Impact of bivalent human papillomavirus (HPV) vaccination upon the risk of acquisition of other HPV types

Abstract

The first human papillomavirus (HPV) vaccine was approved by the FDA in 2006, protecting against 2 carcinogenic types, 16 and 18. There is limited information about whether that vaccination may affect the prevalence of other HPV types in a community over time. Our study aimed to examine this issue. We used published data from The Costa Rica Vaccine Trial. Women were vaccinated with either the Cervarix® HPV vaccine or the hepatitis A vaccine (the control group). Based upon laboratory data, each of these arms can be divided into a naïve cohort with no evidence of HPV infection on trial entry and the rest considered a non-naïve cohort with baseline infection with HPV.

At the end of a 4-year follow-up, women in the combined non-naïve cohort had higher rates of HPV (mean number of HPV types detected per woman = 1.22) compared to the combined naïve cohort (mean number of HPV types detected per woman = 0.917), consistent with their presumptive higher risk profile upon study entry.  However, there was considerable variation by HPV type. Comparing those vaccinated against HPV to the control group, those in the non-naïve cohort had an increased prevalence of HPV type 35 (RR=1.37, 95%CI [0.88, 2.11]), and significantly increased prevalence of HPV types 51 and 59 (RR=1.34 [1.03, 1.74] and 1.86 [1.15, 3.00]) respectively). This could be the result of increased HPV infection with these types after HPV 16/18 vaccination or HPV infection before trial entry.  Among the naïve cohort, there was decreased prevalence in the HPV vaccine arm of HPV type 31 (RR=0.30, 95% CI [.12, .79]) and type 45 (RR=0.18, 95% CI [0.02, 1.46]), suggesting possible cross protection to these HPV types.

The protective effects from the bivalent HPV vaccine in the non-naïve cohort further support the current public health practice of administering the HPV vaccine to young women, without regard to baseline HPV status.

On December 10, 2014, the FDA approved Gardasil 9®, which vaccinates for an additional 5 carcinogenic types of HPV (31, 33, 45, 52 and 58). The addition of these types to the HPV vaccine will be beneficial in decreasing infection with these HPV types, but our analysis looks at the possible implications of how this can affect acquisition types not featured in the vaccine. There was a high prevalence of HPV types 16, 51 and 52 among the total cohort. Types 51 and 52 are not in the prior HPV vaccines and type 51 is not in the new Gardasil 9® vaccine, so type 51 may continue at a high prevalence in some geographic areas and may pose especial risk in the coming years. The prevalences of the various HPV types differ, sometimes substantially, by region. Therefore, further studies examining the prevalence of HPV types over time by geography will be critical for providing guidance to vaccine manufacturers, as well as screening guidelines for clinicians.