Abstract: Deaths due to metastatic hepatocellular carcinoma (HCC) continue to mount due to a low success of clinical intervention. Targeted cancer therapy eliminates all malignant tumor-initiating stem-like cells ( TICs) to prevent relapse and metastasis. Development of HCC is associated with alcohol and/or HCV infection and frequently observed in patients with gene aberrations in component complex SWI/SNF. Our synthetic lethal gene screening of genome-wide CRISPR-based knockout (GeCKO)-lentivirus library in ARID1A-mutant cells of our humanized mouse models identified components (EZH2 and EED) of polycomb suppressor complex 2 (PRC2). We demonstrated that pluripotency transcription factor NANOG metabolically reprograms TICs to promote self-renewal via inhibition of oxidative phosphorylation (OXPHOS) and activation of fatty acid oxidation (FAO). We characterized alcohol-associated human HCCs and alcohol-fed, HCV NS5A transgenic (Tg) mice for HCC development and a critical role of NANOG as a core stem cell factor following TLR4 activation as a crucial component needed for genesis and maintenance of TICs. These findings support our hypothesis that alcohol-mediated NANOG induction and ARID1A mutations cooperatively generate chemoresistant TICs in alcohol-associated HCC. Therefore, targeting ARID1A, a of chromatin remodeling NANOG-PRC2 complex and FAO should eliminate the TIC subpopulation. We propose three specific aims. In Aim 1, we will test if NANOG with PRC2 complex inhibits OXPHOS and generates sorafenib-resistant TICs with mutant ARID1A in alcohol mediated human HCCs. In Aim 2, we will examine how inhibition of FAO inhibit self-renewal and promote apoprtosis in sorafenib-resistant mutant ARID1A tumors. In Aim 3, we will test if combination therapy of sorafenib plus PRC2/FAO inhibition induces metabolic reprogramming and apoptosis in HCC. We will test if targeting synthetic lethality genes (EZH2) eliminates TICs with ARID1A mutations by use of alcohol Western diet-fed and/or HCV-infected FRG-hu-Hep/HSC humanized mice and patient-derived xenograft. In summary our genotyped HCC model will streamline clinical trials for disease modeling/toxicity evaluation by offering “a clinical trial in a humanized mouse”. Use of PDX and humanized mouse systems will provide a path to paradigm-shifting personalized medicine for an accurate synthetic lethality targeting strategy. The timely eradication of chemoresistant TICs arising from chronic alcohol exposure will be beneficial and cost-effective for patients suffering from an otherwise incurable disease. This proposal will establish a personalized therapy strategy targeting a synthetic lethal pathway (PRC2)-NANOG complex and FAO in mutant-ARID1A HCCs. A successful outcome will lead to a paradigm shift, thus facilitating the future development of personalized treatment strategies targeted towards NANOG+ TICs appearing in alcohol-related HCC.