Stronger-Winner Design of Randomized Phase II Clinical Trials with Binary Outcome

• Activity Number: 657
• Type: Contributed
• Date/Time: Thursday, August 2, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Zhiying You*+
• Companies: The University of Alabama at Birmingham
• Keywords: Stronger-winner design ; phase II clinical trials ; binary outcome ; comparative study

Abstract

Randomized designs have been increasingly used in phase II trials of cancer studies and are becoming popular. The pick-the-winner selection designs formalized by Simon et al. directly compares treatments or regimens to prioritize and can be conducted with modest sample size with a disadvantage of large Type I error rate. In this paper, a flexible design with binary outcome is proposed for the comparison of two groups. Investigators can control the Type I and Type II error rate by selecting appropriate values of r and sample size for given values of the assumed response rate of the two treatments, where r is the minimum number of responses or successes by which the number of responses in the new treatment group is greater than the control. Because the larger the r, the stronger the winner or a treatment, we call it the stronger-winner design. Under the stronger-winner design, the pick-the-winner selection design is a special case when r = 1 in the comparison of two groups, except that Simon et al. considered to select one of the two at random if they are equal. The proposed design it most applicable to the comparison between new treatments with a standard control.