Genital infection with human papillomaviruses (HPVs) can lead to cervical, vulvar, vaginal, anal, and other cancers . Co-infection with HIV amplifies the burden of HPV infection. HPV prevalence and short-term persistence are high among women with HIV [2–6]. Further, women with HIV have a 77% cumulative risk of abnormal cervical cytology after up to 10 years of observation . Long-term cumulative HPV detection in women with HIV may be substantially higher, since many HPV infections do not result in abnormal cytology, but long-term surveillance studies have not been reported. Long-term type-specific infection rates are unknown.
The main purpose of this analysis was to estimate the cumulative detection of HPV infection in a cohort of HIV-seropositive women and at-risk comparison of HIV-seronegative women. Although prevalence and short-term incidence data have been published previously, we have not assessed long-term cumulative detection. In addition, this analysis included women recruited during a second enrollment round, with follow-up extended up to 8 years.
Materials and methods
The study was part of the Women's Interagency HIV Study (WIHS), an ongoing observational multicenter cohort study of the health of HIV-seropositive women and at-risk HIV-uninfected comparison women. Enrollment was initially conducted between October 1994 and November 1995 (2055 HIV-seropositive, 569 seronegative women), and a second cohort was similarly enrolled during 2002 (738 HIV-seropositive, 406 seronegative women) [8,9]. After the local human subjects committees reviewed and approved the study, all participants gave written informed consent for data collection. Follow-up continues, but this analysis includes information obtained between 1 October 1994 and 30 March 2005.
Every 6 months, participants had a physical examination that included cervicovaginal lavage (CVL) with 10 ml of saline. The current data reflect HPV DNA testing of over 34 000 CVL specimens conducted to assess the natural history of HPV infection in HIV-seropositive and seronegative women, after which more targeted testing was conducted which focused on the development of severe cervical neoplasia. Briefly, in the Natural History Study, routine testing of all available samples from all WIHS women enrolled during 1994–1995 continued through 8.5 years of follow-up (>27 000 CVLs), and then continued in a random sample of 300 women up to 12 years. For those enrolled during 2001–2002, testing in all samples from all women continued for 3.5 years with a random sample of 200 through 5 years. The number of women who contributed data to each time point by HIV status is shown in Fig. 1.
Protocols for HPV testing have been described previously [2,3]. Briefly, MY09/MY11 consensus primers PCR amplification was followed by hybridization with consensus and HPV type-specific probes. Successful amplification of the β-globin gene during PCR was used to assess specimen adequacy. All β-globin-negative specimens were excluded, and rates were calculated based on the number of β-globin-positive results to avoid positive bias in cumulative prevalence estimates. Results were classified as defined by the International Association for Research on Cancer, including any carcinogenic type (types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68), for any type, and negative for HPV. These investigational results were not used in patient management.
We defined the baseline visit for any individual as the first visit with an adequate (i.e. β-globin-positive) HPV result, regardless of the time of entry to WIHS or chronological date. We excluded women with no interpretable HPV results, HIV-seronegative women who seroconverted during follow-up, and those who reported a prior hysterectomy at their baseline visit. Women were not censored at the time of cervical disease treatment, since they remained at risk for new or recurrent cervical HPV infection; however, they were censored if they had a hysterectomy during follow-up.
Contingency tables were generated to assess baseline patient characteristics by HIV serostatus. Pearson's chi-square tests were used to compare baseline characteristics between HIV-seropositive and seronegative women, including baseline HPV prevalence. Mantel–Haenszel chi-square tests were used to compare HPV prevalence at baseline across HIV/CD4+ strata. Wilcoxon rank-sum tests were used to compare medians. We assessed the cumulative detection of HPV across visits, defined as the proportion of women ever positive for any HPV or for a given HPV type. Cumulative detection of HPV across sequential mid-visit times was estimated using the Kaplan–Meier method (1 – proportion surviving free of HPV); results for women with and without HIV were compared. Normal approximations to the log ratio of two cumulative detections were used to obtain 95% confidence intervals (CIs) for each HPV type and for all HPV types. P values based on the same approximations were given for overall cumulative detections. To assess how annual detection changed with age, we charted annual HPV cumulative detection across age strata for all HPV types and for HPV16/non-16 carcinogenic types/noncarcinogenic types. Generalized estimating equation (GEE) models with logit link were used to study the multivariable association of risk factors with cumulative detection of HPV. These models focused on the 5-year cumulative prevalence, the mid-point of the 10-year follow-up period, and the last time for which point data were available from both the 1994–95 and 2002 subcohorts. All tests were two-sided with significance set at P value less than 0.05.
