|
Activity Number:
|
334
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #308832 |
|
Title:
|
Adaptive Treatment Selection in Seamless Phase II/III Designs
|
|
Author(s):
|
Xiaoling Wu*+
|
|
Companies:
|
Bristol-Myers Squibb Company
|
|
Address:
|
5 Research Parkway, Wallingford, CT, 06492,
|
|
Keywords:
|
seamless adaptive design ; dose selection ; multiple testing
|
|
Abstract:
|
An adaptive seamless design combines a learn phase and a confirm phase together to reject or accept the global hypothesis. Adaptive treatment/dose selection is an important applications of seamless designs as poor dose selection and failure to collect information from early phase studies can cause negative result of a phase III clinical trial. Our work applies several closed multiple testing procedures to control the type I error rate in the setting of an adaptive two-stage design in which our primary interest is to compare multiple treatments with a control. We first review the general procedure to integrate Bonferroni-Holm type and Sims's test type multiple testing into a two stage adaptive dose selection design. Secondly, we use simulated examples to demonstrate advantages and disadvantages of such flexible design over traditional design (splitting learn phase and confirm phase).
|
