53–55, 61, 62, 64, 66, 67, 69–72, 81–84, 89 (CP6108), and IS39.
Type-specific oral HPV infection was classified as prevalent if it was detected at baseline and as incident if it was first detected after a negative type-specific test at baseline. Clearance of type-specific HPV infection was analyzed on the basis of 2 definitions: 1) requiring either a single negative test or 2) requiring 2 consecutive negative tests. Infections that met either definition were considered to have been cleared at the time of the first negative test.
Participant characteristics were compared utilizing χ tests for categorical variables and Mann-Whitney tests for median values for continuous variables. Cumulative incidence and clearance curves were estimated through the Kaplan-Meier method and were utilized to explore the incidence and time to clearance of oral HPV. For the 15% of infections with missing intermittent visits, we assumed that the HPV results were the same at the missing intermittent visit as at the previous visit. Both prevalent infections and re-detected infections (e.g., +, −, −, +) were excluded from incident analyses. All other newly detected infections were classified as incident, although we performed sensitivity analyses comparing sexually abstinent and sexually active participants to examine whether some incidentally detected infections were acquired prior to the study and reactivated during the study. We also conducted sensitivity analyses requiring incident infections to have at least 2 negative tests prior to the first detection of oral HPV, and found the associations with risk factors to be similar.
Risk factors were explored using unadjusted and adjusted Wei-Lin-Weissfeld models, stratifying by HIV status and/or sex. Variables that were significant (P < 0.05) in unadjusted models and variables considered relevant based on previous literature were included in adjusted Wei-Lin-Weissfeld models. Covariates that were strongly correlated were considered in separate models but were also considered in combination to determine which variables to include in the final model. In a sensitivity analysis carried out to explore intermittently detected infections that were variably detected throughout the study (e.g., +, −, +, −, +), we categorized all infections with at least 3 follow-up visits into discrete patterns (persistent, intermittent, and cleared) as previously described. Results were also considered after restricting the data to oncogenic types only and to HPV16 alone, since HPV16 is known to cause most HPV-positive oropharyngeal cancers. All statistical tests were 2-sided, and results were considered significant at an α level of 0.05.
Methods
Study Population and Design
The Persistent Oral Human Papillomavirus Study (POPS) is an ongoing prospective study nested within 2 longitudinal studies of HIV infection. We included participants from the Chicago, Illinois; Washington, DC/Baltimore, Maryland; and Pittsburgh, Pennsylvania sites of the Multicenter AIDS Cohort Study (MACS) and participants from the Chicago, Bronx (New York City), and Brooklyn (New York City) sites of the Women's Interagency HIV Study (WIHS). A convenience sample of HIV-infected persons and HIV-uninfected persons (who were at risk for HIV and were similar to HIV-infected persons in terms of demographic and behavioral characteristics) was enrolled in the POPS between October 2009 and March 2011, as previously reported. Enrollment was stratified by study site, by HIV status, and by ever use of combination antiretroviral therapy (cART), also known as highly active antiretroviral therapy (HAART). We restricted the current analysis to POPS data collected between 2010 and 2013. POPS participants had similar demographic, behavioral, and biological characteristics as the MACS and WIHS participants, except that more of them were cART-naive. This study's definition of cART was reported use of 3 or more antiretroviral medications. The POPS protocol was approved by the executive committees and institutional review boards of all study sites. All of the participants provided written informed consent.
Participants were followed semiannually for up to 3 years, through March 2013. At each visit (approximately 6 months apart), exfoliated epithelial cells were collected by means of a 30-second oral rinse and gargle sample using Scope mouthwash (Procter & Gamble, Cincinnati, Ohio), which has shown a strong reliability and a higher sensitivity than alternative methods. Participants who preferred not to use Scope (<5% of participants) used saline instead to collect oral exfoliated cells, since both solutions have been shown to have strong DNA yields, DNA quality, and stability at room temperature. Risk factor data were collected semiannually through computer-assisted self-interview in the MACS and through interviewer-administered questionnaires in the WIHS. The study's definition for the number of oral sex partners included all male or female partners that the participant had performed oral sex on (fellatio or cunnilingus). Recent behaviors were defined as those performed in the past 6 months.
Laboratory Analysis
Oral rinse samples were stored at 4°C for up to 2 weeks until processed. DNA was purified from the oral rinse using a magnetic bead-based automated platform (QIAsymphony SP; QIAGEN, Germantown, Maryland), as previously described. Purified DNA was evaluated for 37 different HPV DNA genotypes utilizing PGMY09/11 polymerase chain reaction primer pools and primers for β-globin, followed by reverse line blot hybridization to the Roche LINEAR ARRAY HPV Genotyping Test (Roche Molecular Systems, Pleasanton, California). HPV types were classified as either oncogenic (high-risk) or nononcogenic (low-risk) according to the criteria of the International Agency for Research on Cancer. Oncogenic HPV types included types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, and 73, while nononcogenic types included types 6, 11, 26, 40, 42, 53–55, 61, 62, 64, 66, 67, 69–72, 81–84, 89 (CP6108), and IS39.
Statistical Analyses
Type-specific oral HPV infection was classified as prevalent if it was detected at baseline and as incident if it was first detected after a negative type-specific test at baseline. Clearance of type-specific HPV infection was analyzed on the basis of 2 definitions: 1) requiring either a single negative test or 2) requiring 2 consecutive negative tests. Infections that met either definition were considered to have been cleared at the time of the first negative test.
Participant characteristics were compared utilizing χ tests for categorical variables and Mann-Whitney tests for median values for continuous variables. Cumulative incidence and clearance curves were estimated through the Kaplan-Meier method and were utilized to explore the incidence and time to clearance of oral HPV. For the 15% of infections with missing intermittent visits, we assumed that the HPV results were the same at the missing intermittent visit as at the previous visit. Both prevalent infections and re-detected infections (e.g., +, −, −, +) were excluded from incident analyses. All other newly detected infections were classified as incident, although we performed sensitivity analyses comparing sexually abstinent and sexually active participants to examine whether some incidentally detected infections were acquired prior to the study and reactivated during the study. We also conducted sensitivity analyses requiring incident infections to have at least 2 negative tests prior to the first detection of oral HPV, and found the associations with risk factors to be similar.
Risk factors were explored using unadjusted and adjusted Wei-Lin-Weissfeld models, stratifying by HIV status and/or sex. Variables that were significant (P < 0.05) in unadjusted models and variables considered relevant based on previous literature were included in adjusted Wei-Lin-Weissfeld models. Covariates that were strongly correlated were considered in separate models but were also considered in combination to determine which variables to include in the final model. In a sensitivity analysis carried out to explore intermittently detected infections that were variably detected throughout the study (e.g., +, −, +, −, +), we categorized all infections with at least 3 follow-up visits into discrete patterns (persistent, intermittent, and cleared) as previously described. Results were also considered after restricting the data to oncogenic types only and to HPV16 alone, since HPV16 is known to cause most HPV-positive oropharyngeal cancers. All statistical tests were 2-sided, and results were considered significant at an α level of 0.05.
Source...