Multi-omic and functional analysis for classification and treatment of sarcomas with FUS-TFCP2 or EWSR1-TFCP2 fusions

Linking clinical multi-omics with mechanistic studies may improve the understanding of rare cancers. We leverage two precision oncology programs to investigate rhabdomyosarcoma with FUS/EWSR1-TFCP2 fusions, an orphan malignancy without effective therapies. All tumors exhibit outlier ALK expression, partly accompanied by intragenic deletions and aberrant splicing resulting in ALK variants that are oncogenic and sensitive to ALK inhibitors. Additionally, recurrent CKDN2A/MTAP co-deletions provide a rationale for PRMT5-targeted therapies. Functional studies show that FUS-TFCP2 blocks myogenic differentiation, induces transcription of ALK and truncated TERT, and inhibits DNA repair. Unlike other fusion-driven sarcomas, TFCP2-rearranged tumors exhibit genomic instability and signs of defective homologous recombination. DNA methylation profiling demonstrates a close relationship with undifferentiated sarcomas. In two patients, sarcoma was preceded by benign lesions carrying FUS-TFCP2, indicating stepwise sarcomagenesis. This study illustrates the potential of linking precision oncology with preclinical research to gain insight into the classification, pathogenesis, and therapeutic vulnerabilities of rare cancers.

NCT/DKFZ MASTER handbook of interpreting whole-genome, transcriptome, and methylome data for precision oncology

Analysis of selected cancer genes has become an important tool in precision oncology but cannot fully capture the molecular features and, most importantly, vulnerabilities of individual tumors. Observational and interventional studies have shown that decision-making based on comprehensive molecular characterization adds significant clinical value. However, the complexity and heterogeneity of the resulting data are major challenges for disciplines involved in interpretation and recommendations for individualized care, and limited information exists on how to approach multilayered tumor profiles in clinical routine. We report our experience with the practical use of data from whole-genome or exome and RNA sequencing and DNA methylation profiling within the MASTER (Molecularly Aided Stratification for Tumor Eradication Research) program of the National Center for Tumor Diseases (NCT) Heidelberg and Dresden and the German Cancer Research Center (DKFZ). We cover all relevant steps of an end-to-end precision oncology workflow, from sample collection, molecular analysis, and variant prioritization to assigning treatment recommendations and discussion in the molecular tumor board. To provide insight into our approach to multidimensional tumor profiles and guidance on interpreting their biological impact and diagnostic and therapeutic implications, we present case studies from the NCT/DKFZ molecular tumor board that illustrate our daily practice. This manual is intended to be useful for physicians, biologists, and bioinformaticians involved in the clinical interpretation of genome-wide molecular information.

Abstract 926: Genomics-based personalized oncology of advanced thymic epithelial tumors

Introduction: Thymic epithelial tumors (TETs) are very rare. Thymoma A and AB have a better prognosis than more aggressive thymoma B, thymic carcinoma (TC) and neuroendocrine tumors of the thymus (NET). While previous efforts such as TCGA have mainly characterized thymomas (Radovich et al., Cancer Cell 2018), the molecular landscape of TCs and NETs is still elusive.

Patients and Methods: Between 03/2014 and 07/2020, we enrolled 44 TET patients (27 TCs, 11 thymomas, 6 NETs) in a prospective observational study (MASTER) conducted by the National Center for Tumor Diseases (NCT) Heidelberg, NCT Dresden and the German Cancer Consortium (DKTK). MASTER applied whole genome/exome sequencing (WGS, n=22; WES, n=22), transcriptome (n=40) and germline analysis to inform therapy recommendations by a dedicated molecular tumor board (MTB). We systematically gathered follow-up data to evaluate outcome and compared progression-free survival (PFS) of the first treatment according to an MTB recommendation (PFS2) to the last prior systemic treatment (PFS1) in each patient (PFS ratio).

Results: Tumor mutational burden (TMB) was low (median=0.99 mutations/Mb, range 0.08-3.48) but higher than in TCGA (p<0.05). TMB was higher in TCs than in thymoma (p<0.05). Most frequently mutated genes were TP53 (30%), CYLD (16%), SETD2 (14%) and KIT (14%). Germline analysis revealed (likely) pathogenic germline alterations in 25% of patients (MUTYH, n=3; BRCA1, n=2; BRCA2, BAP1, CHEK2, FANCA, TP53, MEN1, n=1). A comprehensive analysis of candidate biomarkers of homologous recombination repair (HRR) defects revealed a subgroup of TET patients with a rationale for PARP inhibitor therapy. Unsupervised clustering of RNA sequencing data mainly revealed clustering that correlated with WHO classification. Additionally, TCs clustered in two subgroups that we identified as immunologically hot and cold tumors using immunohistochemistry validation. Overall survival of patients with hot tumors was significantly longer (p<0.05). The MTB recommended therapies for 42 patients (95%), which were implemented in 24 cases (57%). Five patients had a PFS2 > 6 months and a PFS ratio > 1.3. The best outcome was achieved using imatinib in a patient with a KIT mutation (p.W557R). After progression, the MTB recommended ponatinib based on a secondary KIT mutation (p.V654A). The patient was still on ponatinib when the observation period ended.

Conclusion: We demonstrate that comprehensive molecular analysis provides clinically relevant information in a subgroup of TET patients. Thymoma, TCs, and NETs present with different molecular characteristics. Distinction between immunologically hot and cold TCs may have value for risk stratification and therapeutic strategies. PARP inhibition could be a potential new treatment option in a small subgroup of TETs. Molecular testing of KIT, germline analysis and genetic counseling should be recommended for all patients with advanced TETs.

Citation Format: Lino Möhrmann, Lysann Rostock, Małgorzata Oleś, Arne Jahn, Marie Arlt, Nagarajan Paramasivam, Korinna Jöhrens, Luise Rupp, Marc Schmitz, Daniela Richter, Sebastian Uhrig, Martina Fröhlich, Barbara Hutter, Jennifer Hüllein, Elena E. Wolf, Dorothea Hanf, Laura Gieldon, Simon Kreutzfeldt, Christoph E. Heilig, Veronica Teleanu, Daniel B. Lipka, Andreas Mock, Ivan Jelas, Damian T. Rieke, Marcel Wiesweg, Melanie Boerries, Anna L. Illert, Alexander Desuki, Thomas Kindler, Angela M. Krackhardt, C. Benedikt Westphalen, Heidrun Grosch, Leonidas Apostolidis, Albrecht Stenzinger, Irina A. Kerle, Christoph Heining, Daniel Hübschmann, Evelin Schröck, Stefan Fröhling, Hanno Glimm. Genomics-based personalized oncology of advanced thymic epithelial tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 926.

Abstract 4544: Genomic, transcriptomic, functional, and mechanistic characterization of rhabdomyosarcoma with FUS-TFCP2 or EWSR1-TFCP2 fusions

Rhabdomyosarcoma (RMS) is a soft-tissue sarcoma subtype composed of malignant immature precursor cells with myogenic differentiation defined by aberrant expression of the transcription factors MYOD1 and MYOG. Four subtypes are distinguished, characterized by considerable clinical, histologic, and genetic heterogeneity. RMS with fusions of the transcription factor TFCP2 to either FUS or EWSR1 has only recently been observed, but its classification and pathogenesis are unclear. We studied the clinical course, histopathology, and molecular landscape of 12 cases of this new RMS type and determined the functional properties of tumor-specific genetic alterations. Unusually for gene fusion-driven sarcomas, most tumors had highly rearranged genomes, including chromothripsis, and signs of defective homologous recombination DNA repair. All tumors were characterized by extremely high expression of a truncated TERT variant and the receptor tyrosine kinase ALK. The latter was additionally affected by intragenic deletions (33%), which resulted, together with aberrant splicing events, in the expression of shortened ALK variants (58%). Three ALK variants were oncogenic in immortalized cells in vitro and after xenotransplantation in mice and responded variably to different ALK inhibitors. Additional recurrent alterations included CDKN2A/MTAP co-deletions (67%) and mutations in PAPPA2 (25%) encoding an IGFBP5-specific proteinase. DNA methylation analysis of FUS/EWSR1-TFCP2 RMS, along with 19 other soft-tissue sarcoma types, revealed a close relationship with undifferentiated sarcoma but not with other RMS subtypes, suggesting that FUS/EWSR1-TFCP2 RMS is a distinct sarcoma entity possibly arising from a different cell of origin than other RMS types. Transduction of TFCP2 fusions into immortalized human cells conferred anchorage-independent growth and blocked late myogenic differentiation. Genes significantly induced in these cells were also highly expressed in patient tumors, including ALK, TERT, and two known regulators of skeletal muscle cells, IGFBP5 and PTH1R. ACT-seq demonstrated direct binding of FUS-TFCP2 to the ALK and TERT gene loci outside their regular promoters, which correlated with the expression of alternative transcript variants. Finally, FUS-TFCP2 appeared to induce a defect in DNA double-strand repair in immortalized cells, rendering them sensitive to treatment with cisplatin. Together, our study gives insights into the pathogenesis of a new RMS subtype defined by FUS-TFCP2 or EWSR1-TFCP2 fusions and suggests entry points for therapeutic intervention with DNA-damaging agents, ALK inhibitors, and, in the case of additional CDKN2A/MTAP co-deletion, drugs targeting PRMT5.

