Leukemic Stem Cells Frequency Research Articles

WT1 facilitates the self-renewal of leukemia-initiating cells through the upregulation of BCL2L2: WT1-BCL2L2 axis as a new acute myeloid leukemia therapy target

BackgroundOverexpression of Wilms’ tumor-1 (WT1) transcription factor facilitates proliferation in acute myeloid leukemia (AML). However, whether WT1 is enriched in the leukemia-initiating cells (LICs) and leukemia stem cells (LSCs) and facilitates the self-renewal of LSCs remains poorly understood.MethodsMLL-AF9-induced murine leukemia model was used to evaluate the effect of knockdown of wt1 on the self-renewal ability of LSC. RNA sequencing was performed on WT1-overexpressing cells to select WT1 targets. Apoptosis and colony formation assays were used to assess the anti-leukemic potential of a deubiquitinase inhibitor WP1130. Furthermore, NOD/SCID-IL2Rγ (NSG) AML xenotransplantation and MLL-AF9-induced murine leukemia models were used to evaluate the anti-leukemogenic potential of WP1130 in vivo.ResultsWe found that wt1 is highly expressed in LICs and LSCs and facilitates the maintenance of leukemia in a murine MLL-AF9-induced model of AML. WT1 enhanced the self-renewal of LSC by increasing the expression of BCL2L2, a member of B cell lymphoma 2 (BCL2) family, by direct binding to its promoter region. Loss of WT1 impaired self-renewal ability in LSC and delayed the progression of leukemia. WP1130 was found to modify the WT1-BCL2L2 axis, and WP1130-induced anti-leukemic activity was mediated by ubiquitin proteasome-mediated destruction of WT1 protein. WP1130 induced apoptosis and decreased colony formation abilities of leukemia cells and prolonged the overall survival in the THP1-based xenograft NSG mouse model. WP1130 also decreased the frequency of LSC and prolonged the overall survival in MLL-AF9-induced murine leukemia model. Mechanistically, WP1130 induced the degradation of WT1 by positively affecting the ubiquitination of WT1 protein.ConclusionsOur results indicate that WT1 is required for the development of AML. WP1130 exhibits anti-leukemic activity by inhibiting the WT1-BCL2L2 axis, which may represent a new acute myeloid leukemia therapy target.

A small molecule compound IX inhibits telomere and attenuates oncogenesis of drug-resistant leukemia cells.

Drug resistance is a common obstacle in leukemia treatment and failing to eradicate leukemia stem cells is the main cause of leukemia relapse. Previous studies have demonstrated that telomerase activity is associated with deregulated self-renewal of leukemia stem cells (LSCs). Here, we identified a novel compound IX, an imatinib derivative with a replacement fragment of a telomerase inhibitor, which can effectively eradicate LSCs but had no influence on normal hematopoietic stem cells (HSCs) survival. We showed that compound IX can decrease the viability of drug-resistant K562/G cells and blast crisis CML primary patient cells. Besides, IX can affect LSC survival, inhibit the colony-forming ability, and reduce LSC frequency. In vivo results showed that IX can relieve the tumor burden in patient-derived xenograft (PDX) model and prolong the lifespan. We observed that compound IX can not only decrease telomerase activity, but also affect the alternative lengthening of telomeres. In addition, IX can inhibit both the canonical and non-canonical Wnt pathways. Our data suggested this novel compound IX as a promising candidate for drug-resistant leukemia therapy.

CD34+CD38-CD123+ Leukemic Stem Cell Frequency Predicts Outcome in Older Acute Myeloid Leukemia Patients Treated by Intensive Chemotherapy but Not Hypomethylating Agents.

The prognostic impact of immunophenotypic CD34+CD38−CD123+ leukemic stem cell (iLSC) frequency at diagnosis has been demonstrated in younger patients treated by intensive chemotherapy, however, this is less clear in older patients. Furthermore, the impact of iLSC in patients treated by hypomethylating agents is unknown. In this single-center study, we prospectively assessed the CD34+CD38−CD123+ iLSC frequency at diagnosis in acute myeloid leukemia (AML) patients aged 60 years or older. In a cohort of 444 patients, the median percentage of iLSC at diagnosis was 4.3%. Significant differences were found between treatment groups with a lower median in the intensive chemotherapy group (0.6%) compared to hypomethylating agents (8.0%) or supportive care (11.1%) (p <0.0001). In the intensive chemotherapy group, the median overall survival was 34.5 months in patients with iLSC ≤0.10% and 14.6 months in patients with >0.10% (p = 0.031). In the multivariate analyses of this group, iLSC frequency was significantly and independently associated with the incidence of relapse, event-free, relapse-free, and overall survival. However, iLSC frequency had no prognostic impact on patients treated by hypomethylating agents. Thus, the iLSC frequency at diagnosis is an independent prognostic factor in older acute myeloid patients treated by intensive chemotherapy but not hypomethylating agents.

Leukemic Stem Cells Persistence Measured By Multiparametric Flow Cytometry Is a Biomarker of Poor Prognosis in Adult Patients with Acute Myeloid Leukemia

Introduction: Despite the recent advances in chemotherapy regimens, relapse still substantially affects prognosis of intensively treated adult acute myeloid leukemia (AML) patients. There is growing evidence that a residual populations of leukemic cells may survive chemotherapy and outgrow, eventually causing relapse. These chemo-resistant cells are particularly abundant in the fraction of leukemic stem cells (LSC), which are endowed with pronounced self-renewal properties allowing to initiate and maintain leukemic clone. These cells can be detected, by high sensitivity multiparametric flow-cytometry (MFC), in the CD34+/CD38- fraction of the leukemic populations and can be distinguished from normal hematopoietic stem cells by the expression of specific markers. In recent clinical trials, LSC have been demonstrated to represent a biomarker of poor prognosis when detected at diagnosis but also during treatment course. Moreover, the combined estimate of measurable residual disease (MRD) and LSC refines the prognostic assessment as determined by the sole application of MRD detection. Aim: We analyzed a series of patients (pts) treated in the context of GIMEMA trials, in whom the LSC frequency was assessed by MFC at diagnosis. Pts with measurable levels of LSC were tested again after the consolidation cycle. At the same timepoint "standard" MRD was also determined. The purpose of the study was to demonstrate a correlation between LSC burden at baseline and prognosis in terms of overall (OS) and disease-free survival (DFS). Furthermore, we wanted to investigate the relationship between LSC and "standard" MRD persistence (&gt;0.035%) after consolidation, and possible correlation with outcome. Methods: LSC were evaluated by MFC as described elsewhere (Terwijn, PLoS 2014). LSC were quantified exploiting the expression of the C-type lectin-like molecule-1 (CLL1) and applying a sequential gating strategy that contained the CD34+/CD38- population. Pts were defined as LSC negative (LSCneg) in case of zero LSC count, LSClow or LSChigh when LSC were &gt;0&lt;0,03% or &gt;0.03%, respectively. After consolidation, any level &gt;0 was considered as a LSC persistence. Methods of analysis and thresholds were set according to previous publications (Zeijlemaker, Leukemia 2019). Results: We analyzed 130 pts with de novo AML, in whom LSC determination was available at the baseline. Fifty-nine (45,4%) pts were LSCneg, 49 (37,7%) LSClow, 22 (16,9%) LSChigh. We did not observe any correlation between baseline LSC level and genetic/cytogenetic risk at diagnosis. There was not a significant difference in terms of OS duration according to the 3 LSC levels, however, pts who were LSChigh had the shortest OS (36-month estimate OS of 71.5% vs. 65.4 % vs 52.4 % for the LSCneg, LSClow and LSChigh categories respectively; p=0.21). A statistically significant difference, regardless of the belonging to the LSClow or LSChigh category was observed when we focus on the subgroup of 30 pts with intermediate-risk AML, with a 36-month estimate OS of 76% vs. 77.8% vs 25% for the LSCneg, LSClowand LSChigh categories respectively (p=0.023) (Figure 1A). In 19 patients, LSC persistence was assessed at the post-consolidation time-point. Nine LSChigh pts who failed to eradicate residual LSC at this timepoint had a worse outcome as compared to those belonging to the same category but achieving a LSC clearance or those who were LSClow (36-month OS of 62.5% vs. 59.2% vs. 66.7% vs. 25% for the LSClow converted into LSCneg, LSClow not converted into LSCneg, LSChigh converted into LSCneg and LSChigh not converted into LSCneg categories, respectively; P=0.062) (Figure 1B). In 27 pts LSC and "standard" MRD determination was available. LSC persistence determined a worse 3-years OS both in MRD negative (66.7% vs 85.7%, p=0.44) and MRD positive pts (&lt;20% vs 75.0%, p=0.041). Conclusions: In line with the experience of other European groups, we demonstrated that MFC monitoring of LSC is feasible and provides prognostic information when performed at diagnosis and during treatment course. MFC assessment of LSC also offers the opportunity to monitor pts who lack aberrant phenotypes suitable for "standard" MRD investigation. When the 2 approaches - standard "MRD" and LSC assessment - are combined together, the prognosis prediction of AML can be further refined. Finally, LSC assessment can potentially represent an effective tool to monitor the effect of LSC targeting agents. Disclosures Buccisano: Astellas: Membership on an entity's Board of Directors or advisory committees; Janssen: Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Venditti:Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees; Daiichi-Sankyo: Consultancy, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees; Astellas: Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy.

