Search

Mutations in chromatin regulator ASXL1 are frequently identified in myeloid malignancies, in particular ∼40% of patients with chronic myelomonocytic leukemia (CMML). ASXL1 mutations are associated with poor prognosis in CMML and significantly co-occur with NRAS mutations. Here, we show that concurrent ASXL1 and NRAS mutations defined a population of CMML patients who had shorter leukemia-free survival than those with ASXL1 mutation only. Corroborating this human data, Asxl1−/− accelerated CMML progression and promoted CMML transformation to acute myeloid leukemia (AML) in NrasG12D/+ mice. NrasG12D/+;Asxl1−/− (NA) leukemia cells displayed hyperactivation of MEK/ERK signaling, increased global levels of H3K27ac, upregulation of Flt3. Moreover, we find that NA-AML cells overexpressed all the major inhibitory immune checkpoint ligands: programmed death-ligand 1 (PD-L1)/PD-L2, CD155, and CD80/CD86. Among them, overexpression of PD-L1 and CD86 correlated with upregulation of AP-1 transcription factors (TFs) in NA-AML cells. An AP-1 inhibitor or short hairpin RNAs against AP-1 TF Jun decreased PD-L1 and CD86 expression in NA-AML cells. Once NA-AML cells were transplanted into syngeneic recipients, NA-derived T cells were not detectable. Host-derived wild-type T cells overexpressed programmed cell death protein 1 (PD-1) and T-cell immunoreceptor with immunoglobulin and ITIM domains (TIGIT) receptors, leading to a predominant exhausted T-cell phenotype. Combined inhibition of MEK and BET resulted in downregulation of Flt3 and AP-1 expression, partial restoration of the immune microenvironment, enhancement of CD8 T-cell cytotoxicity, and prolonged survival in NA-AML mice. Our study suggests that combined targeted therapy and immunotherapy may be beneficial for treating secondary AML with concurrent ASXL1 and NRAS mutations.

Introduction: Lenalidomide (LEN) is approved for treatment of Myelodysplastic Syndrome (MDS) with chromosome 5q deletion (del 5q) as it has shown to induce disease remission and transfusion independence in close to 65% of this population. For patients (pts) without del 5q, LEN has shown to reduce transfusion needs in one-quarter of pts, but its use is limited by significant thrombocytopenia, and pts with platelet counts (PLTs) Methods: Pts aged ≥18 years with low/int risk MDS per IPSS, or non-proliferative chronic myelomonocytic leukemia (CMML), with bone marrow blasts Results: 51 pts were accrued, with 44 evaluable pts. Of evaluable pts, most were male (70%), median age was 71 years (range 34-93). 42 (95%) pts had MDS and 2 (5%) CMML. 24 (55%) pts were low-risk per IPSS-R score. 28 (64%) pts were treatment-naïve and 16 (36%) had received ≥1lines of treatment, including erythropoiesis stimulating agents. 24 (55%) pts were assigned to arm A and 20 (45%) pts to arm B. ELT/LEN was well tolerated. Non-hematological grade 3-4 treatment-related adverse events were few, including G3 hyperbilirubinemia (7%), G3 transaminitis (2%), G3 diarrhea (2%), G3 arthralgias (2%). 2 pts had major bleeding events; there were 2 deaths, one attributable to pneumonia and one to gallbladder cancer. 1 of 31 pts who received ELT had a reversible increase in peripheral blasts while on treatment, and 1 pt had development of marrow fibrosis after 6 years of ELT. ITT analysis of total population (51) revealed an objective response (ORR) of 35%; ORR of 32% (9/28) in arm A (PLTs ≥50,000),and 39% (9/23) in arm B (PLTs 17 pts received ELT alone, among these ORR was 41% (7/17), with 29% achieving bilineage responses, 1 CR. 13 pts received LEN alone, ORR was 46% (6/13), of these 100% achieved RBC-TI. 14 pts received ELT/LEN combination, ORR was 36%, 2 (14%) achieved bilineage responses and 2 (14%) CR. There were 2 HIV-positive MDS pts, both achieved sustained CRs, one with ELT alone and one with ELT/LEN combination. Mean time to response was 9.4 wks (range 6-12.4) for ELT alone, 10.9 wks (range 2.4-16) for LEN alone, and 9.9 wks (range 2-20) for ELT/LEN combination. Mean duration of response was 102 wks for ELT (range 8-ongoing), 63 wks (range 25-ongoing) for LEN, and 66 wks (range 8.3-ongoing) for ELT/LEN as of last follow-up. Conclusions: Treatment with ELT/LEN demonstrates good efficacy with ORR of 40.9%, response durability and an acceptable safety profile for evaluable pts with low/int risk MDS. ELT can lead to single agent responses with an acceptable safety profile and sizable proportion of multilineage responses. One patient developed bone marrow fibrosis and one patient had a transient increase in blasts, hence, undermining these pre-existing safety concerns. Lastly, using our study schema 32% of pts were able to initiate or continue LEN achieving a 36% ORR whereas in prior studies these pts would have been excluded. Figure 1 Figure 1. Disclosures Yacoub: Agios: Membership on an entity's Board of Directors or advisory committees; Acceleron Pharma: Membership on an entity's Board of Directors or advisory committees; CTI Biopharma: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees; Incyte: Speakers Bureau. Gritsman: iOnctura: Research Funding. Shastri: Kymera Therapeutics: Research Funding; Onclive: Honoraria; GLC: Consultancy; Guidepoint: Consultancy. Verma: Celgene: Consultancy; Acceleron: Consultancy; Novartis: Consultancy; Stelexis: Consultancy, Current equity holder in publicly-traded company; Eli Lilly: Research Funding; Curis: Research Funding; Medpacto: Research Funding; Incyte: Research Funding; Stelexis: Current equity holder in publicly-traded company; Throws Exception: Current equity holder in publicly-traded company; GSK: Research Funding; BMS: Research Funding. OffLabel Disclosure: Lenalidomide has shown to be effective achieving RBC Transfusion Independence in Myelodysplastic Syndromes without deletion 5q Eltrombopag is a TPO-R agonist, its purpose in this trial is to evaluate if it would increase platelet counts in patients with Myelodisplastic Syndrome.

