|
Activity Number:
|
154
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #305481 |
|
Title:
|
A Two-Stage Adaptive Design for Phase III Trials To Establish Noninferiority and Superiority
|
|
Author(s):
|
Yulan Li*+ and Qing Liu and Jeffrey Maca
|
|
Companies:
|
Novartis Pharmaceuticals Corporation and Johnson & Johnson and Novartis Pharmaceuticals Corporation
|
|
Address:
|
One Health Plaza, East Hanover, NJ, 07936,
|
|
Keywords:
|
two-stage adaptive design ; adaptive closed-testing procedure ; adaptation guidelines ; interim analysis ; operational bias ; conditional power
|
|
Abstract:
|
We consider a phase III trial in which two doses of an experimental drug are compared to an active control to establish a noninferiority claim in efficacy. Based on phase II results, the high dose also is expected to be superior to the active control in efficacy. But its safety profiles are uncertain over the course of treatment in the phase III trial. Due to practical constraints, a sponsor for drug development may not be willing to commit up front to power the trial for demonstrating superiority of the high dose. We propose an exact two-stage adaptive design, by which the sponsor can decide to pursue the superiority objective after an interim analysis only if the high dose still holds up the efficacy expectation and its safety data do not raise major concerns. The procedure strongly controls type I error rates. The issue of operational bias will be discussed for real-life applications.
|
