3D Free-Breathing Fat and Iron Corrected T1 Mapping

Project Summary Non-alcoholic fatty liver disease (NAFLD) has a global prevalence of ~25% and is a leading etiology of chronic liver disease worldwide. Nonalcoholic steatohepatitis (NASH), as an advanced form of NAFLD, involves steatosis, inflammation and concomitant iron deposition in the liver. Due to chronic cellular injury, NASH can also cause liver fibrosis, which is a common trigger of more severe liver complications such as cirrhosis, portal hypertension, or hepatocellular carcinoma (HCC). As a result, simultaneous measures of these clinical indicators are crucial to provide more accurate and comprehensive assessment of NASH. Magnetic resonance imaging (MRI) is a well- accepted modality for evaluating NASH due to its excellent soft-tissue contrast resolution and abundant contrast mechanisms without radiation exposure, allowing for multiparametric assessment of the liver. Several MRI- related biomarkers, such as proton density fat fraction (PDFF), T2*/R2*, and elastography-derived stiffness, have been developed and established for quantitative assessment of liver fat, iron, and fibrosis, respectively, but none has been demonstrated for evaluating inflammation, the most important clinical hallmark throughout the disease course of NASH. T1 is an MRI parameter that can potentially be a useful marker for NASH, and there has long been an interest in using T1 changes to characterize liver inflammation and/or fibrosis. However, existing standard T1 mapping methods suffer from several limitations when applied to NASH patients. First, and most important, standard T1 mapping only measures composite T1 (water-fat-iron-mixed T1) of the liver, which can result in substantial bias due to the increased fat and iron content in the liver of NASH patients. This is because both fat and iron have short T1, while conversely, inflammation/fibrosis tends to prolong T1. Second, T1 mapping of the liver, particularly with 3D whole-liver coverage, remains challenging due to respiratory motion and the slow imaging speed of MRI, which can reduce measurement accuracy and reproducibility. These problems represent major barriers for evaluating fatty liver diseases and associated disease heterogeneity using T1 mapping. The overarching goal of this application is to propose and test a novel 3D liver T1 mapping method that could address these challenges. The new method, called Magnetization-Prepared Dixon Golden-angle RAdial Sparse Parallel (MP-Dixon-GRASP) MRI, features adaptive inversion recovery-prepared multi-echo stack-of-stars acquisition in combination with advanced model-based image reconstruction, which will enable free-breathing fat/iron- corrected T1 mapping to estimate the “true T1” of underlying liver parenchyma. Our main hypothesis to be tested in this R21 proposal is that by removing the influence of fat and iron, more accurate T1 can be estimated as a better marker of liver inflammation/fibrosis. If successful, this technique can be further combined with concomitant T2* and PDFF quantification to ultimately provide a new free-breathing 3D multiparametric MRI technique for assessment of NASH and potentially other chronic liver diseases of high clinical impact.