Abstract
Purpose Patients with high-risk prostate cancer after radical prostatectomy are at risk for death. Adjuvant androgen-deprivation therapy (ADT) may reduce this risk. We hypothesized that the addition of mitoxantrone and prednisone (MP) to adjuvant ADT could reduce mortality compared with adjuvant ADT alone. Methods Eligible patients had cT1-3N0 prostate cancer with one or more high-risk factors after radical prostatectomy (Gleason score [GS] ≥ 8; pT3b, pT4, or pN+ disease; GS 7 and positive margins; or preoperative prostate-specific antigen [PSA] > 15 ng/mL, biopsy GS score > 7, or PSA > 10 ng/mL plus biopsy GS > 6. Patients with PSA ≤ 0.2 ng/mL after radical prostatectomy were stratified by pT/N stage, GS, and adjuvant radiation plan and randomly assigned to ADT (bicalutamide and goserelin for 2 years) or ADT plus six cycles of MP. The primary end point was overall survival (OS). Median OS was projected to be 10 years in the ADT arm, requiring 680 patients per arm to detect a hazard ratio of 1.30 with 92% power and one-sided α = .05. Results Nine hundred sixty-one eligible intent-to-treat patients were randomly assigned to ADT or ADT + MP from October 1999 to January 2007, when the Data Safety Monitoring Committee recommended stopping accrual as a result of higher leukemia incidence with ADT + MP. Median follow-up was 11.2 years. The 10-year OS estimates were 87% with ADT (expected 50%) and 86% with ADT + MP (hazard ratio, 1.06; 95% CI, 0.79 to 1.43). The 10-year estimate for disease-free survival was 72% for both arms. Prostate cancer was the cause of death in 18% of patients in the ADT arm and 22% in the ADT + MP arm. More patients in the MP arm died of other cancers (36% v 18% in ADT alone arm). Conclusion MP did not improve OS and increased deaths from other malignancies. The DFS and 10-year OS in these patients treated with 2 years of ADT were encouraging compared with historical estimates, although a definitive conclusion regarding value of ADT may not be made without a nontreatment control arm.