|
Activity Number:
|
134
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #308335 |
|
Title:
|
Statistical Methods for Assessing Long-Term Stability of Compounds in Biological Matrices
|
|
Author(s):
|
David M. Hoffman*+ and Robert Kringle
|
|
Companies:
|
sanofi-aventis and sanofi-aventis
|
|
Address:
|
1041 Route 202206, Bridgewater, NJ, 08807,
|
|
Keywords:
|
stability ; variance components ; mixed model ; regression
|
|
Abstract:
|
The purpose of a long term stability study is to demonstrate that clinical samples are stable over the maximum period of sample storage expected in future clinical trials. Because these studies often last two years or longer, a particular difficulty is dealing with frequently large random month-to-month variability. Various study designs, analysis methods, and decision procedures have been proposed in the literature. We propose a new design-analysis procedure that utilizes a concurrent control sample to adjust for random month-to-month fluctuations in a mixed model regression analysis. Performance of the new procedure is evaluated and compared to other procedures by simulation and by application to historical studies.
|
