23014 results for "ega"

in 53.12 milliseconds.

• Targeted de-methylation of the FOXP3-TSDR

  CD4+ regulatory T cells (Tregs) are key mediators of immunological tolerance and promising effector cells for immuno-suppressive adoptive cellular therapy to fight autoimmunity and chronic inflammation. Their functional stability is critical for their clinical utility and has been correlated to the demethylated state of the TSDR/CNS2 enhancer element in the Treg lineage transcription factor FOXP3. However, proof for a causal contribution of the TSDR de-methylation to FOXP3 stability and Treg induction is so far lacking. We here established a powerful transient-transfection CRISPR-Cas9-based epigenetic-editing method for the selective de-methylation of the TSDR within the endogenous chromatin environment of a living cell. The induced de-methylated state was stable over weeks in clonal T cell proliferation cultures even after expression of the editing complex had ceased. Epigenetic editing of the TSDR resulted in FOXP3 expression, even in its physiological isoform distribution, proving a causal role for the de-methylated TSDR in FOXP3 regulation. However, successful FOXP3 induction was not associated with a switch towards a functional Treg phenotype, in contrast to what has been reported from FOXP3 overexpression approaches. Thus, TSDR de-methylation is required, but not sufficient for a stable Treg phenotype induction. Therefore, targeted demethylation of the TSDR may be a critical addition to published in vitro Treg induction protocols which so far lack FOXP3 stability.

  Study EGAS00001004867

• Single cell transcriptomics of hESC-derived midbrain dopaminergic neurons generated by a new human development-based protocol

  Recent advances in stem cell technologies provide new opportunities for regenerative medicine and Parkinson’s disease modelling. However, the precise composition and quality of human embryonic stem cell (hESC)-derived preparations at the single cell level, compared to endogenous brain standards, remains to be determined. In addition, multiple developmental pathways important for midbrain dopaminergic (mDA) neuron development remain to be implemented in hESCs. Here we use mouse developmental knowledge and human single cell RNA-sequencing (scRNA-seq) data to generate a new development-based protocol to differentiate hESCs into mDA neurons. We found that Laminin 511, dual WNT activation with CHIR99021 and WNT5A, together with FGF8b, improved different aspects of midbrain patterning. Moreover, mDA neurogenesis and differentiation were enhanced by activation of liver X receptors and inhibition of fibroblast growth factor signaling. Analysis of hESC-derived midbrain cultures by scRNA-seq revealed multiple hESC-derived clusters, which resembled those in the endogenous human ventral midbrain. Moreover, high quality mDA neurons were detected and found to progressively acquire functionality in vitro. Thus, we hereby provide a hESC differentiation protocol that recapitulates several aspects of endogenous mDA neuron development, and a strategy to determine cell composition and quality of hESC-derived preparations that can be useful for disease modeling and cell therapy.

  Study EGAS00001006313

• Intratumoural heterogeneity and immune modulation in lung adenocarcinoma of female smokers and never smokers

  Lung cancer is still the leading cause of cancer death worldwide despite declining smoking prevalence in industrialised countries. Although lung cancer is highly associated with smoking status, a significant proportion of lung cancer cases develop in patients who never smoked, with an observable bias towards female never smokers. A better understanding of lung cancer heterogeneity and immune system involvement during tumour evolution and progression in never smokers is therefore highly warranted. We employed single nucleus transcriptomics of surgical lung adenocarcinoma (LADC) and normal lung tissue samples from patients with or without smoking history. Immune cells as well as fibroblasts and endothelial cells respond to tobacco smoke exposure by inducing a highly inflammatory state in normal lung tissue. In the presence of LADC, we identified differentially expressed transcriptional programmes in macrophages and cancer-associated fibroblasts, providing insight into how the niche favours tumour progression. Within tumours, we distinguished eight subpopulations of neoplastic cells in female smokers and never smokers. Through pseudotemporal ordering, we inferred a trajectory towards two differentiated tumour cell states implicated in cancer progression and invasiveness. A proliferating cell population sustaining tumour growth exhibits differential immune modulating signatures in both patient groups. Our results resolve cellular heterogeneity and immune interactions in LADC, with a special emphasis on female never smokers and implications for the design of therapeutic approaches.

  Study EGAS00001006331

• 5- FU treated organoids

  5-Fluorouracil (5-FU) is a chemotherapeutic drug component that is commonly used for the treatment of solid cancers. The anticancer properties of 5-FU have been attributed to interference with nucleotide synthesis and through direct incorporation into the DNA as being a pyrimidine analog. As both mechanisms of action may have a mutational impact on surviving tumor cells, we performed three independent analyses to characterize the genomic consequences of 5-FU treatment: i) in vitro treatment of intestinal organoids with 5-FU followed by whole genome sequencing, ii) genome-wide mutation analyses in tumor samples from 5-FU treated patients with breast or colorectal cancer, and iii) analysis of paired biopsies from patients treated with 5-FU between biopsies. Our results demonstrate that both in vitro and in vivo 5-FU causes a mutational pattern that is dominated by T>G substitutions in a CTT context and with strong resemblance to COSMIC signature 17. Interestingly, this signature is also found in non 5-FU treated patients, most prominently in esophageal tumors, indicating that distinct endogenous and exogenous triggers can converge into highly similar mutational signatures. Furthermore, our results suggest that 5-FU may have adverse mutagenic effects on healthy cells and contributes to genetic variability in surviving cancer cells thereby contributing to tumor evolution.

  Study EGAS00001003592

• Real-time analysis of the cancer genome and fragmentome from plasma and urine short and long cell-free DNA using Nanopore sequencing

  Cell-free DNA (cfDNA) can be isolated from blood and/or urine of cancer patients and analyzed with sequencing. Unfortunately, most conventional short-read sequencing methods are technically challenging, labor intensive and time consuming, requiring several days but more typically weeks to obtain interpretable data which are limited by a bias for short cfDNA fragments. Here, we demonstrate that with Oxford Nanopore Technologies sequencing  we can achieve economical and ultra-fast delivery of clinical data from liquid biopsies.   Our ‘ITSFASTR’ approach is able to deliver copy number aberrations, and cfDNA fragmentation profiles in less than 24 hours from sample collection. The tumor-derived cfDNA fraction calculated from lung cancer patient plasma and urine from bladder cancer patients was highly correlated (R=0.98) to the tumor fraction calculated from conventional short-read sequencing of the same samples. cfDNA size profile and fragmentation patterns in plasma and urine exhibited the typical cfDNA features yet with a significantly higher proportion of informative fragments that exceed 300bp, exhibiting similar tumor fraction than shorter size fragments. Notably, comprehensive fragment-end composition and nucleosome profiling near transcription start sites can be determined from the same data.  We propose that ITSFASTR is the first point-of-care solution for obtaining genomic and fragmentomic results from liquid biopsies.

  Study EGAS00001006591

• Genomic insights into the pathogenesis of Epstein-Barr virus-associated diffuse large B-cell lymphoma by whole-genome and targeted amplicon-based sequencing

  Epstein–Barr virus (EBV)-associated diffuse large B-cell lymphoma not otherwise specified (DLBCL NOS) constitutes a distinct clinicopathological entity in the current World Health Organization (WHO) classification, yet its genomic features remain sparsely characterized. Combining whole-genome sequencing (WGS), targeted amplicon sequencing (tNGS) and fluorescence in-situ hybridization (FISH) of tumours from 47 cases of EBV+ DLBCL we delineate the genomic landscape of this rare disease. Interrogating WGS data by sequential gene-set enrichment and network-diffusion we describe an accumulation of mutations affecting the nuclear factor κB (NFκB) pathway (CSNK2A2, CARD10), IL-6/JAK/STAT (SOCS1/3, STAT3) and WNT signalling (FRAT1, SFRP5) alongside aberrations in immunological processes like interferon-response. Further, WGS uncovered structural aberrations including 6q deletions (5/8 patients), which were subsequently validated by FISH (14/32 cases). Integrated WGS and tNGS analysis emphasized a profile distinct from EBV-negative DLBCL enriched for mutations in ARID1A (45%), KMT2A/D (32/30%), ANKRD11 (32%), NOTCH2 (32%). Expanding on previous reports, we identified recurrent alterations in CCR6 (15%), DAPK1 (15%), TNFRSF21 (13%), CCR7 (11%) and YY1 (6%). Our comprehensive description of the mutational landscape in EBV+ DLBCL substantiates the evidence of its pathobiological independence. Further, these observations aid in the refinement of the molecular taxonomy for aggressive lymphomas and guides future therapeutic strategies.

  Study EGAS00001004941

• Genetic insights into the biological mechanisms governing human ovarian ageing

  Reproductive longevity is critical for fertility and impacts healthy ageing in women, yet insights into the underlying biological mechanisms and treatments to preserve it are limited. Here, we identify 290 genetic determinants of ovarian ageing, assessed using normal variation in age at natural menopause (ANM) in ~200,000 women of European ancestry. These common alleles influence clinical extremes of ANM; women in the top 1% of genetic susceptibility have an equivalent risk of premature ovarian insufficiency to those carrying monogenic FMR1 premutations. Identified loci implicate a broad range of DNA damage response (DDR) processes and include loss-of-function variants in key DDR genes. Integration with experimental models demonstrates that these DDR processes act across the life-course to shape the ovarian reserve and its rate of depletion. Furthermore, we demonstrate that experimental manipulation of DDR pathways highlighted by human genetics increase fertility and extend reproductive life in mice. Causal inference analyses using the identified genetic variants indicates that extending reproductive life in women improves bone health and reduces risk of type 2 diabetes, but increases risks of hormone-sensitive cancers. These findings provide insight into the mechanisms governing ovarian ageing, when they act across the life-course, and how they might be targeted by therapeutic approaches to extend fertility and prevent disease.