Both Arid1b and Arid2 Are Tumor Suppressors in MLL-AF9 Leukemogenesis

ATP-dependent chromatin remodeling complexes, including BAF (BRG1 associated factor) and PBAF (Polybromo-associated BRG1 associated factor), play major roles in development and disease. These complexes contain multiple subunits and utilize ATP to remodel nucleosomes and, as a result, affect downstream processes. Subunits of the BAF and PBAF complexes are frequently mutated in various cancers, including in acute myeloid leukemia. However, the role of distinct members of these complexes in leukemogenesis is poorly understood. BAF and PBAF complexes are distinguished from each other by unique AT-rich interacting domain (ARID) subunits, ARID1A/B and ARID2, respectively. Recently, Arid2 was suggested to have tumor suppressive functions in MLL-AF9 leukemogenesis, however the role of Arid1b is unknown. In this study, we further evaluated the tumor suppressor roles of Arid2 and Arid1b in MLL-AF9 leukemogenesis. To study the effect of the loss of Arid2 and Arid1b on MLL-AF9 leukemia in vitro, hematopoietic stem and progenitor cells from Arid1bfl/fl, Arid2fl/fl and wild type (control) mice were transformed by retroviral expression of MLL-AF9. Arid1b or Arid2 was then deleted by retroviral delivery of Cre-recombinase and cellular phenotypes were characterized by flow cytometry. In vitro characterization revealed Arid1b-/-and Arid2-/-MLL-AF9 cells have similar growth rates to control cells and no changes in cell cycle status and apoptosis. To study leukemogenesis in vivo, two hematopoietic specific conditional knockout mouse models, Vav1Cre, a constitutively active Cre-recombinase, and Mx1Cre, an inducible Cre-recombinase, were used. Primary and secondary transplant recipient mice were assessed for survival, disease burden, and leukemia stem cell frequency. In primary transplantation, leukemic initiation was not altered by the loss of the Arid2. Overall survival of primary recipients and disease burden in moribund mice (percentage of GFP positive MLL-AF9 cells in bone marrow/spleen and spleen weight) were similar to control counterparts. However, in limiting dilution secondary transplantation of Arid2fl/flVav1Cre (Arid2-/-) and wild type Vav1Cre (control) MLL-AF9, median survival of the Arid2-/- cohort decreased 1 to 2 weeks in comparison to the control cohort and there was a two-fold increase in leukemia initiating cell frequency upon deletion of Arid2. Together these results support the previous indication that Arid2 is a tumor suppressor in MLL-AF9 leukemic initiation. To gain insight into the function of Arid1b in MLL-AF9 leukemogenesis, we performed primary transplantation of Arid1bfl/flcells transduced with MLL-AF9. Our data showed that loss of Arid1b enhanced leukemic initiation in the MLL-AF9 leukemia model. One-month post induction with pIpC, Arid1bfl/fl Mx1Cre MLL-AF9 cells accumulated faster in the peripheral blood and primary recipients exhibited decreased median survival compared with wild type Mx1Cre controls. Percentage of GFP+ leukemia cells was higher in the spleen of moribund mice in the Arid1bfl/fl Mx1Cre cohort, compared with the control cohort. Taken together, our data suggests Arid1b, in addition to Arid2, acts as a potential novel tumor suppressor in MLL-AF9 leukemogenesis. To study the mechanism of Arid1b and Arid2's function in MLL-AF9 leukemogenesis, RNA-sequencing of leukemic bone marrow cells was conducted. Consistent with the previous study, Arid2 null leukemia cells had increased Gata2 expression and decreased p57 expression. Gene Set Enrichment Analysis of Arid1b null cells showed enrichment of oncogenic signatures including ATM, Cyclin D and macrophage development as potentially contributing to the underlying tumor suppressive mechanism of Arid1b. In summary, our study uncovered Arid1b, in addition to Arid2, as a potential novel tumor suppressor in MLL-AF9 leukemia. Our study, together with published studies on the requirement of other subunits (SMARCA4, SMARCD2 and DPF2) of the BAF and PBAF complexes in MLL-AF9 leukemia, reveal opposing roles of the components of the BAF and PBAF complexes in leukemogenesis, the mechanism of which require further investigation. Disclosures No relevant conflicts of interest to declare.

