|
Activity Number:
|
139
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #307511 |
|
Title:
|
An Experimental Design for Clinical Trials Evaluating Combination Agents
|
|
Author(s):
|
Christine McLaren*+ and Vernon M. Chinchilli and Wen-Pin Chen and Frank L. Meyskens
|
|
Companies:
|
University of California, Irvine and The Pennsylvania State University and Chao Family Comprehensive Cancer Center and Chao Family Comprehensive Cancer Center
|
|
Address:
|
Epidemiology Division, Irvine, CA, 92697-7550,
|
|
Keywords:
|
2x2 factorial design ; Cochran-Armitage trend test ; cancer prevention ; simple loop alternative
|
|
Abstract:
|
For clinical trials that intend to test combination agents, new designs are needed to achieve results with fewer patients and less expense. We propose a 2x2 balanced factorial design to test whether a combination of agents is better than the individual treatments or placebo with regard to reducing the proportion of recurrent adenomas, precursors to colorectal cancer. The Cochran-Armitage test of trend can be used to compare the proportion of patients developing at least one adenoma between placebo and treatment groups. We derive a pair of two-sided tests for the ordered alternative of increasing or decreasing proportions across treatment groups and a method for power and sample-size calculations. Simulations demonstrate type I and II error rates. The design achieves a reduction in sample size of more than 30% and promises economy of patients and expense.
|
