Oncogenic KRAS Requires Complete Loss of BAP1 Function for Development of Murine Intrahepatic Cholangiocarcinoma

Intrahepatic cholangiocarcinoma (ICC) is a primary liver cancer that currently has limited treatment options. As a result, patients with this disease generally have a poor prognosis. Previous studies have demonstrated that mutations in KRAS and loss of BRCA-1-associated protein 1 (BAP1) are frequently found in ICC. We developed a mouse model for ICC that incorporates KRAS and BAP1 mutations in a liver-dependent fashion to aid in the improvement of our understanding of this devastating disease. Our findings suggest that complete loss of BAP1 function combined with mutant KRAS appears to be a requirement for inducing ICC formation within the liver. Intrahepatic cholangiocarcinoma (ICC) is a primary biliary malignancy that harbors a dismal prognosis. Oncogenic mutations of KRAS and loss-of-function mutations of BRCA1-associated protein 1 (BAP1) have been identified as recurrent somatic alterations in ICC. However, an autochthonous genetically engineered mouse model of ICC that genocopies the co-occurrence of these mutations has never been developed. By crossing Albumin-Cre mice bearing conditional alleles of mutant Kras and/or floxed Bap1, Cre-mediated recombination within the liver was induced. Mice with hepatic expression of mutant Kras(G12D) alone (KA), bi-allelic loss of hepatic Bap1 (B(homo)A), and heterozygous loss of Bap1 in conjunction with mutant Kras(G12D) expression (B(het)KA) developed primary hepatocellular carcinoma (HCC), but no discernible ICC. In contrast, mice with homozygous loss of Bap1 in conjunction with mutant Kras(G12D) expression (B(homo)KA) developed discrete foci of HCC and ICC. Further, the median survival of B(homo)KA mice was significantly shorter at 24 weeks when compared to the median survival of & GE;40 weeks in B(het)KA mice and approximately 50 weeks in B(homo)A and KA mice (p < 0.001). Microarray analysis performed on liver tissue from KA and B(homo)KA mice identified differentially expressed genes in the setting of BAP1 loss and suggests that deregulation of ferroptosis might be one mechanism by which loss of BAP1 cooperates with oncogenic Ras in hepato-biliary carcinogenesis. Our autochthonous model provides an in vivo platform to further study this lethal class of neoplasm.