Preoperative Ct and Inflammatory Index–Based Risk Scores for Predicting Short-Term Recurrence Following Resection of Pancreatic Ductal Adenocarcinoma Publisher Pubmed



Mansouri Tehrani M M ; Zamani F ; Iraji H ; Safarnezhad Tameshkel F ; Ghasemi M
Source: BMC Surgery Published:2026

Abstract

Background: Pancreatic ductal adenocarcinoma (PDAC) carries a poor prognosis, with nearly half of resected patients relapsing within one year. Current National Comprehensive Cancer Network (NCCN) and American Joint Committee on Cancer (AJCC) staging systems inadequately reflect tumor biology. This study aimed to develop and validate preoperative risk scores integrating computed tomography (CT) features and systemic inflammatory markers to predict 12-month recurrence. Methods: In this two-center retrospective cohort, consecutive patients with histologically confirmed resectable/borderline PDAC who underwent pancreaticoduodenectomy (2017–2024) were included. Standardized preoperative multiphase CT and laboratory data within 28 days before surgery were reviewed. Quantitative and qualitative CT parameters and inflammatory indices were analyzed using multiple imputation and pathway-specific logistic regression for up-front surgery (US) and neoadjuvant therapy (NT) groups. Model performance was evaluated by discrimination, calibration, and decision-curve analysis. Results: Of 544 resections, 133 met eligibility (US = 81; NT = 52). Final models retained tumor size, necrosis, fat stranding, and the C-reactive protein-to-albumin ratio (CAR) and neutrophil-to-lymphocyte ratio (NLR) for US, and tumor size, hepatic-artery contact ≥ 90°, CAR, NLR, and carcinoembryonic antigen (CEA) for NT. Development area under the curve (AUC) values were 0.84 (US) and 0.81 (NT), and test AUCs were 0.82 and 0.78, respectively, outperforming AJCC 8ᵗʰ staging (ΔAUC + 0.19 and + 0.20). Conclusions: Compact, pathway-specific preoperative scores combining CT descriptors and inflammatory indices improved prediction of early recurrence compared with AJCC staging, supporting their potential use for individualized surgical and perioperative decision-making. © The Author(s) 2026.