Cancer Genome Interpreter Research Articles

Whole-Exome Sequencing Reveals Novel Candidate Driver Mutations and Potential Druggable Mutations in Patients with High-Risk Neuroblastoma.

Neuroblastoma is the most prevalent solid tumor in early childhood, with a 5-year overall survival rate of 40-60% in high-risk cases. Therefore, the identification of novel biomarkers for the diagnosis, prognosis, and therapy of neuroblastoma is crucial for improving the clinical outcomes of these patients. In this study, we conducted the whole-exome sequencing of 48 freshly frozen tumor samples obtained from the Biobank. Somatic variants were identified and selected using a bioinformatics analysis pipeline. The mutational signatures were determined using the Mutalisk online tool. Cancer driver genes and druggable mutations were predicted using the Cancer Genome Interpreter. The most common mutational signature was single base substitution 5. MUC4, MUC16, and FLG were identified as the most frequently mutated genes. Using the Cancer Genome Interpreter, we identified five recurrent cancer driver mutations spanning MUC16, MUC4, ALK, and CTNND1, with the latter being novel and containing a missense mutation, R439C. We also identified 11 putative actionable mutations including NF1 Q1798*, Q2616*, and S636X, ALK F1174L and R1275Q, SETD2 P10L and Q1829E, BRCA1 R612S, NOTCH1 D1670V, ATR S1372L, and FGFR1 N577K. Our findings provide a comprehensive overview of the novel information relevant to the underlying molecular pathogenesis and therapeutic targets of neuroblastoma.

Genomic landscape of early-stage prostate adenocarcinoma in Mexican patients: an exploratory study

BackgroundHealth disparities have been highlighted among patient with prostate adenocarcinoma (PRAD) due to ethnicity. Mexican men present a more aggressive disease than other patients resulting in less favorable treatment outcome. We aimed to identify the mutational landscape which could help to reduce the health disparities among minority groups and generate the first genomics exploratory study of PRAD in Mexican patients.MethodsParaffin-embedded formalin-fixed tumoral tissue from 20 Mexican patients with early-stage PRAD treated at The Instituto Nacional de Cancerología, Mexico City from 2017 to 2019 were analyzed. Tumoral DNA was prepared for whole exome sequencing, the resulting files were mapped against h19 using BWA-MEM. Strelka2 and Lancet packages were used to identify single nucleotide variants (SNV) and insertions or deletions. FACETS was used to determine somatic copy number alterations (SCNA). Cancer Genome Interpreter web interface was used to determine the clinical relevance of variants.ResultsPatients were in an early clinical stage and had a mean age of 59.55 years (standard deviation [SD]: 7.1 years) with 90% of them having a Gleason Score of 7. Follow-up time was 48.50 months (SD: 32.77) with recurrences and progression in 30% and 15% of the patients, respectively. NUP98 (20%), CSMD3 (15%) and FAT1 (15%) were the genes most frequently affected by SNV; ARAF (75%) and ZNF419 (70%) were the most frequently affected by losses and gains SNCA’s. One quarter of the patients had mutations useful as biomarkers for the use of PARP inhibitors, they comprise mutations in BRCA, RAD54L and ATM. SBS05, DBS03 and ID08 were the most common mutational signatures present in this cohort. No associations with recurrence or progression were identified.ConclusionsThis pilot study reveals the mutational landscape of early-stage prostate adenocarcinoma in Mexican men, providing a first approach to understand the mutational patterns and actionable mutations in early prostate cancer can inform personalized treatment approaches and reduce the underrepresentation in genomic cancer studies.

INTEGRATIVE ANALYSIS OF DIFFERENTIALLY REGULATED PROTEINS IN MENINGIOMA

Abstract AIMS Meningiomas are common primary intracranial tumours that arise from meninges. loss of chromosome 22q is common genetic event observed in meningiomas. About 50% of meningiomas harbour mutations in NF2 gene that resides on chromosome 22. Moreover, frequent mutations have been observed in AKT1, KLF4, TRAF7, and SMO genes. Meningiomas are divided into different genomic subgroups that express unique proteomic profiles. This study aims to investigate genomic background of meningioma and to determine the impact of mutations on underlying protein expression to identify druggable targets. METHOD Total 96 meningioma samples were sequenced using Illumina TruSight Oncology 500 panel. The raw sequence data was analyzed using TSO500 App. The variant data was annotated using Cancer Genome Interpreter and ANNOVAR. Clinical enrichment analysis was performed to investigate common pathways. Genomic data was correlated with LC-MS based proteomic data to determine variants effect on downstream processes. RESULTS Results showed an average of 1083 total and 7 driver mutations per sample. NF2 gene was found among highly mutated genes in tumour samples. Almost all NF2 mutations were frameshift, splice site and nonsense mutations, previously reported to be associated with high rate of multiple and recurring meningiomas. Additionally, driver mutations were identified in known meningioma genes including AKT1, KLF4, and TRAF7 along with previously unpublished mutations in ERBB2, MET, ARID5B, etc. The results correlate with tumour location, grade, and histopathological subtype. CONCLUSIONS The study identified common as well as unique driver mutations in meningioma samples. After functional validation, novel driver mutations could further enhance personalized treatment for meningioma patients.

