|
Activity Number:
|
380
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #305639 |
|
Title:
|
Adaptive Prediction in Genomic Signatures--Based Clinical Trials
|
|
Author(s):
|
Yang Xie*+ and Guanghua Xiao and Chul W. Ahn and Luc Girard and John Minna
|
|
Companies:
|
The University of Texas Southwestern Medical Center at Dallas and The University of Texas Southwestern Medical Center at Dallas and The University of Texas Southwestern Medical Center at Dallas and The University of Texas Southwestern Medical Center at Dallas and The University of Texas Southwestern Medical Center at Dallas
|
|
Address:
|
5323 Harry Hines Blvd, Dallas, TX, 75390-8551,
|
|
Keywords:
|
Adaptive Prediction ; Personalized Medicine ; Clinical Trials
|
|
Abstract:
|
Personalized medicine is defined by predicting the patients' clinical outcomes using their molecular profiling data and other clinical information before treatment and thereby selecting the best possible therapies. To prove the worth of personalized medicine and bring it into clinical practice, well-designed clinical trials are essential steps. I will present a procedure that builds prediction model based on training data, uses this model to predict the best treatment for individual patients enrolled in the trial, and then updates the model once the outcome of patient is available. The updating is conducted through a re-weighted random forest model accounting for the heterogeneity between training and testing data. Both simulation data sets and oncology data sets are used to show the performance of the method.
|
