R. K. Marcus1,2,4, W. Foo3, A. Maitra1, S. Gupta1 1University Of Texas MD Anderson Cancer Center,Sheikh Ahmed Bin Zayed Al Nahyan Center For Pancreatic Cancer Research,Houston, TX, USA 2University Of Texas MD Anderson Cancer Center,Department Of Surgical Oncology,Houston, TX, USA 3University Of Texas MD Anderson Cancer Center,Department Of Pathology & Laboratory Medicine,Houston, TX, USA 4Medical College Of Wisconsin,Department Of Surgery,Milwaukee, WI, USA
Introduction: Cholangiocarcinoma (CCA) is a poorly understood cancer of the biliary epithelium that represents the second most common type of liver cancer. Surgical resection is currently the only treatment associated with prolonged disease free survival. Unfortunately, most patients with CCA present with unresectable disease and have poor responses to standard chemotherapy, resulting in a median survival of less than one year. While elucidating the molecular pathogenesis of CCA may enable identification of pathways that form the basis for targeted therapies and improve early detection efforts, inquiry into this disease has been hampered by a paucity of genetically faithful animal models. The Cancer Genome Atlas analysis and other recent studies have revealed discrete epigenetic perturbations amongst CCAs originating from different anatomic sites. Intrahepatic cholangiocarcinoma (ICC) arises from the intrahepatic bile ducts and its incidence is currently rising in the United States. A subset of ICCs is characterized by loss-of-function mutations in genes encoding for chromatin regulatory factors such as BRCA associated protein 1 (BAP1). This gene encodes a nuclear deubiquitinase with a role in chromatin remodeling. BAP1 is mutated in 3-33% of ICCs. Loss of this protein may be associated with global epigenomic and transcriptomic alterations that contribute to tumor progression and metastasis dissemination.
Methods: A genetically engineered mouse model that incorporates a deletion of BAP1 with an activating mutation of Kras was developed. An Albumin-Cre promoter was used to induce hepatoblast-specific mutations in these genes.
Results: Conditionally deleted BAP1 mice develop focal biliary precursor lesions and frank ICC within a background of hepatocellular carcinoma (HCC). True mixed ICC-HCC lesions also develop. Tissue-specific activation of Kras alone results in the development of HCC with low penetrance and long latency. By combining Kras activation with heterozygous deletion of BAP1, the presence of biliary disease is induced and disease latency shortened. Homozygous deletion further shortens disease latency.
Conclusion: As hepatoblasts are bipotential, differentiating into both cholangiocytes and hepatocytes, the ICC phenotype of our model may be enhanced by biliary tree-specific deletion of BAP1 using adenoviral Cre enzyme (Ad-Cre) to achieve combinatorial specificity. A novel surgery enabling retrograde biliary tree administration of Ad-Cre was developed. To establish a proof-of-principle, Ad-Cre injection into ROSAmT/mG mice to confirm appropriate localization of Cre recombinase expression within cholangiocytes is ongoing. We will build on this by performing Ad-Cre injections in BAP1fl/fl;K-rasLSL-G12D/+ mice to activate Kras and knock out BAP1. Ideally, this model will develop ICC in a timely manner that recapitulates features of human ICC, and, as such, enables interrogation of the oncogenic pathways involved in development of this disease.