Comparison of actionable events detected in cancer genomes by whole-genome sequencing, in silico whole-exome and mutation panels

BackgroundNext-generation sequencing is used in cancer research to identify somatic and germline mutations, which can predict sensitivity or resistance to therapies, and may be a useful tool to reveal drug repurposing opportunities between tumour types. Multigene panels are used in clinical practice for detecting targetable mutations. However, the value of clinical whole-exome sequencing (WES) and whole-genome sequencing (WGS) for cancer care is less defined, specifically as the majority of variants found using these technologies are of uncertain significance.Patients and methodsWe used the Cancer Genome Interpreter and WGS in 726 tumours spanning 10 cancer types to identify drug repurposing opportunities. We compare the ability of WGS to detect actionable variants, tumour mutation burden (TMB) and microsatellite instability (MSI) by using in silico down-sampled data to mimic WES, a comprehensive sequencing panel and a hotspot mutation panel.ResultsWe reveal drug repurposing opportunities as numerous biomarkers are shared across many solid tumour types. Comprehensive panels identify the majority of approved actionable mutations, with WGS detecting more candidate actionable mutations for biomarkers currently in clinical trials. Moreover, estimated values for TMB and MSI vary when calculated from WGS, WES and panel data, and are dependent on whether all mutations or only non-synonymous mutations were used. Our results suggest that TMB and MSI thresholds should not only be tumour-dependent, but also be sequencing platform-dependent.ConclusionsThere is a large opportunity to repurpose cancer drugs, and these data suggest that comprehensive sequencing is an invaluable source of information to guide clinical decisions by facilitating precision medicine and may provide a wealth of information for future studies. Furthermore, the sequencing and analysis approach used to estimate TMB may have clinical implications if a hard threshold is used to indicate which patients may respond to immunotherapy.

Abstract 2808: cfDNA multi-omics profiling and tissue of origin predictions in cancers of unknown primary

Abstract Cancers of Unknown Primary (CUP) comprise a heterogeneous group of metastatic cancers where the primary site is undetectable. It accounts for 2% of all new UK cancer diagnoses and is the 6th leading cause of cancer death. Favourable subsets of CUP exist, based on clinico-pathological features, and although these patients have improved clinical outcomes, they make up only 20% of patients with CUP. The evolution of molecular profiling and tissue-of-origin (TOO) classifiers has the potential to transform diagnosis and management of patients with CUP. Reliable molecular profiling requires good quality and quantity tumour tissue material, often scarce in patients with CUP. Liquid biopsies offer an alternative, minimally invasive approach. We developed a novel cell-free DNA (cfDNA)-based multi-omics assay combining methylation profiling with mutation and copy number alterations (CNA) detection, to enable TOO prediction and biomarker discovery to aid treatment stratification. Methods: Two novel library preparation methods were applied to 41 CUP cfDNA samples, 80 non-cancer controls and 80 cfDNA samples from known tumour types (breast, lung, colorectal cancer): 1) Targeted (641 gene panel) next generation sequencing (NGS); 2) A multiplex genome-wide methylation capture assay (T7-MBDSeq). Both sequencing outputs can estimate CNA. Cancer Genome Interpreter (CGI) was used to evaluate the potential actionability of molecular alterations detected. A novel TOO classifier, generated utilising in silico dilution mixture sets derived from The Cancer Genome Atlas (TCGA) methylation data, demonstrated an AUROC of 0.96 across 15 different cancer groups. This was applied to cfDNA methylation data from known tumour types to test performance accuracy and subsequently applied to CUP cfDNA samples. Results: Multi-omics profiling from cfDNA was completed for 34/41 patients with CUP. cfDNA mutation profiling and estimated CNA analysis so far revealed >350 alterations across 225 genes. 17/34 (50%) of patients harboured potentially actionable alterations by CGI and, where performed, 68% of tumour mutations were concordant in the cfDNA. cfDNA methylation profiling revealed tumour-specific methylation changes in breast, lung and colorectal cancers. Our preliminary TOO classifier performance in these tumour types had an overall sensitivity and specificity of 85% and 98%, respectively. In patients with CUP, a predicted TOO was made in 26/41 (63%) patients and for 6 patients, where a subsequent primary tumour diagnosis was made, the TOO prediction was correct. Conclusions: cfDNA multi-omics profiling revealed potential therapeutic benefit in 77% of this retrospective CUP cohort and illustrates the potential to re-classify patients to primary tumour types where alternative therapies may improve clinical outcomes. Further multi-omics cfDNA profiling of the CUP cohort and validation in known primary cancer types are ongoing. Citation Format: Alicia-Marie Conway, Alexandra Clipson, Francesca Chemi, Simon Pearce, Katarzyna Kamieniecka, Saba Ferdous, Sophie Richardson, Elaine Kilgour, Matthew G. Krebs, Alastair Kerr, Dominic Rothwell, Natalie Cook, Caroline Dive. cfDNA multi-omics profiling and tissue of origin predictions in cancers of unknown primary [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2808.

DrugCVar: a platform for evidence-based drug annotation for genetic variants in cancer

Targeted therapy for cancer-related genetic variants is critical for precision medicine. Although several databases including The Clinical Interpretation of Variants in Cancer (CIViC), The Oncology Knowledge Base (OncoKB), The Cancer Genome Interpreter (CGI) and My Cancer Genome (MCG) provide clinical interpretations of variants in cancer, the clinical evidence was limited and miscellaneous. In this study, we developed the DrugCVar database, which integrated our manually curated cancer variant-drug targeting evidence from literature and the interpretations from the public resources. In total, 7830 clinical evidences for cancer variant-drug targeting were integrated and classified into 10 evidence tiers. Searching and browsing functions were provided for quick queries of cancer variant-drug targeting evidence. Also, batch annotation module was developed for user-provided massive genetic variants in various formats. Details, such as the mutation function, location of the variants in gene and protein structures and mutation statistics of queried genes in various tumor types, were also provided for further investigations. Thus, DrugCVar could serve as a comprehensive annotation tool to interpret potential drugs for cancer variants especially the massive ones from clinical cancer genomics studies. The database is available at http://drugcvar.omicsbio.info. Supplementary data are available at Bioinformatics online.

