Molecular Classification of Hepatocellular Adenoma Associates With Risk Factors, Bleeding, and Malignant Transformation

BACKGROUND & AIMS: Hepatocellular adenomas (HCAs) are benign liver tumors that can be assigned to molecular subtypes based on inactivating mutations in hepatocyte nuclear factor 1A, activating mutations in beta-catenin, or activation of inflammatory signaling pathways. We aimed to update the classification system for HCA and associate the subtypes with disease risk factors and complications. METHODS: We analyzed expression levels of 20 genes and sequenced exon regions of 8 genes (HNF1A, IL6ST, CTNNB1, FRK, STAT3, GNAS, JAK1, and TERT) in 607 samples of 533 HCAs from 411 patients, collected from 28 centers mainly in France from 2000 and 2014. We performed gene expression profile, RNA sequence, whole-exome and genome sequence, and immunohistochemical analyses of select samples. Molecular data were associated with risk factors, histopathology, bleeding, and malignant transformation. RESULTS: Symptomatic bleeding occurred in 14% of the patients (85% of cases were female, median age, 38 years); 7% of the nodules were borderline between HCA and hepatocellular carcinoma, and 3% of patients developed hepatocellular carcinoma from HCA. Based on molecular features, we classified HCA into 8 subgroups. One new subgroup, composed of previously unclassified HCA, represented 4% of HCAs overall and was associated with obesity and bleeding. These tumors were characterized by activation of sonic hedgehog signaling, due to focal deletions that fuse the promoter of INHBE with GLI1. Analysis of genetic heterogeneity among multiple HCAs, from different patients, revealed a molecular subtype field effect; multiple tumors had different mutations that deregulated similar pathways. Specific molecular subtypes of HCA associated with various HCA risk factors, including imbalances in estrogen or androgen hormones. Specific molecular subgroup of HCA with beta-catenin and sonic hedgehog activation associated with malignant transformation and bleeding, respectively. CONCLUSIONS: Using sequencing and gene expression analyses, we identified a subgroup of HCA characterized by fusion of the INHBE and GLI1 genes and activation of sonic hedgehog pathway. Molecular subtypes of HCAs associated with different patients' risk factors for HCA, disease progression, and pathology features of tumors. This classification system might be used to select treatment strategies for patients with HCA.