Citation Format: Julia Schoepf, Sebastian Uhrig, Christoph E. Heilig, Kwang-Seok Lee, Tatjana Walther, Alexander Carazzato, Anna Maria Dobberkau, Dieter Weichenhan, Christoph Plass, Mark Hartmann, Gaurav Diwan, Zunamys Carrero, Claudia R. Ball, Tobias Hohl, Thomas Kindler, Patricia Rudolph-Hähnel, Anna Nilsson, Ingrid Øra, Roland Imle, Ana Banito, Robert Russell, Barbara C. Jones, Daniel B. Lipka, Hanno Glimm, Daniel Hübschmann, Wolfgang Hartmann, Stefan Fröhling, Claudia Scholl. Genomic, transcriptomic, functional, and mechanistic characterization of rhabdomyosarcoma with FUS-TFCP2 or EWSR1-TFCP2 fusions. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4544.

DNA methylation landscape of tumor-associated macrophages reveals pathways, transcription factors and prognostic value relevant to triple-negative breast cancer patients

The accumulation of myeloid cells, particularly tumor-associated macrophages (TAMs), characterizes the tumor microenvironment (TME) of many solid cancers, including breast cancer. Compared to healthy tissue-resident macrophages, TAMs acquire distinct transcriptomes and tumor-promoting functions by largely unknown mechanisms. Here, we hypothesize the involvement of TME signaling and subsequent epigenetic reprogramming of TAMs. Using the 4T1 mouse model of triple-negative breast cancer, we demonstrate that the presence of cancer cells significantly alters the DNA methylation landscape of macrophages and, to a lesser extent, bone marrow-derived monocytes (BMDMs). TAM methylomes, dissected into BMDM-originating and TAM-specific epigenetic programs, implicated transcription factors (TFs) and signaling pathways involved in TAM reprogramming, correlated with cancer-specific gene expression patterns. Utilizing published single-cell gene expression data, we linked microenvironmentally-derived signals to the cancer-specific DNA methylation landscape of TAMs. These integrative analyses highlighted the role of altered cytokine production in the TME (eg, TGF-β, IFN-γ and CSF1) on the induction of specific TFs (eg, FOSL2, STAT1 and RUNX3) responsible for the epigenetic reprogramming of TAMs. DNA methylation deconvolution identified a TAM-specific signature associated with the identified signaling pathways and TFs, corresponding with severe tumor grade and poor prognosis of breast cancer patients. Similarly, immunosuppressive TAM functions were identified, such as induction of the immune inhibitory receptor-ligand PD-L1 by DNA hypomethylation of Cd274. Collectively, these results provide strong evidence that the epigenetic landscapes of macrophages and monocytes are perturbed by the presence of breast cancer, pointing to molecular mechanisms of TAM reprogramming, impacting patient outcomes.

DNA methylation biomarkers for noninvasive detection of triple-negative breast cancer using liquid biopsy

Noninvasive detection of aberrant DNA methylation could provide invaluable biomarkers for earlier detection of triple-negative breast cancer (TNBC) which could help clinicians with easier and more efficient treatment options. We evaluated genome-wide DNA methylation data derived from TNBC and normal breast tissues, peripheral blood of TNBC cases and controls and reference samples of sorted blood and mammary cells. Differentially methylated regions (DMRs) between TNBC and normal breast tissues were stringently selected, verified and externally validated. A machine-learning algorithm was applied to select the top DMRs, which then were evaluated on plasma-derived circulating cell-free DNA (cfDNA) samples of TNBC patients and healthy controls. We identified 23 DMRs accounting for the methylation profile of blood cells and reference mammary cells and then selected six top DMRs for cfDNA analysis. We quantified un-/methylated copies of these DMRs by droplet digital PCR analysis in a plasma test set from TNBC patients and healthy controls and confirmed our findings obtained on tissues. Differential cfDNA methylation was confirmed in an independent validation set of plasma samples. A methylation score combining signatures of the top three DMRs overlapping with the SPAG6, LINC10606 and TBCD/ZNF750 genes had the best capability to discriminate TNBC patients from controls (AUC = 0.78 in the test set and AUC = 0.74 in validation set). Our findings demonstrate the usefulness of cfDNA-based methylation signatures as noninvasive liquid biopsy markers for the diagnosis of TNBC.

Dynamic DNA methylation reveals novel cis-regulatory elements in mouse hematopoiesis

Differentiation of hematopoietic stem and progenitor cells to terminally differentiated immune cells is accompanied by large-scale remodeling of the DNA methylation landscape. Although significant insights into the molecular mechanisms of hematopoietic tissue regeneration were derived from mouse models, profiling of DNA methylation has been hampered by high cost or low resolution using available methods. The recent development of the Infinium Mouse Methylation BeadChip (MMBC) array facilitates methylation profiling of the mouse genome at a single CpG resolution at affordable cost. We extended the RnBeads package to provide a computational framework for the analysis of MMBC data. This framework was applied to a newly generated reference map of mouse hematopoiesis encompassing nine different cell types. Analysis of dynamically regulated CpG sites showed progressive and unidirectional DNA methylation changes from hematopoietic stem and progenitor cells to differentiated hematopoietic cells and allowed the identification of lineage- and cell type-specific DNA methylation programs. Comparison with previously published catalogs of cis-regulatory elements (CREs) revealed 12,856 novel putative CREs that were dynamically regulated by DNA methylation (mdCREs). These mdCREs were predominantly associated with patterns of cell type-specific DNA hypomethylation and could be identified as epigenetic control regions regulating the expression of key hematopoietic genes during differentiation. In summary, we established an analysis pipeline for MMBC data sets and provide a DNA methylation atlas of mouse hematopoiesis. This resource allowed us to identify novel putative CREs involved in hematopoiesis and will serve as a platform to study epigenetic regulation of normal and malignant hematopoiesis.

Usp22 is an intracellular regulator of systemic emergency hematopoiesis

Emergency hematopoiesis is a concerted response aimed toward enhanced protection from infection, involving multiple cell types and developmental stages across the immune system. Despite its importance, the underlying molecular regulation remains poorly understood. The deubiquitinase USP22 regulates the levels of monoubiquitinated histone H2B (H2Bub1), which is associated with activation of interferon responses upon viral infection. Here, we show that in the absence of infection or inflammation, mice lacking Usp22 in all hematopoietic cells display profound systemic emergency hematopoiesis, evident by increased hematopoietic stem cell proliferation, myeloid bias, and extramedullary hematopoiesis. Functionally, loss of Usp22 results in elevated phagocytosis by neutrophilic granulocytes and enhanced innate protection against Listeria monocytogenes infection. At the molecular level, we found this state of emergency hematopoiesis associated with transcriptional signatures of myeloid priming, enhanced mitochondrial respiration, and innate and adaptive immunity and inflammation. Augmented expression of many inflammatory genes was linked to elevated locus-specific H2Bub1 levels. Collectively, these results demonstrate the existence of a tunable epigenetic state that promotes systemic emergency hematopoiesis in a cell-autonomous manner to enhance innate protection, identifying potential paths toward immune enhancement.

Model soups improve performance of dermoscopic skin cancer classifiers

BACKGROUND

Image-based cancer classifiers suffer from a variety of problems which negatively affect their performance. For example, variation in image brightness or different cameras can already suffice to diminish performance. Ensemble solutions, where multiple model predictions are combined into one, can improve these problems. However, ensembles are computationally intensive and less transparent to practitioners than single model solutions. Constructing model soups, by averaging the weights of multiple models into a single model, could circumvent these limitations while still improving performance.

OBJECTIVE

To investigate the performance of model soups for a dermoscopic melanoma-nevus skin cancer classification task with respect to (1) generalisation to images from other clinics, (2) robustness against small image changes and (3) calibration such that the confidences correspond closely to the actual predictive uncertainties.

METHODS

We construct model soups by fine-tuning pre-trained models on seven different image resolutions and subsequently averaging their weights. Performance is evaluated on a multi-source dataset including holdout and external components.

RESULTS

We find that model soups improve generalisation and calibration on the external component while maintaining performance on the holdout component. For robustness, we observe performance improvements for pertubated test images, while the performance on corrupted test images remains on par.