DIVIS: Integrated and Customizable Pipeline for Cancer Genome Sequencing Analysis and Interpretation

Next-generation sequencing (NGS) has drastically enhanced human cancer research, but diverse sequencing strategies, complicated open-source software, and the identification of massive numbers of mutations have limited the clinical application of NGS. Here, we first presented GPyFlow, a lightweight tool that flexibly customizes, executes, and shares workflows. We then introduced DIVIS, a customizable pipeline based on GPyFlow that integrates read preprocessing, alignment, variant detection, and annotation of whole-genome sequencing, whole-exome sequencing, and gene-panel sequencing. By default, DIVIS screens variants from multiple callers and generates a standard variant-detection format list containing caller evidence for each sample, which is compatible with advanced analyses. Lastly, DIVIS generates a statistical report, including command lines, parameters, quality-control indicators, and mutation summary. DIVIS substantially facilitates complex cancer genome sequencing analyses by means of a single powerful and easy-to-use command. The DIVIS code is freely available at https://github.com/niu-lab/DIVIS, and the docker image can be downloaded from https://hub.docker.com/repository/docker/sunshinerain/divis.

Next-Generation Sequencing Targeted Chronic Lymphocytic Leukemia Panel: A Pilot Study

Background: Rapid progress in next-generation sequencing (NGS) technologies make it possible to spell out the mutational status, the genetic and epigenetic variability of chronic lymphocytic leukemia (CLL). The identification of driver mutations allows us to expand understanding of the pathogenesis of CLL, to identify prognostic groups and to select potential targets for therapy, contributing to the development and implementation of new targeted drugs and its combinations. Nevertheless, the course of CLL does not always correspond to the existing prognostic risk groups, assessed by "standard" cytogenetic and molecular genetic methods. The NGS technology admits establishing markers of an unfavorable course of the disease. Aim: To assess the mutational status of patients (pts) with CLL using the developed Lymphoid Targeted NGS Panel and to study the possible correlation of the mutational status with the clinical characteristics of the disease. Method: In this prospective study were included 24 pts with CLL: treatment-naïve (n=8) and relapsed/refractory (n=16). The diagnosis of CLL was established according to iwCLL criteria (iwCLL, Hallek et al., 2018) and show only typical immunophenotype. The pts were divided into three prognostic groups according to cytogenetic assay: favorable (n=14), neutral (n=2), and unfavorable (n=8). Although, 20/24 pts were divided into two prognostic groups taking into account the data on the mutational status of the immunoglobulin heavy-chain variable (IGHV) region: favorable (n=8) and unfavorable (n=12). Four patients have no available data on IGHV mutational status. All patients had indications for starting treatment: FCR (n=7), RB (n=6), ibrutinib (n=3), venetoclax (n=1), acalabrutinib (n=5), combination of venetoclax and acalabrutinib (n=2). DNA samples were extracted from peripheral B-cell lymphocytes via the standard phenol-chloroform method. Average reading depth of 1000x is produced on a MiSeq platform (Illumina, USA). The 2% threshold of allele frequency (VAF) was used. The clinical significance of mutations was established using the following databases: COSMIC, ClinVar, gnomAD with application in silico analysis (Cscape, Cancer Genome Interpreter, SNPs&Go). The Lymphoid Panel includes 117 genes, part of which is involved in the main 8 cellular signaling pathways underlying the pathogenesis of CLL. We have completed a pilot NGS study using the developed Lymphoid Targeted Gene Panel on DNA samples of six treatment-naïve pts. Results: Genetic aberrations were identified in all DNA samples using NGS. Somatic mutations were detected in 82.9% of cases, in an amount from 15 to 37. In four pts (4/6) with an unfavorable prognosis (cytogenetics and unmutated IGHV), known pathogenic variants of mutations were identified: JAK3 V722L, NOTCH1 P2514fs*4, IDH2 T352P, TP53 Lys120Glu, BRAF D594G. The existing approach to the interpretation of the results does not allow making an unambiguous conclusion about the clinical significance of variants in the IDH2 and JAK3 genes, despite the known pathogenic effect of the variants. The detected variant of the mutation (Lys120Glu) in the TP53 gene is often associated with the presence of a 17p13 deletion, which was confirmed by the FISH assay and correlated with the unfavorable clinical course of the disease in patient CLL-024. Twenty-two mutations were identified, the pathogenicity of which has not yet been determined, in the amount of 2 to 5 (median=3.5) mutations per patient. It should be noted that two patients (CLL-023, CLL-024) with unfavorable prognosis had mutations both in BCR gene and in NOTCH2 gene of unknown significance. Conclusion: The data obtained from a pilot study demonstrate the possibility of using NGS technology in clinical practice. The assessment of the mutational status of pts with CLL using NGS correlates with the clinical parameters of pts. Considering that there is currently no information about prognostic significances of identified mutations, additional research is required. Disclosures Martynkevich: Pfizer: Honoraria, Speakers Bureau; Novartis: Honoraria, Speakers Bureau; BMS: Honoraria, Speakers Bureau. Shuvaev:Pfizer: Honoraria, Speakers Bureau; BMS: Honoraria, Speakers Bureau; Novartis: Honoraria, Speakers Bureau. Voloshin:Novartis: Honoraria, Speakers Bureau.

