An Analysis of Patients with Clinically Localized High‐Risk Prostate Carcinoma

Objectives: The objectives of this study are the following: (1) to determine the outcome of patients with at least two unfavorable prognostic factors, as defined in the literature (tumor Stage ≤T2c, pretreatment PSA level >10 ng/ml, Gleason score ≥7); and (2) to define the impact of conformal radiotherapy (CRT), whole pelvic radiation, and hormonal therapy in the treatment of these patients. Materials and Methods: Between January 1, 1987, and December 31, 1995, 594 evaluable patients were treated with definitive radiotherapy for localized prostate carcinoma at the University of California, San Francisco and associated institutions. One hundred eighty-two patients had clinically localized high-risk prostate carcinomas defined as having at least two of the following adverse risk features: (1) tumor Stage ≥T2c; (2) pretreatment PSA level >10 ng/ml; and (3) Gleason score ≥7. One hundred sixty-four patients had >12 months of PSA follow-up and formed the cohort of this study. Fifty-eight percent of the patients had pretreatment PSA levels >20 ng/ml, 31% had Gleason scores of 8–10, and 60% had Stage T3 disease. Radiotherapy was delivered at 1.8 Gy/fraction/day, 5 days/week. The maximum tumor dose ranged from 60 to 82.4 Gy (median 73.7 Gy). Sixty-two percent of the group had elective whole-pelvic radiotherapy (WPRT), and 34% had androgen suppression therapy (AST). The median PSA follow-up was 39 months. PSA failure was defined by the consensus definition of the American Society for Therapeutic Radiotherapy and Oncology. Results: The 4-year estimate of biochemical freedom from relapse of the 164 patients with clinically localized high-risk prostate cancer was 39%. The median time to PSA failure was 18 months. The 4-year estimate of PSA control was 51% for patients with two adverse risk factors and 16% for those with three adverse risk factors. On univariate analysis, the number of adverse risk factors (p = 0.004) and WPRTs (p = 0.04) were significant prognostic factors for PSA control. The use of CRT (p = 0.08) and AST (p = 0.10) were of borderline significance. On multivariate analysis, the most significant independent prognostic factor for PSA control was the number of risk factors present (favoring two factors, p = 0.002). Treatment with WPRT (p = 0.03) was the next independent predictor. AST was of borderline significance (p = 0.10). Eight percent of patients (14 of 164) had Grade 1–2 cystitis and proctitis. There was no Grade 3–4 toxicity. Conclusions: The combination of pretreatment PSA level, Gleason score, and disease stage could reliably predict the prognosis of patients with localized high-risk prostate carcinoma treated with definitive radiotherapy. The use of prophylactic WPRT improved PSA control in patients with clinical Stage N0 disease who are at high-risk for nodal involvement. Patients with three adverse risk factors (PSA level >10 ng/ml, Gleason Score ≥7, and tumor Stage ≥2c) had a very poor prognosis when treated with conventional therapy and should be considered for novel effective systemic treatment.