Risk assessment of metastatic recurrence in patients with prostate cancer by using the Cancer of the Prostate Risk Assessment score: results from 2937 European patients

Abstract

Study Type--Prognosis (case series) Level of Evidence 4. What's known on the subject? and What does the study add? Different tools allow the individual estimation of the various endpoints in patients with prostate cancer. The Cancer of the Prostate Risk Assessment (CAPRA) score is an easy to calculate prediction tool, based on a large population based database. However, little is known about the performance of this prediction tool in European patients. The data obtained in the present study demonstrate differences in tumour characteristics between European patients and the initial development cohort from the USA. However, the concordance index of the CAPRA scores for predicting biochemical recurrence and metastatic recurrence was 76.2 and 78.5, respectively, in European patients. Therefore, the CAPRA score also allows reliable prediction of the examined endpoints in European patients.

Objectives: • To assess the ability of the Cancer of the Prostate Risk Assessment Score (CAPRA) score for predicting biochemical recurrence (BCR) and metastatic recurrence (MR) by using a large cohort of European patients with prostate cancer. • The CAPRA score was initially developed using patients treated in community-based hospitals in the USA and allows a prediction of the risk of different clinical endpoints, without incorporating the surgical margin status.

Patients and methods: • BCR and metastatic recurrence rates were studied in 2937 patients who underwent radical prostatectomy in a tertiary referral centre after a mean (median, range) follow-up of 49 (56, 12-220) months. • The association between the examined endpoints, individual CAPRA scores and pathological features was analyzed by using Kaplan-Meier, proportional hazard and logistic regressions analyses. • Graphical representation assessed the calibration of the CAPRA score for predicting both endpoints.

Results: • Compared to the initial development cohort, worse tumour characteristics and a lower overall positive surgical margin rate (17.2% vs 32.4%) were detected in the European cohort. • Overall, 530 (18.4%) and 58 (1.9%) of patients developed BCR and MR. Increasing CAPRA scores were related to less favourable pathological characteristics and higher BCR and metastatic recurrence rates. • For example, the 5-year BCR and metastatic recurrence rates were markedly different at the extremes of 0-1 vs ≥ 8 (9.2% vs 70.8% and 0.7% vs 16.4%, respectively). • The concordance index for the prediction of BCR and metastatic recurrence was 76.2 and 78.5, respectively.

Conclusions: • Despite differences between the present cohort and the initial development cohort with respect to clinical features and the outcomes achieved, the data obtained in the present study shows the generalizability of the CAPRA score. • Specifically, the data allow the precise identification of those European patients who are at high risk for BCR and MR.