Probabilistic mapping of lymph node metastasis in epithelial ovarian cancer: a retrospective cohort study using...

To characterize the anatomical distribution, potential metastatic pathways, and clinical factors associated with lymph node metastasis (LNM) in epithelial ovarian cancer (EOC), providing evidence to guide individualized surgical management.

A retrospective analysis was performed on 654 EOC patients who underwent radical surgery. Metastatic lymph nodes were categorized into 14 anatomical regions, with LNM rates calculated and distribution maps generated. Bootstrap-derived Bayesian methods were used to construct a metastatic pathway model. Univariate and multivariate regression analyses were conducted to identify risk factors for LNM, and Bayesian conditional probability was applied to analyze the association between these factors and LNM extent.

LNM was identified in 291 of 654 patients (44.50%), with the highest involvement in the left para-aortic region (54.3%) and the lowest in the right superficial inguinal region (10.9%). The Bayesian model revealed three core metastatic pathways: 1) internal/external iliac lymph nodes → common iliac lymph nodes; 2) primary tumor → para-aortic lymph nodes; 3) obturator lymph nodes → inguinal/presacral lymph nodes. Incorporating clinical variables into the Bayesian framework highlighted key determinants of metastatic extent: tumor size, histological grade, and patient age. This integrated analysis quantified node-specific metastasis probabilities and identified high-risk lymphatic routes, providing a patient-specific roadmap for targeted lymphadenectomy and precision surgical planning.

This study provides a probabilistic characterization of lymph node involvement patterns in EOC based on retrospective observational data. The findings may support hypothesis generation for risk-adapted nodal assessment and warrant further validation in prospective studies.