Genomics England Research Articles

Association between tumour somatic mutations and venous thromboembolism in the 100,000 Genomes Project cancer cohort: a study protocol

Venous thromboembolism (VTE) is a common cause of morbidity and mortality in patients with cancer. There is evidence that specific aberrations in tumour biology contribute to the pathophysiology of this condition. We plan to examine the association between tumour somatic mutations and VTE in an existing cohort of patients with cancer, who were enrolled to the flagship Genomics England 100,000 Genomes Project. Here, we outline an a-priori analysis plan to address this objective, including details on study cohort selection, exposure and outcome definitions, annotation of genetic variants and planned statistical analyses. We will assess the effect of 1) deleterious somatic DNA variants in each gene; 2) tumour mutational burden and 3) tumour mutational signatures on the rate of VTE (outcome) in a pan-cancer cohort. Sensitivity analyses will be performed to examine the robustness of any associations, including adjustment for potentially correlated co-variates: tumour type, stage and systemic anti-cancer therapy. We hope that results from this study may help to identify key genes which are implicated in the development of cancer associated thrombosis, which may shed light on related mechanistic pathways and/or provide data which can be integrated into genetic risk prediction models for these patients.

Genomic landscape of adult testicular germ cell tumours in the 100,000 Genomes Project

Testicular germ cell tumours (TGCT), which comprise seminoma and non-seminoma subtypes, are the most common cancers in young men. In this study, we present a comprehensive whole genome sequencing analysis of adult TGCTs. Leveraging samples from participants recruited via the UK National Health Service and data from the Genomics England 100,000 Genomes Project, our results provide an extended description of genomic elements underlying TGCT pathogenesis. This catalogue offers a comprehensive, high-resolution map of copy number alterations, structural variation, and key global genome features, including mutational signatures and analysis of extrachromosomal DNA amplification. This study establishes correlations between genomic alterations and histological diversification, revealing divergent evolutionary trajectories among TGCT subtypes. By reconstructing the chronological order of driver events, we identify a subgroup of adult TGCTs undergoing relatively late whole genome duplication. Additionally, we present evidence that human leukocyte antigen loss is a more prevalent mechanism of immune disruption in seminomas. Collectively, our findings provide valuable insights into the developmental and immune modulatory processes implicated in TGCT pathogenesis and progression.

The co-occurrence of genetic variants in the TYR and OCA2 genes confers susceptibility to albinism

Although rare genetic conditions are mostly caused by DNA sequence alterations that functionally disrupt individual genes, large-scale studies using genome sequencing have started to unmask additional complexity. Understanding how combinations of variants in different genes shape human phenotypes is expected to provide important insights into the clinical and genetic heterogeneity of rare disorders. Here, we use albinism, an archetypal rare condition associated with hypopigmentation, as an exemplar for the study of genetic interactions. We analyse data from the Genomics England 100,000 Genomes Project alongside a cohort of 1120 individuals with albinism, and investigate the effect of dual heterozygosity for the combination of two established albinism-related variants: TYR:c.1205 G > A (p.Arg402Gln) [rs1126809] and OCA2:c.1327 G > A (p.Val443Ile) [rs74653330]. As each of these changes alone is insufficient to cause disease when present in the heterozygous state, we sought evidence of synergistic effects. We show that, when both variants are present, the probability of receiving a diagnosis of albinism is significantly increased (odds ratio 12.8; 95% confidence interval 6.0 – 24.7; p-value 2.1 ×10−8). Further analyses in an independent cohort, the UK Biobank, support this finding and highlight that heterozygosity for the TYR:c.1205 G > A and OCA2:c.1327 G > A variant combination is associated with statistically significant alterations in visual acuity and central retinal thickness (traits that are considered albinism endophenotypes). The approach discussed in this report opens up new avenues for the investigation of oligogenic patterns in apparently Mendelian disorders.

Prognostic determinants in cancer survival: a multidimensional evaluation of clinical and genetic factors across 10 cancer types in the participants of Genomics England’s 100,000 Genomes Project

BackgroundCancer is a complex disease, caused and impacted by a combination of genetic, demographic, clinical, environmental and lifestyle factors. Analysis of cancer characteristics, risk factors, treatment options and the heterogeneity across cancer types has been the focus of medical research for years. The aim of this study is to describe and summarise genetic, clinicopathological, behavioural and demographic characteristics and their differences across ten common cancer types and evaluate their impact on overall survival outcomes.MethodsThis study included data from 9977 patients with bladder, breast, colorectal, endometrial, glioma, leukaemia, lung, ovarian, prostate, and renal cancers. Genetic data collected through the 100,000 Genomes Project was linked with clinical and demographic data provided by the National Cancer Registration and Analysis Service (NCRAS), Hospital Episode Statistics (HES) and Office for National Statistics (ONS). Descriptive and Kaplan Meier survival analyses were performed to visualise similarities and differences across cancer types. Cox proportional hazards regression models were applied to identify statistically significant prognostic factor associations with overall survival.Results161 clinical and 124 genetic factors were evaluated for prognostic association with overall survival. Of these, 116 unique factors were found to have significant prognostic effect for overall survival across ten cancer types when adjusted for age, sex and stage. The findings confirmed prognostic associations with overall survival identified in previous studies in factors such as multimorbidity, tumour mutational burden, and mutations in genes BRAF, CDH1, NF1, NRAS, PIK3CA, PTEN, TP53. The results also identified new prognostic associations with overall survival in factors such as mental health conditions, female health-related conditions, previous hospital encounters and mutations in genes FANCE, FBXW7, GATA3, MSH6, PTPN11, RB1, RNF43.ConclusionThis study provides a comprehensive view of clinicopathological and genetic prognostic factors across different cancer types and draws attention to less commonly known factors which might help produce more precise prognosis and survival estimates. The results from this study contribute to the understanding of cancer disease and could be used by researchers to develop complex prognostic models, which in turn could help predict cancer prognosis more accurately and improve patient outcomes.