CONCLUSIONS

Overall, souping for skin cancer classifiers has a positive effect on generalisation, robustness and calibration. It is easy for practitioners to implement and by combining multiple models into a single model, complexity is reduced. This could be an important factor in achieving clinical applicability, as less complexity generally means more transparency.

The antileukemic activity of decitabine upon PML/RARA-negative AML blasts is supported by all-trans retinoic acid: in vitro and in vivo evidence for cooperation

The prognosis of AML patients with adverse genetics, such as a complex, monosomal karyotype and TP53 lesions, is still dismal even with standard chemotherapy. DNA-hypomethylating agent monotherapy induces an encouraging response rate in these patients. When combined with decitabine (DAC), all-trans retinoic acid (ATRA) resulted in an improved response rate and longer overall survival in a randomized phase II trial (DECIDER; NCT00867672). The molecular mechanisms governing this in vivo synergism are unclear. We now demonstrate cooperative antileukemic effects of DAC and ATRA on AML cell lines U937 and MOLM-13. By RNA-sequencing, derepression of >1200 commonly regulated transcripts following the dual treatment was observed. Overall chromatin accessibility (interrogated by ATAC-seq) and, in particular, at motifs of retinoic acid response elements were affected by both single-agent DAC and ATRA, and enhanced by the dual treatment. Cooperativity regarding transcriptional induction and chromatin remodeling was demonstrated by interrogating the HIC1, CYP26A1, GBP4, and LYZ genes, in vivo gene derepression by expression studies on peripheral blood blasts from AML patients receiving DAC + ATRA. The two drugs also cooperated in derepression of transposable elements, more effectively in U937 (mutated TP53) than MOLM-13 (intact TP53), resulting in a “viral mimicry” response. In conclusion, we demonstrate that in vitro and in vivo, the antileukemic and gene-derepressive epigenetic activity of DAC is enhanced by ATRA.

Gene expression-based prediction of pazopanib efficacy in sarcoma

BACKGROUND

The multi-receptor tyrosine kinase inhibitor pazopanib is approved for the treatment of advanced soft-tissue sarcoma and has also shown activity in other sarcoma subtypes. However, its clinical efficacy is highly variable, and no reliable predictors exist to select patients who are likely to benefit from this drug.

PATIENTS AND METHODS

We analysed the molecular profiles and clinical outcomes of patients with pazopanib-treated sarcoma enrolled in a prospective observational study by the German Cancer Consortium, DKTK MASTER, that employs whole-genome/exome sequencing and transcriptome sequencing to inform the care of young adults with advanced cancer across histology and patients with rare cancers.

RESULTS

Among 109 patients with available whole-genome/exome sequencing data, there was no correlation between clinical parameters, specific genetic alterations or mutational signatures and clinical outcome. In contrast, the analysis of a subcohort of 62 patients who underwent molecular analysis before pazopanib treatment and had transcriptome sequencing data available showed that mRNA levels of NTRK3 (hazard ratio [HR] = 0.53, p = 0.021), IGF1R (HR = 1.82, p = 0.027) and KDR (HR = 0.50, p = 0.011) were independently associated with progression-free survival (PFS). Based on the expression of these multi-receptor tyrosine kinase genes, i.e. the features NTRK3-high, IGF1R-low and KDR-high, we developed a pazopanib efficacy predictor that stratified patients into three groups with significantly different PFS (p < 0.0001). Application of the pazopanib efficacy predictor to an independent cohort of patients with pazopanib-treated sarcoma from DKTK MASTER (n = 43) confirmed its potential to separate patient groups with significantly different PFS (p = 0.02), whereas no such association was observed in patients with sarcoma from DKTK MASTER (n = 97) or The Cancer Genome Atlas sarcoma cohort (n = 256) who were not treated with pazopanib.

CONCLUSION

A score based on the combined expression of NTRK3, IGF1R and KDR allows the identification of patients with sarcoma and with good, intermediate and poor outcome following pazopanib therapy and warrants prospective investigation as a predictive tool to optimise the use of this drug in the clinic.

MGMT inactivation as a new biomarker in patients with advanced biliary tract cancers

Biliary tract cancers (BTCs) have poor prognosis and limited therapeutic options. The impact of O⁶ -methylguanine-DNA methyltransferase (MGMT) inactivation in advanced BTC patients is not established. We investigated the prevalence, prognostic and predictive impact of MGMT inactivation in two multicenter cohorts. MGMT inactivation was assessed through PCR and immunohistochemistry (IHC) in an Italian cohort; the results were then externally validated using RNA sequencing (RNA-seq) data from the BTC subcohort of the MASTER (Molecularly Aided Stratification for Tumor Eradication Research) precision oncology program of the National Center for Tumor Diseases Heidelberg and the German Cancer Consortium. Among 164 Italian cases, 18% presented MGMT promoter hypermethylation (>14%) and 73% had negative MGMT protein expression. Both were associated with worse overall survival (OS) (HR 2.31; p<.001 and HR 1.99, p=0.012, respectively). In the MASTER cohort, patients with lower MGMT mRNA expression showed significantly poorer OS (mOS 20.4 vs 31.7 months, unadjusted HR 1.89; p=0.043). Our results suggest that MGMT inactivation is a frequent epigenetic alteration in BTC, with a significant prognostic impact, and provide the rationale to explore DNA-damaging agents in MGMT-inactivated BTCs.

Explainable artificial intelligence in skin cancer recognition: A systematic review

BACKGROUND

Due to their ability to solve complex problems, deep neural networks (DNNs) are becoming increasingly popular in medical applications. However, decision-making by such algorithms is essentially a black-box process that renders it difficult for physicians to judge whether the decisions are reliable. The use of explainable artificial intelligence (XAI) is often suggested as a solution to this problem. We investigate how XAI is used for skin cancer detection: how is it used during the development of new DNNs? What kinds of visualisations are commonly used? Are there systematic evaluations of XAI with dermatologists or dermatopathologists?

METHODS

Google Scholar, PubMed, IEEE Explore, Science Direct and Scopus were searched for peer-reviewed studies published between January 2017 and October 2021 applying XAI to dermatological images: the search terms histopathological image, whole-slide image, clinical image, dermoscopic image, skin, dermatology, explainable, interpretable and XAI were used in various combinations. Only studies concerned with skin cancer were included.

RESULTS

37 publications fulfilled our inclusion criteria. Most studies (19/37) simply applied existing XAI methods to their classifier to interpret its decision-making. Some studies (4/37) proposed new XAI methods or improved upon existing techniques. 14/37 studies addressed specific questions such as bias detection and impact of XAI on man-machine-interactions. However, only three of them evaluated the performance and confidence of humans using CAD systems with XAI.

CONCLUSION

XAI is commonly applied during the development of DNNs for skin cancer detection. However, a systematic and rigorous evaluation of its usefulness in this scenario is lacking.

DNMT1 Deficiency Impacts on Plasmacytoid Dendritic Cells in Homeostasis and Autoimmune Disease

Dendritic cells (DCs) are heterogeneous immune regulators involved in autoimmune diseases. Epigenomic mechanisms orchestrating DC development and DC subset diversification remain insufficiently understood but could be important to modulate DC fate for clinical purposes. By combining whole-genome methylation assessment with the analysis of mice expressing reduced DNA methyltransferase 1 levels, we show that distinct DNA methylation levels and patterns are required for the development of plasmacytoid DC and conventional DC subsets. We provide clonal in vivo evidence for DC lineage establishment at the stem cell level, and we show that a high DNA methylation threshold level is essential for Flt3-dependent survival of DC precursors. Importantly, reducing methylation predominantly depletes plasmacytoid DC and alleviates systemic lupus erythematosus in an autoimmunity mouse model. This study shows how DNA methylation regulates the production of DC subsets and provides a potential rationale for targeting autoimmune disease using hypomethylating agents.

