Search

BackgroundThe stimulation and expansion of antigen-specific T cells ex vivo enables the targeting of a multitude of cancer antigens. However, clinical scale T-cell expansion from rare precursors requires repeated stimulations ex vivo leading to T-cell terminal effector differentiation and exhaustion that adversely impact therapeutic potential. We leveraged immune checkpoint blockade relevant to antigen-specific CD8+ human T cells to improve the expansion and function of T cells targeting clinically relevant antigens.MethodsA clinically-compliant protocol relying on peptide-pulsed monocyte-derived dendritic cells and cytokines was used to expand antigen-specific CD8+ targeting the oncogenic Epstein-Barr virus (EBV) and the tumor associated antigen (TAA) Wilms Tumor 1 (WT1) protein. The effects of antibody-mediated blockade of immune checkpoints applied to the cultures (T-cell expansion, phenotypes and function) were assessed at various time points. Genomic studies including single cell RNA (scRNA) sequencing and T-cell receptor sequencing were performed on EBV-specific T cells to inform about the impact of immune checkpoint blockade on the clonal distribution and gene expression of the expanded T cells.ResultsSeveral immune checkpoints were expressed early by ex vivo expanded antigen-specific CD8+ T cells, including PD-1 and TIM-3 with co-expression matching evidence of T-cell dysfunction as the cultures progressed. The introduction of anti-PD-L1 (expressed by the dendritic cells) and anti-TIM-3 antibodies in combination (but not individually) to the culture led to markedly improved antigen-specific T-cell expansion based on cell counts, fluorescent multimer staining and functional tests. This was not associated with evidence of T-cell dysfunction when compared to T cells expanded without immune checkpoint blockade. Genomics studies largely confirmed these results, showing that double blockade does not impart specific transcriptional programs or patterns on TCR repertoires. However, our data indicate that combined blockade may nonetheless alter gene expression in a minority of clonotypes and have donor-specific impacts.ConclusionsThe manufacturing of antigen-specific CD8+ T cells can be improved in terms of yield and functionality using blockade of TIM-3 and the PD-L1/PD-1 axis in combination. Overcoming the deleterious effects of multiple antigenic stimulations through PD-L1/TIM-3 blockade is a readily applicable approach for several adoptive-immunotherapy strategies.

Acute myeloid leukemias (AML) are characterized by mutations of tumor suppressor and oncogenes, involving distinct genes in adults and children. While certain mutations have been associated with the increased risk of AML relapse, the genomic landscape of primary chemotherapy-resistant AML is not well defined. As part of the TARGET initiative, we performed whole-genome DNA and transcriptome RNA and miRNA sequencing analysis of pediatric AML with failure of induction chemotherapy. We identified at least three genetic groups of patients with induction failure, including those with NUP98 rearrangements, somatic mutations of WT1 in the absence of apparent NUP98 mutations, and additional recurrent variants including those in KMT2C and MLLT10. Comparison of specimens before and after chemotherapy revealed distinct and invariant gene expression programs. While exhibiting overt therapy resistance, these leukemias nonetheless showed diverse forms of clonal evolution upon chemotherapy exposure. This included selection for mutant alleles of FRMD8, DHX32, PIK3R1, SHANK3, MKLN1, as well as persistence of WT1 and TP53 mutant clones, and elimination of FLT3, PTPN11, and NRAS mutant clones. These findings delineate genetic mechanisms of primary chemotherapy resistance in pediatric AML, which should inform improved approaches for its diagnosis and therapy.

Neomorphic missense mutations affecting crucial lysine residues in histone H3 genes significantly contribute to a variety of solid cancers. Despite the high prevalence of H3K27M mutations in pediatric glioblastoma and their well-established impact on global histone H3 lysine 27 di- and trimethylation (H3K27me2/3), the relevance of these mutations has not been studied in acute myeloid leukemia (AML). Here, we report the first identification of H3K27M and H3K27I mutations in patients with AML. We find that these lesions are major determinants of reduced H3K27me2/3 in these patients and that they are associated with common aberrations in the RUNX1 gene. We demonstrate that H3K27I/M mutations are strong disease accelerators in a RUNX1-RUNX1T1 AML mouse model, suggesting that H3K27me2/3 has an important and selective leukemia-suppressive activity in this genetic context.

Purpose: RUNX1-mutated (RUNX1mut) acute myeloid leukemia (AML) is associated with adverse outcome, highlighting the urgent need for a better genetic characterization of this AML subgroup and for the design of efficient therapeutic strategies for this disease. Toward this goal, we further dissected the mutational spectrum and gene expression profile of RUNX1mut AML and correlated these results to drug sensitivity to identify novel compounds targeting this AML subgroup. Experimental Design: RNA-sequencing of 47 RUNX1mut primary AML specimens was performed and sequencing results were compared to those of RUNX1 wild-type samples. Chemical screens were also conducted using RUNX1mut specimens to identify compounds selectively affecting the viability of RUNX1mut AML. Results: We show that samples with no remaining RUNX1 wild-type allele are clinically and genetically distinct and display a more homogeneous gene expression profile. Chemical screening revealed that most RUNX1mut specimens are sensitive to glucocorticoids (GCs) and we confirmed that GCs inhibit AML cell proliferation through their interaction with the glucocorticoid receptor (GR). We observed that specimens harboring RUNX1 mutations expected to result in low residual RUNX1 activity are most sensitive to GCs, and that coassociating mutations as well as GR levels contribute to GC sensitivity. Accordingly, acquired glucocorticoid sensitivity was achieved by negatively regulating RUNX1 expression in human AML cells. Conclusions: Our findings show the profound impact of RUNX1 allele dosage on gene expression profile and glucocorticoid sensitivity in AML, thereby opening opportunities for preclinical testing which may lead to drug repurposing and improved disease characterization. Clin Cancer Res; 23(22); 6969–81. ©2017 AACR.

Compared to solid tumors, antibody-based immunotherapy is underexploited in acute myeloid leukemia (AML), likely due to the limited availability of targetable surface antigens specifically expressed on AML cells. While targeted immunotherapies hold great promise, their development inexorably depends on the identification of cell surface antigens selectively expressed by cancer cells. The identification of such antigens has been challenging due to the characteristic hydrophobicity of plasma membrane proteins. Another level of complexity is brought by the fact that AML is a heterogeneous disease comprised of different subgroups with associated prognosis ranging from favorable to adverse. In order to identify novel AML antigens to be targeted using immunotherapeutic approaches, we have analysed the surface proteome of a collection of 100 primary human AML specimens, enriched in poor prognosis AML samples. These analyses revealed that AML subgroups have distinct surfaceome signatures. We also confirmed the expression of known AML markers at the surface of specific AML subgroups. Our search for novel AML antigens focused on four poor prognosis AML subgroups: MLL-rearranged, complex karyotype, RUNX1-mutated and normal karyotype with NPM1, DNMT3A and FLT3 (ITD) mutations. We identified proteins selectively expressed at the surface of AML cells from these subgroups. Analysis of the whole cohort of specimens also identified known and novel pan AML antigens. Single cell RNA-sequencing analysis of aforementioned subgroup-specific and pan AML antigens revealed that many of these are expressed by the vast majority of blast populations within primary AML specimens, including most primitive ones. Overall, these results suggest that studying the surface proteome of AML cells has the potential to reveal promising AML antigens to be targeted using immunotherapeutic approaches.

Developmental processes underlying normal tissue regeneration have been implicated in cancer, but the degree of their enactment during tumor progression and under the selective pressures of immune surveillance, remain unknown. Here we show that human primary lung adenocarcinomas are characterized by the emergence of regenerative cell types, typically seen in response to lung injury, and by striking infidelity among transcription factors specifying most alveolar and bronchial epithelial lineages. In contrast, metastases are enriched for key endoderm and lung-specifying transcription factors, SOX2 and SOX9, and recapitulate more primitive transcriptional programs spanning stem-like to regenerative pulmonary epithelial progenitor states. This developmental continuum mirrors the progressive stages of spontaneous outbreak from metastatic dormancy in a mouse model and exhibits SOX9-dependent resistance to natural killer cells. Loss of developmental stage-specific constraint in macrometastases triggered by natural killer cell depletion suggests a dynamic interplay between developmental plasticity and immune-mediated pruning during metastasis. Single-cell analysis of lung cancer progression uncovers developmental and regenerative programs co-opted by cancer cells and immune-mediated pruning during metastatic outbreak

Transcriptional regulators, including the cohesin complex member STAG2, are recurrently mutated in cancer. The role of STAG2 in gene regulation, hematopoiesis, and tumor suppression remains unresolved. We show Stag2 deletion in hematopoietic stem/progenitor cells (HSPC) results in altered hematopoietic function, increased self-renewal, and impaired differentiation. ChIP-sequencing revealed that while Stag2 and Stag1 can bind the same loci, a component of Stag2 binding sites are unoccupied by Stag1 even in Stag2-deficient HSPCs. While concurrent loss of Stag2 and Stag1 abrogated hematopoiesis, Stag2 loss alone decreased chromatin accessibility and transcription of lineage-specification genes, including Ebf1 and Pax5, leading to blunted HSPC commitment to the B-cell lineage. Our data illustrate a role for Stag2 in transformation and transcriptional dysregulation distinct from its shared role with Stag1 in chromosomal segregation.One Sentence SummaryStag1 rescues topologically associated domains in the absence of Stag2, but cannot restore chromatin architecture required for hematopoietic lineage commitment

In acute myeloid leukemia (AML), risk stratification based on cytogenetics and mutation profiling is essential but remains insufficient to select the optimal therapy. Accurate biomarkers are needed to improve prognostic assessment. We analyzed RNA sequencing and survival data of 430 AML patients and identified HMGA2 as a novel prognostic marker. We validated a quantitative PCR test to study the association of HMGA2 expression with clinical outcomes in 358 AML samples. In this training cohort, HMGA2 was highly expressed in 22.3% of AML, mostly in patients with intermediate or adverse cytogenetics. High expression levels of HMGA2 (H + ) were associated with a lower frequency of complete remission (58.8% vs 83.4%, P 0.001), worse 3-year overall survival (OS, 13.2% vs 43.5%, P 0.001) and relapse-free survival (RFS, 10.8% vs 44.2%, P 0.001). A positive HMGA2 test also identified a subgroup of patients unresponsive to standard treatments. Multivariable analyses showed that H + was independently associated with significantly worse OS and RFS, including in the intermediate cytogenetic risk category. These associations were confirmed in a validation cohort of 260 patient samples from the UK NCRI AML17 trial. The HMGA2 test could be implemented in clinical trials developing novel therapeutic strategies for high-risk AML.

