Preoperative CT-based node-RADS scoring for risk stratification in pancreatic head cancer: diagnostic accuracy and...

To evaluate the diagnostic performance and prognostic value of the Node Reporting and Data System (Node-RADS) for predicting lymph node metastasis and survival outcomes in patients with pancreatic head adenocarcinoma.

This retrospective single-center study included patients with pancreatic head adenocarcinoma undergoing surgical resection. Preoperative contrast-enhanced CT images were evaluated using the Node-RADS classification. Diagnostic performance for lymph node metastasis was assessed by ROC analysis, and associations with overall and disease-free survival were analyzed using Kaplan-Meier and Cox regression methods.

Node-RADS scoring demonstrated moderate discriminatory ability for detecting lymph node metastasis, with an area under the curve of 0.686, sensitivity of 75.0%, and specificity of 62.2%. The negative predictive value (76.7%) was notably higher than the positive predictive value (60.0%). In multivariate analysis, Node-RADS ≥ 3 emerged as an independent predictor of overall survival (hazard ratio: 2.59, p = 0.023), maintaining significance after adjustment for tumor grade, lymphovascular invasion, and pathologic nodal status.

Node-RADS demonstrates moderate diagnostic accuracy for detecting lymph node metastasis in pancreatic head adenocarcinoma but provides independent prognostic information for overall survival beyond conventional clinicopathological factors. Systematic Node-RADS scoring may enhance preoperative risk stratification and facilitate personalized treatment planning in pancreatic cancer patients.