Multicentre Validation of the 2019 Briganti Nomogram: One Threshold Does Not Fit All

The 2019 Briganti nomogram is widely used to guide the indication for pelvic lymph node dissection (LND) at the time of radical prostatectomy in patients with localized prostate cancer. Although previously validated, its generalizability across distinct clinical settings remains uncertain.

We conducted a multicentre external validation of the nomogram in 481 patients from three French academic institutions (Centre A n=198, Centre B n=183 and Centre C n=100). Discrimination, calibration, and clinical utility were assessed. Spared LNDs and missed lymph node invasions (LNIs) were evaluated across risk thresholds.

The overall area under the receiver operating characteristics curve (AUC) was 0.733 but varied across centres (0.580-0.768). Calibration was acceptable overall but showed systematic overestimation in low-prevalence centres. At the 7% recommended threshold, the proportion of spared LNDs ranged from 51% to 76%, while missed LNIs ranged from 0% to 8.9%. Decision curve analysis revealed that the optimal threshold differed between centres.

These results underscore the need for local validation and population-specific threshold adjustment before clinical implementation. Fixed thresholds may lead to under- or overtreatment depending on institutional case mix. Nomogram-based decision-making should be individualized based on local performance and patient-centred risk tolerance.