FLT3, DNMT3A, and NPM1 are the most frequently mutated genes in cytogenetically normal acute myeloid leukemia (AML), but little is known about how these mutations synergize upon cooccurrence. Here we show that triple-mutated AML is characterized by high leukemia stem cell (LSC) frequency, an aberrant leukemia-specific GPR56highCD34low immunophenotype, and synergistic upregulation of Hepatic Leukemia Factor (HLF). Cell sorting based on the LSC marker GPR56 allowed isolation of triple-mutated from DNMT3A/NPM1 double-mutated subclones. Moreover, in DNMT3A R882-mutated patients, CpG hypomethylation at the HLF transcription start site correlated with high HLF mRNA expression, which was itself associated with poor survival. Loss of HLF via CRISPR/Cas9 significantly reduced the CD34+GPR56+ LSC compartment of primary human triple-mutated AML cells in serial xenotransplantation assays. HLF knockout cells were more actively cycling when freshly harvested from mice, but rapidly exhausted when reintroduced in culture. RNA sequencing of primary human triple-mutated AML cells after shRNA-mediated HLF knockdown revealed the NOTCH target Hairy and Enhancer of Split 1 (HES1) and the cyclin-dependent kinase inhibitor CDKN1C/p57 as novel targets of HLF, potentially mediating these effects. Overall, our data establish HLF as a novel LSC regulator in this genetically defined high-risk AML subgroup.

In this study, we analyzed RNA-sequencing data of 14 samples characterized by biallelic CEBPA (CEBPA(bi)) mutations included in the Leucegene collection of 415 primary acute myeloid leukemia (AML) specimens, and describe for the first time high frequency recurrent mutations in the granulocyte colony-stimulating factor receptor gene CSF3R, which signals through JAK-STAT proteins. Chemical interrogation of these primary human specimens revealed a uniform and specific sensitivity to all JAK inhibitors tested irrespective of their CSF3R mutation status, indicating a general sensitization of JAK-STAT signaling in this leukemia subset. Altogether, these results identified the co-occurrence of mutations in CSF3R and CEBPA in a well-defined AML subset, which uniformly responds to JAK inhibitors and paves the way to personalized clinical trials for this disease.

To the editor: RUNX1 (also known as AML1 or CBFA2 ) and CBFB encode the α and β subunits of a heterodimeric core binding transcription factor complex involved in the development of normal hematopoiesis (reviewed by de Bruijn and Speck[1][1]). Both genes are rearranged in acute myeloid leukemia (

Identification of MYC mutations in acute myeloid leukemias with NUP98 – NSD1 translocations

Complex karyotype (CK) acute myeloid leukemia (AML), especially those with TP53 anomalies, are particularly resistant to current therapies and remain one of the most heterogeneous and poorly characterized genetic group. Using global transcriptomic analyses, we identified the oncofetal HMGA2 gene as one of the most upregulated gene in CK AML. HMGA2 is a known stem cell gene and we showed, at single cell level, that HMGA2 is indeed expressed in normal hematopoietic stem cells (HSCs) but also in primitive CD34+ leukemic cells. Abnormal re-expression of HMGA2 has been reported in several tumor types and linked to chemo-resistance, advanced tumor grade and poor prognosis. Using a CRISPR/Cas9 genomic screening approach, we identified vulnerabilities in leukemic cells overexpressing HMGA2. Among them, we found that targeting the G2/M cell cycle phase transition was deleterious to those cells. We confirmed these observations using chemical inhibitors of the G2/M regulators ATR, CHK1 and WEE1 and found that these compounds were preferentially active on HMGA2+ CK AML specimens. Our findings thus suggest that HMGA2 is a key functional determinant in AML associated with stem cell features, G2/M perturbation and related drug sensitivity.

In acute myeloid leukemia (AML), risk stratification based on cytogenetics and mutation profiling is essential but remains insufficient to select the optimal therapy. Accurate biomarkers are needed to improve prognostic assessment. We analyzed RNA sequencing and survival data of 430 AML patients and identified HMGA2 as a novel prognostic marker. We validated a quantitative PCR test to study the association of HMGA2 expression with clinical outcomes in 358 AML samples. In this training cohort, HMGA2 was highly expressed in 22.3% of AML, mostly in patients with intermediate or adverse cytogenetics. High expression levels of HMGA2 (H + ) were associated with a lower frequency of complete remission (58.8% vs 83.4%, P < 0.001), worse 3-year overall survival (OS, 13.2% vs 43.5%, P < 0.001) and relapse-free survival (RFS, 10.8% vs 44.2%, P < 0.001). A positive HMGA2 test also identified a subgroup of patients unresponsive to standard treatments. Multivariable analyses showed that H + was independently associated with significantly worse OS and RFS, including in the intermediate cytogenetic risk category. These associations were confirmed in a validation cohort of 260 patient samples from the UK NCRI AML17 trial. The HMGA2 test could be implemented in clinical trials developing novel therapeutic strategies for high-risk AML.

Using next-generation sequencing of primary acute myeloid leukemia (AML) specimens, we identified to our knowledge the first unifying genetic network common to the two subgroups of KMT2A (MLL)-rearranged leukemia, namely having MLL fusions or partial tandem duplications. Within this network, we experimentally confirmed upregulation of the gene with the most subtype-specific increase in expression, LOC100289656, and identified cryptic MLL fusions, including a new MLL-ENAH fusion. We also identified a subset of MLL fusion specimens carrying mutations in SPI1 accompanied by inactivation of its transcriptional network, as well as frequent RAS pathway mutations, which sensitized the leukemias to synthetic lethal interactions between MEK and receptor tyrosine kinase inhibitors. This transcriptomics-based characterization and chemical interrogation of human MLL-rearranged AML was a valuable approach for identifying complementary features that define this disease.

The genetic and transcriptional signature of EVI1 (ecotropic viral integration site 1)-rearranged (EVI1-r) acute myeloid leukemias (AMLs) remains poorly defined. We performed RNA sequencing of 12 EVI1-r AMLs and compared the results with those of other AML subtypes (n = 139) and normal CD34+ cells (n = 17). Results confirm high frequencies of RAS and other activated signaling mutations (10/12 AMLs) and identify new recurrent mutations in splicing factors (5/12 AMLs in SF3B1 and 2/12 AMLs in U2AF1), IKZF1 (3/12 AMLs), and TP53 (3/12 AMLs). Mutations in IKZF1, a gene located on chromosome 7, and monosomy 7 are mutually exclusive in this disease. Moreover IKZF1 expression is halved in monosomy 7 leukemias. EVI-r AMLs are also characterized by a unique transcriptional signature with high expression levels of MECOM, PREX2, VIP, MYCT1, and PAWR. Our results suggest that EVI1-r AMLs could be molecularly defined by specific transcriptomic anomalies and a hitherto unseen mutational pattern. Larger patient cohorts will better determine the frequency of these events.

