External validation of Stockholm3 in a retrospective German clinical cohort

Abstract

Background and objective: Stockholm3 is a comprehensive blood test amalgamating protein biomarkers, genetic indicators, and clinical data to predict clinically significant prostate cancer risk (International Society of Urological Pathology grade ≥2 upon biopsy). Our study aims to externally validate Stockholm3 and compare its performance with the use of prostate-specific antigen (PSA) and the Rotterdam Prostate Cancer Risk Calculator (RPCRC) for clinically significant prostate cancer detection. Methods: We gathered data from men subjected to prostate biopsies at the Martini-Klinik, Germany, between 2014 and 2017. Participants were selected based on elevated PSA levels or suspicious digital rectal examinations, all undergoing a 10–12-core systematic biopsy without a magnetic resonance imaging–targeted biopsy. We assessed Stockholm3 and RPCRC performance for clinically significant prostate cancer detection. Furthermore, we compared the proportion of men recommended for biopsy and biopsy outcomes with Stockholm3 and RPCRC against PSA ≥3 ng/ml. Key findings and limitations: Our study encompassed 405 biopsied men, with a median age of 66 yr (interquartile range [IQR]: 60–72), PSA levels at 7 ng/ml (IQR: 5.2–10.8), and Stockholm3 scores at 18 (IQR: 10–34). Among them, 128 men (31%) received clinically significant prostate cancer diagnoses. Employing the recommended Stockholm3 threshold (≥15) could have reduced unnecessary biopsies by 52%, while detecting 92% of clinically significant cases compared with using PSA ≥3 ng/ml as a biopsy criterion. Both Stockholm3 and RPCRC exhibited strong discrimination, with area under the curve values of 0.80 (95% confidence interval [CI]: 0.76–0.85) and 0.75 (95% CI: 0.70–0.80), respectively. Stockholm3 demonstrated good calibration, while RPCRC underestimated the risk compared with observed outcomes. Moreover, Stockholm3 yielded positive clinical net benefits, whereas RPCRC yielded negative net benefits for clinically relevant thresholds. Conclusions and clinical implications: Stockholm3 utilization could detect 92% of clinically significant prostate cancer cases while simultaneously reducing unnecessary biopsies by 52%, compared with the PSA ≥3 ng/ml criterion, based on our analysis within a cohort of men who underwent systematic biopsies. Patient summary: In a German clinical cohort of 405 men, Stockholm3, a blood test for early prostate cancer detection, exhibited favorable clinical benefits. It identified a substantial number of clinically significant cases while reducing unnecessary biopsies by over half in men without the disease and those with clinically nonsignificant prostate cancer. © 2024