A Genomics England haplotype reference panel and imputation of UK Biobank

We built a reference panel with 342 million autosomal variants using 78,195 individuals from the Genomics England (GEL) dataset, achieving a phasing switch error rate of 0.18% for European samples and imputation quality of r2 = 0.75 for variants with minor allele frequencies as low as 2 × 10−4 in white British samples. The GEL-imputed UK Biobank genome-wide association analysis identified 70% of associations found by direct exome sequencing (P < 2.18 × 10−11), while extending testing of rare variants to the entire genome. Coding variants dominated the rare-variant genome-wide association results, implying less disruptive effects of rare non-coding variants.

Rare Sequence Variation Underlying Suspected Familial Cerebral Small-Vessel Disease.

Cerebral small-vessel disease (cSVD) is the leading monogenic cause of stroke. Despite genetic screening in routine diagnosis, many cases remain without a known causative variant. Using a cohort with suspected familial cSVD and whole-genome sequencing, we screened for variants in genes associated with monogenic cSVD and searched for novel variants associated with the disease. Rare variants were identified in whole-genome sequencing data from the NBR (National Institute for Health Research BioResource Rare Disease) study. Pathogenic variants in known monogenic cSVD genes were identified. Gene-based burden tests and family analysis were performed to identify novel variants associated with familial cSVD. A total of 257 suspected cSVD cases (mean ± SD age, 56.2 ± 16.1 years), and 13 086 controls with other nonstroke diseases (5874 [44.9%] men) were studied. A total of 8.9% of the cases carried a variant in known cSVD genes. Excluding these known causes, 23.6% of unrelated subjects with cSVD carried predicted deleterious variants in the Genomics England gene panel, but no association was found with cSVD in burden tests. We identified potential associations with cSVD in noncoding genes, including RP4-568F9.3 (adjusted P = 7.1 × 10-25), RP3-466I7.1 (adjusted P = 8.9 × 10-16), and ZNF209P (adjusted P = 1.0 × 10-15), and matrisomal genes (adjusted P = 5.1 × 10-6), including FAM20C, INHA, LAMC1, and VWA5B2. Predicted deleterious variants in known cSVD genes were present in 23.6% of unrelated cases with cSVD, but none of the genes were associated with the disease. Rare variants in noncoding and matrisomal genes could potentially contribute to cSVD development. These genes could play a role in tissue development and brain endothelial cell function. However, further studies are needed to confirm their pathophysiological roles.

A Comparison of Structural Variant Calling from Short-Read and Nanopore-Based Whole-Genome Sequencing Using Optical Genome Mapping as a Benchmark.

The identification of structural variants (SVs) in genomic data represents an ongoing challenge because of difficulties in reliable SV calling leading to reduced sensitivity and specificity. We prepared high-quality DNA from 9 parent-child trios, who had previously undergone short-read whole-genome sequencing (Illumina platform) as part of the Genomics England 100,000 Genomes Project. We reanalysed the genomes using both Bionano optical genome mapping (OGM; 8 probands and one trio) and Nanopore long-read sequencing (Oxford Nanopore Technologies [ONT] platform; all samples). To establish a "truth" dataset, we asked whether rare proband SV calls (n = 234) made by the Bionano Access (version 1.6.1)/Solve software (version 3.6.1_11162020) could be verified by individual visualisation using the Integrative Genomics Viewer with either or both of the Illumina and ONT raw sequence. Of these, 222 calls were verified, indicating that Bionano OGM calls have high precision (positive predictive value 95%). We then asked what proportion of the 222 true Bionano SVs had been identified by SV callers in the other two datasets. In the Illumina dataset, sensitivity varied according to variant type, being high for deletions (115/134; 86%) but poor for insertions (13/58; 22%). In the ONT dataset, sensitivity was generally poor using the original Sniffles variant caller (48% overall) but improved substantially with use of Sniffles2 (36/40; 90% and 17/23; 74% for deletions and insertions, respectively). In summary, we show that the precision of OGM is very high. In addition, when applying the Sniffles2 caller, the sensitivity of SV calling using ONT long-read sequence data outperforms Illumina sequencing for most SV types.

Progressive Kidney Failure by Angiotensinogen Inactivation in the Germline.

