Click here to go to the previous page
Looking Closer into the Utility of Adaptive Approaches
Program Code:
259
Date:
Tuesday, June 25, 2013
Time:
1:45 PM to 3:15 PM
EST
CHAIR
:
Sue-Jane Wang, PhD,MA,MS (SCHAGY), Assoc. Dir., Adaptive Design & Pharmacogenomics, OB, OTS, CDER, FDA, United States
SPEAKER
(S):
Raina Duan, Aptiv Solutions, United States
Valerii Fedorov,
DrSc (SPKNON), Vice President, Predictive Analytics, Innovation,
Quintiles Transnational Corp, United States
Dr. Federov is VP & Head of the Predictive Analytics, Innovation, at Quintiles. Previously he was head of Research Statistics Unit at GSK and Sr Research Statistician at Oak Ridge National Lab. Author of +200 publications/books in various areas of statistics & biostatistics. ASA Fellow.
|
Jonathan (Yoni) D. Lebowitsch, DrSc (SPKNON), Product Manager, Medidata Solutions Worldwide, United States
Description
This symposium will discuss the advantages and disadvantages of a few methodologies that are directly related to implementation of adaptive designed trials. A weighted average approach on the type I error and type II error will be introduced and will be compared with the standard approach where type I error is fixed and type II error is to be minimized. As for randomization, a hybrid randomization algorithm will be presented, which will be contrasted with conventional randomization to illustrate the pros and cons among the algorithms. Finally, adaptive dose finding designs including penalized D-optimal design, CRM design and Bayesian adaptive designs will be compared. In addition, the tradeoff between the desire to maximize information and various constraints will be discussed while gaining information under ethical/cost constraints in adaptive dose finding studies.