Two-Stage Winner Design with Survival Outcomes: Methods and Features

• Activity Number: 515
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Fang Fang*+ and Yong Lin and Yunlin Li and Jay Yang and Xiaosha Sherman Zhang
• Companies: Bayer Healthcare Pharmaceuticals and University of Medicine and Dentistry of New Jersey and K&L Consulting Services, Inc. and K&L Consulting Services, Inc. and Astellas Pharma Global Development, Inc.
• Keywords: oncology ; two-stage winner design ; survival endpoints ; correlation

Abstract

In oncology clinical studies, a phase 2 trial alone may not be ideal due to long follow-up or high consumption of resources. It may be necessary to evaluate more than one promising regimens at the same time in the phase 3 trial. In this case, an interim analysis is often used to drop the inferior arm(s) and to avoid the high cost, long term trial conduction, and unnecessary exposure to ineffective treatments. This approach is considered as combining the two phases into one study: phase 2 portion will be carried out by the interim analysis of phase 3 trial. When appropriate surrogate endpoints exist, such as progression free survival in oncology trials, they can be used at the interim analysis to accelerate the drug development process. The statistical frameworks of the two-stage winner design are available in the literature for continuous endpoints and will be generalized to survival endpoints here. The asymptotic correlation of log-rank statistics for the design is derived and used to calculate the operational characteristics. Some features of the two-stage winner design survival trial using the same endpoint or different endpoints at interim analysis are explored.