Characterization of intratumor heterogeneity in a porcine hepatocellular carcinoma model

In 2022, there are expected to be ~41,000 new liver cancer cases and ~30,500 deaths, of which 75% will be due to hepatocellular carcinoma (HCC). The Oncopig is a promising translational large animal cancer model that can recapitulate human HCC phenotypes due to its similarities with humans in terms of size, anatomy, liver metabolism, and genetics. Previously, with whole genome sequencing, we identified Oncopig HCC intratumor heterogeneity that resembled signatures observed in human HCC tumors{1}. In this study, we designed a whole exome sequencing kit to decrease sequencing costs and aimed to further investigate genes and relevant pathways implicated in HCC carcinogenesis in our Oncopig HCC model.

Materials and Methods:

An Oncopig HCC tumor was induced following intrahepatic autologous injection of Oncopig HCC cells into a cirrhotic liver microenvironment. The Oncopig HCC tumor was biopsied at 5 different locations within the tumor to characterize intratumor heterogeneity. Tumor fragments, the injected HCC cell line, and a control kidney sample were sequenced and aligned to the pig reference genome (Sscrofa11.1) using BWA MEM. Duplicate reads were removed using the GATK MarkDuplicates function. Strelka 2.9 was used to call somatic single nucleotide variants (SNVs). Variants with a sufficient empirical variant score in at least one sample passed the filter. The functional impact of identified SNVs was predicted using SnpEff. Ingenuity Pathway Analysis was used to identify pathways enriched for genes impacted by SNVs.

Results: 10,735 total SNVs were identified within different regions of the Oncopig HCC tumor associated with 39,634 effects with approximately 671 (1.7%) high impact effect. 8,456 (66.2%) effects were missense and 650 (5.1%) were nonsense. 13,042 (32.9%) effects were identified in exon regions. We identified variants in the TP53BP1, CDKN2AIP, TERT, and RPS6KA3 genes all found to be previously related to HCC driver genes{2}. Over 1000 genes with moderate to high impact variants were identified to be involved with hepatic fibrosis, liver hyperplasia, hepatitis, hepatic steatosis, type II diabetes mellitus signaling, and cell death and survival. TP53 and KRAS were identified as potential upstream regulators of 271 and 113 genes with moderate to high variant effects respectively.

Conclusion: In this study, we showed intratumor heterogeneity in the Oncopig recapitulates common phenotypes expressed in human HCC proving further validation of our model with the goal of studying HCC tumor biology and testing therapeutic options.