Diffusion kurtosis imaging to assess response to treatment in hypervascular hepatocellular carcinoma.

Abstract

The objective of our study was to compare diffusion kurtosis imaging (DKI) with conventional diffusion-weighted imaging (DWI) for assessing the response to treatment in hypervascular hepatocellular carcinoma (HCC). Sixty-two consecutive patients with treated or untreated hypervascular HCC underwent MRI of the liver including DKI (b values of 0, 100, 500, 1000, 1500, and 2000 s/mm(2)). The mean kurtosis (MK) and apparent diffusion coefficient (ADC) values of the hepatic parenchyma and of the HCCs were computed. The detectability of viable HCC based on MK and ADC values was compared. We also assessed the correlation between Child-Pugh grades and MK or ADC values. For a total of 112 HCC nodules (viable, n = 63; nonviable, n = 49), the MK value was significantly higher for the viable group (mean ± SD, 0.81 ± 0.11) than for the non-viable group (0.57 ± 0.11) (p < 0.001). The mean ADC value was significantly lower for the viable group (1.44 ± 0.42 × 10(-3) mm(2)/s) than for the nonviable group (1.94 ± 0.52 × 10(-3) mm(2)/s) (p < 0.001). The sensitivity, specificity, and AUC of the ROC curve for the assessment of HCC viability were greater (p < 0.001) using MK (85.7%, 98.0%, and 0.95, respectively; cutoff value = 0.710) than using ADC (79.6%, 68.3%, and 0.77, respectively; cutoff value = 1.535 × 10(-3) mm(2)/s). Although the ADC of hepatic parenchyma was lower in patients with Child-Pugh grade B or C disease than in those with grade A disease (p = 0.02), no significant difference in MK (p = 0.45) was found among the Child-Pugh grades. DKI can be a new option for the assessment of posttherapeutic response in HCC.