Autosomal recessive renal tubular dysgenesis is a rare, usually fatal inherited disorder of the renin-angiotensis system (RAS). Herein, we report an adolescent individual experiencing an unknown chronic kidney disease and aim to provide novel insights into disease mechanisms. Exome sequencing for a gene panel associated with renal disease was performed. The RAS was assessed by comprehensive biochemical analysis in blood. Renin expression was determined in primary tubular cells by quantitative polymerase chain reaction and in situ hybridization on kidney biopsy samples. Allele frequencies of heterozygous and biallelic deleterious variants were determined by analysis of the Genomics England 100,000 Genomes Project. The patient was delivered prematurely after oligohydramnios was detected during pregnancy. Postnatally, he recovered from third-degree acute kidney injury but developed chronic kidney disease stage G3b over time. Exome sequencing revealed a previously reported pathogenic homozygous missense variant, p.(Arg375Gln), in the AGT (angiotensinogen) gene. Blood AGT concentrations were low, but plasma renin concentration and gene expression in kidney biopsy, vascular, and tubular cells revealed strong upregulation of renin. Angiotensin II and aldosterone in blood were not abnormally elevated. Renal tubular dysgenesis may present as chronic kidney disease with a variable phenotype, necessitating broad genetic analysis for diagnosis. Functional analysis of the RAS in a patient with AGT mutation revealed novel insights regarding compensatory upregulation of renin in vascular and tubular cells of the kidney and in plasma in response to depletion of AGT substrate as a source of Ang II (similarly observed with hepatic AGT silencing for the treatment of hypertension).

Health Care Costs After Genome-Wide Sequencing for Children With Rare Diseases in England and Canada

Etiologic diagnoses for rare diseases can involve a diagnostic odyssey, with repeated health care interactions and inconclusive diagnostics. Prior studies reported cost savings associated with genome-wide sequencing (GWS) compared with cytogenetic or molecular testing through rapid genetic diagnosis, but there is limited evidence on whether diagnosis from GWS is associated with reduced health care costs. To measure changes in health care costs after diagnosis from GWS for Canadian and English children with suspected rare diseases. This cohort study was a quasiexperimental retrospective analysis across 3 distinct English and Canadian cohorts, completed in 2023. Mixed-effects generalized linear regression was used to estimate associations between GWS and costs in the 2 years before and after GWS. Difference-in-differences regression was used to estimate associations of genetic diagnosis and costs. Costs are in 2019 US dollars. GWS was conducted in a research setting (Genomics England 100 000 Genomes Project [100KGP] and Clinical Assessment of the Utility of Sequencing and Evaluation as a Service [CAUSES] Research Clinic) or clinical outpatient setting (publicly reimbursed GWS in British Columbia [BC], Canada). Participants were children with developmental disorders, seizure disorders, or both undergoing GWS between 2014 and 2019. Data were analyzed from April 2021 to September 2023. GWS and genetic diagnosis. Annual health care costs and diagnostic costs per child. Study cohorts included 7775 patients in 100KGP, among whom 788 children had epilepsy (mean [SD] age at GWS, 11.6 [11.1] years; 400 female [50.8%]) and 6987 children had an intellectual disability (mean [SD] age at GWS, 8.2 [8.4] years; 2750 female [39.4%]); 77 patients in CAUSES (mean [SD] age at GWS, 8.5 [4.4] years; 33 female [42.9%]); and 118 publicly reimbursed GWS recipients from BC (mean [SD] age at GWS, 5.5 [5.2] years; 58 female [49.2%]). GWS diagnostic yield was 143 children (18.1%) for those with epilepsy and 1323 children (18.9%) for those with an intellectual disability in 100KGP, 47 children (39.8%) in the BC publicly reimbursed setting, and 42 children (54.5%) in CAUSES. Mean annual per-patient spending over the study period was $5283 (95% CI, $5121-$5427) for epilepsy and $3373 (95% CI, $3322-$3424) for intellectual disability in the 100KGP, $724 (95% CI, $563-$886) in CAUSES, and $1573 (95% CI, $1372-$1773) in the BC reimbursed setting. Receiving a genetic diagnosis from GWS was not associated with changed costs in any cohort. In this study, receiving a genetic diagnosis was not associated with cost savings. This finding suggests that patient benefit and cost-effectiveness should instead drive GWS implementation.

Analysis-ready VCF at Biobank scale using Zarr.

Variant Call Format (VCF) is the standard file format for interchanging genetic variation data and associated quality control metrics. The usual row-wise encoding of the VCF data model (either as text or packed binary) emphasises efficient retrieval of all data for a given variant, but accessing data on a field or sample basis is inefficient. Biobank scale datasets currently available consist of hundreds of thousands of whole genomes and hundreds of terabytes of compressed VCF. Row-wise data storage is fundamentally unsuitable and a more scalable approach is needed. We present the VCF Zarr specification, an encoding of the VCF data model using Zarr which makes retrieving subsets of the data much more efficient. Zarr is a cloud-native format for storing multi-dimensional data, widely used in scientific computing. We show how this format is far more efficient than standard VCF based approaches, and competitive with specialised methods for storing genotype data in terms of compression ratios and calculation performance. We demonstrate the VCF Zarr format (and the vcf2zarr conversion utility) on a subset of the Genomics England aggV2 dataset comprising 78,195 samples and 59,880,903 variants, with a 5X reduction in storage and greater than 300X reduction in CPU usage in some representative benchmarks. Large row-encoded VCF files are a major bottleneck for current research, and storing and processing these files incurs a substantial cost. The VCF Zarr specification, building on widely-used, open-source technologies has the potential to greatly reduce these costs, and may enable a diverse ecosystem of next-generation tools for analysing genetic variation data directly from cloud-based object stores.

