Abstract Details
Activity Number:
|
428
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biopharmaceutical Section
|
Abstract - #309934 |
Title:
|
Phase II/III Seamless Adaptive Dose-Selection Design for Longitudinal Patient Data
|
Author(s):
|
Caitlyn Ellerbe*+ and Jordan Elm and Viswanathan Ramakrishnan and Bruce Turnbull and Stacia DeSantis and Edward Jauch and Valerie Durkalski
|
Companies:
|
Medical University of South Carolina and Medical University of South Carolina and Medical University of South Carolina and Cornell University and The University of Texas Health Sciences and Medical University of South Carolina and Medical University of South Carolina
|
Keywords:
|
Adaptive Seamless Design ;
Clinical Trial ;
Dose Selection ;
Longitudinal Data ;
Selection Bias ;
generalized linear mixed model (GLMM)
|
Abstract:
|
Adaptive designs offer investigators the ability to modify trial parameters to promote safety and trial efficiency. However, in longitudinal phase II/III trials, adaptive designs are limited in the ability to use partial information without inflating the type I error. We propose a two-stage design for a continuous endpoint measured at several visits after enrollment. In stage I several doses of interest are compared to a control, and the optimal dose is selected using all available data. In stage II the efficacy of the selected dose relative to a placebo is tested using data from new subjects as well as the data used for the dose selection in stage I. We propose a correction to the test statistic, which we show, theoretically and through simulations controls the type I error rate, for selecting one of two doses observed over two time points. For more general situations, where there are multiple doses and more than two visits, a bootstrap procedure is proposed to control the type I error. This procedure provides a mechanism to generalize the design to more complex situations that include physician guided dose selection and dose-response models.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2013 program
|
2013 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.