Magnetic resonance elastography (MRE) is a noninvasive imaging technology for measuring tissue stiffness that is beginning to see widespread clinical use for assessing hepatic fibrosis as an alternative to liver biopsy. During the first funding period, we established that an MRE-assessed damping ratio or loss modulus can be used to detect inflammation before the onset of fibrosis in five different liver disease models. In this proposal, we will further advance the multiparametric hepatic MRE technique and evaluate new imaging biomarkers, including viscoelastic (VE), poroelastic (PE) and nonlinear (NL) mechanical properties, to more fully characterize the pathophysiologic stages of nonalcoholic fatty liver disease (NAFLD). We have two hypotheses: (1) VE, PE and NL properties are biomarkers that will further increase our ability to quantify pathologic processes in nonalcoholic steatohepatitis (NASH) such as inflammation and hepatocellular ballooning, individually or in combination with other mechanical properties; and (2) measurements of fat fraction, VE, PE and NL properties can be integrated into a composite metric that will serve as a reliable noninvasive surrogate for the histologic NAFLD activity score (NAS), the most widely accepted metric for disease state. Our long-term goals are to: (1) noninvasively quantify inflammation and ballooning; and (2) quantify disease progression and regression in preclinical animal models and clinical patients. In this proposed work, an advanced multiparametric, self-navigating, and hybrid radial-Cartesian XD-MRE technique will be developed for the in vivo characterization of the VE, PE and NL properties of the liver. XD-MRE provides 3D vector MRE high- frequency (30-200 Hz) wave information at multiple phases of low-frequency (0.1-5 Hz) motion obtained using free-breathing, endogenous motion or standard breath-held, external motion. We will implement a 3D direct inversion method to calculate VE parameters at high frequencies, a 3D finite-element method to solve the PE equations at low frequencies, and quantify NL effects from VE changes due to tissue deformation (or volumetric strain change) from the low-frequency perturbation (or vibration) of the liver. An efficient ?virtual NAS? (vNAS) imaging protocol, acquiring multiple mechanical parameters and fat fraction, will be developed to predict the NAS. Development and refinement of the vNAS algorithm will be performed in a specifically designed NASH mouse model with disease progression and regression. A clinical vNAS imaging protocol will be developed for NAS prediction in NAFLD patients. We will evaluate the repeatability and reproducibility for measuring the new MRE imaging biomarkers in healthy volunteers and biopsy-proven NAFLD patients using a test-retest strategy. A pilot study will be performed in NAFLD patients to establish the statistical properties of the NAS score predictors, which are essential for future clinical trial design. This work will make it possible for our team and many other clinical investigators who use these advanced techniques to cross-validate ?vNAS? and to explore this promising new application of hepatic MRE to advance the precision of diagnosing NAFLD.
This research will develop and explore hepatic magnetic resonance elastography (MRE) that provides diagnostic information that cannot be obtained using conventional imaging techniques. The research has already shown that multiparametric MRE has a promising role in the detection of the broad pathophysiologic spectrum of chronic liver diseases, an important health problem in the US and worldwide, as an alternative to invasive biopsy. The success of this proposed study will maximize the clinically relevant information provided by hepatic MRE to substantially advance our understanding and ability to diagnose disease progress or regress.
Showing the most recent 10 out of 16 publications
