Detecting a Cutoff Point for Predictive Biomarkers in Clinical Trials Using Contrast Tests

• Activity Number: 238
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Jianliang Zhang*+
• Companies: MedImmune
• Keywords: predictive biomarker ; contrast test

Abstract

Predictive biomarkers are used to identify patients who are likely or unlikely to benefit from a specific therapy. The biomarker is usually based on a measurement involving a single gene or protein, or a combination of many variables. A cutoff point in the measurement to categorize patients according to their likelihood of benefiting from the therapy needs to be established in early phase drug development. Such predictive biomarker information will then be used and confirmed in late phase clinical development. A statistical approach using contrast tests for determining such a cutoff is developed in this presentation. The method is expanded to various types of endpoints with power and sample size calculations as well.