|
Activity Number:
|
8
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #318304
|
|
|
Title:
|
Adjusting for Enrichment Effects When Estimating Oncology Biomarker Clinical Utility
|
|
Author(s):
|
Jared Lunceford*
|
|
Companies:
|
Merck Research Laboratories
|
|
Keywords:
|
biomarker ;
oncology ;
enrichment ;
missing data
|
|
Abstract:
|
Evaluating the potential utility of oncology biomarkers to identify patients most likely to respond to investigational therapies involves estimating clinical outcomes above and below the putative cut-offs on the biomarker assay. Bias in the estimation of response rates or survival time related quantities can occur under various constraints on the underlying biomarker study. In some cases, enrichment strategies may have been used to selectively enroll oncology patients into clinical trials to more efficiently demonstrate therapeutic benefit. In other cases the biomarker itself is selectively assessed in retrospective analyses using only a subset of enrolled patients based on clinical outcome. We evaluate these different forms of selective enrichment and the potential for bias and propose strategies for mitigating such bias using a missing data framework and propose both frequentist and Bayesian methods that could be useful, illustrating their relative performance through simulation.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2016 program
|
