Magnetic resonance elastography and proton density fat fraction predict adverse outcomes in hepatocellular carcinoma

Magnetic resonance (MR) elastography and proton density fat fraction (PDFF) are emerging techniques for non-invasive assessment of liver stiffness and steatosis, respectively. We investigated the role of MR metrics in pre-treatment prognostication of hepatocellular carcinoma (HCC).Patients with newly diagnosed HCC were prospectively recruited. Pre-treatment MR elastography and PDFF were performed on tumor and non-tumor regions. HCC treatment was categorized as potentially curative (resection/ablation) or non-curative (locoregional/systemic therapy). HCC recurrence, liver-related complications (ascites/ variceal bleeding/ hepatic encephalopathy) and mortality were monitored.Of the 158 recruited patients (mean age 62.9 years, 84.2% male, 82.9% viral hepatitis), 58.2% (n = 92) and 41.8% (n = 66) received potentially curative and non-curative therapy, respectively. Pre-treatment non-tumor liver stiffness independently predicted liver-related complications, regardless of treatment type (HR 1.384, 95% CI 1.067-1.796, p = 0.014). In the potentially curative therapy group, non-tumor stiffness and non-tumor PDFF were independently associated with HCC recurrence (HR 1.308, 95% CI 1.022-1.673 & HR 1.080, 95% CI 1.009-1.156 respectively, both p < 0.05); and non-tumor PDFF predicted mortality (HR 1.160, 95% CI 1.038-1.296, p = 0.009). In the non-curative group, tumor stiffness independently predicted liver-related complications (HR 1.299, 95% CI 1.023-1.651, p = 0.032), and a combination of tumor stiffness ≥ 5.7 kPa plus non-tumor stiffness ≥ 3.7 kPa was associated with a two-fold risk of liver-related complications (86.7% vs 40.0%, p < 0.001).Pre-treatment MR elastography and PDFF over tumor and non-tumor regions demonstrated prognosticating roles in HCC. Simultaneous measurements of both metrics during conventional MR liver should be considered in the diagnostic workup of HCC.