Human papillomavirus (HPV) disproportionately affects women living with HIV (WLWH) resulting in a much larger burden of HPV-associated disease, such as cervical cancer, than that seen in the general population. The prevalence of HPV infection among WLWH is approximately 50%, which is twice the prevalence in women without HIV.1 The rate of persistent HPV infection among WLWH is approximately 20%–24%, making WLWH 3–6 fold more likely to have a persistent HPV infection than women without HIV.2,3 The disparity between rates of cervical cancer is equally wide; within a North American population of WLWH, the incidence rate (IR) of invasive cervical cancer (ie, cervical cancer that has invaded into deeper layers of the cervix beyond the surface) was 16 per 100,000 person-years (PY), compared with only 5 per 100,000 PY in women without HIV.4
In addition to higher rates of HPV-related infection and disease, WLWH also experience infection with a wider range of HPV types,5 which has important implications for vaccine and cervical screening programming. HPV16 is well known to be the most carcinogenic of HPV types. However, it is less affected by increased immunodeficiency than other oncogenic HPVs and is also seen in a reduced proportion among WLWH.6
Although HPV vaccines are now available and have promising safety and immunogenicity findings in WLWH to date,7–10 it is critical to identify the residual burden of oncogenic HPV within WLWH to inform postvaccination cervical screening needs for this population. In this study, we assessed rates of new persistent infection with oncogenic HPV types not contained in the quadrivalent HPV (qHPV) vaccine in our cohort of qHPV-vaccinated WLWH.
As part of a longitudinal study of HPV vaccine immunogenicity and efficacy, girls and WLWH aged 9 and greater were recruited from 14 clinics serving WLWH across Canada between 2008 and 2012, with long-term follow-up to 8 years. All participants or guardians, as appropriate, provided informed consent to enroll in the study. The study population and methods of enrollment have previously been described.7 Participants were scheduled to receive 3 doses of qHPV vaccine intramuscularly at month 0/2/6. Serology for anti-qHPV antibodies was performed by competitive Luminex immunoassay at Merck Research Laboratories. Pelvic examination was performed on participants who were postmenarchal and sexually active at the discretion of the care provider and in accordance with time- and geographic-specific clinical recommendations. For participants undergoing pelvic examination, cervical cytology and cervicovaginal HPV DNA samples were collected by a health care provider at the screening visit and at months 0/6/12/18/24/36/48/60/72/84/96. The ThinPrep Pap test was used for collection of cervical cytology samples using a cytobrush, and results were classified by Bethesda Criteria centrally at the British Columbia Cancer Agency Cervical Cancer Screening Laboratory. Aliquots of the PreservCyt from Pap tests were processed and typed for 36 HPV genotypes by Linear array assay (Roche Molecular Systems, Pleasanton, CA).11
For this analysis, the primary outcome was rate of persistent HPV infection with oncogenic, nonquadrivalent vaccine HPV types (ie, oncogenic types not including HPV16/18) within our cohort of qHPV-vaccinated WLWH. Persistent HPV infection was defined as the detection of the same incident HPV type in samples collected at 2 or more consecutive study visits (>6 months apart) or detection of an HPV type at the last available visit. Although this definition of persistent infection is an accepted definition used within the HPV vaccine literature,12,13 it is known to overestimate the true number of persistent infections by including cases where HPV is present only in the last sample; however, it is accepted as it errs on the side of caution since some of these infections will persist. Because of this, a sensitivity analysis was also conducted where only the confirmed persistent cases (ie, detection of the same HPV type at 2 or more consecutive study visits) were considered. The final subanalysis presented herein determines the incidence of persistent infection with HPV types contained only in the nonavalent vaccine (HPV31/33/45/52/58) as compared to the incidence of persistent infection with oncogenic HPV types not contained within available vaccines (HPV35/39/51/56/59/68). To be eligible for this analysis, participants had to have received at least one dose of vaccine and had to have at least one HPV DNA result after vaccination. For ascertainment of HPV cases, participants were required to be DNA negative to the relevant HPV type at the screening and baseline visits. The HPV types considered in this analysis were HPV31/33/35/39/45/51/52/56/58/59/68; these HPV types were selected for consideration due to their oncogenic potential.14
A total of 284 participants were eligible for interim analysis with 1205 PY of follow-up and a median follow-up time of 4 years per person. Eligible population characteristics at baseline are described in Table 1. The median age was 38 years (interquartile range: 32–44). Participants were predominantly of black (41%) and white (36%) ethnicity. The median CD4 count at first vaccination was 499 cells/mm3 (interquartile range: 375–680), and 71% of participants had HIV plasma viral loads <50 copies/mL. A total of 267 participants (94%) received all 3 doses of vaccine. The vaccine was safe and highly immunogenic within this population, as previously described.7
The IRs of persistent HPV types are shown in Figure 1. The most frequently documented persistent infections were infections with HPV51 [IR: 1.4 per 100 person-years (/100 PY), 95% confidence interval (CI): 0.8 to 2.3]. The second and third most common types contributing to persistent infection were HPV52 (IR: 1.2/100 PY, 95% CI: 0.6 to 2.1) and HPV39 (IR: 1.1/100 PY, 95% CI: 0.6 to 1.9), respectively. These types were followed by HPV45 (IR: 0.9/100 PY, 95% CI: 0.4 to 1.7) and HPV35 (IR: 0.7/100 PY, 95% CI: 0.3 to 1.4) being fourth and fifth most common, respectively. Overall, 40% of persistent infections were cases in which the HPV type was detected in at least 2 consecutive samples while HPV was detected in the last sample in 60% of cases. This 40%/60% split between confirmed persistent and last sample cases was also consistent within HPV types.
