|
Activity Number:
|
429
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #305241 |
|
Title:
|
Multiple Testing Procedures for Multi-Dose Clinical Trials
|
|
Author(s):
|
Gang Li*+ and Peter Ouyang and Yining Wang
|
|
Companies:
|
Johnson & Johnson PRD and Johnson & Johnson and Johnson & Johnson
|
|
Address:
|
920 Route 202, P.O. Box 300 , Raritan, NJ, 08869,
|
|
Keywords:
|
likelihood ratio test ; family-wise error rate
|
|
Abstract:
|
Phase 3 clinical trials are likely to include two or more doses of study drug because the recommended doses of newly approved drugs often differ by geographic regions. In this article, two multiple testing procedures are proposed for multi-dose Phase 3 clinical trials. One is motivated from multivariate likelihood ratio test (LRT); another one starts with testing the doses combined. Both procedures strongly control family-wise error rate (FWER). The performances of the two procedures are evaluated against the commonly used multiple testing procedures such as Dunnett's step down and Hochberg by simulation.
|
