|
Activity Number:
|
140
- Design and Analysis of Cancer Immunotherapy Trials
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, July 30, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #326465
|
Presentation
|
|
Title:
|
Use of Piecewise Weighted Log-Rank Test for Trials with Delayed Treatment Effect
|
|
Author(s):
|
Boguang Zhen* and Zhenzhen Xu and Bin Zhu and Yongsoek Park
|
|
Companies:
|
FDA and FDA and NIH/NCI and Unversity of Pittsburgh
|
|
Keywords:
|
delayed treatment effect ;
weighted log-rank test ;
clinical trials
|
|
Abstract:
|
If a new product has a delayed start of treatment effect in a clinical trial designed to compare time-to-event endpoint, the use of the regular log-rank test with equal weight across all time points may lead to wrong estimates of study duration and statistical power due to the violation of the proportional hazards assumption. Although many methods have been developed to deal with delayed treatment effect based on various classes of weighted log-rank tests, the current practice in most clinical trials to account for delayed effect is still either ignoring it or increasing the sample size, and then performs the analysis using the regular log-rank test. This may be partially due to lack of awareness of the new methods as well as due to the concern on the suitability of using these methods in clinical trials. In this paper, we evaluate any potential concerns particularly related to the piecewise weighted log-rank test (PWLRT) as proposed by Xu et al. and discuss the practical use of it in order to address these concerns on the use of PWLRT for clinical trials with delayed effect. The evaluation results can also be generalized to other weighted methods that deal with delayed effect.
|
Authors who are presenting talks have a * after their name.
