Real world prevalence of pelvic lymph node involvement in prostate cancer in Asia: do we need a rethink on normograms?

TL;DR

This study evaluates how well a prostate cancer lymph node risk calculator works in Asian patients, finding it moderately effective but suggesting possible improvements for diverse populations.

Contribution

The study provides new insights into the performance of the Briganti 2019 nomogram in an Asian population, highlighting potential limitations and the need for recalibration.

Findings

The Briganti 2019 nomogram had a moderate discriminatory ability (AUC of 0.77) in predicting lymph node involvement in an Asian cohort.

Independent predictors of lymph node involvement included percentage of positive biopsy cores and extraprostatic extension on MRI.

Abstract

The Briganti 2019 nomogram stratifies risk of lymph node involvement (LNI) in prostate cancer, reducing unnecessary pelvic lymph node dissection (PLND) during radical prostatectomy (RP). However the applicability of the nomogram in diverse populations remains under-explored, with only one external validation study performed in an Asian population to date. We aim to evaluate the performance of the nomogram in a large tertiary Asian institution.

A retrospective cohort study was conducted, with analysis of the cancer registry in our tertiary institution of all patients who underwent RP with PLND between 1988 and 2023. The Briganti 2019 nomogram score was retrospectively calculated for each patient, and post-operative data was analyzed to determine rates of LNI in order to determine the performance of the nomogram in our cohort.

437 patients were included, with a median Briganti score of 11.2% (IQR 3.9–28.5%). The mean number of lymph nodes excised per patient was 15.1±12. 292 (66.8%) patients had a Briganti score greater than 7%, but only 8.6% were noted to harbor pN1 disease after RP. In our Asian cohort, the 2019 Briganti nomogram only had a moderate discriminatory ability with an area under the receiver operating characteristic curve (AUC) of 0.77. On multivariate analysis, independent predictors of LNI in our population included percentage of positive biopsy cores [Odds Ratio (OR) 1.02, 95%CI 1.01–1.04, p=0.01] and extraprostatic extension on MRI prostate (OR 3.00, 95%CI 1.20–7.56, p=0.02).

The Briganti 2019 nomogram, while effective in many settings, only had a moderate ability to identify patients with pN1 disease in our Asian cohort. With potential limitations in its generalizability to multiple populations, a re-evaluation of its thresholds and further calibration to other populations might be required.

Figures