#787 Kidney genetics—a reproducible and curated database of kidney-disease associated genes

Abstract Background and Aims Genetic insights are becoming increasingly influential in the understanding and treatment of various kidney diseases (KD). Hundreds of genes associated with monogenic kidney disease have been identified, providing valuable insights into their diagnosis, management, and monitoring. However, the lack of a unified and standardized database of genes assigned to kidney diseases has led to diagnostic blind spots and comparability issues among current studies of kidney genetics. To address this gap, we created “Kidney-Genetics”, a regularly updated, automated and publicly accessible database, which aims to provide a comprehensive list of all relevant genes associated with kidney disease. Methods Kidney disease-associated gene information was compiled from various sources, including: (1) Genomics England and Australia PanelApp [1], (2) a comprehensive literature review of published gene lists, (3) clinical diagnostic panels for kidney disease, (4) a Human Phenotype Ontology (HPO)-based [2] search in rare disease databases (OMIM, Orphanet), and (5) a PubTator [3] API-based automated literature extraction from PubMed. An evidence-scoring system was developed to distinguish high- evidence genes from low evidence genes and candidate genes. High-evidence genes were defined as those present in two or more of the five resources and were manually curated based on predetermined criteria or, in the case of existing ClinGen [4] curation, their data and scores were used. Genes with a score of one or less were accordingly classified as candidate genes. Additionally, genes were grouped into different categories for later genotype-phenotype correlation matching (see Fig. 1). Results The Kidney-Genetics database currently contains detailed information on 3025 kidney-associated genes with detailed annotations on gene expression, kidney phenotype, inheritance mode, disease onset, possible syndromic disease mode and genetic variation. The number of genes extracted from the five analyzed sources of information are as follows: (1) 534, (2) 822, (3) 956, (4) 789, and (5) 2158 (see Fig. 2). Notably, 598 genes (19.8%) of the total 3025 genes are present in two or more of the analyzed information sources, thus meeting our evidence criteria, indicating high confidence and their potential for diagnostic use. Of these high-evidence genes, 526 (88.0%) are present in at least one, and 56 (9.4%) are present in all 10 comprehensive diagnostic laboratory panels. To ensure currency, Kidney-Genetics will be updated regularly and automatically on a monthly basis. We will also provide phenotypic and functional clustering results to facilitate gene grouping. Conclusion By utilizing Kidney-Genetics, clinicians, geneticists, and researchers can examine genomic data and improve their understanding of the genetic components of diverse KDs. The code and results are freely available on GitHub (kidney-genetics.org) [5]. A standardized pipeline and automated system keep our database on the cutting edge of kidney research and diagnostics. Screening efforts toward manual curation (such as through the ClinGen initiative) and assignment of diagnostic genes to kidney disease groups (e.g., syndromic vs. isolated; adult vs. pediatric; cystic, nephrotic, etc.) are currently under development.

#1761 Whole-genome sequencing reveals contribution of rare and common variation to structural kidney and urinary tract malformations

Abstract Background and Aims Structural kidney and urinary tract malformations are the commonest cause of kidney failure in children and young adults. Targeted and whole-exome sequencing has identified over 50 monogenic causes for these phenotypically heterogeneous conditions but have largely focused on cohorts enriched for familial, syndromic, or consanguineous disease. We sought to better characterise the genetic architecture of these conditions using whole genome sequencing (WGS) data from the UK's 100 000 Genomes Project. Method 992 unrelated individuals with structural kidney and urinary tract malformations were assessed using the Genomics England clinical interpretation pipeline to determine a genetic diagnosis. Statistically robust case-control association testing was performed using WGS data from a subset of 813 patients and 25 205 ancestry-matched unaffected individuals seeking enrichment of common and rare single-nucleotide and structural variants (SNV/SVs) on a genome-wide, per-gene, and cis-regulatory element basis. Heritability analysis was carried out in 623 cases and 20 060 controls of European ancestry with polygenic risk scores (PRS) derived for each of the six phenotypes. The PUV-PRS was validated in an independent cohort of 77 patients from the AGORA data- and biobank and an additional 2 746 European controls. Results The overall diagnostic yield was 4.3% (43/992; Fig. 1) with age &amp;lt; 18 years (P = 2.0 × 10−4), family history (P = 7.5 × 10−3) and extra-renal features (P = 1.9 × 10−4) all independent predictors of a genetic diagnosis. The diagnostic yield was higher in cases with cystic kidney dysplasia (10.7%) or kidney anomalies (5.9%). HNF1B was the most frequently identified cause. Exome-wide rare variant analysis identified enrichment of SNV/indels affecting AUTS2 (P = 3.7 × 10−6), ARHGAP5 (P=6.0 × 10−6), known kidney malformation gene HNF1B (P = 6.1 × 10−6), and ZNF879 (P=7.0 × 10−6), all of which have been previously implicated in developmental and/or cancer pathways. An increased burden of rare SVs affecting HNF1B (P = 3.1 × 10−4) was observed. There was no evidence of enrichment for cis-regulatory element disrupting SVs. Genome-wide testing of 19 million common and low-frequency variants (minor allele frequency &amp;gt; 0.1%) did not detect any individual variants that were significantly associated in the total cohort, but together these variants were estimated to contribute to 23% (standard error 11%) of the heritability of kidney and urinary tract malformations in cases with European ancestry. Comparison of phenotype-specific PRS showed considerable overlap between kidney anomalies, cystic kidney dysplasia, obstructive uropathy and vesico-ureteric reflux but the polygenic contribution to PUV and bladder exstrophy was distinct (Fig. 2). A PUV-PRS consisting of 36, 106 variants was validated in an independent European cohort of 77 cases and 2 746 controls (empirical P = 1 × 10−4). Conclusion We show that non-Mendelian genomic factors are important for the pathogenesis of structural kidney and urinary tract malformations, supported by the observation that a minority of patients in this large unselected cohort have a monogenic diagnosis. In addition, a significant proportion of heritability can be attributed to common and low-frequency variation and there is overlap in the polygenic contributors to upper urinary tract malformations with distinct polygenic signatures relevant to PUV and bladder exstrophy. Larger cohorts of patients are needed to increase power to detect variants contributing to this heritability and identify biological mechanisms and pathways important for kidney and urinary tract development.

