Activity Number:
|
153
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 12, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biometrics Section*
|
Abstract - #301463 |
Title:
|
Predicting the Future Profile of Prostate-specific Antigen Levels Following Radiation Therapy for Prostate Cancer.
|
Author(s):
|
Dirk Moore*+ and Alexandra Hanlon
|
Affiliation(s):
|
Temple University and Fox Chase Cancer Center
|
Address:
|
Speakman Hall 006-00, Philadelphia, Pennsylvania, 19122,
|
Keywords:
|
PSA ; prostate cancer ; non-linear random effects
|
Abstract:
|
Prostate-specific antigen (PSA) levels fall sharply after treatment of a cancerous prostate with radiation therapy. When the treatment is successful, the PSA levels remain low indefinitely. But in some patients the PSA levels rise, sometimes sharply, after the initial decline. We have developed a random effects quadratic-linear spline model for the post-treatment PSA profile from a large database of prostate cancer patients followed for up to ten years. We use the fitted model to predict the future PSA trajectory of a new patient, and to continuously update the predicted trajectory as new PSA measurements are collected on that patient.
|