Acute myeloid leukemia (AML) is the most frequent leukemia in elderly individuals with a median age at diagnosis of 67 years (Juliusson et al., Blood 2009). It arises in a step-wise process and originates from hematopoietic stem cells (HSC) (Jan et al.,Sci Transl Med. 2012). Genetic and epigenetic alterations drive the formation of pre-leukemic HSC clones with altered function, which can gain dominance and eventually give rise to AML upon the acquisition of cooperating lesions (Jan et al.,Sci Transl Med. 2012). However, it is currently impossible to predict which healthy elderly individuals with clonal hematopoiesis will eventually develop myeloid malignancies, as the pathways to leukemia are unknown. Heterozygous inactivating mutations of the epigenetic regulator Ten-Eleven Translocation-2 (TET2) are commonly found in patients with AML, yet also in a remarkable fraction of healthy elderly individuals in whom it is associated with clonal hematopoiesis (Busque, et al Nat Genet. 2012). These observations and studies in Tet2-deficient mice strongly suggest that TET2 inactivation is an early event in the pathogenesis of myeloid malignancies, but is not sufficient to fully transform HSC (Moran-Crusio et al., Cancel Cell 2011). TET2 cooperates with several transcription factors to regulate hematopoiesis (Rasmussen et al., Genome Res 2019), one of which is PU.1 (de la Rica et al., Genome Biol. 2013), an essential transcription factor governing normal hematopoiesis (Iwasaki et al., Blood 2005). In humans, PU.1 activity or expression is only moderately impaired in the majority of AML patients, and remarkably, also in aged HSC (Will et al., Nat Med. 2015), underscoring the essentiality of PU.1. Importantly, PU.1 target genes are frequently found hypermethylated in AML (Sonnet et al., Genome Med. 2014, Kaasinen et al., Nat Commun. 2019), suggesting a profound epigenetic inactivation of the PU.1 network. We hypothesized that moderate impairment of PU.1 abundance, as found in AML, can cooperate with loss-of-function mutations of Tet2 to initiate malignancy. We developed a novel tissue-specific compound mutant mouse model carrying heterozygous deletion of an upstream regulatory element (URE) of Pu.1 along with Tet2 deletion (Vav-iCre+ PU.1URΕ∆/+Tet2+/flox; Vav-iCre+ PU.1URΕ∆/+Tet2flox/flox). While none of the single mutant mice developed AML, compound mutant mice developed aggressive myeloid leukemia whose penetrance and latency exhibited Tet2 dose dependency. The disease presented with leukocytosis, anemia and splenomegaly. By cell morphology analysis of the peripheral blood, bone marrow and spleen, the leukemic mice exhibited accumulation of differentiation-blocked myeloblasts, myelocytes and/or metamyelocytes, that was confirmed using detailed myeloid differentiation markers, distinguishing the disease in immature or mature AML. Furthermore, gold standard in vitro and in vivo assays, assessing both self-renewal and differentiation capacity of double mutant mice-derived cells, revealed that the expanded differentiation-blocked stem and progenitor cells bear aberrant self-renewal and disease-initiating capacities. Comprehensive molecular profiling by next generation sequencing of disease-initiating cells uncovered a substantial overlap with human AML, such as functional GF1b loss with concomitant overexpression of CD90/Thy1 (Thivakaran et al., Haematologica 2018). Importantly, our analyses also revealed transcriptional dysregulation, hypermethylation of PU.1 regulated enhancers with concomitant loss of enhancer activity and alterations in chromatin accessibility of particularly genes co-bound by PU.1 and TET2. Current efforts focus on identifying key effectors of the dysregulated PU.1/TET2 sub-network driving malignant transformation in clonal hematopoiesis. Our collected data provide proof of concept that moderate PU.1 dose impairment can functionally cooperate with the inactivation of Tet2 in the initiation of myeloid leukemia and uncovers a likely unifying AML pathomechansim. Disclosures Will: Novartis Pharmaceuticals: Other: Service on advisory boards, Research Funding.

Acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) are associated with disease-initiating stem cells that are not eliminated by conventional therapies. Novel therapeutic targets against preleukemic stem cells need to be identified for potentially curative strategies. We conducted parallel transcriptional analysis of highly fractionated stem and progenitor populations in MDS, AML, and control samples and found interleukin 8 (IL8) to be consistently overexpressed in patient samples. The receptor for IL8, CXCR2, was also significantly increased in MDS CD34(+) cells from a large clinical cohort and was predictive of increased transfusion dependence. High CXCR2 expression was also an adverse prognostic factor in The Cancer Genome Atlas AML cohort, further pointing to the critical role of the IL8-CXCR2 axis in AML/MDS. Functionally, CXCR2 inhibition by knockdown and pharmacologic approaches led to a significant reduction in proliferation in several leukemic cell lines and primary MDS/AML samples via induction of G0/G1 cell cycle arrest. Importantly, inhibition of CXCR2 selectively inhibited immature hematopoietic stem cells from MDS/AML samples without an effect on healthy controls. CXCR2 knockdown also impaired leukemic growth in vivo. Together, these studies demonstrate that the IL8 receptor CXCR2 is an adverse prognostic factor in MDS/AML and is a potential therapeutic target against immature leukemic stem cell-enriched cell fractions in MDS and AML.

Epigenetic regulators are recurrently mutated and aberrantly expressed in acute myeloid leukemia (AML). Targeted therapies designed to inhibit these chromatin-modifying enzymes, such as the histone demethylase lysine-specific demethylase 1 (LSD1) and the histone methyltransferase DOT1L, have been developed as novel treatment modalities for these often refractory diseases. A common feature of many of these targeted agents is their ability to induce myeloid differentiation, suggesting that multiple paths toward a myeloid gene expression program can be engaged to relieve the differentiation blockade that is uniformly seen in AML. We performed a comparative assessment of chromatin dynamics during the treatment of mixed lineage leukemia (MLL)-AF9-driven murine leukemias and MLL-rearranged patient-derived xenografts using 2 distinct but effective differentiation-inducing targeted epigenetic therapies, the LSD1 inhibitor GSK-LSD1 and the DOT1L inhibitor EPZ4777. Intriguingly, GSK-LSD1 treatment caused global gains in chromatin accessibility, whereas treatment with EPZ4777 caused global losses in accessibility. We captured PU.1 and C/EBPα motif signatures at LSD1 inhibitor-induced dynamic sites and chromatin immunoprecipitation coupled with high-throughput sequencing revealed co-occupancy of these myeloid transcription factors at these sites. Functionally, we confirmed that diminished expression of PU.1 or genetic deletion of C/EBPα in MLL-AF9 cells generates resistance of these leukemias to LSD1 inhibition. These findings reveal that pharmacologic inhibition of LSD1 represents a unique path to overcome the differentiation block in AML for therapeutic benefit.

