Quantitative characterization of the liver-pancreas axis in diabetes via multiparametric magnetic resonance elastography

PROJECT SUMMARY / ABSTRACT Nonalcoholic fatty liver disease (NAFLD) is closely associated with the impairment of many metabolic pathways, including decreased hepatic insulin sensitivity and secretion, increased glucagon, and the risk of developing type 2 diabetes mellitus (T2DM). Conversely, patients with diabetes have a higher prevalence of steatohepatitis and end- stage liver disease. As an intermediate state of hyperglycemia easily detectable in clinical settings, prediabetes is the condition that precedes T2DM but the progression is variable, which could increase over time with associated risk factors. To date, the mechanisms underlying prediabetic progression are still not understood. We think simultaneous imaging of the abdominal organs’ abnormalities along the liver-pancreas axis and measures of insulin action, secretion, and hepatic extraction can provide insights into T2DM development, a better understanding of the relationship between NAFLD and diabetes, and an opportunity to intervene in prediabetic progression. The overall goal of this work is to develop an advanced multiparametric abdominal MRE method for characterizing pathophysiologic state of the liver-pancreas axis and prediabetic progression in NAFLD. • In Aim 1, A multifrequency, self-navigating, and hybrid radial-Cartesian 3D vector magnetic resonance elastography (MRE) technology will be developed and expanded on the liver, pancreas and fat data acquisition and image reconstruction. Ten healthy volunteers will be recruited during MRE driver, imaging sequence and reconstruction development and optimization. Benchmark for technical success will be evaluated with a test-retest repeatability and measurement agreement study on ten patients with diagnosed NAFLD. • In Aim 2, We will perform quantitative imaging assessments of the liver-pancreas axis in a cross-sectional study in 150 patients with diagnosed NAFLD with varying risk or progression to diabetes (i.e., 50 patients without diabetes, 50 patients with prediabetes, 50 patients with diabetes). Laboratory and clinical data, as well as insulin secretion and action measurements, will be collected within 30 days of MRI/MRE. Precursory tissue abnormalities and integrative metabolic-mechanical models will be trained and evaluated to assess the risk of developing T2DM. • In Aim 3, assuming 20% withdrawal rate, we plan to have one-year follow-up examinations in 120 patients out of those who have baseline examinations in Aim 2. The models trained in Aim 2 will be further tuned with longitudinal MRI/MRE changes for assessing bi-directional treatment effects on tissue composition, structural and metabolic features, bridging the cellular, gland, enzymes, and hormones to the liver-pancreas axis, thereby predicting progression or regression direction of diabetes development in NAFLD. We anticipate that this program will provide initial validation of multiparametric MRI/MRE and the derived surrogate to assess the liver-pancreas axis of diabetes progression in NAFLD. This project’s success will also provide a valuable noninvasive assessment tool for emerging therapeutic interventions.