Validation of a treatment policy for patients with prostate specific antigen failure after three-dimensional conformal prostate radiation therapy

BACKGROUND. The objective of this report was to present an outcomes validation for the Fox Chase Cancer Center (FCCC) management policy for patients who demonstrate prostate specific antigen (PSA) failure after receiving three-dimensional conformal radiation therapy (3DCRT). METHODS. Eligible patients included 248 men with T1-T3N0M0 prostate carcinoma who demonstrated PSA failure (according to the American Society for Therapeutic Radiology and Oncology definition) after completing definitive 3DCRT alone or with androgen deprivation (AD) therapy between May 1989 and November 1997. The primary endpoint evaluated was freedom from distant metastasis (FDM). The secondary endpoints evaluated included cause specific survival (CSS) and overall survival (OS). The variables evaluated in the multivariate analyses (MVA) included initial PSA, Gleason score, T classification, dose, PSA nadir, time to PSA failure, PSA doubling time (PSADT), initial use of AD therapy, and the use of AD therapy upon PSA failure. RESULTS. The 5-year FDM, CSS, and OS rates for the entire group were 76%, 92%, and 76%, respectively. It was found that four variables were independent predictors of FDM: Gleason score (P = 0.0039), PSA nadir (P = 0.0001), PSADT (P = 0.0001), and the use of AD on PSA failure (P = 0.0001). One hundred forty-eight men demonstrated a PSADT < 12 months. AD therapy was started in 59 men, and 89 men refused AD therapy and were observed. The use of AD therapy was associated with a significant improvement in the 5-year FDM rate (57% vs. 78%; P = 0.0026). In the group of men with PSADT < 12 months, the median time to distant failure was significantly longer in the men who received AD therapy (6 months vs. 25 months; P = 0.02). Of the 100 men with a PSADT ? 12 months, 89 men were observed, and 11 men received AD therapy. There was no improvement in the 5-year FDM rate with the use of AD therapy compared with observation (88% vs. 92%, respectively; P = 0.74). CONCLUSIONS. The current results validate the use of PSADT as an indicator of patients who may be observed expectantly or treated with AD therapy for PSA failure after 3DCRT. Prospective trials are needed to define further the optimal treatment for these patients.