A parsimonious 3-gene signature predicts clinical outcomes in an acute myeloid leukemia multicohort study

LSP1 promotes the progression of acute myelogenous leukemia by regulating KSR/ERK signaling pathway and cell migration

We aimed to investigate the role and mechanism of LSP1 in the progression of acute myelogenous leukemia. In this study, we established shLSP1 cell line to analyze the function of LSP1 in AML. We observed high expression of LSP1 in AML patients, whereas it showed no expression in normal adults. Furthermore, we found that LSP1 expression was associated with disease prognosis. Our results indicate that LSP1 plays a crucial role in mediating proliferation and survival of leukemia cells through the KSR/ERK signaling pathway. Additionally, LSP1 promotes cell chemotaxis and homing by enhancing cell adhesion and migration. We also discovered that LSP1 confers chemotactic ability to leukemia cells in vivo. Finally, our study identified 12 genes related to LSP1 in AML, which indicated poor survival outcome in AML patients and were enriched in Ras and cell adhesion signaling pathways. Our results revealed that the overexpression of LSP1 is related to the activation of the KSR/ERK signaling pathway, as well as cell adhesion and migration in AML patients. Reducing LSP1 expression impair AML progression, suggesting that LSP1 may serve as a potential drug therapy target for more effective treatment of AML.

Machine learning based analysis of single-cell data reveals evidence of subject-specific single-cell gene expression profiles in acute myeloid leukaemia patients and healthy controls

Acute Myeloid Leukaemia (AML) is characterized by uncontrolled growth of immature myeloid cells, disrupting normal blood production. Treatment typically involves chemotherapy, targeted therapy, and stem cell transplantation but many patients develop chemoresistance, leading to poor outcomes due to the disease's high heterogeneity. In this study, we used publicly available single-cell RNA sequencing data and machine learning to classify AML patients and healthy, monocytes, dendritic and progenitor cells population. We found that gene expression profiles of AML patients and healthy controls can be classified at the individual level with high accuracy (>70 %) when using progenitor cells, suggesting the existence of subject-specific single cell transcriptomics profiles. The analysis also revealed molecular determinants of patient heterogeneity (e.g. TPSD1, CT45A1, and GABRA4) which could support new strategies for patient stratification and personalized treatment in leukaemia.

Genetic deletion of JAM-C in preleukemic cells rewires leukemic stem cell gene expression program in AML

ABSTRACT

The leukemic stem cell (LSC) score LSC-17 based on a stemness-related gene expression signature is an indicator of poor disease outcome in acute myeloid leukemia (AML). However, it is not known whether "niche anchoring" of LSC affects disease evolution. To address this issue, we conditionally inactivated the adhesion molecule JAM-C (Junctional Adhesion Molecule-C) expressed by hematopoietic stem cells (HSCs) and LSCs in an inducible mixed-lineage leukemia (iMLL)-AF9-driven AML mouse model. Deletion of Jam3 (encoding JAM-C) before induction of the leukemia-initiating iMLL-AF9 fusion resulted in a shift from long-term to short-term HSC expansion, without affecting disease initiation and progression. In vitro experiments showed that JAM-C controlled leukemic cell nesting irrespective of the bone marrow stromal cells used. RNA sequencing performed on leukemic HSCs isolated from diseased mice revealed that genes upregulated in Jam3-deficient animals belonged to activation protein-1 (AP-1) and tumor necrosis factor α (TNF-α)/NF-κB pathways. Human orthologs of dysregulated genes allowed to identify a score that was distinct from, and complementary to, the LSC-17 score. Substratification of patients with AML using LSC-17 and AP-1/TNF-α genes signature defined 4 groups with median survival ranging from <1 year to a median of "not reached" after 8 years. Finally, coculture experiments showed that AP-1 activation in leukemic cells was dependent on the nature of stromal cells. Altogether, our results identify the AP-1/TNF-α gene signature as a proxy of LSC anchoring in bone marrow niches, which improves the prognostic value of the LSC-17 score. This trial was registered at www.ClinicalTrials.gov as #NCT02320656.

The novel prognostic analysis of AML based on ferroptosis and cuproptosis related genes

BACKGROUND

Acute myeloid leukemia (AML) is a hematological malignancy. The aim of this research was to develop a ferroptosis and cuproptosis related novel prognostic signature associated with AML.

METHODS

The ferroptosis and cuproptosis related genes correlated with the prognosis of AML were identified by univariate Cox analysis. The consistent cluster analysis was performed for 150 AML patients in TCGA dataset. The key module genes associated with GSVA score of ferroptosis and cuproptosis were identified by WGCNA. univariate Cox and LASSO regression analysis were adopted to build a ferroptosis and cuproptosis AML prognostic signature. Finally, the expression of five prognostic genes in clinical tissue samples were verified by RT-qPCR.

RESULTS

A grand total of 27 FCRGs associated with AML prognosis were identified.Then, two AML sub-types with significantly different survival were obtained. We found 3 significantly differential expressed immune cells (naive CD4 cells, regulatory T cells and resting mast cells) between two risk sub-groups. Meanwhile, 'IL6 JAK STAT3 signaling' and 'P53 pathway' were enriched in low-risk group. A ferroptosis and cuproptosis related prognostic signature was build based on 8 prognostic genes. RT-qPCR results indicated that there was no significant difference in the expression of OLFML2A and CD109 between AML and normal samples. However, compared to the control group, LGALS1, SOCS1, and RHOC showed significantly lower expression in the AML group.

CONCLUSION

The prognostic signature comprised of OLFML2A, LGALS1, ABCB11, SOCS1, RHOC, CD109, RD3L and PTPN13 based on ferroptosis and cuproptosis was established, which provided theoretical basis for the research of AML.

Machine learning and integrative multi-omics network analysis for survival prediction in acute myeloid leukemia

BACKGROUND

Acute myeloid leukemia (AML) is the most common malignant myeloid disorder in adults and the fifth most common malignancy in children, necessitating advanced technologies for outcome prediction.

METHOD

This study aims to enhance prognostic capabilities in AML by integrating multi-omics data, especially gene expression and methylation, through network-based feature selection methodologies. By employing artificial intelligence and network analysis, we are exploring different methods to build a machine learning model for predicting AML patient survival. We evaluate the effectiveness of combining omics data, identify the most informative method for network integration and compare the performance with standard feature selection methods.