The age of the bone marrow microenvironment influences B-cell acute lymphoblastic leukemia progression via CXCR5-CXCL13

B-cell acute lymphoblastic leukemia (B-ALL) occurs most commonly in children, whereas chronic myeloid leukemia is more frequent in adults. The myeloid bias of hematopoiesis in elderly individuals has been considered causative, but the age of the bone marrow microenvironment (BMM) may be contributory. Using various murine models of B-ALL in young vs old mice, we recapitulated B-ALL preponderance in children vs adults. We showed differential effects of young vs old BM macrophages on B-ALL cell function. Molecular profiling using RNA- and ATAC-sequencing revealed pronounced differences in young vs old BMM-derived macrophages and enrichment for gene sets associated with inflammation. In concordance with the role of C-X-C motif chemokine (CXCL) 13 for disease-associated B-cell chemoattraction, we found CXCL13 to be highly expressed in young macrophages on a translational compared with a transcriptional level. Inhibition of CXCL13 in BM macrophages impaired leukemia cell migration and decreased the proliferation of cocultured B-ALL cells, whereas recombinant CXCL13 increased pAKT and B-ALL cell expansion. Pretreatment of B-ALL-initiating cells with CXCL13 accelerated B-ALL progression. Deficiency of Cxcr5, the receptor for CXCL13, on B-ALL-initiating cells prolonged murine survival, whereas high expression of CXCR5 in pediatric B-ALL may predict central nervous system relapse. CXCL13 staining was increased in bone sections from pediatric compared with adult patients with B-ALL. Taken together, our study shows that the age of the BMM and, in particular, BM macrophages influence the leukemia phenotype. The CXCR5-CXCL13 axis may act as prognostic marker and an attractive novel target for the treatment of B-ALL.

Control of APOBEC3B induction and cccDNA decay by NF-κB and miR-138-5p

Background & Aims

Immune-mediated induction of cytidine deaminase APOBEC3B (A3B) expression leads to HBV covalently closed circular DNA (cccDNA) decay. Here, we aimed to decipher the signalling pathway(s) and regulatory mechanism(s) involved in A3B induction and related HBV control.

Methods

Differentiated HepaRG cells (dHepaRG) knocked-down for NF-κB signalling components, transfected with siRNA or micro RNAs (miRNA), and primary human hepatocytes ± HBV or HBVΔX or HBV-RFP, were treated with lymphotoxin beta receptor (LTβR)-agonist (BS1). The biological outcomes were analysed by reverse transcriptase-qPCR, immunoblotting, luciferase activity, chromatin immune precipitation, electrophoretic mobility-shift assay, targeted-bisulfite-, miRNA-, RNA-, genome-sequencing, and mass-spectrometry.

Results

We found that canonical and non-canonical NF-κB signalling pathways are mandatory for A3B induction and anti-HBV effects. The degree of immune-mediated A3B production is independent of A3B promoter demethylation but is controlled post-transcriptionally by the miRNA 138-5p expression (hsa-miR-138-5p), promoting A3B mRNA decay. Hsa-miR-138-5p over-expression reduced A3B levels and its antiviral effects. Of note, established infection inhibited BS1-induced A3B expression through epigenetic modulation of A3B promoter. Twelve days of treatment with a LTβR-specific agonist BS1 is sufficient to reduce the cccDNA pool by 80% without inducing significant damages to a subset of cancer-related host genes. Interestingly, the A3B-mediated effect on HBV is independent of the transcriptional activity of cccDNA as well as on rcDNA synthesis.

Conclusions

Altogether, A3B represents the only described enzyme to target both transcriptionally active and inactive cccDNA. Thus, inhibiting hsa-miR-138-5p expression should be considered in the combinatorial design of new therapies against HBV, especially in the context of immune-mediated A3B induction.

Lay summary

Immune-mediated induction of cytidine deaminase APOBEC3B is transcriptionally regulated by NF-κB signalling and post-transcriptionally downregulated by hsa-miR-138-5p expression, leading to cccDNA decay. Timely controlled APOBEC3B-mediated cccDNA decay occurs independently of cccDNA transcriptional activity and without damage to a subset of cancer-related genes. Thus, APOBEC3B-mediated cccDNA decay could offer an efficient therapeutic alternative to target hepatitis B virus chronic infection.

•

Impairment of NF-κB signalling prevents APOBEC3B induction and cccDNA decay.

•

APOBEC3B is post-transcriptionally regulated by the hsa-miR-138-5p.

•

Over-expression of the hsa-miR-138-5p inhibits APOBEC3B expression and cccDNA decay.

•

A3B timely induces cccDNA decay without damage to cancer-related genes.

•

APOBEC3B-mediated cccDNA decay is independent of cccDNA transcriptional activity.

Integrating Patient Data Into Skin Cancer Classification Using Convolutional Neural Networks: Systematic Review

Background

Recent years have been witnessing a substantial improvement in the accuracy of skin cancer classification using convolutional neural networks (CNNs). CNNs perform on par with or better than dermatologists with respect to the classification tasks of single images. However, in clinical practice, dermatologists also use other patient data beyond the visual aspects present in a digitized image, further increasing their diagnostic accuracy. Several pilot studies have recently investigated the effects of integrating different subtypes of patient data into CNN-based skin cancer classifiers.

Objective

This systematic review focuses on the current research investigating the impact of merging information from image features and patient data on the performance of CNN-based skin cancer image classification. This study aims to explore the potential in this field of research by evaluating the types of patient data used, the ways in which the nonimage data are encoded and merged with the image features, and the impact of the integration on the classifier performance.

Methods

Google Scholar, PubMed, MEDLINE, and ScienceDirect were screened for peer-reviewed studies published in English that dealt with the integration of patient data within a CNN-based skin cancer classification. The search terms skin cancer classification, convolutional neural network(s), deep learning, lesions, melanoma, metadata, clinical information, and patient data were combined.

Results

A total of 11 publications fulfilled the inclusion criteria. All of them reported an overall improvement in different skin lesion classification tasks with patient data integration. The most commonly used patient data were age, sex, and lesion location. The patient data were mostly one-hot encoded. There were differences in the complexity that the encoded patient data were processed with regarding deep learning methods before and after fusing them with the image features for a combined classifier.

Conclusions

This study indicates the potential benefits of integrating patient data into CNN-based diagnostic algorithms. However, how exactly the individual patient data enhance classification performance, especially in the case of multiclass classification problems, is still unclear. Moreover, a substantial fraction of patient data used by dermatologists remains to be analyzed in the context of CNN-based skin cancer classification. Further exploratory analyses in this promising field may optimize patient data integration into CNN-based skin cancer diagnostics for patients’ benefits.

Comprehensive Genomic and Transcriptomic Analysis for Guiding Therapeutic Decisions in Patients with Rare Cancers

The clinical relevance of comprehensive molecular analysis in rare cancers is not established. We analyzed the molecular profiles and clinical outcomes of 1,310 patients (rare cancers, 75.5%) enrolled in a prospective observational study by the German Cancer Consortium that applies whole-genome/exome and RNA sequencing to inform the care of adults with incurable cancers. Based on 472 single and six composite biomarkers, a cross-institutional molecular tumor board provided evidence-based management recommendations, including diagnostic reevaluation, genetic counseling, and experimental treatment, in 88% of cases. Recommended therapies were administered in 362 of 1,138 patients (31.8%) and resulted in significantly improved overall response and disease control rates (23.9% and 55.3%) compared to previous therapies, translating into a progression-free survival ratio >1.3 in 35.7% of patients. These data demonstrate the benefit of molecular stratification in rare cancers and represent a resource that may promote clinical trial access and drug approvals in this underserved patient population.

Hotspot DNMT3A mutations in clonal hematopoiesis and acute myeloid leukemia sensitize cells to azacytidine via viral mimicry response

Somatic mutations in DNA methyltransferase 3A (DNMT3A) are among the most frequent alterations in clonal hematopoiesis (CH) and acute myeloid leukemia (AML), with a hotspot in exon 23 at arginine 882 (DNMT3A^R882). Here, we demonstrate that DNMT3A^R882H-dependent CH and AML cells are specifically susceptible to the hypomethylating agent azacytidine (AZA). Addition of AZA to chemotherapy prolonged AML survival solely in individuals with DNMT3A^R882 mutations, suggesting its potential as a predictive marker for AZA response. AML and CH mouse models confirmed AZA susceptibility specifically in DNMT3A^R882H-expressing cells. Hematopoietic stem cells (HSCs) and progenitor cells expressing DNMT3A^R882H exhibited cell autonomous viral mimicry response as a result of focal DNA hypomethylation at retrotransposon sequences. Administration of AZA boosted hypomethylation of retrotransposons specifically in DNMT3A^R882H-expressing cells and maintained elevated levels of canonical interferon-stimulated genes (ISGs), thus leading to suppressed protein translation and increased apoptosis.

Combining CNN-based histologic whole slide image analysis and patient data to improve skin cancer classification

BACKGROUND

Clinicians and pathologists traditionally use patient data in addition to clinical examination to support their diagnoses.

OBJECTIVES

We investigated whether a combination of histologic whole slides image (WSI) analysis based on convolutional neural networks (CNNs) and commonly available patient data (age, sex and anatomical site of the lesion) in a binary melanoma/nevus classification task could increase the performance compared with CNNs alone.

METHODS

We used 431 WSIs from two different laboratories and analysed the performance of classifiers that used the image or patient data individually or three common fusion techniques. Furthermore, we tested a naive combination of patient data and an image classifier: for cases interpreted as 'uncertain' (CNN output score <0.7), the decision of the CNN was replaced by the decision of the patient data classifier.

