|
Activity Number:
|
64
- Statistical Issues Specific to Therapeutic Areas
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #311018
|
|
|
Title:
|
An Extension of Bayesian Principal Stratum Analysis of Clinical Disability Progression in SPMS
|
|
Author(s):
|
Zhe Chen* and Katherine Riester
|
|
Companies:
|
Biogen and Biogen
|
|
Keywords:
|
SPMS;
MS relapse;
Estimand;
Principal stratification;
Subgroup identification;
Bayesian analysis
|
|
Abstract:
|
A recent analysis of data from a large randomized placebo-controlled trial in the secondary progressive multiple sclerosis (SPMS) patients assessed the efficacy of an MS drug versus placebo in a subgroup of patients who would not relapse during the trial using a Bayesian principal stratum (PS) method. The purpose of the analysis was to study the efficacy of the drug in delaying the MS clinical disability progression (CDP) after removing its confounding effect in reducing the MS relapse. The estimand of interest is the relative risk of developing CDP between the drug and placebo among a counterfactual stratum of non-relapsers who would never experience a relapse irrespective of treatment to be received. This analysis used the fixed time intervals (e.g. 12 or 24 months after randomization) to define the PS. With this approach a relatively small fixed time interval might be insufficient to capture the PS for the whole study follow-up. We explore an extension of this approach by modelling the annualized relapse rate in defining the PS and present illustrations on simulated MS data.
|
Authors who are presenting talks have a * after their name.