RESULTS

Our findings demonstrate that integrating gene expression and methylation data significantly improves prediction accuracy compared to single omics data. Among network integration methods, our study identifies the best approach that improves informative feature selection for predicting patient outcomes in AML. Comparative analyses demonstrate the superior performance of the proposed network-based methods over standard techniques.

CONCLUSIONS

This research presents an innovative and robust methodology for building a survival prediction model tailored to AML patients. By leveraging multilayer network analysis for feature selection, our approach contributes to improving the understanding and prognostic capabilities in AML and laying the foundation for more effective personalized therapeutic interventions in the future.

A 4-gene prognostic index for enhancing acute myeloid leukaemia survival prediction

Despite advancements in utilizing genetic markers to enhance acute myeloid leukaemia (AML) outcome prediction, significant disease heterogeneity persists, hindering clinical management. To refine survival predictions, we assessed the transcriptome of non-acute promyelocytic leukaemia chemotherapy-treated AML patients from five cohorts (n = 975). This led to the identification of a 4-gene prognostic index (4-PI) comprising CYP2E1, DHCR7, IL2RA and SQLE. The 4-PI effectively stratified patients into risk categories, with the high 4-PI group exhibiting TP53 mutations and cholesterol biosynthesis signatures. Single-cell RNA sequencing revealed enrichment for leukaemia stem cell signatures in high 4-PI cells. Validation across three cohorts (n = 671), including one with childhood AML, demonstrated the reproducibility and clinical utility of the 4-PI, even using cost-effective techniques like real-time quantitative polymerase chain reaction. Comparative analysis with 56 established prognostic indexes revealed the superior performance of the 4-PI, highlighting its potential to enhance AML risk stratification. Finally, the 4-PI demonstrated to be potential marker to reclassified patients from the intermediate ELN2017 category to the adverse category. In conclusion, the 4-PI emerges as a robust and straightforward prognostic tool to improve survival prediction in AML patients.

A CD38-directed, single-chain T-cell engager targets leukemia stem cells through IFN-γ-induced CD38 expression

ABSTRACT

Treatment resistance of leukemia stem cells (LSCs) and suppression of the autologous immune system represent major challenges to achieve a cure in acute myeloid leukemia (AML). Although AML blasts generally retain high levels of surface CD38 (CD38pos), LSCs are frequently enriched in the CD34posCD38neg blast fraction. Here, we report that interferon gamma (IFN-γ) reduces LSCs clonogenic activity and induces CD38 upregulation in both CD38pos and CD38neg LSC-enriched blasts. IFN-γ-induced CD38 upregulation depends on interferon regulatory factor 1 transcriptional activation of the CD38 promoter. To leverage this observation, we created a novel compact, single-chain CD38-CD3 T-cell engager (BN-CD38) designed to promote an effective immunological synapse between CD38pos AML cells and both CD8pos and CD4pos T cells. We demonstrate that BN-CD38 engages autologous CD4pos and CD8pos T cells and CD38pos AML blasts, leading to T-cell activation and expansion and to the elimination of leukemia cells in an autologous setting. Importantly, BN-CD38 engagement induces the release of high levels of IFN-γ, driving the expression of CD38 on CD34posCD38neg LSC-enriched blasts and their subsequent elimination. Critically, although BN-CD38 showed significant in vivo efficacy across multiple disseminated AML cell lines and patient-derived xenograft models, it did not affect normal hematopoietic stem cell clonogenicity and the development of multilineage human immune cells in CD34pos humanized mice. Taken together, this study provides important insights to target and eliminate AML LSCs.

Towards In Silico Identification of Genes Contributing to Similarity of Patients' Multi-Omics Profiles: A Case Study of Acute Myeloid Leukemia

We propose a computational framework for selecting biologically plausible genes identified by clustering of multi-omics data that reveal patients' similarity, thus giving researchers a more comprehensive view on any given disease. We employ spectral clustering of a similarity network created by fusion of three similarity networks, based on mRNA expression of immune genes, miRNA expression and DNA methylation data, using SNF_v2.1 software. For each cluster, we rank multi-omics features, ensuring the best separation between clusters, and select the top-ranked features that preserve clustering. To find genes targeted by DNA methylation and miRNAs found in the top-ranked features, we use chromosome-conformation capture data and miRNet2.0 software, respectively. To identify informative genes, these combined sets of target genes are analyzed in terms of their enrichment in somatic/germline mutations, GO biological processes/pathways terms and known sets of genes considered to be important in relation to a given disease, as recorded in the Molecular Signature Database from GSEA. The protein-protein interaction (PPI) networks were analyzed to identify genes that are hubs of PPI networks. We used data recorded in The Cancer Genome Atlas for patients with acute myeloid leukemia to demonstrate our approach, and discuss our findings in the context of results in the literature.

TP53-Mutated Myelodysplasia and Acute Myeloid Leukemia

TP53-mutated myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) form a distinct and heterogeneous group of myeloid malignancies associated with poor outcomes. Studies carried out in the last years have in part elucidated the complex role played by TP53 mutations in the pathogenesis of these myeloid disorders and in the mechanisms of drug resistance. A consistent number of studies has shown that some molecular parameters, such as the presence of a single or multiple TP53 mutations, the presence of concomitant TP53 deletions, the association with co-occurring mutations, the clonal size of TP53 mutations, the involvement of a single (monoallelic) or of both TP53 alleles (biallelic) and the cytogenetic architecture of concomitant chromosome abnormalities are major determinants of outcomes of patients. The limited response of these patients to standard treatments, including induction chemotherapy, hypomethylating agents and venetoclax-based therapies and the discovery of an immune dysregulation have induced a shift to new emerging therapies, some of which being associated with promising efficacy. The main aim of these novel immune and nonimmune strategies consists in improving survival and in increasing the number of TP53-mutated MDS/AML patients in remission amenable to allogeneic stem cell transplantation.

Identifying a prognostic model and screening of potential natural compounds for acute myeloid leukemia

Background

Acute myeloid leukemia (AML) is one of the most common hematologic malignancies with a poor prognosis and high recurrence rate. The discovery of new predictive models and therapeutic agents plays a crucial role.

Methods

The differentially expressed gene that was explicitly highly expressed in The Cancer Genome Atlas (TCGA) and GSE9476 transcriptome databases were screened and included in the least absolute shrinkage and selection operator (LASSO) regression model to derive risk coefficients and build a risk score model. Functional enrichment analysis was conducted on the screened hub genes to explore the potential mechanisms. Subsequently, critical genes were incorporated into a nomogram model based on risk scores to analyze prognostic value. Finally, this study combined network pharmacology to find potential natural compounds for hub genes and used molecular docking to verify the binding ability of molecular structures to natural compounds to explore drug development for possible efficacy in AML.