RESULTS

The CNN on its own achieved the best performance (mean ± standard deviation of five individual runs) with AUROC of 92.30% ± 0.23% and balanced accuracy of 83.17% ± 0.38%. While the classification performance was not significantly improved in general by any of the tested fusions, naive strategy of replacing the image classifier with the patient data classifier on slides with low output scores improved balanced accuracy to 86.72% ± 0.36%.

CONCLUSION

In most cases, the CNN on its own was so accurate that patient data integration did not provide any benefit. However, incorporating patient data for lesions that were classified by the CNN with low 'confidence' improved balanced accuracy.

Methrix: an R/bioconductor package for systematic aggregation and analysis of bisulfite sequencing data

MOTIVATION

Whole genome bisulfite sequencing (WGBS), measures DNA methylation at base pair resolution resulting in large bedGraph like coverage files. Current options for processing such files are hindered by discrepancies in file format specification, speed and memory requirements.

RESULTS

We developed methrix, an R package, which provides a toolset for systematic analysis of large datasets. Core functionality of the package includes a comprehensive bedGraph or similar tab-separated text file reader - which summarizes methylation calls based on annotated reference indices, infers and collapses strands, and handles uncovered reference CpG sites while facilitating a flexible input file format specification. Additional optimized functions for quality control filtering, sub-setting, and visualization allow user-friendly and effective processing of WGBS results. Easy integration with tools for differentially methylated region (DMR) calling and annotation further eases the analysis of genome-wide methylation data. Overall, methrix enriches established WGBS workflows by bringing together computational efficiency and versatile functionality.

AVAILABILITY AND IMPLEMENTATION

Methrix is implemented as an R package, made available under MIT license at https://github.com/CompEpigen/methrix and can be installed from the Bioconductor repository.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

International Consensus Definition of DNA Methylation Subgroups in Juvenile Myelomonocytic Leukemia

PURPOSE

Known clinical and genetic markers have limitations in predicting disease course and outcome in juvenile myelomonocytic leukemia (JMML). DNA methylation patterns in JMML have correlated with outcome across multiple studies, suggesting it as a biomarker to improve patient stratification. However, standardized approaches to classify JMML on the basis of DNA methylation patterns are lacking. We, therefore, sought to define an international consensus for DNA methylation subgroups in JMML and develop classification methods for clinical implementation.

EXPERIMENTAL DESIGN

Published DNA methylation data from 255 patients with JMML were used to develop and internally validate a classifier model. Accuracy across platforms (EPIC-arrays and MethylSeq) was tested using a technical validation cohort (32 patients). The suitability of both methods for single-patient classification was demonstrated using an independent cohort (47 patients).

RESULTS

Analysis of pooled, published data established three DNA methylation subgroups as a de facto standard. Unfavorable prognostic parameters (PTPN11 mutation, elevated fetal hemoglobin, and older age) were significantly enriched in the high methylation (HM) subgroup. A classifier was then developed that predicted subgroups with 98% accuracy across different technological platforms. Applying the classifier to an independent validation cohort confirmed an association of HM with secondary mutations, high relapse incidence, and inferior overall survival (OS), while the low methylation subgroup was associated with a favorable disease course. Multivariable analysis established DNA methylation subgroups as the only significant factor predicting OS.

CONCLUSIONS

This study provides an international consensus definition for DNA methylation subgroups in JMML. We developed and validated methods which will facilitate the design of risk-stratified clinical trials in JMML.

AmpliconDesign - an interactive web server for the design of high-throughput targeted DNA methylation assays

Targeted analysis of DNA methylation patterns based on bisulfite-treated genomic DNA (BT-DNA) is considered as a gold-standard for epigenetic biomarker development. Existing software tools facilitate primer design, primer quality control or visualization of primer localization. However, high-throughput design of primers for BT-DNA amplification is hampered by limits in throughput and functionality of existing tools, requiring users to repeatedly perform specific tasks manually. Consequently, the design of PCR primers for BT-DNA remains a tedious and time-consuming process. To bridge this gap, we developed AmpliconDesign, a webserver providing a scalable and user-friendly platform for the design and analysis of targeted DNA methylation studies based on BT-DNA, e.g. deep amplicon bisulfite sequencing (ampBS-seq) or EpiTYPER MassArray. Core functionality of the web server includes high-throughput primer design and binding site validation based on in silico bisulfite-converted DNA sequences, prediction of fragmentation patterns for EpiTYPER MassArray, an interactive quality control as well as a streamlined analysis workflow for ampBS-seq.

Molecular tumor classification using DNA methylome analysis

Tumor classifiers based on molecular patterns promise to define and reliably classify tumor entities. The high tissue- and cell type-specificity of DNA methylation, as well as its high stability, makes DNA methylation an ideal choice for the development of tumor classifiers. Herein, we review existing tumor classifiers using DNA methylome analysis and will provide an overview on their emerging impact on cancer classification, the detection of novel cancer subentities and patient stratification with a focus on brain tumors, sarcomas and hematopoietic malignancies. Furthermore, we provide an outlook on the enormous potential of DNA methylome analysis to complement classical histopathological and genetic diagnostics, including the emerging field of epigenomic analysis in liquid biopsies.

Epigenomic technologies for precision oncology

Epigenetic patterns in a cell control the expression of genes and consequently determine the phenotype of a cell. Cancer cells possess altered epigenomes which include aberrant patterns of DNA methylation, histone tail modifications, nucleosome positioning and of the three-dimensional chromatin organization within a nucleus. These altered epigenetic patterns are potential useful biomarkers to detect cancer cells and to classify tumor types. In addition, the cancer epigenome dictates the response of a cancer cell to therapeutic intervention and, therefore its knowledge, will allow to predict response to different therapeutic approaches. Here we review the current state-of-the-art technologies that have been developed to decipher epigenetic patterns on the genomic level and discuss how these methods are potentially useful for precision oncology.

Germline SDHB-inactivating mutation in gastric spindle cell sarcoma

Gastrointestinal stromal tumors (GISTs) are the most frequent mesenchymal tumors of the gastrointestinal tract. Inactivating mutations or epigenetic deregulation of succinate dehydrogenase complex (SDH) genes are considered defining features of a subset of GIST occurring in the stomach. Based on comprehensive molecular profiling and biochemical analysis within a precision oncology program, we identified hallmarks of SDH deficiency (germline SDHB-inactivating mutation accompanied by somatic loss of heterozygosity, lack of SDHB expression, global DNA hypermethylation, and elevated succinate/fumarate ratio) in a 40-year-old woman with undifferentiated gastric spindle cell sarcoma that did not meet the diagnostic criteria for other mesenchymal tumors of the stomach, including GIST. These data reveal that the loss of SDH function can be involved in the pathogenesis of non-GIST sarcoma of the gastrointestinal tract.

Effects of Label Noise on Deep Learning-Based Skin Cancer Classification

Recent studies have shown that deep learning is capable of classifying dermatoscopic images at least as well as dermatologists. However, many studies in skin cancer classification utilize non-biopsy-verified training images. This imperfect ground truth introduces a systematic error, but the effects on classifier performance are currently unknown. Here, we systematically examine the effects of label noise by training and evaluating convolutional neural networks (CNN) with 804 images of melanoma and nevi labeled either by dermatologists or by biopsy. The CNNs are evaluated on a test set of 384 images by means of 4-fold cross validation comparing the outputs with either the corresponding dermatological or the biopsy-verified diagnosis. With identical ground truths of training and test labels, high accuracies with 75.03% (95% CI: 74.39–75.66%) for dermatological and 73.80% (95% CI: 73.10–74.51%) for biopsy-verified labels can be achieved. However, if the CNN is trained and tested with different ground truths, accuracy drops significantly to 64.53% (95% CI: 63.12–65.94%, p < 0.01) on a non-biopsy-verified and to 64.24% (95% CI: 62.66–65.83%, p < 0.01) on a biopsy-verified test set. In conclusion, deep learning methods for skin cancer classification are highly sensitive to label noise and future work should use biopsy-verified training images to mitigate this problem.

Methylome-based cell-of-origin modeling (Methyl-COOM) identifies aberrant expression of immune regulatory molecules in CLL

Background

In cancer, normal epigenetic patterns are disturbed and contribute to gene expression changes, disease onset, and progression. The cancer epigenome is composed of the epigenetic patterns present in the tumor-initiating cell at the time of transformation, and the tumor-specific epigenetic alterations that are acquired during tumor initiation and progression. The precise dissection of these two components of the tumor epigenome will facilitate a better understanding of the biological mechanisms underlying malignant transformation. Chronic lymphocytic leukemia (CLL) originates from differentiating B cells, which undergo extensive epigenetic programming. This poses the challenge to precisely determine the epigenomic ground state of the cell-of-origin in order to identify CLL-specific epigenetic aberrations.