In a sensitivity analysis that limited to only the confirmed persistent cases (not including cases of HPV detection in the last sample), the most frequently documented HPV type remained as HPV51 (IR: 0.6/100 PY, 95% CI: 0.2 to 1.2), followed by HPV52 (IR: 0.5/100 PY, 95% CI: 0.2 to 1.1) and HPV39 (IR: 0.4/100 PY, 95% CI: 0.1 to 1.0), respectively.
In a subanalysis pooling HPV types into categories of nonavalent (HPV31/33/45/52/58) or oncogenic HPV types not contained within available vaccines (HPV35/39/51/56/59/68), the composite endpoints yielded an IR of 2.4/100 PY (95% CI: 1.6 to 3.5) for persistent infection with nonavalent HPV types and an IR of 3.6/100 PY (95% CI: 2.6 to 4.9) for persistent infection with HPV types not contained within vaccines.
Of the top 5 persistent HPV types observed in this cohort, only HPV52 and 45 are contained within the nonavalent vaccine. In addition, the persistent infection with the HPV types added in the nonavalent vaccine that are not present in the quadrivalent vaccine had an IR of 2.4/100 PY while persistent infection with HPV types not contained within any available vaccine resulted in a higher IR of 3.6/100 PY. This implies that the nonavalent vaccine could further assist in the protection of WLWH, but gaps in protection for this population would remain. Although the HPV types that are not contained within any currently available vaccine contribute less to disease in the general population, they are carcinogenic, and the effect of HIV infection on the pathogenicity of these specific HPV types has not been completely elucidated. Description of HPV types associated with CIN3+ in women without HIV and WLWH has shown that the contribution of HPV51 and 39 toward dysplasia in WLWH is greater than in women without HIV.15 Meta-analysis has also shown that WLWH who have high-grade squamous intraepithelial lesions are less likely to be infected with HPV16 than the general population and more likely to be infected with HPV51, among other types, or to have multiple HPV-type infection.6 HIV is known to disrupt epithelial tight junctions, which may facilitate HPV entry to the basal epithelial layer.16 It is also known that the HIV tat protein enhances HPV transcription.17 This could be a mechanism explaining the potential oncogenic effects of HPV serotypes that could differentially affect WLWH. As the HPV types not contained within available vaccines may cause disease in this way, it is important to note that the infectivity and carcinogenic potential of these HPV types are enhanced in WLWH. Further study is needed to more clearly describe the contribution of these HPV types to cervical dysplasia among WLWH.
The high rate of persistent infection with HPV51 validates previous data indicating that there is a high burden of HPV51 in WLWH, and that this type would be very important in WLWH after vaccination.18 We observed less persistent HPV31 and HPV33 than reported in some previous studies of North American WLWH.6,19 However, we did see relatively high rates of persistent HPV52 and HPV58, which is consistent with previous literature in WLWH.6 We might hypothesize the differences could be a result of some cross-protection against HPV31, which is closely related to HPV16 within the alpha-9 phylogenetic group, and HPV33, which is also an alpha-9 HPV type. Evidence of cross-protection against HPV31 and HPV33 by the qHPV vaccine has previously been documented.20
The main analysis was conservative in nature and provides an overestimate of the incidence of persistent infection as not all cases of HPV detected at the last visit will go on to truly persist. The sensitivity analysis provided the opposite scenario of an underestimate of incidence of persistent infection because it only included cases where the HPV type was documented at 2 consecutive visits. Taken together, these analyses were consistent in demonstrating that HPV51, 52, and 39 contribute the largest burden of persistent infection among this vaccinated population, and they demonstrate the upper and lower limits within which the true value of incident persistent infections lies.