Abstract 184: SureSelect sequencing panels and algorithms to detect copy number variations (CNVs), DNA rearrangements, microsatellite Instability and tumor mutational burden (TMB) in FFPE specimens

Abstract Background: The current standard of genomic profiling of cancer tissues relies upon multiple separate technologies to aid in tumor characterization. Here we describe an NGS panel and analysis workflow for cancer samples that can detect single nucleotide polymorphisms (SNPs), insertions and deletions (INDEL), somatic copy number alterations (SCNAs), and translocations (TLs), Tumor Mutation Burden (TMB) and Microsatellite Instability (MSI) in a single assay. Methods: We have developed a SureSelect target enrichment sequencing panels for comprehensive profiling of single nucleotide variations (SNP/INDEL), gene copy number variations (CNV), and DNA translocations (TL) and relevant microsatellite regions in a single assay. The TMB panel gene list was curated based on genes found in cancer gene databases such as Clinical Interpretations of Variants in Cancer (CIViC), Cancer Genome Interpreter (CGIdb), Catalog of Somatic Mutations in Cancer Census (COSMIC), Precision Oncology Knowledge Base (OncoKB). An accompanying data analysis pipeline will provide highly sensitive and accurate detection of variants and allows for the determination TMB and MSI status. Probes have been selected for over 500 genes to detect SCNAs and translocation in several cancer subtypes. The probes are optimized with the SureSelect XTHS protocol, which enables sensitive and accurate detection of rare mutation events within a heterogeneous sample with molecular barcode-mediated error correction. The SCNAs probes target regions in the genome that help in decreasing the noise for SCNAs calling. The data analysis utilizes either a matched normal or standard control DNA to compute read depth log ratios. Aberrant regions of constant ploidy are first identified followed by the application of a variational streaming algorithm on log ratio and SNP allele frequencies to determine the major clones present in the tumor. Translocation detection was verified on lung cancer samples by analyzing split reads across the fusion breakpoints. Results: By diluting samples with known copy-number aberrations into normal DNA, a limit of detection of ⇐2.3 copies/cell was demonstrated, equivalent to the detection of 3 copies at 30% tumor fraction. A similar approach demonstrated an ability to detect EML4-ALK and SLC24A2-ROS1 translocations at 3% allele frequency. Performance of the assay was further evaluated using 100+ FFPE samples harboring rearrangements, gene amplifications, and SNP/Indels. Concordance to the reference methods (FISH, WES, WGS) was >90%. TMB and MSI algorithms were benchmarked using in silico analysis of TCGA data and performance of the assay was evaluated on FFPE samples with orthogonally-determined MSI/TMB status. Conclusions: This work represents an important advancement in the development of a single assay to detect copy number variation, DNA rearrangement and mutations, TMB and MSI status of a FFPE sample. Citation Format: Arjun Vadapalli, Akanksha Khare, Hanjun Shin, Ashutosh Ashutosh, Linus Forsmark, Anne Lucas, Scott Happe, Gilbert Amparo, Carlos Pabon, Jayati Ghosh, Tracy Liu, Jimmy Jin, Mike Ruvolo, Douglas Roberts. SureSelect sequencing panels and algorithms to detect copy number variations (CNVs), DNA rearrangements, microsatellite Instability and tumor mutational burden (TMB) in FFPE specimens [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 184.

Abstract 5469: iCVA- A knowledge-based cancer variation annotation application

Abstract A typical genomic sequencing experiment results in a long list of statistically significant variant candidates among protein-coding genes without any unifying biological theme. This leads to a daunting task of identifying the causal variants and genes to accurately diagnose the disease for clinical or research utility. While a number of downstream tools exist to aid in cancer genome annotation and interpretation, including Clinical Interpretations of Variants in Cancer (CIViC), ClinGen, Database of Curated Mutations (DoCM), Oncotator, and Variant Interpretation for Cancer (VIC), these only provide selective information (somatic/germline/gene-variant specific) or are time/resource consuming. To address this challenge, we developed a software called iCVA using the following methodology: 1) we obtained a priori defined gene sets for different cancer-inducing mechanisms such as pro-oncogenesis, tumor-suppression, DNA repair, angiogenesis, inflammation, metabolism, hypoxia, cell cycle and immune system. 2) genomic variation data were integrated from 1000 Genome, Exome Sequencing Project, Exome Aggregation Consortium (ExAC), ClinVar, dbSNP, GNOMAD, The Cancer Genome Atlas (TCGA), the Catalogue of Somatic Mutations in Cancer (COSMIC) and the International Cancer Gene Census (ICGC), and re-classified into a consensus classification according to ACMG guidelines. 3) Finally, we devised a reporting system to process mutation data from a sequencing experiment to utilize the harmonized pathway and mutational information integrated in a local database to identify and classify cancer-specific gene mutations. The iCVA reporting system classifies mutations as germline or somatic and then sub-classifies these into different cancer-specific mechanisms, including those known to be targetable through existing therapies. To our knowledge, iCVA is the first tool to provide a comprehensive report on cancer variants in a simplified and faster manner, to accelerate the genomic characterization of cancer samples analyzed by high-throughput DNA sequencing. Citation Format: Gaurav Kumar, Melanie Kelly, Paolo Fortina, Adam Ertel. iCVA- A knowledge-based cancer variation annotation application [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5469.