Methods

We developed a linear regression model, methylome-based cell-of-origin modeling (Methyl-COOM), to map the cell-of-origin for individual CLL patients based on the continuum of epigenomic changes during normal B cell differentiation.

Results

Methyl-COOM accurately maps the cell-of-origin of CLL and identifies CLL-specific aberrant DNA methylation events that are not confounded by physiologic epigenetic B cell programming. Furthermore, Methyl-COOM unmasks abnormal action of transcription factors, altered super-enhancer activities, and aberrant transcript expression in CLL. Among the aberrantly regulated transcripts were many genes that have previously been implicated in T cell biology. Flow cytometry analysis of these markers confirmed their aberrant expression on malignant B cells at the protein level.

Conclusions

Methyl-COOM analysis of CLL identified disease-specific aberrant gene regulation. The aberrantly expressed genes identified in this study might play a role in immune-evasion in CLL and might serve as novel targets for immunotherapy approaches. In summary, we propose a novel framework for in silico modeling of reference DNA methylomes and for the identification of cancer-specific epigenetic changes, a concept that can be broadly applied to other human malignancies.

Electronic supplementary material

Supplementary information accompanies this paper at 10.1186/s13073-020-00724-7.

Histone methyltransferase SETDB1 contributes to melanoma tumorigenesis and serves as a new potential therapeutic target

Alterations in histone modifications play a crucial role in the progression of various types of cancer. The histone methyltransferase SETDB1 catalyzes the addition of methyl groups to histone H3 at lysine 9. Here, we describe how overexpression of SETDB1 contributes to melanoma tumorigenesis. SETDB1 is highly amplified in melanoma cells and in the patient tumors. Increased expression of SETDB1, which correlates with SETDB1 amplification, is associated with a more aggressive phenotype in in vitro and in vivo studies. Mechanistically, SETDB1 implements its effects via regulation of thrombospondin 1, and the SET-domain of SETDB1 is essential for the maintenance of its tumorigenic activity. Inhibition of SETDB1 reduces cell growth in melanomas resistant to targeted treatments. Our results indicate that SETDB1 is a major driver of melanoma development and may serve as a potential future target for the treatment of this disease.

Upfront azacitidine (AZA) in juvenile myelomonocytic leukemia (JMML): Interim analysis of the prospective AZA-JMML-001 study

10031

Background: Hematopoietic stem cell transplantation (HSCT) is the only curative therapy for JMML patients (pts). Novel therapies controlling the disorder prior to HSCT are urgently needed. A phase 2, multicenter, open-label study was conducted to evaluate safety and anti-leukemia activity of AZA monotherapy prior to HSCT in pts with newly diagnosed (ND) JMML. Methods: AZA (75 mg/m2 IV) was administered once daily on days 1–7 of each 28-day cycle (C). Primary endpoint was number of pts with clinical complete remission or clinical partial remission (cPR) at C3 day (D) 28 (C3D28). Results: 18 JMML pts (13 PTPN11-, 3 NRAS-, 1 KRAS-, 1 NF1-mutated) were enrolled from 09/2015 to 11/2017. Median (range) white blood cell and platelet (Plt) counts were 19.7 (4.3–59.0) × 109/L and 28 (7–85) × 109/L, respectively. DNA methylation class (MC) was high, intermediate (int), or low in 11, 5, and 2 pts, respectively. 16 pts completed C3 and 5 pts C6. 2 pts discontinued treatment (Tx) pre-C3D28 due to disease progression (PD). 6 pts (33%) had ≥ 1 grade (Gr) 3–4 manageable adverse event (AE) related to AZA. Most common Gr 3–4 AEs related to AZA were neutropenia (2) and anemia (2). 11 pts (61%) were in cPR at C3D28; 7 had PD at C3D28 or prior. All 7 pts of the int/low MC and 4/11 in high MC achieved cPR. 17 pts received HSCT at median of 58 days (37–518) from last AZA dose; 14 were leukemia-free at a median follow-up of 15.7 months (0.1–31.7) after HSCT. 2 pts (high MC) given HSCT relapsed after allograft. 16/18 pts were alive at a median follow-up of 19.8 months (2.6–37.3) from diagnosis. 1 pt discontinuing Tx prior to C3 died from PD; 1 non-responder died from transplant-related causes. Plt response in pts with cPR prompted retrospective comparison of Plt counts at time of HSCT with a historical registry control cohort. Pts with NF1-mutated JMML with higher Plt counts versus other genetic subtypes were excluded. While 7/16 (44%) study pts had Plt counts ≥ 100 × 109/L at HSCT, only 10/58 (17%) historical cohort pts reached this cutoff ( P < 0.01). Conclusions: This study shows that AZA monotherapy was well tolerated in pts with ND JMML. Although the long-term advantage of AZA Tx remains to be fully assessed, responses show it was effective in JMML and provided clinical benefit to pts in this study. Clinical trial information: NCT02447666.

Integrative Analysis Defines Distinct Prognostic Subgroups of Intrahepatic Cholangiocarcinoma

Intrahepatic cholangiocarcinoma (iCCA) is the second most common primary liver cancer. It is defined by cholangiocytic differentiation and has poor prognosis. Recently, epigenetic processes have been shown to play an important role in cholangiocarcinogenesis. We performed an integrative analysis on 52 iCCAs using both genetic and epigenetic data with a specific focus on DNA methylation components. We found recurrent isocitrate dehydrogenase 1 (IDH1) and IDH2 (28%) gene mutations, recurrent arm-length copy number alterations (CNAs), and focal alterations such as deletion of 3p21 or amplification of 12q15, which affect BRCA1 Associated Protein 1, polybromo 1, and mouse double minute 2 homolog. DNA methylome analysis revealed excessive hypermethylation of iCCA, affecting primarily the bivalent genomic regions marked with both active and repressive histone modifications. Integrative clustering of genetic and epigenetic data identified four iCCA subgroups with prognostic relevance further designated as IDH, high (H), medium (M), and low (L) alteration groups. The IDH group consisted of all samples with IDH1 or IDH2 mutations and showed, together with the H group, a highly disrupted genome, characterized by frequent deletions of chromosome arms 3p and 6q. Both groups showed excessive hypermethylation with distinct patterns. The M group showed intermediate characteristics regarding both genetic and epigenetic marks, whereas the L group exhibited few methylation changes and mutations and a lack of CNAs. Methylation-based latent component analysis of cell-type composition identified differences among these four groups. Prognosis of the H and M groups was significantly worse than that of the L group. Conclusion: Using an integrative genomic and epigenomic analysis approach, we identified four major iCCA subgroups with widespread genomic and epigenomic differences and prognostic implications. Furthermore, our data suggest differences in the cell-of-origin of the iCCA subtypes.

Differential roles of STAT1 and STAT2 in the sensitivity of JAK2V617F- vs. BCR-ABL-positive cells to interferon alpha

BACKGROUND

Interferon alpha (IFNa) monotherapy is recommended as the standard therapy in polycythemia vera (PV) but not in chronic myeloid leukemia (CML). Here, we investigated the mechanisms of IFNa efficacy in JAK2V617F- vs. BCR-ABL-positive cells.

METHODS

Gene expression microarrays and RT-qPCR of PV vs. CML patient PBMCs and CD34+ cells and of the murine cell line 32D expressing JAK2V617F or BCR-ABL were used to analyze and compare interferon-stimulated gene (ISG) expression. Furthermore, using CRISPR/Cas9n technology, targeted disruption of STAT1 or STAT2, respectively, was performed in 32D-BCR-ABL and 32D-JAK2V617F cells to evaluate the role of these transcription factors for IFNa efficacy. The knockout cell lines were reconstituted with STAT1, STAT2, STAT1Y701F, or STAT2Y689F to analyze the importance of wild-type and phosphomutant STATs for the IFNa response. ChIP-seq and ChIP were performed to correlate histone marks with ISG expression.

RESULTS

Microarray analysis and RT-qPCR revealed significant upregulation of ISGs in 32D-JAK2V617F but downregulation in 32D-BCR-ABL cells, and these effects were reversed by tyrosine kinase inhibitor (TKI) treatment. Similar expression patterns were confirmed in human cell lines, primary PV and CML patient PBMCs and CD34+ cells, demonstrating that these effects are operational in patients. IFNa treatment increased Stat1, Stat2, and Irf9 mRNA as well as pY-STAT1 in all cell lines; however, viability was specifically decreased in 32D-JAK2V617F. STAT1 or STAT2 knockout and reconstitution with wild-type or phospho-deficient STAT mutants demonstrated the necessity of STAT2 for IFNa-induced STAT1 phosphorylation in BCR-ABL- but not in JAK2V617F-expressing cells. STAT1 was essential for IFNa activity in both BCR-ABL- and JAK2V617F-positive cells. Furthermore, ChIP experiments demonstrate higher repressive and lower active chromatin marks at the promoters of ISGs in BCR-ABL-expressing cells.

