Bellmunt Risk Score as a Prognostic Tool in Metastatic Castration-Resistant Prostate Cancer Survival.

Abstract

Metastatic castration-resistant prostate cancer (mCRPC) is an incurable disease with a highly variable clinical course, necessitating robust prognostic tools to guide individualized patient management. To validate the prognostic ability of the Bellmunt Risk Score (BRS) for predicting survival in patients with mCRPC. For this prognostic study, a post hoc analysis was conducted of 2 international, multicenter, phase 3 randomized clinical trials of men (aged ≥18 years) with mCRPC (ACIS [first-line setting] and ELM-PC-5 [postdocetaxel setting]). The ACIS trial enrolled patients between December 2014 and August 2016. The ELM-PC-5 trial enrolled patients between October 2010 and February 2013. Data analyses were conducted between September 2024 and March 2025. Baseline BRS (range, 0-3), calculated by assigning 1 point for each of 3 factors (Eastern Cooperative Oncology Group Performance Status score ≥1, hemoglobin level <10 g/dL, and liver metastases). The primary outcomes, overall survival (OS) and radiographic progression-free survival, were assessed via Kaplan-Meier and multivariable Cox proportional hazards regression analysis. This analysis included 678 evaluable male participants from the ACIS trial (median age, 71 years [range, 48-92 years]) and 1078 from the ELM-PC-5 trial (median age, 70 years [range, 43-89 years]), for a total of 1756 participants. The median follow-up was 54.8 months (IQR, 51.5-58.4 months) in the ACIS trial and 10.7 months (IQR, 0.4-27.1 months) in the ELM-PC-5 trial. Having a higher BRS was consistently associated with unfavorable OS in both cohorts. In the ACIS trial, the median OS decreased from 42.2 months (95% CI, 35.7-46.7 months) for patients with a BRS of 0 to 9.1 months (95% CI, 3.7 months to not reached) for those with a BRS of 3. Adjusted hazard ratios (AHRs) for OS were 1.37 (95% CI, 1.12-1.67) for patients with a BRS of 1, 2.64 (95% CI, 1.89-3.69) for a BRS of 2, and 8.29 (95% CI, 2.57-26.78) for a BRS of 3 compared with those with a BRS of 0. In the ELM-PC-5 trial, the median OS decreased from 23.0 months (95% CI, 21.50 months to not reached) for patients with a BRS of 0 to 3.2 months (95% CI, 1.4-8.6 months) for those with a BRS of 3. AHRs for OS were 1.65 (95% CI, 1.31-2.07) for patients with a BRS of 1, 2.93 (95% CI, 2.22-3.87) for those with a BRS of 2, and 4.43 (95% CI, 2.65-7.40) for those with a BRS of 3 compared with those with a BRS of 0. The BRS remained a robust and independent prognostic factor for both outcomes in all multivariable models. In this prognostic study of patients with mCRPC, the BRS was validated as a prognostic tool that provides clinically meaningful information across different treatment lines. Due to its simplicity, the BRS is a practical aid for guiding treatment decisions and facilitating prognostic discussions in routine clinical care.