Mixture Representation of Efficacy Measures in Biomarker Studies

• Activity Number: 130
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Jason Hsu*+ and Szu-Yu Tang
• Companies: Ohio State University and Ventana Medical Systems, Inc.
• Keywords: Subgroup identification ; Biomarker ; Relative risk ; Mixture representation ; Efficacy ; Odds ratio

Abstract

In personalized medicine, the patient population is thought of as a mixture of two or more subgroups that might derive differential efficacy from a drug. A decision to make is which subgroup or union of the subgroups should the drug be developed for. Interestingly, some common measures of efficacy are such that its value for a mixture population may not be representable as a function of efficacy for the subgroups and their characteristics, unless the study is properly designed. (In extreme cases, Simpson's Paradox can occur.) This presentation describes designs of study that would lead to probabilistic models so that efficacy measured as a difference of means, a relative risk, or an odds ratio for a mixture population can be represented as a function of efficacy for the subgroups and their characteristics.