|
Activity Number:
|
494
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 10, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #306254 |
|
Title:
|
Futility Analysis for a Randomized Trial with a Time-to-Event Endpoint
|
|
Author(s):
|
Corina M. Sirbu*+ and Kay M. Larholt
|
|
Companies:
|
Boston Scientific Corporation and Boston Scientific Corporation
|
|
Address:
|
100 Boston Scientific Way, Marlborough, MA, 01752-1242,
|
|
Keywords:
|
futility analysis ; clinical trials ; conditional power ; stochastic curtailing ; interim analysis
|
|
Abstract:
|
Consider a double blind, randomized study with a primary endpoint of subject's time to intervention. The endpoint usually is reached within 10 days of device placement. An interim analysis is planned after approximately 25% of patients have been enrolled in the study and reached the study end time point. With a time-to-event endpoint, the essential information is gained through the number of events in addition to the number of patients enrolled. The purpose of the proposed interim analysis is to determine if the trial is unlikely to achieve success (futility analysis) using conditional power. Conditional power is the probability that the reference test will reject the null hypothesis, given the data observed at the interim point. We will present, through simulations, various methods for futility stopping based on conditional power for the primary endpoint.
|