Cell-type specific transcriptional programs are regulated by the activity of tissue-specific transcription factors and enhancer elements within structurally defined topologically associating domains (TADs). The coordinated and dynamic changes in chromatin architecture are highly regulated as transcriptional output is influenced by chromatin accessibility, histone modification, promoter enhance interactions, and recruitment of transcriptional co-activator complexes. The genes which contribute to transcriptional regulation, including members of the cohesin complex, are frequently mutated in human cancers, including leukemias, Ewing sarcoma, and glioblastoma multiforme. Despite this, the mechanistic role of STAG2 in gene regulation, hematopoietic function, and tumor suppression has not been delineated. We show that somatic Stag2 deletion in hematopoietic stem/progenitor cells (HSPC) results in altered hematopoietic function, increased self-renewal, and impaired differentiation across all three lineages consistent with myelodysplasia. Chromatin immunoprecipitation sequencing of Stag2-deficient HSPCs revealed that Stag2 and Stag1 have both shared and non-redundant cistromes with Stag1/2 common binding sites enriching at TAD boundaries with CTCF occupancy. This maintains TAD integrity in the setting of Stag2 loss of function which we confirmed with Hi-C chromosome capture. High resolution of the Hi-C data at 10kB resolution identified a specific role for Stag2, but not Stag1, in maintaining short-range chromatin interactions, specifically at genes with PU.1 and IRF8 motifs. While co-deletion of Stag2 and Stag1 resulted in synthetic lethality, Stag2 loss alone resulted in decreased chromatin accessibility and reduced transcriptional output at key PU.1 target loci involved in lineage-specification, including reduced Ebf1 and Pax5 expression resulting in impaired B-lineage differentiation. We investigated whether expression of PU.1 could overcome the non-permissive chromatin state at these key downstream targets and rescue hematopoietic differentiation; however, PU.1 expression could not restore Ebf1 expression or B cell differentiation and did not attenuate the serial replating capacity of Stag2 deficient hematopoietic stem/progenitor cells (HSPCs). Given this transcriptional "choke-point" we investigated whether expression of Ebf1 would have a more significant impact on Stag2 deficient cells. Indeed Ebf1 rescue restored B cell colony formation/differentiation in vitro and in vivo and abrogated serial replating of Stag2 deficient HSPCs. These data highlight the non-hierarchical and non-redundant relationship between transcription factors and chromatin architecture and demonstrate a key role for Stag2-regulated local interactions in transcription factor output and hematopoietic differentiation. Nonetheless, the mechanistic underpinnings of the structural basis for transcriptional regulation remain associative. We have recently been able to reduce the cell input for Hi-C assays such that we can now analyze the chromatin architecture of purified populations and model the structural transition from Lin- Sca-1+ Kit+ hematopoietic stem cells to committed granulocyte macrophage precursor cells both in normal hematopoiesis and in the Stag2 deficient setting. Previous studies using in vitro systems have shown that complete cohesin depletion results in the loss of structural loop components; however, cohesin levels are reduced, but not absent, in cancer cells. As such, our studies highlight a key role for locus-specific alterations in gene regulation and DNA interactions in Stag2 deficient cells, which results in altered gene expression and contributes to transformation. Taken together, these data identify a key role for Stag2 loss in transcriptional dysregulation distinct from its shared role with Stag1 in chromosomal segregation. Moreover, we illustrate a critical link between cohesin, chromosomal contacts, and gene regulation that contributes to hematopoietic transformation. Disclosures Viny: Mission Bio: Other: Sponsored travel; Hematology News: Membership on an entity's Board of Directors or advisory committees. Dekker:Arima Genomics: Membership on an entity's Board of Directors or advisory committees. Levine:Imago Biosciences: Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria; Lilly: Honoraria; Gilead: Consultancy; Novartis: Consultancy; Prelude Therapeutics: Research Funding; Roche: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; C4 Therapeutics: Membership on an entity's Board of Directors or advisory committees; Isoplexis: Membership on an entity's Board of Directors or advisory committees; Qiagen: Membership on an entity's Board of Directors or advisory committees; Loxo: Membership on an entity's Board of Directors or advisory committees.

Summary Transcriptional regulators, including the cohesin complex member STAG2, are recurrently mutated in cancer. The role of STAG2 in gene regulation, hematopoiesis, and tumor suppression remains unresolved. We show that Stag2 deletion in hematopoietic stem and progenitor cells (HSPCs) results in altered hematopoietic function, increased self-renewal, and impaired differentiation. Chromatin immunoprecipitation (ChIP) sequencing revealed that, although Stag2 and Stag1 bind a shared set of genomic loci, a component of Stag2 binding sites is unoccupied by Stag1, even in Stag2-deficient HSPCs. Although concurrent loss of Stag2 and Stag1 abrogated hematopoiesis, Stag2 loss alone decreased chromatin accessibility and transcription of lineage-specification genes, including Ebf1 and Pax5, leading to increased self-renewal and reduced HSPC commitment to the B cell lineage. Our data illustrate a role for Stag2 in transformation and transcriptional dysregulation distinct from its shared role with Stag1 in chromosomal segregation.

Summary Dendritic cells (DCs) play a critical role in orchestrating adaptive immune responses due to their unique ability to initiate T cell responses and direct their differentiation into effector lineages. Classical DCs have been divided into two subsets, cDC1 and cDC2, based on phenotypic markers and their distinct abilities to prime CD8 and CD4 T cells. While the transcriptional regulation of the cDC1 subset has been well characterized, cDC2 development and function remain poorly understood. By combining transcriptional and chromatin analyses with genetic reporter expression, we identified two principal cDC2 lineages defined by distinct developmental pathways and transcriptional regulators, including T-bet and RORγt, two key transcription factors known to define innate and adaptive lymphocyte subsets. These novel cDC2 lineages were characterized by distinct metabolic and functional programs. Extending our findings to humans revealed conserved DC heterogeneity and the presence of the newly defined cDC2 subsets in human cancer., Graphical Abstract, Highlights • Single-cell analyses reveal novel dendritic cell subsets • Major cDC2 subsets differentially express T-bet and RORγt • Distinct pro- and anti-inflammatory potential of T-bet+ and Tbet– cDC2s • Transcriptional basis for cDC2 heterogeneity conserved across mouse and human, Single-cell analyses of dendritic cells reveals new subsets with distinct pro- and anti-inflammatory potential.

Appropriate culture methods for the interrogation of primary leukemic samples were hitherto lacking and current assays for compound screening are not adapted for large-scale investigation of synergistic combinations. In this study, we report a novel approach that efficiently distills synthetic lethal interactions between small molecules active on primary human acute myeloid leukemia (AML) specimens. In single-dose experiments and under culture conditions preserving leukemia stem cell activity, our strategy considerably reduces the number of tests needed for the identification of promising compound combinations. Initially conducted with a selected library of 5000 small molecules and 20 primary AML specimens, it reveals 5 broad classes of sensitized therapeutic target pathways along with their synergistic patient-specific fingerprints. This novel method opens new avenues for the development of AML personalized therapeutics and may be generalized to other tumor types, for which in vitro cancer stem cell cultures have been developed.

Genome-wide transcriptome profiling has enabled non-supervised classification of tumours, revealing different sub-groups characterized by specific gene expression features. However, the biological significance of these subtypes remains for the most part unclear. We describe herein an interactive platform, Minimum Spanning Trees Inferred Clustering (MiSTIC), that integrates the direct visualization and comparison of the gene correlation structure between datasets, the analysis of the molecular causes underlying co-variations in gene expression in cancer samples, and the clinical annotation of tumour sets defined by the combined expression of selected biomarkers. We have used MiSTIC to highlight the roles of specific transcription factors in breast cancer subtype specification, to compare the aspects of tumour heterogeneity targeted by different prognostic signatures, and to highlight biomarker interactions in AML. A version of MiSTIC preloaded with datasets described herein can be accessed through a public web server (http://mistic.iric.ca); in addition, the MiSTIC software package can be obtained (github.com/iric-soft/MiSTIC) for local use with personalized datasets.

Based on transcriptomic analyses of thousands of samples from The Cancer Genome Atlas, we report that expression of constitutive proteasome (CP) genes (PSMB5, PSMB6, PSMB7) and immunoproteasome (IP) genes (PSMB8, PSMB9, PSMB10) is increased in most cancer types. In breast cancer, expression of IP genes was determined by the abundance of tumor infiltrating lymphocytes and high expression of IP genes was associated with longer survival. In contrast, IP upregulation in acute myeloid leukemia (AML) was a cell-intrinsic feature that was not associated with longer survival. Expression of IP genes in AML was IFN-independent, correlated with the methylation status of IP genes, and was particularly high in AML with an M5 phenotype and/or MLL rearrangement. Notably, PSMB8 inhibition led to accumulation of polyubiquitinated proteins and cell death in IPhigh but not IPlow AML cells. Co-clustering analysis revealed that genes correlated with IP subunits in non-M5 AMLs were primarily implicated in immune processes. However, in M5 AML, IP genes were primarily co-regulated with genes involved in cell metabolism and proliferation, mitochondrial activity and stress responses. We conclude that M5 AML cells can upregulate IP genes in a cell-intrinsic manner in order to resist cell stress.

