Introduction: Biochemical recurrence (BCR) after radical prostatectomy (RP) or radiation therapy (RT) for prostate cancer (PC) is associated with substantially different cancer-specific mortality (CSM) outcomes, commonly attributed to patient selection bias. We compared CSM after BCR in RP vs. RT patients while accounting for baseline other-cause mortality (OCM) disparities. Methods: We identified patients with BCR after initial RP or RT for PC at a single US health system between 1995 and 2023. We compared CSM using Fine-Gray competing risk regression adjusting for patient and disease characteristics. As a sensitivity analysis, we performed 1:1 OCM risk matching using predicted 5-year OCM to create cohorts with comparable baseline health status, then evaluated CSM differences adjusting for European Association of Urology (EAU) BCR risk classification. Results: We identified 1,047 men with BCR (717 RP, 330 RT). Median follow-up from BCR was 5.5 (2.7–10.5) years. Cumulative incidence of OCM (10-year: 43.5% vs. 13.3%) and CSM (10-year: 28.9% vs. 6.8%) was significantly higher in RT patients compared to RP (Gray’s test p < 0.001 for both). After multivariable adjustment, RT was associated with substantially increased CSM relative hazard (HR 4.95, 95% CI 3.10–7.89, p < 0.001). OCM-matching created a smaller cohort of 248 patients (124 per group) with comparable 5-year OCM (10.8% vs. 12.0%, p = 0.2). In this cohort, RT patients had a 2.18-fold increased CSM relative hazard compared to RP patients (95% CI 1.20–3.96, p = 0.01) even after adjustment for EAU BCR risk. Conclusions: A non-negligible portion of CSM after BCR disparities between RT and RP seems to be related to OCM differences. However, substantial CSM differences persist even after OCM matching, suggesting that BCR definitions after RP and RT are not clinically equivalent for prognostic assessment or therapeutic decision-making.
Cancer-specific mortality differences between radical prostatectomy and radiotherapy after biochemical recurrence: analysis of a health system-wide cohort / C. Silvani, A. Santangelo, A. Stephens, J. Considine, A. Msska, M. Perry, A. Sood, M. Catanzaro, A. Briganti, A. Salonia, F. Montorsi, N. Nicolai, E. Montanari, C. Rogers, F. Abdollah. - In: WORLD JOURNAL OF UROLOGY. - ISSN 1433-8726. - 44:1(2026 Apr), pp. 309.1-309.8. [10.1007/s00345-026-06387-0]
Cancer-specific mortality differences between radical prostatectomy and radiotherapy after biochemical recurrence: analysis of a health system-wide cohort
C. SilvaniPrimo
;E. Montanari;
2026
Abstract
Introduction: Biochemical recurrence (BCR) after radical prostatectomy (RP) or radiation therapy (RT) for prostate cancer (PC) is associated with substantially different cancer-specific mortality (CSM) outcomes, commonly attributed to patient selection bias. We compared CSM after BCR in RP vs. RT patients while accounting for baseline other-cause mortality (OCM) disparities. Methods: We identified patients with BCR after initial RP or RT for PC at a single US health system between 1995 and 2023. We compared CSM using Fine-Gray competing risk regression adjusting for patient and disease characteristics. As a sensitivity analysis, we performed 1:1 OCM risk matching using predicted 5-year OCM to create cohorts with comparable baseline health status, then evaluated CSM differences adjusting for European Association of Urology (EAU) BCR risk classification. Results: We identified 1,047 men with BCR (717 RP, 330 RT). Median follow-up from BCR was 5.5 (2.7–10.5) years. Cumulative incidence of OCM (10-year: 43.5% vs. 13.3%) and CSM (10-year: 28.9% vs. 6.8%) was significantly higher in RT patients compared to RP (Gray’s test p < 0.001 for both). After multivariable adjustment, RT was associated with substantially increased CSM relative hazard (HR 4.95, 95% CI 3.10–7.89, p < 0.001). OCM-matching created a smaller cohort of 248 patients (124 per group) with comparable 5-year OCM (10.8% vs. 12.0%, p = 0.2). In this cohort, RT patients had a 2.18-fold increased CSM relative hazard compared to RP patients (95% CI 1.20–3.96, p = 0.01) even after adjustment for EAU BCR risk. Conclusions: A non-negligible portion of CSM after BCR disparities between RT and RP seems to be related to OCM differences. However, substantial CSM differences persist even after OCM matching, suggesting that BCR definitions after RP and RT are not clinically equivalent for prognostic assessment or therapeutic decision-making.| File | Dimensione | Formato | |
|---|---|---|---|
|
s00345-026-06387-0.pdf
accesso riservato
Tipologia:
Publisher's version/PDF
Licenza:
Nessuna licenza
Dimensione
1.2 MB
Formato
Adobe PDF
|
1.2 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.




