Radical prostatectomy or radiotherapy reduce prostate cancer mortality in elderly patients: a population-based propensity score adjusted analysis

Purpose: Contemporary data regarding the effect of local treatment (LT) vs. non-local treatment (NLT) on cancer-specific mortality (CSM) in elderly men with localized prostate cancer (PCa) are lacking. Hence, we evaluated CSM rates in a large population-based cohort of men with cT1-T2 PCa according to treatment type. Methods: Within the SEER database (2004–2014), we identified 44,381 men ≥ 75 years with cT1-T2 PCa. Radical prostatectomy and radiotherapy patients were matched and the resulting cohort (LT) was subsequently matched with NLT patients. Cumulative incidence and competing risks regression (CRR) tested CSM according to treatment type. Analyses were repeated after Gleason grade group (GGG) stratification: I (3 + 3), II (3 + 4), III (4 + 3), IV (8), and V (9-10). Results: Overall, 4715 (50.0%) and 4715 (50.0%) men, respectively, underwent NLT and LT. Five and 7-year CSM rates for, respectively, NLT vs. LT patients were 3.0 and 5.4% vs. 1.5 and 2.1% for GGG II, 4.5 and 7.2% vs. 2.5 and 2.8% for GGG III, 7.1 and 10.0% vs. 3.5 and 5.1% for GGG IV, and 20.0 and 26.5% vs. 5.4 and 9.3% for GGG V patients. Separate multivariable CRR also showed higher CSM rates in NLT patients with GGG II [hazard ratio (HR) 3.3], GGG III (HR 2.6), GGG IV (HR 2.4) and GGG V (HR 2.6), but not in GGG I patients (p = 0.5). Conclusions: Despite advanced age, LT provides clinically meaningful and statistically significant benefit relative to NLT. Such benefit was exclusively applied to GGG II to V but not to GGG I patients.