The added value of F-FDG PET/MRI multimodal imaging in hepatocellular carcinoma for identifying cytokeratin 19 status.

Purpose: In hepatocellular carcinoma (HCC), cytokeratin 19(CK19) has been proven to be associated with clinical aggressiveness. Therefore, this study aimed to explore the added value of F-FDG PET/MRI in predicting CK19 status in HCC.

Methods: Sixty-six patients who underwent whole-body or abdominal F-FDG PET/MRI after conventional PET/CT for HCC were retrospectively enrolled. The maximal standard uptake value (T-SUVmax) and the mean apparent diffusion coefficient (T-ADCmean) of the tumor (T), as well as those of the normal liver tissues (L) were derived, followed by calculations of the T-SUVmax/L-SUVmax (SUVmax-T/L) and the T-ADCmean/L-ADCmean (ADCmean-T/L) ratios. Combined with the postoperative pathological results, the performance in predicting the CK19 status in HCC was evaluated using receiver operating characteristic analysis (ROC).

Results: The areas under the ROC curve (AUCs) for T-SUVmax, SUVmax-T/L, T-ADCmean, and ADCmean-T/L in predicting the CK19-positive HCC were 0.700, 0.717, 0.717, and 0.735, respectively. In the logistic regression analysis, the T-SUVmax was an independent and significant factor to predict CK19-positive HCC, with an odds ratio of 1.27. In addition, no significant differences were found in the pathological grading, microvascular invasion, liver capsular invasion, Hepatitis B virus (HBV) infection, alpha fetoprotein (AFP) level, and tumor diameter between the CK19-positive and CK19-negative groups, except the recurrent rate.

Conclusions: The radiomic features derived from F-FDG PET/MRI can be used to predict the CK19 status of HCC. T-SUVmax and T-ADCmean were significant indicators, whereas T-SUVmax was an independent predictor.