Results

A total of 33 highly expressed genes may be associated with poor prognosis of AML patients. After LASSO and multivariate Cox regression analysis of 33 critical genes, Rho-related BTB domain containing 2 (RHOBTB2), phospholipase A2 (PLA2G4A), interleukin-2 receptor-α (IL2RA), cysteine and glycine-rich protein 1 (CSRP1), and olfactomedin-like 2A (OLFML2A) were found to played a significant role in the prognosis of AML patients. CSRP1 and OLFML2A were independent prognostic factors of AML. The predictive power of these 5 hub genes in combination with clinical features was better than clinical data alone in predicting AML in the column line graphs and had better predictive value at 1, 3, and 5 years. Finally, through network pharmacology and molecular docking, this study found that diosgenin in Guadi docked well with PLA2G4A, beta-sitosterol in Fangji docked well with IL2RA, and OLFML2A docked well with 3,4-di-O-caffeoylquinic acid in Beiliujinu.

Conclusions

The predictive model of RHOBTB2, PLA2G4A, IL2RA, CSRP1, and OLFML2A combined with clinical features can better guide the prognosis of AML. In addition, the stable docking of PLA2G4A, IL2RA, and OLFML2A with natural compounds may provide new options for treating AML.

Challenging gold standard hematology diagnostics through the introduction of whole genome sequencing and artificial intelligence

The diagnosis of hematological malignancies is rather complex and requires the application of a plethora of different assays, techniques and methodologies. Some of the methods, like cytomorphology, have been in use for decades, while other methods, such as next-generation sequencing or even whole genome sequencing (WGS), are relatively new. The application of the methods and the evaluation of the results require distinct skills and knowledge and place different demands on the practitioner. However, even with experienced hematologists, diagnostic ambiguity remains a regular occurrence and the comprehensive analysis of high-dimensional WGS data soon exceeds any human's capacity. Hence, in order to reduce inter-observer variability and to improve the timeliness and accuracy of diagnoses, machine learning based approaches have been developed to assist in the decision making process. Moreover, to achieve the goal of precision oncology, comprehensive genomic profiling is increasingly being incorporated into routine standard of care.

Leveraging IFNγ/CD38 regulation to unmask and target leukemia stem cells in acute myelogenous leukemia

Elimination of drug-resistant leukemia stem cells (LSCs) represents a major challenge to achieve a cure in acute myeloid leukemia (AML). Although AML blasts generally retain high levels of surface CD38 (CD38pos), the presence of CD34 and lack of CD38 expression (CD34posCD38neg) are immunophenotypic features of both LSC-enriched AML blasts and normal hematopoietic stem cells (HSCs). We report that IFN-γ induces CD38 upregulation in LSC-enriched CD34posCD38neg AML blasts, but not in CD34posCD38neg HSCs. To leverage the IFN-γ mediated CD38 up-regulation in LSCs for clinical application, we created a compact, single-chain CD38-CD3-T cell engager (CD38-BIONIC) able to direct T cells against CD38pos blasts. Activated CD4pos and CD8pos T cells not only kill AML blasts but also produce IFNγ, which leads to CD38 expression on CD34posCD38neg LSC-enriched blasts. These cells then become CD38-BIONIC targets. The net result is an immune-mediated killing of both CD38neg and CD38pos AML blasts, which culminates in LSC depletion.

Artificial intelligence in hematological diagnostics: Game changer or gadget?

The future of clinical diagnosis and treatment of hematologic diseases will inevitably involve the integration of artificial intelligence (AI)-based systems into routine practice to support the hematologists' decision making. Several studies have shown that AI-based models can already be used to automatically differentiate cells, reliably detect malignant cell populations, support chromosome banding analysis, and interpret clinical variants, contributing to early disease detection and prognosis. However, even the best tool can become useless if it is misapplied or the results are misinterpreted. Therefore, in order to comprehensively judge and correctly apply newly developed AI-based systems, the hematologist must have a basic understanding of the general concepts of machine learning. In this review, we provide the hematologist with a comprehensive overview of various machine learning techniques, their current implementations and approaches in different diagnostic subfields (e.g., cytogenetics, molecular genetics), and the limitations and unresolved challenges of the systems.

Mitochondrial gene expression signature predicts prognosis of pediatric acute myeloid leukemia patients

Introduction

Gene expression profile of mitochondrial-related genes is not well deciphered in pediatric acute myeloid leukaemia (AML). We aimed to identify mitochondria-related differentially expressed genes (DEGs) in pediatric AML with their prognostic significance.

Methods

Children with de novo AML were included prospectively between July 2016-December 2019. Transcriptomic profiling was done for a subset of samples, stratified by mtDNA copy number. Top mitochondria-related DEGs were identified and validated by real-time PCR. A prognostic gene signature risk score was formulated using DEGs independently predictive of overall survival (OS) in multivariable analysis. Predictive ability of the risk score was estimated along with external validation in The Tumor Genome Atlas (TCGA) AML dataset.

Results

In 143 children with AML, twenty mitochondria-related DEGs were selected for validation, of which 16 were found to be significantly dysregulated. Upregulation of SDHC (p<0.001), CLIC1 (p=0.013) and downregulation of SLC25A29 (p<0.001) were independently predictive of inferior OS, and included for developing prognostic risk score. The risk score model was independently predictive of survival over and above ELN risk categorization (Harrell's c-index: 0.675). High-risk patients (risk score above median) had significantly inferior OS (p<0.001) and event free survival (p<0.001); they were associated with poor-risk cytogenetics (p=0.021), ELN intermediate/poor risk group (p=0.016), absence of RUNX1-RUNX1T1 (p=0.027), and not attaining remission (p=0.016). On external validation, the risk score also predicted OS (p=0.019) in TCGA dataset.

Discussion

We identified and validated mitochondria-related DEGs with prognostic impact in pediatric AML and also developed a novel 3-gene based externally validated gene signature predictive of survival.

A DNA Damage Response Related Signature to Predict Prognosis in Patients with Acute Myeloid Leukemia

The prognosis of acute myeloid leukemia (AML) is disappointing in most subtypes and varies widely. DNA damage response (DDR) is associated with prognosis and immunotherapy in multiple cancers. Here, we identify a signature of eight DDR-related genes associated with overall survival, which stratifies AML patients into high- and low-risk groups. Patients in low-risk group were more likely to respond to sorafenib. The signature could be an independent prognostic predictor for patients treated with ADE and ADE plus gemtuzumab ozogamicin. Therefore, this DDR prognostic signature might be applied to prognostic stratification and treatment selection in AML patients, which warrants further studies.