Splenic CD24low Red Pulp Macrophages Provide an Alternate Niche for Chronic Myeloid Leukemia Stem Cells

Chronic myeloid leukemia (CML) is a typical stem-cell driven malignancy, driven by leukemia stem cells (LSCs). LSCs are resistant to conventional therapies. This resistance is mediated by cell-intrinsic mechanisms and interactions with the microenvironment. LSCs depend on signals from a specialized microenvironment, a so called niche, to maintain their stem cell characteristics. In CML the bone marrow (BM) as a niche is well-investigated and several therapeutic targets, which aim at LSCs by interrupting their interaction with the BM-niche are under investigation. However, even though splenomegaly is a hallmark of CML the contribution of the splenic microenvironment to CML development has not been studied so far. This project aims to investigate the role of the splenic microenvironment as an independent secondary LSC niche and its contribution to disease development. To induce a CML-like disease in mice we retrovirally transduced FACS-sorted Lineage- Sca-1+ cKit+ BM cells with pMSCV-p210BCR/ABL-IRES-GFP and injected the transduced cells into non-irradiated mice. To find out if the spleen contributes to disease development we induced CML in splenectomized and sham operated mice. Splenectomized mice survived significantly longer compared to sham operated controls (median survival 31 vs. 22 days; p=0.0006) with 20% of the splenectomized mice surviving longer than 90 days. Moreover, the number of LSCs in the BM of splenectomized mice was reduced 3.7-fold (p=0.002). Flowcytometric analysis of the spleen and BM compartments of CML bearing mice revealed that the majority of the leukemic stem and progenitor cells (LSPCs) were located in the spleen (19-fold more LSCs in the spleen; p =0.007). Moreover we found the leukemic compartment in the spleen to be enriched for LSPCs compared to the BM (20 % spleen vs. 10 % BM; % LSPCs of total leukemic cells; p=0.01). To confirm this phenotypic observation functionally we performed a limiting dilution transplantation of leukemic cells from spleen and BM. In line with the phenotypic observation we found a higher frequency of LSCs in the spleen compared to the BM (1/41'703 vs. 1/432'594; p=0.02). We next analyzed the gene expression of LSPCs from spleen and BM. We found that the gene expression profile of splenic LSPCs showed higher expression of stemness-related genes and reduced expression of myeloid differentiation genes compared to BM LSPCs, indicating that the spleen is more supportive of primitive LSPCs. Knowing that the spleen contributes to disease development by providing an alternate niche for LSCs we next analyzed the spleens using confocal microscopy. We found that the LSCs resided exclusively in the red pulp. Previous studies have shown that HSCs reside in direct contact with red pulp macrophages (RPMs) during extramedullary hematopoiesis (Dutta et al., JEM, 2015). In addition we found that in spleens from human CML patients CD34+ leukemia cells localized together with macrophages (p=0.001). Furthermore we could show that RPMs are capable of producing both SCF and G-CSF. To test the role of RPMs as a potential niche component in vitro we co-incubated LSCs and RPMs overnight before plating the LSCs in a colony formation assay. We found that the co-incubation with RPMs improved the colony formation capacity of LSCs (CFUs 166 vs. 138; p=0.0356). To test the role of RPMs in vivo we depleted macrophages in CML mice using clodronate liposomes. This resulted in significantly reduced splenomegaly (867mg vs. 249mg; p&lt;0.0001) and reduced numbers of splenic LSCs (23-fold; p=0.001). This was further confirmed in a genetically engineered mouse model lacking RPMs (Spic-/-) (735mg vs. 450mg; p=0.0112 and 22-fold fewer LSCs; p&lt;0.0001). Finally, while in naïve mice RPMs can be differentiated into a CD24low and CD24high population, the CD24high population is lost in CML bearing mice. In summary, we found that the spleen provides an alternate niche for LSCs, thereby contributing to CML development. Compared to the BM niche the splenic niche is more supportive of primitive LSPCs, as shown by the higher frequency of LSCSs found in the spleen and the higher expression of stemness related genes in splenic LSCs compared to BM LSCs. Moreover we identified CD24low RPMs as a unique and central component of the splenic LSC niche. Even though we could show that RPMs are capable of producing SCF and G-CSF the exact mechanisms by which RPMs support LSCs remains to be investigated. Disclosures No relevant conflicts of interest to declare.

The First-in-Class Anti-CD47 Antibody Magrolimab (5F9) in Combination with Azacitidine Is Effective in MDS and AML Patients: Ongoing Phase 1b Results

The First-in-Class Anti-CD47 Antibody Magrolimab (5F9) in Combination with Azacitidine Is Effective in MDS and AML Patients: Ongoing Phase 1b Results

Sirpαfc Treatment Targets Human Acute Myeloid Leukemia Stem Cells

Acute myeloid leukemia (AML) is an aggressive hematologic malignancy and is the most common type of acute leukemia in adults. Although a majority of patients achieve remission following cytotoxic chemotherapy, most will relapse and ultimately die. Therapy resistance and relapse are driven by leukemia stem cells (LSC). Evidence of genetic and functional heterogeneity in the LSC compartment underscores the importance of developing therapeutic strategies that will target all subclones effectively. We previously showed that LSCs in AML depend on CD47-SIRPα interaction to evade immune surveillance (Theocharides et al, JEM 2012). CD47 acts as a "do not eat me" signal that binds to the inhibitory receptor SIRPα on macrophages and masks cancer cells from macrophage-mediated phagocytosis. TTI-621 (Trillium Therapeutics Inc., Ontario, Canada) is a human SIRPαFc protein formed by fusing the IgV doman of human SIRPα to a human IgG1-Fc moiety; it is designed to bind CD47 on leukemia cells and disrupt its interaction with SIRPα on host macrophages. Our previous studies in AML cell lines and a small number of primary AML samples demonstrated increased phagocytosis in vitro and decreased engraftment in xenotransplant models following SIRPαFc treatment (Theocharides et al, JEM 2012, Petrova et al, Clin Cancer Res 2017). Here, we tested the efficacy of TTI-621 against a broad panel of primary AML samples in xenotransplantation models to determine efficacy and response rates in this heterogeneous disease. Bulk cells obtained from the peripheral blood of 30 AML patients representing a broad range of cytogenetic and molecular subtypes were transplanted intrafemorally into sublethally-irradiated NSG mice. After a 2-week engraftment period, mice were treated with either SIRPαFc or control IgG by intraperitoneal injection 3×/week for 4 weeks, following which leukemic engraftment was determined by flow cytometry. In all but 1 sample, a significant reduction in AML engraftment was seen in SIRPαFc-treated mice compared to controls. For 23 samples defined as good responders, SIRPαFc treatment resulted in 91% (range 53-100%, p&lt;0.0001) and 98% (range 88-100%, p&lt;0.0001) reduction of leukemic engraftment in the injected femur and in non-injected bones, respectively, compared to controls. Six samples demonstrated a lesser response that was largely observed in the non-injected bones, with relative reduction of 69% (range 43-93%, p=0.03); these samples were defined as partial responders. The majority of samples from patients with unfavorable features such as age &gt;60, adverse cytogenetic risk, and secondary AML, as well as samples obtained from relapsed/resistant patients, were classified as good responders. Notably, 20 of 23 good responders had a high LSC17 score, which we have shown is associated with poor initial therapy response and short survival following standard treatments (Ng et al, Nature 2016). To determine whether SIRPαFc treatment killed LSCs, we transplanted leukemia cells harvested from primary treated mice into untreated secondary recipients at limiting dilution. For four independent samples, including three partial responders and the one non-responder, we observed a significantly lower LSC frequency (3.9-10.3 fold, p=0.002-0.024) in mice transplanted with SIRPαFc-treated cells compared to controls, indicating that SIRPαFc treatment reduced LSC numbers in primary mice, despite partial or no reduction of bulk disease. Our data demonstrate that SIRPαFc effectively targets LSCs in a human AML xenotransplantion model with high response rates across a heterogeneous cohort of primary AML samples, including samples with unfavorable risk features. SIRPαFc may be most effective in the remission setting as maintenance therapy for patients with detectable residual disease, to eradicate residual LSCs and prevent relapse. Disclosures Jin: Trillium Therapeutics: Other: licensing agreement. Wong:Trillium Therapetuics: Employment. Uger:Trillium Therapetuics: Employment. Minden:Trillium Therapetuics: Other: licensing agreement. Danska:Trillium Therapeutics: Other: licensing agreement, Research Funding. Wang:Pfizer AG Switzerland: Honoraria, Other: Travel and accommodation; Pfizer International: Honoraria, Other: Travel and accommodation; Trilium therapeutics: Other: licensing agreement, Research Funding; NanoString: Other: Travel and accommodation.

