Abstract: Objective To investigate the value of preoperative CT signs combined with clinical indicators in predicting the expression of cytokeratin 19 (CK19) in patients with hepatocellular carcinoma (HCC). Methods HCC patients admitted to Guangxi Medical University Cancer Hospital from January 2019 to December 2021 were enrolled as the studied objects. The clinical data and CT signs were collected and randomly divided into training group and validation group according to a ratio of 7: 3. The independent predictors of CK19 expression was analyzed by Logistic backward stepwise regression and a predictive model was constructed. The receiver operating characteristic (ROC) curve, calibration curve, and decision curve were used to evaluate the predictive efficacy of the model. Results A total of 380 HCC patients were enrolled, of which 68 (17.9%) were CK19 positive. Age, AFP, rim arterial phase hyperenhancement, delayed central enhancement, and intratumoral artery were independent predictors of CK19 expression in HCC (all P<0.05). The area under the curve (AUC) of CK19 expression predicted by the clinical imaging model in the training group and validation group was 0.789 and 0.748, respectively. The calibration curve showed a good fitting degree of the clinical imaging model. The decision curve showed that the clinical image model had good clinical application value. Conclusions Preoperative CT signs combined with age and AFP effectively predict CK19 expression in HCC patients.

Key words: Hepatocellular carcinoma, CT signs,  Cytokeratin 19