A Multiple Imputation Approach for Surrogate Markers Evaluation in the Causal Inference Framework

• Activity Number: 132
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Xiaopeng Miao*+ and Xiaoming Li and Ivan Chan
• Companies: Boston University School of Public Health and Gilead Sciences and Merck
• Keywords: Casual treatment effect ; Multiple imputations ; Surrogate endpoints ; Principal stratification ; Biomarkers

Abstract

The concept of principal surrogate developed in the causal inference framework (Frangakis and Rubin, 2002; Gilbert and Hudgens, 2008) has drawn much attention in the field of biomarker research. Principal surrogates are defined based on the causal treatment effects in principal strata, which are constructed based on the joint distribution of the potential surrogate outcomes. The challenge of evaluating principal surrogates lies in the fact that half of these potential surrogate outcomes cannot be observed in most clinical trials. Therefore assessing the principal surrogacy of biomarkers is essentially a missing data problem. We propose a multiple imputation approach to evaluate candidate principal surrogate markers. The proposed method employs baseline variables to impute the missing potential surrogate outcomes. The stratum-specific causal treatment effects on the clinical endpoint are then estimated for each imputed dataset and the inference for surrogacy of a biomarker is based on the combined results over multiple imputations. Simulation studies are performed to evaluate the performance of the proposed and the implementation of the method is illustrated using a vaccine study.