External validation of the prostascore model in patients with metastatic hormone-sensitive prostate cancer recruited to the CHAARTED study.

Abstract

To externally validate 'prostascore' in patients with metastatic hormone-sensitive prostate cancer recruited to the phase III CHAARTED study. We conducted a retrospective analysis of the prospectively collected data from patients with metastatic hormone-sensitive prostate cancer in the CHAARTED study, a phase III multicentre study conducted between 2006 and 2014. The main outcome of the present analysis was overall survival, assessed using Kaplan-Meier analysis or log-rank testing, in the whole cohort according to different prostascores. In addition, patients with different scores were compared according to treatment arm. A total of 702 cases had complete baseline data, allowing calculation of prostascores and inclusion in the present analysis. Overall survival was assessed according to prostascores in the entire cohort and the P value for overall survival trend was significant (P < 0.001). Likewise, progression-free survival was assessed according to prostascores in the entire cohort and the P value for progression-free survival trend was also significant (P < 0.001). Overall survival comparisons according to treatment arm were evaluated among different prostascores. Notably, the P value for overall survival difference was not significant for a prostascore = 2 (P = 0.702), but was significant for scores of 3, 4 and 5 (P < 0.05). The cause-specific hazard ratio for cancer-specific survival (adjusted for treatment arm used) was also evaluated. The P value for pairwise comparisons between different scores was significant (P < 0.01) except for the comparison between scores 4 and 5. The present study further confirms the role of prostascore in predicting the outcomes of patients with metastatic hormone-sensitive prostate cancer and also highlights its potential role in therapeutic decision-making.