Differentiating Cytokeratin 19 expression of hepatocellular carcinoma by using multi-b-value diffusion-weighted MR imaging with mono-exponential, stretched exponential, intravoxel incoherent motion, diffusion kurtosis imaging and fractional order calculus models

Purpose: To evaluate Cytokeratin 19 expression in hepatocellular carcinoma (HCC) with diffusion parameters derived from mono-exponential model (MEM), stretched exponential model (SEM), diffusion kurtosis imaging (DKI), intravoxel incoherent motion (IVIM) imaging, and fractional order calculus (FROC) model and compare their predictive performance.

Method: 61 patients with pathologically confirmed primary HCC were included in this prospective study. All the DWIs were acquired using a 3.0 T MR scanner with 10b-values (0-2000 s/mm²). The apparent diffusion coefficient (ADC), distributed diffusion coefficient (DDC), heterogeneity index (α), apparent kurtosis coefficient (AK), apparent diffusion coefficient (AD), pseudo-diffusion coefficient (D_p), true diffusion coefficient (D_t), perfusion fraction (f), diffusion coefficient (D), fractional order parameter (β), and a microstructural quantity (μ) were calculated. The diagnostic efficacy of various diffusion parameters for predicting CK19 expression of HCC was compared.

Results: ADC, DDC, D_t, D_p, AD, and D were significantly lower in CK19-positive HCCs than in CK19-negative HCCs (P ≤ 0.05). β was significantly higher in CK19-positive group (P = 0.001). AD (AUC = 0.845) had the greatest AUC values in differentiating CK19-positive and CK19-negative HCC with individual parameters. The combination of β, AD, and D_p generated the highest area under the ROC curve (AUC = 0.881) compared with individual parameters.

Conclusions: ADC, DDC, D_t, D_p, D, and β may serve as noninvasive and quantitative imaging markers and holds great potential in accurately accessing CK19 status of HCC. More importantly, the combination of different diffusion parameters yielded better diagnostic performance.