|
Activity Number:
|
109
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #300799 |
|
Title:
|
Identification of Promising Subgroups in the Retrospective Analysis of Clinical Trials
|
|
Author(s):
|
Ilya Lipkovich*+ and Alex Dmitrienko and Eric Su and Jonathan Denne and Gregory Enas
|
|
Companies:
|
Eli Lilly and Company and Eli Lilly and Company and Eli Lilly and Company and Eli Lilly and Company and Eli Lilly and Company
|
|
Address:
|
Lilly Corporate Center, Indianapolis, IN, 46285,
|
|
Keywords:
|
Recursive partitioning ; Classification trees ; Data mining ; Retrospective data analysis
|
|
Abstract:
|
Large amounts of data from clinical studies with a nonsignificant primary analysis are often underutilized and drug development programs are terminated early without a rigorous examination of subgroups of patients that could potentially benefit from the treatment. In this talk we propose a novel methodology for identifying such promising subgroups, based on a recursive tree partitioning algorithm. The proposed procedure incorporates a built-in mechanism for protection against false discovery by adjusting for multiplicity, tuning a complexity penalty for the data splitting criterion using cross-validation, and replication across validation datasets. In some cases these protections may not only generate hypotheses, but test and confirm hypotheses to drive regulatory decisionmaking. The results of a simulation study are presented and an application to real clinical trials is discussed.
|
