|
Activity Number:
|
177
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #302094 |
|
Title:
|
Adaptive Dose Selection Using Interim Analyses When Fitting a Bayesian Emax Model to Clinical Trial Data
|
|
Author(s):
|
Neal Thomas*+ and Byron Jones and Patrick Johnson and Helen Richardson
|
|
Companies:
|
Pfizer, Inc. and Pfizer, Inc. and Pfizer, Inc. and Pfizer, Inc.
|
|
Address:
|
50 Pequot Avenue, New London, CT, 06320 ,
|
|
Keywords:
|
Emax model ; dose adaptive design ; dose ranging clinical trial ; Bayes ; stopping rules
|
|
Abstract:
|
The presentation will describe an adaptive dose response study to compare the mean effects of a set of doses with placebo using a nonlinear (Emax) function. The design has two parts: the first part is a small parallel group trial followed by a second part which is adaptive in nature. The initial selection of doses is made using optimal design theory with power to detect a difference of a specified size between the highest dose and placebo. Subsequent patients are recruited in cohorts and the data are reanalysed using a Bayesian nonlinear model. The doses for each cohort are selected using Bayesian posterior probabilities of satisfying various criteria for success. The Bayesian posterior probabilities will also be used to stop the trial early for success or futility if necessary. The operating characteristics of the adaptive design will be assessed.
|