Among all the 3766 women in the WIHS (2791 HIV-seropositive, 975 seronegative), 22 HIV seroconverters were excluded; 249 women (206 HIV-seropositive, 43 seronegative) were excluded due to hysterectomy prior to their baseline visits; 57 women (42 HIV-seropositive, 15 seronegative) were excluded because they had no HPV results from any visit. The number of women eligible for analysis of prevalence and cumulative detection was thus 3438 (2543 HIV-seronegative, 895 seronegative). Median follow-up in those enrolled during 1994–1995 was 6.8 years among HIV-seropositive and 6.9 among HIV-seronegative patients, whereas it was 2.4 years for HIV-seropositive and seronegative women enrolled during 2002. Overall, the mean (median/minimum) percentage (%) of: WIHS women who had a CVL obtained was 95% (95%/91%); women with a CVL who were HPV-tested at each visit was 95% (97%/76%); HPV tests that were adequate at each visit was 96% (96%/92%).
The characteristics of women included in this study group at their first visits with valid HPV test results are shown in Table 1. The median age of HIV-seropositive women was 35 years, whereas that of seronegative women was 32 years (P < 0.0001). Although the distribution of lifetime number of sex partners in HIV-seropositive and seronegative women differed, there was no linear pattern to this, and the median number was 10 in both groups (P = 0.25). The number of sex partners in the past 6 months, however, was greater in HIV-seronegative than in HIV-seropositive women. Compared to HIV-seronegative women, seropositive women were less likely to smoke, but were more likely to have ever used intravenous drugs. The low rate of use of HAART by HIV-seropositive women reflects treatment standards at the time of enrollment (0.4% in those recruited in 1994–1995); it has increased with time .
Figure 1 shows HPV cumulative prevalence at baseline and at each subsequent visit through the 8-year semiannual visit sequence. The prevalence of any HPV infection at the first visit with HPV results was 1333/2543 (52%) among HIV-seropositive women and 199/895 (22%) among seronegative women (P < 0.0001). When examined by CD4+ stratum among HIV-seropositive women, the prevalence of HPV of any type was 297/814 (36%) among women with CD4+ cell counts greater than 500/μl, 559/1052 (53%) among women with CD4+ cell counts between 200 and 500/μl, and 434/604 (72%) among women with CD4+ cell counts less than 200/μl (P < 0.0001). The prevalence of carcinogenic HPV was 716/2543 (28%) among HIV-seropositive women and 93/895 (10%) among seronegative women (P < 0.0001). Carcinogenic HPV types were found in 131/814 (16%) HIV-seropositive women with CD4+ cell counts greater than 500/μl, 299/1052 (28%) of those with CD4+ cell counts between 200 and 500/μl, and 262/604 (43%) of those with CD4+ cell counts less than 200/μl (P < 0.0001). The prevalence of noncarcinogenic HPV was 1038/2543 (41%) among HIV-seropositive women and 136/895 (15%) among seronegative women (P < 0.0001).
Cumulative HPV detection is shown in Fig. 1, censoring at loss to follow-up or hysterectomy, showing a progressive rise in cumulative detection both for any and for specifically carcinogenic HPV detection among both HIV-seropositive and seronegative women. The cumulative incident detection of HPV among seronegative women lagged that of seropositive women by several years. Among HIV-seropositive women, the cumulative detection of any HPV infection was 68% at 1 year, but rose to 75% at 2 years, 86% at 5 years, and 92% at 8 years; similar rates in HIV-seronegative women were 34, 43, 57, and 66% (all P < 0.0001 vs. HIV-seropositive women). In HIV-seropositive women, cumulative detection of carcinogenic HPV rose to 40% at 1 year, 47% at 2 years, 61% at 5 years, and 67% at 8 years; similar rates in HIV-seronegative women were 16, 20, 20, and 36% (again, all P < 0.0001 vs. HIV-seropositive women). Cumulative detection of noncarcinogenic HPV types is not shown in the figure, but among HIV-seropositive women it was 59% at 1 year, 67% at 2 years, 82% at 5 years, and 89% at 8 years; similar cumulative detection rates among HIV-seronegative women were 25, 35, 48, and 56% (P < 0.0001).