  Study EGAS00001004947

• Control of Focal Adhesion Kinase Activation by RUNX1-regulated miRNAs in high-risk AML

  We recently described a 16-gene expression signature for improved risk stratification of acute myeloid leukemia (AML) patients called the AML Prognostic Score (APS). A subset of APS high-risk AML patients showed increased levels of focal adhesion kinase (FAK), encoded by the Protein Tyrosine Kinase 2 (PTK2) gene, which was correlated with RUNX1 mutations. RUNX1 mutant cells are more sensitive to PTK2 inhibitors. As we were not able to detect RUNX1 binding sites in the PTK2 promoter, we hypothesized that RUNX1 might regulate micro(mi)RNAs that repress PTK2, such that loss-of-function RUNX1 mutations would result in reduced miRNA expression and derepression of PTK2. Examination of paired RNA-seq and miRNA-seq data from 301 AML cases revealed two miRNAs that positively correlated with RUNX1 expression, contained RUNX1 binding sites in their promoters and were predicted to target PTK2. We show that the hsa-let7a-2-3p and hsa-miR-135a-5p promoters are regulated by RUNX1, and that PTK2 is a direct target of both miRNAs. Even in the absence of RUNX1 mutations, hsa-let7a-2-3p and hsa-miR-135a-5p regulate PTK2 expression, and reduced expression of these two miRNAs sensitizes AML cells to PTK2 inhibition. These data explain how RUNX1 regulates PTK2, and identify potential miRNA biomarkers for targeting AML with PTK2 inhibitors.

  Study EGAS00001006491

• BinDel: software tool for detecting clinically significant microdeletions in low-coverage WGS-based NIPT samples

  Clinically pathogenic chromosomal microdeletions causing genetic disorders such as DiGeorge syndrome are rare genetic aberrations that can cause clinically relevant fetal and childhood developmental deficiencies. Clinical severity of such deficiencies depend on the exact genomic location and genes affected by the fetal chromosomal aberration. Here we present the BinDel, a novel region-aware microdeletion detection software package developed to infer clinically relevant microdeletion risk in low-coverage whole-genome sequencing NIPT data. To test BinDel, we quantified the impact of sequencing coverage, fetal DNA fraction, and region length on microdeletion risk detection accuracy. We also estimated BinDel accuracy on known microdeletion samples and clinically validated aneuploidy samples. BinDel identified each positive control sample as high risk. We also determined that it is critical to take into account that the sample with a detected high microdeletion risk does not have a full chromosome aneuploidy, as the latter can cause erroneous high microdeletion risk findings. We observed that lower sequencing coverage resulted in reduced microdeletion detection accuracy, and higher fetal fractions considerably increased the microdeletion detection accuracy, with coverage becoming increasingly relevant as fetal DNA fraction decreased. In conclusion, we developed an R package-based software tool BinDel for inferring fetal microdeletion risks, which accurately identified all positive control samples with microdeletion or -duplication aberrations as high-risk samples.

  Study EGAS00001006663

• Uncovering tumor intrinsic and extrinsic factors that regulate hepatocellular carcinoma growth using patient derived xenograft assays

  Several major factors limit our understanding of hepatocellular carcinoma (HCC). First, human HCCs are infrequently biopsied for diagnosis and thus are not often biologically interrogated. Second, HCC initiation and progression are strongly influenced by the cirrhotic microenvironment, and the exact contributions of intrinsic and extrinsic tumor factors are unclear. A powerful approach to examine the personalized biology of liver cancers and the influence of host tissues is with patient derived xenograft (PDX) models. In Asia, HCCs from hepatitis B virus patients have been efficiently converted into PDXs, but few parallel efforts from the west have been reported. In a large-scale analysis, we implanted 93 HCCs and 8 cholangiocarcinomas (CCAs) to systematically analyze host factors and to define an optimized platform for PDX development from both surgical and biopsy samples. NSG mice that had undergone partial hepatectomy (PHx) represented the best combination of engraftability, growth, and passageability, but overall rates were low and indicative of a unique intrinsic biology for HCCs in the US. PDX models preserved the histology and genetic features of parental tumors, and ultimately, 9 new models were usable for pre-clinical studies. Intriguingly, HCC PDXs were differentially sensitive to regorafenib and sorafenib and CCA PDXs were also highly sensitive to regorafenib. PDX models functionalize early and advanced stage HCCs and revealed unique biological features of liver cancers from the US.

  Study EGAS00001004020

• Cell-free DNA fragmentation patterns and personalized sequencing reveal circulating tumor DNA in urine and plasma of glioma patients

  Glioma-derived cell-free DNA (cfDNA) is challenging to detect using liquid biopsy as levels in body fluids are low. We determined the glioma-derived DNA fractions in tumor biopsies, in cerebrospinal fluid (CSF), plasma and urine samples, using deep sequencing of personalized capture panels. By sequencing cfDNA across thousands of mutations identified individually in each patient’s matched tumor we detected tumor-derived DNA in plasma (10/12) and urine samples (8/11). The median tumor fraction was 6.4x10-3 in CSF, 3.1x10-5 in plasma and 4.7x10-5 in urine. We identified a shift in the size distribution for mutant cfDNA fragments in these body fluids. Next, we analyzed cfDNA fragment sizes with paired-end shallow whole genome sequencing (WGS) in urine samples from 35 patients with gliomas, 8 individuals with non-malignant brain disorders, and 26 controls (n=69 individuals, 96 samples). cfDNA in urine of glioma patients was significantly more fragmented compared to urine from patients with non-malignant brain disorders (t-test, p=1.7x10-2) and compared to urine of controls (t-test, p=5.2x10-9). The proportion of DNA fragments <60 bp was higher in glioma patients urine and could be used for classification (AUC=0.93). Machine learning models integrating fragment lengths could identify urine samples from glioma patients (AUC=0.97 in cross-validation).

  Study EGAS00001004355

• Loss of LGR4/GPR48 causes severe neonatal salt-wasting due to disrupted WNT signaling altering adrenal zonation

  Disorders of isolated mineralocorticoid deficiency causing potentially life-threatening salt-wasting crisis early in life have been associated with gene variants of aldosterone biosynthesis or resistance, but in some patients no such variants are found. WNT/β-catenin signaling is crucial for differentiation and maintenance of the aldosterone producing adrenal zona glomerulosa (zG). We describe a highly consanguineous family with multiple perinatal deaths or infants presenting at birth with failure to thrive, severe salt-wasting crises associated with isolated hypoaldosteronism, nail anomalies, short stature, and deafness. Whole exome sequencing revealed a homozygous splice variant in the R-SPONDIN receptor LGR4 gene (c.618-1G>C) regulating WNT signaling. The resulting transcripts affected protein function and stability, and resulted in loss of Wnt/β-catenin signaling in vitro. The impact of LGR4 inactivation was analyzed by adrenal cortex specific ablation of Lgr4, using Lgr4Flox/Flox mated with Sf1:Cre mice. Inactivation of Lgr4 within the adrenal cortex in the mouse model caused decreased WNT signaling, aberrant zonation with deficient zG and reduced aldosterone production. Thus, human LGR4 mutations establish a direct link between LGR4 inactivation and decreased canonical WNT signaling with abnormal zG differentiation and endocrine function. Therefore, variants in WNT signaling and its regulators should systematically be considered in familial hyperreninemic hypoaldosteronism.

  Study EGAS00001006808

• Dynamics of sequence and structural cell-free DNA landscapes in small-cell lung cancer

  Patients with small-cell lung cancer (SCLC) have an exceptionally poor prognosis, calling for improved real-time non-invasive biomarkers of therapeutic response. We performed targeted error-correction sequencing on 171 serial plasmas and matched white blood cell (WBC) DNA from 33 patients with metastatic SCLC who received systemic treatment with chemotherapy or immunotherapy-containing regimens. Tumor-derived sequence alterations and plasma aneuploidy were evaluated serially and combined to assess changes in total cell-free tumor load (cfTL). Longitudinal dynamic changes in cfTL were monitored to determine circulating cell-free tumor DNA (ctDNA) molecular response during therapy. Combined tiered analyses of tumor-derived sequence alterations and plasma aneuploidy allowed for assessment of ctDNA molecular response in all patients. Molecular responses captured the therapeutic effect and long-term clinical outcomes in a more accurate and rapid manner compared to radiographic imaging. Patients with sustained molecular responses had longer overall (log-rank p=0.0006) and progression-free (log-rank p<0.0001) survival, with molecular responses detected on average 4 weeks earlier than imaging. ctDNA analyses provide a rapid approach for the assessment of early on-therapy molecular responses and have important implications for the management of patients with SCLC, including the development of improved strategies for real-time tumor burden monitoring.

  Study EGAS00001006831

• cfDNA dataset from extracelular vesicles and paired non-fractionated samples

  Liquid biopsies contain multiple analytes that can be mined to improve the detection and management of cancer. Beyond cell-free DNA (cfDNA), mutations have been detected in DNA associated with extracellular vesicles (EV-DNA). The genome-wide composition and structure of EV-DNA remains poorly characterized, and whether circulating EVs are enriched in tumor signal compared to unfractionated cfDNA is still unclear. Here, using whole genome sequencing from selected lung cancer patients with a relatively high cfDNA tumor content >5%, we determined that the tumor fraction and heterogeneity are comparable between DNA associated with small (<200 nm) EVs and matched plasma cfDNA. DNA in EV fractions, obtained with standardized size-exclusion chromatography, are comprised of short ~150-180 bp fragments and long >1000 bp fragments that are poor in tumor signal. Other fractions only exhibit short fragments with similar tumor DNA content. The composition in bases at the end of EV-DNA fragments, as well as their fragmentation patterns are similar to total plasma cfDNA. Mitochondrial DNA however is relatively enriched in EV fractions. Our results suggests that cfDNA in plasma may be of dual nature, either bound to proteins (including the nucleosome) but also associated to EVs. cfDNA associated to small EV (including exosomes) is however not preferentially enriched in tumor signal.