Immune dysfunction signatures predict outcomes and define checkpoint blockade-unresponsive microenvironments in acute myeloid leukemia

BackgroundImmune exhaustion and senescence are dominant dysfunctional states of effector T cells and major hurdles for the success of cancer immunotherapy. In the current study, we characterized how acute myeloid leukemia (AML) promotes the generation of senescent-like CD8+ T cells and whether they have prognostic relevance.METHODSWe analyzed NanoString, bulk RNA-Seq and single-cell RNA-Seq data from independent clinical cohorts comprising 1,896 patients treated with chemotherapy and/or immune checkpoint blockade (ICB).ResultsWe show that senescent-like bone marrow CD8+ T cells were impaired in killing autologous AML blasts and that their proportion negatively correlated with overall survival (OS). We defined what we believe to be new immune effector dysfunction (IED) signatures using 2 gene expression profiling platforms and reported that IED scores correlated with adverse-risk molecular lesions, stemness, and poor outcomes; these scores were a more powerful predictor of OS than 2017-ELN risk or leukemia stem cell (LSC17) scores. IED expression signatures also identified an ICB-unresponsive tumor microenvironment and predicted significantly shorter OS.ConclusionThe IED scores provided improved AML-risk stratification and could facilitate the delivery of personalized immunotherapies to patients who are most likely to benefit.TRIAL REGISTRATIONClinicalTrials.gov; NCT02845297.FUNDINGJohn and Lucille van Geest Foundation, Nottingham Trent University's Health & Wellbeing Strategic Research Theme, NIH/NCI P01CA225618, Genentech-imCORE ML40354, Qatar National Research Fund (NPRP8-2297-3-494).

A six-gene prognostic signature for both adult and pediatric acute myeloid leukemia identified with machine learning

BACKGROUND

Although it is well-known that adult and pediatric acute myeloid leukemias (AMLs) are genetically distinct diseases, they still share certain gene expression profiles. The age-related genetic heterogeneities of AMLs have been well-studied, but the common prognostic signatures and molecular mechanisms of adult and pediatric AMLs are less investigated.

AIM

To identify genes and pathways that are associated with both pediatric and adult AMLs and discover a gene signature for overall survival (OS) prediction.

METHODS

Through mining the transcriptome profiles of The Cancer Genome Atlas (TCGA) data sets of adult cancers and The Therapeutically Applicable Research to Generate Effective Treatments (TARGET) data of pediatric cancers, we identified genes that are commonly dysregulated in both pediatric and adult AMLs, further discovered a common gene signature, and built two risk score models for TCGA and TARGET cohorts, respectively with L ₀ regularized global AUC (area under the receiver operating characteristic curve) summary maximization.

RESULTS

We identified 57 genes that are differentially expressed and prognostically significant in both adult and childhood AMLs. The top 4 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways enriched with those 57 genes include transcriptional misregulation, focal adhesion, PI3K-Akt signaling pathway, and signaling pathways regulating pluripotency of stem cells. We further identified a 6-gene signature including genes of ADAMTS3, DNMT3B, NYNRIN, SORT1, ZFHX3, and ZG16B for risk prediction. We constructed a risk score model with one dataset (either TCGA or TARGET) and evaluated its performance with the other. The test AUCs for the risk prediction of TCGA data with a 2-year and 5-year OS cutoffs are 0.762 (P = 2.33e-13, 95% CI: 0.69-0.83) and 0.759 (P = 7.26e-08, 95% CI: 0.66-0.85), respectively, while the test AUCs of TARGET data with the same cutoffs are 0.71 (P = 3.3e-07, 95% CI: 0.62-0.79) and 0.72 (P= 5.25e-09, 95% CI: 0.65-0.80), respectively. We further stratified patients into 3 equal sized prognostic subtypes with the 6-gene risk scores. The P-values of the tertile partitions are 1.74e-07 and 3.28e-08 for the TARGET and TCGA cohorts, respectively, which are significantly better than the standard cytogenetic risk stratification of both cohorts (TARGET: P = 1.64e-06; TCGA: P = 1.79e-05). When validated with two other independent cohorts, the 6-gene risk score models remain a significant predictor for OS. Investigating the common gene expression program is significant in that we may extrapolate the findings from adults to children and avoid unnecessary pediatric clinical trials.

A 13-gene signature to predict the prognosis and immunotherapy responses of lung squamous cell carcinoma

Lung squamous cell carcinoma (LUSC) comprises 20–30% of all lung cancers. Immunotherapy has significantly improved the prognosis of LUSC patients; however, only a small subset of patients responds to the treatment. Therefore, we aimed to develop a novel multi-gene signature associated with the immune phenotype of the tumor microenvironment for LUSC prognosis prediction. We stratified the LUSC patients from The Cancer Genome Atlas dataset into hot and cold tumor according to a combination of infiltration status of immune cells and PD-L1 expression level. Kaplan–Meier analysis showed that hot tumors were associated with shorter overall survival (OS). Enrichment analyses of differentially expressed genes (DEGs) between the hot and cold tumors suggested that hot tumors potentially have a higher immune response ratio to immunotherapy than cold tumors. Subsequently, hub genes based on the DEGs were identified and protein–protein interactions were constructed. Finally, we established an immune-related 13-gene signature based on the hub genes using the least absolute shrinkage and selection operator feature selection and multivariate cox regression analysis. This gene signature divided LUSC patients into high-risk and low-risk groups and the former inclined worse OS than the latter. Multivariate cox proportional hazard regression analysis showed that the risk model constructed by the 13 prognostic genes was an independent risk factor for prognosis. Receiver operating characteristic curve analysis showed a moderate predictive accuracy for 1-, 3- and 5-year OS. The 13-gene signature also performed well in four external cohorts (three LUSC and one melanoma cohorts) from Gene Expression Omnibus. Overall, in this study, we established a reliable immune-related 13-gene signature that can stratify and predict the prognosis of LUSC patients, which might serve clinical use of immunotherapy.

