|
Activity Number:
|
277
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:30 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #310022 |
|
Title:
|
An Application of Adaptive Randomization Using Hierarchical Bayes Model in a Prospective Biomarker-Based Clinical Trial
|
|
Author(s):
|
Suyu Liu*+ and Edward S. Kim and Xian Zhou and Ignacio Wistuba and Roy Herbst and Jeffrey Lewis and J. Jack Lee
|
|
Companies:
|
The University of Texas M.D. Anderson Cancer Center and The University of Texas M.D. Anderson Cancer Center and Genentech, Inc. and The University of Texas M.D. Anderson Cancer Center and The University of Texas M.D. Anderson Cancer Center and The University of Texas M.D. Anderson Cancer Center and The University of Texas M.D. Anderson Cancer Center
|
|
Address:
|
1515 Holcombe Blvd Unit 447, Houston, TX, 77030,
|
|
Keywords:
|
adaptive randomization ; clinical trial ; hierarchical Bayes model ; operating characteristics
|
|
Abstract:
|
To identify best-matched treatments for patients, we implement an outcome-based adaptive randomization in BATTLE trial (Biomarker-based Approaches of Targeted Therapy for Lung Cancer Elimination). It consists of 4 parallel phase II studies each with targeted therapies. Patients require a core biomarker biopsy prior to randomization. A hierarchical Bayes model is used to characterize efficacy rates among the 4 treatments for each biomarker profile. Based on the posterior clinical efficacy, patients are adaptively randomized according to their real-time biomarker status. The operating characteristics based on simulations indicate that the design can accurately identify the effective biomarker-treatment combinations, and allocates more patients to more efficacious treatments - a step toward "personalizing medicine". Examples of data realization and practical considerations will be given.
|