PS980 GRIK5 DRIVES SELF RENEWAL PATHWAYS IN HUMAN ACUTE MYELOID LEUKEMIA CELLS

Background:Acute myeloid leukemia (AML) is characterized by a block in differentiation and uncontrolled proliferation of immature hematopoietic precursor cells. Self‐renewing leukemia stem cells (LSCs) are supposed to initiate and promote the disease. In our previous study, we had identified a set of genes associated with high LSC frequency in normal karyotype AML. Among these, we identified the Glutamate receptor ionotropic, kainate 5 (GRIK5) coding for a transmembrane protein, which is known to be expressed in the central nervous system as a ligand‐gated ion channel receptor.Aims:In this study, we aimed to understand the functional role of GRIK5 in normal HSPCs and AML cells. In particular, we investigated how GRIK5 drives self‐renewal pathways in AML and whether GRIK5 might represent a novel cell surface target for anti‐leukemic therapy.Methods:We performed knockdown (KD) experiments using small hairpin (sh) RNAs to determine whether GRIK5 was essential for proliferation, differentiation, and survival of primary human cord blood CD34+ HSPCs, leukemic cell lines, and primary human‐derived AML xenograft cells. Furthermore, lentivirally transduced HSPCs and leukemia cells were analyzed in vitro for proliferative output, viability, immunophenotype, and colony formation potential. RNA sequencing (RNA‐Seq) and whole cell mass spectrometry followed by enrichment analysis were used to identify key signaling pathways affected by GRIK5 KD.Results:We found that knockdown of GRIK5 expression strongly impaired leukemic cell proliferation and totally abrogated human leukemic engraftment in immunocompromised mice. In the healthy system, however, GRIK5 KD predominantly reduced the proliferative output of more committed progenitors, while more immature HSPCs were less affected. RNA‐Seq data points towards a major involvement in several key self‐renewal pathways.Summary/Conclusion:Here we found that the surface receptor GRIK5 is overexpressed in the majority of AML samples compared to healthy HSPCs. GRIK5 plays a pivotal role in proliferation, differentiation, and engraftment capacity of primary human leukemia cells in xenotransplantation assays and might, therefore, be used as a novel therapeutic target for anti‐leukemic therapies.

The first-in-class anti-CD47 antibody Hu5F9-G4 is active and well tolerated alone or with azacitidine in AML and MDS patients: Initial phase 1b results.

7009 Background: Hu5F9-G4 (5F9) is an antibody targeting CD47, a macrophage immune checkpoint and “don’t eat me” signal on cancers. CD47 blockade induces tumor phagocytosis and eliminates leukemia stem cells (LSC) in AML models. Azacitidine (AZA) synergizes with 5F9 by inducing “eat me” signals on AML, enhancing phagocytosis. This trial explored the safety/efficacy of 5F9 alone or with AZA in AML/MDS patients (pts). Methods: This Phase 1b treated: r/r AML/MDS pts with 5F9; and untreated AML (induction chemo ineligible) and higher risk MDS pts with 5F9+AZA. A 5F9 priming/intrapatient dose escalation regimen (1-30 mg/kg weekly) was used to mitigate on target anemia. Results: 10 (6 AML, 4 MDS) r/r pts received 5F9 (median 2 prior therapies (range 1-6). 24 untreated pts (15 AML, 9 MDS) received 5F9+AZA. In total, median age was 73, 62% of AML pts were intermediate or poor cytogenetic risk (38% unknown), all MDS pts were intermediate or high risk by IPSS-R. 5F9 alone or with AZA was well-tolerated with no MTD reached. 5F9 did not potentiate AZA toxicities. Treatment-related AEs ( &gt; 10% of pts) for 5F9+AZA were anemia (25%), thrombocytopenia (20%), and infusion reactions (15%). In 25 efficacy evaluable pts, 8/15 (53%) untreated AML/MDS pts had a CR/CRi to 5F9+AZA (5/10 (50%) in AML, 3/5 (60%) in MDS). 1/10 (10%) r/r AML/MDS pts had a response (MLFS) to 5F9 alone. LSC frequency was reduced/eliminated in most 5F9+aza responders; 50% of responders were MRD negative by flow cytometry. 4/10 (40%) AML pts became RBC transfusion independent and 4/5 (80%) MDS pts had hematologic improvement. Time to response was more rapid (median 1.9 mos) than expected for AZA alone. As of Jan 2019, no responder has relapsed (median follow-up of 3.4 mos (range 1.1 – 6.8 mos). 2 pts had successful allogeneic transplant. Conclusions: 5F9+AZA is a novel immunotherapy blocking a key macrophage checkpoint. It has been well tolerated with robust activity in AML/MDS pts with rapid CRs and MRD negativity. Adding 5F9 to cytotoxic agents may be a promising treatment strategy. An expansion cohort is ongoing. Funded by Forty Seven and the California Institute for Regenerative Medicine. Clinical trial information: NCT03248479.

The side population enriches for leukemia-propagating cell activity and Wnt pathway expression in zebrafish acute lymphoblastic leukemia.

Cancer stem cells have been strongly linked to resistance and relapse in many malignancies. However, purifying them from within the bulk tumor has been challenging, so their precise genetic and functional characteristics are not well defined. The side population assay exploits the ability of some cells to efflux Hoechst dye via ATP-binding cassette transporters. Stem cells have increased expression of these transporters and this assay has been shown to enrich for stem cells in various tissues and cancers. This study identifies the side population within a zebrafish model of acute lymphoblastic leukemia and correlates the frequency of side population cells with the frequency of leukemia stem cells (more precisely referred to as leukemia-propagating cells within our transplantation model). In addition, the side population within the leukemia evolves with serial transplantation, increasing in tandem with leukemia-propagating cell frequency over subsequent generations. Sorted side population cells from these tumors are enriched for leukemia-propagating cells and have enhanced engraftment compared to sorted non-side population cells when transplanted into syngeneic recipients. RNA-sequencing analysis of sorted side population cells compared to non-side population cells identified a shared expression profile within the side population and pathway analysis yielded Wnt-signaling as the most overrepresented. Gene set enrichment analysis showed that stem cell differentiation and canonical Wnt-signaling were significantly upregulated in the side population. Overall, these results demonstrate that the side population in zebrafish acute lymphoblastic leukemia significantly enriches for leukemia-propagating cells and identifies the Wnt pathway as a likely genetic driver of leukemia stem cell fate.

CD34+CD38- leukemic stem cell frequency to predict outcome in acute myeloid leukemia.