RUNX1 is an essential transcription factor for definite hematopoiesis and plays important roles in immune function. Mutations in RUNX1 occur in 5-13% of Acute Myeloid Leukemia (AML) patients (RUNX1mut ) and are associated with adverse outcome, highlighting the need for better genetic characterization of this AML subgroup and for the design of efficient therapeutic strategies for patients with RUNX1mut AML. Toward this goal, we performed RNA-sequencing of a cohort of RUNX1mut primary AML specimens and used a chemogenomic approach combining mutational and gene expression analysis of these specimens with assessment of their drug sensitivity profile through chemical screening. Sequencing data analysis demonstrated that samples with no remaining RUNX1 wild-type allele are clinically and genetically distinct and display a more homogeneous gene expression profile. Interestingly, these studies also unveiled the increased susceptibility of RUNX1-mutated specimens to Glucocorticoids (GCs) and revealed that RUNX1 allele dosage dictates sensitivity to these compounds in AML patient cells, unravelling a new role for RUNX1 in the Glucocorticoid Receptor (GR) pathway. GR is a nuclear receptor that modulates the expression of thousands of genes involved in several biological processes such as metabolism, immune function, skeletal growth, etc. GCs are commonly used to treat cancers of the lymphoid system, however their potential benefits for AML treatment have never been assessed formally and the mechanism of action of these drugs is not fully understood. Transcriptome analyses identified NR3C1 (encoding the GR) as one of the genes whose expression is determined by RUNX1 allele dosage, with increased expression in RUNX1mut specimens, indicating that RUNX1 inactivation could lead to GR upregulation, which might explain the increased sensitivity to GCs. We previously showed that RUNX1 silencing sensitizes human AML cell lines to GCs and that this acquired sensitivity is accompanied by the upregulation of the GR both at the transcript and protein levels. However, basal levels of GR could not explain GC sensitivity in all cases, indicating that other mechanisms are involved in the GC response. By performing co-immunoprecipitations (co-IP), we demonstrated that RUNX1 and GR physically interact in AML cells. Overexpression of FLAG-tagged RUNX1 mutants in HEK293 cells followed by co-IP identified the C-terminal inhibitory domain of RUNX1 as essential for the interaction with GR. Our results suggest that RUNX1 could be part of the GR transcriptional complex and could modulate the transcription of genes involved in the response to GCs. To identify regulators of the GC response, we are currently performing genome-wide CRISPR-Cas9-based genetic screens in AML cell lines. We generated RUNX1-deficient (GCs sensitive) AML cell lines that uniformly express Cas9 under a doxycycline-inducible promoter. These cell lines, along with parental cell lines (GC resistant) were used to conduct screens in GCs-supplemented media for the identification of genes that confer sensitivity or resistance to GCs. Mechanistic insights gained from these experiments will allow the design of additional therapeutic strategies to potentiate the effect of GCs on poor outcome AML. Disclosures No relevant conflicts of interest to declare.

Background: Acute myeloid leukemia (AML) with complex karyotype (CK) is associated with adverse prognosis with current therapies, in particular in the presence of TP53 mutations. Identification of novel therapeutic strategies is therefore urgently needed for this subgroup of patients. Methods and aims: As part of the Leucegene project, we RNA sequenced a cohort of 415 clinically annotated primary human AML specimens. Comparative transcriptomic analysis of the 68 CK AML specimens with the non-CK samples of the cohort was performed to identify features associated with the CK AML subgroup. Based on these results, we performed a chemical screen aimed at identifying active agents toward CK AML. Gene expression analysis, primary AML specimen culture and chemical screening were performed as previously described by our group (Lavallée et al., Nature Genetics, 2015; Pabst et al., Nature Methods, 2014; Simon et al., Clinical cancer research, 2017). Results: CK AML is a heterogeneous disease by definition and accordingly, Multidimensional Scaling (MDS) analysis fails to identify a distinct transcriptional signature. Nonetheless, comparative transcriptome analysis of CK and non-CK AML specimens revealed highly deregulated genes between these subgroups (Figure 1A). Among them, EDA2R, a known target to p53, is specifically downregulated in CK AML. This reflects the frequent TP53 alterations observed in CK AML. Most importantly, we found that the recently identified AML prognostic gene HMGA2 is overexpressed in the majority of these patients, independently of the TP53 status. HMGA2 is an oncofetal gene expressed in hematopoietic stem cell (HSC) -enriched populations while its expression decreases in progenitors and is nearly absent in mature cells. Interestingly, Gene Set Enrichment Analysis (GSEA) revealed an enrichment for HSC genes in AML specimens with high HMGA2 expression levels, which could reflect a stem-cell origin or a more immature state for these leukemias. Global transcriptomic analysis of CK AML also identified a G2/M checkpoint signature for HMGA2-high specimens (Figure 1B). Accordingly, we observed that chemical inhibitors of the G2/M regulators ATR, CHK1 and WEE1 are preferentially active on HMGA2-high CK specimens and that engineered overexpression of HMGA2 in different leukemia cell lines enhanced sensitivity to G2/M inhibition (Figure 1C). This specific inhibition was independent of the TP53 status. Conclusions: Here we report a comprehensive transcriptomic analysis of the CK AML specimens of the Leucegene cohort and identified aberrant expression of the HMGA2 oncogene in close to 80% of these samples. Chemical interrogation of primary AML specimens revealed a TP53-independent but HMGA2-mediated sensitization of CK AML to G2/M checkpoint inhibition. Thus, our results reveal a novel vulnerability for AML expressing high levels of HMGA2 and identify potential biological targets for these high-risk patients. Since several CHK1 and PLK1 inhibitors have been or are currently under evaluation in clinical trials, our findings suggest that HMGA2-high AML patients could benefit from this therapeutic approach. Figure 1: A/ Volcano plot representing the genes differentially expressed in CK versus non-CK AML. Samples with low group median expression (2; n=39) and HMGA2 null (RPKM=0; n=83) AML specimens. C/ Dose-response curves and IC50 values for representative compounds in K562 cells infected with control YFP vector (blue) or HMGA2-YFP expressing vector (red). Disclosures Sauvageau: ExCellThera: Employment, Equity Ownership.

Introduction: Acute myeloid leukemia (AML) evolution is a multistep process in which cells evolve from hematopoietic stem and progenitor cells (HSPCs) that acquire genetic anomalies, such as chromosomal rearrangements and mutations, which define distinct subgroups. Mutations in Nucleophosmin 1 (NPM1), which occur in ~30% patients, are the most frequent subgroup-defining mutations in AML and appear to be a late driver event in this disease. Bulk RNA-sequencing studies have identified differentially expressed genes between AML subgroups, but they are uninformative of the composition of cell types populating each sample. Large scale Single-cell RNA sequencing (scRNA-seq) technologies now enable a detailed characterization of intra tumoral heterogeneity, and could help to better understand the stepwise evolution from normal to malignant cells. Methods: Twelve primary human AML specimens from MSKCC and Quebec Leukemia Cell Bank, including 8 with NPM1 mutations, were included in this cohort. Cells were subjected to scRNA-seq using 10X Genomics Chromium Single Cell 3' protocols and libraries were sequenced on Illumina HiSeq or NovaSeq platforms. FASTQ files were processed using SEQC pipeline (Azizi E et al, Cell 2018), resulting in a carefully filtered count matrix of > 100,000 single cells (4877 to 11532 cells per sample). Results: Using euclidean distance metrics and t-Distributed Stochastic Neighbor Embedding (t-SNE) visualization, we explored the phenotypic overlap between samples and showed that leukemia cells from different patients were mostly dissimilar, suggesting inter-sample heterogeneity. However, samples with similar morphology and similar NPM1 mutational status were phenotypically closer (Fig A), as anticipated from bulk RNA-sequencing data (TCGA, NEJM 2013). We partitioned cells into distinct clusters using Phenograph (Levine J et al, Cell 2015) (Fig B) and measured the diversity of samples per cluster using Shannon's entropy metric, revealing that mature cell types (B/plasma cells, T/NK and erythroid cells, Fig C), presumably excluded from the tumor bulk, are transcriptionally similar across samples. Most notably, the next most diverse cluster (C36), comprising 438 cells from 11/12 samples, contains cells with a HSPC-like phenotype, as suggested by i) highest correlation of the centroid of this cluster with HSC1 (lin-/CD133+/CD34dim) population from sorted bulk RNA-sequencing data (Novershtern N et al, Cell 2011), and ii) marked GSEA enrichment for stem cell signatures (top enrichment: Jaatinen_hematopoeitic_stem_cell_up, NES = 9.04, FDR q-val = 0). To study the extent to which NPM1 or other mutations drive heterogeneity in leukemia populations, we interrogated 3'-derived single-cell sequences for all recurrent mutations in AML and found that NPM1 gene has unique features (e.g. relatively high single-cell expression and 3' localization) that allow specific identification of mutations in 5 to 34% of cells per mutated sample. To control for the high frequency of false negatives caused by dropouts in scRNA-seq data, we normalized the abundance of mutated vs wild-type cells to provide an estimation of mutation frequency in different cell types (Fig D). As expected, NPM1 mutations were rare in B and T/NK lymphoid cells (also observed using RT-qPCR in sorted populations by Dvorakova D et al, Leuk Lymphoma 2013) and were found in the majority of leukemia and myeloid cells. Interestingly, these mutations were detected at various frequencies in erythroid cells, suggesting that NPM1 mutations are acquired in cells with different lineage commitment in different patients. Most notably, the HSPC-like cluster C36 also contained a subpopulation of cells that have acquired NPM1 mutations and are transcriptionally different from wild-type cells. Conclusion: This study presents a first comprehensive single-cell map of primary AML, and the first 3'-based interrogation of mutations in single cells. It led to the identification phenotypically distinct cells presenting a HSPC-like expression profile which were sub-clonally harboring NPM1 mutations, providing the means to identify deregulated genes in these important leukemia subpopulations. Figure Figure. Disclosures Levine: Epizyme: Patents & Royalties; Celgene: Consultancy, Research Funding; Janssen: Consultancy, Honoraria; Isoplexis: Equity Ownership; C4 Therapeutics: Equity Ownership; Prelude: Research Funding; Gilead: Honoraria; Imago: Equity Ownership; Novartis: Consultancy; Roche: Consultancy, Research Funding; Loxo: Consultancy, Equity Ownership; Qiagen: Equity Ownership, Membership on an entity's Board of Directors or advisory committees.

