|
Activity Number:
|
473
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #307061 |
|
Title:
|
A Statistical Method To Integrate Independent Review and Investigator Review in Clinical Cancer Trial
|
|
Author(s):
|
Xiaolong Luo*+
|
|
Companies:
|
Johnson & Johnson Pharmaceutical R&D
|
|
Address:
|
920 Route 202, S., Raritan, NJ, 08869,
|
|
Keywords:
|
clinical trial ; cancer ; survival analysis ; informative censoring ; independent review ; tumor assessment
|
|
Abstract:
|
In open label cancer trials, investigators are aware of treatment group assignment. This knowledge could consciously or unconsciously bias treating physician's assessment of the patient's treatment outcome. In order to avoid this bias, many phase III oncology trials are implemented with an independent review for tumor assessment. At the same time, a patient's treatment plan is driven by the protocol, leading to potential informative censoring. Furthermore, the data with informative censoring cannot be correctly analyzed and interpreted with procedures such as Kaplan Meier method and log rank test. The dilemma would be either taking investigator assessment with bias from knowing the treatment assignment or taking independent review with bias from the analysis procedures. We will illustrate this situation and solve the problem with actual clinical trials data and simulation analyses.
|
- 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 2006 program |
