|
Activity Number:
|
79
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303897 |
|
Title:
|
An Efficient Algorithm to Determine the Optimal Two-Stage Randomized Multinomial Designs in Oncology Clinical Trials
|
|
Author(s):
|
Yong Zhang*+ and William L. Mietlowski and Bee Chen and Yibin Wang
|
|
Companies:
|
University of Michigan and Novartis Oncology and Novartis Oncology and Novartis Oncology
|
|
Address:
|
Dept. of Biostatistics, 1420 Washington Heights, Ann Arbor, MI, 48109,
|
|
Keywords:
|
Oncology clinical trials ; Two-stage randomized multinomial design ; Response rate ; Early progression rate
|
|
Abstract:
|
In this paper we develop an efficient algorithm to determine the optimal two-stage randomized multinomial designs in Phase II oncology clinical trials. Sun et al. (2007) proposed an optimal two-stage randomized multinomial design simultaneously considering response rate and early progression rate. The existing software to compute the sample size and boundaries at each stage is time consuming to execute. We develop an algorithm for the two-treatment case using an approximation method. Simulations show that our approximation is quite accurate while saving more than 90% in computation time. This enables the evaluation of multiple scenarios of alternative hypotheses and/or error rates in a timely manner thereby increasing the utility of the randomized two-stage multinomial design in practice.
|
