Abstract
Introduction
Cholangiocarcinomas (CCA) are epithelial malignancies categorized as intrahepatic cholangiocarcinoma (iCCA) or extrahepatic cholangiocarcinoma (eCCA). 20-30% of iCCA and more than 65% of eCCA are unresectable. Of the resectable tumors, 5-year survival is ~30% for both cancers. Currently, they are managed using the same treatment algorithm, but studies show different possible gene targets. In this study, we aim to identify clinical and molecular differences between iCCA and eCCA to better differentiate tumor subtypes and potential treatment targets.
Methods
The Cancer Genome Atlas (TCGA) cholangiocarcinoma dataset was accessed through cBioPortal and cases were classified as iCCA (n=384) or eCCA (n=420). Clinicopathologic characteristics, including tumor stage, grade, recurrence, and overall survival, and genomic characteristics, including tumor mutational burden (TMB), fraction of the genome altered (FGA), and gene-specific alterations, were compared between groups using the cBioPortal analytical platform. Statistical analyses were performed to evaluate differences between iCCA and eCCA. A two-sided P value of 0.05 was considered significant.
Results
Compared with eCCA, iCCA was characterized by larger mean tumor size (p=10-10), higher mean FGA (p=9.807×10⁻⁶), and higher mean BMI (p=0.002), as well as more frequent stage IV disease at diagnosis (p=0.0397), poorer tumor differentiation (p=4.239x10-3), multifocal disease (p=10-10), and underlying cirrhosis (p=4.8x10-4). Compared with iCCA, eCCA had a higher TMB (p=3.372×10⁻⁴). Genomic alterations enriched in iCCA include BICC1, IDH1, BAP1, IDH2, FGFR2, NRBF2, and FOXP1. Those enriched in eCCA include TP53, SMAD4, KRAS, NSD2, BRCA2, STK11, and AMER1. Survival was not different between iCCA and eCCA.
Conclusion
iCCA is harder to detect, thus allowing time to dedifferentiate and progress in stage, causing multifocal disease and increased FGA. Genes associated with iCCA and eCCA are mutually exclusive. Currently, three FDA-approved, targeted therapies for CCA act on FGFR2 and IDH1, which are associated with iCCA. No therapy acts on genes associated with eCCA. If targeted therapies progress, genomic diagnosis of CCA could lead to better patient outcomes. iCCA and eCCA are clinically and genomically different cancers. Their mutually exclusive genes can be used diagnostically, improving treatment for patients by using potential targeted therapies. Moving forward, mutant-vs-wild type gene analysis will cross-examine mutations with clinical outcomes.