The value of 18F-FDG PET/MRI in prediction of microvascular invasion in hepatocellular carcinoma

Rationale and Objectives: Investigating the association between maximum standardized uptake value (SUV_max), peritumoral and intratumoral apparent diffusion coefficient (ADC) values and whether these parameters are useful in predicting the preoperative microvascular invasion (MVI) of hepatocellular carcinoma (HCC). Materials and methods: Forty-four patients [8 women and 36 men, median age of 62 (21–76)] with single HCCs (≥2 cm) who underwent preoperative 18F-FDG PET/MRI were retrospectively evaluated. The peritumoral and intratumoral ADC values were evaluated on diffusion-weighted images using Image J an open software and the intratumoral SUV_max values were measured on fusion 18F-FDG PET/MRI images. Univariate and multivariate logistic regression analyses were performed to determine the most influential factor predicting MVI. Interobserver agreement was checked using the intraclass correlation coefficient (ICC). Results: Univariate analysis showed that the histologic grade, tumor size, maximum peritumoral ADC (PT_band ADC_max), mean peritumoral ADC (PT_band ADC_mean), mean intratumoral ADC (IT ADC_mean), and maximum SUV (SUV_max) correlated with MVI (p < 0.05). On multivariate analysis, the SUV_max was the only independent risk factor for the MVI of HCC [OR, 1.68; 95% CI (1.04–2.70); p = 0.032)]. An AUC value of 0.896, 95% CI, 0.786–1.0) had a sensitivity of 75% and specificity of 97% using the best cut-off SUV_max 5.85 to differentiate MVI-positive HCCs from MVI-negative. The PPV, NPV, and accuracy were 92.3%, 87%, and 89%, respectively. The ICC values were 0.95–0.99, a nearly perfect level of agreement. Conclusion: 18F-FDG PET/MRI is a useful noninvasive imaging tool for predicting the MVI of HCC.