|
Activity Number:
|
27
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303649 |
|
Title:
|
A Group Sequential Design with Potential Futility Stop and Sample Size Adjustment
|
|
Author(s):
|
Xiao J. Sun*+ and Joshua Chen and Kuang-Kuo (Gordon) Lan
|
|
Companies:
|
Merck Research Laboratories and Merck Research Laboratories and Johnson & Johnson PRD
|
|
Address:
|
UG1CD-44, North Wales, PA, 19454,
|
|
Keywords:
|
group sequential ; futility stop ; sample size adjustment ; type I error rate
|
|
Abstract:
|
In a group sequential trial with potential futility stop and sample size adjustment, the interim analysis may result in four potential decisions: 1) stop early due to a convincing benefit; 2) stop early due to futility; 3)continue the trial without sample size re-adjustment; and 4)adjust the sample size and continue the trial. It is well known that increasing the sample size based upon unblinded interim data may inflate the false positive rate. However, if we set up a futility stopping rule, by accepting the null hypothesis early, futility stopping will decrease the chance of rejecting the null hypothesis and will therefore reduce the type I error rate. In this paper, we extend the Lan-Trost approach to group sequential studies. Our simulation studies show that the type I error rate can be well controlled in a group sequential study with potential futility stop and sample size adjustment.
|