  Study EGAS00001006848

• CYP2C19 long-read sequencing

  Inter-individual differences in drug response based on genetic variations can lead to drug toxicity and treatment inefficacy. A large part of this variability is caused by genetic variants in pharmacogenes. Unfortunately, the Single Nucleotide Variant (SNV) panels currently used in clinical pharmacogenomic (PGx) testing are unable to detect all genetic variability in these genes. Long-read sequencing, on the other hand, has been shown to be able to resolve complex (pharmaco)genes. In this study we have assessed the added value of long-read sequencing for PGx focusing on the clinically important and highly polymorphic CYP2C19 gene. With a capture-based long-read sequencing panel we were able to characterize the entire region and assign variants to their allele of origin (phasing), resulting in the identification of 813 unique variants in 37 samples. To assess the clinical utility of this data we have compared the performance of three different *-allele tools (Aldy, PharmCat and PharmaKU) which are specifically designed to assign clinical haplotypes to pharmacogenes based on all input variants. We conclude that long-read sequencing can improve our ability to characterize the CYP2C19 locus, help to identify novel haplotypes and that *-allele tools are an useful asset in phenotype prediction. Ultimately, this approach could help to better predict an individual’s drug response and improve therapy outcomes.

  Study EGAS00001006929

• ATAC-seq of a selected group of AML cases

  ATAC-seq data were generated for a number of selected patients to investigate changes in enhancer and promoter regions. ATAC-seq was essentially carried out as described in31. Briefly, prior to transposition the viability of the cells was assessed and 1 × 106 cells were treated in culture medium with DNase I (Sigma) at a final concentration of 200 U ml−1 for 30 minutes at 37 °C. After Dnase I treatment, cells were washed twice with ice-cold PBS, and cell viability and the corresponding cell count were assessed. 5 × 104 cells were aliquoted into a new tube and spun down at 500 × g for 5 minutes at 4 °C, before the supernatant was discarded completely. The cell pellet was resuspended in 50 µl of ATAC-RSB buffer (10 mM Tris-HCl pH 7.4, 10 mM NaCl, 3 mM MgCl2) containing 0.1% NP-40, 0.1% Tween-20, and 1% Digitonin (Promega), and was incubated on ice for 3 minutes to lyse the cells. Lysis was washed out with 1 ml of ATAC-RSB buffer containing 0.1% Tween-20. Nuclei were pelleted at 500 × g for 10 minutes at 4 °C. The supernatant was discarded carefully and the cell pellet was resuspended in 50 µl of transposition mixture (25 µl 2× tagment DNA buffer, 2.5 µl transposase (100 nM final; Illumina), 16.5 µl PBS, 0.5 µl 1% digitonin, 0.5 µl 10% Tween-20, 5 µl H2O) by pipetting up and down six times. The reaction was incubated at 37 °C for 30 minutes with mixing before the DNA was purified using the Monarch PCR & DNA Cleanup Kit (NEB) according to the manufacturer’s instructions. Purified DNA was eluted in 20 µl elution buffer (EB) and 10 µl purified sample was objected to a ten-cycle PCR amplification using Nextera i7- and i5-index primers (Illumina). Purification and size selection of the amplified DNA were carried out with Agencourt AMPure XP beads. For purification the ratio of sample to beads was set to 1:1.8, whereas for size selection the ratio was set to 1:0.55. Purified samples were eluted in 15 µl of EB. Quality and concentration of the generated ATAC libraries were analyzed using the High Sensitivity D1000 ScreenTape Kit (Agilent) and libraries were sequenced paired-end on a NovaSeq (Illumina). ATAC-seq reads were aligned to the human reference genome build hg19 with bowtie2

  Dataset EGAD00001007583

• WES data (cfTrack study) from NSCLC patients and OC patients

  We also collected samples from 8 NSCLC patients and 4 ovarian cancer patients and. For all 8 NSCLC patients, a tumor biopsy sample, a WBC sample, and three plasma samples were collected. For all 4 ovarian cancer patients, a WBC sample and two serum samples were collected. We collected tumor tissue sample from one ovarian cancer patient (OV4). The cfDNA was extracted from their plasma samples using the QIAamp circulating nucleic acid kit from QIAGEN (Germantown, MD). For serum cfDNA, ampure XP beads size selection was further performed to eliminate gDNA contamination. In brief, 0.5 volume of beads were first added to the cfDNA samples. After incubation, the supernatant was transferred to a new tube and an additional 2.0 volume of beads were added. After 80% ethanol wash, cfDNA was eluted from the beads. FA assays (Agilent Technologies) were performed to rule out the contamination of gDNA in the size selected samples. The cfDNA WES library of all patients and the genomic DNA WES library of the 4 ovarian cancer patients were constructed with the SureSelect XT HS kit from Agilent Technologies (Santa Clara, CA) according to the manufacturer’s protocol. In brief, 10ng of cfDNA was used as input material. After end repair/dA-tailing of cfDNA, the adaptor was ligated. The ligation product was purified with Ampure XP beads (Beckman-Coulter, Atlanta, GA) and the adaptor-ligated library was amplified with index primer in 10-cycle PCR. The amplified library was purified again with Ampure XP beads, and the amount of amplified DNA was measured using the Qubit 1xdsDNA HS assay kit (ThermoFisher, Waltham, MA). 700-1000 ng of DNA sample was hybridized to the capture library and pulled down by streptavidin-coated beads. After washing the beads, the DNA library captured on the beads was re-amplified with 10-cycle PCR. The final libraries were purified by Ampure XP beads. The library concentration was measured by Qubit, and the quality was further examined with Agilent Bioanalyzer before the final step of 2x150bp paired-end sequencing at an average coverage of 200. Whole-exome capture libraries of genomic DNA from the 8 NSCLC patients were constructed via Roche SeqCap EZ Exome V6 (Roche). Enriched exome libraries were sequenced on the Illumina HiSeq 3000 platform (Illumina) to generate 2x100bp paired-end reads at an average coverage of 200.

  Dataset EGAD00001008454

• Single cell transcriptional characterization of human megakaryocyte lineage commitment and maturation

  Single cell RNA-sequencing of sternal bone marrow reciding Hematopoietic Stem Cells (HSCs) and Megakaryocytes (MKs) from individuals undergoing elective open heart valve replacement. HSCs were defined as Lineage-, CD34+, CD38-, CD45RA-, CD90+, CD49f+ cells. MKs where CD41a+, CD42b+ and ploidy was determined with Hoechst. A sternal bone marrow scraping was taken directly following median sternotomy using a Volkmann’s spoon. The sample was collected into an EDTA Vacutainer tube containing 1.8mg/ml EDTA. 4mL of Dulbecco’s phosphate buffered saline (PBS, Sigma) containing 10% human serum albumin (HSA, Gemini Bio Products) was added and the whole volume was resuspended by pipetting 2-3 times. The sample was then put on metallic thermal beads (ThermoFisher Scientific) at a temperature between 0-4°C and transported to the University of Cambridge for further processing. For HSC isolation the cells were stained with the following antibody cocktail: PECy5 conjugated anti-lineage specific antibodies: CD2 (BD), CD3 (BD), CD10 (BD), CD11b (BD), CD11c (BD), CD19 (BD), CD20 (BD), CD56 (BD), biotinylated CD42b (Pab5, NHS Blood and Transplant, International Blood Group Reference Laboratory [IBGRL]), biotinylated GP6 (Pab5, NHS Blood and Transplant, International Blood Group Reference Laboratory [IBGRL]) used in combination with PECy5 conjugated streptavidin (Biolegend). Alexa Fluor 700 conjugated anti-CD34 (BD), PerCP-Cy5.5 conjugated anti-CD38 (BD), Pacific Blue conjugated anti-CD45RA (Invitrogen), PECy7 conjugated anti-CD90 (BD),PE conjugated anti-CD49f (BD). After staining cells were kept at 4°C before sorting using a FACS Aria Fusion flow sorter (BD). Single HSCs defined as Lineage-, CD34+, CD38-, CD45RA-, CD90+, CD49f+ cells were sorted by FACS directly into individual wells of a 96-well plate. Index sort data was collected for each single cell. For MK isolation the cells were stained for surface MK markers with mouse anti-human CD41a APC conjugated antibody (BD) and mouse anti-human CD42b PE conjugated antibody (BD) and for ploidy analysis with 1ug/ml Hoechst 33342 (Invitrogen). After incubation at 37°C for 30 minutes, the cells were kept at 4°C before sorting using a FACS Aria Fusion flow sorter (BD). Single cells and MK pools of 20 cells were sorted by FACS according to ploidy level using a 100uM nozzle directly into individual wells of a 96-well plate. cDNA synthesis and poly(A) enrichment was performed following the G&T-seq protocol (Macaulay et al. 2015), a variation of the Smart-seq2 protocol1. ERCC spike-in RNA (Ambion) was added to the lysis buffer in a dilution of 1:4,000,000.