#3020 A novel copy number analysis identifies human patients with NPHP1 whole gene deletions in previously genetically unsolved cases

Abstract Background and Aims Genome sequencing is rapidly becoming a routine approach to investigating inherited human diseases in clinical practice. Nevertheless, data generated from genome sequencing requires extensive bioinformatic analysis and filtering of variants to determine pathogenicity and remains challenging. Despite well-established workflows and software in place to process raw data, the analysis of genome sequencing data is still complex. Copy Number Variations (CNVs) are increasingly recognised to be pathogenic, especially where this leads to a whole gene being deleted. Software tools to decipher CNVs across genome sequencing datasets are being utilised but are not completely accurate. Method Using a bespoke CNV analysis of Genomics England 100,000 Genomes Project (100 kGP) WGS data from 11754 parent-child trios, we searched for homozygous deletions overlapping with annotated genes, across the 22 autosomes. Results Homozygous ∼150 kb whole gene deletion of NPHP1-MALL on chromosome 1, which aligned with a known region containing segmental duplications (SD), was the most frequent observation among all autosomal recessive (AR) OMIM-morbid genes. Clinical phenotypes were consistent with this genotype including nephronophthisis and retinal dystrophy. We extended our analysis to look for NPHP1 whole gene deletions in 18877 additional singletons within the 100 kGP. This reverse genetics approach identified additional patients, up to this point genetically unsolved, with whole gene deletions and matching phenotypes. The frequency of heterozygous NPHP1 whole gene deletion was assessed in a control cohort of 15440 cancer patients within 100 kGP and was found to be 0.35%. Conclusion In conclusion, NPHP1 was the most frequently observed AR OMIM-morbid gene knockout in our analysis. Homozygous whole gene deletions in NPHP1 may be easily missed using standard WGS pipelines which don't assess fully for CNVs and a targeted read-depth approach to identify CNVs in parent-patient trios and singletons is valuable in identifying these deletions.

Increasing the diagnostic yield of childhood glaucoma cases recruited into the 100,000 Genomes Project

Childhood glaucoma (CG) encompasses a heterogeneous group of genetic eye disorders that is responsible for approximately 5% of childhood blindness worldwide. Understanding the molecular aetiology is key to improving diagnosis, prognosis and unlocking the potential for optimising clinical management. In this study, we investigated 86 CG cases from 78 unrelated families of diverse ethnic backgrounds, recruited into the Genomics England 100,000 Genomes Project (GE100KGP) rare disease cohort, to improve the genetic diagnostic yield. Using the Genomics England/Genomic Medicine Centres (GE/GMC) diagnostic pipeline, 13 unrelated families were solved (13/78, 17%). Further interrogation using an expanded gene panel yielded a molecular diagnosis in 7 more unrelated families (7/78, 9%). This analysis effectively raises the total number of solved CG families in the GE100KGP to 26% (20/78 families). Twenty-five percent (5/20) of the solved families had primary congenital glaucoma (PCG), while 75% (15/20) had secondary CG; 53% of this group had non-acquired ocular anomalies (including iris hypoplasia, megalocornea, ectopia pupillae, retinal dystrophy, and refractive errors) and 47% had non-acquired systemic diseases such as cardiac abnormalities, hearing impairment, and developmental delay. CYP1B1 was the most frequently implicated gene, accounting for 55% (11/20) of the solved families. We identified two novel likely pathogenic variants in the TEK gene, in addition to one novel pathogenic copy number variant (CNV) in FOXC1. Variants that passed undetected in the GE100KGP diagnostic pipeline were likely due to limitations of the tiering process, the use of smaller gene panels during analysis, and the prioritisation of coding SNVs and indels over larger structural variants, CNVs, and non-coding variants.

Phenome-wide identification of therapeutic genetic targets, leveraging knowledge graphs, graph neural networks, and UK Biobank data.

