Pretreatment PSA predicts for biochemical disease free survival in patients treated with post-prostatectomy external beam irradiation

Objective: To assess the outcome and determine prognostic factors for patients treated with external beam radiotherapy following radical prostatectomy. Methods and Materials: Forty-four patients were treated after prostatectomy with radiotherapy between March 1988 and October 1993. All patients were free from clinically or radiographically suspicious local or distant disease. One patient underwent neoadjuvant hormonal therapy, but no other patients received hormonal therapy prior to radiation. Pre-radiotherapy PSA and follow-up PSA data were available in all patients. Four patients had undetectable PSA (<0.2) and the remainder had detectable PSA at the time of radiation (median 2.9 range 0-24.9). These two groups were analyzed together. Median follow-up was 48 months for all patients and 48 months for disease-free patients. A median dose of 60Gy (range 58-66) was given to the prostate bed. Forty-three percent of patients were pT2, 52% were pT3, and 5% were pT4; 55% had close (< 1mm) or positive margins, 27% had seminal vesicle invasion, 32% had Gleasons grade >7, and 11% had nodal involvement. Survival was analyzed using the life table method. Actuarial freedom from biochemical (BCM) failure, defined as a rise of greater than 10% or an undetectable PSA becoming detectable, was the primary endpoint studied. Results: Fifty-nine percent of patients had a detectable PSA return to undetectable levels after XRT. The actuarial five year freedom from biochemical failure for all patients was 24%. A significant difference in BCM disease free survival was seen for patients irradiated with a pre-XRT PSA ≤2.7 versus a pre-XRT PSA >2.7 (p=0.0001). Sixty percent of the former group were BCM disease free versus 0% in the latter. Biochemical disease free survival was not affected by preoperative PSA level, presence of undetectable PSA after surgery, surgery to radiation interval, seminal vesicle invasion, clinical stage, pathologic stage, Gleasons grade, or total dose. There were no symptomatic or clinically suspicious local failures, and there were no grade 3, 4, or 5 acute or late complications. There were 69% grade 1 and 2 acute reactions and one grade 2 late complication. Conclusions: Pelvic radiotherapy for patients with a PSA of ≤2.7 after prostatectomy was effective in biochemically controlling 60% of the patients with four years median follow up. To our knowledge these data represent the longest follow-up for this patient population