|
Abstract:
|
There is an increased interest in finding predictive biomarkers that can guide treatment options for both mutation carriers and non-carriers. The statistical assessment of variation in treatment benefit (TB) according to the biomarker carrier status (i.e., differential TB) plays an important role in evaluating predictive biomarkers. For time to event endpoints, the hazard ratio (HR) for interaction between treatment and a biomarker from a Proportional Hazards regression model is commonly used as a measure of differential TB. While this can be easily obtained using available statistical software packages, the interpretation of HR is not straightforward. Therefore, we propose different summary measures of differential TB for evaluating a predictive biomarker based on clinically useful functions of survival probabilities. We examine the operating characteristics of our proposed measures, illustrate their use and interpretation using data from completed clinical trials, and provide suggestions for a practically useful measure.
|
