|
Activity Number:
|
220
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
ENAR
|
| Abstract - #307852 |
|
Title:
|
A Strategic Approach to Designing Early Phase Cancer Trials
|
|
Author(s):
|
Tze Leung Lai*+ and Philip Lavori and Mei-Chiung Shih
|
|
Companies:
|
Stanford University and Stanford University and Stanford University
|
|
Address:
|
Department of Statistics, Stanford, CA, 94305-4065,
|
|
Keywords:
|
adaptive design ; cancer trials ; safety ; efficacy ; futility stopping ; internal pilot
|
|
Abstract:
|
Early phase cancer trials are typically small in size and relatively short in duration. There are also ethical constraints in not subjecting patients to excessive toxicity. The goals and scope of such trials are necessarily limited. We propose a strategic approach that incorporates the longer-term goal towards establishing the efficacy and safety of the new therapy in the design of early phase trials. In particular, by making use of recent advances in adaptive design and inference, our strategic approach can enable the results of the Phase II trial to be used as an internal pilot of a later Phase III trial. Early stopping for futility will also save resources if the longer-term goal is found unlikely to be met. These ideas for Phase II/III trials can also be modified for Phase I/II studies. Institutional issues in implementing these trials are also discussed.
|