  Dataset EGAD00001006677

• Target gene sequencing for human normal endometrial glands

  The raw fastq files target sequencing of 112 genes for 1,298 endometrial glands and matched blood samples. The paired-end sequencing data sets (R1 and R2) are deposited. ABCC1, ACRC, ANK3, ARHGAP35, ARID1A, ARID5B, ATCAY, ATM, ATR, BARD1, BCOR, BRCA1, BRCA2, BRD4, BRIP1, CAMTA1, CDC23, CDYL, CFAP54, CHD4, CHEK1, CHEK2, CTCF, CTNNB1, CUX1, DGKA, DISP2, DYNC2H1, EMSY, FAAP24, FAM135B, FAM175A, FAM65C, FANCA, FANCB, FANCC, FANCD2, FANCE, FANCF, FANCG, FANCI, FANCL, FANCM, FAT1, FAT3, FBN2, FBXW7, FGFR2, FRG1, GPR50, HEATR1, HIST1H4B, HNRNPCL1, HOOK3, KIAA1109, KIF26A, KMT2B, KMT2C, KRAS, LAMA2, LRP1B, MLH1, MON2, MRE11A, MSH2, MSH6, MTOR, NBN, PALB2, PHEX, PIK3CA, PIK3R1, PLXNB2, PLXND1, PMS2, POLE, POLR3B, PPP2R1A, PTEN, PTPN13, RAD50, RAD51, RAD51B, RAD51C, RAD51D, RAD52, RAD54B, RAD54L, RICTOR, SACS, SIGLEC9, SLC19A1, SLX4, SPEG, STT3A, TAF1, TAF2, TAS2R31, TFAP2C, TNC, TONSL, TP53, TTC6, UBA7, VNN1, WT1, XIRP2, ZBED6, ZC3H13, ZFHX3, ZFHX4, ZMYM4.

  Dataset EGAD00001008465

• RRBS data (cfSort study) from noncancer tissue DNA

  The RRBS libraries of the genomic DNA from the 521 tissue samples were constructed following the standard RRBS protocol. 100-200 ng of intact genomic DNA in the volume of 21.5 µl was used as input material. Restriction digestion was done with 2.5 µl 10xCutSmart buffer and 1 µl MspI (NEB) for 18 h at 37 oC and 20 min at 65 oC. 0.5 µl 10xCutSmart buffer, 0.3 µl dACGTP mixture (100 mM dATP, 10 mM dCTP, 10 mM dGTP), 1 µl Klenow (exo-, 5U/µl, NEB) and 2.6 µl RT-PCR water, 0.6 µl 50 mM DTT (ThermoFisher) was added to the mixture for end repair and A-overhang addition with the program 30 oC for 20 min, 37 oC for 1 h and 75 oC for 20 min. Adapter ligation was then performed with 1 µl 10xThermoFisher HC T4 ligase buffer, 0.4 µl 100 mM ATP (ThermoFisher), 0.2 µl 50 mM DTT, 1 µl ThermoFisher HC T4 DNA ligase (30 Weiss Unit/µl), 30 ng home-made duplex UMI adapter with all the cytosines methylated (protocol adopted from Kennedy et al.) at 16 oC for 20 h and 65 oC for 20 min. Bisulfite conversion of the adapter-ligated product was carried out with QIAGEN EpiTect plus DNA bisulfite kit following their protocol for two rounds of conversion. The converted product was purified with Qiagen MinElute spin column and eluted with 20 µl RT-PCR water. PCR amplification was done using the NEBNext Multiplex Oligos for Illumina (2.5 µl of universal and index primer each) and 25 µl KAPA HiFi HotStart Uracil+ ReadyMix (Roche) with the following cycling conditions: 98 oC for 45 s, 9 cycles of 98 oC for 15 s, 60 oC for 30 s and 72 oC for 30 s, followed by a final extension at 72 oC for 5 min. The PCR product was purified with 1x AmpureXP beads and eluted with 30 µl EB buffer. DNA concentration was measured by Qubit 1xdsDNA HS assay. 5% TBE-UREA PAGE and bioanalyzer assay was performed as quality control on each library before sequencing.

  Dataset EGAD00001010880

• RNA-sequencing of human skin samples obtained from SSc patients and healthy controls.

  Patient information SSc patients were recruited at the Department of Rheumatology of the Leiden University Medical Center (Leiden, The Netherlands). All patients met the American Rheumatism Association classification criteria for SSc (Subcommittee for scleroderma criteria 1980), and were classified according to LeRoy and Medsger criteria as either limited or diffuse cutaneous disease (LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA Jr, Rowell N 1988). Institutional review board approval and written informed consent was obtained before patients entered this study. Two 4 mm skin biopsies were taken from a standardized location on the most proximal part of the lower arm, distal from the elbow. In 10 patients the skin biopsy came from a clinically affected area and in 4 patients the skin was locally unaffected. One sample was used for RNA sequencing and one sample was used for immunohistochemistry. Skin biopsies from healthy individuals were commercially sourced (Tissue Solutions, UK) and collected from donors undergoing skin resection surgery and after informed consent. To match the healthy skin with patients as much as possible, skin biopsies from healthy controls were also taken from a similar position (the under-arm (for 4 controls) and leg (for 2 controls)). Healthy skin donors were selected to match the age and sex of the SSc patient cohort. Biopsies from patients and controls were equally treated and were both stored at -80°C until RNA isolation was performed. RNA from frozen skin biopsies was isolated using RNeasy kit from fibrous tissue (Qiagen, the Netherlands). RNA quantity was determined by using SimplyNano 2000 and quality was assessed on Tapestation (Agilent, the Netherlands). All samples included in the study had a RIN score above 7.0. Transcriptome characterisation and analysis RNA sequencing was performed using polyA selection and a stranded protocol using Ion Torrent next generation sequencing technology (Service XS, The Netherlands). The Ion PI Template OT2 200 Kit v3 and Ion PI Sequencing 200 Kit v3 were used according to the manufacturer’s instructions. 20 samples were run on 11 PI chips. PI chip analyses, base calling and quality checks were performed using the Torrent Server Suite. An average of 42 million 100 bp reads was generated per sample. Following quality control, reads were aligned to the human genome (Homo sapiens GRh38.78) using Bowtie2 and STAR (Dobin et al. 2013; Langmead and Salzberg 2012). Reads were first aligned with STAR. For the unmapped reads from STAR, a second alignment step was performed using bowtie2 (local very sensitive options)

  Dataset EGAD00001003832

• Whole Genome Sequencing of clonal expansions of single healthy somatic cells (human, female)

  HSC73_clone: Bone marrow mononuclear cells from the healthy 73 years old female were thawed and labeled with Alexa-Fluor 488-conjugated anti-CD34 (581, Biolegend), Alexa-Fluor 700-conjugated anti-CD38 (HIT2, eBioscience), a cocktail of APC-conjugated lineage antibodies consisting of anti-CD4 (RPA-T4), anti-CD8 (RPA-T8), anti-CD11b (ICRF44), anti-CD20 (2H7), anti-CD56 (B159, all BD Biosciences), anti-CD14 (61D3), anti-CD19 (HIB19) and anti-CD235a (HIR2, all eBiocience) and 1 micro-gram/ml propidium iodide (Sigma). Using a BD FACSAria cell sorter, single Lin-CD34+CD38-PI- cells were individually sorted into low-adhesion 96-well tissue culture plates (Corning) containing 100micro-litre of StemSpan Serum-Free Expansion Medium (Stemcell technologies) supplemented with 100ng/ml of human SCF and FLT-3L, 50ng/ml of human TPO, 20ng/ml of human IL-3, IL-6 and G-CSF (all cytokines from Peprotech) and 50U/ml of penicillin and 50μg/ml of streptomycin (Sigma). Cells were incubated at 37 degrees C in a humidified atmosphere with 5% CO2 in air. After 5 days in culture, another 100micro litres of cytokine-containing medium were added. 13 days after seeding, clones B6 and G2 had expanded to approx. 105 cells and were selected for whole genome sequencing (2x101bp, paired-end, Illumina HiSeq2500) after tagmentation-based library preparation (see Extended Experimental Procedures) for clone B6 and standard library preparation for clone G2. For germline-control ~106 unsorted bone marrow mononuclear cells from the same donor were used for sequencing. An average of 30-fold sequence coverage for each the clones and the matching control were obtained.L4clone: A progenitor cell clone was raised from a peripheral blood sample of the 39 year old healthy female. Frozen peripheral blood mononuclear cells (PBMCs) were isolated from 2 ml heparinised peripheral blood via Ficoll Paque density centrifugation. A methylcellulose assay was performed as described earlier (Weisse et al., 2012). In brief, non-adherent mononuclear cells were incubated in the presence of the recombinant human cytokines IL-3, IL-5 and GM-CSF (R&D systems) over 14 days to induce colony formation. Colonies were detected under an inverted light microscope, and plucked by a pipette when colonies had approximately 10,000 cells/CFU. Each colony was washed three times in PBS and finally frozen as a cell pellet in -80 degrees C. Genomic DNA was isolated using the QIAamp DNA micro kit according to the instructions of the manufacturer (Qiagen, Hilden, Germany). Whole genome sequencing (2x101bp, paired-end, Illumina HiSeq2500) was performed for colony 4 after tagmentation-based library preparation and resulted in 15-fold sequence coverage for each the colony and the matching whole blood.

