Preoperative prediction of microvascular invasion in intrahepatic cholangiocarcinoma and its prognostic implications: A multicenter study

Previous studies on preoperative predictors of microvascular invasion (MVI) in intrahepatic cholangiocarcinoma (ICCA) have presented inconsistent results. This study aimed to identify preoperative clinical and magnetic resonance imaging (MRI) factors that can predict MVI in ICCA and to evaluate their prognostic utility using a multicenter cohort. A multicenter cohort of 446 patients who underwent preoperative MRI and surgical resection for ICCA at six tertiary referral institutions between 2009 and 2018 was analyzed for clinical, pathologic, and MR imaging characteristics. Univariable and multivariable logistic regression analyses were performed to identify significant predictors of pathologically confirmed MVI, which were subsequently used to stratify patients into low-, intermediate-, and high-risk groups based on the number of predictors identified. Kaplan-Meier survival analysis and log-rank test were conducted to assess long-term survival and early recurrence among the three groups. Among the 446 patients (mean age, 63.0 ± 9.9 years; 277 men), 234 (52.5%) had MVI on pathology. Independent predictors of MVI included serum carbohydrate antigen 19-9 levels (≥80 U/mL; odds ratio [OR]: 2.68, 95% confidence interval [CI]: 1.64-4.37, p < 0.001), tumor size (≥4 cm; OR: 1.60, 95% CI: 1.01-2.54, p = 0.046), tumor multiplicity (OR: 2.84, 95% CI: 1.58-5.12, p < 0.001), and arterial phase peritumoral enhancement (OR: 2.83, 95% CI: 1.81-4.42, p < 0.001). Stratifying patients by MVI risk - low (no predictors), intermediate (1-3 predictors), and high (4 predictors) - revealed a significant decrease in both recurrence-free and overall survival rates (p < 0.001), along with a corresponding increase in early recurrence rates (p < 0.001) as the risk level increased. Risk stratification utilizing four key predictors can effectively assess the risk of MVI in ICCA and is associated with postoperative outcomes.