Acute myeloid leukemia (AML) is a genetically heterogeneous hematologic malignancy, which is initiated and driven by a rare fraction of leukemia stem cells (LSCs). Despite the difficulties of identifying a common LSC phenotype, there is increasing evidence that high expression of stem cell gene signatures is associated with poor clinical outcome. Identification of functionally distinct subpopulations in this disease is therefore crucial to dissecting the molecular machinery underlying LSC self-renewal. Here, we combined next-generation sequencing technology with in vivo assessment of LSC frequencies and identified the adhesion G protein-coupled receptor 56 (GPR56) as a novel and stable marker for human LSCs for the majority of AML samples. High GPR56 expression was significantly associated with high-risk genetic subgroups and poor outcome. Analysis of GPR56 in combination with CD34 expression revealed engraftment potential of GPR56(+)cells in both the CD34(-)and CD34(+)fractions, thus defining a novel LSC compartment independent of the CD34(+)CD38(-)LSC phenotype.

Background:Acute promyelocytic leukemia (APL) is a favorable-risk subgroup of AML characterized by the t(15;17) translocation. The leading cause of early death in APL is uncontrolled bleeding mostly attributed to aberrant expression of tissue factor (F3) and annexin A2 (ANXA2) on leukemic promyelocytes leading to disseminated intravascular coagulation and hyperfibrinolysis, respectively. To prevent or treat such complications, early suspicion of APL and rapid initiation of therapy and supportive measures are critical. Podoplaninor PDPN is a surface glycoprotein expressed in most cell types (http://www.gtexportal.org), but not in blood cells. CLEC-2, the PDPN receptor, is expressed on normal platelets and was found to be necessary for the separation of blood and lymphatic vessels during embryogenesis. PDPN expression (whether endogenous or ectopic) in cell lines induces platelet aggregation, which can be inhibited by chemical tool compounds or by monoclonal antibodies (Chang et al, Oncotarget, 2015, Kato et al, Biochem. Biophys. Res. Commun., 2006). Aims and Methods:We analyzed the transcriptome of 30 APL comprised in the Leucegene 430 AML cohort, aiming to identify clinically useful markers and to better understand the hemostasis-related transcriptomic landscape of this subgroup. Patient cohorts and sorted normal hematopoietic cell populations (n=63) were previously reported (Lavallée et al, Nature Genetics, 2015 and Lavallée et al, Blood, 2016). Comparative analysis of gene expression and mutations were performed as previously described. Results:Our analytical pipeline identified several mutated genes in this cohort, most of which are non-specific and previously identified. CEBPE mutations were the only exception and were specific to APL specimens in this cohort (2/30 vs 0/400, p= 0.005). Most interestingly, we identified PDPN as the single most differentially overexpressed gene in APL (median 2.6 vs 0 RPKM, q = 7 x 10-29, Fig A-B). We also found that PDPN is not expressed in whole blood, bone marrow and in any sorted cell subpopulations from these normal tissues, including promyelocytes (median PDPN expression = 0 RPKM, range 0-0.018). This indicates that platelets are never exposed to PDPN in the adult vasculature and reveals that this gene is ectopically expressed in APL promyelocytes. Accordingly, our hypothesis is that aberrant PDPN expression on leukemic promyelocytes contributes to abnormal platelet aggregation in APL patients. We found that high PDPN expression is associated with lower platelet counts at presentation (18 vs 34 x 1012/L, median PDPN expression ≥ 10 vs < 10 RPKM, p = 0.016, Fig C). Furthermore, a strong inverse correlation was observed between the number of estimated circulating PDPN+ promyelocytes and platelet counts (leukocyte count x [% PDPN+ cells] > 1 x 109/L, Fig C, p=0.007). By incorporating anti-PDPN antibody (clone NC-08, Biolegend) in the EuroFlow protocol, we observed that PDPN expression test was 90% sensitive and 100% specific for APL (n= 48 and 50 APL and non-APL primary AML, respectively). Of note, 5 APL cases considered positive expressed low levels of PDPN. Interestingly, by comparing expression of all coagulation and fibrinolysis genes in APL (n=30) to that of non-APL specimens (n=400), we found that PDPN was the most discriminatory transcript (Fig. A). This result stands in sharp contrast with that found with F3 and ANXA2 which largely overlap in these APL versus non-APL human AML (Fig. A-B). This reinforces the hypothesis that aberrant PDPN expression may be a strong and unappreciated contributor to platelet consumption in APL. Conclusion:I) CEPBE is recurrently mutated in a small subset of APL patients. II) PDPN is the most specific aberrant transcript in this disease and is a new biomarker for APL. Systematic assessment of podoplanin by flow cytometry in newly diagnosed AML could lead to an earlier detection of unsuspected APL cases. III) PDPN expression by leukemic promyelocytes likely contributes to defective primary hemostasis, representing a new mechanism of APL-related bleeding. This provides a strong rationale for evaluating PDPN-CLEC-2 axis inhibitors in this setting. Figure Figure. Disclosures No relevant conflicts of interest to declare.

Background: Acute myeloid leukemias (AML) with complex karyotype (CK) are associated with an adverse patient prognosis with current therapies, in particular when they include TP53 mutations. Identification of novel therapeutic strategies is urgently needed for this subgroup of patients. Methods and aims:We have performed an RNA-Sequencing transcriptomic analysis of 68 CK AML included in the Leucegene cohort comprising 415 primary AML samples and performed correlative targeted chemical screening aiming at the identification of active agents in this subgroup. Cell culture and analysis of mutations, gene expression and chemical screening were performed as previously described (Pabst et al., Nature Methods, 2014; Lavallée et al, Nature Genetics, 2015). Results:Genes mutated at a frequency greater than 5% in the CK cohort were: TP53 (43/68, 63%), NRAS (9/68, 13%), DNMT3A (8/68, 12%), TET2 and NF1 (7/68 each, 10%), RUNX1 (6/68), FLT3, MLL and PTPN11 (5/68 each) and JAK2 and U2AF1 (4/68 each). Of those, only TP53 (p < 0.0001) and NF1 (p = 0.02) were preferentially associated with CK AML. HMGA2 (High Mobility Group AT-hook 2) was the most differentially expressed gene in CK AML compared to other AML (median: 1.35 vs 0.008 RPKM, q = 2 x 10-14). TP53-mutated samples were in addition characterized by low expression levels of Ectodysplasin A2 Receptor or EDA2R, a known target of TP53. We next tested in dose-response studies 267 compounds enriched in approved anti-cancer drugs on 27 primary CK AML samples and in 11 intermediate-risk karyotype controls. Four percent of these compounds were very active (median IC50 < 10 nM) in CK AML (red dots in Fig A). Surprisingly, this small subset of compounds comprised the only 2 inhibitors of Polo-Like Kinase 1 (PLK1), volasertib (median IC50: 6.6 nM) and rigosertib (median IC50: 8.5 nM). Rigosertib was, in addition, more active in CK AML than in intermediate-risk AML (p = 0.0035) supporting the hypothesis that choice of treatment, rather than genetic anomalies, determine prognosis (Fig. B-C). Most interestingly, expression levels of PLK1 correlated with the measured sensitivity to both rigosertib and volasertib raising the possibility that PLK1 expression determines sensitivity to these inhibitors. PLK1 is more highly expressed in CK AML (p=0.009) and in TP53-mutated AML (p = 0.015) than in control samples. A threshold of 5 RPKM was particularly predictive for drug sensitivity (rigosertib: p < 0.001, Fig C; volasertib: p = 0.037) suggesting that it could become a companion biomarker. Rigosertib and volasertib also potently inhibit PLK2 and PLK3 in cell-free assays, but no association was observed between PLK2/PLK3 expression and response to inhibitors, suggesting that cell lethality was mediated by PLK1 inhibition. PLK1 is a protein kinase involved in cell cycle. Accordingly, expression of this gene in our dataset defines a cluster of very highly correlated genes comprising mostly cell cycle components (e.g. CDC20, CCNB2, CENPA and CTSE1: > 0.90). Volasertib was recently studied in a phase 2 clinical trial in AML patients in combination with low-dose cytarabine (LDAC) (Döhner et al, Blood 2014). In this analysis, and in line with our data, responses were seen across all genetic groups including in adverse-risk AML (1/14 in LDAC vs 5/14 in LDAC + volasertib). Conclusion:Our chemo-transcriptomic analysis revealed that CK AML are characterized by high levels of HMGA2 expression and, in addition for the TP53 mutated subset, low levels of EDA2R. Most importantly our results show that CK AML with high expression levels of PLK1 are uniformly and preferentially sensitive to PLK1 inhibitors. Our data thus support the hypothesis that sensitivity to PLK1 inhibitors is associated to PLK1 expression, and that this gene may represent a promising biomarker to predict biologicalresponse to these agents. Our analysis provides for the first time a strong rationale for investigating PLK1 inhibitors in the context of CK and/or TP53 mutated AML in which correlative PLK1 expression analyses are commanded. Figure. Figure. Disclosures No relevant conflicts of interest to declare.