Transforming growth factor beta is an important regulator of hematopoiesis and its overactivation has been shown to occur in myelodysplastic syndromes. Even though TGF-beta leads to activation of the transcription factor SMAD2, the exact gene expression program and chromatin changes stimulated in human hematopoiesis are not well studied. We generated primary erythroid progenitors from human CD34+ stem cells and assessed genome occupancy of SMAD2 upon TGF-beta stimulation by Chip-seq analysis. We observed widespread occupancy of important stem cell and differentiation regulators with approximately 7000 binding regions. Parallel RNA-seq identified gene expression changes that are associated with smad2 binding at respective promoters and enhancers. Genes involved in stem cell maintenance as well as erythroid differentiation were prominently altered by TGF-beta and contained prominent SMAD2 peaks in hematopoietic progenitors. Next, to determine the relative contribution of various TGF-beta ligands in bone marrow failure, we analyzed a large cohort of MDS serum samples and found TGF-beta 1 (TGF-b1) to be the predominantly elevated ligand in 30% samples. Exposure to MDS patient serum led to erythroid differentiation blocks in primary hematopoietic stem cells in clonogenic cultures. Finally, we analyzed the effect of constitutive secretion of TGF-b1 in a transgenic mouse containing TGF-b1 gene driven by an albumin promoter. These mice constitutively secrete TGF-b1 and become cytopenic and show marrow histological changes that mimic human MDS. We observed stem cell expansion and block at the stage of common myeloid progenitor (CMP) in these mice; findings that are similar to stem and progenitors alterations seen in human MDS patients. To inhibit TGF-b1 signaling we next used a novel ALK5 (TGF-beta receptor I kinase) inhibitor that has been shown to be safe for human use. Vactosertib is specific small molecule inhibitor of ALK5 with a IC50 between 1 x 10-8 M and 1 x 10-9 M and has been shown to have an acceptable safety profile in phase I trial in solid tumors (NCT02160106). Vactosertib was able to inhibit smad2 phosphorylation and activation in a variety of hematopoietic cell lines. Vactosertib treatment led to reversal of hematopoietic alterations upon exposure to MDS serum in clonogenic assays. Vactosertib dosing in vivo also led to increased blood counts in TGF/alb transgenic mice and resulted in normalization of stem and progenitor changes. In conclusion, we demonstrate increased TGF-b1 levels in subset of MDS samples. We also determined TGF-b1 stimulated smad2 binding in important regulatory regions of the genome in primary human erythroid progenitors. We also demonstrate the preclinical efficacy of novel TGF-beta receptor 1 kinase inhibitor in vitro and in vivo. These studies provide preclinical rationale for using this ALK5 inhibitor in trials. Disclosures Huh: MedPacto Inc: Employment. Hwang:MedPacto Inc: Employment. Seong-Jin:MedPacto Inc: Employment. Will:Novartis Pharmaceuticals: Research Funding. Steidl:BayerHealthcare: Consultancy, Research Funding; Novartis: Consultancy, Research Funding; GlaxoSmithKline: Research Funding; Celgene: Consultancy; Aileron Therapeutics: Consultancy, Research Funding; Stelexis Therapeutics: Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Other: Scientific Co-Founder; Pieries Pharmaceuticals: Consultancy. Verma:Stelexis: Equity Ownership, Honoraria; Acceleron: Honoraria; Celgene: Honoraria; BMS: Research Funding; Janssen: Research Funding.

Introduction The World Trade Center (WTC) disaster exposed first responders to high levels of aerosolized carcinogens (Lioy et. al. Env. Health Perspect 2002). Clonal hematopoiesis is associated with exposure to smoking and genotoxic stimuli (Jaiswal et. al. NEJM 2014; Genovese et. al. NEJM 2015). We sought to determine its incidence in WTC-exposed first responders. We also assessed the effect of WTC particulate matter (WTC-PM) on genome integrity in vitro, and in murine studies. Methods Deep targeted sequencing was performed on blood collected from 481 first responders (429 WTC-exposed firefighters, 52 WTC-exposed emergency medical service workers) and 52 non-exposed first responders. Samples were analyzed for 237 genes mutated in hematologic malignancies and interpreted using reference databases. Non synonymous somatic mutations were annotated and analyzed. Results In the WTC-exposed cohort, 57 individuals with 66 somatic mutations of expected pathogenic potential were identified (overall prevalence 11.9%). In the non-exposed cohort, only one pathogenic mutation was found in the IDH2 gene (overall prevalence 1.9%). There was a strong association between increasing age and prevalence of mutations in the WTC-exposed cohort (Fig 1A). DNMT3A (16/66), TET2 (7/66), SF3B1 and SRSF2 (3/66 each) were the most common genes identified in the WTC-exposed cohort (Fig 1B). Median VAF was 12% and missense mutations were most frequent alteration. Aging, smoking, DNA repair and alkylating agent exposure related mutational signatures were observed with a cytosine to thymine (C→T) transition being most common. Next, we assessed the effect of WTC-PM on genome integrity and replication in vitro. WTC-PM that was collected in the first three days after 9/11 was used in concentrations mimicking exposure levels. Lymphocytes exposed to WTC-PM demonstrated a significant increase in phosphorylated H2AX foci accumulation, suggesting a DNA damage response (Fig 2). Since common fragile sites (CFSs) detect basal levels of stress in the cell, and activate DNA damage response (DDR), we profiled DNA replication dynamics at CFS-FRA16D at very high resolution using the single molecule analysis of replicated DNA (SMARD) assay. Treatment with WTC-PM significantly altered replication at two common fragile sites (regions 1 and 2 of FRA16D, Fig 3A) with replication pausing being observed at multiple sites (Fig 3B-I, white rectangles). Striking increase in replication initiation was seen, characterized as dormant origins activated to rescue replication pausing (Fig 3E, J). These alterations were accompanied by a corresponding increase in replication speed, conditions that lead to DNA replication errors and mutagenesis (Fig 3F, K). Next, we treated mice with WTC-PM via the oropharyngeal route to mimic first responder exposures, and then harvested and analyzed their bone marrow compartments. Significant expansion of hematopoietic stem cells (Kit+, Sca1+, Lineage-ve, KSL) was seen in WTC-PM treated mice (Fig 4A,B). Whole genome sequencing of sorted stem cells showed a significant increase in non-synonymous SNPs, deletions and indels in the WTC-PM treated samples when compared to control (Fig 4C-E). These genomic alterations were found to occur at low VAF throughout the whole genome, demonstrating widespread genotoxic effects of WTC-PM on hematopoietic stem cells in vivo (Fig 4F). Discussion We report a high burden of mutations in 11.9% (57/481) WTC-exposed first responders compared to the non-exposed cohort (1.9%, 1/52). The frequency of the somatic mutations was many fold higher than in previous studies (Jaiswal et. al. NEJM 2014; Genovese et. al. NEJM, 2015). In the 50-59 year age group, 10% of WTC-exposed individuals carried somatic mutations, compared to the frequency of 2.5% reported by Jaiswal et. al. for the same age group. Despite deeper sequencing performed in our study, the median VAF in our study was 12%, indicating that the difference in technique did not bias our study towards increased detection of small, subclinical clones when compared to previous studies. Furthermore, we demonstrate that WTC-PM can perturb DNA replication and increased genomic instability in vivo, potentially leading to higher burden of clonal hematopoiesis in WTC-exposed first responders. These results demonstrate adverse environmental exposures can be associated with a high rate of clonal hematopoiesis. Disclosures Landgren: Sanofi: Membership on an entity's Board of Directors or advisory committees; Adaptive: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Merck: Other: IDMC; Theradex: Other: IDMC; Abbvie: Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Karyopharm: Membership on an entity's Board of Directors or advisory committees. Fletcher:Genoptix/Neogenomics: Employment. Ebert:Broad Institute: Other: Contributor to a patent filing on this technology that is held by the Broad Institute.; Celgene: Research Funding; Deerfield: Research Funding. Steidl:GlaxoSmithKline: Research Funding; Celgene: Consultancy; Aileron Therapeutics: Consultancy, Research Funding; Stelexis Therapeutics: Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Other: Scientific Co-Founder; Pieries Pharmaceuticals: Consultancy; BayerHealthcare: Consultancy, Research Funding; Novartis: Consultancy, Research Funding. Will:Novartis Pharmaceuticals: Research Funding. Verma:Stelexis: Equity Ownership, Honoraria; Acceleron: Honoraria; Celgene: Honoraria; BMS: Research Funding; Janssen: Research Funding.

