Evaluating Marker-Guided Treatment Selection Strategies

• Activity Number: 130
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Junlong Li*+ and Roland Matsouaka and Tianxi Cai
• Companies: Harvard University and Harvard University and Harvard University
• Keywords: Personalized Medicine ; Subgroup Analysis ; Biomarker-analysis Design ; Non-parametric estimation ; Perturbation-resampling ; Cross-validation

Abstract

A potential strategy to improve healthcare efficiency is to effectively tailor strategies for disease prevention, detection, treatment or management by incorporating patient level predictor information such as environmental exposure, biological and genetic marker measurements. With more potential biomarker guided treatment strategies, it is important to quantify the value of such marker guided selection strategies. In this paper, we consider a general setting with multiple markers and propose non- parametric methods to make inference about the value of a class of marker guided treatment selection strategies. We also propose procedures for comparing different treatment selection strategies, which can be used to quantify the incremental value of new markers in improving treatment selection. While working models may be used to derive treatment selection rules, the evaluation procedure is fully non-parametric which ensures the validity of the inference. Our proposals are examined through a simulation study and illustrated with the data from a clinical trial for evaluating the efficacy of a 3-drug combination against a standard 2-drug combination for treating HIV-1 infected patients.