Design and Estimation for Evaluating Principal Surrogate Markers in Vaccine Trials

• Activity Number: 491
• Type: Topic Contributed
• Date/Time: Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Ying Huang*+ and Peter B Gilbert
• Companies: Fred Hutchinson Cancer Research Center and University of Washington/Fred Hutchinson Cancer Research Center
• Keywords: close-out placebo vaccination ; estimated likelihood ; immune correlates ; principal surrogate ; pseudo-score ; study design

Abstract

Identification of immune biomarkers predictive of vaccine efficacy is useful for guiding vaccine development in our battle against infectious diseases. To alleviate the non-identifiability problems associated with using a potential-outcomes framework, novel trial designs have been proposed for evaluating principal surrogates, which utilize baseline covariates correlated with the immune biomarker(s) and/or augment the trial by vaccinating uninfected placebo recipients at the end of the trial. If not used efficiently, however, the additional information generated by the augmented component can lead to counter-intuitive results regarding the precision of estimation. In this paper, we propose using a pseudo-score type estimator for evaluating principal surrogacy in the augmented vaccine trial design, and compare its performance with existing approaches. We study asymptotic properties of the new estimator in a two-phase sampling setting, based on which we discuss considerations for optimal sampling of immune biomarkers in vaccine efficacy trials with respect to estimation of the vaccine efficacy curve and the predicted vaccine effect.