Cancer-specific mortality after radical prostatectomy versus radiotherapy in incidental prostate cancer
Francesco Di Bello, Lukas Scheipner, Andrea Baudo, Mario de Angelis, Letizia Maria Ippolita Jannello, Carolin Siech, Zhe Tian, Kira Vitucci, Jordan A. Goyal, Claudia Collà Ruvolo, Gianluigi Califano, Massimiliano Creta, Simone Morra, Pietro Acquati, Fred Saad, Shahrokh F. Shariat, Luca Carmignani, Ottavio De Cobelli, Sascha Ahyai, Alberto Briganti, Felix K.‐H. Chun, Nicola Longo, Pierre I. Karakiewicz
Abstract
• Incidental prostate cancer (IPCa) is a rare entity and with a not well-known natural history. When active treatment is contemplated, IPCA patients undergo radical prostatectomy (RP) or radiotherapy (RT). • In actively treated incidental prostate cancer (IPCa) patients, radical prostatectomy (RP) is associated with a protective hazard ratio (HR = 0.35, P = .01) compared to radiotherapy (RT). • The magnitude of protective effect recorded for RP over RT applied only in GS 8-10 patients (HR = 0.31, P = .039). To test for cancer specific mortality (CSM) differences after either radical prostatectomy (RP) or radiotherapy (RT) in incidental prostate cancer (IPCa) patients. Within the Surveillance, Epidemiology, and End Results (SEER) database (2004–2015), IPCa patients were identified. Cumulative incidence plots as well as competing risks regression (CRR) models were fitted to address CSM after adjustment for other-cause mortality (OCM). Furthermore, a subgroup analysis was performed to test for CSM differences between RP and RT according to Gleason sum (GS 6,7, and 8–10). Of 1,466 IPCa patients, 770 (53%) underwent RP vs. 696 (47%) RT. Incidental PCa RT patients were older, and exhibited higher PSA, higher proportion of Gleason sum 8–10, and higher clinical T stage. In cumulative incidence plots, 5-year CSM rates adjusted for OCM were 0.9 for RP vs. 6.8% for RT (Δ = 5.9%). After multivariable adjustment for clinical characteristics (age, PSA, Gleason sum, and clinical T stage) as well as for OCM, RP was associated with a protective hazard ratio (HR) of 0.35 (95% confidence interval [CI] 0.15 – 0.78, p value = .01). Within Gleason sum 8–10 IPCA patients, RP was associated with a protective HR of 0.31 ( P = .039). Incidental PCa RT-treated patients exhibited less favorable clinical characteristics than their RP counterparts. Despite full adjustment, RP was associated with a protective effect relative to RT. This effect exclusively applied to the Gleason sum 8–10 subgroup. In consequence, IPCa patients harboring Gleason sum 8–10 should ideally be considered for RP instead of RT.