Novel Method Enhances Vaccine Efficacy Evaluation Using Immune Correlate Data
Key Insights
A new method leverages immune correlate data in time-to-event models to improve the precision of vaccine efficacy (VE) evaluation.
The approach identifies immune response biomarkers as correlates of risk (CoR) and protection (CoP) using data from vaccine and placebo recipients.
Application to zoster and dengue (search) vaccine trials demonstrates the method's ability to refine VE estimates by accounting for individual immunogenicity.
A novel statistical method has been developed to enhance the precision of vaccine efficacy (VE) evaluation by incorporating immune correlate data into time-to-event models. This approach, detailed in a recent study, leverages immunogenicity data from clinical trials to identify immune response biomarkers that serve as correlates of risk (CoR) and correlates of protection (CoP), ultimately refining VE estimates.
The method utilizes data from both vaccine and placebo recipients in time-to-event models to assess immunogenicity as a CoR, providing a more comprehensive evaluation of the association between immune responses and disease risk. By applying the Prentice criterion of conditional independence, the study determines whether immunogenicity fully mediates the vaccine effect.
Application to Zoster Vaccine
The study applied this method to data from a double-blind, placebo-controlled efficacy trial of a high-potency live-attenuated herpes zoster (search) (HZ) vaccine. The trial enrolled 1395 subjects, with 32 cases of HZ observed between vaccination and the end of the follow-up period. The analysis included 1326 subjects with available data on vaccination status, disease status, time to disease, age, and VZV antibody (search) titers.
Dengue Vaccine Analysis
Furthermore, the method was applied to anonymized data from the phase 3 CYD-TDV (search) dengue (search) vaccine trial in Asia. The cohort included 611 participants (417 CYD-TDV recipients and 194 placebo recipients) followed for over 6 years after their third dose. The analysis incorporated serotype-specific log2 PRNT50 values measured around day 28 post-vaccination.
Statistical Methods
The study employed Cox proportional hazards (PH) models and Fine-Gray subdistribution hazards models to evaluate the effect of immunogenicity on time-to-event outcomes, accounting for competing risks such as virologically confirmed dengue (search) disease caused by different serotypes (DENV1 (search)-4). The risk curve, defined as the hazard ratio as a function of immunogenicity, was used to represent the hazard relative to a reference datapoint.
Immunogenicity-Based Vaccine Efficacy Estimation
To estimate immunogenicity-based VE, the study utilized subgroup-specific risk curves obtained from covariate-adjusted models and subgroup-specific immunogenicity data. This approach allows for the estimation of VE in different covariate-defined subgroups, providing a more nuanced understanding of vaccine performance.
Implications for Vaccine Development
By enhancing the precision of VE evaluation, this method offers a valuable tool for vaccine developers and public health officials. The ability to identify CoRs and CoPs can inform vaccine design and development, while refined VE estimates can support more accurate assessments of vaccine performance and better-informed public health decisions.