Abstract 3222: The Virtual Molecular Tumor Board of the Variant Interpretation for Cancer Consortium: A systematic gateway connecting cancer genome interpretation and progress in genomic knowledgebases in cancer

Abstract Comprehensive interpretation of a patient's cancer genome is a cross-disciplinary process requiring the expertise of oncologists, molecular pathologists, genetic counselors, genomic scientists, and bioinformaticians. The Variant Interpretation for Cancer Consortium (VICC) has created a non-clinical forum for the variant interpretation process by connecting experts from multiple academic institutions and industry to create a Virtual Molecular Tumor Board (VMTB). During VICC-VMTB sessions, experts retrospectively discuss de-identified, challenging patient cases in the context of genomic interpretation guidelines to evaluate a patient's profile across several distinct components: genomic variants; disease context; therapeutic history; family history; demographic information; genomic assay information; and variant allele frequency. These components are processed through the prism of publically available knowledge and expertise of VMTB members, resulting in a broad spectrum of clinically relevant assertions about the patient's disease that may be of prognostic, diagnostic, or predictive relevance. Through these evaluations, the VICC-VMTB primarily works as a means of illuminating challenges in genomic interpretation and advancing the objectives of domain-specific VICC working groups. Examples of challenges that have been highlighted by the VICC-VMTB forum include: 1) inconsistent disease and therapy ontologies across resources are being addressed by Disease Harmonization & Drug Harmonization working groups, respectively; 2) use of machine-learning and natural language processing into clinical-grade variant interpretation are being addressed by the AI-Assisted Curation working group; 3) aggregation and harmonization of knowledgebase data into a centralized resource is being addressed by the Data Licensing and Variant Harmonization working groups; 4) inefficiencies in knowledgebase search functions are being addressed by the Search working group; 5) determination of accurate prediction of biological consequences of rare variants is being addressed by the In Silico Interpretation working group; and 6) standardization of genomic variant curation is being addressed by the Knowledge Curation and Interpretation Standards working group. These VICC working groups provide dedicated expert assemblies intended to specifically tackle these challenges, further increasing the breadth for which genomic information can be made useful in cancer variant interpretation. The VICC-VMTB has been a pertinent link between cancer genomes and clinical interpretation. We are confident VICC-VMTB will continue to be a gateway of acceleration and progress needed in the field of cancer interpretation to provide accurate and efficient answers for our patients in the future. Citation Format: Beth A. Pitel, Shruti Rao, Catherine Del Vecchio Fitz, Subha Madhavan, Rodrigo Dientsmann, Peter Horak, Ian King, Susan M. Mockus, Gordana Raca, Damian T. Rieke, Peter Rogan, Dmitriy Sonkin, David Tamborero, Ioannis S. Vlachos, Brian Walsh, Jeremy L. Warner, Malachi Griffith, Obi L. Griffith, Debyani Chakravarty, Alex H. Wagner. The Virtual Molecular Tumor Board of the Variant Interpretation for Cancer Consortium: A systematic gateway connecting cancer genome interpretation and progress in genomic knowledgebases in cancer [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3222.

Germline Genetic Findings Which May Impact Therapeutic Decisions in Families with a Presumed Predisposition for Hereditary Breast and Ovarian Cancer.

In this study, we aim to gain insight in the germline mutation spectrum of ATM, BARD1, BRIP1, ERCC4, PALB2, RAD51C and RAD51D in breast and ovarian cancer families from Spain. We have selected 180 index cases in whom a germline mutation in BRCA1 and BRCA2 was previously ruled out. The importance of disease-causing variants in these genes lies in the fact that they may have possible therapeutic implications according to clinical guidelines. All variants were assessed by combined annotation dependent depletion (CADD) for scoring their deleteriousness. In addition, we used the cancer genome interpreter to explore the implications of some variants in drug response. Finally, we compiled and evaluated the family history to assess whether carrying a pathogenic mutation was associated with age at diagnosis, tumour diversity of the pedigree and total number of cancer cases in the family. Eight unequivocal pathogenic mutations were found and another fourteen were prioritized as possible causal variants. Some of these molecular results could contribute to cancer diagnosis, treatment selection and prevention. We found a statistically significant association between tumour diversity in the family and carrying a variant with a high score predicting pathogenicity (p = 0.0003).

Collaborative, Multidisciplinary Evaluation of Cancer Variants Through Virtual Molecular Tumor Boards Informs Local Clinical Practices.

PURPOSEThe cancer research community is constantly evolving to better understand tumor biology, disease etiology, risk stratification, and pathways to novel treatments. Yet the clinical cancer genomics field has been hindered by redundant efforts to meaningfully collect and interpret disparate data types from multiple high-throughput modalities and integrate into clinical care processes. Bespoke data models, knowledgebases, and one-off customized resources for data analysis often lack adequate governance and quality control needed for these resources to be clinical grade. Many informatics efforts focused on genomic interpretation resources for neoplasms are underway to support data collection, deposition, curation, harmonization, integration, and analytics to support case review and treatment planning.METHODSIn this review, we evaluate and summarize the landscape of available tools, resources, and evidence used in the evaluation of somatic and germline tumor variants within the context of molecular tumor boards.RESULTSMolecular tumor boards (MTBs) are collaborative efforts of multidisciplinary cancer experts equipped with genomic interpretation resources to aid in the delivery of accurate and timely clinical interpretations of complex genomic results for each patient, within an institution or hospital network. Virtual MTBs (VMTBs) provide an online forum for collaborative governance, provenance, and information sharing between experts outside a given hospital network with the potential to enhance MTB discussions. Knowledge sharing in VMTBs and communication with guideline-developing organizations can lead to progress evidenced by data harmonization across resources, crowd-sourced and expert-curated genomic assertions, and a more informed and explainable usage of artificial intelligence.CONCLUSIONAdvances in cancer genomics interpretation aid in better patient and disease classification, more streamlined identification of relevant literature, and a more thorough review of available treatments and predicted patient outcomes.

