Model construction and validation of dual-energy CT multi-parameters combined with clinical indicators for predicting lymph node metastasis in pancreatic ductal adenocarcinoma

Abstract

Lymph node metastasis (LNM) is a critical prognostic factor in pancreatic ductal adenocarcinoma (PDAC), but preoperative prediction remains challenging. This study aimed to evaluate the predictive value of combining dual-energy CT (DECT) multiparameters with clinical indicators for LNM in PDAC, and to develop a preoperative model to support clinical decision-making. A total of 126 pathologically confirmed PDAC patients were retrospectively enrolled and divided into training (n=87) and test (n=39) cohorts. Univariate and multivariate Logistic regression were used to identify independent predictors of LNM and construct a combined model. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). Multivariate logistic regression identified CA19-9, intratumoral necrosis, and the slope of the energy attenuation curve (λ) in venous phase (λ_VP) as independent predictors of LNM. The combined model yielded an AUC of 0.816 (95% CI: 0.730–0.874) in the training cohort and 0.761 (95% CI: 0.613–0.909) in the test cohort, with satisfactory calibration (Hosmer-Lemeshow test: P  = 0.722 and P  = 0.604, respectively). DCA confirmed superior net benefit within threshold probabilities of 0–0.78 (training cohort) and 0–0.80 (test cohort). The combined model integrating DECT multiparameters and clinical indicators can effectively predict LNM in PDAC, with favourable diagnostic performance and promising clinical application value.