  Dataset EGAD00001000666

• Ultrasensitive detection of minor allele fractions in maternal samples based on microhaplotype analysis

  Background: Tools for quantifying and studying genetic variation in heterogeneous cell populations have a limited sensitivity. However, many biological problems require ultrasensitive genetic deconvolution of cell mixtures. The detection of fetal cells in cervical samples from pregnant women presents such a challenge. Methods: Using deep sequencing of microhaplotype loci and analysis of the error profiles in genetically pure and mixed samples, we increase the sensitivity for minor allele detection and demonstrate the potential for paternal allele detection in cervical samples from pregnant women. Results: Using artificial DNA mixtures, we show that unique alleles can be readily identified in mixtures with a ratio of 1 in 10,000 (0.01%). By applying this method on ten non-invasive cervical samples from pregnant women we could identify the presence of paternal alleles in half of the samples. Unexpectedly, we demonstrate those alleles not to be of fetal origin but of sperm. Conclusions: We developed a novel method that can deconvolute genetic mixtures to at least 0.01%, which can be applied to study minor allele fractions in genetic mixtures and, for example, could aid in trophoblast cell detection in maternal blood or cervix. Our results demonstrate the importance of discriminating fetal and paternal alleles when analyzing cervical fetal cells and may pinpoint a potential cause of erroneous results in previous trophoblast isolation studies.

  Study EGAS00001007057

• PD-L1 blockade immunotherapy rewires cancer-induced emergency myelopoiesis

  Introduction Immune checkpoint blockade (ICB) immunotherapy has revolutionized cancer treatment, demonstrating exceptional clinical responses in a wide range of cancers. Despite the success, a significant proportion of patients still fail to respond, highlighting the existence of unappreciated mechanisms of immunotherapy resistance. Delineating such mechanisms is paramount to minimize immunotherapy failures and optimize the clinical benefit. Methods In this study, we treated tumour-bearing mice with PD-L1 blockage antibody (aPD-L1) immunotherapy, to investigate its effects on cancer-induced emergency myelopoiesis, focusing on bone marrow HSPCs. We examined the impact of aPD-L1 treatment on HSPC quiescence, proliferation, transcriptomic profile, and functionality. Results Herein, we reveal that aPD-L1 in tumour-bearing mice targets the hematopoietic stem and progenitor cells (HSPCs) in the bone marrow (ΒΜ), mediating their exit from quiescence and promoting their proliferation. Notably, disruption of the PDL1/PD1 axis induces transcriptomic reprogramming in HSPCs, observed in both individuals with Hodgkin lymphoma (HL) and tumour-bearing mice, shifting towards an inflammatory state. Furthermore, HSPCs from aPDL1-treated mice demonstrated resistance to cancer-induced emergency myelopoiesis, evidenced by a lower generation of MDSCs compared to control-treated mice. Discussion: Our findings shed light on unrecognized mechanisms of action of ICB immunotherapy in cancer, which involves targeting of BM-driven HSPCs and reprogramming of cancer-induced emergency myelopoiesis.

  Study EGAS00001007873

• Double mutant DNMT3A AML: a unique subtype

  Mutation of DNMT3A, encoding a de novo methyltransferase essential for cytosine methylation, is a common early event in clonal hematopoiesis (CH) and adult acute myeloid leukemia (AML). Spontaneous deamination of methylated cytosines damages DNA, which is repaired by the base excision repair (BER) enzymes MBD4 and TDG. Congenital MBD4-deficiency has been linked to early-onset CH and AML, and is marked by exceedingly high levels of DNA damage and mutation of DNMT3A. Strikingly, wildtype (WT) DNMT3A binds TDG, thereby potentiating its repair activity. Since TDG is the only remaining BER enzyme in MBD4-deficient AML patients capable of repairing methylation damage, we investigated whether mutant DNMT3A negatively affects the repair function of TDG. We found that, whereas WT DNMT3A stimulates TDG function, mutant DNMT3A impairs TDG-mediated repair of DNA damage in vitro. In light of this finding and to extrapolate our observations to the broader AML patient population, we investigate here the genetic profiles and survival outcomes of AML patients with single (SM) versus double mutant (DM) DNMT3A. DM DNMT3A AML patients show a characteristic driver mutation landscape and reduced overall survival when compared to SM DNMT3A AML patients. Importantly, whole-genome sequencing showed a trend for increased DNA damage in primary DM DNMT3A AML samples, especially when DNMT3A mutations are located at the DNMT3A-TDG interaction interface.

  Study EGAS00001007966

• Genome wide variation in the Angolan Namib desert reveals unique Pre-Bantu ancestry

  The populations of the Angolan Namib Desert have been largely neglected in previous surveys of the genomic landscape of southern Africa. Although at present, the Namib is culturally dominated by Southwest Bantu-speaking cattle-herders, the region exhibits an extraordinary ethnographic diversity which includes an array of semi-nomadic peoples whose subsistence strategies fall outside the traditional division between foraging and food production and can thus be referred to as “peripatetic”. Among these small-scale populations are the last speakers of the Kwadi branch of the Khoe-Kwadi language family associated with the introduction of pastoralism into southern Africa (Kwepe), as well as a range of groups whose origins remain enigmatic (Kwisi, Twa and Tjimba). Using genome-wide data from 208 individuals belonging to nine ethnically diverse groups from the Angolan Namib and adjacent areas (Kwepe, Kwisi, Twa, Tjimba, !Xun, Kuvale, Himba, Nyaneka, Ovimbundu) in combination with published data from other regions of Africa, we reconstruct in detail the histories of contact emerging from pre-historic migrations to southern Africa and show that peripatetic groups from southwestern Angola stand out for exhibiting elevated levels of an unique, regionally-specific and highly divergent Pre-Bantu ancestry. These findings highlight the importance of the Namib for understanding the deep genetic structure of Africa.

  Study EGAS00001007011

• Germline whole genome sequencing of patients with Li-Fraumeni syndrome

  Five hundred ng to 1 ug of genomic DNA was submitted to The Centre for Applied Genomics (TCAG) at The Hospital for Sick Children for genomic library preparation and whole genome sequencing. DNA samples were quantified using the Qubit High Sensitivity Assay and purity was assessed using the Nanodrop OD 260/280 ratio. Approximately 500-700 ng of DNA was used as input material for library preparation using the Illumina TruSeq PCR-free DNA Library Prep Kit following the manufacturer’s recommended protocol. In brief, DNA was fragmented to 400 bp on average using sonication on a Covaris LE220 instrument. Fragmented DNA was end-repaired, A-tailed and indexed TruSeq Illumina adapters with overhang-T were added to the DNA. Libraries were then validated on a Bioanalyzer DNA High Sensitivity chip to check for size and absence of primer dimers and quantified by qPCR using Kapa Library Quantification Illumina/ABI Prism Kit protocol (KAPA Biosystems). Validated libraries were pooled in equimolar quantities and paired-end sequenced on an Illumina HiSeq X platform following Illumina’s recommended protocol to generate paired-end reads of 150-bases in length and an average depth of 40X. FASTQ files were aligned to the hg19 reference genome using BWA-mem v0.78. PCR duplicates were marked with Picard MarkDuplicates v1.1.08, and base recalibration and realignment was performed using GATK v2.8.1.

  Study EGAS00001007061

• BMP4 and temozolomide synergize in the majority of patient derived glioblastoma cultures

  The poor prognosis of patients with glioblastoma is partly due to resistance to standard treatment, which combines chemoradiation using temozolomide (TMZ). The concept of cancer stem cells provided new insight into tumor resistance and management, and isolated Glioblastoma stem-like cells (GSCs) which contribute to this therapy resistance. Bone morphogenetic protein 4 (BMP4) is a differentiation factor that stimulates astroglial differentiation of such GSCs, thereby reducing their self-renewal. Exposing GSCs to BMP4 could therefore sensitize them to chemotherapy with TMZ, but previous studies with various treatment paradigms have produced contradictory results in this matter. Therefore, we re-defined the optimal treatment paradigm for the combination of TMZ and BMP4 and subsequently re-assessed the inter-tumoral variability in response to co-treatment with TMZ+BMP4 in vitro. First, the simultaneous treatment was found more effective than sequential therapy. Second, after application of our optimized treatment protocol, 70% of a total of 20 patient-derived glioblastoma cultures showed synergy when the treatments were combined. The synergistic effect was a result of increased cellular apoptosis and not of decreased proliferation. Comparative RNA-sequencing suggested that combination treatment results in decreased MAPK signaling. We conclude that addition of BMP4 to standard therapy with TMZ could enlarge the group of patients with good therapeutic response.

  Study EGAS00001007095

• Structural variant analysis of homologous recombination-deficient genomes

  Homologous recombination (HR) deficiency causes DNA breaks and cytogenetic aberrations. Paradoxically, the types of DNA rearrangements specifically associated with HR-deficient cancers only minimally impact chromosomal structure. Addressing this, we combined a genome graph analysis of short-read whole genome sequencing (WGS) profiles across thousands of tumors with deep linked-read (LR) WGS of 46 BRCA1 or BRCA2 mutant breast cancers to discover a distinct class of HR deficiency-enriched rearrangements called reciprocal pairs. LR WGS showed that reciprocal pairs with identical rearrangement orientations gave rise to one of two distinct chromosomal outcomes, distinguishable only with long molecule data. While one (cis) outcome corresponded to the copy and pasting of a small segment to a distant site, a second (trans) outcome was a quasi-balanced translocation or multi-megabase inversion with substantial (10kb) duplications at each junction. The full spectrum of reciprocal pair outcomes could be explained by an HR-independent replication restart repair mechanism. LR WGS additionally identified single-strand annealing (SSA) as a BRCA2-deficiency specific repair pathway in human cancers. Replication restart- and SSA-associated SVs improved BRCA1- vs. BRCA2- deficiency classification and identified metastatic cancer cases with favorable chemotherapy responses. Our data reveal classes of BRCA1- and BRCA2-deficiency specific rearrangements as drivers of cytogenetic aberrations in HR deficient cells.

