|
Activity Number:
|
322
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #305497 |
|
Title:
|
Comparing CRM with Frequentist t-Statistic Method for Phase I Clinical Trials in Oncology
|
|
Author(s):
|
Amit Phansalkar*+
|
|
Companies:
|
Cytel, Inc.
|
|
Address:
|
675 Massachusetts ave, Cambridge , MA, 02139,
|
|
Keywords:
|
adaptive ; CRM ; Bayesian ; frequentist ; oncology ; simulation
|
|
Abstract:
|
There are several ways to design and analyze Phase I clinical trials in oncology where the primary objective is to estimate the maximum tolerated dose (MTD). We will describe results from simulation experiments that compare the traditional 3+3 design, the Bayesian Continual reassessment method (CRM) design and a frequentist t-statistic method of Ivanova, Bolognese et. al.(2008). The CRM design implemented here includes extensions that use the power model to estimate the posterior probability of response at each dose level. The methods will be compared using data generated from a series of dose-response curves reflecting high early response as well as late rising mean response and the null case. Effectiveness of the methods will be examined along the dimensions such as power, probability of selecting MTD at the target dose, the expected sample size and the probability of stopping early.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