Current risk algorithms are primarily based on pre-treatment factors and imperfectly predict outcome in acute myeloid leukemia (AML). We introduce and validate a post-treatment approach of leukemic stem cell (LSC) assessment for prediction of outcome. LSC containing CD34+CD38- fractions were measured using flow cytometry in an add-on study of the HOVON102/SAKK trial. Predefined cut-off levels were prospectively evaluated to assess CD34+CD38-LSC levels at diagnosis (n = 594), and, to identify LSClow/LSChigh (n = 302) and MRDlow/MRDhigh patients (n = 305) in bone marrow in morphological complete remission (CR). In 242 CR patients combined MRD and LSC results were available. At diagnosis the CD34+CD38- LSC frequency independently predicts overall survival (OS). After achieving CR, combining LSC and MRD showed reduced survival in MRDhigh/LSChigh patients (hazard ratio [HR] 3.62 for OS and 5.89 for cumulative incidence of relapse [CIR]) compared to MRDlow/LSChigh, MRDhigh/LSClow, and especially MRDlow/LSClow patients. Moreover, in the NPM1mutant positive sub-group, prognostic value of golden standard NPM1-MRD by qPCR can be improved by addition of flow cytometric approaches. This is the first prospective study demonstrating that LSC strongly improves prognostic impact of MRD detection, identifying a patient subgroup with an almost 100% treatment failure probability, warranting consideration of LSC measurement incorporation in future AML risk schemes.

OPN Maintains Leukemia Stem Cells Survival By Suppressing PUMA

Acute myeloid leukemia (AML) is the most common acute leukemia in adults. In the era of chemotherapy, the cure rate of AML is 35 to 40% of adult patients who are 60 years of age or younger while it is only 5 to 15% of patients who are older than 60 years of age could achieve relative long term survival. Leukemia stem cells (LSCs) population is a small proportion in leukemia cells but plays an important role in initiation of AML and is the main cause of the recurrence after chemotherapy due to the ability of maintaining quiescence and other mechanisms to sustain survival. How to eradicate LSCs was investigated intensively during past years, but how LSCs maintain survival is still elusive. Here we show that osteopontin(OPN), a member of extracellular matrix protein, is essential for AML stem cells survival and could be a novel and potent therapeutic target for eradicating LSCs.We used RNA interference system to knock down OPN in a mouse model of MLL-AF9 induced AML. We examined the survival of AML mice and found that OPN knockdown (KD) group showed a significant prolongation of survival. Colony forming assay showed that OPN KD AML cells formed colonies in a lower frequency than that of control AML cells. Flow cytometry showed the frequency of LSC in KD AML cells was lower than that in control AML cells(0.04±0.02% vs 3.61±0.32%) and the median fluorescence density of c-Kit is lower in KD AML cells than of control AML cells(759.2±80.8 vs 1353.5±27.5). Brdu assay indicated an increased quiescent fraction in G0/G1 phases (64.9±8.9% vs 53.6±3.4%) of KD AML cells. The apoptosis of KD AML cells is increased than that in the control AML cells(14.82±3.4% vs 4.9±1.2%). Most interestingly, the change of phenotype is much more prominent in the LSCs population, ( comparing KD group and control group, frequency of G0/G1 in LSCs: 50.2±7.5% vs 26.4±3.7%; frequency of apoptotic cells in LSCs: 58.42±5.38% vs 7.45±1.57%). Limiting dilution assay confirmed that lower frequency of LSCs in the KD AML cells than in control(1/1204 vs 1/214). Thus we concluded that OPN plays a very important role in maintain survival of LSCs.To determine whether OPN knockdown could induce the same phenotype in human AML cells. We constructed OPN knockdown cell-line model by RNA interference system. Human AML cell-lines (SKM-1, ML2, OCI-AML3, SKM-1 FL3-ITD knock in) presented the same phenotype such as the increase of apoptosis and the decrease of the fraction of cells entering cell cycle. Then we performed Realtime PCR and western blot in mouse model to reveal the underlying mechanisms. Realtime PCR showed a significantly up-regulation of P16, P18, P21, which is related to the cell cycle arrest; western blot showed a significantly increase of PUMA and cleaved-caspase3, which indicated that knockdown OPN could promote apoptosis through repress the function of BCL2 by up-regulate PUMA in protein level. Interestingly, incubation of recombinant OPN protein with OPN KD cells did not rescue the phenotype suggesting that intracellular function of OPN in LSCs is important for its survival. The detailed mechanism of how OPN regulate PUMA is still under investigation.In our current study, it showed that OPN knock down can significantly prolong the survival of leukemic mice and decrease the leukemia burden by decreasing the stemness of LSC and promoting LSC apoptosis by inducing PUMA protein function. Thus it could be a good therapeutic target for eradicating LSCs in AML. DisclosuresNo relevant conflicts of interest to declare.

Sphingosine-1-Phosphate Receptor 3 (S1PR3) Promotes Myeloid Commitment of Human Hematopoietic and Leukemic Stem Cells

Understanding the mechanisms underlying the abnormal differentiation of human acute myeloid leukemia (AML) may reveal novel therapies to eradicate leukemic stem cells (LSC), which are often resistant to standard treatments and contribute to relapse. Cellular metabolism is recognized as a hallmark of cancer and is known to be distinct between hematopoietic stem cells (HSC) and downstream progenitors. In particular, sphingosine-1-phosphate (S1P) is a bioactive lipid produced from sphingolipid metabolism that regulates proliferation and survival and is implicated in HSC egress, lymphocyte trafficking and mouse lymphoid lineage determination primarily through binding to S1PR1, one of five S1P G-protein coupled receptors. However, sphingolipids are understudied in human LSC and HSC. We recently found that sphingolipid perturbation governs HSC self-renewal and influences lineage outcome. Here, we show that S1PR3 governs myeloid commitment of LSC in a subset of human AML and is thus an attractive therapeutic target.Lipidomic analysis of LSC+ and LSC‒ fractions derived from AML patient samples and validated by xenotransplantation assays showed distinct sphingolipid alterations when compared to each other and to normal human cord blood (CB) CD34+CD38‒ stem cells and CD34+CD38+ progenitor populations. Interestingly, LSC+ fractions have increased S1P levels, prompting us to wonder if S1P signaling contributes to LSC maintenance. To gain a better understanding of the role of S1P in stemness, we examined S1P signaling in normal CB. Analysis of gene expression of S1PR1-5 and S1P transporters (SPNS2 and MFSD2B) within a comprehensive transcriptional roadmap of human hematopoiesis comprising thirteen normal populations from HSC to mature blood lineages demonstrated distinct expression patterns in specific blood lineages. S1PR1 and MFSD2B were most highly expressed in lymphoid and erythroid lineages, respectively, consistent with murine data. Notably, S1PR3 expression is specific to mature monocytes and granulocytes in human CB. S1PR3 protein was absent from the surface of long-term (LT) and short-term (ST)-HSC as determined by flow cytometry. Remarkably, lentiviral overexpression (OE) of S1PR3 was sufficient to promote myelopoiesis at the expense of erythropoiesis from LT- and ST-HSC in vitro in liquid culture, single cell assays and colony forming assays. To ascertain the mechanisms promoting myeloid commitment, we performed gene expression profiling by RNA-seq of LT- and ST-HSC following S1PR3 OE. This yielded a subset of shared genes similarly upregulated following S1PR3 OE relative to controls, including the known myeloid differentiation and AML-associated factors Early Growth Response 1 and 2 (EGR1/2) and Tribbles 1 (TRIB1). Thus, promiscuous expression of S1PR3 promotes a myeloid fate program in human HSC.S1PR3 protein expression was higher in AML patient samples relative to human CB and bone marrow cells by flow cytometry. Bioinformatic analysis of three independent AML cohorts revealed that AML patient samples with high S1PR3 gene expression also had high EGR2 and TRIB1 expression. Limiting dilution xenotransplantation assays of LSC-containing fractions sorted based on surface expression of S1PR3 demonstrated lower LSC frequency in S1PR3high versus S1PR3low/- LSC fractions. Moreover, S1PR3 OE in LSC+ fractions virtually abolished leukemic engraftment in xenotranplantation assays. These results suggest that higher S1PR3 levels in LSC are associated with a more mature myeloid state and that further increasing S1PR3 levels disrupts LSC maintenance. Treatment of mice bearing primary AML xenografts with FTY720, a S1P mimetic that targets S1P receptors including S1PR3, decreased leukemia burden in a subset of patient samples tested, including those obtained from relapsed and treatment-refractory patients. 70% of AML samples tested showed decreased LSC frequency in serial repopulation assays following FTY720 treatment. Importantly, treatment with FTY720 did not alter normal hematopoietic xenografts, demonstrating the existence of a therapeutic window. Collectively, our results provide the first direct evidence that sphingolipids govern myeloid commitment in human HSC and LSC, and demonstrate the potential of S1PR3 as a novel therapeutic target in AML for eradication of LSC while sparing HSC. DisclosuresNo relevant conflicts of interest to declare.

