Abstract 3944: Characterization of molecular heterogeneity in hepatocellular carcinoma: Trunk and branch drivers

Introduction: Molecular heterogeneity occurs in Hepatocellular carcinoma (HCC), but its implications in clinical decision-making are unknown. The clonal evolution model explains that trunk alterations arise at early stages and are shared by all malignant cells, whereas branch alterations occur in subclonal tumoral cells. We aim to characterize the genomic landscape of HCC through the identification of trunk driver alterations and the study of its distribution in intra- and inter-tumor heterogeneity. Methods: 153 HCC samples representing the multiple steps of hepatocarcinogenesis were analyzed by deep targeted-sequencing covering exonic and promoter regions of the most frequently mutated drivers in HCC. Genes mutated in early lesions [39 dysplastic nodules and 54 early HCCs (eHCC) defined as <2cm, without satellites or vascular invasion] were classified as candidate trunk genes. Candidate trunk genes were further explored in two additional cohorts: a) intra-tumor heterogeneity cohort: 42 tumor regions of 21 tumors >4cm (2-3 regions/tumor); and 2) inter-tumor heterogeneity cohort: 39 tumors from 17 patients with multinodular lesions (2-3 nodules/patient). Transcriptome and copy-number variations (CNVs) were analyzed using expression and SNP arrays, respectively. Results: A total of 46 mutations were identified in the cohort of early lesions. Average number of mutations and CNV aberrations were higher in eHCCs than in dysplastic nodules [1.1 vs 0.5, mutations/patient (p=0.03), and 8% vs 0.6% of aberrant chromosomal arms (p<0.0003), respectively]. Overall, 72% (23/32) of the sequenced eHCCs presented at least 1 trunk mutation, being TERT, TP53 and CTNNB1 the most frequent (21/23, 91%). In the intra-tumor heterogeneity cohort, 81% (17/21) tumors showed at least 1 shared mutation in TERT, TP53 and/or CTNNB1 between different tumor regions (trunk drivers). In the inter-tumor heterogeneity cohort, the similarity of the CNV-profile of multinodular tumors was used to classify them as clonal (intra-hepatic metastasis) or non-clonal (synchronic tumors). 6/17 (35%) of patients harbored clonal tumors according to their CNV profiles (Pearson p<0.05). Clonality classification was further confirmed by gene expression-based hierarchical clustering. 82% (9/11) of the sequenced clonal tumors shared TERT, TP53 and/or CTNNB1 as trunk alterations. In contrast, no trunk mutations were shared across non-clonal tumors. Conclusions: TERT, TP53 and CTNNB1 are trunk drivers mutated in early HCC tumors that remained as trunk aberrations across different regions of the same tumor and between primary and metastatic nodules. These mutations are early trunk drivers that can be captured with single biopsies and could represent ideal therapeutic targets in the future. Citation Format: Sara Torrecilla, Daniela Sia, Andrew N. Harrington, Zhongyang Zhang, Genis Camprecios, Agrin Moeini, Toffanin Sara, Isabel Fiel, Ke Hao, Monica Higuera, Laia Cabellos, Helena Cornella, Milind Mahajan, Yujin Hoshida, Augusto Villanueva, Sander Florman, Myron Schwartz, Josep Llovet. Characterization of molecular heterogeneity in hepatocellular carcinoma: Trunk and branch drivers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3944. doi:10.1158/1538-7445.AM2017-3944