A cellular hierarchy framework for understanding heterogeneity and predicting drug response in acute myeloid leukemia

The treatment landscape of acute myeloid leukemia (AML) is evolving, with promising therapies entering clinical translation, yet patient responses remain heterogeneous, and biomarkers for tailoring treatment are lacking. To understand how disease heterogeneity links with therapy response, we determined the leukemia cell hierarchy makeup from bulk transcriptomes of more than 1,000 patients through deconvolution using single-cell reference profiles of leukemia stem, progenitor and mature cell types. Leukemia hierarchy composition was associated with functional, genomic and clinical properties and converged into four overall classes, spanning Primitive, Mature, GMP and Intermediate. Critically, variation in hierarchy composition along the Primitive versus GMP or Primitive versus Mature axes were associated with response to chemotherapy or drug sensitivity profiles of targeted therapies, respectively. A seven-gene biomarker derived from the Primitive versus Mature axis was associated with response to 105 investigational drugs. Cellular hierarchy composition constitutes a novel framework for understanding disease biology and advancing precision medicine in AML.

Multi-Faceted Effects of ST6Gal1 Expression on Precursor B-Lineage Acute Lymphoblastic Leukemia

Normal early human B-cell development from lymphoid progenitors in the bone marrow depends on instructions from elements in that microenvironment that include stromal cells and factors secreted by these cells including the extracellular matrix. Glycosylation is thought to play a key role in such interactions. The sialyltransferase ST6Gal1, with high expression in specific hematopoietic cell types, is the only enzyme thought to catalyze the terminal addition of sialic acids in an α2-6-linkage to galactose on N-glycans in such cells. Expression of ST6Gal1 increases as B cells undergo normal B-lineage differentiation. B-cell precursor acute lymphoblastic leukemias (BCP-ALLs) with differentiation arrest at various stages of early B-cell development have widely different expression levels of ST6GAL1 at diagnosis, with high ST6Gal1 in some but not in other relapses. We analyzed the consequences of increasing ST6Gal1 expression in a diagnosis sample using lentiviral transduction. NSG mice transplanted with these BCP-ALL cells were monitored for survival. Compared to mice transplanted with leukemia cells expressing original ST6Gal1 levels, increased ST6Gal1 expression was associated with significantly reduced survival. A cohort of mice was also treated for 7 weeks with vincristine chemotherapy to induce remission and then allowed to relapse. Upon vincristine discontinuation, relapse was detected in both groups, but mice transplanted with ST6Gal1 overexpressing BCP-ALL cells had an increased leukemia burden and shorter survival than controls. The BCP-ALL cells with higher ST6Gal1 were more resistant to long-term vincristine treatment in an ex vivo tissue co-culture model with OP9 bone marrow stromal cells. Gene expression analysis using RNA-seq showed a surprisingly large number of genes with significantly differential expression, of which approximately 60% increased mRNAs, in the ST6Gal1 overexpressing BCP-ALL cells. Pathways significantly downregulated included those involved in immune cell migration. However, ST6Gal1 knockdown cells also showed increased insensitivity to chemotherapy. Our combined results point to a context-dependent effect of ST6Gal1 expression on BCP-ALL cells, which is discussed within the framework of its activity as an enzyme with many N-linked glycoprotein substrates.

CALCRL Gene is a Suitable Prognostic Factor in AML/ETO+ AML Patients

Introduction

The t(8 ; 21) translocation is the most common chromosomal abnormality in human acute myeloid leukemia (AML) subtype 2 (M2), which forms the AML/ETO fusion gene. However, AML/ETO alone does not necessarily cause leukemia. Other factors are thought to contribute to the disease. Calcitonin receptor-like (CALCRL), a G-protein-coupled neuropeptide receptor, is involved in various biological processes, such as colony formation and drug resistance.

Methods

First, The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases were used to determine any differences in CALCRL expression in AML patients with and without AML/ETO and the prognostic significance of CALCRL expression in AML patients was further evaluated. Next, we detected the CALCRL expression level in 67 AML/ETO⁺ AML patients and 16 patients with nonmalignant hematological diseases using qRT-PCR and identified its prognostic relevance.

Results

Individuals in the group expressing low levels of CALCRL had a longer median survival time. In AML/ETO⁺ AML patients, higher mRNA levels of CALCRL were observed before treatment, which decreased after the complete remission that followed multiple chemotherapy sessions. Clinical features indicated that more patients in the CALCRL^high group also had c-kit mutations compared with patients in other groups. Overall survival (OS) was longer in patients with lower levels of CALCRL expression, especially in patients with c-kit mutations or with more blast cells in bone marrow (BM). In addition, a longer OS was observed in the CALCRL^low group after hematopoietic stem cell transplantation (HSCT).

Conclusions

This preliminary study indicates that CALCRL could serve as a suitable prognostic factor in AML/ETO⁺ AML patients.

Release of IFN-γ by acute myeloid leukemia cells remodels bone marrow immune microenvironment by inducing regulatory T cells

PURPOSE

The stromal and immune bone marrow (BM) landscape is emerging as a crucial determinant for acute myeloid leukemia (AML). Regulatory T cells (Tregs) are enriched in the AML microenvironment, but the underlying mechanisms are poorly elucidated. Here, we addressed the effect of IFN-γ released by AML cells in BM Tregs induction and its impact on AML prognosis.

EXPERIMENTAL DESIGN

BM aspirates from AML patients were subdivided according to IFNG expression. Gene expression profiles in INFGhigh and IFNGlow samples were compared by microarray and NanoString analysis and used to compute a prognostic index. The IFN-g release effect on the BM microenvironment was investigated in mesenchymal stromal cell (MSC)/AML cell co-cultures. In mice, AML cells silenced for IFN-γ expression were injected intrabone.

RESULTS

IFNGhigh AMLsamples showed an upregulation of inflammatory genes, usually correlated with a good prognosis in cancer. By contrast, in AML patients, high IFNG expression associated with poor overall survival. Notably, IFN-g release by AML cells positively correlated with a higher BM suppressive Tregs' frequency. In co-culture experiments, IFNGhigh AML cells modified MSC transcriptome by up-regulating IFN-γ-dependent genes related to Treg induction, including indoleamine 2,3-dioxygenase 1 (IDO1). IDO1 inhibitor abrogated the effect of IFN-γ release by AML cells on MSC-derived Treg induction. Invivo, the genetic ablation of IFN-γ production by AML cells reduced MSC IDO1 expression and Treg infiltration, hindering AML engraftment.

CONCLUSIONS

IFN-g release by AML cells induces an immune-regulatory program in MSCs and remodels BM immunological landscape toward Treg induction, contributing to an immunotolerant microenvironment.

