Statistical Methods for Assessing Immune Correlates of Risk and Protection Using Flexible Two-Phase Sampling Designs that Enrich Longitudinal Samples

About This Grant

Project Summary/Abstract Immune correlates of protection (CoP) are biomarkers that predict vaccine-induced protection against dis- eases and play a crucial role in the design and development of effective vaccines. The U.S. government (USG)-led initiative to identify CoPs for COVID vaccines highlighted the importance of neutralizing antibody titers as surrogate endpoints, significantly impacting vaccine recommendations and approvals. To effec- tively measure these immune biomarkers, researchers utilize two-phase designs, such as case-cohort or case-control studies, to boost statistical power and enhance representation. This proposal aims to develop novel two-phase sampling designs that allow enrichment of longitudinal immune response marker mea- surements in immune correlates studies. The proposal also aims to develop advanced statistical methods for datasets collected under sampling designs that would introduce bias if analyzed using conventional in- verse probability-weighted methods. Aim 1 focuses on the analysis of the immune response biomarkers measured at the peak immunogenicity time point, while Aim 2 delves into the study of the decaying im- mune response biomarkers. The final product will feature a user-friendly software implementation of the proposed methods, along with its application to analyze COVID correlates datasets from past and ongoing USG-sponsored vaccine efficacy trials.