CONCLUSIONS

JAK2V617F but not BCR-ABL sensitizes MPN cells to interferon, and this effect was dependent on STAT1. Moreover, STAT2 is a survival factor in BCR-ABL- and JAK2V617F-positive cells but an IFNa-sensitizing factor solely in 32D-JAK2V617F cells by upregulation of STAT1 expression.

From Basic Knowledge to Effective Therapies

In this issue of Cancer Cell, Ott et al. use integrative analysis of histone ChIP-seq and ATAC-seq to describe enhancer-based regulatory circuits in chronic lymphocytic leukemia. This work identified and validated transcription factor PAX5 as main driver of an oncogenic circuitry, which can be disrupted by BET bromodomain inhibition.

LaminA/C regulates epigenetic and chromatin architecture changes upon aging of hematopoietic stem cells

BACKGROUND

The decline of hematopoietic stem cell (HSC) function upon aging contributes to aging-associated immune remodeling and leukemia pathogenesis. Aged HSCs show changes to their epigenome, such as alterations in DNA methylation and histone methylation and acetylation landscapes. We previously showed a correlation between high Cdc42 activity in aged HSCs and the loss of intranuclear epigenetic polarity, or epipolarity, as indicated by the specific distribution of H4K16ac.

RESULTS

Here, we show that not all histone modifications display a polar localization and that a reduction in H4K16ac amount and loss of epipolarity are specific to aged HSCs. Increasing the levels of H4K16ac is not sufficient to restore polarity in aged HSCs and the restoration of HSC function. The changes in H4K16ac upon aging and rejuvenation of HSCs are correlated with a change in chromosome 11 architecture and alterations in nuclear volume and shape. Surprisingly, by taking advantage of knockout mouse models, we demonstrate that increased Cdc42 activity levels correlate with the repression of the nuclear envelope protein LaminA/C, which controls chromosome 11 distribution, H4K16ac polarity, and nuclear volume and shape in aged HSCs.

CONCLUSIONS

Collectively, our data show that chromatin architecture changes in aged stem cells are reversible by decreasing the levels of Cdc42 activity, revealing an unanticipated way to pharmacologically target LaminA/C expression and revert alterations of the epigenetic architecture in aged HSCs.

RAS-pathway mutation patterns define epigenetic subclasses in juvenile myelomonocytic leukemia

Juvenile myelomonocytic leukemia (JMML) is an aggressive myeloproliferative disorder of early childhood characterized by mutations activating RAS signaling. Established clinical and genetic markers fail to fully recapitulate the clinical and biological heterogeneity of this disease. Here we report DNA methylome analysis and mutation profiling of 167 JMML samples. We identify three JMML subgroups with unique molecular and clinical characteristics. The high methylation group (HM) is characterized by somatic PTPN11 mutations and poor clinical outcome. The low methylation group is enriched for somatic NRAS and CBL mutations, as well as for Noonan patients, and has a good prognosis. The intermediate methylation group (IM) shows enrichment for monosomy 7 and somatic KRAS mutations. Hypermethylation is associated with repressed chromatin, genes regulated by RAS signaling, frequent co-occurrence of RAS pathway mutations and upregulation of DNMT1 and DNMT3B, suggesting a link between activation of the DNA methylation machinery and mutational patterns in JMML.

Addition of rituximab to standard therapy improves response rate and progression-free survival in relapsed or refractory thrombotic thrombocytopenic purpura and autoimmune haemolytic anaemia

Treatment of relapsed or refractory autoimmune mediated haemolytic syndromes, such as autoimmune haemolytic anaemia (AIHA) and thrombotic thrombocytopenic purpura (TTP), represents a therapeutic challenge. Here we report on our experience with the monoclonal anti-CD20 antibody rituximab (R) compared to standard treatment in these diseases. Patients with non-familialTTP orAIHA and no underlying malignancy were included in our analysis. Safety and efficacy of R-treatment were compared to results obtained in standard treatment approaches. Altogether, 27 patients were analyzed, comprising 15 patients withTTP and 12 patients with AIHA. The patients’ average age at the time of diagnosis was 54 years. Eleven patients received antibody treatment (8 TTP, 3 AIHA). No acute or late WHO grade III/IV toxicity associated with rituximab was noted. With standard therapy, the overall response rate (ORR) was 66.7% for AIHA and 65.8% for TTP, respectively. For the R-containing regimens the ORR was 100%. In patients with TTP, median progression free survival (PFS) with R-treatment was 3.8 years, as compared to 0.1 years in the standard-treatment group. In patients with AIHA median PFS was not reached for R-containing treatment; all patients are in sustained remissions with a median follow up of 12.5 months. In the absence of prospective trials, our data underline the safety and efficacy of rituximab in relapsed and refractory autoimmune anaemias with favourable response rates and promising long-term progressionfree survival. Therefore, prospective clinical trials evaluating rituximab as salvage- and first-line-therapy are clearly warranted.

Erratum: DNMT and HDAC inhibitors induce cryptic transcription start sites encoded in long terminal repeats

This corrects the article DOI: 10.1038/ng.3889.

Identification of Cell Type-Specific Differences in Erythropoietin Receptor Signaling in Primary Erythroid and Lung Cancer Cells

Lung cancer, with its most prevalent form non-small-cell lung carcinoma (NSCLC), is one of the leading causes of cancer-related deaths worldwide, and is commonly treated with chemotherapeutic drugs such as cisplatin. Lung cancer patients frequently suffer from chemotherapy-induced anemia, which can be treated with erythropoietin (EPO). However, studies have indicated that EPO not only promotes erythropoiesis in hematopoietic cells, but may also enhance survival of NSCLC cells. Here, we verified that the NSCLC cell line H838 expresses functional erythropoietin receptors (EPOR) and that treatment with EPO reduces cisplatin-induced apoptosis. To pinpoint differences in EPO-induced survival signaling in erythroid progenitor cells (CFU-E, colony forming unit-erythroid) and H838 cells, we combined mathematical modeling with a method for feature selection, the L₁ regularization. Utilizing an example model and simulated data, we demonstrated that this approach enables the accurate identification and quantification of cell type-specific parameters. We applied our strategy to quantitative time-resolved data of EPO-induced JAK/STAT signaling generated by quantitative immunoblotting, mass spectrometry and quantitative real-time PCR (qRT-PCR) in CFU-E and H838 cells as well as H838 cells overexpressing human EPOR (H838-HA-hEPOR). The established parsimonious mathematical model was able to simultaneously describe the data sets of CFU-E, H838 and H838-HA-hEPOR cells. Seven cell type-specific parameters were identified that included for example parameters for nuclear translocation of STAT5 and target gene induction. Cell type-specific differences in target gene induction were experimentally validated by qRT-PCR experiments. The systematic identification of pathway differences and sensitivities of EPOR signaling in CFU-E and H838 cells revealed potential targets for intervention to selectively inhibit EPO-induced signaling in the tumor cells but leave the responses in erythroid progenitor cells unaffected. Thus, the proposed modeling strategy can be employed as a general procedure to identify cell type-specific parameters and to recommend treatment strategies for the selective targeting of specific cell types.

A major challenge in the development of therapeutic interventions is the selective inhibition of a signal transduction pathway in one cell type such as a cancer cell leaving the other cell type such as a healthy cell as unaffected as possible. Here, we propose a new approach that combines mathematical modeling based on quantitative experimental data with statistical methods. We demonstrate based on simulated data that our approach can determine which parameters are the same and which parameters differ in two exemplary cell types. We compare a lung cancer cell line to the precursor cells of red blood cells. We show that the same signal transduction network induced by erythropoietin (EPO), a hormone that is frequently employed to treat anemia in cancer patients, regulates survival of both cell types. Based on our experimental data in combination with our computational approach, we identify seven cell type-specific differences in this signaling pathway. Our strategy allows predicting therapeutic targets that could be inhibited to interfere with survival of lung cancer cells while leaving production of red blood cells unaffected.

CREBBP is a target of epigenetic, but not genetic, modification in juvenile myelomonocytic leukemia

BACKGROUND

Juvenile myelomonocytic leukemia (JMML) is a myeloproliferative neoplasm of childhood whose clinical heterogeneity is only poorly represented by gene sequence alterations. It was previously shown that aberrant DNA methylation of distinct target genes defines a more aggressive variant of JMML, but only few significant targets are known so far. To get a broader picture of disturbed CpG methylation patterns in JMML, we carried out a methylation screen of 34 candidate genes in 45 patients using quantitative mass spectrometry.