Acute myeloid leukemia (AML) is associated with poor clinical outcome and the development of more effective therapies is urgently needed. G protein-coupled receptors (GPCRs) represent attractive therapeutic targets, accounting for approximately 30% of all targets of marketed drugs. Using next-generation sequencing, we studied the expression of 772 GPCRs in 148 genetically diverse AML specimens, normal blood and bone marrow cell populations as well as cord blood-derived CD34-positive cells. Among these receptors, 30 are overexpressed and 19 are downregulated in AML samples compared with normal CD34-positive cells. Upregulated GPCRs are enriched in chemokine (CCR1, CXCR4, CCR2, CX3CR1, CCR7 and CCRL2), adhesion (CD97, EMR1, EMR2 and GPR114) and purine (including P2RY2 and P2RY13) receptor subfamilies. The downregulated receptors include adhesion GPCRs, such as LPHN1, GPR125, GPR56, CELSR3 and GPR126, protease-activated receptors (F2R and F2RL1) and the Frizzled family receptors SMO and FZD6. Interestingly, specific deregulation was observed in genetically distinct subgroups of AML, thereby identifying different potential therapeutic targets in these frequent AML subgroups.

Introduction. The mechanisms of haemoglobin (Hb) subtype switch remain largely unknown even though it has a therapeutic potential (Bauer et al., 2012). In fact, reactivation of fetal haemoglobin (HbF) production in adult erythrocytes (F-cells) is protective in many Hb disorders. Pregnancy may represent the only physiological condition where HbF production increases transiently (Ibrahim et al, 2009; Yamada et al., 2012). Moreover, flow cytometry (FC), by intracellular staining using anti-HbF antibodies, has improved the evaluation of maternal F-cells (Corcoran et al., 2014; Chen et al., 2000). In this study, the objectives are: 1) to document HbF expression prospectively peri- and post-partum and confirm its maternal origin; 2) to assess potential causes of HbF modulation throughout pregnancy (i.e. β-hCG, EPO) and; 3) to identify gene differences between HbF-positive and HbF-negative pregnant women. Methods. In this observational, single institution, prospective study, 879 pregnant women were screened. A total of 345 participants were included in the study: the first consecutive 176 women with negative total HbF expression ( Results. At the time of analysis, 93 HbF-positive and 147 HbF-negative women had completed follow-up. Results revealed that, among all pregnant women screened, 22.07% (n = 194) had positive HbF expression. Of this percentage, 90.72% (n = 176) filled the inclusion criteria. Of all women with positive HbF expression during 1 or more of the first 3 visits, 72.04% showed a post-partum decrease of HbF below 1%. Furthermore, FC analyses in HbF-positive women confirmed the maternal origin of HbF in all cases and revealed an average of 13.28% (sd = 5.38%) F-cells when total HbF was ≥ 1%. Only modest correlations could be found so far between HbF levels and β-hCG (ρ = 0.25) or EPO (ρ = 0.07). Comparative DNA analyses of HbF-positive and HbF-negative women are currently underway and will be presented at the meeting. Conclusions. To our knowledge, this represents the largest prospective study evaluating HbF expression throughout pregnancy, as well as the first to evaluate post-partum HbF dynamics and the genetic background of HbF-negative and HbF-positive pregnant women. Preliminary data shows that: 1) a significant percentage of pregnant women have a transient elevation of HbF which cannot be explained by fetomaternal hemorrhage nor hereditary persistence of HbF; 2) the proportion of F-cells in women remains constant peri- and post-partum when HbF ≥ 1%; 3) a large proportion of maternal erythrocytes express HbF at low levels and; 4) HbF modulation during pregnancy is only modestly correlated to β-hCG and EPO. These data may lead to the study of a new paradigm of Hb subtype switch as well as HbF-induction therapeutics for the benefit of Hb disorders. Disclosures No relevant conflicts of interest to declare.

Insights into the complex clonal architecture of acute myeloid leukemia (AML) unravelled by deep sequencing technologies have challenged the concept of AML as a hierarchically organised disease initiated and driven by rare self-renewing leukemic stem cells (LSCs). In contrast to normal human hematopoietic stem cells (HSCs), which are highly enriched in the CD34+ CD38- population, LSCs have also been found in the CD34- and the CD38+ fractions questioning the existence of a consistent LSC surface marker profile for AML. Besides, low LSC frequencies in primary samples, rapid onset of differentiation upon ex vivo culture, and genetic inter-specimen heterogeneity hamper the dissection of the molecular machinery that drives LSC self-renewal. We performed RNA-Sequencing of primary human AML samples and assessed LSC frequencies by limiting dilution analyses for 56 of these in NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) mice. By comparing gene expression profiles between high vs low LSC frequency leukemias, we identified the G-protein coupled receptor 56 (GPR56) has significantly more expressed in high LSC frequency leukemias. We validated the RNA-seq data with protein expression by FACS and found an excellent correlation. To determine whether GPR56 positive cells overlapped with the known LSC-associated phenotype CD34+ CD38-, we stained 45 AML samples with CD34, CD38, GPR56, and antibodies against other described LSC markers. Although CD34+ GPR56+ and CD34+ CD38- compartments identified the same population in some samples, we found in the majority of samples that GPR56 further subdivided the CD34+ CD38- compartment. Accordingly, not only the proportions of total GPR56+ and CD34+ GPR56+ cells were significantly higher in LSChigh versus LSClow samples, but also the proportion of GPR56+ cells within the CD34+ CD38- compartment was significantly different between the groups indicating that GPR56 might be of additional value to what is currently considered the best described LSC phenotype. The percentage of total CD34 positive cells did not correlate with LSC frequency clearly distinguishing GPR56 from CD34 or CD38, which are only suitable LSC markers when used in combination. We analysed other potential LSC markers (TIM3, CD96, CD44, CD123, CLL1 and CD47) in our RNA-Seq dataset and by FACS analysis in combination with CD34 as we did for GPR56 and none of them correlated with LSC frequency in our sample collection. To determine whether GPR56 discriminates engrafting LSCs from non-LSCs, we sorted GPR56+ and GPR56- cells within the CD34-positive and -negative compartments from selected specimens with known engraftment potential. We found that GPR56 identified the engrafting fraction in CD34positive AML samples, with a >50 fold enrichment in LSC in the CD34+GRP56+ fraction vs the CD34+GPR56- fraction within the same sample, demonstrating that GPR56 is a good LSC marker. Specimens with high molecular or cytogenetic risk such as chromosome 5 or 7 anomalies and EVI1- rearrangementexpressed high levels of both, GPR56 and CD34, while samples with coexistent FLT3 -ITD, DNMT3A, and NPM1 mutations displayed a unique CD34low GPR56high profile. Moreover, we found a divergent distribution of variant allele frequencies in GPR56+ versus GPR56- fractions identifying GPR56 as a discriminator of leukemic sub-clones with high and low NSG engrafting capacity. Analysis of engrafted cells re-sorted based on GPR56 after being harvested from mouse bone marrow revealed reduced complexity of the clonal composition. Most importantly, GPR56 positive cells differentiated to GPR56 negative cells in mice, which did not happen in the human niche, in which GPR56 positive and negative fractions represented two independently evolved subclones. In summary our work identifies GPR56 as a novel LSC marker in AML and also shows that GPR56 readily identifies a functionally distinct LSC-rich subclone in the majority of human AML patients and reveals hitherto unforeseen complexity in the interaction between human LSCs and the NSG mouse environment. Disclosures No relevant conflicts of interest to declare.

Acute myeloid leukemia (AML) is associated with poor overall survival and the development of more effective therapies is urgently needed. G Protein-Coupled Receptors (GPCRs) represent the largest family of membrane receptors, with an estimated 800 members in humans, and are attractive therapeutic targets, accounting for approximately 30% of targets of marketed drugs. These receptors are key transducers that bind a vast diversity of ligands (e.g. glycoproteins, peptides, amino acids, nucleotides, nucleosides, ions) allowing the cells to adapt to their environment by regulating a wide variety of physiological processes including the control of blood pressure, heart rate, digestive processes, hormone secretion, cell growth and migration, as well as vision and olfaction. Binding to their ligands leads to conformational rearrangements promoting the engagement and modulation of many distinct downstream signaling effectors that are both G protein-dependent and independent. Several GPCRs are critical for cell proliferation and survival, and can be aberrantly expressed in cancer cells. For example, the chemokine receptor CXCR4 plays an important role in metastasis and angiogenesis in breast cancer and many other types of tumors. In AML, CXCR4 overexpression has been associated with poor outcome. Moreover, in vivo mouse studies have shown that the use of a small molecule antagonist of CXCR4 increases the mobilization of AML cells into the peripheral blood and improves the apoptotic effects of chemotherapy. This activity has been explored in a phase I/II clinical study showing that the addition of CXCR4 antagonists to chemotherapy in AML might improve the remission rate. The role of GPCRs in mouse leukemogenesis has also been suggested in a transcriptome analysis of two related leukemia clones that differ in their stem cell frequency. In this study, GPCRs were the most differentially expressed class of genes between the two clones. Currently, an extensive assessment of GPCR expression in human AML is lacking. To address this issue, we studied the expression of GPCRs in a large cohort of AML samples, as well as in normal blood cells, bone marrow cell populations, and cord blood-derived CD34+ cells as controls.The 772 GPCRs analyzed in this study consist of all the GPCR members included in the International Union of Basic and Clinical Pharmacology (IUPHAR) database, as well as 370 olfactory, 24 taste and 4 vomeronasal receptors. RNA sequencing data analysis was performed as previously described (Lavallée et al, Blood 2015 Jan 1;125(1):140-3) and revealed that 240 GPCRs are expressed in cells from this AML cohort. Among these receptors, 30 are upregulated and 19 are downregulated in AML samples compared to CD34+ normal cells. Upregulated GPCRs are enriched in chemokine (CCR1, CXCR4, CCR2, CX3CR1, CCR7, and CCRL2), adhesion (CD97, EMR1, EMR2, and GPR114) and purine (including P2RY2 and P2RY13) receptor sub-families. The downregulated members include adhesion GPCRs such as LPHN1, GPR125, GPR56, CELSR3, and GPR126, protease-activated receptors (F2R and F2RL1), and the Frizzled family receptors SMO and FZD6. Interestingly, specific deregulation was observed in genetically distinct subgroups of AML, a subset of GPCRs being differentially expressed in normal karyotype AML with NPM1 or FLT3 -ITD mutations, and in specimens with Core Binding Factor and MLL rearrangements, thereby representing promising therapeutic targets or diagnostic markers. In conclusion, our results demonstrate that several GPCR members are deregulated in AML with a clear enrichment in distinct classes, providing the rationale for functional assays using available agonists or antagonists to leukemia-enriched GPCRs. Since these receptors are the targets of several US Federal and Drug Administration approved drugs, our results pave the way to explore selected GPCRs as novel AML therapeutic targets. Disclosures Bouvier: American Society of Nephrology: Other: speaker; Domain Therapeutics: Other: Company SAB member, Research Funding; Vertex Pharmaceutical: Research Funding; Ontario Genomic Institute: Other: SAB meeting; BMS: Research Funding; DalCor Pharmaceutics: Other: Company SAB member; Pfizer: Other: Speaker, Research Funding; Novo-Nordisc: Research Funding.