The ongoing expansion of human genomic datasets propels therapeutic target identification; however, extracting gene-disease associations from gene annotations remains challenging. Here, we introduce Mantis-ML 2.0, a framework integrating AstraZeneca's Biological Insights Knowledge Graph and numerous tabular datasets, to assess gene-disease probabilities throughout the phenome. We use graph neural networks, capturing the graph's holistic structure, and train them on hundreds of balanced datasets via a robust semi-supervised learning framework to provide gene-disease probabilities across the human exome. Mantis-ML 2.0 incorporates natural language processing to automate disease-relevant feature selection for thousands of diseases. The enhanced models demonstrate a 6.9% average classification power boost, achieving a median receiver operating characteristic (ROC) area under curve (AUC) score of 0.90 across 5220 diseases from Human Phenotype Ontology, OpenTargets, and Genomics England. Notably, Mantis-ML 2.0 prioritizes associations from an independent UK Biobank phenome-wide association study (PheWAS), providing a stronger form of triaging and mitigating against underpowered PheWAS associations. Results are exposed through an interactive web resource.

Abstract PL-01: Breast Cancer Evolution, Immune Evasion and Metastasis Driven by Chromosomal Instability

Abstract Chromosomal instability (CIN) is a poor prognostic feature with high prevalence in breast cancer and is responsible for driving Somatic Copy Number Aberrations and Intratumour Heterogeneity (ITH) within subclones during breast cancer evolution. We have observed that SCNAs are enriched in metastatic samples, including gains in chromosome 11q13.3 (encompassing CCND1) in HER2+ breast cancer. However, there have thus far been no prospective, comprehensive studies of intratumor heterogeneity in breast cancer. We will present data from TRACERx Breast/SCANDARE a prospective, multicentre study of triple-negative breast cancer (TNBC) in which multi-region biopsies have been collected from primary TNBC prior to neoadjuvant treatment, and at longitudinal time points including surgery and relapse. We find prevalent intratumor heterogeneity in somatic mutations and SCNAs in the primary tumor prior to treatment. We have leveraged the rich clinical annotation in this cohort to correlate intratumor heterogeneity scores with key clinical outcomes, including treatment response (pathological complete response to neoadjuvant chemotherapy) and survival. We find that CIN manifests as oncogenic amplification on extrachromosomal DNA (ecDNA), which plays a pivotal role in driving drug resistance and tumor evolution. Through analysis of Whole Genome Sequencing data from 2936 breast tumors from the Genomics England breast cancer cohort, we have identified focal amplifications driven by ecDNA in 46.4% of HER2+ breast cancers. EcDNA were enriched in metastatic tumors and was associated with poor clinical outcomes. In addition to oncogenic amplifications such as HER2 derived from ecDNA during cancer evolution, ecDNAs contained immunomodulatory genes associated with reduced T cell infiltration. Further Immune-modulation by CIN results from loss of heterozygosity of the HLA locus (HLA LOH) and is a prevalent subclonal event in breast cancer. CIN is further drives haploid LOH, resulting in the high prevalence of whole genome doubling and further CIN in TNBC. Tumors are heterogeneous compositions of distinct clones with different compliments of SCNAs and varying levels of fitness, hence measuring the proliferation of individual clones to predict future evolutionary outcomes is likely to be critical. We have developed SPRINTER (Single-cell Proliferation Rate Inference in Non-homogeneous Tumours through Evolutionary Routes), a novel computational method to measure proliferation rates in individual tumour clones using single-cell whole-genome DNA sequencing data. We applied SPRINTER to 42,009 triple negative breast cancer cells, demonstrating significant proliferative heterogeneity between distinct tumor clones. Importantly, we show a correlation between higher proliferation rates and increased rates of somatic variants, suggesting proliferation links to clone evolvability. Taken together, these complimentary analyses provide evidence for the importance of CIN in driving ITH, selection, metastasis and immune evasion and provide a framework to determine recurrent evolutionary patterns in breast cancer evolution. Citation Format: C. Swanton. Breast Cancer Evolution, Immune Evasion and Metastasis Driven by Chromosomal Instability [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PL-01.

Abstract LB014: The role of genetic alterations in metastatic development insights from a large-scale genomic database

