|
Activity Number:
|
467
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #310100 |
|
Title:
|
Bayesian Optimal Design in Phase II Studies
|
|
Author(s):
|
Gary Rosner*+ and Peter Mueller and Meichun Ding
|
|
Companies:
|
The University of Texas M. D. Anderson Cancer Center and The University of Texas M.D. Anderson Cancer Center and Amgen Inc.
|
|
Address:
|
1515 Holcombe Boulevard, Houston, TX, 77030,
|
|
Keywords:
|
Optimal design ; Phase II clinical study ; Sequential design
|
|
Abstract:
|
In anticancer drug development, phase II studies screen out new therapies if they show activity. The treatments that pass the phase II test will undergo further evaluation, such as in a phase III clinical trial. In general, these phase II designs consider each treatment in isolation, with historical information entering the design through specification of null and alternative hypotheses. We propose a systematic decision-making approach to the phase II screening process. We discuss optimal and approximately optimal Bayesian sequential designs for phase II studies screening for active drugs. Computer simulations show that the methodology leads to high probability of discarding treatments with low success rates and moving treatments with high success rates to phase III trial.
|