Leukemia Stem Cells in Relapsed AML Patients Are Uniquely Dependent on Nicotinamide Metabolism

Most AML patients who receive intensive chemotherapy achieve a significant clinical response; however, the majority will relapse and succumb to their disease, indicating that leukemia stem cells (LSCs) are not effectively targeted. Further, it has recently been shown that LSC frequency and phenotypic diversity are increased at relapse (Ho et al. Blood, 2016), thereby creating an even more challenging clinical scenario. Thus, novel therapies specifically designed to target LSCs in relapsed AML patients are urgently needed. Previously, we have shown that LSCs can be targeted by perturbing energy metabolism (Lagadinou et al. Cell Stem Cell, 2013). Therefore, the goal of the current study was to identify and target metabolic dependencies of relapsed LSCs, with the hope that this would allow improved efficacy for AML patients with relapsed disease.To achieve this objective we first measured metabolic differences in LSCs isolated from de novo and relapsed patients. This analysis revealed that relapsed LSCs have significantly increased levels of nicotinamide compared to de novo LSCs (Figure 1A). Nicotinamide is a precursor of NAD+, an essential coenzyme in energy metabolism. We hypothesized that relapsed LSCs are dependent on nicotinamide metabolism to maintain energy metabolism. To test this hypothesis, we targeted nicotinamide metabolism with the small molecule APO866, an inhibitor of Nampt, the rate-limiting enzyme for conversion of nicotinamide to NAD+. This resulted in a significant decrease in NAD+ in LSCs isolated from both de novo and relapsed AML specimens (data not shown). However, strikingly, inhibition of nicotinamide metabolism only decreased viability and colony-forming ability of LSCs isolated from relapsed AML patients, not LSCs from untreated patients (Figure 1B). To verify that inhibition of Nampt was targeting functional LSCs, we treated a relapsed AML patient specimen with APO866 for 24 hours and measured the ability of the leukemia cells to engraft into immune deficient mice. We observed a significant reduction in leukemia engraftment upon APO866 treatment (data not shown). Importantly, inhibition of nicotinamide metabolism did not affect normal hematopoietic stem cell frequency or colony forming ability (data not shown). Altogether, these data suggest that inhibition of nicotinamide metabolism specifically targets relapsed LSCs.We next sought to understand the mechanism by which inhibiting nicotinamide metabolism targets relapsed LSCs. To this end we measured changes in the major energy metabolism pathways (oxidative phosphorylation [OXPHOS] and glycolysis) in LSCs isolated de novo and relapsed AML patient specimens. Upon APO866 treatment, we observed a significant decrease in OXPHOS and OXPHOS capacity in relapsed LSCs but not de novo LSCs (Figure 1C). Furthermore, no change in glycolysis was observed (data not shown). These data demonstrate that inhibition of nicotinamide metabolism targets OXPHOS specifically in relapsed LSCs. To determine how APO866 reduced OXPHOS, we measured stable isotope metabolic flux of amino acids, the fatty acid palmitate, and glucose into the TCA cycle after APO866 treatment. We observed an increased accumulation of citrate, malate, and α-ketoglutarate from amino acids and palmitate, consistent with decreased activity of the NAD+ dependent enzymes isocitrate dehydrogenase, α-ketoglutarate dehydrogenase and malate dehydrogenase (data not shown). Through direct measurement of enzyme activity, we confirmed that isocitrate dehydrogenase, α-ketoglutarate dehydrogenase and malate dehydrogenase activity were each significantly decreased upon APO866 treatment (Figure 1D). Consistent with our previous findings we did not observe any changes in glycolysis or glucose contribution to the TCA cycle (data not shown). Overall, these data suggest that inhibition of nicotinamide metabolism through Nampt inhibition results in decreased OXPHOS through decreased TCA cycle activity.In conclusion, we have shown that relapsed LSCs have distinct metabolic properties including increased levels of nicotinamide, which can be selectively targeted to eradicate relapsed LSCs. We propose that therapeutic strategies designed to target nicotinamide metabolism may be useful for relapsed AML patients and may allow for broad efficacy such as that observed when LSCs are targeted in the up-front treatment setting. [Display omitted] DisclosuresNemkov:Omix Technologies inc: Equity Ownership. Pollyea:Curis: Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; AbbVie: Consultancy, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy; Karyopharm: Membership on an entity's Board of Directors or advisory committees.

IKZF2 Drives Self-Renewal and Blocks Myeloid Differentiation Programs in Myeloid Leukemic Stem Cells

