|
Activity Number:
|
29
- Statistical Issues Specific to Therapeutic Areas, Power and Sample Size Calculations, and Trial Monitoring
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 8, 2021 : 1:30 PM to 3:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #318209
|
|
|
Title:
|
Sample Size Reestimation for the Finkelstein and Schoenfeld Test Statistic for a Composite Endpoint with Two Components
|
|
Author(s):
|
Thomas Zhou* and Joseph Massaro
|
|
Companies:
|
Boston University and Boston University
|
|
Keywords:
|
Prioritized outcomes;
adaptive clinical trial;
interim analyses;
conditional power;
promising zone
|
|
Abstract:
|
Calculating power for a composite endpoint is often performed through simulations, where computational costs rise with increasing sample size and complexity of the endpoint. Mis-specifications in initial estimates of effect size for each component of the composite may also result in the study being underpowered. While many methods have been developed to adaptively increase sample size at some interim stage by boosting the conditional power to a desired target power, this process for composite endpoints typically involve extensive simulations, which share in the aforementioned computational limitations. In this study, we propose an analytical solution to calculate conditional power for the Finkelstein-Schoenfeld test for prioritized composite endpoints with two components. Conditional power formulas are derived assuming the population-level underlying distributions in each of the component outcomes are the same as those observed in the interim data. Sample size re-estimation is performed through the Mehta-Pocock (2011) promising zone algorithm. Monte-Carlo simulations demonstrate consistency in performance of the formulas and robustness to mis-specified distributional assumptions.
|
Authors who are presenting talks have a * after their name.
