Activity Number:
|
30
- Statistical Considerations in Adaptive Designs
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, August 7, 2022 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract #322665
|
|
Title:
|
A Note on Multiplicity Control for Trials with Adaptation Based on Intermediate Endpoint
|
Author(s):
|
Wenjun Zheng* and Chenguang Wang and Bret Musser
|
Companies:
|
Regeneron Pharmaceuticals and Regeneron Pharmaceuticals and Regeneron Pharmaceuticals
|
Keywords:
|
Type I error;
Adaptive design;
Intermediate endpoint;
Multiplicity control
|
Abstract:
|
In the past decade, Adaptive designs (ADs) have grown in popularity in the pharmaceutical industry, which have been accompanied by an intense discussion on ADs in the statistical literatures. ADs with adaptations based on an intermediate endpoint are considered as a special topic in the FDA AD guidance and have received relatively less attention compared to designs with adaptations to sample size or with treatment selections. Existing literatures about ADs with adaptations based on an intermediate endpoint are mostly for oncology trials with time to event intermediate and primary endpoints. The potential inflation on the Type I error for such AD is currently a less well-studied phenomenon than other ADs. We considered an AD with an intermediate composite endpoint and a primary endpoint that is a subset of events of the intermediate endpoint. We explored various multiplicity control strategies and showed the advantages and caveats with different options. Our findings were applied to a hypothetical seamless phase 2-3 study for demonstration.
|
Authors who are presenting talks have a * after their name.