Epigenetic regulation of individual components of combined hepatocellular-cholangiocarcinoma

Combined hepatocellular carcinoma-cholangiocarcinoma (HCC-CCA) is a rare liver tumor comprising histologic features of both HCC and CCA. Due to its heterogeneous nature, treatment of combined HCC-CCA is a significant clinical challenge and prognosis remains poor. Therefore, further understanding of the tumor biology underlying the individual subtypes of this mixed tumor is required to improve treatment stratification and optimize treatment strategies. This study sought to identify altered epigenetic regulation and gene expression patterns in the individual components of combined HCC-CCA. Formalin fixed paraffin embedded (FFPE) tumor specimens from 9 patients diagnosed with combined HCC-CCA were utilized in this study. Hematoxylin and eosin (H&E) staining was performed for each sample, and regions representative of the individual HCC and CCA components were delineated. Adjacent unstained slides were cut and dissected to separate HCC and CCA components. DNA and RNA extraction was performed for each sample for DNA methylation (n = 7 HCC and 7 CCA) and gene expression (n = 7 HCC and 8 CCA) profiling via reduced representation bisulfite sequencing (RRBS) and RNA-seq, respectively. Samples did not cluster by tumor type when comparing genome-wide DNA methylation or gene expression patterns. Of the 5 patients with DNA methylation data available for both subtypes, 4 clustered by patient as opposed to cancer subtype, suggesting similar epigenetic regulatory patterns arising from development in the same microenvironment and genetic background. Differential analysis resulted in the identification of 57 differentially expressed genes (DEGs) and 808 differentially methylated regions (DMRs) between the HCC and CCA subtypes. Genes associated with DMRs were associated with Wnt signaling, voltage-gated channels, metal binding, and cellular regulation. Finally, increased expression of several genes previously implicated in tumor aggressiveness, prognosis, and treatment responses were identified. These results highlight the potential importance of accounting for underlying HCC and CCA tumor biology when determining the optimal course of treatment for this deadly disease.