Myeloid leukemic stem cells are maintained by programs that drive self-renewal and block myeloid differentiation through both genetic and epigenetic mechanisms. Previously, we found the chromatin remodeler IKZF2 as a target of RNA binding protein MSI2 which is a central regulator of translation in stem cell programs. In contrast to being commonly deleted in hypodiploid B-cell Acute Lymphoblastic Leukemia and acting as a tumor suppressor, we propose that IKZF2 is required for myeloid leukemia. Although IKZF2 is highly expressed in hematopoietic stem cells (HSC), we found that it is dispensable for HSC function utilizing IKZF2 deficient mice. IKZF2 is also highly expressed in leukemic stem cells (LSCs) in a murine MLL-AF9 model. Conditional deletion of Ikzf2 in the hematopoietic system with Vav-Cre system, significantly impaired LSC function as assessed through limiting dilution assays (LSC frequency is 1:7,697 in Ikzf2-deficient cells versus 1:122 cells in wildtype cells) and serial transplantations. IKZF2 deletion with a tamoxifen inducible Cre (Cre-ER) in established leukemias resulted in reduced colony formation, increased differentiation and apoptosis while delaying leukemogenesis. Furthermore, shRNA depletion of IKZF2 in another murine AML model using the oncogene AML1-Eto9a also showed reduced colony formation and delayed leukemogenesis, suggesting that IKZF2 is required for myeloid leukemia.Similar to the mouse HSCs, shRNA depletion of IKZF2 in human CD34+ enriched cord blood HSPCs resulted in no overt phenotype in colony formation, differentiation and apoptosis. Intracellular flow cytometry for IKZF2 revealed that IKZF2 is highly expressed in the CD34+CD38- fraction compared to the CD34- fraction in nine AML patients. Notably, IKZF2 depletion with shRNAs resulted in reduced frequency of CD34+CD38- fraction and reduced colony formation in AML patient samples. Depletion of IKZF2 in five human AML cell lines (MOLM-13, KCL-22, KASUMI-1, NOMO-1 and NB-4) with different oncogenes also resulted in reduced proliferation, increased differentiation and increased apoptosis. These data suggest that IKZF2 is differentially required in myeloid leukemia cells compared to normal cells.Mechanistically, ATAC-sequencing (assay for transposase-accessible chromatin with sequencing) in MLL-AF9 LSCs revealed that a substantial portion of the decreased accessibility changes occur in the intronic regions (34.65% for open peaks compared with 45.95% for closed peaks) whereas more promoter regions are opened than closed (21.26% for open peaks; 12.77% for closed peaks) when IKZF2 is lost. This suggests that IKZF2 loss leads to reduced accessibility preferentially occurring in intronic enhancers whereas increased accessibility was found at promoters. Motif enrichment analysis from the combinatorial assessment of RNA-sequencing, chromatin accessibility by ATAC-seq and direct binding of IKZF2 by the cut and run method in MLL-AF9 LSCs identified the C/EBPδ and C/EBPε as the most accessible motifs whereas HOXA9 motif became less accessible in the Ikzf2 deleted LSCs. More specifically, we found 13 genes bound by IKZF2 that contained C/EBP motifs that had also increased accessibility (Log2FC>1, pval<0.05) and increased gene expression (Log2FC>0.75, pval<0.05) in Ikzf2 deleted MLL-AF9 LSCs. Using the cre-ER expressing MLL-AF9 LSCs, we validated that C3, Fpr2, S100a8 and S100a9 were upregulated after acute deletion. These direct targets and CEBP expression correlated with IKZF2 expression in the TCGA AML patient cohort. Furthermore, forced HOXA9 expression could partially rescue the colony formation, differentiation and apoptosis effects after Ikzf2 was deleted by tamoxifen treatment. Additionally, CEBPE depletion by shRNAs partially rescued the effects of IKZF2 deletion. Thus, we demonstrate that IKZF2 is dispensable for normal hematopoiesis but required for maintaining LSC function. We find that IKZF2 can act as a chromatin remodeler that regulates the self-renewal HOXA9 gene expression program and inhibits C/EBP driven differentiation program in LSCs. Our study provides the rationale to therapeutically target IKZF2 in myeloid leukemia. DisclosuresNo relevant conflicts of interest to declare.

PTPN11 Mutations Confer Unique Metabolic Properties and Increase Resistance to Venetoclax and Azacitidine in Acute Myelogenous Leukemia

Previous studies have demonstrated the importance of energy metabolism as it relates to numerous aspects of leukemia stem cell (LSC) biology. Specifically, in acute myelogenous leukemia (AML), it has been shown that LSCs have a unique reliance on oxidative phosphorylation (OXPHOS), and that inhibition of B-cell lymphoma 2 (BCL-2) acts to down-regulate OXPHOS and eradicate LSCs in pre-clinical models. In the clinical setting, when BCL-2 inhibitor venetoclax is combined with azacitidine, high response rates (~80%) in elderly de novo AML patients have been observed (PMID: 29339097). Nonetheless, a significant portion of these patients ultimately experience disease progression. The mechanism of resistance and characteristics of these patients is poorly understood.Our preliminary data shows that the venetoclax + azacitidine (ven/aza) regimen targets LSCs through alteration of energy metabolism. Specifically, the regimen disrupts the TCA cycle leading to decreases in ATP production and inhibition of OXPHOS. This metabolic targeting is central to ven/aza efficacy and we hypothesize that resistance and progression of AML patients is due to compensatory mechanisms that restore sufficient levels of OXPHOS (PMID:3333149). Investigation of such mechanisms led us to explore the potential activity of MCL1. Previous studies have shown MCL1 can influence venetoclax resistance, however little is known about MCL1's role in metabolism, although a report in breast cancer cells suggests MCL1 modulates OXPHOS (PMID: 28978427). In leukemia, MCL1 expression has been shown to be partially upregulated through mutations in protein tyrosine phosphatase non- receptor type 11 (PTPN11), and PTPN11 mutations have been shown to increase LSC frequency. Thus, we hypothesized that PTPN11 mutations may confer resistance to venetoclax-based regimens at least partially by up-regulation of MCL1.To test this hypothesis, we investigated the relationship between PTPN11 mutations, MCL1, and the metabolic phenotype. In comparison to specimens with a wild type allele, LSCs isolated from PTPN11 mutant patient specimens showed increased levels of OXPHOS as well as glycolysis, amino acids, and fatty acids, suggesting an ability to utilize multiple energy sources for survival. PTPN11 mutant specimens also show decreased sensitivity to venetoclax, suggesting OXPHOS is not affected by venetoclax to the same degree as PTPN11 wild type specimens (fig. 1). Furthermore, when we introduced a mutated allele of PTPN11 into a primary AML specimen we observed increased oxidative phosphorylation and glycolysis, which correlated with decreased in vitro sensitivity to venetoclax (fig. 2).To test the potential role of MCL1 in PTPN11 mutant specimens, we employed a small molecule MCL-1 inhibitor. Metabolic analysis of specimens treated with the MCL-1 inhibitor showed decreased OXPHOS in PTPN11 mutant specimens (fig 3). Further, PTPN11 mutant specimens exhibit increased sensitivity to the MCL-1 inhibitor (fig. 4).To investigate a potential mechanistic link to clinical observations, we next examined 45 older, previously untreated AML patients from our institution who received ven/aza, both in the context of the multi-institutional study NCT02203773 (N=33) and with off-label use (N=12). Of 12 variables examined, only the presence of PTPN11 predicted shorter response duration (table 1). In addition, of the 9 patients who progressed ven/aza, 2 (22%) acquired PTPN11 mutations upon progression, further suggesting PTPN11 may represent a resistance mechanism to this regimen. Notably, PTPN11 is not preferentially detected in patients who progress after regimens other than ven/aza.In conclusion, AML containing PTPN11 mutations exhibit a unique energy metabolism profile. These specimens also appear to have increased sensitivity to MCL-1 inhibitors. The presence of PTPN11 mutations represents both a novel method for predicting response to ven/aza and a potential strategy for targeting patients who progress. We propose that addition of an MCL-1 inhibitor for treatment of AML patients bearing PTPN11 or related mutations may increase therapeutic responses. DisclosuresSavona:Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Incyte: Membership on an entity's Board of Directors or advisory committees, Research Funding; Boehringer Ingelheim: Consultancy. Fesik:Boehringer Ingelheim: Consultancy. Pollyea:Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Curis: Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Research Funding; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy.

