Electronic Thesis and Dissertation Repository


Master of Science


Epidemiology and Biostatistics


Dr. George Rodrigues


Purpose: This study aims to determine prognostic factors for overall survival (OS) after biochemical failure (BF), and identify a risk stratification system to predict OS for prostate cancer patients with BF.

Methods: Univariable and multivariable Cox proportional hazards regression analyses, and recursive partitioning analysis (RPA) were conducted using data from 1246 patients who experienced BF in the Genitourinary Radiation Oncologists of Canada (GUROC) Prostate Cancer Risk Stratification (ProCaRS) database. Two thirds of patients were randomized into a Training cohort (n=831), and the final third into a Validating cohort (n=415).

Results: Age, baseline PSA, T stage, Gleason score, hormone therapy, radiation therapy, nadir PSA, TTBF and pre-BF PSADT were significant (p

Conclusions: The 2-class post-treatment risk stratification system allows for the identification of high risk and low risk patients in terms of OS after BF to help guide patient selection for future clinical trials and clinical treatment decision-making.