Related Experiment Video
Updated: Jul 4, 2025
Author Spotlight: Advancing Hepatic Fibrosis Diagnosis Using Magnetic Resonance Elastography and AI
Published on: July 21, 2023
A gadoxetic acid-enhanced MRI-based model using LI-RADS v2018 features for preoperatively predicting Ki-67 expression
Yingying Liang1, Fan Xu2, Qiuju Mou3
1Department of Radiology, Guangzhou First People's Hospital, School of Medicine, South China University of Technology, 1Panfu Road, Guangzhou, Guangdong Province, 510180, China.
Purpose:
To construct a gadoxetic acid-enhanced MRI (EOB-MRI) -based multivariable model to predict Ki-67 expression levels in hepatocellular carcinoma (HCC) using LI-RADS v2018 imaging features.
Methods:
A total of 121 patients with HCC who underwent EOB-MRI were enrolled in this study. The patients were divided into three groups according to Ki-67 cut-offs: Ki-67 ≥ 20% (n = 86) vs. Ki-67 < 20% (n = 35); Ki-67 ≥ 30% (n = 73) vs. Ki-67 < 30% (n = 48); Ki-67 ≥ 50% (n = 45) vs. Ki-67 < 50% (n = 76). MRI features were analyzed to be associated with high Ki-67 expression using logistic regression to construct multivariable models. The performance characteristic of the models for the prediction of high Ki-67 expression was assessed using receiver operating characteristic curves.
Results:
The presence of mosaic architecture (p = 0.045), the presence of infiltrative appearance (p = 0.039), and the absence of targetoid hepatobiliary phase (HBP, p = 0.035) were independent differential factors for the prediction of high Ki-67 status (≥ 50% vs. < 50%) in HCC patients, while no features could predict high Ki-67 status with thresholds of 20% (≥ 20% vs. < 20%) and 30% (≥ 30% vs. < 30%) (p > 0.05). Four models were constructed including model A (mosaic architecture and infiltrated appearance), model B (mosaic architecture and targetoid HBP), model C (infiltrated appearance and targetoid HBP), and model D (mosaic architecture, infiltrated appearance and targetoid HBP). The model D yielded better diagnostic performance than the model C (0.776 vs. 0.669, p = 0.002), but a comparable AUC than model A (0.776 vs. 0.781, p = 0.855) and model B (0.776 vs. 0.746, p = 0.076).
Conclusions:
Mosaic architecture, infiltrated appearance and targetoid HBP were sensitive imaging features for predicting Ki-67 index ≥ 50% and EOB-MRI model based on LI-RADS v2018 features may be an effective imaging approach for the risk stratification of patients with HCC before surgery.