Abstract 1114: IGFBP7 reduces acute myeloid leukemia stem cell survival without affecting normal hematopoiesis

Abstract Only 30-40% of acute myeloid leukemia (AML) patients survive five years after diagnosis. This poor prognosis is mainly caused by treatment failure due to chemotherapy resistance and relapse. Leukemic stem cells (LSCs) are thought to be the major determinants of AML relapse due their potential for self-renewal and chemotherapy resistance. It has been demonstrated that LSC frequency and expression of LSC gene expression signatures is highly predictive of therapy failure in AML patients, indicating the clinical importance of LSCs. LSCs co-reside with residual normal hematopoietic stem cells (HSCs) in the diseased bone marrow. Increasing the dose of chemotherapy might eliminate LSCs, however will inevitable result in non-specific elimination of HSCs, delaying or preventing recovery of normal hematopoiesis after treatment. To significantly improve AML patients outcome, there is an urgent need for identification of alternative therapies that specifically eliminate LSCs while sparing HSC. We generated gene expression profiles of HSCs, LSCs and leukemic progenitors all derived from the same AML bone marrow and identified insulin growth factor binding protein 7 (IGFBP7) as one of the top differentially expressed genes. We found that low IGFBP7 is a feature of LSCs, associated with reduced chemotherapy sensitivity and that IGFBP7 is frequently downregulated at relapse as compared to AML diagnosis. Together, these results suggest that there is a survival advantage of IGFBP7low AML cells during chemotherapy treatment. To test whether increasing IGFBP7 levels might be a strategy to deplete leukemic (stem) cells, we overexpressed IGFBP7 or added recombinant human IGFBP7 (rhIGFBP7) to primary AML cells. Strikingly, IGFBP7 overexpression or addition of rhIGFBP7 led to induction of differentiation and cell death in AML patient cells. Moreover, rhIGFBP7 reduces leukemic engraftment in an AML xenograft mouse model, reverses a stem-like gene signature, and inhibits AML blast and leukemic stem/progenitor cell survival in vitro and in vivo, while it had no influence on normal hematopoietic (stem) cell survival. Our findings define an association between chemotherapy sensitivity and IGFBP7 expression levels in primary AML cells, and indicate that treatment of AML patients with a combination of rhIGFBP7 and chemotherapy might reduce AML residual disease and LSC survival. Citation Format: Han J. Verhagen, Noortje van Gils, Anna van Rhenen, Arjo Rutten, Marjon Smit, Mei-Ling Tsui, Louise L. de Vos Klootwijk, Renee X. Menezes, Meyram Cil, Margaretha G. Roemer, Fabio Brocco, Fedor Denkers, Eline Vermue, Stan Heukelom, Sonja Zweegman, Jeroen J. Janssen, Gert J. Ossenkoppele, Gerrit Jan Schuurhuis, Linda Smit. IGFBP7 reduces acute myeloid leukemia stem cell survival without affecting normal hematopoiesis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1114.

Lsd1 inhibition induces differentiation and decreases leukemic stem cell frequency in Hoxa9/Meis1-driven AML

Lysine specific demethylase 1 (LSD1) represents a promising epigenetic target in the treatment of Acute Myeloid Leukemia (AML) and inhibition of LSD1 has been shown to facilitate a response of AML cells to all-trans retinoic acid (ATRA). In this project, we have now modeled the effect of pharmacological or genetic inactivation of LSD1 in two different murine leukemia models based on the retroviral overexpression of either Hoxa9/Meis1 (H/M) or MN1. We transformed bone marrow cells from conditional LSD1 knock-out (KO) mice.

Mir-99 Regulates Normal and Leukemic Stem Cell Function

Acute myeloid leukemia (AML) is composed of functionally heterogeneous cells including leukemic stem cells (LSCs), which exhibit the ability to self-renew and propagate disease. It is thought that failure of common chemotherapy regimens is due to insufficient eradication of LSCs. However, the mechanisms that maintain stem cell function in the hematopoietic system are not well understood. MicroRNAs play an important role in the regulation of normal and malignant hematopoietic stem cells. Our studies showed that miR-99, a miRNA highly expressed in AML patient cell populations enriched for LSC activity, is among the most highly expressed miRNAs in hematopoietic stem cells (HSCs), suggesting that miR-99 plays a role in regulating normal HSCs as well as LSCs.To test the role of miR-99 in normal hematopoiesis, we knocked down (KD) miR-99 in mouse HSCs (Lin-cKit+Sca1+CD34-SLAM+), which resulted in ~3 fold reduced methylcellulose colony formation upon secondary plating (P=0.01), as well as accelerated granulopoiesis as demonstrated by increased Gr1+Mac1+ cells 7 days after culture initiation (P<0.01), suggesting that miR-99 functions to suppresses differentiation. Consistent with this model, transplantation assays demonstrated >10-fold reduction in long-term engraftment capacity of miR-99 KD compared to scrambled controls (P=0.0004). In addition, Ki-67/DAPI staining of stably engrafted miR-99 KD hematopoietic stem and progenitor cells (HSPCs) showed increased cell cycling, demonstrating that miR-99 also maintains HSPC quiescence. Gene set enrichment analysis (GSEA) of RNA-sequencing data generated from stably engrafted Lin-Sca-1+c-Kit+ cells revealed that miR-99 KD induces significant depletion of LT-HSC gene signatures (P<0.001) and induction of a late progenitor signature (P<0.001), providing further evidence that miR-99 normally functions to maintain HSPCs in the undifferentiated state.To test whether miR-99 maintains LSCs, we performed miR-99 KD experiments using the MLL-AF9 retroviral mouse model. miR-99 KD resulted in a significant extension in survival in secondary transplants compared to scrambled controls (median 92 days vs. 48 days, P<0.001). Evaluation of the bone marrow at the time of death revealed ~2.5 fold decrease in the frequency of LSCs (P<0.01) and ~2 fold increase in the percentage of cycling LSCs (in SG2M) (P<0.001). Analysis of RNA-seq data from miR-99 KD LSCs revealed induction of a differentiated normal progenitor signature (P<0.001) and depletion of a shared HSC/LSC gene signature (P=0.05). Giemsa staining of peripheral blood showed miR-99 KD also induced a significant increase in the number of differentiated myeloid precursors in the peripheral blood (P<0.001), reminiscent of AML differentiation-inducing agents used in the clinic such as ATRA. Consistent with a role in regulating leukemic blast differentiation, microRNA-Seq data from the 153 AML patients in the TCGA database revealed that miR-99 expression inversely correlated with their French-American-British classifications, with low expression levels associated with M4 and M5 subtypes. Compatible with a role in maintaining LSCs, miR-99 KD in a primary AML sample reduced long-term engraftment upon xenotransplantation into NSG mice, and the engrafting cells displayed increased CD14 expression. Together, these data demonstrate that similar to normal HSPCs, miR-99 maintains LSCs function.As miR-99 functions to maintain both LSCs and HSCs, we asked which miR-99 target genes mediate miR-99 KD phenotypes. To address this question, we performed a shRNA library-based forward genetic screen designed to rescue the reduced HSC function following miR-99 KD. We designed 180 shRNAs against 45 predicted miR-99 targets that we identified as upregulated upon acute miR-99 KD in mouse HSPCs. Among the conserved miR-99 targets, Hoxa1, a member of the Hox family of transcription factors, was among the top hits, with all 4 shRNAs being enriched compared to controls. Ectopic expression of Hoxa1 in MonoMac6 AML cells was sufficient to induce differentiation, a phenotype similar to miR-99 KD. These data indicate that Hoxa1 is an important downstream mediator of miR-99 function. DisclosuresNo relevant conflicts of interest to declare.