Unraveling the immunogenetic architecture of cholangiocarcinoma

Cholangiocarcinoma (CCA) is a type of biliary tract cancer (BTC) that remains notoriously aggressive, characterized by profound heterogeneity and conferring a dismal prognosis. Recent advances in genomic profiling have broadened our understanding of its pathogenesis, uncovered targetable tumor somatic alterations, and ushered in an era of precision medicine. These efforts have identified actionable drivers such as FGFR2 fusions, IDH1 mutations, BRAF V600E, and HER2 amplification, for which effective targeted therapies are now available in advanced CCA.1 Against this backdrop, the immunogenetic architecture of CCA and its clinical subtypes has remained relatively poorly defined.2 In this issue of Hepatology, Han et al.3 reported the first large-scale genome-wide association study (GWAS) of CCA and its clinical subtypes, analyzing 2366 cases and 11,750 controls of European ancestry to systematically map inherited susceptibility to this aggressive malignancy. The authors identified 5 new single-nucleotide polymorphisms (SNPs) reaching genome-wide significance (p<5×10−8) across CCA subtypes and 12 additional loci with suggestive associations (p<5×10−7), underscoring the heterogeneous and pleiotropic genetic architecture of this malignancy. One of the most notable insights from this work is the distinctly different patterns of susceptibility variants between primary sclerosing cholangitis (PSC) and non-PSC-related CCA. PSC-related CCA seems to be a largely human leukocyte antigen (HLA)–mediated, genetically distinct subtype (odds ratio 3.5). In stratified analyses, the main genome‑wide significant signal, rs2395184 in the HLA region on chromosome 6, was detected exclusively in PSC‑related cases. Moreover, PSC‑associated tumors are predominantly extrahepatic, implying that this immunogenetic predisposition manifests most strongly in the extrahepatic biliary epithelium. Another genome‑wide significant SNP, rs535777, located near HLA‑DRB1 and HLA‑DQA1 at 6p21.32, was strongly associated with overall CCA risk. Additional signals are associated with extrahepatic but not intrahepatic tumors, highlighting a genomic difference based on anatomical location. As for non‑PSC-related CCA, a rare intronic variant, rs142674434, in THSD7A at 7p21.3 was identified as a susceptibility locus. Although recent studies have shown that pathogenic germline alterations in BRCA1/2, ATM, and BAP1 genes may play a role in BTC susceptibility,4 most cases of CCA have been considered sporadic and shaped mainly by environmental and polygenic risk factors. In this study, the discovery of risk variants clustered around HLA-DRB1/DQA1/DRA/DRB5 suggests that certain HLA class II alleles and their peptides are central to PSC-related CCA pathogenesis. HLA has long played a key role in immune dysregulation, cancer immunoediting, and contributes to the tumor microenvironment.5 The enrichment of susceptibility signals in the HLA locus provides mechanistic support to the long-held model in which chronic immune activation and inflammation in the biliary tree lead to malignant transformation.6 This concept is further supported by the study’s genetic data that mirror long‑standing clinical observations linking CCA to immune‑mediated inflammatory disorders, particularly inflammatory bowel disease.7 Furthermore, with Mendelian randomization analysis, Han and colleagues also identified causal links between metabolic traits such as high body mass index, triglyceride levels, and fasting insulin with an increased susceptibility to CCA. These signals reinforce that CCA often arises from a background of chronic inflammation and metabolic dysfunction, complementing epidemiological data showing that metabolic syndrome and metabolic dysfunction–associated fatty liver disease (MAFLD) are associated with a higher risk of CCA.8 Understanding the pathways that connect immune dysregulation, metabolic dysfunction, to CCA carcinogenesis can open avenues for precision prevention. Aggressive metabolic risk modification in MAFLD patients and potential immunomodulatory approaches in high-risk PSC cohorts warrant further investigation. Unfortunately, this cohort is confined to European ancestry, limiting generalizability to East Asian populations where CCA burden is highest. Liver fluke-related CCA, which remains a major etiological subtype in parts of Asia and drives a distinct set of genetic mutations and biliary microbiome, is grossly underrepresented.9,10 Whether the HLA risk variants described here are also relevant to fluke‑related CCA remains unknown and will require dedicated studies in endemic regions. This highlights the need to expand genomic studies to more diverse populations to fully capture the global genomic landscape of CCA. Despite these exciting developments, there is still considerable work to do before insights from this study can be translated into tangible clinical benefits. Potential future areas of research include the development of HLA-based polygenic risk scores to stratify the risk of CCA in patients with PSC to identify those who would benefit from more intensified surveillance, or even enrollment in chemoprevention trials in the future.11 Another interesting angle would be whether specific HLA risk variants are linked to different responses to immunotherapy or other targeted treatments.12 By pinpointing HLA class II alleles and their peptides that confer high risk for PSC-related CCA, this potentially provides a genetic framework for antigen discovery that could pave the way for designing personalized HLA neoantigen or mRNA vaccines, a novel treatment avenue that is still in its infancy, but which has already demonstrated feasibility and promise in isolated case reports and small trials for biliary tract cancer.13,14 This GWAS study delineates the germline architecture of CCA susceptibility, identifying PSC-related CCA as an HLA-driven entity, highlighting the genetic heterogeneity between intrahepatic and extrahepatic tumors, and identifying metabolic risk factors that predispose to CCA development. Taken together, these findings move the field beyond purely descriptive epidemiology and provide a framework for risk prediction, mechanistic insight, and prevention. The challenge now is to translate these genetic insights into actionable prevention strategies, refine surveillance models, establish druggable targets, and ultimately pave the way into a comprehensive precision medicine framework.