15-year cause specific and all-cause survival following brachytherapy for prostate cancer: negative impact of long-term hormonal therapy.

ABSTRACT

PURPOSE: We analyzed factors influencing 15-year cause specific and all-cause survival in men treated with prostate brachytherapy. MATERIALS AND METHODS: A total of 1,669 men with a median age of 66 years who had T1-T3 prostate cancer were treated with prostate brachytherapy and followed a mean of 10 years. Treatments were implant alone, implant plus hormone therapy, or external beam irradiation or implant plus hormone therapy plus external beam irradiation. Hormone therapy was administered in 898 men (53.8%) for a median of 6 months. Cause specific and all-cause survival were estimated by the Kaplan-Meier method with comparisons made by logistic regression and Cox proportions hazard rates. RESULTS: The 15-year cause specific survival rate was 94.1%. Cause specific survival in the 3 NCCN® risk groups was 96.3%, 97.5% and 85.2% (p <0.001). Hormone therapy did not positively impact cause specific survival. The 15-year all-cause survival rate was 57%. Cox regression revealed age (HR 1.09, p <0.001), hormone therapy (HR 1.04, p = 0.032), diabetes (HR 1.86, p = 0.013), atrial fibrillation (HR 2.90, p = 0.041), smoking (HR 1.42, p = 0.030) and emphysema (HR 8.20, p = 0.040) as significant associations. At 15 years hormone therapy decreased all-cause survival from 60.3% to 54.9% (p = 0.009). All-cause survival was not reduced when hormone therapy was limited to 6 months or less (p = 0.005). This difference was present in men 66 years old or younger (p = 0.017) and in older men (p = 0.05). CONCLUSIONS: Prostate brachytherapy yields favorable 15-year cause specific survival, especially in patients at high risk. All-cause survival is less in patients with preexisting diabetes, atrial fibrillation and emphysema. Hormone therapy for longer than 6 months has a negative effect on all-cause survival even in younger patients without an apparent beneficial effect on cause specific survival.