|
Activity Number:
|
8
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #318040
|
|
|
Title:
|
Bias in Lifetime Data Analysis for Validation of Biomarkers
|
|
Author(s):
|
Tinghui Yu* and Yabing Mai
|
|
Companies:
|
FDA/CDRH and Merck Research Laboratories
|
|
Keywords:
|
Aggregate data ;
Combined hazard ratio ;
Cox proportional hazard model ;
Heterogeneous survival data ;
Misspecified models
|
|
Abstract:
|
Personalized medicines are developed to target patient population harboring specific biomarker expressions. Responses to these medicine can vary dramatically with the patients' biological profile. Therefore, the estimated efficacy of the medicine is subject to impacts of the imprecision in the companion diagnostics used for patient selection as well as the statistical methods chosen for data analyses. We developed two different methods to provide precise measurement of the bias and variation in the hazard ratio estimates when heterogeneity is observed in the enrolled patients. We will demonstrate these novel tools for life-time-data analyses using real life examples.
|
Authors who are presenting talks have a * after their name.
