PROJECT SUMMARY Nonalcoholic fatty liver disease (NAFLD) affects approximately one-third of the general population in the United States. Nonalcoholic steatohepatitis (NASH) - a more severe form of NAFLD - develops in 5-10% of NAFLD patients (i.e., 5-10 million people in the US) with the hallmark of hepatic inflammation in the setting of hepatic steatosis. Differentiation of NASH from simple hepatic steatosis is vital for patient management in NAFLD as NASH is associated with accelerated progression into end-stage liver diseases (liver failure and liver cancer) and with much higher liver-related mortality than the simple fatty liver. While noninvasive imaging methods have been developed to quantify liver fat and liver fibrosis, there is an unmet need for new imaging methods to detect and grade liver inflammation. The goal of this project is to develop a positron emission tomography (PET) method to enable noninvasive assessment of liver inflammation and NASH diagnosis. Using the widely accessible radiotracer 18F-fluorodeoxyglucose (FDG), we propose a liver parametric PET method using dynamic scanning and tracer kinetic modeling. We hypothesize that glucose transport measured by liver parametric PET can accurately characterize the liver inflammation underlying NASH. One major challenge with enabling this technique is that the liver is a unique organ with dual blood supplies from the hepatic artery and portal vein, requiring the knowledge of dual-blood input function (DBIF) for kinetic modeling, which is difficult to obtain in human PET studies. Our preliminary work has addressed this technical challenge by developing a novel optimization-derived DBIF modeling approach. With improved kinetic modeling, we have discovered in a pilot clinical study that the blood-to-tissue glucose transport rate K1 was strongly associated with histological liver inflammation grades. The focus of this proposal is to further develop and optimize the methodology of liver parametric PET and validate its effectiveness for assessing liver inflammation and diagnosing NASH in patients with NAFLD. More specifically, we will (1) develop the technical method of liver parametric PET; (2) Evaluate glucose transport rate K1 as an effective and unique PET biomarker of liver inflammation; (3) Combine liver parametric PET with CT or MR methods for multiparametric NASH diagnosis; (4) Shorten the scan time of liver parametric PET from one hour to fifteen minutes. The integrated outcome of these specific aims will be a new technical capability and validation of 18F-FDG PET for liver inflammation assessment and for detecting and grading NASH. The proposed method will fill the critical gap for noninvasive assessment of liver inflammation in fatty liver disease. Successful completion of this project has the potential to profoundly impact the clinical management of NAFLD patients and clinical trials of new NASH treatments.