Title
|
* ! Practical Aspects of Adaptive Designs in Clinical Trials
|
Date / Time / Room
|
Sponsor
|
Type
|
08/12/2002
8:30 AM -
10:20 AM
Room: H-West Ballroom
|
Biometrics Section*, Biopharmaceutical Section*, ENAR
|
Topic Contributed
|
Organizer:
|
Paul Gallo, Novartis Pharmaceuticals
|
Chair:
|
Paul Gallo, Novartis Pharmaceuticals
|
Discussant:
|
9:55 AM - Lloyd Fisher, University of Washington
|
Floor Discussion
|
10:15 AM
|
Description
During the past decade, considerable progress has been made in developing designs for clinical trials which can provide robustness against incorrect initial assumptions, reduce the number of patients exposed to inefficient or even hazardous treatments, and increase the changes for tials to provide conclusive results. These designs allow knowledge gained from the accumulating information to be used for implementing data-driven modifications during a trial, and are called adaptive, flexible, or self-learning. The challenge is to guarantee in a frequentist framework that predefined levels of hypothesis tests are maintained even if important design features are modified. Such features may include, in addition to termination of a trial or of treatment arms at interim analyses, the following: sample size reassessments, adaption of test statistics (e.g., weights in contrasts), modification of multiple testing strategies, change of primary efficacy parameters, or changes in inclusion / exclusion criteria.
The aim of this session is to present an integrated overview of recent developments in this area, both in Europe and the U.S., and in particular to discuss practical aspects in the design, conduct, and analysis of adaptive clinical trials, as well as in estimation and decision making.
|
|