We looked specifically at how the cumulative detection of HPV varied between HIV-seropositive and seronegative women across hierarchical groupings of carcinogenicity. The ratio of 8-year cumulative detection of HPV16 among HIV seropositive to seronegative women was 2.26 (95% CI 1.63, 3.14) and for HPV18 was 2.44 (95% CI 1.69, 3.53). For non-16/18 carcinogenic HPV types, the ratio was 3.21 (95% CI 2.69, 3.73), which is higher than for HPV16/18 (P = 0.03). The ratio was also higher for noncarcinogenic types than for HPV16/18, at 4. (95% CI 3.62, 5.56, P = 0.01).
In multivariable regression analyses, as shown in Table 2, enrollment characteristics correlated with a higher cumulative detection of any HPV by 5 years included CD4+ stratum among HIV-seropositive women, younger age, and enrollment cohort. Current smoking increased and former smoking decreased cumulative detection of HPV. Lifetime number of sexual partners was not associated with cumulative HPV detection. In a second analysis, we additionally assessed the impact of enrollment period on these findings by incorporating it as a covariate in our models. Although there was a small association between this enrollment period and cumulative prevalence [odds ratio (OR) 1.15; 95% CI 1.04, 1.26], its inclusion in the model had no discernable impact on other effect estimates in the model (data not shown), and enrollment period is not shown as a variable in Table 2.
Type-specific 8-year cumulative HPV detection rates are shown in Table 3. The most common HPV types in HIV-seropositive women were noncarcinogenic: type 53 (25.3%), type 61 (24.2%), type 81 (20.9%), and type 62(20.7%). In contrast, the HPV types most commonly detected over 8 years of follow-up in HIV-seronegative women included the noncarcinogenic type 53 (9.0%) and type 81 (7.5%), but also the carcinogenic type 56 (7.7%). HPV16 was detected over 8 years in 15.2% of the HIV-seropositive and 6.7% of seronegative women, whereas HPV18 was detected in 15.0% of HIV-seropositive and 6.1% of seronegative women (P < 0.0001 for both). Other types were detected in 66.2% of HIV-seropositive and 35.1% of seronegative women. Differences in cumulative HPV detection between HIV-seropositive and seronegative women were significant at P value less than 0.05 for all types except type 6969.
Human papillomavirus acquisition is very common among sexually active women, and HIV-seropositive women have among the highest HPV rates reported. These results expand upon our previous reports that HIV infection increases HPV detection [2,3]. With follow-up for some women now extending to 8 years, cumulative HPV detection among HIV-seropositive women surpassed 90%. Two-thirds of the HIV-seropositive women had carcinogenic HPV detected over 8 years.
Despite high cumulative HPV detection and a high frequency of abnormal Pap tests, prior study has shown that most abnormal Paps in HIV-seropositive women are atypical or low grade . In registry studies, cervical cancer risk among these women is only modestly higher than that in the general population . In WIHS, which included aggressive screening and treatment of precursors and expert pathology review of reported cervical cancers, the increased risk of cervical cancer in HIV-seropositive compared to seronegative women did not reach statistical significance .
Across time, almost 90% of HIV-seropositive women in this study experienced noncarcinogenic HPV infections, significantly greater cumulative detection than was observed for carcinogenic HPVs. Clifford et al. have reported in a meta-analysis that the noncarcinogenic HPV types 53, 61, and 83 were more common than carcinogenic types in North American women with HIV. They also reported that women with HIV and high-grade Pap test abnormalities were more likely to have HPV types other than HPV16. In the general population, HPV types 16 and 18 account for the majority of cancers and true precancers, whereas noncarcinogenic types do not increase cancer risk [14,15].However, data in the WIHS have shown that the prevalence of HPV16 is the least affected by changes in immune status of any carcinogenic HPV among HIV-seropositive women [4,16]. Most cervical cytologic lesions in HIV-seropositive women who are in care and have regular Pap testing are atypical or low-grade, whereas high-grade lesions are not common and cancers are rare [7,12]. A high cumulative detection of noncarcinogenic HPV may contribute to this difference, although correlative studies linking individual cytology and HPV results are needed.