Limitations to this analysis include the fact that our findings may not be generalizable to other global settings due to multiple factors including availability and engagement in HPV vaccination programs, cervical screening programs, and HIV care including antiretroviral use, as well as the geographic distribution of different HPV types. In addition, these findings pertain to women vaccinated with the qHPV vaccine, which is now largely replaced by the more recently available nonavalent vaccine. However, our findings do break down the persistent HPV types based on availability in current vaccines, including the nonavalent vaccine, which addresses this shift in vaccine valency. Nonetheless, findings of the residual burden of oncogenic HPV types 35/39/51/56/59/68 in WLWH vaccinated with the nonavalent vaccine may differ.
To the best of our knowledge, only one other article has described HPV infection with non-vaccine HPV types after vaccination within a population of WLWH, but women were only followed for 1 year after vaccination.21 Similar to our findings, they reported a higher frequency of the non-vaccine HPV types 51 and 52 detected at 28 and 52 weeks. In contrast to our findings, they detected a relatively high frequency of HPV31 at the 52-week time point and HPV68 at both time points, but not a higher frequency of HPV39.21 Given the broad range of oncogenic HPV types seen in vaccinated WLWH, cervical screening will remain important in this population. Additional data are needed to inform programs using HPV DNA testing as a screening modality for WLWH due to the likely altered carcinogenicity of certain HPV types in WLWH.
Our findings add critical data to the literature regarding persistent HPV infection with extended follow-up after vaccination in WLWH. WLWH who have been vaccinated with the qHPV vaccine remain vulnerable to a clinically significant burden of persistent HPV infections. The frequency with which these strains lead to cervical dysplasia and cancer requires ongoing study. Although the nonavalent vaccine has the potential to eliminate a portion of that burden, many of the persistent HPV infections that WLWH face are due to HPV types not contained within any currently available vaccine. Our findings support the continued regular cervical screening of WLWH regardless of their HPV vaccine history and validate the need for a multipronged approach to elimination of cervical cancer.
The authors acknowledge the CTN 236 HPV in HIV Study Team, in alphabetical order: Ariane Alimenti, MD (University of British Columbia); Arezou Azampanah, MSc (Women's Health Research Institute); Ari Bitnun, MD (University of Toronto); Sandra Blitz, MSc (University Health Network); Jason Brophy, MD (University of Ottawa); Jan Christilaw, MD (University of British Columbia); Jeffrey Cohen, MD (Windsor Regional Hospital HIV Care Program); Andrew Coldman, PhD (British Columbia Cancer Agency); Simon Dobson, MD (Vaccine Evaluation Centre); Laurie Edmiston (Canadian AIDS Treatment Information Exchange); Catherine Hankins, MD, PhD (Amsterdam Institute for Global Health and Development); Christos Karatzios, MD (McGill University Health Centre); Mel Krajden, MD (British Columbia Centre for Disease Control); Normand Lapointe, MD (CHU Sainte Justine); Jessica McAlpine, MD (University of British Columbia); Dianne Miller, MD (University of British Columbia); Dirk van Niekerk, MD (British Columbia Cancer Agency); Gina Ogilvie, MD, DrPH (University of British Columbia); Neora Pick, MD (University of British Columbia); Lindy Samson, MD (University of Ottawa); Julie van Schalkwyk, MD (University of British Columbia); David Scheifele, MD (Vaccine Evaluation Centre); Joel Singer, PhD (CIHR Clinical Trials Network); Sarah Stone, MD (British Columbia Centre for Excellence in HIV/AIDS); Gavin Stuart, MD (University of British Columbia); Marcie Summers (Positive Women's Network); Laura Vicol, MN, NP (University of British Columbia); and Melissa Watt (Women's Health Research Institute). The authors thank all the additional clinicians and research staff for their important contributions to participant enrollment and study visits. The authors also thank the participants without whom this research would not be possible.
1. Blitz S, Baxter J, Raboud J, et al. Evaluation of HIV and highly active antiretroviral therapy on the natural history of human papillomavirus infection and cervical cytopathologic findings in HIV-positive and high-risk HIV-negative women
. J Infect Dis. 2013;208:454.