Hematopoietic stem cells (HSC) harbor extensive self-renewal capabilities along with multilineage differentiation plasticity to sustain blood production over a lifetime. Maintenance of a fully functional proteome that can rapidly change to adjust to the status of HSC activation is facilitated in part by two major intracellular proteolytic pathways, the ubiquitin proteasome system and the autophagy/lysosomal system. Alterations in HSC proteostasis have been associated with a number of degenerative and malignant diseases, underscoring the importance of elucidating the precise contribution of components of the cellular proteostasis network to HSC maintenance. In this work, we have focused in a highly selective type of autophagy, known as chaperone-mediated autophagy (CMA), whereby individual proteins bearing a unique pentapeptide motif (KFERQ-like) are targeted for degradation in lysosomes upon binding to the heat shock cognate protein of 70 kDa (HSC70). Substrate proteins are directly translocated into the lysosomal lumen through a dedicated multiprotein translocation complex. The main component of this complex is the lysosome-associated membrane protein type 2A (LAMP-2A) that also serves as substrate receptor and is the limiting component of CMA. Using novel mouse models that allow for CMA tracking (KFERQ-Dendra mice) and for selective depletion of LAMP-2A in hematopoietic cells (Vav-iCre:LAMP2Af/f mice), we have investigated the physiological role of CMA in HSCs during steady-state and upon activation and gained novel insights on the consequences of CMA failure in these cells in aging. Our work revealed that the basal CMA activity detected in quiescent HSC under steady-state conditions is significantly stimulated upon HSC activation following 5-fluorouracil (5-FU) in vivo exposure. This upregulation of CMA is necessary to ensure HSCs persistence during activation because, upon serial 5-FU injections, CMA-deficient HSCs had a significantly reduced multilineage reconstitution ability with premature bone marrow failure. We found reduced long-term colony formation of CMA-deficient HSCs in serial colony formation assays and demonstrated that these cells had a significant and progressive disadvantage of repopulating lethally-irradiated congenic recipient mice upon serial bone marrow transplantation. We also found that CMA becomes increasingly important for the maintenance of functional HSCs in aging, since as mice age, CMA-deficient HSCs showed an even greater functional defect compared to age-mated control-derived HSCs. Using comparative transcriptomics and metabolomics on HSCs from control and LAMP-2A-deficient mice, we found evidence for metabolic alterations and dysfunctional redox signaling. We confirmed that CMA-deficient HSCs have reduced rates of glycolysis, lower ATP production and higher reactive oxygen species levels than control cells. Deficient cellular energetics and increased oxidative stress are important consequences of CMA failure in HSC, since supplementation with pyruvate or treatment with anti-oxidant agents (i.e. N-acetyl-cysteine) was sufficient to restore CMA-deficient HSC function, as measured by serial colony formation. Proteomic analysis of CMA-deficient Lin-Scal+c-Kit+ (LSK) cells revealed an overall increase of acetylated and oxidized proteins, including key metabolic enzymes and proteins required for the cellular response to oxidative stress. We propose that, upon CMA failure, the inability of HSC cells to timely turning over these regulatory proteins, favors accumulation of unwanted post-translational modifications that interfere with their normal functioning. Together, our findings suggest that CMA upregulation during HSCs activation is required for the proteome remodeling that facilitates transition from quiescent to activated cells. By assuring timely turnover of selected proteins, CMA sustains the metabolic adaptation required to meet these cells' energetic needs and assures an efficient cellular response to stress. Disclosures No relevant conflicts of interest to declare.

