Wheeler, David A.

Abstract Background and Aims Although many studies revealed transcriptomic subtypes of HCC, concordance of the subtypes are not fully examined. We aim to examine a consensus of transcriptomic subtypes and correlate them with clinical outcomes. Approach and Results By integrating 16 previously established genomic signatures for HCC subtypes, we identified five clinically and molecularly distinct consensus subtypes. STM (STeM) is characterized by high stem cell features, vascular invasion, and poor prognosis. CIN (Chromosomal INstability) has moderate stem cell features, but high genomic instability and low immune activity. IMH (IMmune High) is characterized by high immune activity. BCM (Beta‐Catenin with high Male predominance) is characterized by prominent β‐catenin activation, low miRNA expression, hypomethylation, and high sensitivity to sorafenib. DLP (Differentiated and Low Proliferation) is differentiated with high hepatocyte nuclear factor 4A activity. We also developed and validated a robust predictor of consensus subtype with 100 genes and demonstrated that five subtypes were well conserved in patient‐derived xenograft models and cell lines. By analyzing serum proteomic data from the same patients, we further identified potential serum biomarkers that can stratify patients into subtypes. Conclusions Five HCC subtypes are correlated with genomic phenotypes and clinical outcomes and highly conserved in preclinical models, providing a framework for selecting the most appropriate models for preclinical studies.

Abstract The completion of The Cancer Genome Atlas (TCGA) project for colorectal cancer (CRC) has enabled a new, focused phase of sequencing tumor samples for which genome-wide genetic, epidemiological, clinical and lifestyle data have been collected. By combining somatic mutational profiles with these aforementioned data, we may identify and report effective prevention and treatment approaches for a broader population of individuals. The advent of targeted deep sequencing approaches using DNA obtained from formalin fixed paraffin embedded tissues has made possible the genetic characterization of the large numbers of patients needed to make such an effort relevant. As a first step, we describe a custom gene panel designed from large-scale studies for targeted deep sequencing, and its application to over 4,200 CR tumors, collected by the Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO). This study is designed to identify recurrent somatic mutations and copy number alterations (CNAs) for association testing with germline genetic and lifestyle and environmental risk factors for CRC, and thereby identify relevant approaches to impact cancer prevention. The targeted CRC panel includes 205 genes. Genes were primarily selected as significantly mutated genes identified from the Nurses’ Health Study and Health Professional's Follow-up Study (n = 700), and TCGA (n = 525). We also included 15 genes associated with high penetrance germline mutations and augmented the list to include genes in pathways of somatically altered genes, identified by literature review, from public databases and known to be associated with loss of heterozygosity. For these 205 genes, amplification primers were designed to include all coding regions of transcripts that are listed in the UCSC Genome Browser. For regions with CNAs, the TCGA dataset was analyzed to include regions with greater than or equal to 2 copy focal gains or losses that were found in more than 4 or 3 tumors, respectively. Candidate genes in regions with significant CNAs from the TCGA CRC publication (Nature 2012) were also included. For CNAs, 6 to 12 amplicons were designed for each of the 56 selected regions (32 gains and 24 losses). Additional target regions include: 1) 25 microsatellite loci previously used to identify defective DNA mismatch repair and 212 homoploymer repeats. These include microsatellite loci recommended by the NCI Consensus Panel for identifying MSI; 2) amelogenin (for gender); and 3) Fusobacterium to detect a putative CRC-associated pathogen in tumor biopsies. At the AACR annual meeting, we expect to present results for deep sequencing (∼1,000X) of the first 1,000 CR tumors, including any preliminarily identified pathways and subtypes that may provide the basis for association testing with germline genetic and lifestyle and environmental risk factors needed for inferring better approaches to prevention and treatment of CRC. Citation Format: Syed H. Zaidi, Catie Grasso, Jasmine Mu, Eve Shinbrot, Marios Giannakis, Charles Connolly, Ivan Borozan, Hermann Brenner, Peter Campbell, Andrew Chan, Jenny Chang-Claude, Mengmeng Du, Vincent Ferretti, Amy French, Charles Fuchs, Steven Gallinger, Levi Garraway, Andrea Gsur, Marc Gunter, Tabitha Harrison, Michael Hoffmeister, Li Hsu, Wen-Yi Huang, Jeroen Huyghe, Mathieu Lemire, Elaine Mardis, John McPherson, Polly Newcomb, Lincoln Stein, Wei Sun, Lee Timms, Quang Trinh, David Wheeler, Christina Yung, Niha Zubair, Shuji Ogino, Stephen Thibodeau, Ulrike Peters, Thomas Hudson. Linking the molecular profile of colorectal tumors to germline genetic and environmental risk factors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5221.