Mutational spectrum and classification of novel mutations in patients with metastatic gastrointestinal stromal tumours.

In total, ~85% of malignant gastrointestinal stromal tumours (GISTs) harbour activating mutations in one of the genes KIT or PDGFRA, while 10-15% of all GISTs have no detectable KIT or PDGFRA mutations, but could have alterations in genes of the succinate dehydrogenase complex or in BRAF, PIK3CA or rarely RAS family genes. The clinical benefit of tyrosine kinase inhibitors, such as imatinib, depends on the GIST genotype, therefore molecular characterization of GIST has a crucial role in overall management of GIST. The aim of the present study was to molecularly characterize a cohort of 70 patients with metastatic GISTs from the Slovenian Cancer Registry (National Cancer Registry) treated between January 2002 and December 2011. Exons 9, 11, 13 and 17 of the KIT gene and exons 12, 14 and 18 of the PDGFRA gene were analysed by direct Sanger sequencing. All KIT/PDGFRA wild-type GISTs were tested for the presence of mutations in hot spot regions of KRAS, NRAS, BRAF, PIK3CA and AKT1 genes. Novel variants were characterized and classified using Cancer Genome Interpreter and according to The American College of Medical Genetics and Genomics/Association for Molecular Pathology guidelines. In total, 60 (85.7%) patients had mutations in KIT and 2 (2.9%) in PDGFRA. Whereas, 8 (11.4%) patients with GIST had no mutation in either of the analysed genes. The majority of GIST cases (n=52) had a mutation in KIT exon 11, where 40 different mutations were detected. Eight of the variants were novel: c.1652_1672del, c.1653_1660delinsAA, c.1665_1672delinsCC, c.1668_1686del, c.1676_1720del, c.1715_1756dup, c.1721_1765dup, and c.1722_1766dup. Mutation frequencies of KIT and PDGFRA genes observed in Slovenian patients are comparable with those in other European populations. In the present group of patients analysed, the most frequently mutated region was exon 11 in the KIT gene, responsible for coding juxtamembrane domain of KIT protein. In this region, eight novel mutations were identified and classified as likely pathogenic driver variants. In addition, the present study identified 6 patients with secondary KIT mutation and 1 patient with double mutant GIST, who had two different mutations in PDGFRA exon 14.

Holistic cancer genome profiling for every patient.

Technological advances in the ability to read the human genome have accelerated the speed of sequencing, such that today we can perform whole genome sequencing (WGS) in one day. Until recently, genomic studies have largely been limited to seeking novel scientific discoveries. The application of new insights gained through cancer WGS into the clinical domain, have been relatively limited. Looking ahead, a vast amount of data can be generated by genomic studies. Of note, excellent organisation of genomic and clinical data permits the application of machine-learning methods which can lead to the development of clinical algorithms that could assist future clinicians and genomicists in the analysis and interpretation of individual cancer genomes. Here, we describe what can be gleaned from holistic whole cancer genome profiling and argue that we must build the infrastructure and educational frameworks to support the modern clinical genomicist to prepare for a future where WGS will be the norm.

Detection of actionable variants in various cancer types reveals value of whole-genome sequencing over in-silico whole-exome and hotspot panel sequencing

Abstract Background Declining costs of next-generation sequencing technologies has led to their more widespread use for the identification of targetable mutations. Nonetheless, while multigene panels are being assimilated into clinical practice, the use of whole-exome and whole-genome sequencing (WES and WGS) remains limited, as its added value is unclear in the clinic. Methods In our study, we have used the Cancer Genome Interpreter (CGI) to identify the actionable variants that are detected by WGS and compared this to in-silico down-sampled regions of WES and a hotspot panel. Results We show that within some tumour types, WGS is invaluable in identifying actionable variants that involve copy number defects and structural aberrations. Moreover, WGS successfully identifies many resistance biomarkers that would have otherwise been missed by WES and panels, particularly in the breast cancer dataset studied. WGS is more valuable of the detection of actionable variants in some tumour types, whereas hotspot panels perform well for other tumour types. Additionally, we show that a tumour type-agnostic approach to selecting genomic biomarker-based drug allocation increases the number of possible FDA-approved and clinical trial-based drug prescriptions for any given genomic biomarker in the patient cohorts studied. Conclusions Taken together, we show that WGS has the potential to impact patient care by identifying more targetable mutations and therefore expand drug options available for the patient. Furthermore, we illustrate the potential of tumour type-agnostic drug repurposing in nine cancer datasets. Legal entity responsible for the study QIMR Berghofer Medical Research Institute. Funding QIMR Berghofer Medical Research Institute. Disclosure All authors have declared no conflicts of interest.

