|
Activity Number:
|
233
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303716 |
|
Title:
|
A Fixed Study Design with Conditional Serial Assessment of Co-Primary Endpoints: An Application to a Single-Arm Pilot Oncology Trial
|
|
Author(s):
|
Jonathan Mahnken*+ and Jo Wick and Byron J. Gajewski and Matthew S. Mayo
|
|
Companies:
|
The University of Kansas Medical Center and The University of Kansas Medical Center and The University of Kansas and The University of Kansas Medical Center
|
|
Address:
|
MSN 1026, 3901 Rainbow Blvd, Kansas City, KS, 66160,
|
|
Keywords:
|
Clinical trials ; Multiple endpoints ; Operating characteristics ; Pilot study ; Power ; Sample size
|
|
Abstract:
|
Optimizing study designs is a major responsibility for applied statisticians in academic health centers. We describe the study design for an early-phase, oncology trial with two co-primary endpoints; however, evaluation of the second endpoint was conditional on a positive response to the first endpoint, so these dichotomous measures were collected conditionally and in series. We presented the probability density function for the distribution of this study design, and derived parameters that described this distribution. We also derived the maximum likelihood estimates for the unknown response probability parameters of primary interest to the trial investigator. Operating characteristics were derived theoretically and then validated via a simulation study for the motivating example.
|
