Online Program Home
My Program

Abstract Details

Activity Number: 491
Type: Contributed
Date/Time: Wednesday, August 3, 2016 : 8:30 AM to 10:20 AM
Sponsor: Section on Statistics in Genomics and Genetics
Abstract #319031
Title: Pathway-Structured Predictive Model for Cancer Survival Prediction
Author(s): Xinyan Zhang* and Yan Li and Omotomilayo F. Akinyemiju and Akinyemi I. Ojesina and Phillip Buckhaults and Bo Xu and Nengjun Yi
Companies: University of Alabama at Birmingham and University of Alabama at Birmingham and University of Alabama at Birmingham and University of Alabama at Birmingham and University of South Carolina and Southern Research Institute and University of Alabama at Birmingham
Keywords: pathway ; prediction ; C-index ; breast cancer ; penalized Cox regression ; hierarchical Cox model
Abstract:

Heterogeneity of prognosis and prediction based on clinical factors or molecular signatures in cancer treatment across patients has been a persisted problem over decades. One of the main shortcomings of the previous studies is the failure to incorporate pathway-based genetic structure of cancer into the predictive model. To address this problem, we propose a two-stage procedure to incorporate pathway information into the predictive modeling using large-scale gene expression data. In the first stage, we fit all predictors within each pathway using penalized Cox model and Bayesian hierarchical Cox model. In the second stage, we combine the cross-validated prognostic scores of all pathways obtained in the first stage as new predictors to build an integrated prognostic model for prediction. We used the proposed method to analyze a breast cancer data set from The Cancer Genome Atlas (TCGA) project for predicting overall survival using gene expression profiling. The results show that the proposed approach not only improves survival prediction compared with the alternative analysis that ignores the pathway information, but also identifies significant biological pathways.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

 
 
Copyright © American Statistical Association