Apparent Diffusion Coefficient Combined with LI-RADS Imaging Features to Predict Ki-67 Expression of Hepatocellular Carcinoma

Abstract

Abstract Background: The optimal cut-off value of Ki-67 to guide the clinical management of patients with HCC remains undetermined, although previous studies have shown that high Ki-67 expression is associated with tumor differentiation, lymph node metastasis, and poor prognoses. The aim of this study was to investigate the value of apparent diffusion coefficient (ADC) and Liver Imaging Reporting and Data System (LI-RADS) v2018 imaging features at different Ki-67 labeling index (LI) levels in preoperative prediction of Ki-67 expression in hepatocellular carcinoma (HCC). Methods: A total of 124 patients with HCC who underwent enhanced-MRI and were enrolled in this study. The patients were divided into three groups according to Ki-67 cut-offs: Ki-67 LI<10% (n=33) vs. Ki-67 LI≥10% (n=91); Ki-67 LI< 25% (n=61) vs. Ki-67 LI≥25% (n=63); Ki-67 LI<50% (n=92) vs. Ki-67 LI≥50% (n=32). A multivariable model was established using ADC and LI-RADS v2018 imaging features. The performance of the multivariable model was assessed using receiver operating characteristic curves. Results: When Ki-67 LI≥10%, ADC (odds ratio [OR]=0.018, 95% confidence interval [CI]: 0.001–0.343, P=0.008) was an independent predictor of HCC, with an AUC of 0.684 (0.594–0.764), sensitivity and specificity of 90.91% and 46.15%, respectively. Similarly, when Ki-67 LI≥25%, ADC (OR=0.002, 95% CI: 0.000–0.048, P<0.001) was an independent predictor, with an AUC of ADC of 0.784 (0.701–0.853), sensitivity and specificity of 93.44% and 61.90%, respectively. When ki-67 LI≥50%, mosaic architecture (OR=6.844, 95% CI: 1.815–25.811, P=0.005) and ADC (OR=0.001, 95% CI: 0.000–0.015, P<0.001) were independent predictors of HCC. The sensitivity, specificity and AUC of the model based on ADC and mosaic architecture were 94.57%, 74.12% and 0.883 (0.813–0.934), respectively. The diagnostic performance of the combined model was better than that of ADC (0.883 vs. 0.825, P=0.02) and mosaic architecture (0.883 vs. 0.704, P<0.001). Conclusions: When Ki-67 LI≥50%, ADC combined with Mosaic architecture has good differential performance, and may be an effective imaging approach for clinical decision-making and prognostic assessment in patients with HCC.