Management of hematologic disease- and therapy-related thrombocytopenia remains a serious clinical issue, especially in patients with myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML). The ribonucleoside and DNA-demethylating agent azacytidine (AZA), has proven useful for the treatment of patients with MDS and AML not eligible for stem cell transplantation. While low-dose AZA therapy induces clinical remissions in up to 50% of treated patients, it comes at the cost of aggravating pre-existing thrombocytopenia which is observed in a subset of patients; this can lead to increased bleeding and bleeding-associated mortality, and importantly, often requires dose modifications and delays of therapy. Thus, identification of strategies alleviating ineffective megakaryopoiesis will likely lead to increased therapeutic efficacy for patients with MDS/AML. Eltrombopag (EP), a second-generation small molecule thrombopoietin receptor (TPO-R) agonist was effective in raising platelet counts in patients with MDS as a single agent, as well as in combination with certain standard of care therapies. However, it failed to stimulate platelet production during the first four cycles of AZA treatment as uncovered by a recent phase III placebo-controlled clinical study (SUPPORT; NCT02158936). The goals of this study were to identify the cellular and molecular underpinnings of AZA-associated inhibition of megakaryopoiesis and to assess the ineffectiveness of EP in mitigating AZA treatment-associated thrombocytopenia. Our results demonstrate that at a clinically-equivalent and non-cytotoxic dose, AZA rapidly induces transient activation of interferon type I (IFN-I) signaling in various hematopoietic cell types, including stem and lineage-committed progenitor cells (HSPCs). We detected IFNα and IFNβ production and release using ELISA and intracellular flow cytometry on primary total mononuclear cell- and purified CD34-positive HSPC populations derived from cord blood, bone marrow from healthy volunteers or patients with MDS/AML. AZA-mediated activation of Type I IFNs in healthy control- and MDS/AML cells was preceded by an accumulation of double-stranded RNA (dsRNA) species and decreased total RNA cytosine methylation measured by immunocytochemistry and intracellular FACS analysis; this suggested that AZA triggered the accumulation of immunogenic RNA species which elicit an IFN-I response. In support, we found Toll like receptor 3 (TLR3) activation and phosphorylation of STAT1 in CD34+ HPSC, along with premature activation of Suppressor of Cytokine Signaling 1 (SOCS1), a well-known JAK/STAT-dependent signaling attenuator. This rapid AZA-induced viral mimicry response led to abrogation of thrombopoietin (TPO) or EP-stimulated TPO-R signaling and inhibition of ex vivo megakaryocyte progenitor proliferation quantified by colony formation in semi-solid medium. Importantly, inhibition of IFN-I signal activation using the JAK3 inhibitor decernotinib, the IFNα/β-blocking peptide, B18R, or RNA interference-mediated knock-down of SOCS1 counteracted the inhibitory effects of AZA on TPO-R stimulation and restored megakaryopoiesis. Given these observations, we pre-clinically tested a revised treatment protocol, in which primary cells were first exposed to AZA for four days followed by TPO-R stimulation using TPO or EP. This new treatment strategy alleviated AZA's inhibitory effects at the molecular and cellular levels, demonstrating that upon resolution of the AZA-mediated vial mimicry response, EP and TPO can effectively stimulate TPO-R signaling and megakaryopoiesis. Together, our data reveal a mechanistic basis of AZA-mediated inhibition of megakaryopoiesis in patients with MDS/AML. Additionally, we show that EP cannot overcome the megakaryopoiesis-inhibitory effects of acute IFN-I signaling activation upon AZA exposure. Findings of our study are consistent with and provide a molecular explanation for the observations made in the context of the SUPPORT study. In the future, it will be critical to better understand and potentially counteract the megakaryopoiesis-inhibitory effects by IFN-I pathway activation upon AZA therapy in patients with MDS/AML. Disclosures Okoye-Okafor: Novartis Pharmaceuticals: Research Funding. Pallaud:Novartis Pharmaceuticals: Employment. Marques Ramos:Novartis Pharmaceuticals: Employment. Verma:Janssen: Research Funding; BMS: Research Funding; Celgene: Honoraria; Stelexis: Equity Ownership, Honoraria; Acceleron: Honoraria. Heckman:Celgene: Research Funding; Novartis: Research Funding; Oncopeptides: Research Funding; Orion Pharma: Research Funding. Will:Novartis Pharmaceuticals: Research Funding.

SF3B1 mutations are the most frequently occurring splicing factor mutations in MDS and AML, however the misspliced genes that contribute the malignant state in SF3B mutant MDS or AML remains unclear. We determined that SF3B1 mutant cases of MDS express a longer, active isoform of interleukin 1 receptor associated kinase (IRAK4). IRAK4 is a serine/threonine kinase that is downstream of toll-like receptor (TLR) signaling and leads to activation of oncogenic signaling states, including NF-kB and MAPK. Examination of IRAK4 by RNA sequencing showed that normal cells predominantly express small IRAK4 isoforms resulting from exclusion of the part of exon 6. These isoforms are targeted for proteosomal degradation leading to diminished IRAK4 expression and activation in normal cells. In contrast, a large proportion of MDS/AML samples with SF3B1 mutation show increased expression of an IRAK4 isoform that retains full exon 6, encoding the full-length protein (IRAK4-Long). Consequently, we show that expression of mutant SF3B1-K700E in leukemic cells is associated with increased NF-kB activity, suggesting that mutations in SF3B1 instruct expression of IRAK4 RNA isoforms with maximal functional potential. Furthermore, SF3B1 mutant MDS and AML cells exhibited a block in hematopoietic differentiation in clonogenic assays. This differentiation block was ameliorated with pharmacologic inhibition of IRAK4 with CA-4948, a potent oral clinically useful small-molecule inhibitor of IRAK4. CA-4948 blocked TLR-stimulated cytokine release in various cell models and also led to decreased leukemic burden in mice xenografted with SF3B1 mutant MDS/AML cells. Finally, we determined that SF3B1 mutation induced IRAK4 activation led to TRAF6 mediated K63 ubiquitination of critical cell cycle and regulatory proteins directly implicated in oncogenesis. We had recently shown that U2AF1 mutations can lead to IRAK4 activation via retention of exon 4 (Smith et al, Nat Cell Bio, 2019). Our data now demonstrate that SF3B1 leads to overactivation of IRAK4 via retention of a different exon (exon 6), thus reinforcing that IRAK/TRAF6 activation is a common downstream oncogenic pathway in splicing factor mutated MDS/AML. Taken together, in this study, we find that mutations in SF3B1 induce expression of therapeutically targetable "active" IRAK4 isoforms and provide a genetic link between a spliceosome mutation and oncogenic innate immune signaling in MDS and AML. Disclosures Booher: Curis: Employment. Ramachandra:Aurigene: Employment. Samson:Curis: Employment. Will:Novartis Pharmaceuticals: Research Funding. Steidl:BayerHealthcare: Consultancy, Research Funding; Novartis: Consultancy, Research Funding; GlaxoSmithKline: Research Funding; Celgene: Consultancy; Stelexis Therapeutics: Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Other: Scientific Co-Founder; Pieries Pharmaceuticals: Consultancy; Aileron Therapeutics: Consultancy, Research Funding. Starczynowski:Kurome Therapeutics: Consultancy. Verma:Janssen: Research Funding; BMS: Research Funding; Celgene: Honoraria; Stelexis: Equity Ownership, Honoraria; Acceleron: Honoraria.

The activating mutation JAK2 V617F plays a central role in the pathogenesis of polycythemia vera, essential thrombocythemia, and primary myelofibrosis. Inhibition of JAK2 activity leads to growth inhibition and apoptosis in cells with mutated JAK2. However, the proapoptotic proteins involved in JAK2 inhibition-induced apoptosis remain unclear. In this study, we show that JAK2 inhibition-induced apoptosis correlated with up-regulation of the nonphosphorylated form of the BH3-only protein Bim in hematopoietic cell lines bearing JAK2 mutations. Knockdown of Bim dramatically inhibited apoptosis induced by JAK2 inhibition, which was reversed by the BH3 mimetic agent ABT-737. In addition, ABT-737 enhanced the apoptosis induced by JAK2 inhibition in JAK2 V617F+ HEL and SET-2 cells. The combination of JAK inhibitor I and ABT-737 reduced the number of erythroid colonies derived from CD34+ cells isolated from JAK2 V617F+ polycythemia vera patients more efficiently than either drug alone. These data suggest that Bim is a key effector molecule in JAK2 inhibition-induced apoptosis and that targeting this apoptotic pathway could be a novel therapeutic strategy for patients with activating JAK2 mutations.

