|
Activity Number:
|
233
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #304446 |
|
Title:
|
On the Three-Arm Noninferiority Trials with Binary Endpoints
|
|
Author(s):
|
Gang Li*+ and Shan Gao
|
|
Companies:
|
GlaxoSmithKline and Techdata LLC
|
|
Address:
|
1250 S. Collegeville Rd, Collegeville , PA, 19426,
|
|
Keywords:
|
Effect preservation test ; Group sequential design ; Noninferiority trial ; Three-arm trial ; Type-I error ; Power
|
|
Abstract:
|
Noninferiority trials are usually conducted for two objectives. First of all the trial should establish the effectiveness of the test treatment. Secondly the trial seeks to determine whether the test treatment is no worse than an active control. Whenever it is ethically justifiable to expose patients to placebo, the three-arm design with a test treatment, an active control and a placebo group is the preferred design for these two objective. In this paper, we first consider the sample size estimation problem for these two objectives under a fixed sample design. Next we turn to a modified group sequential design for the three-arm noninferiority trial. We compare the performance of two statistical testing methods and three sample size determinations methods for the effect preservation test via Monte Carlo simulations.
|
