Activity Number:
|
29
- Bayesian Methods and Applications in Clinical Trials (I)
|
Type:
|
Contributed
|
Date/Time:
|
Sunday, July 30, 2017 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract #324646
|
|
Title:
|
Borrowing Historical Information for Clinical Development - a Case Study
|
Author(s):
|
Guochen Song* and Peng Sun and John Zhong and Jie Li and Baoguang Han and Stacy Lindborg
|
Companies:
|
Biogen and Biogen and Biogen and Biogen and Biogen and Biogen
|
Keywords:
|
Bayesian ;
Survival ;
borrowing historical information
|
Abstract:
|
Nusinersen is the first FDA approved treatment for spinal muscular atrophy (SMA). Infantile-onset SMA is a rare fatal disease. The phase III study of nusinersen was ended early based on a positive interim analysis and to allow patients to gain early access to treatment. As a consequence the second primary endpoint of the study, ventilation free survival, was not expected to be sufficiently powered in the final analysis. A covariate-adjusted Bayesian hierarchical piece-wise exponential model was used to borrow information from the phase II trial as well as from two observational natural history trials. In this presentation, we will report the results of the planned Bayesian analysis to demonstrate that borrowing historical information narrows the creditable interval of the hazard ratio.
|
Authors who are presenting talks have a * after their name.