Differentiating cholangiocyte phenotype hepatocellular carcinoma and assessing early recurrence risk using contrast-enhanced ultrasound features and clinical indicators.

objectiveTo establish a clinical prediction model for cholangiocyte phenotype hepatocellular carcinoma (HCC) using contrast-enhanced ultrasound (CEUS) and clinical indicators; evaluate its diagnostic performance; and assess the risk of early recurrence after hepatectomy.

methodsWe conducted a retrospective analysis of clinical data from 150 patients with HCC who were treated at our institution between July 2022 and December 2023. All patients underwent preoperative CEUS, and immunohistochemical analysis of surgical specimens served as the reference standard for histopathological diagnosis. We systematically evaluated and compared clinical presentations, conventional ultrasound characteristics and CEUS imaging features between cholangiocyte phenotype HCC (n = 79) and classical HCC (n = 71). Using multivariate logistic regression, we identified independent predictors of cholangiocyte phenotype HCC and developed a predictive model. Model performance-including discriminative accuracy and clinical net benefit-was rigorously assessed using receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA). Finally, Kaplan-Meier survival analysis was performed to evaluate the risk of early recurrence in patients with cholangiocyte phenotype HCC.

resultsThis retrospective study enrolled 150 patients with HCC. Independent predictors of cholangiocyte phenotype HCC comprised three ultrasound features-ill-defined tumor margins, heterogeneous arterial-phase enhancement and early washout-as well as three clinical variables: multifocality, serum alpha-fetoprotein (AFP) ≥ 400 ng/ml, and alkaline phosphatase (ALP) ≥ 150 U/L. A predictive model incorporating these six factors yielded an area under the receiver operating characteristic curve (AUC) of 0.891 (95% confidence interval [CI]: 0.841-0.942), with a sensitivity of 73.0% and specificity of 90.0%. Patients with cholangiocyte phenotype HCC exhibited significantly shorter recurrence-free survival (RFS) following hepatectomy, with a median RFS of 29.5 months (95% CI: 20.50-29.50) (p < 0.001). Among molecular subgroups, the CK7-/CK19 + cholangiocyte phenotype HCC cohort showed the poorest RFS, with median RFS of 17.00 months (95% CI: 5.50-22.00).

conclusionCEUS combined with clinical indicators may aid in predicting cholangiocyte phenotype HCC. Patients with cholangiocyte phenotype HCC exhibit a poorer prognosis following hepatectomy, and CK19 positivity may serve as a prognostic indicator in this subtype.