Defining biochemical cure for prostate carcinoma patients treated with external beam radiation therapy

BACKGROUND. The authors retrospectively reviewed their institution's long t erm experience with conventional external beam radiation therapy (RT) for l ocalized prostate carcinoma to identify criteria associated with long term biochemical cure. METHODS, Between January 1987 and December 1994, 871 patients were treated with external beam RT alone for clinically localized prostate carcinoma at William Beaumont Hospital, Royal Oak, Michigan. Ail patients received only external beam RT to a median total dose of 66.6 grays (Gy) (range, 59.4-70. 4 Gy). No patient received hormonal therapy unless treatment failure was do cumented. The median follow-up was 5.0 years (range, 0.2-11.8 years). Bioch emical failure was defined according to the American Society for Therapeuti c Radiology and Oncology Consensus Panel definition. RESULTS. In the entire study group, 380 patients experienced biochemical fa ilure at a median interval of 1.5 years after the completion of RT. The 5-y ear and 7-year actuarial rates of biochemical control were 50% and 48%, res pectively. On multivariate analysis, a higher pretreatment prostate specifi c antigen (PSA) level, higher Gleason score, higher clinical T classificati on, higher nadir level, and shorter time interval to nadir all were associa ted significantly with biochemical failure (P < 0.001). The median interval s to biochemical failure for patients with pretreatment PSA levels less tha n or equal to 3.9 ng/mL, 4.0-19.9 ng/mL, and greater than or equal to 20.0 ng/mL were 2.2 years, 1.5 years, and 1.2 years, respectively (P < 0.001). T he median intervals to biochemical failure for patients with Gleason scores of 2-4, 5-7, and 8-10 were 1.8 years, 1.5 years, and 1.1 years, respective ly (P < 0.001). Only 6 patients failed beyond 5 years after treatment even though 136 patients were at risk for failure beyond this point. When restri cting analysis to 643 patients (74%) with greater than or equal to 3 years of PSA follow-up, the median nadir level for biochemically controlled patie nts was 0.6 ng/mL and occurred at a median interval of 1.9 years after RT v ersus a median nadir level of 1.3 ng/mL (P = 0.002) occurring at a median i nterval of 1.0 years (P < 0.001) in those patients who experienced biochemi cal failure. Patients were divided into subgroups based on their PSA nadir level and time to nadir. The 5-year actuarial biochemical control rates for patients with nadir values of less than or equal to 0.4 ng/mL, 0.5-0.9 ng/ mL, 1.0-1.9 ng/mL, 2.0-3.9 ng/mL, and greater than or equal to 4.0 ng/mL we re 78%, 60%, 50%, 20%, and 9%, respectively (P < 0.001). The 5-year actuari al biochemical control rates for patients who reached their nadir at < 1.0 years, 1.0-1.9 years, 2.0-2.9 years, and greater than or equal to 3.0 years were 30%, 52%, 64%, and 92%, respectively (P < 0.001). All 52 patients who achieved a nadir of less than or equal to 0.4 ng/mL and required greater t han or equal to 2.0 years to reach this nadir had biochemically controlled disease. CONCLUSIONS. These results suggest that a patient has a high likelihood of biochemical cure after treatment for prostate carcinoma with conventional d oses of external beam RT if he has not demonstrated biochemical failure wit hin 5 years of treatment. Patients with lower pretreatment PSA levels and l ower Gleason scores may require longer follow-up than those with less favor able characteristics to achieve the same certainty of cure. Patients who ac hieve a PSA nadir less than or equal to 0.4 ng/mL and require greater than or equal to 2.0 years to reach this nadir have the highest probability of c ure. (C) 1999 American Cancer Society.