Functionally critical decreases in levels or activity of the ETS family transcription factor PU.1 are present in approximately 2/3 of patients with acute myeloid leukemia (AML), across different AML subtypes (Sive, Leukemia 2016) including at the stem cell level (Steidl, Nat Genet 2006; Will, Nat Med 2015). Thus, targeting PU.1 could be an appealing option for treatment. As complete loss of PU.1 leads to stem cell failure (Iwasaki, Blood 2005), we hypothesized that PU.1 inhibition could eradicate leukemic cells harboring already low levels of PU.1, with modest effects on normal cells. We initially tested this hypothesis using 3 different shRNAs, and found that PU.1 inhibition led to a significant decrease in proliferation and clonogenicity, and increased apoptosis of mouse and human leukemic cell lines with low PU.1 levels, as well as the majority of primary human AML cells tested. We demonstrated that these effects were indeed due to decreased PU.1 levels by retroviral add-back experiments. The direct pharmacologic targeting of transcription factors has proven challenging in the past. Besides the core ETS binding motif (GGAA) in the DNA major groove, PU.1 binding to chromatin depends on additional minor groove contacts enriched for AT nucleotides upstream of the ETS motif, which determine selectivity for PU.1. Using an integrated screening strategy utilizing biosensor surface plasmon resonance, DNA footprinting, and cell-based dual-color PU.1 reporter assays, we developed novel small molecules of the heterocyclic diamidine family acting as first-in-class PU.1 inhibitors. Targeted occupancy by our compounds in the minor groove induces perturbations in DNA conformation that are transmitted to the PU.1 site in the major groove and thus inhibits PU.1 binding via an allosteric mechanism. Consistent with this, the inhibitory effects were selective for PU.1 versus other ETS transcription factors. Treatment with 3 different compounds led to cell growth inhibitory effect with respect to PU.1 level and preferentially affects PU.1low AML cells. Similarly to what we observed with shRNAs, treatment with our novel inhibitors led to decreased proliferation and colony forming capacity, increased apoptosis, and disrupted serial replating capacity of PU.1low AML cells and a majority of primary AML cell samples. Targeted ChIP and expression analysis showed that the compounds disrupt PU.1-promoter interaction and lead to downregulation of canonical PU.1 transcriptional targets in AML cells, confirming on-target activity in AML cells. Genome-wide analysis showed highly significant enrichment of known transcriptional targets of PU.1, and selectivity over genes regulated by other ETS family members. Comparison with published transcriptomic and PU.1 ChIP-seq data sets, as well as ARACNe analysis of the PU.1 regulon in primary AML cells, demonstrated that the inhibitors antagonize PU.1-regulated pathways at a genome-wide level. ChIP-seq performed in PU.1low AML cells confirmed a genome-wide decrease of PU.1 peaks after treatment and provides novel insight into the molecular mechanisms mediating the anti-leukemic effects of pharmacological PU.1 inhibition. To test the effects of PU.1 inhibition on normal hematopoiesis, we treated normal hematopoietic stem/progenitors cells (HSPC) in colony forming assays and saw decreased production of mature granulo-monocytic cells, consistent with PU.1's known role in this lineage. However, this effect was reversible upon drug removal, and serial replating capacity was not affected suggesting no significant effects on more immature HSPC. Congenic transplantation assays of treated normal bone marrow cells led to no change in myeloid and T-cells and only a modest decrease in B-cell numbers. Lastly, in vivo treatment with PU.1 inhibitors in mouse and human AML (xeno)transplantation models significantly decreased tumor burden and increased survival. To conclude, our study provides proof-of-principle for PU.1 inhibition as a novel therapeutic strategy in AML. Furthermore, we present the development of first-in-class PU.1 inhibitors acting via an allosteric minor groove-mediated mechanism. Our work shows that the specific pharmacological targeting of the DNA interaction of transcription factors such as PU.1 is feasible in principle, and may open the way for targeting of other transcription factors through minor groove-directed approaches. Disclosures Will: Novartis Pharmaceuticals: Consultancy, Research Funding. Steidl: Celgene: Consultancy; Aileron Therapeutics: Consultancy, Research Funding; Novartis: Research Funding; GlaxoSmithKline: Research Funding; Bayer Healthcare: Consultancy.

While the TP53 tumor suppressor gene is mutated in more than 50% of human tumors, in Acute Myeloid Leukemia (AML) TP53 mutations are rare, occurring in less than 10% of cases. Yet, functional inactivation of wild-type p53 due to non-mutational abnormalities occurs frequently in AML and other hematological malignancies. A major mechanism of p53 inactivation results from the overexpression of its endogenous inhibitors MDMX (also known as MDM4, HDMX, and HDM4) and MDM2 which are frequently overexpressed in various p53 wild-type human cancers, including AML. Strikingly, MDMX has been reported to be overexpressed in over 92% of AML cases, while MDM2 overexpression is less frequent. Pharmacological disruption of both these interactions has long been sought after as an attractive strategy to fully restore p53-dependent tumor suppressor activity in AML and other cancers with wild-type p53. Nonetheless, selective targeting of this pathway has thus far been limited to MDM2-only small molecule inhibitors which lack affinity for MDMX. ALRN-6924 is an optimized alpha helical p53 stapled peptide and first-in-class dual MDMX/MDM2 inhibitor which has recently entered phase I/II clinical testing (NCT02264613, NCT02909972) in solid tumors and lymphomas with, thus far, excellent tolerability and objective responses as single agent1. The goal of our study was to evaluate the molecular, cellular, and biochemical mechanisms of action of ALRN-6924 in AML. We used biochemical affinity studies as well as single molecule FISH and live single cell imaging to assess MDMX/MDM2 binding as well as p21 transactivation by p53 in response to ALRN-6924. Effects on cellular proliferation, apoptosis, DNA repair, cell cycle, clonogenic capacity, and serial replating were determined using AML cell lines and primary human AML patients' cells, including in leukemic stem (CD34+ CD38-) and progenitor (CD34+CD38+) cells. Genome-wide molecular effects were determined by RNA seq. P53 activity in a patient undergoing treatment with ALRN-6924 was measured by intracellular staining of p53 and its target gene p21 in CD34+ cells by flow cytometry. Furthermore, we evaluated ALRN-6924 activity in a xeno-transplantation model of human AML in NSG mice. We found that MDMX is significantly overexpressed in highly fractionated leukemic stem (Lin-CD34+CD38-CD90-) and progenitor (Lin-CD34+CD38+CD123+CD45+) cells in AML patients compared to identically sorted, age-matched healthy controls (p Our study provides insight into the molecular, cell biological, and in vivo effects of pharmacological dual MDMX/MDM2 inhibition in AML, which may have important implications for other MDMX/MDM2-related cancers. Our findings provide proof-of-concept for ALRN-6924 as a novel therapeutic in p53-wildtype AML, and provide a rationale for its further preclinical and clinical development in AML and other cancers. Furthermore, the success of targeting p53 raises the intriguing prospect that the same development path is possible for other helix-in-groove targets, and may thus pave the way for a new class of targeted therapeutics. 1 ASCO 2017 annual meeting: J Clin Oncol 35, 2017 (suppl; abstr 2505) Disclosures Guerlavais: Aileron: Employment. Annis: Aileron Therapeutics: Employment. Will: Novartis Pharmaceuticals: Consultancy, Research Funding. Aivado: Aileron: Employment. Steidl: Novartis: Research Funding; Celgene: Consultancy; Bayer Healthcare: Consultancy; GlaxoSmithKline: Research Funding; Aileron Therapeutics: Consultancy, Research Funding.