Survival prediction in acute myeloid leukemia using gene expression profiling

BACKGROUND

Acute myeloid leukemia (AML) is a genetically heterogeneous blood disorder. AML patients are associated with a relatively poor overall survival. The objective of this study was to establish a machine learning model to accurately perform the prognosis prediction in AML patients.

METHODS

We first screened for prognosis-related genes using Kaplan-Meier survival analysis in The Cancer Genome Atlas dataset and validated the results in the Oregon Health & Science University dataset. With a random forest model, we built a prognostic risk score using patient's age, TP53 mutation, ELN classification and normalized 197 gene expression as predictor variable. Gene set enrichment analysis was implemented to determine the dysregulated gene sets between the high-risk and low-risk groups. Similarity Network Fusion (SNF)-based integrative clustering was performed to identify subgroups of AML patients with different clinical features.

RESULTS

The random forest model was deemed the best model (area under curve value, 0.75). The random forest-derived risk score exhibited significant association with shorter overall survival in AML patients. The gene sets of pantothenate and coa biosynthesis, glycerolipid metabolism, biosynthesis of unsaturated fatty acids were significantly enriched in phenotype high risk score. SNF-based integrative clustering indicated three distinct subsets of AML patients in the TCGA cohort. The cluster3 AML patients were characterized by older age, higher risk score, more frequent TP53 mutations, higher cytogenetics risk, shorter overall survival.

CONCLUSIONS

The random forest-based risk score offers an effective method to perform prognosis prediction for AML patients.

Identification and validation of CALCRL-associated prognostic genes in acute myeloid leukemia

In the past few decades, several advances have been made in the field of acute myeloid leukemia (AML), especially in the development of novel drugs. However, the overall survival rate remains particularly disappointing due to a high rate of chemotherapy resistance and relapse. The calcitonin receptor-like receptor (CALCRL) is a novel promising therapeutic target of AML and has been indicated to be strongly correlated with chemotherapy resistance and relapse driven by leukemic stem cells. Nevertheless, the CALCRL downstream genes associated with the drug resistance and relapse of AML remain to be elucidated. Within this study, we used multiple gene expression datasets from the Gene Expression Omnibus (GEO) database and cBioPortal to explore the candidate CALCRL-associated genes that could potentially mediate the chemoresistance and relapse of AML. Then, we investigated the prognostic value, coexpression relationship with CALCRL, and expression characteristics of these genes using independent data from The Cancer Genome Atlas (TCGA). Eventually, three genes were screened out as CALCRL-associated prognostic genes. The expression of AGTPBP1 and LYST was negatively correlated with CALCRL, high expression of which was associated with favorable prognosis in AML. In contrast, the expression of ETS2 was positively correlated with CALCRL, high expression of which was associated with poor prognosis in AML. The results indicated that the three prognostic genes are potential CALCRL downstream genes that mediate drug resistance and relapse in AML. This study helps to further explore the role and molecular pathways of CALCRL in mediating drug resistance and relapse of AML.

A novel 4-mRNA signature predicts the overall survival in acute myeloid leukemia

Acute myeloid leukemia (AML) is an aggressive cancer of myeloid cells with high levels of heterogeneity and great variability in prognostic behaviors. Cytogenetic abnormalities and genetic mutations have been widely used in the prognostic stratification of AML to assign patients into different risk categories. Nevertheless, nearly half of AML patients assigned to intermediate risk need more precise prognostic schemes. Here, 336 differentially expressed genes (DEGs) between AML and control samples and 206 genes representing the intratumor heterogeneity of AML were identified. By applying a LASSO Cox regression model, we generated a 4-mRNA prognostic signature comprising KLF9, ENPP4, TUBA4A and CD247. Higher risk scores were significantly associated with shorter overall survival, complex karyotype, and adverse mutations. We then validated the prognostic value of this 4-mRNA signature in two independent cohorts. We also proved that incorporation of the 4-mRNA-based signature in the 2017 European LeukemiaNet (ELN) risk classification could enhance the predictive accuracy of survival in patients with AML. Univariate and multivariate analyses showed that this signature was independent of traditional prognostic factors such as age, WBC count, and unfavorable cytogenetics. Finally, the molecular mechanisms underlying disparate outcomes in high-risk and low-risk AML patients were explored. Therefore, our findings suggest that the 4-mRNA signature refines the risk stratification and prognostic prediction of AML patients.

Survival prognostic factors in patients with acute myeloid leukemia using machine learning techniques

This paper identifies prognosis factors for survival in patients with acute myeloid leukemia (AML) using machine learning techniques. We have integrated machine learning with feature selection methods and have compared their performances to identify the most suitable factors in assessing the survival of AML patients. Here, six data mining algorithms including Decision Tree, Random Forrest, Logistic Regression, Naive Bayes, W-Bayes Net, and Gradient Boosted Tree (GBT) are employed for the detection model and implemented using the common data mining tool RapidMiner and open-source R package. To improve the predictive ability of our model, a set of features were selected by employing multiple feature selection methods. The accuracy of classification was obtained using 10-fold cross-validation for the various combinations of the feature selection methods and machine learning algorithms. The performance of the models was assessed by various measurement indexes including accuracy, kappa, sensitivity, specificity, positive predictive value, negative predictive value, and area under the ROC curve (AUC). Our results showed that GBT with an accuracy of 85.17%, AUC of 0.930, and the feature selection via the Relief algorithm has the best performance in predicting the survival rate of AML patients.

Immune landscapes predict chemotherapy resistance and immunotherapy response in acute myeloid leukemia

Acute myeloid leukemia (AML) is a molecularly and clinically heterogeneous hematological malignancy. Although immunotherapy may be an attractive modality to exploit in patients with AML, the ability to predict the groups of patients and the types of cancer that will respond to immune targeting remains limited. This study dissected the complexity of the immune architecture of AML at high resolution and assessed its influence on therapeutic response. Using 442 primary bone marrow samples from three independent cohorts of children and adults with AML, we defined immune-infiltrated and immune-depleted disease classes and revealed critical differences in immune gene expression across age groups and molecular disease subtypes. Interferon (IFN)-γ-related mRNA profiles were predictive for both chemotherapy resistance and response of primary refractory/relapsed AML to flotetuzumab immunotherapy. Our compendium of microenvironmental gene and protein profiles provides insights into the immuno-biology of AML and could inform the delivery of personalized immunotherapies to IFN-γ-dominant AML subtypes.

