P K Chiu1, M J Roobol2, D Nieboer3, J Y Teoh1, S K Yuen4, S M Hou1, M K Yiu4, C F Ng1,5. 1. Division of Urology, Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, Hong Kong. 2. Department of Urology, Erasmus University Medical Centre, Rotterdam, The Netherlands. 3. Department of Public Health, Erasmus University Medical Centre, Rotterdam, The Netherlands. 4. Division of Urology, Department of Surgery, Queen Mary Hospital, University of Hong Kong, Hong Kong, Hong Kong. 5. Department of Surgery, SH Ho Urology Centre, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong, Hong Kong.
Abstract
BACKGROUND: To adapt the well-performing European Randomized Study of Screening for Prostate Cancer (ERSPC) risk calculator to the Chinese setting and perform an external validation. METHODS: The original ERSPC risk calculator 3 (RC3) for prostate cancer (PCa) and high-grade PCa (HGPCa) was applied to a development cohort of 3006 previously unscreened Hong Kong Chinese men with initial transrectal biopsies performed from 1997 to 2015, age 50-80 years, PSA 0.4-50 ng ml-1 and prostate volume 10-150 ml. A simple adaptation to RC3 was performed and externally validated in a cohort of 2214 Chinese men from another Hong Kong hospital. The performance of the models were presented in calibration plots, area under curve (AUC) of receiver operating characteristics (ROCs) and decision curve analyses. RESULTS: PCa and HGPCa was diagnosed in 16.7% (503/3006) and 7.8% (234/3006) men in the development cohort, and 20.2% (447/2204) and 9.7% (214/2204) men in the validation cohort, respectively. The AUCs using the original RC3 model in the development cohort were 0.75 and 0.84 for PCa and HGPCa, respectively, but the calibration plots showed considerable overestimation. In the external validation of the recalibrated RC3 model, excellent calibration was observed, and discrimination was good with AUCs of 0.76 and 0.85 for PCa and HGPCa, respectively. Decision curve analyses in the validation cohort showed net clinical benefit of the recalibrated RC3 model over PSA. CONCLUSIONS: A recalibrated ERSPC risk calculator for the Chinese population was developed, and it showed excellent discrimination, calibration and net clinical benefit in an external validation cohort.
BACKGROUND: To adapt the well-performing European Randomized Study of Screening for Prostate Cancer (ERSPC) risk calculator to the Chinese setting and perform an external validation. METHODS: The original ERSPC risk calculator 3 (RC3) for prostate cancer (PCa) and high-grade PCa (HGPCa) was applied to a development cohort of 3006 previously unscreened Hong Kong Chinese men with initial transrectal biopsies performed from 1997 to 2015, age 50-80 years, PSA 0.4-50 ng ml-1 and prostate volume 10-150 ml. A simple adaptation to RC3 was performed and externally validated in a cohort of 2214 Chinese men from another Hong Kong hospital. The performance of the models were presented in calibration plots, area under curve (AUC) of receiver operating characteristics (ROCs) and decision curve analyses. RESULTS: PCa and HGPCa was diagnosed in 16.7% (503/3006) and 7.8% (234/3006) men in the development cohort, and 20.2% (447/2204) and 9.7% (214/2204) men in the validation cohort, respectively. The AUCs using the original RC3 model in the development cohort were 0.75 and 0.84 for PCa and HGPCa, respectively, but the calibration plots showed considerable overestimation. In the external validation of the recalibrated RC3 model, excellent calibration was observed, and discrimination was good with AUCs of 0.76 and 0.85 for PCa and HGPCa, respectively. Decision curve analyses in the validation cohort showed net clinical benefit of the recalibrated RC3 model over PSA. CONCLUSIONS: A recalibrated ERSPC risk calculator for the Chinese population was developed, and it showed excellent discrimination, calibration and net clinical benefit in an external validation cohort.
Authors: Marinus J Hagens; Piter J Stelwagen; Hans Veerman; Sybren P Rynja; Martijn Smeenge; Vincent van der Noort; Ton A Roeleveld; Jolien van Kesteren; Sebastiaan Remmers; Monique J Roobol; Pim J van Leeuwen; Henk G van der Poel Journal: World J Urol Date: 2022-10-16 Impact factor: 3.661