PROJECT SUMMARY This application, “Highly penetrant and immunogenic mouse models of non-viral HCC that are suitable for evaluation of immune checkpoint inhibitors”, is submitted in response to FOA PAR-17-245 “Research Projects to Enhance Applicability of Mammalian Models for Translational Research”. The goal of this project is to develop accurate, innovative, and immunogenic mouse models of hepatocellular carcinoma (HCC) that closely mimic the main etiologies of human non-viral HCC in their pathogenic, transcriptomic, and genomic profiles. These etiologies include non-alcoholic (NASH) and alcoholic (ASH) steatohepatitis, and type 2 diabetes (T2D). Although the current major etiologies that underlie HCC development are hepatitis virus B and C (HBV, HCV) infections, non-viral HCC is predicted to become the major form of this aggressive cancer in the US and Europe in the not too distant future. While the incidence of non-viral HCC and its associated mortality continue to grow at an alarming rate, progress in HCC treatment has been disappointingly slow. Recently, however, inhibitors of the PD-1:PD-L1 checkpoint were found to be much more effective in HCC treatment than any other targeted or non-targeted therapeutic used in the past. Nonetheless, with objective response rates around 20%, there is much room for future improvement. Such improvement depends on better understanding of how immune checkpoint inhibition leads to HCC regression and the identification of adjuvants that enhance the response to this new class of drugs. Both objectives are dependent on the availability of immunogenic mouse models of human HCC. We will bank on our recent success in developing an immunogenic mouse model of NASH-driven HCC that is responsive to PD-1/PD-L1 blockade to develop new and improved mouse models of non-viral HCC that show rapid and synchronized tumor development, making them highly useful for translational research. In addition to NASH-driven HCC, we will develop new immunogenic models of ASH-driven HCC. We will also generate a series of mouse HCC-derived cell lines that give rise to synchronized orthotopic tumors, whose growth and response to treatment can be monitored by in vivo imaging. To determine and demonstrate the human relevance of these models, they will be subjected to extensive genomic and transcriptomic characterization and immunoprofiling. The results of these analyses will be compared to the genomic and transcriptomic profiles of human HCC using innovative computational tools. The following Specific Aims will be pursued: 1) Use the MUP- uPA model of NASH driven HCC to compare the carcinogenic efficacy of different NASH-related diets; 2) Use the MUP-uPA mouse to develop new models of ASH-induced HCC that do not involve HFD feeding; 3) Compare mouse and human HCC genomic, transcriptomic, and immune profiles; and 4) Generate cell lines from the different mouse HCC models that will give rise to synchronized orthotopic tumors that are useful for drug testing.

PROJECT NARRATIVE We will develop new mouse models of non-viral hepatocellular carcinoma (HCC) that depend on accumulation of random mutations as well as HCC cell lines that can be used to generate orthotopic tumors. Our preliminary studies show that such models are immunogenic and highly suitable for the evaluation of immune checkpoint inhibitors and adjuvants that further potentiate the response to checkpoint blockade. We will carry out extensive transcriptomic, genomic, and immunophenotypic analyses and will use novel computational tools to establish the similarity between the mouse models and human HCC of different etiologies.