How artificial intelligence might disrupt diagnostics in hematology in the near future

Artificial intelligence (AI) is about to make itself indispensable in the health care sector. Examples of successful applications or promising approaches range from the application of pattern recognition software to pre-process and analyze digital medical images, to deep learning algorithms for subtype or disease classification, and digital twin technology and in silico clinical trials. Moreover, machine-learning techniques are used to identify patterns and anomalies in electronic health records and to perform ad-hoc evaluations of gathered data from wearable health tracking devices for deep longitudinal phenotyping. In the last years, substantial progress has been made in automated image classification, reaching even superhuman level in some instances. Despite the increasing awareness of the importance of the genetic context, the diagnosis in hematology is still mainly based on the evaluation of the phenotype. Either by the analysis of microscopic images of cells in cytomorphology or by the analysis of cell populations in bidimensional plots obtained by flow cytometry. Here, AI algorithms not only spot details that might escape the human eye, but might also identify entirely new ways of interpreting these images. With the introduction of high-throughput next-generation sequencing in molecular genetics, the amount of available information is increasing exponentially, priming the field for the application of machine learning approaches. The goal of all the approaches is to allow personalized and informed interventions, to enhance treatment success, to improve the timeliness and accuracy of diagnoses, and to minimize technically induced misclassifications. The potential of AI-based applications is virtually endless but where do we stand in hematology and how far can we go?

Gene expression changes contribute to stemness and therapy resistance of relapsed acute myeloid leukemia: roles of SOCS2, CALCRL, MTSS1, and KDM6A

Relapse is associated with therapy resistance and is a major cause of death in acute myeloid leukemia (AML). It is thought to result from the accretion of therapy-refractory leukemic stem cells. Genetic and transcriptional changes that are recurrently gained at relapse are likely to contribute to the increased stemness and decreased therapy responsiveness at this disease stage. Despite the recent approval of several targeted drugs, chemotherapy with cytosine arabinoside and anthracyclines is still the mainstay of AML therapy. Accordingly, a number of studies have investigated genetic and gene expression changes between diagnosis and relapse of patients subjected to such treatment. Genetic alterations recurrently acquired at relapse were identified, but were restricted to small proportions of patients, and their functional characterization is still largely pending. In contrast, the expression of a substantial number of genes was altered consistently between diagnosis and recurrence of AML. Recent studies corroborated the roles of the upregulation of SOCS2 and CALCRL and of the downregulation of MTSS1 and KDM6A in therapy resistance and/or stemness of AML. These findings spur the assumption that functional investigations of genes consistently altered at recurrence of AML have the potential to promote the development of novel targeted drugs that may help to improve the outcome of this currently often fatal disease.

TP53 abnormalities correlate with immune infiltration and associate with response to flotetuzumab immunotherapy in AML

Somatic TP53 mutations and 17p deletions with genomic loss of TP53 occur in 37% to 46% of acute myeloid leukemia (AML) with adverse-risk cytogenetics and correlate with primary induction failure, high risk of relapse, and dismal prognosis. Herein, we aimed to characterize the immune landscape of TP53-mutated AML and determine whether TP53 abnormalities identify a patient subgroup that may benefit from immunotherapy with flotetuzumab, an investigational CD123 × CD3 bispecific dual-affinity retargeting antibody (DART) molecule. The NanoString PanCancer IO360 assay was used to profile 64 diagnostic bone marrow (BM) samples from patients with TP53-mutated (n = 42) and TP53-wild-type (TP53-WT) AML (n = 22) and 45 BM samples from patients who received flotetuzumab for relapsed/refractory (R/R) AML (15 cases with TP53 mutations and/or 17p deletion). The comparison between TP53-mutated and TP53-WT primary BM samples showed higher expression of IFNG, FOXP3, immune checkpoints, markers of immune senescence, and phosphatidylinositol 3-kinase-Akt and NF-κB signaling intermediates in the former cohort and allowed the discovery of a 34-gene immune classifier prognostic for survival in independent validation series. Finally, 7 out of 15 patients (47%) with R/R AML and TP53 abnormalities showed complete responses to flotetuzumab (<5% BM blasts) on the CP-MGD006-01 clinical trial (NCT #02152956) and had significantly higher tumor inflammation signature, FOXP3, CD8, inflammatory chemokine, and PD1 gene expression scores at baseline compared with nonresponders. Patients with TP53 abnormalities who achieved a complete response experienced prolonged survival (median, 10.3 months; range, 3.3-21.3 months). These results encourage further study of flotetuzumab immunotherapy in patients with TP53-mutated AML.

Flotetuzumab as salvage immunotherapy for refractory acute myeloid leukemia

Approximately 50% of acute myeloid leukemia (AML) patients do not respond to induction therapy (primary induction failure [PIF]) or relapse after <6 months (early relapse [ER]). We have recently shown an association between an immune-infiltrated tumor microenvironment (TME) and resistance to cytarabine-based chemotherapy but responsiveness to flotetuzumab, a bispecific DART antibody-based molecule to CD3ε and CD123. This paper reports the results of a multicenter, open-label, phase 1/2 study of flotetuzumab in 88 adults with relapsed/refractory AML: 42 in a dose-finding segment and 46 at the recommended phase 2 dose (RP2D) of 500 ng/kg per day. The most frequent adverse events were infusion-related reactions (IRRs)/cytokine release syndrome (CRS), largely grade 1-2. Stepwise dosing during week 1, pretreatment dexamethasone, prompt use of tocilizumab, and temporary dose reductions/interruptions successfully prevented severe IRR/CRS. Clinical benefit accrued to PIF/ER patients showing an immune-infiltrated TME. Among 30 PIF/ER patients treated at the RP2D, the complete remission (CR)/CR with partial hematological recovery (CRh) rate was 26.7%, with an overall response rate (CR/CRh/CR with incomplete hematological recovery) of 30.0%. In PIF/ER patients who achieved CR/CRh, median overall survival was 10.2 months (range, 1.87-27.27), with 6- and 12-month survival rates of 75% (95% confidence interval [CI], 0.450-1.05) and 50% (95% CI, 0.154-0.846). Bone marrow transcriptomic analysis showed that a parsimonious 10-gene signature predicted CRs to flotetuzumab (area under the receiver operating characteristic curve = 0.904 vs 0.672 for the European LeukemiaNet classifier). Flotetuzumab represents an innovative experimental approach associated with acceptable safety and encouraging evidence of activity in PIF/ER patients. This trial was registered at www.clinicaltrials.gov as #NCT02152956.