Abstract Metastasis remains a major challenge in cancer treatment, with somatic genetic alterations being a crucial feature in progression. This study harnesses the whole genome sequencing data in the Genomics England biobank (GEL), cataloging of somatic alterations in the context of cancer metastasis. Our research aims to uncover the genetic underpinnings that facilitate the spread of cancer from its primary site to secondary locations by analyzing clinicogenomic data from 3047 cancer patients with metastatic progression. In this study, we analyzed the complex relationship between somatic alterations and cancer progression to uncover significant interactions among these alterations in a broad distribution of common tumor types among well-established oncogenic and tumor suppressor genes. In our study, we employed a Bayesian network approach, enabling us to not only pinpoint crucial genetic contributors to metastasis but also to consider other clinical variables, including patient treatment, sex, age, tumor grade and cancer types. In our analysis, LILRB2 was found to directly increase metastatic risk in hematologic cancer, with prior evidence suggesting its potential as a therapeutic biomarker in colorectal cancer and adversely related to patient prognosis in lung cancer. Similarly, MYL1, initially linked to tumor metastasis and immune infiltration in head and neck squamous cell carcinoma, was also found to increase the risk of metastasis in renal cancer, while interacting with the transcription factor E2F4, marker of poor prognosis in hepatocellular carcinoma. Our study not only confirmed the numerous key somatic gene alterations significantly linked to metastatic behavior across various cancers, but also established their significant roles.Our work highlights the potential of leveraging extensive genomic databases such as GEL to enhance our understanding of cancer metastasis. GEL’s paired germline and somatic data allows us to not only identify mutations unique to cancer cells but also to uncover germline variants that are associated with an elevated risk of cancer and the presence of more aggressive tumor characteristics. By characterizing metastatic risk features, we can explore the opportunities to stratify high-risk patients in clinical trials, with similar effort has been applied to brain metastasis in lung adenocarcinoma. Although tumor heterogeneity poses a treatment challenge due to cell diversity, previous work has demonstrated value in targeting the metastatic potential. For instance, trastuzumab target HER2-positive breast cancers prone to metastasis, markedly improving patient outcomes. Similarly, vemurafenib effectively controls metastatic Melanoma progression by targeting specific mutations like BRAFV600E. Understanding the genetic basis of metastasis could help identify high risk patients and lead to personalized combination therapies addressing various tumor subpopulations. Future research based on these findings could lead to the development of personalized combination treatment strategies that improve patient outcomes. Citation Format: Songjun Xu, Danyang Yu, Shuwei Li, Christopher Moy, Julio Molineros. The role of genetic alterations in metastatic development insights from a large-scale genomic database [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB014.

Abstract 4337: Illuminating the complex rearrangement landscape of sarcoma

Abstract Structural rearrangements and in particular, gene fusions, are key mutational processes in bone and soft tissue tumors, used for both disease classification and as biomarkers. However, the mutational processes and rearrangement processes underlying many of these events remain poorly characterized. Recent data indicate that sarcomas show particularly high frequencies of complex rearrangement events, including patterns which do not fit those of known mutational mechanisms. As the largest whole genome sequencing (WGS) cohort of sarcomas to date, the Genomics England (GE) 100,000 Genomes project represents a unique dataset in which to profile these events. Structural variants (SVs) were identified in whole genome sequencing data for 978 samples from the Genomics England 100,000 Genomes Project using an optimized approach based on 5 specialized SV callers. SVs identified by at least 2 callers were filtered to retain only those which passed the quality filter, and to remove variants present in either matched germline samples or &amp;gt;1% of a panel of normal variants. Chromothriptic events, extrachromosomal DNA (ecDNA) and genes enriched for SVs were identified using established algorithms. The prevalence of SVs varies across sarcoma, in keeping with previous reports, with particularly high rates observed in myxofibrosarcoma and liposarcoma. As described previously, liposarcomas showed the highest rates of chromothripsis. We have examined some tumor types at higher granularity than previously available, demonstrating directly that the rates of complex rearrangement events vary across subtypes. For example, chromothriptic events are identified in 100% of well-differentiated liposarcomas, but less than 10% of myxoid liposarcomas. Similarly, the presence of ecDNA varies by tumor type, with the novel observation of particularly high rates in angiosarcoma. Using these data, we have catalogued the rearrangement patterns generating canonical gene fusions in fusion-driven soft tissue tumor types. Despite more recent advances in histological classification, survival for patients with sarcoma has remained largely unchanged for 40 years. Characterization of the mutational processes underlying these rearrangements will shed light on the pathogenesis of these tumors. Citation Format: Sara Waise, Tom Lesluyes, Jonas Demeulemeester, Maxime Tarabichi, Adrienne Flanagan, Nischalan Pillay, Peter Van Loo. Illuminating the complex rearrangement landscape of sarcoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4337.

Abstract 1746: KRAS mutant lung adenocarcinoma is associated with distinct mutational signature profiles