Background Acute myeloid leukemias (AML) with t(8;21)(q22;q22);RUNX1-RUNX1T1 and inv(16)(p13.1q22) or t(16;16)(p13.1;q22);CBFB-MYH11 are recurrent genetic entities commonly designated as core binding factor (CBF) AML. Both subgroups have distinct gene expression signatures and are characterized by recurrent mutations in KIT, FLT3, and RAS pathway genes. More recently, ASXL1 and ASXL2 mutations have been identified in t(8;21) AML. The TCGA study has analyzed a limited number of AML with CBF rearrangements, but to date more comprehensive CBF AML cohorts have not been unbiasedly explored by next-generation sequencing. Therefore, we have performed RNA sequencing of 415 genetically diverse AML specimens, including 48 CBF AML samples. In this analysis, we compared the mutational profile and transcriptomic landscape of both CBF subgroups to that of non-CBF AML. Methods Analysis of mutations and gene expression was performed as previously described (Lavallée et al, Nature Genetics, doi:10.1038/ng.3371). Mutations in all genes that are recurrently mutated in hematological malignancies are reported. In order to identify novel acquired recurrent mutations in CBF subgroups, genes with variants in ≥ 3 samples were systematically confirmed by Sanger sequencing of non-tumoral DNA. Results Genes mutated in the t(8;21) cohort are: KIT (8/20, 40%), FLT3 and ASXL2 (4/20 each, 20%), ASXL1, NRAS, ZBTB7A, TET2, SMC1A (3/20 each, 15%), DNMT3A (2/20), and JAK2, SMC3, STAG2, WT1, and CSF3R (1/20 each). Mutations in inv(16) AML are found in the following genes: KIT (14/28, 50%), NRAS (12/28, 43%), FLT3 (8/28, 29%), PRRC2B (3/28, 11%), KRAS (2/28, 7%) and BCORL1, DNMT3A, GATA2 and NF1 (1/28). The most frequent mutations were found in activated signaling genes (KIT, NRAS, KRAS, FLT3, JAK2, CSF3R), identified in 14/20 (70%) and 25/28 (89%) of t(8;21) and inv(16) AML samples respectively. 38% of mutated samples contained 2 to 5 such mutations, and the sum of their variant allele frequencies never exceeded ~50%, suggesting that each mutation occurs in a different subclone. This result supports the hypothesis that these mutations and CBFfusion genes are strong collaborators in AML. Several novel observations emerged from these analyses. First, we identified 2 frameshift and 1 missense novel acquired mutations in ZBTB7A, which are specific to the t(8;21) subgroup (3/20 vs 1/395, p = 0.0004). ZBTB7A encodes a transcription factor of the POK/ZBTB family and other genes encoding this family of transcription factors, such as BCL6 and PLZF, are rearranged in hematological malignancies. Second, we established that ASXL2 mutations are very rare (2/395) in t(8;21) negative samples and thus specifically associated to RUNX1 -RUNX1T1 fusions (p < 0.0001). Third, mutually exclusive mutations in cohesin complex genes (SMC1A, SMC3 and STAG2) are frequent in t(8;21) AML (5/20, 25%). Lastly, a novel acquired PRRC2B A1506S missense mutation was identified in 3 inv(16) AML samples. PRRC2B, a gene with poorly described functions, was the only non-activated signaling gene recurrently mutated in this subgroup. Using the most significantly and differentially expressed genes, we identified signatures of 145 and 127 genes specific to t(8;21) and inv(16) groups, respectively. 78% and 81% of these genes have not been previously described in gene set enrichment analyses of CBF AML, and are potential novel CBF diagnostic markers or genes that are functionally related to CBF fusions. Using gene signatures and principal component analyses (PCA), CBF subgroups homogeneously clustered together with one sole exception: a sample harboring a t(16;21);RUNX1-CBFA2T3 unambiguouslygrouped with t(8;21) specimens. The rare but recurrent RUNX1-CBFA2T3 chimeric proteins are known to share similar structural characteristics with RUNX1-RUNX1T1, and our observations now unify the transcriptomic networks of these 2 genetic entities. We also identified and characterized 8 additional RUNX1 fusions in our AML cohort, including 6 novel fusions, which share a different transcriptomic profile compared to RUNX1-RUNX1T1 positive samples, thereby suggesting that they might have distinct functional consequences. Conclusion Our comprehensive RNA sequencing analysis substantially contributes to a better understanding of mutations and gene expression profiles in CBF AML, and reports a unity between RUNX1 -RUNX1T1 and RUNX1-CBFA2T3 genetic networks. Disclosures No relevant conflicts of interest to declare.

Background Recent development in sequencing technologies with deep coverage for mutation analysis has enabled the identification of clonal architecture in some cancers. RAS mutations are observed in a large proportion of MLL leukemias. Our hypothesis is that determination of RAS mutation status in MLL leukemias should provide insights into the clonal make up of this disease and clues about the nature of clones that overcome therapy. Methods We combined exome and transcriptome sequencing in 32 adult MLL leukemias and results were compared to our cohort of 48 normal karyotype (NK) AML. Exome capture and paired-end sequencing (2 x 100bp, Illumina HiSeq 2000) were performed using TruSeq (Illumina) protocols. Mean coverage was 165X for transcriptome and 42X for exome. Initial analysis was focused on 25 known AML-associated genes and excluded all other novel mutations. Average transcriptome and exome coverage for N/KRAS alleles was 287X (25-846) and 42X (9-151), respectively. Clones were defined based on the identification of N/KRAS mutations in at least 1% of the reads. Results Figure 1 shows mutation status, MLL partners and FAB classification for each MLL leukemia. No mutations were observed in NPM1, FLT3 (ITD), CEBPA (biallelic), RUNX1, DNMT3A, IDH1, KIT, BCOR, SF3B1, U2AF1 or RAD21. On average, 1 mutated gene (range: 0-4) per sample was found compared to 3 (range 0-5) in NK-AML (p < 0.0001). We observed that 13/32 MLL leukemias (which include 2 paired samples) harbored N/KRAS mutations. There were no association between RAS mutation status and MLL partner, FAB classification, age, white blood cell count and overall survival. RAS mutations were found in 15% of NK-AML which contained on average 2.3 additional mutations in leukemia-associated genes compared to only 0.3 (p Excluding 2 paired relapse specimens, a total of 24 N/KRAS mutated clones were identified in 11 of the 30 MLL leukemias. The first sample included 5 clones each containing different NRAS mutations (e.g. G13R, G13D, etc.) contributing to 17, 9, 4, 2 and 2 % of the reads. Since RAS mutations are mostly heterozygous, we estimated that the contribution of each clones varied between 34 (i.e. 17% x 2) to 4%. A similar analysis revealed 4 clones in another specimen, contributing to 22, 12, 12 and 4 % of the cells. In 4 additional samples, the proportions of N/KRAS mutated clones were 1) 42, 38 and 8% 2) 92, 4 and 2 %, 3) 78 and 12% and 4) 52 and 32%, establishing that 20% (6/30) of these MLL leukemias were oligo- to polyclonal. In comparison, our NK-AML cohort of 48 patients included 7 specimens mutated for N/KRAS in which a total of 10 different clones were identified for an average of 0.2 RAS mutated clones per NK-AML versus 0.8 in MLL leukemias (p=0.007). This result further strengthens the hypothesis that RAS and MLL-fusion genes are strong collaborators in human AML. Grossmann et al recently showed that RAS mutated clones can be lost at relapse (Leukemia, 2013), possibly suggesting that other genes are at play in collaborating with MLL-fusions and causing drug resistance. To identify such genes, we further analyzed paired diagnosis and relapse samples in 2 patients. In the first patient, the KRAS mutation that was found in 66% of the cells at diagnosis was identified in all cells at relapse. In the second patient, while KRAS G12V and G12D mutations were found in 78% and 12 % of the cells at diagnosis, only the G12V clone was detected in 100% of the cells at relapse indicating in vivo clonal selection in both cases. We then performed a comparative analysis of mutated/wild type allele ratios for other coding genes. This analysis enabled us to identify a subset of mutations in candidate genes that are present at relapse in the dominant clone but that were undetectable or at lower frequency at presentation, indicating they might be specifically involved into occurrence of relapse (i.e. drug resistance). Conclusion NRAS and KRAS are mutated in 37% of MLL leukemias in this cohort. In contrast to NK-AML, these leukemias are frequently oligo- to polyclonal and contain few additional mutations suggesting that RAS activation may be sufficient to induce AML in the presence of MLL fusions. Evidence from our limited number of relapse patients, and that of others, suggests that RAS does not confer drug resistance which could be explained by novel mutations in genes that were specifically detected in the dominant clones at relapse. Disclosures: No relevant conflicts of interest to declare.