  Study EGAS00001007186

• deep-learning-powered tissue deconvolution for cfDNA

  Plasma cell-free DNA (cfDNA) is a noninvasive biomarker for cell death of all organs. Deciphering the tissue origin of cfDNA can reveal abnormal cell death because of diseases, which has great clinical potential in disease detection and monitoring. Despite the great promise, the sensitive and accurate quantification of tissue-derived cfDNA remains challenging to existing methods due to the limited characterization of tissue methylation and the reliance on unsupervised methods. To fully exploit the clinical potential of tissue-derived cfDNA, here we present one of the largest comprehensive and high-resolution methylation atlas based on 521 noncancer tissue samples spanning 29 major types of human tissues. We systematically identified fragment-level tissue-specific methylation patterns and extensively validated them in orthogonal datasets. Based on the rich tissue methylation atlas, we develop the first supervised tissue deconvolution approach, a deep-learning-powered model, cfSort, for sensitive and accurate tissue deconvolution in cfDNA. On the benchmarking data, cfSort showed superior sensitivity and accuracy compared to the existing methods. We further demonstrated the clinical utilities of cfSort with two potential applications: aiding disease diagnosis and monitoring treatment side effects. The tissue-derived cfDNA fraction estimated from cfSort reflected the clinical outcomes of the patients. In summary, the tissue methylation atlas and cfSort enhanced the performance of tissue deconvolution in cfDNA, thus facilitating cfDNA-based disease detection and longitudinal treatment monitoring.

  Study EGAS00001007213

• ATAC-Seq of CD4 T cell subsets

  Blood samples were taken from 4 age-matched healthy volunteers (two males and two females), and naïve CD4 T cells were isolated using the naïve CD4+ T Cell Isolation Kit II, human (Miltenyi Biotec). Two hundred thousand cells were plated in a 96-well plate, activated with Dynabeads Human T-activator CD3/CD28 (bead-to-cell ratio of 1:1, as recommended by the manufacturer) (ThermoFisher Scientific) and polarized towards CD4 T cell subtypes (Teff, Th1, Th2, rTh17, pTh17, and Treg cells) with specific cocktails of recombinant cytokines and neutralizing antibodies (see details in the material and methods) for each subset. At day 4 post-stimulation, forty thousand cells were used to perform ATAC-Seq on CD4 T cell subtypes. Again, crude nuclei of live CD4+ T cell subsets were treated with Tagment DNA buffer and Tagment DNA Enzyme (Nextera DNA Library Prep Kit, Illumina), and then the DNA was purified by MinElute PCR Purification Kit (Qiagen). Transposed DNA fragments were amplified using specific adapters followed by purification with MinElute PCR Purification Kit (Qiagen). Fragments from 240-360pb were selected in the PippinHT system (Sage Science). The quality of the library and its DNA concentration were assessed by Bioanalyzer instruments (Agilent Technologies) and ultimately submitted for sequencing using Illumina HiSeq 2500 sequencer, V4 chemistry.

  Study EGAS00001007345

• Towards standardized whole exome sequencing (WES) for cancer patients: lessons from a multicentric pilot study

  A growing number of druggable targets and national initiatives for precision oncology necessitate broader genomic profiling for many cancer patients. Whole exome sequencing (WES) offers unbiased analysis of the entire coding sequence, segmentation-based detection of copy number alterations (CNAs), and accurate determination of complex biomarkers including tumor mutational burden (TMB), homologous recombination repair deficiency (HRD), and microsatellite instability (MSI). To assess the inter-institution variability of clinical WES, we performed a comparative pilot study between German Centers of Personalized Medicine (ZPMs) from five participating institutions. Tumor and matched normal DNA from 30 patients were analyzed using custom sequencing protocols and bioinformatic pipelines. Calling of somatic variants was highly concordant with a sensitivity between 91-95% and a positive predictive value (PPV) of 82-95% compared with a three-institution consensus and full agreement for 16 of 17 druggable targets. Deviations were observed for cases with low VAF or coverage, differing annotations, and different filter protocols. CNAs showed overall agreement in 76% for the genomic sequence with high wet-lab variability. Complex biomarkers correlated strongly between institutions (HRD: 0.79-1, TMB: 0.97-0.99) and all institutions agreed on microsatellite instability. This study will contribute to the development of quality control frameworks for comprehensive genomic profiling and sheds light onto parameters that require stringent standardization.

  Study EGAS00001007363

• Machine Learning Signal Enrichment for Ultrasensitive Plasma Tumor Burden Monitoring

  In solid tumor oncology, circulating tumor DNA (ctDNA) is poised to transform care through accurate assessment of minimal residual disease (MRD) and therapeutic response monitoring. To overcome the sparsity of ctDNA fragments in low tumor fraction (TF) settings and increase MRD sensitivity, we previously leveraged genome-wide mutational integration through plasma whole genome sequencing (WGS). We now introduce MRD-EDGE, a composite machine learning-guided WGS ctDNA single nucleotide variant (SNV) and copy number variant (CNV) detection platform designed to increase signal enrichment. MRD-EDGESNV uses deep learning and a ctDNA-specific feature space to increase SNV signal-to-noise enrichment in WGS by 300X compared to previous WGS error suppression. MRD-EDGECNV also reduces the degree of aneuploidy needed for ultrasensitive CNV detection through WGS from 1 Gb to 200 Mb, thereby expanding its applicability to a wider range of solid tumors. We harness the improved performance to identify MRD following surgery in multiple cancer types, track changes in tumor burden in response to neoadjuvant immunotherapy in non-small cell lung cancer (NSCLC) and demonstrate ctDNA shedding in precancerous colorectal adenomas. Finally, the radical signal-to-noise enrichment in MRD-EDGESNV enables de novo mutation calling in melanoma and small cell lung cancer (SCLC) without matched tumor, yielding clinically informative TF monitoring for patients on immune checkpoint inhibition (ICI).

  Study EGAS00001007451

• Machine learning guided signal enrichment for ultrasensitive plasma tumor burden monitoring

  In solid tumor oncology, circulating tumor DNA (ctDNA) is poised to transform care through accurate assessment of minimal residual disease (MRD) and therapeutic response monitoring. To overcome the sparsity of ctDNA fragments in low tumor fraction (TF) settings and increase MRD sensitivity, we previously leveraged genome-wide mutational integration through plasma whole genome sequencing (WGS). We now introduce MRD-EDGE, a composite machine learning-guided WGS ctDNA single nucleotide variant (SNV) and copy number variant (CNV) detection platform designed to increase signal enrichment. MRD-EDGESNV uses deep learning and a ctDNA-specific feature space to increase SNV signal-to-noise enrichment in WGS by 300X compared to previous WGS error suppression. MRD-EDGECNV also reduces the degree of aneuploidy needed for ultrasensitive CNV detection through WGS from 1 Gb to 200 Mb, thereby expanding its applicability to a wider range of solid tumors. We harness the improved performance to identify MRD following surgery in multiple cancer types, track changes in tumor burden in response to neoadjuvant immunotherapy in non-small cell lung cancer (NSCLC) and demonstrate ctDNA shedding in precancerous colorectal adenomas. Finally, the radical signal-to-noise enrichment in MRD-EDGESNV enables de novo mutation calling in melanoma and small cell lung cancer (SCLC) without matched tumor, yielding clinically informative TF monitoring for patients on immune checkpoint inhibition (ICI).

  Study EGAS00001007545

• Cell culture differentiation and proliferation conditions influence the in vitro regeneration of the human airway epithelium

  The human airway mucociliary epithelium can be recapitulated in vitro using primary cells cultured in an Air-Liquid Interface (ALI), a reliable surrogate to perform pathophysiological studies. As tremendous variations exist between media used for ALI-cultured human airway epithelial cells, our study aimed to evaluate at a single-cell level the impact of several media (e.g. BEGMTM, PneumaCultTM, “Half&Half” and “Clancy”) on cell type distribution. Our work revealed the impact of these media on cell composition, gene expression profile, cell signaling or epithelial shape. We found higher proportions of multiciliated cells in PneumaCultTM-ALI and Half&Half, stronger EGF signaling from basal cells in BEGMTM-ALI, differential expression of the SARS-CoV-2 entry factor ACE2, and distinct secretomes transcripts depending on media used. We also established that proliferation in PneumaCultTM-Ex Plus favored secretory cells fate, showing the key influence of proliferation media on late differentiation epithelial characteristics. Altogether, our data offer a comprehensive repertoire for evaluating the effects of culture conditions on airway epithelial differentiation and will help to choose the most relevant medium according to the phenomenon to be investigated such as cilia or mucus biology or viral infection. We detail useful parameters that should be explored to document airway epithelial cell fate and shaping.

  Study EGAS00001007572

• Patient-derived tumoroids from CIC::DUX4 rearranged sarcoma identify MCL1 as a therapeutic target

  High-risk sarcomas, such as metastasized and relapsed Ewing and CIC-rearranged sarcomas, currently lack effective curative systemic treatment options. Precision medicine approaches offer hope and among them ex vivo drug response profiling of patient-derived tumor cells emerges as a promising novel tool to identify effective drugs for individual patients. Here, we established ex vivo culture conditions to propagate Ewing and CIC::DUX4-sarcomas as tumoroids in a co-clinical manner. These models retained their original molecular and functional characteristics, including recurrent ARID1A mutations, and serve as tumor avatars suitable for large-scale drug testing. Indeed, screening a large drug library on a small living biobank of such tumors not only revealed distinct response differences between the two entities, but also identified an exceptional dependency of CIC::DUX4-sarcoma cells on MCL1. Mechanistically. MCL1 was identified as a direct target gene of the CIC::DUX4 fusion protein. Genetic and pharmacological inhibition of MCL1 induced rapid apoptosis in CIC::DUX4-sarcoma cells and inhibited tumor growth in a xenograft model. Thus, MCL1 represents a potential novel therapeutic target for CIC::DUX4-sarcoma. Overall, our study highlights the feasibility of co-clinical drug response profiling for individual sarcoma cases and suggests that prospective tests, to evaluate its clinical benefit are highly warranted.

