Investigating Susceptibility of ꞵ‐catenin‐mutated Hepatocellular Carcinoma to Checkpoint Inhibitors

Abstract

Hepatocellular Carcinoma (HCC) is the 5th most common cause of cancer‐related death in the United States with an estimated 32,000 annual deaths. The standard of care for HCC as established by the IMBrave150 trial is a combination of the PD‐L1 inhibitor, Atezolizumab, and an antiangiogenic, Bevacizumab. While this combination extends overall survival to 19.2 months in unresectable HCC, only 52% of patients survive to 18 months. Furthermore, objective response rates are around 32% and hence clarifying underlying mechanisms regulating the immune response in HCC oncogenesis might strengthen immunotherapy efficacy. Especially important is to identify response rates of these agents in specific molecular subclasses of HCC may help in better patient selection. The oncogene ꞵ‐catenin is consistently mutated and activated among 30% of HCC patients. The Wnt/ꞵ‐catenin pathway has been approached as a target for precision medicine in HCC. We recently modeled HCCs with ꞵ‐catenin mutations in mice up to 69% genetic similarity. Here, we combine hydrodynamic tail vein injections with the Sleeping Beauty Transposase which is a non‐viral method of expressing genes of interest. Our model co‐expresses an S45Y point mutant of ꞵ‐catenin with the tyrosine kinase receptor MET (B+M). This combination represents ~10% of patient HCCs and provides a clinically relevant tool for the investigation of personalized HCC therapeutics.A recent study demonstrated HCCs driven by a Δ90‐truncation mutant of ꞵ‐catenin in combination with the Myc oncogene are shown to lack sufficient immune cell surveillance. These HCCs have poor cytotoxic T cell and dendritic cell recruitment while responding poorly to checkpoint inhibitors. Since activating point mutations in CTNNB1 gene are common in patient HCCs, we asked whether poor immune surveillance extends to HCC models harboring point mutations. We first checked and found the presence of CD45+, F4/80+, and CD11b+ cells both 2 and 4 weeks after B+M tumor induction. This indicates the presence of both adaptive and innate immune cells in the B+M tumor microenvironment. We next asked whether checkpoint inhibition will effectively reduce tumor burden in the B+M model. Therefore, we treated B+M mice with either IgG or anti‐PD1 monoclonal antibodies 4 weeks after tumor induction. Tissues were harvested 3.5 weeks after treatment. We found a modest 20% reduction in liver weight to body weight ratio, a measure of tumor burden, in B+M mice treated with anti‐PD1 compared to IgG controls. Our results demonstrate that HCCs with missense ꞵ‐catenin mutations do respond albeit marginally to checkpoint inhibitors. In the future, combination of checkpoint inhibitors with other inhibitors will be tested in these models to determine their efficacy in treating ꞵ‐catenin‐mutated HCCs.