Abstract 3823 Background: Somatic mutations of splicing factor 3B, subunit 1 (SF3B1) are prevalent in myelodysplastic syndromes (MDS) with ring sideroblasts (RS). Little is known about the clinical phenotype associated with these mutations. Initial studies have reported an excellent event-free survival (EFS) in patients with SF3B1 mutations but subsequent reports did not confirm that observation. Therefore, its clinical significance remains uncertain. Our objective was to determine if the SF3B1 mutational status predicts distinctive morphological, cytogenetic or clinical characteristics in patients with MDS with RS. Methods and patients: Twenty-seven patients diagnosed with MDS and ≥ 15% RS in bone marrow aspiration between 2000 and 2011 at our institution and other hospitals affiliated to the Quebec Leukemia Cell Bank (BCLQ) and with available material for genetic analyses were included in this study. Morphology of samples was revised. After DNA extraction, polymerase chain reaction was performed to amplify SF3B1 mutational hotspots in exons 13 to 15. Amplicons were sequenced by the Sanger method. Clinical and laboratory data were collected retrospectively from hospital medical records and BCLQ database. Results: Point mutations of SF3B1 were found in 48% (13/27) with the following distribution: refractory anemia with RS (RARS): 7/10, refractory cytopenia with multilineage dysplasia (RCMD): 5/10 and refractory anemia with excess blasts (RAEB): 1/7. All mutations were heterozygous nonsense mutations and K700E exon 15 mutations accounted for 85% (11/13) of mutations. The remaining mutations were located in exon 14 at position 666. In the entire cohort, SF3B1 mutations were associated with an older median age at diagnosis (69 vs 60 years, p= 0.007). They were associated with favorable cytogenetics whereas unmutated SF3B1 was associated with unfavorable cytogenetics (8 cases vs 0 with favorable cytogenetics and 0 cases vs 11 with unfavorable cytogenetics for mutated and wild type SF3B1 respectively, p < 0.0001). Median overall survival (OS) was similar in the 2 groups (not reached vs 27 months, p=0.7) but median EFS was higher in patients with mutated SF3B1 (not reached vs 5 months, p=0.02). To evaluate the influence of higher bone marrow blasts on outcomes, analyses were restricted to the RARS and RCMD-RS categories. In that subgroup, SF3B1 mutations were correlated with higher median platelet counts at diagnosis (201 vs 70 × 109/L, p=0,04). In addition, SF3B1 mutations remained associated with favorable cytogenetics, and unmutated SF3B1 with unfavorable cytogenetics (p=0.004). Moreover, SF3B1 mutations were more frequently detected within favorable or intermediate-1 categories of the International Prognostic Scoring System (IPSS) compared to wild type SF3B1 (p=0.04). However, in patients with < 5% bone marrow blasts, EFS was similar between mutated and unmutated SF3B1 (not reached vs 46 months, p =0.4). Conclusion: In this series, SF3B1 mutations are frequent in MDS with RS and they are strongly associated with favorable cytogenetics. They are mostly detected among low-risk groups of the World Health Organisation classification (RARS and RCMD-RS), but they are not independently associated with an increased overall or event-free survival. Disclosures: No relevant conflicts of interest to declare.

Solid tumors at the advanced stages are often characterized by the overexpression of tyrosine kinase receptors that stimulate growth through the MAP kinase pathway and activate antiapoptotic signaling through the PI3K kinase pathway. One such receptor is the epidermal growth factor receptor (EGFR) that is overexpressed in many tumors including brain, lung, breast, ovarian and prostate carcinomas. EGFR can synergize with other tyrosine kinases such as Src to promote invasion and metastasis. However, despite the implication of EGFR in multiple processes, inhibitors of its tyrosine kinase activity such as ZD1839 or erlotinib showed moderate antiproliferative activity against certain tumor types in the clinic. It has been reported that EGFR expressing cells that harbor dysfunctional PTEN, are moderately sensitive to EGFR inhibitors. Given the complexity of cell response the latter class of inhibitors, we thought it of interest to investigate agents that are not only directed at EGFR but also to divergent targets such as Src or DNA, with the purpose of producing single compounds with greater potency than their single inhibitor counterpart. These molecules termed “combi-molecules” were designed to remain small enough for their quinazoline moiety to be able to bind to the ATP site of EGFR and to block a divergent target such as Src or DNA. Using molecular modeling, a structure-based drug design program was used to identify a linker that could be placed between the quinazoline moiety required for binding in the ATP site and the appendage directed at the divergent target. A solvent exposed ASP residue near the opening of the EGFR ATP binding pocket was used to optimize interaction with ionizable linkers. This interaction was found to be tolerant of bulky substituents and was exploited to append other pharmacophores to the combi-molecules. The results showed that in the category of mixed EGFR-DNA targeting molecules, the EGFR inhibitory potency of EGFR was in the low micromolar to nM range and the compounds also retained significant DNA damaging potential. In the category of EGFR-Src targeting molecules, we identified SB163 that contained a quinazoline moiety, the ionizable spacer and an analogue of PP2. SB163 showed significant antiproliferative and antimetastatic property in a Boyden Chamber assay and its activity was superior to that of a combination of known Src inhibitor PP2 + ZD1839, a clinical inhibitor of EGFR. Furthermore, sub-cellular distribution studies, using a fluorescent probe containing the optimized ionizable spacer showed that their biodistribution is unique with preferential localization in the perinuclear region. The results in toto suggests that in the design of promiscuous molecules, an ionizable basic arm linked to the aminoquinazoline moiety preserves EGFR inhibitory potency and allow freedom to append other moiety directed at the cross- target. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):A142.

Introduction: α-Thalassemia results from an heterogeneous group of disorders involving the α-Globin genes. Common deletions are responsible for 95% of cases of α-thalassemia, whereas Constant Spring mutation (Hb CS), one of the known mutations involving the termination codon of α2-globin gene, is the most common non-deletional cause of α-thalassemia worldwide. Its prevalence has never been studied in occidental countries. Patients and methods: as part of our neonatal hemoglobinopathy screening program, fetal cord blood samples are systematically tested by HPLC. When an abnormal Hb or HbH is found, α-Globin deletion is carried by multiplex-PCR. A two part observational study was done at our center based on the available data from our hemoglobinopathy screening program available since 2002. Firstly, we aimed to determine the level of Hb H present and to correlate it to a specific type of α-Globin gene deletion by m-PCR sequencing (table 1). 306 samples were analyzed and demonstrated a statistically significant difference between categories of α-globin gene deletions: 2 del cis vs 1 del, 2 del trans vs 1 del, 2 del cis vs 2 del trans (difference of means of 8.07, 4.33 and 3.75 respectively, P Table 1 Deletion/allelic distribution Hb H level (range) Hb H level (means) Std Dev 1 deletion 1.0 – 12.4 2.03 1.12 2 deletions cis - 10.1 2.16 2 deletions trans - 6.35 1.5 Our second goal was to establish the prevalence of α-Globin termination codon mutation in the fetal blood sampling containing ≥1% of Hb H in absence of α-Globin gene deletion by m-PCR. 395 fetal blood samples were analysed. Of these, 59 lacked α-Globin gene deletions. 17 were excluded from our analysis for lack of personal data or DNA. The 42 remaining samples were studied by amplification of the termination codon followed by an enzymatic digestion for the recognition site TTAA. 2 samples (5%) were found to be heterozygotes for α2-globin mutation, presumably CS mutation. The small number of positive cases does not permit the conduct of a powerful statistical analysis, however the two mutated samples had a high level of Hb H (2,9%) whereas only 21% of the 42 studied samples had an Hb H ≥2%. Conclusion: These results represent a step forward in diagnosis of α-thalassemia and in identification of patients requiring particular genetic counselling based on HbH level at birth in cord blood. These results also indicate that 5% of Hb H positive/α-Globin deletion negative samples were caused by termination codon mutation in an occidental urban based population, and that it could be of interest to search for these mutations, especially when the Hb H level is high.