Myelodysplastic syndrome (MDS) & Acute Myeloid Leukemia (AML) arise from a population of aberrant hematopoietic stem cells (HSCs) that have been altered by multiple genetic & epigenetic alterations. A preliminary genomewide transcriptomic and epigenomic analysis of stem and progenitor cells from MDS and AML patients revealed STAT3 upregulation with corresponding loss of promoter methylation in diseased samples (Will et al, Blood, 2012). We now demonstrate that STAT3 is overexpressed in highly purified FACS sorted disease initiating populations in MDS/AML. STAT3 expression was determined to be significantly higher in phenotypic LT-HSC (CD34+/CD38-/CD90+/Lin -ve), ST-HSC (CD34+/CD38-/CD90-/Lin -ve) and GMP (CD34+/CD38+/CD123+/CD45Ra+/Lin -ve) from 10 diseased and 6 control sorted samples. Transcriptomic data from another cohort of CD34+ cells from 183 MDS patients found significantly increased expression of STAT3 in MDS samples when compared to healthy controls and importantly, found that increased STAT3 expression was predictive of significantly adverse prognosis (P value < 0.01, median survival of 2.6 years compared to 5.8 years for group with lower STAT3). Clinical correlations revealed that MDS patients with higher STAT3 expression in stem/progenitor cells were significantly more anemic (Mean Hgb 9.4gm/dl vs 10.2gm/dl, P=0.002) and had higher blast counts (Mean Blast Count 7.1% vs 5.3%, P=0.02), further demonstrating STAT3 to be an adverse prognostic marker in MDS. To functionally determine the role of STAT3 in MDS/AML pathogenesis, we used AZD9150, a Gen 2.5 Antisense Oligonucleotide (ASO) specific inhibitor of STAT3 (Hong et al, SciTM, 2015). This ASO has recently demonstrated safety and single-agent antitumor activity in patients with refractory lymphoma in a phase 1 dose-escalation study. We evaluated its efficacy in multiple leukemic cell lines and primary MDS/AML samples and used an inactive structural analogue oligonucleotide as control. AZD9150 treatment led to significantly decreased viability in numerous AML and MDS derived cell lines. (N=7, P < 0.01). Loss of viability was accompanied by dose dependent increase in apoptosis in leukemic cells. Specific inhibition of STAT3, but not STAT5, was seen in AZD9150 treated leukemic cells in luciferase reporter assays. AZD9150 treatment also led to decreased expression of important oncogenic downstream genes including not only STAT3 itself, but also IL8, IL6, CXCR2 and others. In vivo studies using a leukemia xenograft model in NSG mice demonstrated that treatment with AZD9150 at 50mg/kg/day led to significantly improved survival compared to control oligonucleotide (p-value: 0.004). It is generally challenging to show uptake of oligonucleotide-based therapeutics by primary cancer samples. Here, we treated healthy stem cells and AML primary patient samples with AZD 9150 and determined by intracellular flow cytometry that the ASO were readily taken up by primary patient- or cord blood-derived HSCs, myeloid progenitors and lymphocytes in a time and dose-dependent manner. We then treated a cohort of MDS and AML primary samples (N= 7) with AZD9150 or control and found qualitatively decreased leukemic colony growth, enhanced erythroid colony formation, and increased myeloid differentiation following STAT3 inhibition. In conclusion, we demonstrate that STAT3 is upregulated in highly purified stem and progenitor cells in MDS and AML and is an adverse prognostic marker. Importantly, a clinically applicable oligonucleotide inhibitor of STAT3 shows preclinical in vitro and in vivo efficacy in MDS and AML models, thereby providing a rationale for further development and clinical testing in MDS and AML. Supported by LLS Disclosures No relevant conflicts of interest to declare.

Introduction: While the presence of HIV portends a poor prognosis in many malignancies, its clinical association with MDS and AML has not been well studied, and is known primarily from small case reports and case series. We evaluated the prevalence and clinical course of unexplained cytopenias and clinically proven MDS/AML in a large cohort of HIV-positive (HIV+) patients (pts) seen at a single center, and compared their clinical course to MDS/AML in non-HIV (HIV-) pts. Methods: Data from 15,120 HIV+ pts seen between 1/1/1997 to 6/1/2016 was analyzed for hematologic parameters, HIV status (CD4 count, HIV viral load), chronic disease status and nutrient deficiencies. Anemia was defined as hemoglobin Results: Out of 13,277 HIV+ pts that had evaluable blood counts, 1075 pts (8.1%) had cytopenias affecting more than one lineage. 912 pts had bilineage cytopenias and 163 pts had a pancytopenia. This was significantly higher than the frequency of bi/trilineage cytopenias in a non-HIV elderly population that was recently analyzed; 8.1% in HIV+ pts vs 0.9% in non-HIV pts, p 250 and HIV viral load We also identified 10 MDS and 9 AML pts in our HIV+ cohort. Mean CD4 count for these cases was 340 cells/mm3, with only 4 pts < 200 cells/mm3. Median time to MDS or AML diagnosis from date of HIV diagnosis was 15.6 and 13.7 yrs respectively. All pts were receiving anti-retroviral therapy at the time of MDS or AML diagnosis. Only 3 MDS pts and 1 AML patient had a history of prior exposure to chemotherapy. Clinical characteristics of the HIV+ MDS pts (n=10) were compared to HIV- MDS pts (n=135) from the same institution diagnosed between 1997-2011. HIV+ MDS pts presented at a younger age (59.5 vs 70.6 yrs, p = 0.001) and also had significantly higher risk (Int-2/High) disease by IPSS (87.5% vs 17.8%, p Conclusions: Anemia occurs in most HIV pts at some point in their course, however, bicytopenia and pancytopenia are generally thought to be uncommon, especially in pts with well-controlled disease. We demonstrated in our cohort an unexpectedly high prevalence of unexplained bi/trilieneage cytopenias, not associated with advanced HIV, chronic disease or nutrient deficiencies. This finding suggests the presence of an independent, HIV-specific, mechanism of hematopoietic suppression, acting either at the level of the stem cell, or the bone marrow milieu. We also report on the largest cohort of HIV+ MDS/AML cases in the literature and demonstrate that HIV+ pts who develop MDS and AML do so at a younger age, present with more advanced disease that is resistant to standard therapy, and have worse overall survival than a control cohort of HIV- pts. These data present evidence that suggests that MDS should be considered as a diagnosis early in HIV. Appropriate risk stratification and new therapeutic approaches are needed. Kaplan-Meier curve comparing overall survival in HIV infected and non-HIV infected MDS patients. Kaplan-Meier curve comparing overall survival in HIV infected and non-HIV infected MDS patients. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

