Estimation to Assess Surrogate Marker in Vaccine Trials

• Activity Number: 428
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): SoYoung Kim* and Ying Huang
• Companies: Fred Hutchinson Cancer Research Center and Fred Hutchinson Cancer Research Center
• Keywords: AIDS ; Causal inference ; Clinical trial ; Missing data ; Potential outcomes ; Surrogate marker

Abstract

In HIV vaccine/prevention clinical trials, evaluating vaccine efficacy is generally expensive due to the long trial period to observe clinical outcomes. Thus, it is of interest to find vaccine-induced immune responses as surrogate markers for the true study endpoint, which can be used to select vaccine regimens based on their immunogenicity in phase I/II studies for entering the future efficacy trials. We adopt a potential outcomes framework to evaluate vaccine-induced immune responses as principal surrogate endpoints for vaccine's effect in preventing HIV infection. However, addressing this problem is difficult because immune response to the vaccine is only observed in the vaccine group. Previously, Follmann (2006) proposed baseline irrelevant predictor (BIP) and closeout placebo vaccination for a binary outcome and Qin et al.(2008) extended this approach to discrete survival outcome. However, there have not been any studies on the asymptotic properties of the estimated likelihood estimator in this particular setting. In this paper, I developed asymptotic properties for surrogate endpoint assessment in vaccine trials with a binary outcome as well as time-to-event clinical endpoint