Abstract
OBJECTIVE: To develop a preoperative composite biomarker model integrating FAR and SII/CA19-9 for predicting survival in distal cholangiocarcinoma (DCC) after pancreatoduodenectomy. METHODS: This retrospective cohort study analyzed 238 DCC patients (2010-2023). Optimal cut-off were determined by ROC analysis (FAR: 8.85, AUC = 0.602; SII/CA19-9: 8, AUC = 0.668). Intergroup comparisons demonstrated no significant differences between the groups. Survival analysis validated pronounced survival disparities across these groups. Multivariable Cox regression identified independent prognostic factors, and a nomogram was constructed for survival prediction. a nomogram integrated independent risk factors into a predictive model, then calibration curves and decision curve analysis (DCA) collectively validated the model's prognostic capability. RESULTS: The independent Prognostic Factors were FAR > 8.85 (HR = 1.919, 95% CI: 1.333-2.762), SII/CA19-9 ≤ 8 (HR = 0.522, 95% CI: 0.356-0.765), and R1 resection (HR = 0.523, 95% CI: 0.328-0.834). Low SII/CA19-9 (≤ 8) patients had median OS of 17 months vs. 44 months in high-ratio group (P < 0.001). High FAR (> 8.85) correlated with reduced median OS (20 months vs. 51months, P < 0.001). The composite model outperformed AJCC staging (C-index: 0.72 vs. 0.62) and CA19-9 alone (AUC: 0.68 vs. 0.61), with 31% net benefit gain in decision curve analysis. CONCLUSION: The FAR and SII/CA19-9 composite model enhances preoperative prognostication in DCC, stratifying high-risk patients (low SII/CA19-9 and high FAR) for personalized adjuvant strategies.