Development and validation of a nomogram to predict Ki67 expression in hepatocellular carcinoma based on preoperative contrast-enhanced CT features and clinical factors

To develop a nomogram based on contrast-enhanced CT (CECT) image features and clinical factors for preoperatively predicting the expression level of Ki67 in patients with hepatocellular carcinoma (HCC).

One hundred eighty-three patients diagnosed with HCC were included in this study. All patients underwent CECT scans before surgeries or biopsies. The Ki67 expression was assessed by immunohistochemistry. The nomogram was constructed based on a combination of CECT image features and clinical factors which showed an independent association with Ki67 expression. The area under the receiver operating characteristic curve (AUC), calibration curve and decision curve analysis (DCA) were used to verify the performance of the nomogram.

In multivariate logistic regression, bumpy margin, blurred margin, intra-tumoral vessels and α-fetoprotein (AFP) were identified as independent predictors of Ki67 expression (p < 0.05). Three CECT image features and one clinical factor were selected to construct the nomogram, which showed great discrimination ability in the training and validation cohort with AUCs of 0.932 and 0.870 respectively. The calibration curve and DCA indicated that the nomogram had positive clinical application.

The nomogram, developed based on CECT image features and clinical factors in this study, provide a non-invasive method for accurately predicting the expression level of Ki67, and has the potential to support clinicians in making pretreatment decisions.