The cumulative detection of HPV in HIV-seronegative women in our cohort lagged behind that among HIV-seropositive women by several years; yet it still rose to 66% by 8 years, with carcinogenic HPV found in more than a third during the same period. The high rates of HPV detection in HIV-seropositive women likely represent background HPV exposure rates in sexually active women. At-risk HIV-seronegative women in the WIHS reported more sexual activity and were more likely to report multiple partners than HIV-seropositive women [17,18], so lower exposure risk does not explain lower HPV detection in HIV-seronegative women. Higher HPV detection rates in HIV-seropositive women reflect both more reactivation of prior infections acquired during more sexually active phases of their lives and ongoing sexual exposure. Conversely, lower HPV detection rates in HIV-seronegative women likely reflect lower reactivation rates and similar HPV exposures that result in less detectable infection because host immune mechanisms either prevented infections or cleared them between HPV tests .
These high rates of cumulative HPV detection over time are consistent with prior reports, although our surveillance extends further. Despite excluding women with prevalent cervical disease, Sun et al. found that cumulative detection of HPV after up to 2 years of observation was 83% in HIV-infected women and 62% in uninfected women. Over 3 years of observation, Tornesello et al. found HPV in 39% of HIV-infected and 14% of HIV-uninfected women. Schneider et al. found a cumulative detection of HPV infection of 66% in presumably HIV-uninfected young women despite 5 years of normal Pap results, whereas Winer et al. reported a 32% cumulative incidence in the US university students followed for 2 years.
We found that cumulative detection of HPV was not related to lifetime number of partners. Although this may seem counterintuitive, since HPV is sexually transmitted, HPV incidence is more closely related to recent sexual activity, as even women with impaired immunity can clear HPV over time . Women with large numbers of lifetime partners who are too ill for sex are unlikely to acquire new HPV infections, whereas women with relatively few lifetime partners, but recent new partners, are likely to acquire new infections.
Our study was limited by several factors. At enrollment, most women were over age 30 years and so were beyond the age of highest risk for HPV acquisition, especially acquisition of types 16 and 18 [22,23]. In addition, we only tested for HPV at 6-month intervals, whereas some infections might have been acquired and cleared between visits, and some women were lost to follow-up for variable intervals. Identification and treatment of lesions related to HPV types 16 and 18 prior to WIHS enrollment may have increased the proportion of non-16/18 carcinogenic HPV types detected during follow-up. We were unable to assess the impact of prior treatment on subsequent HPV detection, as this was captured only by self-report and appeared unreliable. We have previously shown that antiretroviral therapy reduces HPV infection, but mainly in women who are highly adherent ; because of the complexity of assessing adherence, we did not explore the impact of antiretroviral therapy on cumulative detection. Women were censored from follow-up at hysterectomy because hysterectomy distorts the pattern of HPV infection, favoring noncarcinogenic types . Nevertheless, women might have continued to contribute after hysterectomy to cumulative HPV detection. Life-long cumulative HPV detection is likely to be even greater than our results indicate.
Persistence rather than incidence of HPV infection appears to determine cancer risk; additional study is needed to define which types are most persistent, how HIV-induced immunosuppression influences HPV persistence as well as incidence, and how persistent infections are correlated with development of precancer and cancer in immunocompromised women. We could not determine whether the higher rate of HPV detection in HIV-seropositive women was due to more persistent infections that were more likely to be detected at semiannual visits, to more frequent reactivation of previously acquired infections, or to greater susceptibility to HPV infection. Furthermore, condom use can decrease HPV acquisition and speed clearance [26,27]; because of limitations to our data on consistency of condom use across partners and time, we did not explore whether condom use alters HPV risk over time. Further studies should investigate the effectiveness of condoms in reducing HPV acquisition among HIV-seropositive women.
The high cumulative detection of HPV in HIV-seropositive women underscores the importance of education for these women and the clinicians who care for them. HPV-related cervical, vaginal, and vulvar lesions are common, often chronic problems in HIV-seropositive women. The most commonly detected HPV types over time in HIV-seropositive women were noncarcinogenic, potentially causing cytologic changes and cervical lesions, but posing minimal risk for cancer. Balancing the importance of treatment for true cancer precursors against the discomfort and disruption of repeated colposcopy and cervical interventions will require a deeper understanding of the natural history of HPV in HIV-seropositive women. Our group is exploring the links between HIV-related immunocompromise, HPV clearance and persistence, and their relationship to progression to cervical and other lower genital tract cancers.
HPV DNA testing and statistical analysis was supported by R01-CA-085178 (H.D.S.) and the Einstein-Montefiore Center for AIDS funded by the NIH (AI-51519) and the Einstein Cancer Research Center (P30CA013330) from the National Cancer Institute.