A combination of resistance mechanisms is frequent in non-small cell lung cancer (NSCLC) that progressed to EGFR tyrosine kinase inhibitors (TKIs)

Abstract Background Despite a primary benefit, almost all patients (pts) develop acquired resistance to EGFR TKIs. Some mechanisms were related to the selective pressure of the different generations (G) EGFR TKIs, but the complete alteration profile at resistance remains to be elucidated. Methods Metastatic NSCLC pts who performed a tissue and plasma biopsy at EGFR TKI progression, and optionally before, were identified from the prospective MATCH-R trial (NCT02517892). Targeted next-generation sequencing (NGS, 74 to 82 genes), comparative genomic hybridization (CGH), whole-exome sequencing (WES), and RNA sequencing (RNA-Seq) were performed on the tissue. All molecular oncogenic alterations were classified as definitive/potential resistance mechanism or concomitant genetic alterations using OncoKB and Cancer Genome Interpreter. Results Among 62 pts included, tissue analysis at TKI progression was successful with NGS (94%), CGH (76%) and WES+RNA-Seq (53%). In the post 1st-2nd G EGFR TKI cohort (31 pts) the most frequent resistance alteration was p.T790M (n = 23, 74%). EGFR amplification (amp) was found in 7 (23%) pts and off-target mutations (mut) at resistance in 9 (29%, 2 KRAS, 2 JAK2, 1 PIK3CA, and 1 MET). Median PFS to EGFR TKI was 17 months (mo) for the pts with p.T790M as the only resistance mechanism vs 8 mo for the other resistance profiles (HR 2.42 [95% CI, 1.04-5.65]; p = 0.028). In the post 3rd G cohort (31 pts), acquired resistance alterations were mostly on-target (n = 12, 39%): 9 p.C797S, 2 p.L718Q and 1 EGFR amp. Off-target resistance alterations were found in 11 (35%) tumors: 5 fusions involving oncogenes (16%, RET, BRAF, ALK, and FGFR3), 2 BRAF p.V600E (6%), then MET amp, HER2 amp, KRAS mut, and PIK3CA mut in 1 pts (3%). p.T790M loss (54%) was associated with a shorter time to osimertinib discontinuation vs maintained (median 13 vs 22 mo; HR 2.16 (95% CI, 1.00-4.73); p = 0.046). More than one resistance mechanism was found in 12 (39%) and 14 (45%) pts in the post 1st-2nd and 3rd G cohort, respectively. Conclusions Genetic alteration landscape at EGFR TKIs progression is complex and heterogeneous since more than one resistance alteration was recognized in 42% of pts. Combination strategy might be needed. Legal entity responsible for the study Gustave Roussy, Cancer Campus. Funding Gustave Roussy, Cancer Campus. Disclosure Y. Loriot: Honoraria (self): Roche, MSD, Astellas, Janssen, AstraZeneca, BMS, Seattle Genetics, Sanofi, Clovis, Pfizer; Honoraria (institution): Roche, MSD, Astellas, Janssen, AstraZeneca, BMS, Seattle Genetics, Sanofi, Clovis, Incyte, Pfizer; Research grant / Funding (institution): Roche, MSD, Sanofi; Travel / Accommodation / Expenses: Roche, MSD, Janssen, AstraZeneca, Seattle Genetics; Licensing / Royalties: Pending patent (USA 62/455211, Europe 17209098.7). D. Planchard: Advisory / Consultancy: AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim, Celgene, Daiichi Sankyo, Eli Lilly, Merck, MedImmune, Novartis, Pfizer, prIME Oncology, Peer CME, Roche; Honoraria (self): AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim, Celgene, Eli Lilly, Merck, Novartis, Pfizer, prIME Oncology, Peer CME, Roche; Research grant / Funding (institution): AstraZeneca, Bristol-Myers Squibb, AbbVie, Boehringer Ingelheim, Eli Lilly, Merck, Novartis, Pfizer, Roche, MedImmune, Sanofi-Aventis, Taiho Pharma, Novocure, Daiichi Sankyo; Travel / Accommodation / Expenses: AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim , Roche, Merck, Novartis, prIME Oncology, Pfizer. S. Michiels: Advisory / Consultancy, Punctual statistical advice : IDDI Belgium, Janssen Cilag France; Advisory / Consultancy, Data and Safety Monitoring Member: Hexal, JJ Principal/sub-Investigator of Clinical Trials for AbbVie, Agios Pharmaceuticals, Amgen, Argen-X; Research grant / Funding (institution): AstraZeneca, BMS, Boehringer Ingelheim, Janssen-Cilag, Merck, Novartis, Pfizer, Roche, Sanofi; Non-remunerated activity/ies: AstraZeneca, Bayer, BMS, Boehringer Ingelheim, Johnson & Johnson, Lilly, MedImmune, Merck, NH TherAGuiX, Pfizer, Roche. J-C. Soria: Advisory / Consultancy: AstraZeneca, Astex, Clovis, GSK, GamaMabs, Lilly, MSD, Mission Therapeutics, Merus, Pfizer, PharmaMar, Pierre Fabre, Roche/Genentech, Sanofi, Servier, Symphogen, and Takeda; Full / Part-time employment: AstraZeneca; Shareholder / Stockholder / Stock options: AstraZeneca and Gritstone. F. Andre: Research grant / Funding (institution): NVS, AZ, Roche, Daiichi, Lilly, Pfizer. G. Vassal: Advisory / Consultancy, Non-remunerated activity/ies: AstraZeneca, Bayer, BMS, Celgene, Incyte, Ipsen, Lilly, MSD, Merck, Pierre-Fabre, Novartis, Pfizer, Roche/Genentech, Servier, Takeda. B. Besse: Research grant / Funding (institution): AbbVie, Amgen, AstraZeneca, Biogen, Blueprint Medicines, BMS, Celgene, Eli Lilly, GSK, Ignyta, Ipsen, Merck KGaA, MSD, Nektar, Onxeo, Pfizer, PharmaMar, Sanofi, Spectrum Pharmaceuticals, Takeda, Tiziana Pharma. All other authors have declared no conflicts of interest.