  Study EGAS00001008039

• This study generated Oxford Nanopore long read sequencing data of cancer cell line mixtures for validating long-read variant calls in cancer genomics

  Long-read sequencing (LRS) improves genome alignment and facilitates resolving variants in complex genomic regions, making it a promising approach for cancer variant detection and biomarker discovery. Here, we evaluate the performance of LRS to detect somatic variants across different tumour purities and sequencing depths by comparing to somatic variants from short-read sequencing. We generated experimental mixtures of cancer cell lines and matched normal cell lines to simulate 10 tumour purities (ranging from 0 to 100% tumour content). This resulted in 22 samples which were sequenced using whole genome LRS to a targeted depth of 60x and somatic variants were identified. We downsampled sequencing data to explore the optimal read depth for somatic variant detection. Our results show that long-read variant calling tools achieve recall rates comparable to short-read gold standards. While a tumour sequencing depth of 30x to 60× is generally sufficient for detecting common variants, particularly structural variants (SVs), sequencing the matched normal sample at adequate depth is crucial for accuracy. Notably, we found variants unique to LRS that may represent real and previously undetected events. This study highlights key factors for optimising cancer genome sequencing with LRS and provides a comprehensive dataset of cell line mixtures for future research and tool development.

  Study EGAS00001008107

• Prognostic relevance of microenvironmental factors CD163 and CD8 combined with EZH2 and chromosome 18 gain in a validation cohort of follicular lymphoma patients of the Lunenburg Lymphoma Biomarker Consortium

  Background: In follicular lymphoma (FL), studies addressing the prognostic value of microenvironment-related immunohistochemical (IHC) markers and tumor cell-related genetic markers have yielded conflicting results, precluding implementation in practice. Therefore, the Lunenburg Lymphoma Biomarker Consortium (LLBC) performed a validation study for published markers. Methods: To maximize sensitivity, an end-of-spectrum design was applied for 122 uniformly immunochemotherapy-treated FL patients retrieved from international trials and registries; early failure (EF): progression or lymphoma-related death <2 years versus long remission: response duration of >5 years. IHC staining for T-cells and macrophages was performed on tissue microarrays from initial biopsy and scored with a validated computer-assisted protocol. Shallow whole-genome and deep targeted sequencing was performed on the same samples. Results: 96/122 cases with complete molecular and immunohistochemical data were included in the analysis. EZH2 wild-type (p=0.006), gain of chromosome 18 (p=0.002), low percentages of CD8+ cells (p=0.011) and CD163+ areas (p=0.038) were associated with EF. No significant differences in other markers were observed, thereby refuting previous claims on their prognostic significance. Conclusion: Using an optimized study design, this LLBC study validates wild-type EZH2 status, gain of chromosome 18, low percentages of CD8+ cells and CD163+ area as predictors of EF to immunochemotherapy in FL.

  Study EGAS00001002049

• The double-hit signature identifies double-hit diffuse large B-cell lymphoma with genetic events cryptic to FISH​

  High-grade B-cell lymphomas with MYC and BCL2 and/or BCL6 rearrangements (HGBL-DH/THs) include a group of diffuse large B-cell lymphomas (DLBCLs) with inferior outcomes after standard chemoimmunotherapy. We recently described a gene expression signature that identifies 27% of germinal center B-cell DLBCLs (GCB-DLBCLs) as having a double-hit-like expression pattern (DHITsig) and inferior outcomes; however, only half of these cases have both MYC and BCL2 translocations identifiable using standard breakapart fluorescence in situ hybridization (FISH). Here, 20 DHITsig+ GCB-DLBCLs apparently lacking MYC and/or BCL2 rearrangements underwent whole-genome sequencing. This revealed 6 tumors with MYC or BCL2 rearrangements that were cryptic to breakapart FISH. Copy-number analysis identified 3 tumors with MYC and 6 tumors with MIR17HG gains or amplifications, both of which may contribute to dysregulation of MYC and its downstream pathways. Focal deletions of the PVT1 promoter were observed exclusively among DHITsig+ tumors lacking MYC translocations; this may also contribute to MYC overexpression. These results highlight that FISH fails to identify all HGBL-DH/THs, while revealing a range of other genetic mechanisms potentially underlying MYC dysregulation in DHITsig+ DLBCL, suggesting that gene expression profiling is more sensitive for identifying the biology underlying poor outcomes in GCB-DLBCL.

  Study EGAS00001004285

• Bulk-tissue RNA-sequencing paired nuclear and cytoplasmic fractions of anterior prefrontal cortex, cerebellar cortex and putamen tissues from post-mortem neuropathologically-confirmed control individuals.

  Gaining insight into the genetic regulation of gene expression in human brain is key to the interpretation of genome-wide association studies for major neurological and neuropsychiatric diseases. Expression quantitative trait loci (eQTL) analyses have largely been used to achieve this, providing valuable insights into the genetic regulation of steady-state RNA in human brain, but not distinguishing between molecular processes regulating transcription and stability. RNA quantification within cellular fractions can disentangle these processes in cell types and tissues which are challenging to model in vitro. We investigated the underlying molecular processes driving the genetic regulation of gene expression specific to a cellular fraction using allele-specific expression (ASE). Applying ASE analysis to genomic and transcriptomic data from paired nuclear and cytoplasmic fractions of anterior prefrontal cortex, cerebellar cortex and putamen tissues from 4 post-mortem neuropathologically-confirmed control human brains, we demonstrate that a significant proportion of genetic regulation of gene expression occurs post-transcriptionally in the cytoplasm, with genes undergoing this form of regulation more likely to be synaptic. These findings have implications for understanding the structure of gene expression regulation in human brain, and importantly the interpretation of rapidly growing single-nucleus brain RNA-sequencing and eQTL datasets, where cytoplasm-specific regulatory events could be missed.

  Study EGAS00001006380

• The Dutch Microbiome Project Batch 1

  The Dutch Microbiome Project (DMP) data includes shotgun metagenomic sequencing of faecal samples 8,208 Dutch individuals. Paired-end sequencing was performed using Illumina HiSeq 2000 platform. Data is archived in two batches to facilitate easier data access and upload to EGA. Batch 1 of DMP includes 4396 samples.

  Dataset EGAD00001007027

• The patterns and dynamics of genomic instability in metastatic pancreatic cancer

  Pancreatic cancer is an aggressive malignancy with a five-year mortality of 97–98%, usually due to widespread metastatic disease. Previous studies indicate that this disease has a complex genomic landscape, with frequent copy number changes and point mutations, but genomic rearrangements have not been characterized in detail. Despite the clinical importance of metastasis, there remain fundamental questions about the clonal structures of metastatic tumours, including phylogenetic relationships among metastases, the scale of ongoing parallel evolution in metastatic and primary sites, and how the tumour disseminates. Here we harness advances in DNA sequencing to annotate genomic rearrangements in 13 patients with pancreatic cancer and explore clonal relationships among metastases. We find that pancreatic cancer acquires rearrangements indicative of telomere dysfunction and abnormal cell-cycle control, namely dysregulated G1-to-S-phase transition with intact G2–M checkpoint. These initiate amplification of cancer genes and occur predominantly in early cancer development rather than the later stages of the disease. Genomic instability frequently persists after cancer dissemination, resulting in ongoing, parallel and even convergent evolution among different metastases. We find evidence that there is genetic heterogeneity among metastasis-initiating cells, that seeding metastasis may require driver mutations beyond those required for primary tumours, and that phylogenetic trees across metastases show organ-specific branches. These data attest to the richness of genetic variation in cancer, brought about by the tandem forces of genomic instability and evolutionary selection.

  Study EGAS00000000064

• Nanopore sequencing enables allelic phasing of FLG loss-of-function variants, intragenic copy number variation and methylation status in atopic dermatitis and ichthyosis vulgaris

  Loss-of-function (LoF) variants in the FLG gene are associated with ichthyosis vulgaris (IV) and atopic dermatitis (AD). IV and AD patients with two FLG LoF variants are reported to have a more severe disease course. However, subsets of compound heterozygous patients display phenotypic heterogeneity despite a prediction of a severe phenotype from their genotype. The underlying genetic mechanism of this phenotypic variability is unknown. One possibility is that two LoF FLG variants could be located in cis (on the same allele), thus altering the predicted gene dosage. To investigate the allelic contribution of compound heterozygous FLG LoF mutations to IV phenotypic variability, we employ long reads from Oxford Nanopore Technologies, utilising two methodologies to phase the entire coding region of FLG into parental alleles: an amplicon-sequencing approach and an adaptive-sampling strategy. We investigated the allelic phasing of FLG LoF variants and intragenic-repeat copy-number variations (CNV) of 21 subjects with known compound heterozygous genotypes and mixed IV severity. Our results suggest that IV severity is not linked to the presence of cis variants in our IV patient cohort. Nanopore reads enabled detection and accurate phasing of CNVs and LoF variants into parental alleles, providing a framework for future in-depth genetic analysis. These results highlight the usefulness of utilising long reads for interrogating the FLG gene and other highly repetitive genes for genetic stratification related to clinical phenotypes.