2. Sun X-W, Kuhn L, Ellerbrock TV, et al. Human papillomavirus infection in women
infected with the human immunodeficiency virus. N Engl J Med. 1997;337:1343–1349.
3. Kriek J-M, Jaumdally SZ, Masson L, et al. Female genital tract inflammation, HIV co-infection and persistent mucosal human papillomavirus (HPV) infections. Virology. 2016;493:247–254.
4. Abraham AG, D'Souza G, Jing YZ, et al. Invasive cervical cancer risk among HIV-infected women
: a North American multicohort collaboration prospective study. J Acquir Immune Defic Syndr. 2013;62:405–413.
5. Strickler HD, Burk RD, Fazzari M, et al. Natural history and possible reactivation of human papillomavirus in human immunodeficiency virus-positive women
. J Natl Cancer Inst. 2005;97:577–586.
6. Clifford GM, Goncalves MAG, Franceschi S; HPV and HIV Study Group. Human papillomavirus types among women
infected with HIV: a meta-analysis. AIDS. 2006;20:2337–2344.
7. Money DM, Moses E, Blitz S, et al. HIV viral suppression results in higher antibody responses in HIV-positive women
vaccinated with the quadrivalent human papillomavirus vaccine. Vaccine. 2016;34:4799–4806.
8. Kojic EM, Kang M, Cespedes MS, et al. Immunogenicity and safety of the quadrivalent human papillomavirus vaccine in HIV-1-infected women
. Clin Infect Dis. 2014;59:127–135.
9. Giacomet V, Penagini F, Trabattoni D, et al. Safety and immunogenicity of a quadrivalent human papillomavirus vaccine in HIV-infected and HIV-negative adolescents and young adults. Vaccine. 2014;32:5657.
10. Kahn JA, Xu J, Kapogiannis BG, et al. Immunogenicity and safety of the human papillomavirus 6, 11, 16, 18 vaccine in HIV-infected young women
. Clin Infect Dis. 2013;57:735–744.
11. Coutlée F, Rouleau D, Petignat P, et al. Enhanced detection and typing of human papillomavirus (HPV) DNA in anogenital samples with PGMY primers and the linear array HPV genotyping test. J Clin Microbiol. 2006;44:1998–2006.
12. Villa LL, Costa RLR, Petta CA, et al. High sustained efficacy of a prophylactic quadrivalent human papillomavirus types 6 11 16 18 L1 virus-like particle vaccine through 5 years of follow-up. Br J Cancer. 2006;95:1459–1466.
13. Villa LL, Costa RLR, Petta CA, et al. Prophylactic quadrivalent human papillomavirus (types 6, 11, 16, and 18) L1 virus-like particle vaccine in young women
: a randomised double-blind placebo-controlled multicentre phase II efficacy trial. Lancet Oncol. 2005;6:271–278.
14. de Martel C, Plummer M, Vignat J, et al. Worldwide burden of cancer attributable to HPV by site, country and HPV type. Int J Cancer. 2017;141:664–670.
15. Massad LS, Xie X, Burk RD, et al. Association of cervical precancer with human papillomavirus types other than 16 among HIV co-infected women
. Am J Obstetrics Gynecol. 2016;214:354.e351–354.e356.
16. Tugizov SM, Herrera R, Chin-Hong P, et al. HIV-associated disruption of mucosal epithelium facilitates paracellular penetration by human papillomavirus. Virology. 2013;446:378–388.
17. Tornesello ML, Buonaguro FM, Bethgiraldo E, et al. Human-immunodeficiency-virus type-1 tat gene enhances human papillomavirus early gene-expression. Intervirology. 1993;36:57–64.
18. Coutlée F, Ratnam S, Ramanakumar AV, et al. Distribution of human papillomavirus genotypes in cervical intraepithelial neoplasia and invasive cervical cancer in Canada. J Med Virol. 2011;83:1034–1041.
19. Massad LS, Xie X, Burk R, et al. Long-term cumulative detection of human papillomavirus among HIV seropositive women
. AIDS. 2014;28:2601–2608.
20. Garland SM, Cornall AM, Brotherton JML, et al. Final analysis of a study assessing genital human papillomavirus genoprevalence in young Australian women
, following eight years of a national vaccination program. Vaccine. 2018;36:3221.
21. Cespedes MS, Kang M, Kojic EM, et al. Anogenital human papillomavirus virus DNA and sustained response to the quadrivalent HPV vaccine in women
living with HIV-1. Papillomavirus Res. 2018;6:15–21.