**Background:** Intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver tumor, and surgery remains the best treatment option. Identifying preoperative predictors of prognosis is critical for patient stratification and management. This study investigated the association between preoperative serological and clinical indicators and postoperative outcomes in patients undergoing resection for ICC.
**Methods:** This retrospective study included 151 patients (62.4% male, median age 63 years) who underwent surgery for ICC at Qingdao University Affiliated Hospital from January 2015 to May 2021. Inclusion criteria were age >18 years, confirmed ICC diagnosis, and available preoperative serological indicators. Exclusion criteria included prior anticancer therapies, liver transplantation, unavailable data, and perioperative death. Data collected included demographics, comorbidities, preoperative blood tests (including tumor markers, liver function, inflammatory/nutritional indices), surgical details, and pathological factors. The primary outcome was overall survival (OS; time from surgery to ICC-related death), and the secondary outcome was tumor-free survival (TFS; time to recurrence). ROC curves determined optimal cutoff values for serological ratio indicators. Univariate analysis used Kaplan-Meier and log-rank tests; multivariate analysis used Cox proportional hazards regression. Statistical significance was set at P < .05.
**Key Results:** Of 151 patients, 124 (83.8%) underwent radical resection. Mean follow-up was 26 months; 99 patients developed recurrence and 91 died. The 0.5-, 1-, and 3-year OS rates were 80.1%, 63.7%, and 30.1%, respectively; TFS rates were 65.4%, 48.9%, and 12.8%. Univariate analysis showed that T-stage, N-stage, M-stage, vascular invasion, hemoglobin, CEA, CA125, CA199, direct bilirubin, alkaline phosphatase, albumin, PT, NLR, PARI, AAPR, AGR, PNI, and incisal margin were significantly associated with both OS and TFS. CA242 (P = .049) and GGT (P = .042) were associated only with OS, while history of biliary surgery (P = .043) was associated only with TFS. Multivariate analysis for OS identified four independent predictors: PT (P = .005; HR 0.301, 95% CI 0.13–0.7), vascular invasion (P = .01; HR 3.02, 95% CI 1.307–6.982), N-stage (P = .03; HR 0.481, 95% CI 0.248–0.93), and incisal margin (P = .035; HR 1.488, 95% CI 1.028–2.153). For TFS, independent predictors were PT (P = .005; HR 0.309, 95% CI 0.136–0.703), vascular invasion (P = .044; HR 2.22, 95% CI 1.021–4.829), N-stage (P = .048; HR 0.507, 95% CI 0.258–0.995), and CEA (P = .05; HR 0.511, 95% CI 0.261–1).
**Clinical Implications:** Preoperative PT, vascular invasion, and N-stage are independent predictors of both OS and TFS in ICC patients after hepatectomy. Incisal margin status independently predicts OS, while CEA level independently predicts TFS. These readily available clinical and laboratory markers may assist clinicians in preoperative risk stratification, postoperative monitoring, and treatment planning. The study also confirms the prognostic relevance of several inflammatory and nutritional indices (NLR, PNI, AAPR, AGR) at the univariate level, though they did not remain independent predictors in multivariate analysis. Limitations include the single-center retrospective design, small sample size (n=151), and lack of a matched control group, warranting larger prospective studies.