|
Activity Number:
|
60
- Adaptive Design and Statistical Consideration in Clinical Trials
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 30, 2017 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #323926
|
|
|
Title:
|
Investigation on Combining Phase II Proof-Of-Concept Trial with Dose Finding Trial
|
|
Author(s):
|
Yutao Liu* and Qiqi Deng and Ken Cheung
|
|
Companies:
|
Columbia University and Boehringer Ingelheim Pharmaceuticals, Inc. and Deptment of Biostatistics, Columbia University
|
|
Keywords:
|
dose finding ;
adaptive design ;
MCPMod
|
|
Abstract:
|
In this project we explore a seamless phase IIa/IIb dose-finding design that provides better balance between efficiency of the development and the risk of large upfront investment. In the 1st stage of the trial, patients will be allocated to highest dose versus placebo. In the 2nd stage, patients will be allocated to multiple doses of interest according to pre-specified overall allocation ratio of the trial. A futility analysis will be conducted once the endpoints of the patients in the 1st stage are available. Allocating all patients to highest dose in active arm will minimize the sample size needed for earlier signal of Go/NoGo decision. Go/NoGo decision in futility analysis is to continue the expansion of full dose ranging or not. The impact of sample size, futility boundary and different allocation ratio between stages and doses were evaluated and compared among various options. By re-using the sample size of PoC portion in analysis, and providing option of stopping at interim, the design reduces the expected overall sample size and time needed for phase II study while maintaining similar type I and power. The impact on the dose estimation behaviors were also explored.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2017 program
|
