Deconstructing the measure of vaccine efficacy against disease irrespective of HPV in HPV vaccine clinical trials

Abstract

Background: Human papillomavirus (HPV) vaccines were licensed by demonstrating prevention of anogenital disease caused by specific HPV types in clinical studies. Measuring the impact of HPV vaccination on the overall burden of anogenital disease (irrespective of HPV) is an important public health question which is ideally addressed in post-licensure epidemiological studies. Attempts were made to use clinical trial data for that purpose. However, the interpretation of vaccine efficacy on the endpoint of disease irrespective of HPV is not widely understood.

Methods: We used the 9-valent HPV vaccine clinical program as a case study to determine the value of measuring vaccine efficacy in such endpoint. This assessment was rigorously performed by heuristic reasoning and through probability calculations.

Results: The measure of vaccine efficacy in the irrespective of HPV endpoint is driven simultaneously in opposite directions by the high estimate of prophylactic efficacy and a numerically negative estimate of risk reduction that is also a reflection of high prophylactic efficacy and no cross-protection.

Conclusions: The vaccine efficacy estimate in the irrespective of HPV endpoint is ambiguous and difficult to interpret. Comparing this estimate across different HPV vaccine studies requires an understanding of the contributions of vaccine HPV type efficacy and the incidence of disease not related to vaccine HPV types for each study. Without such understanding, comparing studies and drawing conclusions from such comparison are highly misleading. Approaches are proposed to divide this endpoint in components that are easier to interpret.

Keywords: HPV; vaccine.