Abstract Background: Lung cancer is the 3rd most common cancer in the UK and leading cause of cancer mortality worldwide. Somatic RAS mutations are the most common oncogenes in human cancer, and KRAS mutations are the largest molecular subset of lung adenocarcinoma (LUAD). Each cancer genome accumulates a unique combination of somatic mutations, and these mutational signatures provide a genetic imprint of mutational processes that have occurred during tumorigenesis. Mutational signatures provide information on tumor aetiology/maintenance and could highlight potential therapeutic targets. This is the largest study to date to profile KRAS mutant LUAD mutational signatures using whole genome sequencing (WGS) data. Methods: Paired tumor and germline WGS data were obtained through the 100 000 Genomes Project (Genomics England). KRAS mutations were profiled using Ensembl Variant Effect Predictor (VEP) v109.0. Single base substitution (SBS), doublet base substitution (DBS), small insertion and deletions (ID) and copy number (CN) mutational signature extraction was performed using SigProfiler (COSMIC version 3.3). Statistical analysis and plotting were performed in R; a two-tailed p-value &amp;lt;0.05 was considered statistically significant. Results: 680 LUAD patients were included; 301 KRAS mutant (44%), 379 KRAS wild type (63%). Common KRAS variants were present, including G12C (38.9%), G12V (15.9%) and G12D (14.0%). 17 SBS, 9 DBS, 10 ID and 9 CN signatures were extracted and present in both KRAS mutant and wild type cohorts. KRAS mutant samples were enriched in smoking related signatures (SBS4, DBS2, ID3; odds (OR) ratio 5.54, 3.80, 4.97 respectively, p &amp;lt;0.001) and diploid CN signature (CN1, OR 1.41, p = 0.03). Conversely, KRAS mutant samples were less likely to have clocklike (ID5, OR 0.61, p = 0.012; ID8, OR 0.44, p &amp;lt;0.001), APOBEC (SBS2, OR 0.49, p &amp;lt;0.001; SBS13; 0.50, p &amp;lt;0.001), chromosomal instability (CN9, OR 0.63, p = 0.019), loss of heterozygosity (CN14, OR 0.47, p = 0.015), or homologous recombination deficiency (CN17, OR 0.58, p &amp;lt;0.001) signatures. Discussion: These results show distinct mutational signature profiles between KRAS mutant and wild type LUAD. As previously described, KRAS mutant LUAD is dominated by smoking related signatures; however, the majority of remaining signatures were more common in the KRAS wild type cohort where there was a statistically significant difference. This data suggests different mutational processes are observed in KRAS mutant compared to wild type LUAD. KRAS mutations are commonly clonal and thought to initiate tumorigenesis: timing signatures relative to KRAS mutations is necessary to further our understanding of KRAS cancer evolution and the subject of future work. Combined with clinical data integration, mutational signatures could provide prognostic information and inform predictive models. Citation Format: Laura Woodhouse, Aliah Hawari, Andreas Gruber, Colin Lindsay, David Wedge. KRAS mutant lung adenocarcinoma is associated with distinct mutational signature profiles [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1746.

Abstract 5051: Evolutionary characterization and refinement of copy-number high endometrial carcinoma

Abstract Copy-number (CN)-high endometrial carcinoma (EC) is characterized by abundant CN-alterations, thought to be driven by TP53 aberrations, and is the subgroup with the poorest prognosis. CN-high EC exhibit substantial heterogeneity in histotypes, molecular markers, and mutational profiles. Here, we aimed to refine this subgroup by assessing the evolution and identifying CN-signatures (Drews et al. Nature) to establish CN-clusters. We used whole-genome sequencing (WGS;100X) data from 146 CN-high cases from Genomics England (GeL) as discovery cohort. We then performed shallow WGS (sWGS;0.1X) of FFPE tumor DNA of n=307 CN-high EC from (non-)trial cohorts. Timing of canonical driver mutations by MutationTimer in CN-high GeL revealed that TP53 mutations were: (i) not ubiquitous and (ii) not consistently identified as early clonal events (Table), challenging conventional belief that TP53 inactivation is necessary for initiation of CN-high EC. Evolutionary analyses (Plackett-Luce) are ongoing to identify carcinogenesis trajectories and assess how driver mutations other than TP53 influence CN-phenotypes. Unsupervised clustering of 307 EC by CN-signatures identified five clusters (E1: 19%; E2: 2%; E3; 20%; E4: 19%; E5: 40%), characterized by CN-signatures linked to biological phenomena. Clusters E1/E4 reflect homologous recombination deficiency (CX3), while cluster E5 seems driven by chromosomal missegregation (CX1). Notably, there was no one-to-one relationship between CN-clusters and histotypes, suggesting CN-evolution is independent of morphology. In conclusion, we have identified TP53-dependent and -independent pathways in CN-high EC evolution. Potentially clinically relevant CN-clusters were identified using cost-effective sWGS. sWGS of additional n=475 CN-high EC are ongoing. Furthermore, we will analyze in-depth clinicopathological relationships of clusters, including prognosis/prediction, using PORTEC trials prior to the meeting. Table. Timing of canonical driver mutations in CN-high EC Clonal - early Clonal - late Subclonal Clonal - NA Total 1. TP53 87 (67%) 4 (3%) 3 (2%) 36 (28%) 130 (100%) 2. PIK3CA 42 (76%) 1 (2%) 1 (2%) 11 (20%) 55 (100%) 3. PPP2R1A 18 (47%) 1 (3%) 2 (5%) 17 (45%) 38 (100%) 4. PTEN 17 (59%) 1 (3%) 2 (7%) 9 (31%) 29 (100%) 5. PIK3R1 11 (46%) 0 (0%) 3 (13%) 10 (42%) 24 (100%) 6. FBXW7 14 (61%) 0 (0%) 0 (0%) 9 (39%) 23 (100%) 7. KRAS 9 (64%) 0 (0%) 2 (14%) 3 (21%) 14 (100%) Abbreviation: NA, timing not applicable. Citation Format: Claire J. Kramer, Dina Ruano, Aliah Hawari, Felix Blanc-Durand, Cor D. de Kroon, Judith R. Kroep, Gitte Ortoft, Estrid Hogdall, Alexandra Leary, Carien L. Creutzberg, Maaike P. Vreeswijk, Nanda Horeweg, David N. Church, David C. Wedge, Tjalling Bosse. Evolutionary characterization and refinement of copy-number high endometrial carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5051.