Abstract Background: Recent genome-wide association studies (GWAS) have identified 28 susceptibility loci for Lung cancer (LC), 32 loci for smoking behavior, 10 loci for chronic obstructive pulmonary disease (COPD), 63 loci for pulmonary function and related phenotypes, totaling 107 GWAS susceptibility loci (as of November 2014). Given the common variants that have been found to be associated with LC in GWAS, targeted-region exome sequencing provides a cost-effective approach for further investigation of high-priority regions of the genome and has the great potential to identify rare causal variants in GWAS loci. Patients and Methods: Using an extreme phenotype approach, we selected 48 sporadic LC patients reporting heavy smoking histories with the presence of severe COPD in whom environmental factors are considered overwhelming, and 54 familial LC cases from families with at least three first-degree relatives with LC who are likely enriched for genomic signal, to search for the disease-causing rare mutations within the target 107 GWAS loci using exome sequencing data. Results: By focusing on exome profiles of these target 107 loci, we identified two disease-causing rare mutations on 6p21.32 TNXB p. Arg504His and 10q25.1 CCDC147 p.Arg696Cys. The homozygous Arg504His mutation presented in four familial and one sporadic LC cases. Notably, the minor allele frequency (MAF) of this variant in Caucasians from 1000 Genomes is very rare (<0.001). The heterozygous CCDC147 Arg696Cys variant were identified in two familial cases and one sporadic case. The MAF for this variant in the Caucasians from 1000 Genomes is 0.004. Both deleterious changes lead to truncation of the putative proteins, which disrupts the predicted domain and alters the cellular localization. We also observed several other suggestive rare mutations on G7C, TNS1, DBH1, and CHRNA5. Conclusion: Our target exome sequencing results demonstrate novel highly disruptive LC risk-conferring TNXB and CCDC147 mutations. Since the rare variants have appreciable effects on disease risk prediction, families and patients may warrant screening for mutations in TNXB and CCDC147 (if validated) to assess their potential LC risk. Citation Format: Yanhong Liu, Farrah Kheradmand, Michael Scheurer, Caleb Davis, David Wheeler, Edwin Silverman, Shete Sanjay, Spiridon Tsavachidis, Georgina Armstrong, Elena Kupert, Marshall Anderson, Yafang Li, Claudio Pikielny, Joan E. Bailey-Wilson, Ming You, Colette Gaba, Mariza DeAndrade, Diptasri Mandal, Claire Simpson, Susan Pinney, Christopher Amos, Margaret Spitz. Target exome sequencing for disease-causing rare mutations in familial and sporadic lung cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4600. doi:10.1158/1538-7445.AM2015-4600

Abstract Background: Hepatocellular carcinomas (HCCs) are heterogenous primary liver cancers with pleomorphic histologic features. We present a comprehensive genetic, genomic, transcriptomic and proteomic analysis of 203 HCCs in The Cancer Genome Atlas Program (TCGA) on behalf of the TCGA Liver analysis working group to provide a deeper understanding of the molecular features of HCCs, to classify them in a clinically-relevant manner, and to provide a public resource that identifies potential targets for emerging therapies.Results: Mutation analyses confirmed the presence of TERT promoter mutations in 40% of HCCs. The analysis also revealed the first reported germline mutation in the TERT promoter in a 30 year old male, possibly reflecting an inherited predisposition to HCC. TERT promoter mutations were enriched in hepatitis C virus infected subjects. Analyses of hepatitis B virus (HBV)-induced HCCs demonstrated recurrent integrations of HBV in the TERT locus. Thus, HBV integration may substitute for TERT promoter mutations as an alternate mechanism of telomerase activation. Other tumor suppressor genes inactivated by mutation included the cell cycle genes TP53 (32% of cases), RB1 (8%), and PTEN (3%), the Wnt/beta catenin pathway gene AXIN1 (4%), and the chromatin remodeling genes ARID1A (9%), BAP1 (6%), ARID1B (4%), ARID2 (4%), and PBRM1 (3%). The oncogene CTNNB1 was mutated in 27% of HCCs. Mutations were also found in the MTOR pathway genes, MTOR (4%), TSC2 (4%) and TSC1 (3%). The patterns of mutational signatures observed included signatures characteristic of the known carcinogens aristolochic acid and aflatoxin B1. Clustering of mRNA, miRNA, or protein expression, DNA hypermethylation and DNA copy number profiles identified five, four, two, five and three clusters, respectively. Analysis of mRNA expression data identified five groups. In particular, Group 1 was associated with Asian race, higher histologic grade, and micro- and macrovascular invasion. Integrated clustering showed a grouping of HCCs into three superclusters: iCluster 1 had a low frequency of CDKN2A silencing, lack of CTNNB1 mutation, a normal-like methylation pattern, and overexpression of proliferation marker genes; iClusters 2 and 3 had a high frequency of CDKN2A silencing, CTNNB1 mutation, and overexpression of oxidative response genes; iCluster 3 tumors had a higher degree of chromosomal instability, high frequency of 17p loss, and a highly altered methylation profile. Copy number aberrations included gains in the MYC, VEGFA, CCND1, TERT, and MET loci and losses in the TP53, RB1, ERFFI1, CDKN2A, ARID1A and PTEN loci. Discussion: Based on integrated analysis of genome, transcriptome, methylome and proteome we showed that HCCs naturally separate into three to five distinct groups with associated molecular and clinical characteristics. Importantly, we identified a novel germline mutation in the TERT promoter that may be associated with an inherited risk for HCC. Note: This abstract was not presented at the meeting. Citation Format: Lewis R. Roberts, David A. Wheeler. Comprehensive integrative characterization of hepatocellular carcinoma: The TCGA HCC project. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3745. doi:10.1158/1538-7445.AM2015-3745