Can DKI-MRI predict recurrence and invasion of peritumoral zone of hepatocellular carcinoma after transcatheter arterial chemoembolization?

BACKGROUND: Hepatocellular carcinoma (HCC) is a major cause of cancer-related mortality worldwide. Transcatheter arterial chemoembolization (TACE) has been performed as a palliative treatment for patients with HCC. However, HCC is easy to recur after TACE. Magnetic resonance imaging (MRI) has clinical potential in evaluating the TACE treatment effect for patients with liver cancer. However, traditional MRI has some limitations. AIM: To explore the clinical potential of diffusion kurtosis imaging (DKI) in predicting recurrence and cellular invasion of the peritumoral liver zone of HCC after TACE. METHODS: Seventy-six patients with 82 HCC nodules were recruited in this study and underwent DKI after TACE. According to pathological examinations or the overall modified response evaluation criteria in solid tumors (mRECIST) criterion, 48 and 34 nodules were divided into true progression and pseudo-progression groups, respectively. The TACE-treated area, peritumoral liver zone, and far-tumoral zone were evaluated on DKI-derived metric maps. Non-parametric U test and receiver operating characteristic curve (ROC) analysis were used to evaluate the prediction performance of each DKI metric between the two groups. The independent t-test was used to compare each DKI metric between the peritumoral and far-tumoral zones of the true progression group. RESULTS: DKI metrics, including mean diffusivity (MD), axial diffusivity (DA), radial diffusivity (DR), axial kurtosis (KA), and anisotropy fraction of kurtosis (Fak), showed statistically different values between the true progression and pseudo-progression groups (P < 0.05). Among these, MD, DA, and DR values were higher in pseudo-progression lesions than in true progression lesions, whereas KA and FAk values were higher in true progression lesions than in pseudo-progression lesions. Moreover, for the true progression group, the peritumoral zone showed significantly different DA, DR, KA, and FAk values from the far-tumoral zone. Furthermore, MD values of the liver parenchyma (peritumoral and far-tumoral zones) were significantly lower in the true progression group than in the pseudo-progression group (P < 0.05). CONCLUSION: DKI has been demonstrated with robust performance in predicting the therapeutic response of HCC to TACE. Moreover, DKI might reveal cellular invasion of the peritumoral zone by molecular diffusion-restricted change.