Testing Strategies for Trial Designs with Prognostic Biomarkers

• Activity Number: 606
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Chris Holland*+ and Catherine Jia and Alicia Zhang
• Companies: Amgen and Amgen and Amgen
• Keywords: biomarker ; trial design ; power ; sample size

Abstract

Biomarkers are typically classified into two groups, predictive and prognostic. With predictive biomarkers, experimental treatments are thought to work differently between patients with and without that biomarker whereas with prognostic biomarkers patients are expected to have differing prognoses regardless of the treatment being given. However, in reality, the interplay between study subjects, their biomarkers, and the experimental treatments they receive in clinical trials can be much more complicated. Particularly in trials with time-to-event endpoints, a number of circumstances must be considered in order to design a clinical trial that allows realistically-powered tests of appropriate patient populations. In this paper, we will present formulas and methods that allow clinical trial designers to make sample size and power calculations under various assumptions regarding the biomarker-positive and -negative patients intended to be evaluated. With such calculations readily available, the most efficient and appropriate study design and testing strategy can be implemented.