|
Activity Number:
|
523
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 10, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304972 |
|
Title:
|
Molecular Classification of Prostate Tumors
|
|
Author(s):
|
Jaya M. Satagopan*+
|
|
Companies:
|
Memorial Sloan-Kettering Cancer Center
|
|
Address:
|
307 E. 63rd Street, New York, NY, 10021,
|
|
Keywords:
|
logistic regression ; risk prediction ; nomogram ; prostate cancer
|
|
Abstract:
|
Approximately 25%--40% of prostate cancer patients undergoing radical prostatectomy experience disease recurrence. Predicting which patients are likely to recur can be valuable for proper disease management. The nomogram estimates the probability of recurrence in a defined time period given the clinical characteristics of a patient. However, the predictive performance of the nomogram needs improvement. Disease outcome, such as prostate cancer recurrence, is a complex combination of various genomic and exposure factors. We develop a prediction tool based on penalized logistic regression to determine five-year disease recurrence among prostate cancer patients using genomic risk factors and other clinical variables. The predictive performance of the resulting model is compared to a model based solely on genomic factors and a model based on clinical variables alone.
|
- 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 |
