|
Activity Number:
|
123
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303642 |
|
Title:
|
Brick Tunnel Randomization: A Way to Accommodate a Problematic Allocation Ratio in Adaptive Design Dose-Finding Studies
|
|
Author(s):
|
Olga M. Kuznetsova*+ and Yevgen Tymofyeyev
|
|
Companies:
|
Merck & Co., Inc. and Merck & Co., Inc.
|
|
Address:
|
RY34-A316, Rahway , NJ, 07065-0900,
|
|
Keywords:
|
allocation in dose-finding studies ; large block size ; constrained randomization ; problematic allocation ratio ; permuted block randomization ; maximal procedure
|
|
Abstract:
|
Adaptive design dose-finding studies commonly allocate small cohorts of subjects to a large number of treatment arms in a problematic allocation ratio, such as 0.2: 0: 0.12: 0.35: 0.25: 0.08: 0. Permuted block randomization will not work well for a small cohort of subjects as the required block size would be huge. A solution, which generalizes the maximal procedure [Berger, Ivanova, and Knoll, 2003] to the case of K=2 treatment arms and any allocation ratio, is offered. Brick tunnel randomization requires the allocation path to venture along the ridges of the k-dimensional unitary cubes that are pierced by the allocation ray that starts at the origin and has slopes proportional to the intended allocation ratio. The transition probabilities at each node within the tunnel are defined in such a way that the unconditional allocation ratio is the same for every allocation step.
|
