|
Activity Number:
|
334
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #309324 |
|
Title:
|
Adaptive Design with Continuous Responses and Binary Utility
|
|
Author(s):
|
Yuehui Wu*+ and Valeri V. Fedorov
|
|
Companies:
|
GlaxoSmithKline and GlaxoSmithKline
|
|
Address:
|
1250 S Collegeville Rd, Collegeville, PA, 19426,
|
|
Keywords:
|
multiple endpoints ; utility function ; optimal design ; dichotomization
|
|
Abstract:
|
In clinical trials, correlated multiple endpoints problems are commonly seen. Sometimes, the observed responses are continuous but a regulatory agency will approve the drug only if the probability that the efficacy measure exceeds a predefined threshold while the toxicity does not exceed another given threshold. Thus the probability of interest (utility) is based on dichotomized responses. We emphasized the usefulness of differentiation between response and utility functions and developed tools to build fully adaptive and composite designs for the corresponding models. We use locally optimal designs as benchmarks. Two criteria: D-optimality and L-optimality (variance of estimated location) are considered. It is also shown that the practice of reporting dichotomized responses leads to a substantial loss in the precision of estimated parameter.
|
