Activity Number:
|
15
- Incorporating Knowledge from Previous Clinical Trials into the Design of New Trials
|
Type:
|
Invited
|
Date/Time:
|
Monday, August 3, 2020 : 10:00 AM to 11:50 AM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract #309390
|
|
Title:
|
Bayesian Synthetic Control Approaches in Pediatric Drug Development: A Case Study in Acute Lymphoblastic Leukemia
|
Author(s):
|
Antara Majumdar* and Jingjing Ye and Rebecca Rothwell and Corinne Ahlberg
|
Companies:
|
Medidata Solutions and FDA and FDA and Acorn AI by Medidata, a Dassault Systèmes company
|
Keywords:
|
Synthetic Controls;
Clinical Trials;
Bayesian Analysis;
Bayesian Synthetic Controls;
External Evidence;
Historical Data
|
Abstract:
|
A fully-powered randomized controlled trial can be challenging to conduct on children because of difficulties in recruitment and enrollment of pediatric patients. One way to improve feasibility of trials in pediatric patients, when clinically appropriate, is through borrowing of information from comparable adult trials with adult approvals. We refer to patient-level, historical clinical trial data as Synthetic Control data. When borrowing information from comparable adult trials to inform a pediatric trial, the adult trial data constitutes the Synthetic Control data. We conducted a case study on Acute Lymphoblastic Leukemia (ALL) single-arm trials to explore the advantages and disadvantages of Bayesian Synthetic Control approaches for borrowing information from adult trials to support the estimation of stable parameter estimates for the pediatric trial. We report comparisons of Bayesian methods and results from this case study. We present considerations for future pediatric trial design.
|
Authors who are presenting talks have a * after their name.