External Validation of Existing Nomograms on Indian Prostate Biopsy Data
Related Experiment Video
Updated: Nov 4, 2026
Enhancing Prostate Tumor Biobanking Reliability with Improved Sampling Technique and Histological Characterization
Published on: November 17, 2023
Mehul Gupta1, Kevin Arulraj1, Rajeev Kumar1
1Department of Urology, All India Institute of Medical Sciences, New Delhi, India.
Objectives:
Risk calculators (RC) are predictive models that individualize the risk of prostate cancer (PCa). The current study aimed to validate existing RC for predicting detection of any PCa and clinically significant prostate cancer (csPCa) on prostate biopsy in the Indian population.
Methods:
Men who underwent transrectal ultrasound-guided prostate biopsy (January 2016-July 2023) at our tertiary care centre were studied. Six tools were evaluated (SWOP-RC, Barcelona-RC, PLUM-RC, PCRC-MRI, Japanese Nomogram, Korean Nomogram) for probability of detection of any PCa and csPCa.
Results:
The study included 602 patients. PCa was detected in 326 patients (54.2%), including csPCa in 248 (41.2%). For predicting any PCa, both PLUM-RC and SWOP-RC showed an AUC of 0.92 for patients with PSA 10-20 ng/mL and PIRADS-3 on MRI. For predicting csPCa, no single RC demonstrated highest AUCs across all PSA strata. On decision curve analysis, the largest net benefit was seen for individuals with PSA 10-20 ng/mL and PIRADS-3 lesions. Within this criterion, SWOP-RC, PCRC-MRI and Japanese nomogram detected > 10 additional cancers per 100 men while avoiding 40%-55% of biopsies at threshold probabilities ≥ 20%. For most other subgroups, a biopsy-all strategy yielded net benefits similar to the RC.
Conclusions:
Current Rc demonstrate good applicability (in terms of calibration analysis and discriminatory ability) for prostate cancer prediction in Indian men. The observed net benefit is greatest for individuals with PSA 10-20 ng/mL and PIRADS-3 lesions, and is shown by SWOP-RC for PCa and PCRC-MRI and the Japanese nomogram for csPCa.
