|
Activity Number:
|
138
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #309210 |
|
Title:
|
Bayesian Variable Selection with Joint Modeling of Categorical and Survival Outcomes: An Application to Individualizing Chemotherapy Treatment in Advanced Colorectal Cancer
|
|
Author(s):
|
Wei Chen*+
|
|
Companies:
|
Karmanos Cancer Institute
|
|
Address:
|
Biostatistic Core, Detroit, MI, 48201,
|
|
Keywords:
|
Multivariate Regression ; Hierarchical Model ; Latent Variable ; Interaction ; MCMC
|
|
Abstract:
|
Clinical studies for colorectal cancer have shown that genetic alterations lead to different responses to same treatment, despite the morphologic similarities of tumors. This article focuses on developing statistical method appropriate for individualizing treatment. The multivariate regression model with latent variables and structured variance covariance matrix considered here accounts for the correlated nature of multiple endpoints and the fact that endpoints have different statistical distributions. The mixture normal hierarchical structure incorporates a variable selection rule for interaction terms. The application to the advanced colorectal cancer study revealed the associations between multiple endpoints and certain alterations of biomarkers, demonstrating the potential of individualizing treatment based on genetic profiles.
|
