RESUMO
INTRODUCTION: Serum prostate specific antigen (PSA) is a well-known prognostic parameter in men with prostate cancer. The treatment of men with very high PSA values and apparently no detectable metastases is not fully established. PATIENTS AND METHODS: Ancillary analysis from the GETUG 12 phase 3 trial. Patients with non-metastatic high-risk prostate cancer by bone and computerized tomography (CT) scan were randomly assigned to receive androgen deprivation therapy (ADT) and docetaxel plus estramustine or ADT alone. Relapse-free survival (RFS), clinical RFS, metastases-free survival (MFS), overall survival (OS), and prostate cancer-specific survival (PCSS) were estimated using the Kaplan-Meier method for different levels of PSA (50 ng/mL, 75 ng/mL, and 100 ng/mL). The relationship between PSA and outcomes was studied using residual-based approaches and spline functions. RESULTS: The median follow-up was 12 years (range: 0-15.3). Baseline PSA (<50 ng/mL, n = 328; ≥50ng/mL, n = 85) was associated with improved RFS (P = .0005), cRFS (P = .0024), and MFS (P = .0068). The 12-year RFS rate was 46.33% (CI 40.59-51.86), 33.59% (CI 22.55-44.97), and 11.76% (1.96-31.20) in men with PSA values <50 ng/mL (n = 328), 50-100 ng/mL (n = 68), and ≥100 ng/mL (n = 17), respectively. Exploratory analyses revealed no deviation from the linear relationship assumption between PSA and the log hazard of events. CONCLUSIONS: Men with apparently localized prostate cancer and a high baseline PSA value have a reasonable chance of being long-term disease-free when treated with curative intent combining systemic and local therapy.