Hilar cholangiocarcinoma (HCCA) has a high propensity for early recurrence. We evaluated the prognostic value of preoperative inflammatory biomarkers, Pan‑Immunological Value (PIV), Platelet‑Lymphocyte Ratio (PLR), Lymphocyte‑Monocyte Ratio (LMR), Neutrophil‑Platelet Ratio (NPR), and Neutrophil‑Lymphocyte Ratio (NLR), in resected HCCA. A retrospective cohort study included 373 HCCA patients undergoing radical resection from January 2017 to May 2022. Patients treated between January 2017 and December 2020 were assigned to the training set (n = 236) for model development. Those treated between January 2021 and May 2022 were assigned to the independent temporal validation set (n = 137), this split approximates a 6:4 ratio. Baseline characteristics were comparable between sets (all P > 0.05). Patients were categorized based on one-year recurrence-free survival (RFS) into early and non-early recurrence groups. Clinical, pathological, and preoperative inflammatory marker data were collected. Statistical analyses included univariate and multivariate Cox proportional hazards regression, Kaplan-Meier curves, and receiver operating characteristic (ROC) curve analysis to assess the prognostic value of the inflammatory markers. In univariate analysis, higher PIV, PLR, NLR, NPR and lower LMR were associated with early recurrence (all P < 0.05). To avoid statistical collinearity, separate multivariable Cox models were constructed, each adjusting for clinicopathological factors but including only one inflammatory marker. In these models, PLR (HR = 1.026, P = 0.026), NLR (HR = 1.523, P = 0.011), NPR (HR = 1.831, P = 0.010), and LMR (HR = 0.483, P = 0.001) remained independent predictors of early recurrence; PIV did not (HR = 1.001, P = 0.425). Kaplan‑Meier curves confirmed shorter RFS for patients with high PIV, PLR, NLR, NPR or low LMR (all log‑rank P < 0.001). The combined model integrating all five markers achieved AUCs of 0.857 (training) and 0.908 (validation), outperforming individual markers (AUCs 0.616–0.637). The combined predictive model integrating inflammatory markers PIV, PLR, NLR, NPR, and LMR shows improved prognostic accuracy in HCCA, potentially providing clinically actionable stratification alongside traditional clinical parameters.
Guo et al. (Mon,) studied this question.