8PD - Automatic interpretation of cancer genomes creates the largest repository of tumour genetic driver events

BackgroundTumor genome sequencing is becoming widely available in the clinical setting. However, the interpretation of tumor somatic variants remains an important challenge to implement precision cancer medicine. MethodsHere, we present IntOGen, a platform aimed for tumor genome interpretation. The first component of the platform is an automatic pipeline for cancer driver gene identification across cohorts. This pipeline implements several bioinformatics methods to detect driver genes, on the basis of signals of positive selection in their mutational pattern across tumor cohorts. It also annotates several oncogenomic features that shed light onto the role of driver genes in the development of each malignancy. The second component of the platform leverages these features to identify the somatic mutations that drive the tumor of each patient providing additional information about the possible effect of driver mutations on treatment response. ResultsWe have applied the framework to more than 50,000 tumors across more than 60 tumors types creating the largest repository of tumor genetic driver events. This comprehensive oncogenomics repository is the third component of IntOGen. ConclusionsAll the results and framework are available online to the cancer genomics research community at www.intogen.org, which we envision can support a broad range of oncology use cases. Legal entity responsible for the studyInstitute for Research in Biomedicine (IRB, Barcelona). FundingMinisterio de Educación y Ciencia de España (SAF-2015-R-66084 and RTI2018-094095-B-I00) Instituto Nacional de Bioinformática (INB) - PT17/0009/0013. DisclosureAll authors have declared no conflicts of interest.

Whole-genome sequencing of triple-negative breast cancers in a population-based clinical study.

SummaryWhole genome sequencing (WGS) brings comprehensive insights to cancer genome interpretation. To explore clinical value of WGS, we sequenced 254 triple negative breast cancers (TNBC) with associated treatment and outcome data collected between 2010-2015 via the population-based Sweden Cancerome Analysis Network-Breast (SCAN-B) project (ClinicalTrials.gov ID:NCT02306096). Applying the HRDetect mutational-signature-based algorithm to classify tumors, 59% were predicted to have Homologous-recombination-repair deficiency (HRDetect-high): 67% explained by germline/somatic mutations of BRCA1/BRCA2, BRCA1 promoter hypermethylation, RAD51C hypermethylation or biallelic loss of PALB2. A novel mechanism of BRCA1 abrogation was discovered via germline SINE-VNTR-Alu retrotransposition. HRDetect provided independent prognostic information, with HRDetect-high patients having better outcome on adjuvant chemotherapy for invasive disease-free survival (Hazard Ratio, HR=0.42, 95% confidence interval, CI=0.2-0.87), and distant relapse-free interval (HR=0.31, CI=0.13-0.76) compared to HRDetect-low, regardless of whether a genetic/epigenetic cause was identified. HRDetect-intermediate, some possessing potentially targetable biological abnormalities, had poorest outcomes. HRDetect-low cancers also had inadequate outcomes: ~4.7% were mismatch-repair-deficient - another targetable defect, not typically sought; and was enriched for (but not restricted to) PIK3CA/AKT1 pathway abnormalities. New treatment options need to be considered for now-discernible HRDetect-intermediate and HRDetect-low categories. This population-based study advocates for WGS of TNBC to better inform trial stratification and improve clinical decision-making.

Comparative Analysis of Public Knowledge Bases for Precision Oncology.

Precision oncology depends on the availability of up-to-date, comprehensive, and accurate information about associations between genetic variants and therapeutic options. Recently, a number of knowledge bases (KBs) have been developed that gather such information on the basis of expert curation of the scientific literature. We performed a quantitative and qualitative comparison of Clinical Interpretations of Variants in Cancer, OncoKB, Cancer Gene Census, Database of Curated Mutations, CGI Biomarkers (the cancer genome interpreter biomarker database), Tumor Alterations Relevant for Genomics-Driven Therapy, and the Precision Medicine Knowledge Base. We downloaded each KB and restructured their content to describe variants, genes, drugs, and gene-drug associations in a common format. We normalized gene names to Entrez Gene IDs and drug names to ChEMBL and DrugBank IDs. For the analysis of clinically relevant gene-drug associations, we obtained lists of genes affected by genetic alterations and putative drug therapies for 113 patients with cancer whose cases were presented at the Molecular Tumor Board (MTB) of the Charité Comprehensive Cancer Center. Our analysis revealed that the KBs are largely overlapping but also that each source harbors a notable amount of unique information. Although some KBs cover more genes, others contain more data about gene-drug associations. Retrospective comparisons with findings of the Charitè MTB at the gene level showed that use of multiple KBs may considerably improve retrieval results. The relative importance of a KB in terms of cancer genes was assessed in more detail by logistic regression, which revealed that all but one source had a notable impact on result quality. We confirmed these findings using a second data set obtained from an independent MTB. To date, none of the existing publicly available KBs on gene-drug associations in precision oncology fully subsumes the others, but all of them exhibit specific strengths and weaknesses. Consideration of multiple KBs, therefore, is essential to obtain comprehensive results.