Data in this manuscript were collected by the Women's Interagency HIV Study (WIHS).
WIHS (Principal Investigators): UAB-MS WIHS (Michael Saag, Mirjam-Colette Kempf, and Deborah Konkle-Parker), U01-AI-103401; Atlanta WIHS (Ighovwerha Ofotokun and Gina Wingood), U01-AI-103408; Bronx WIHS (Kathryn Anastos), U01-AI-035004; Brooklyn WIHS (Howard Minkoff and Deborah Gustafson), U01-AI-031834; Chicago WIHS (Mardge Cohen), U01-AI-034993; Metropolitan Washington WIHS (Mary Young), U01-AI-034994; Miami WIHS (Margaret Fischl and Lisa Metsch), U01-AI-103397; UNC WIHS (Adaora Adimora), U01-AI-103390; Connie Wofsy Women's HIV Study, Northern California (Ruth Greenblatt, Bradley Aouizerat, and Phyllis Tien), U01-AI-034989; WIHS Data Management and Analysis Center (Stephen Gange and Elizabeth Golub), U01-AI-042590; Southern California WIHS (Alexandra Levine and Marek Nowicki), U01-HD-032632 (WIHS I - WIHS IV). The WIHS is funded primarily by the National Institute of Allergy and Infectious Diseases (NIAID), with additional co-funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), the National Cancer Institute (NCI), the National Institute on Drug Abuse (NIDA), and the National Institute on Mental Health (NIMH). Targeted supplemental funding for specific projects is also provided by the National Institute of Dental and Craniofacial Research (NIDCR), the National Institute on Alcohol Abuse and Alcoholism (NIAAA), the National Institute on Deafness and other Communication Disorders (NIDCD), and the NIH Office of Research on Women's Health. WIHS data collection is also supported by UL1-TR000004 (UCSF CTSA) and UL1-TR000454 (Atlanta CTSA).
Roles of authors: All co-authors were involved in conception and design, and drafting and revision of the manuscript for important intellectual content. In addition, X.X did statistical calculations and helped guide analyses. G.D'S. helped focus analyses based on HPV expertise. H.M., A.M.L., M.Y., and J.C. are principal investigators for the WIHS and supervised the overall study. D.H.W. performed critical executive functions for WIHS as a program representative from NICHD. J.P. and R.B. performed HPV testing and provided guidance about terminology and focused discussion. M.J.K. provided patient care and oversaw local site gynecologic management and patient data acquisition critical to the conduct of the work and assisted in describing the relevance of findings to clinical care.
Conflicts of interest
The contents of this publication are solely the responsibility of the authors and do not represent the official views of the United States Government.
There are no conflicts of interest.
1. Schiffman M, Herrero R, DeSalle R, Hildesheim A, Wacholder S, Rodriguez AC, et al. The carcinogenicity of human papillomavirus types reflects viral evolution
2. Palefsky JM, Minkoff H, Kalish LA, Levine A, Sacks HS, Garcia P, et al. Cervicovaginal human papillomavirus infection in human immunodeficiency virus-1 (HIV)-positive and high-risk HIV-negative women
. J Natl Cancer Inst
3. Strickler HD, Burk RD, Fazzari M, Anastos K, Minkoff H, Massad LS, et al. Natural history and possible reactivation of human papillomavirus in human immunodeficiency virus (HIV) positive women
. J Natl Cancer Inst
4. Strickler HD, Palefsky JM, Shah KV, Anastos K, Klein RS, Minkoff H, et al. HPV type 16 and immune status in human immunodeficiency virus seropositive women
. J Natl Cancer Inst
5. Sun XW, Kuhn L, Ellerbrock TV, Chiasson MA, Bush TJ, Wright TC. Human papillomavirus infection in women infected with the human immunodeficiency virus
. New Engl J Med
6. Jamieson DJ, Duerr A, Burk R, Klein RS, Paramsothy P, Schuman P, et al. for the HIV Epidemiology Research Study Group. Characterization of genital human papillomavirus infection in women who have or who are at risk of having HIV infection
. Am J Obstet Gynecol
7. Massad LS, Seaberg EC, Wright RL, Darragh T, Lee YC, Colie C, et al. Squamous cervical lesions in women with human immunodeficiency virus: long-term follow up
