Development and validation of quantitative magnetic resonance imaging (MRI) of nonalcoholic fatty liver disease (NAFLD) ; Développement et validation de l'imagerie par résonance magnétique quantitative dans la stéatohépatite métabolique chez la souris

Nonalcoholic fatty liver disease (NAFLD) has emerged as the most common form of chronic liver diseases. Traditionally, NAFLD includes two subgroups, nonalcoholic fatty liver (NAFL) and a progressive inflammatory subgroup, called nonalcoholic steatohepatitis (NASH), which can progress to fibrosis, cirrhosis and hepatocellular carcinoma. Besides this classification, fibrosis is currently recognized as the main prognostic factor of NAFLD. Assessing NAFLD severity requires liver histological analysis, obtained after liver biopsy which is an invasive procedure with sampling variability and moderate inter and intra-operator reproducibility. Quantitative magnetic resonance imaging (MRI) may be used as a non-invasive alternative for assessing NAFLD. In the first part of this work, the diagnostic value of several quantitative MRI parameters has been assessed in mice models of NAFLD by correlating the MRI data with detailed hepatic histological analysis. The MRI parameters included multifrequency visco-elastic parameters obtained with MR elastography, proton density fat fractions (PDFF) and T1 relaxation times. In the second part of the work, a novel temporal diffusion spectroscopy MRI sequence with oscillating gradient spin echo-echo planar imaging (OGSE-EPI) has been implemented and validated in phantoms and mice to further characterize NAFLD with quantitative MRI. The results of the first mice study showed that the MRE visco-elastic parameters, and particularly the loss modulus, were highly correlated with hepatic fibrosis, and to a lesser degree with inflammation. None of the visco-elastic parameters was correlated with hepatocyte ballooning, an important histopathological feature of NASH. The diagnostic performance of MRE derived parameters was moderate for diagnosing NASH and high for diagnosing significant fibrosis. The results also confirmed PDFF as biomarker of hepatic steatosis. Moreover, the T1 values correlated with fibrosis and inflammation. However, the T1 measurements were affected by hepatic steatosis. ...