  Study EGAS50000000166

• The tissue origin of highly elevated cell-free DNA in patients with and without cancer

  Forty years ago, the first measurements of cell free DNA (cfDNA) from cancer patients revealed that its total concentration was often elevated compared to healthy controls. The basis for this increase, and in particular the source of the extra cfDNA in such patients, has never been determined. To address this issue, we assessed methylation patterns of cfDNA from 57 individuals, 34 of whom had unequivocally elevated cfDNA concentrations (i.e., >15 ng/mL of plasma). We found that the excess cfDNA in cancer patients did not come from either neoplastic cells or surrounding cells from the tumor's tissue of origin. Instead, 80% of the cfDNA was derived from neutrophils, B cells or T cells, with neutrophils accounting for ~2/3. B cells contributed more than T cells despite the great excess of T cells over B cells in the blood. To see if this result was generalizable to other clinical situations in which elevated cfDNA concentrations have been described, we evaluated nine patients before and one day following surgery. After surgical resection, cfDNA concentration increased by an average of 16.1-fold, as expected. The majority of this increase (55%) was contributed by leukocytes, though damaged hepatocytes contributed a substantial minority (31% of increase). These data suggest that the elevated concentrations of cfDNA in cancer patients, as with patients after surgery, arises largely from dying leukocytes—not from the neoplastic cells or the tissue of origin of the cancer.

  Study EGAS00001005400

• METABRIC miRNA landscape

  MicroRNAs (miRNAs) are key players in cancer pathogenesis1. They display differential expression across breast cancer subtypes, and play oncogenic and tumor-suppressive roles2-6. A systematic study of the global miRNA landscape in breast cancer and the factors shaping it has not been undertaken. Here we report the results of profiling miRNA expression in 1,302 breast tumors with detailed clinical annotation and long-term follow-up, and for which matching genomic and mRNA expression data were available7. This provides a comprehensive view of the quantity, distribution and variation of the miRNA population and dissects how much genomic, transcriptional and post-transcriptional events contribute to miRNA expression architecture. The key clinical parameters and cellular pathways related to the miRNA landscape are identified, exemplified by context-dependent interaction with cell adhesion and Wnt signaling. Strikingly, in breast cancer miRNAs appear to act as modulators of mRNA-mRNA interactions rather than molecular switches. We demonstrate an important modulatory role for miRNAs in the biology of breast tumors devoid of somatic copy number aberrations, a common subtype where immune response is prominent. Remarkably, prognostic miRNA signatures derived in this subtype are consistently prognostic across several other subtypes and validate in external cohorts. These findings represent a new framework to study the biology of miRNA in human cancer.

  Study EGAS00000000122

• Machine learning to detect the SINEs of cancer

  We previously described an approach called RealSeqS to evaluate aneuploidy in plasma cell-free DNA (cfDNA) through the amplification of ~350,000 repeated elements with a single primer. We hypothesized that an unbiased evaluation of the large amount of sequencing data obtained with RealSeqS might reveal other differences between plasma samples from patients with and without cancer. This hypothesis was tested through the development of a novel machine-learning approach called Alu Profile Learning Using Sequencing (A-PLUS) and its application to samples from 5108 individuals, 2037 with cancer and the remainder without cancer. Samples from cancer patients and controls were pre-specified into four cohorts used for: 1) model training, 2) analyte integration and threshold determination, 3) validation, and 4) reproducibility. A-PLUS alone provided a sensitivity of 40.5% across 11 different cancer types in the Validation Cohort, at a specificity of 98.5%. Combining A-PLUS with aneuploidy and 8 common protein biomarkers detected 51% of 1167 cancers at 98.9% specificity. We found that part of the power of A-PLUS could be ascribed to a single feature – the global reduction of AluS sub-family elements in the circulating DNA of cancer patients. We confirmed this reduction through the analysis of another independent dataset obtained with a very different approach (whole genome sequencing). The evaluation of Alu elements therefore has the potential to enhance the performance of several methods designed for the earlier detection of cancer.

  Study EGAS00001007169

• Neoadjuvant immune checkpoint blockade in women with mismatch repair deficient endometrial cancer

  Neoadjuvant immune checkpoint blockade (ICB) has shown unprecedented activity in mismatch repair deficient (MMRd) colorectal cancers, but its effectiveness in MMRd endometrial cancer (EC) remains unknown. In this investigator-driven, phase I, feasibility study (NCT04262089), 10 women with MMRd EC of any grade, planned for primary surgery, received two cycles of neoadjuvant pembrolizumab (200 mg IV) every three weeks. A pathologic response (primary objective) was observed in 5/10 patients, with 2 patients showing a major pathologic response. No patient achieved a complete pathologic response. A partial radiologic response (secondary objective) was observed in 3/10 patients, 5/10 patients had stable disease and 2/10 patients were non-evaluable on Magnetic Resonance Imaging (MRI). All patients completed treatment without severe toxicity (exploratory objective). At median duration of follow-up of 22.5 months, two non-responders experienced disease recurrence. In-depth analysis of the loco-regional and systemic immune response (predefined exploratory objective) showed that monoclonal T cell expansion significantly correlated with treatment response. Tumour-draining lymph nodes displayed clonal overlap with intra-tumoural T cell expansion. All pre-specified endpoints, efficacy in terms of pathologic response as primary endpoint, radiologic response as secondary outcome and safety and tolerability as exploratory endpoint, were reached. Neoadjuvant ICB with pembrolizumab proved safe and induced pathologic, radiologic, and immunologic responses in MMRd EC, warranting further exploration of extended neoadjuvant treatment.

  Study EGAS50000000483

• Biopsy-Derived Organoids in Personalised Early Breast Cancer Care: Challenges of Tumour Purity and Normal Cell Overgrowth Cap Their Practical Utility

  The ability to establish organoids composed exclusively of tumour rather than healthy cells is essential for their implementation into clinical practice. Organoids have in recent years emerged as a powerful tool to expand patient material in culture and generate modifiable 3D models derived from humans or animal models. For translational research, they enable the creation of model systems for an ever-increasing number of cell types and diseases. And in personalised medicine, they potentially allow for functional drug testing with high predictive power in certain settings. We found that using biopsy material from untreated, early-stage primary breast cancer patients poses significant challenges for consistently culturing tumour cells as organoids. Specifically, we observed frequent outgrowth of genetically normal, non-cancerous epithelial cells. We analysed >100 biopsy samples from early-stage breast cancer and present our large collection of >70 organoid lines. We also show methods of assessing successful tumour cell culture in a time and cost-efficient manner, proving a high rate (>85%) of normal cell overgrowth in early-stage breast cancer organoids. Finally, we show a number of successful attempts to culture cancer organoids from mastectomy-derived tissue of advanced, metastatic breast cancer. We conclude that the usefulness of organoids from early breast cancer for translational research and personalised medicine, especially guidance of adjuvant or post-surgical maintenance therapy, is strongly limited by the low success rate of culturing cancerous cells under organoid conditions.

  Study EGAS50000000605

• Single cell transcriptomics in expanded Tregs of APS-1 patients

  Aim of study: Exploring single cell transcriptomics in expanded Tregs of with autoimmune polyendocrine syndrome type 1 (APS-1) patients (global gene expression, immune panel, TCR). N= 9 APS-1 patients and 9 healthy controls Protocol 1. Regulatory T cells (Tregs) were isolated from EDTA-blood, activated with anti-CD3/CD28 and IL2, expanded in cell culture for 14 days and stored at -150℃ in AB serum or FBS + 5% DMSO until use. 2. Cells were thawed, run through a Live/Dead column and carefully counted. 10000 Tregs were used as input for the 10x protocol. 3. cDNA was generated and amplified using Single Cell VDJ 5’ Gel Beads, Chromium Next GEM Chip K Single cell Kit, Next GEM Single Cell 5’ GEM Kit v2 and 5’ Feature Barcode Kit (all from 10x genomics). 4. Gene expression (GEX) libraries were generated by using the Library Construction Kit from 10x genomics. 5. T-cell receptor (TCR) libraries were generated using the Single Cell Human TCR Amplification Kit and the Library Construction Kit from 10x genomics. 6. 10x Genomics Target Hybridization Kit and Human Immunology Panel was used with GEX libraries. Only 8 patient samples and 8 control samples were used for the immune panel. 7. All libraries have a unique sample index (Dual Index TT Set A). 8. All samples were sequenced on Illumina NovaSeq 6000 on a NovaSeq S2 flow cell.

  Study EGAS50000000181

• Multi-omics bulk and single-cell profiling of epithelioid sarcoma

  Epithelioid sarcoma (EpS) is an aggressive sarcoma, characterized by the loss of SMARCB1 expression. EpS is traditionally classified as distal or proximal according to clinicopathological features, but its molecular characteristics remain largely unknown. To establish an EpS molecular classification and uncover determinants of inter- and intra-patient heterogeneity in EpS, we used multi-omics profiling and integrated the genomic, transcriptional and methylome landscapes with single-cell RNA sequencing on fresh samples as well as spatial transcriptomics. We identified two molecular subtypes of EpS: “distal-like” and “proximal-like”, which were distinct from the histological subtypes. Distal-like tumors include all morphological distal subtypes and a subset of proximal and hybrid EpS. They express a specific molecular single-cell derived epithelial-to-mesenchymal transition signature, which associates with improved patient survival and includes Desmoglein 2 (DSG2), which we identify as a potential routine immunohistochemical biomarker to improve diagnosis. Distal-like EpS also display increased peri-tumoral CD8+ T cell infiltrates and specific tumor-immune cell interactions. Conversely, proximal-like EpS harbor a higher inter-tumoral molecular heterogeneity with some cases resembling other SMARCB1-deficient tumors by DNA methylation profiling, and increased intra-tumoral pro-tumoral macrophage infiltrates. Our study introduces a novel molecular classification of EpS with prognostic value, associated diagnostic biomarkers and distinct tumor immune microenvironments, which extends beyond traditional clinicopathological classification. This paves the way for precision medicine-based therapeutic strategies for patients with EpS.

  Study EGAS50000000973