. Obstet Gynecol
8. Barkan SE, Melnick SL, Martin-Preston S, Weber K, Kalish LA, Miotti P, et al. The Women's Interagency HIV Study
9. Bacon M, von Wyl V, Alden C, Sharp G, Robison E, Hessol N, et al. for the Women's Interagency HIV Study: an observational cohort brings clinical sciences to the bench
. Clin Diag Lab Immunol
10. Gange SJ, Barron Y, Greenblatt RM, Anastos K, Minkoff H, Young M, et al. Effectiveness of highly active antiretroviral therapy among HIV-1 infected women
. J Epidemiol Commun Health
11. Engels EA, Biggar RJ, Hall HI, Cross H, Crutchfield A, Finch JL, et al. Cancer risk in people infected with human immunodeficiency virus in the United States
. Int J Cancer
12. Massad LS, Seaberg EC, Watts DH, Minkoff H, Levine AM, Henry D, et al. Long-term incidence of cervical cancer in women with HIV
13. Clifford GM, Goncalves MA, Franceschi S. for the HPV and HIV Study Group. Human papillomavirus types among women infected with HIV: a meta-analysis
14. Castle PE, Glass AG, Rush BB, Scott DR, Wentzensen N, Gage JC, et al. Clinical human papillomavirus detection forecasts cervical cancer risk in women over 18 years of follow-up
. J Clin Oncol
15. Munos N, Bosch XF, de Sanjose S, Herrero R, Castellsague X, Shah KV, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer
. N Engl J Med
16. Xue X, Gange SJ, Zhong Y, Burk RD, Minkoff H, Massad LS, et al. Marginal and mixed-effects models in the analysis of human papillomavirus natural history data
. Cancer Epidemiol Biomarkers Prev
17. Wilson TE, Massad LS, Riester KA, Barkan S, Richardson J, Young M, et al. Sexual, contraceptive, and drug use behaviors of women with HIV and those at high risk for infection: results from the Women's Interagency HIV Study
18. Linas BS, Minkoff HM, Cohen MH, Karim R, Cohan D, Wright RL, et al. Relative time to pregnancy among HIV-infected and uninfected women in the Women's Interagency HIV Study, 2002–2009
19. Tornesello ML, Duraturo ML, Giorgi-Rossi P, Sansone M, Piccoli R, Buonaguro L, et al. Human papillomavirus (HPV) genotypes and HPV 16 variants in human immunodeficiency virus-positive Italian women
. J Gen Virol
20. Schneider A, Kirchhoff T, Meinhardt G, Gissmann L. Repeated evaluation of human papillomavirus 16 status in cervical swabs of young women with a history of normal Papanicolaou smears
. Obstet Gynecol
21. Winer RL, Lee SK, Hughes JP, Adam DE, Kiviat NB, Koutsky LA. Genital human papillomavirus infection: incidence and risks factors in a cohort of female university students
. Am J Epidemiol
22. Einstein MH, Martens MG, Garcia FA, Ferris DG, Mitchell AL, Day SP, et al. Clinical validation of the Cervista HPV HR and 16/18 genotyping tests for use in women with ASC-US cytology
. Gynecol Oncol
23. Stoler MH, Wright TC, Sharma A, Zhang G, Apple R, Wright TL, et al. The interplay of age stratification and HPV testing on the predictive value of ASC-US cytology: results from the ATHENA HPV Study
. Am J Clin Pathol
24. Minkoff H, Zhong Y, Burk RD, Palefsky JM, Xue X, Watts DH, et al. Influence of adherent and effective antiretroviral therapy use on human papillomavirus infection and squamous intraepithelial lesions in human immunodeficiency virus-positive women
. J Infect Dis
25. D'Souza G, Burk RD, Zhong Y, Minkoff H, Massad LS, Xue X, et al. Cervicovaginal human papillomavirus (HPV)-infection before and after hysterectomy: evidence of different tissue tropism for carcinogenic and noncarcinogenic HPV types in a cohort of HIV-positive and HIV-negative women
. Int J Cancer
26. Winer RL, Hughes JP, Feng Q, O’Reilly S, Kiviat NB, Holmes KK, et al. Condom use and the risk of genital human papillomavirus infection in young women
. N Engl J Med
27. Shew ML, Fortenberry JD, Tu W, Juliar BE, Batteiger BE, Qadadri B, et al. Association of condom use, sexual behaviors, and sexually transmitted infections with the duration of genital human papillomavirus infection among adolescent women
. Arch Pediatr Adolesc med