Efficient Early Phase Designs in Oncology Based on Evidence Synthesis

• Activity Number: 454
• Type: Topic Contributed
• Date/Time: Wednesday, August 4, 2010 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Jyotirmoy Dey*+ and Wentao Feng
• Companies: Novartis Pharmaceuticals Corporation and Novartis Pharmaceuticals Corporation
• Keywords: Evidence synthesis ; Bayesian ; historical controls ; Early clinical trial ; Oncology

Abstract

In this talk we discuss examples of early phase Oncology designs that seek to gain efficiency through Bayesian evidence synthesis. Clinical study planning frequently rely, albeit informally, on relevant information from experiments in similar clinical context, e.g. same stage and severity of disease, drugs of the same class, etc. Through simple applications in early clinical trials using a Bayesian framework we allow formal incorporation of such information to provide useful benchmarking of efficacy or safety signals. Efficiency is gained mainly through improvement in decision operating characteristics or precision of estimates. Examples range from implementation in both phase I and II, and from utilizing information from historical controls to synthesis across multiple concurrent study arms.