FINDINGS

Five of 34 candidate genes analyzed showed recurrent hypermethylation in JMML. cAMP-responsive element-binding protein-binding protein (CREBBP) was the most frequent target of epigenetic modification (77 % of cases). However, no pathogenic mutations of CREBBP were identified in a genetic analysis of 64 patients. CREBBP hypermethylation correlated with clinical parameters known to predict poor outcome.

CONCLUSIONS

This study supports the relevance of epigenetic aberrations in JMML pathophysiology. Our data confirm that DNA hypermethylation in JMML is highly target-specific and associated with higher-risk features. These findings encourage the development of prognostic markers based on epigenetic alterations, which will be helpful in the difficult clinical management of this heterogeneous disease.

Epigenetic silencing of AKAP12 in juvenile myelomonocytic leukemia

A-kinase anchor protein 12 (AKAP12) is a regulator of protein kinase A and protein kinase C signaling, acting downstream of RAS. Epigenetic silencing of AKAP12 has been demonstrated in different cancer entities and this has been linked to the process of tumorigenesis. Here, we used quantitative high-resolution DNA methylation measurement by MassARRAY to investigate epigenetic regulation of all three AKAP12 promoters (i.e., α, β, and γ) within a large cohort of juvenile myelomonocytic leukemia (JMML) patient samples. The AKAP12α promoter shows DNA hypermethylation in JMML samples, which is associated with decreased AKAP12α expression. Promoter methylation of AKAP12α correlates with older age at diagnosis, elevated levels of fetal hemoglobin and poor prognosis. In silico screening for transcription factor binding motifs around the sites of most pronounced methylation changes in the AKAP12α promoter revealed highly significant scores for GATA-2/-1 sequence motifs. Both transcription factors are known to be involved in the haematopoietic differentiation process. Methylation of a reporter construct containing this region resulted in strong suppression of AKAP12 promoter activity, suggesting that DNA methylation might be involved in the aberrant silencing of the AKAP12 promoter in JMML. Exposure to DNMT- and HDAC-inhibitors reactivates AKAP12α expression in vitro, which could potentially be a mechanism underlying clinical treatment responses upon demethylating therapy. Together, these data provide evidence for epigenetic silencing of AKAP12α in JMML and further emphasize the importance of dysregulated RAS signaling in JMML pathogenesis.

Identification of DNA methylation changes at cis-regulatory elements during early steps of HSC differentiation using tagmentation-based whole genome bisulfite sequencing

Epigenetic alterations during cellular differentiation are a key molecular mechanism which both instructs and reinforces the process of lineage commitment. Within the haematopoietic system, progressive changes in the DNA methylome of haematopoietic stem cells (HSCs) are essential for the effective production of mature blood cells. Inhibition or loss of function of the cellular DNA methylation machinery has been shown to lead to a severe perturbation in blood production and is also an important driver of malignant transformation. HSCs constitute a very rare cell population in the bone marrow, capable of life-long self-renewal and multi-lineage differentiation. The low abundance of HSCs has been a major technological barrier to the global analysis of the CpG methylation status within both HSCs and their immediate progeny, the multipotent progenitors (MPPs). Within this Extra View article, we review the current understanding of how the DNA methylome regulates normal and malignant hematopoiesis. We also discuss the current methodologies that are available for interrogating the DNA methylation status of HSCs and MPPs and describe a new data set that was generated using tagmentation-based whole genome bisulfite sequencing (TWGBS) in order to comprehensively map methylated cytosines using the limited amount of genomic DNA that can be harvested from rare cell populations. Extended analysis of this data set clearly demonstrates the added value of genome-wide sequencing of methylated cytosines and identifies novel important cis-acting regulatory regions that are dynamically remodeled during the first steps of haematopoietic differentiation.

Dysregulated serum response factor triggers formation of hepatocellular carcinoma

The ubiquitously expressed transcriptional regulator serum response factor (SRF) is controlled by both Ras/MAPK (mitogen-activated protein kinase) and Rho/actin signaling pathways, which are frequently activated in hepatocellular carcinoma (HCC). We generated SRF-VP16iHep mice, which conditionally express constitutively active SRF-VP16 in hepatocytes, thereby controlling subsets of both Ras/MAPK- and Rho/actin-stimulated target genes. All SRF-VP16iHep mice develop hyperproliferative liver nodules that progresses to lethal HCC. Some murine (m)HCCs acquire Ctnnb1 mutations equivalent to those in human (h)HCC. The resulting transcript signatures mirror those of a distinct subgroup of hHCCs, with shared activation of oncofetal genes including Igf2, correlating with CpG hypomethylation at the imprinted Igf2/H19 locus.

CONCLUSION

SRF-VP16iHep mHCC reveal convergent Ras/MAPK and Rho/actin signaling as a highly oncogenic driver mechanism for hepatocarcinogenesis. This suggests simultaneous inhibition of Ras/MAPK and Rho/actin signaling as a treatment strategy in hHCC therapy.

Epo-induced erythroid maturation is dependent on Plcγ1 signaling

Erythropoiesis is a tightly regulated process. Development of red blood cells occurs through differentiation of hematopoietic stem cells (HSCs) into more committed progenitors and finally into erythrocytes. Binding of erythropoietin (Epo) to its receptor (EpoR) is required for erythropoiesis as it promotes survival and late maturation of erythroid progenitors. In vivo and in vitro studies have highlighted the requirement of EpoR signaling through Janus kinase 2 (Jak2) tyrosine kinase and Stat5a/b as a central pathway. Here, we demonstrate that phospholipase C gamma 1 (Plcγ1) is activated downstream of EpoR-Jak2 independently of Stat5. Plcγ1-deficient pro-erythroblasts and erythroid progenitors exhibited strong impairment in differentiation and colony-forming potential. In vivo, suppression of Plcγ1 in immunophenotypically defined HSCs (Lin(-)Sca1(+)KIT(+)CD48(-)CD150(+)) severely reduced erythroid development. To identify Plcγ1 effector molecules involved in regulation of erythroid differentiation, we assessed changes occurring at the global transcriptional and DNA methylation level after inactivation of Plcγ1. The top common downstream effector was H2afy2, which encodes for the histone variant macroH2A2 (mH2A2). Inactivation of mH2A2 expression recapitulated the effects of Plcγ1 depletion on erythroid maturation. Taken together, our findings identify Plcγ1 and its downstream target mH2A2, as a 'non-canonical' Epo signaling pathway essential for erythroid differentiation.

Identification of regulatory networks in HSCs and their immediate progeny via integrated proteome, transcriptome, and DNA methylome analysis

In this study, we present integrated quantitative proteome, transcriptome, and methylome analyses of hematopoietic stem cells (HSCs) and four multipotent progenitor (MPP) populations. From the characterization of more than 6,000 proteins, 27,000 transcripts, and 15,000 differentially methylated regions (DMRs), we identified coordinated changes associated with early differentiation steps. DMRs show continuous gain or loss of methylation during differentiation, and the overall change in DNA methylation correlates inversely with gene expression at key loci. Our data reveal the differential expression landscape of 493 transcription factors and 682 lncRNAs and highlight specific expression clusters operating in HSCs. We also found an unexpectedly dynamic pattern of transcript isoform regulation, suggesting a critical regulatory role during HSC differentiation, and a cell cycle/DNA repair signature associated with multipotency in MPP2 cells. This study provides a comprehensive genome-wide resource for the functional exploration of molecular, cellular, and epigenetic regulation at the top of the hematopoietic hierarchy.

RASA4 undergoes DNA hypermethylation in resistant juvenile myelomonocytic leukemia

Aberrant DNA methylation at specific genetic loci is a key molecular feature of juvenile myelomonocytic leukemia (JMML) with poor prognosis. Using quantitative high-resolution mass spectrometry, we identified RASA4 isoform 2, which maps to chromosome 7 and encodes a member of the GAP1 family of GTPase-activating proteins for small G proteins, as a recurrent target of isoform-specific DNA hypermethylation in JMML (51% of 125 patients analyzed). RASA4 isoform 2 promoter methylation correlated with clinical parameters predicting poor prognosis (older age, elevated fetal hemoglobin), with higher risk of relapse after hematopoietic stem cell transplantation, and with PTPN11 mutation. The level of isoform 2 methylation increased in relapsed cases after transplantation. Interestingly, most JMML cases with monosomy 7 exhibited hypermethylation on the remaining RASA4 allele. The results corroborate the significance of epigenetic modifications in the phenotype of aggressive JMML.