Application of machine learning in the management of acute myeloid leukemia: current practice and future prospects

Machine learning (ML) is rapidly emerging in several fields of cancer research. ML algorithms can deal with vast amounts of medical data and provide a better understanding of malignant disease. Its ability to process information from different diagnostic modalities and functions to predict prognosis and suggest therapeutic strategies indicates that ML is a promising tool for the future management of hematologic malignancies; acute myeloid leukemia (AML) is a model disease of various recent studies. An integration of these ML techniques into various applications in AML management can assure fast and accurate diagnosis as well as precise risk stratification and optimal therapy. Nevertheless, these techniques come with various pitfalls and need a strict regulatory framework to ensure safe use of ML. This comprehensive review highlights and discusses recent advances in ML techniques in the management of AML as a model disease of hematologic neoplasms, enabling researchers and clinicians alike to critically evaluate this upcoming, potentially practice-changing technology.

Identification of a 6-gene signature for the survival prediction of breast cancer patients based on integrated multi-omics data analysis

Purpose

To identify a gene signature for the prognosis of breast cancer using high-throughput analysis.

Methods

RNASeq, single nucleotide polymorphism (SNP), copy number variation (CNV) data and clinical follow-up information were downloaded from The Cancer Genome Atlas (TCGA), and randomly divided into training set or verification set. Genes related to breast cancer prognosis and differentially expressed genes (DEGs) with CNV or SNP were screened from training set, then integrated together for feature selection of identify robust biomarkers using RandomForest. Finally, a gene-related prognostic model was established and its performance was verified in TCGA test set, Gene Expression Omnibus (GEO) validation set and breast cancer subtypes.

Results

A total of 2287 prognosis-related genes, 131 genes with amplified copy numbers, 724 gens with copy number deletions, and 280 genes with significant mutations screened from Genomic Variants were closely correlated with the development of breast cancer. A total of 120 candidate genes were obtained by integrating genes from Genomic Variants and those related to prognosis, then 6 characteristic genes (CD24, PRRG1, IQSEC3, MRGPRX, RCC2, and CASP8) were top-ranked by RandomForest for feature selection, noticeably, several of these have been previously reported to be associated with the progression of breast cancer. Cox regression analysis was performed to establish a 6-gene signature, which can stratify the risk of samples from training set, test set and external validation set, moreover, the five-year survival AUC of the model in the training set and validation set was both higher than 0.65. Thus, the 6-gene signature developed in the current study could serve as an independent prognostic factor for breast cancer patients.

Conclusion

This study constructed a 6-gene signature as a novel prognostic marker for predicting the survival of breast cancer patients, providing new diagnostic/prognostic biomarkers and therapeutic targets for breast cancer patients.

A three-gene signature might predict prognosis in patients with acute myeloid leukemia

The identification of effective signatures is crucial to predict the prognosis of acute myeloid leukemia (AML). The investigation aimed to identify a new signature for AML prognostic prediction by using the three-gene expression (octamer-binding transcription factor 4 (OCT4), POU domain type 5 transcription factor 1B (POU5F1B) and B-cell-specific Moloney murine leukemia virus integration site-1 pseudogene 1 (BMI1P1). The expressions of genes were obtained from our previous study. Only the specimens in which three genes were all expressed were included in this research. A three-gene signature was constructed by the multivariate Cox regression analyses to divide patients into high-risk and low-risk groups. Receiver operating characteristic (ROC) analysis of the three-gene signature (area under ROC curve (AUC) = 0.901, 95% CI: 0.821–0.981, P<0.001) indicated that it was a more valuable signature for distinguishing between patients and controls than any of the three genes. Moreover, white blood cells (WBCs, P=0.004), platelets (PLTs, P=0.017), percentage of blasts in bone marrow (BM) (P=0.011) and complete remission (CR, P=0.027) had significant differences between two groups. Furthermore, high-risk group had shorter leukemia-free survival (LFS) and overall survival (OS) than low-risk group (P=0.026; P=0.006), and the three-gene signature was a prognostic factor. Our three-gene signature for prognosis prediction in AML may serve as a prognostic biomarker.

Requirement for LIM kinases in acute myeloid leukemia

Acute myeloid leukemia (AML) is an aggressive disease for which only few targeted therapies are available. Using high-throughput RNA interference (RNAi) screening in AML cell lines, we identified LIM kinase 1 (LIMK1) as a potential novel target for AML treatment. High LIMK1 expression was significantly correlated with shorter survival of AML patients and coincided with FLT3 mutations, KMT2A rearrangements, and elevated HOX gene expression. RNAi- and CRISPR-Cas9-mediated suppression as well as pharmacologic inhibition of LIMK1 and its close homolog LIMK2 reduced colony formation and decreased proliferation due to slowed cell-cycle progression of KMT2A-rearranged AML cell lines and patient-derived xenograft (PDX) samples. This was accompanied by morphologic changes indicative of myeloid differentiation. Transcriptome analysis showed upregulation of several tumor suppressor genes as well as downregulation of HOXA9 targets and mitosis-associated genes in response to LIMK1 suppression, providing a potential mechanistic basis for the anti-leukemic phenotype. Finally, we observed a reciprocal regulation between LIM kinases (LIMK) and CDK6, a kinase known to be involved in the differentiation block of KMT2A-rearranged AML, and addition of the CDK6 inhibitor palbociclib further enhanced the anti-proliferative effect of LIMK inhibition. Together, these data suggest that LIMK are promising targets for AML therapy.

CGRP Signaling via CALCRL Increases Chemotherapy Resistance and Stem Cell Properties in Acute Myeloid Leukemia

The neuropeptide CGRP, acting through the G-protein coupled receptor CALCRL and its coreceptor RAMP1, plays a key role in migraines, which has led to the clinical development of several inhibitory compounds. Recently, high CALCRL expression has been shown to be associated with a poor prognosis in acute myeloid leukemia (AML). We investigate, therefore, the functional role of the CGRP-CALCRL axis in AML. To this end, in silico analyses, human AML cell lines, primary patient samples, and a C57BL/6-based mouse model of AML are used. We find that CALCRL is up-regulated at relapse of AML, in leukemic stem cells (LSCs) versus bulk leukemic cells, and in LSCs versus normal hematopoietic stem cells. CGRP protects receptor-positive AML cell lines and primary AML samples from apoptosis induced by cytostatic drugs used in AML therapy, and this effect is inhibited by specific antagonists. Furthermore, the CGRP antagonist olcegepant increases differentiation and reduces the leukemic burden as well as key stem cell properties in a mouse model of AML. These data provide a basis for further investigations into a possible role of CGRP-CALCRL inhibition in the therapy of AML.