Activation of MAPK signaling is characteristic of many cancers, and approximately 15% of AML patients carry activating RAS mutations. The PI3K pathway is also constitutively activated in AML. Both pathways interact with each other extensively and provide compensatory signaling when one pathway is inhibited. It has been demonstrated in vitro that concurrent inhibition of both pathways is effective in blocking the proliferation of AML cell lines through an immediate decrease in pAkt and pErk, leading to inhibition of pS6 (Gritsman et al, J Clin Invest, 2014, 124(4):1794-1809). The combination of alpelisib (BYL-719), which inhibits PI3K (p110alpha), and binimetinib (MEK-162), which inhibits MEK (MAPK pathway), was tested in various tumor types in the Phase Ib Clinical Trial CMEK162x2109. We present here the clinical data and correlative phosphoprotein analysis of 6 patients with relapsed refractory RAS-mutated AML treated at Massachusetts General Hospital on Extension Arm 4A of this trial. Entry to this trial was restricted to adult patients with RAS-mutated AML previously treated with 1 or 2 prior chemotherapy regimens or for whom there was no known effective therapy. Patients received 200mg QD of alpelisib and 45mg BID of binimetinib concurrently and continuously with adjustments for toxicities. Hematologic toxicity could not be determined in these patients with active AML causing myelosuppression. There were responses in blood and/or bone marrow (BM) in 3 patients, but none made partial response criteria by IWG guidelines, mainly due to platelet counts 50% reduction in a third patient. A rise in absolute neutrophil count was seen in 3 patients, 2 from below 100/ul to above 500/ul, and in 1 from 5,320/ul to 12,797/ul during the first month. BM partial responses were seen in 2 patients, with blast percentages dropping from 39% to 10% in 1 patient and from 12% to 5% in another. We analyzed the effects of this drug combination on signaling targets by collecting peripheral blood samples on day 1 pre-treatment and at 6 and 24 hours post-initiation of treatment. In 4 patients, we performed analysis of the phosphorylation of Akt, ribosomal protein S6, Erk, STAT5, and STAT3 on timed peripheral blood samples by immunoblotting. In all 4 cases at baseline, pAkt, pErk, and pS6 were detectable, while pSTAT5 and pSTAT3 levels were variable. In 3 of 4 cases, we observed a transient decrease in pAkt at 6 hours, but then a rebound at 24 hours. In 3 of 4 cases, we observed a lack of sustained pS6 inhibition. We observed sustained pErk inhibition at 24 hours in 2 cases. One patient who had blood blast clearance, improvement in ANC, and a drop in bone marrow blasts, showed strong inhibition of pAkt and pERK at 6 hours, although pS6 did not decrease. In another case we performed phospho-flow cytometry on timed whole blood samples. We observed an increase in pAkt, pS6, and pErk from baseline at 24 hours, both in CD34+38+blasts, and in the primitive CD34+38- cells. This second patient showed no clinical benefit from the treatment in terms of blood blast count, ANC, or platelets. We also performed immunohistochemistry for pS6, pErk, pAkt, pSTAT5, pSTAT3, p-eIF4E, and Caspase 3 on BM sections obtained at diagnosis and at one month and two months post-initiation of treatment, when available. These studies generally showed either persistence or an increase in the pS6 and p-eIF4E signals, both indicators of mTORC1 activity, in post-treatment BM samples. Levels of pAkt, pErk, pSTAT5 and pSTAT3 were highly variable in the post-treatment bone marrow samples. In conclusion, we demonstrated some initial target inhibition with the concurrent use of alpelisib and binimetinib in a subset of patients. However, inhibition of late downstream targets was not sufficiently sustained to achieve consistent clinical benefit in our patients with RAS-mutated AML. While the strategy of concurrent inhibition of various critical signaling pathways remains interesting, sustained inhibition of downstream signaling may require a different dosing schedule of the two drugs. Given the incomplete inhibition of mTORC1 targets pS6 and p-eIF4E in most cases, the addition of a third agent to inhibit pathways causing cross-activation downstream of mTORC1 may be required. Disclosures Weinstock: Novartis: Consultancy, Research Funding. Fathi:Bexalata: Other: Advisory Board participation; Merck: Other: Advisory Board participation; Celgene: Consultancy, Research Funding; Seattle Genetics: Consultancy, Other: Advisory Board participation, Research Funding; Agios Pharmaceuticals: Other: Advisory Board participation.

Eltrombopag (EP) is a small-molecule, nonpeptide thrombopoietin receptor (TPO-R) agonist that has been approved recently for the treatment of thrombocytopenia in patients with chronic immune thrombocytopenic purpura. Prior studies have shown that EP stimulates megakaryopoiesis in BM cells from patients with acute myeloid leukemia and myelodysplastic syndrome, and the results also suggested that it may inhibit leukemia cell growth. In the present study, we studied the effects of EP on leukemia cell proliferation and the mechanism of its antiproliferative effects. We found that EP leads to a decreased cell division rate, a block in G1 phase of cell cycle, and increased differentiation in human and murine leukemia cells. Because EP is species specific in that it can only bind TPO-R in human and primate cells, these findings further suggested that the antileukemic effect is independent of TPO-R. We found that treatment with EP leads to a reduction in free intracellular iron in leukemic cells in a dose-dependent manner. Experimental increase of intracellular iron abrogated the antiproliferative and differentiation-inducing effects of EP, demonstrating that its antileukemic effects are mediated through modulation of intracellular iron content. Finally, determination of EP's antileukemic activity in vivo demonstrated its ability to prolong survival in 2 mouse models of leukemia.

Thrombocytopenia is a frequent symptom and clinical challenge in patients with myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). Eltrombopag is a small molecule thrombopoietin receptor agonist that might be a new option to treat thrombocytopenia in these diseases, provided that it does not stimulate malignant hematopoiesis. In this work, we studied the effects of Eltrombopag on proliferation, apoptosis, differentiation, colony formation, and malignant self-renewal of bone marrow mononuclear cells of patients with AML and MDS. Malignant bone marrow mononuclear cells did not show increased proliferation, or increased clonogenic capacity at concentrations of Eltrombopag ranging from 0.1 to 30 μg/mL. On the contrary, we observed a moderate, statistically nonsignificant (P = .18), decrease of numbers of malignant cells (mean, 56%; SD, 28%). Eltrombopag neither led to increased 5-bromo-2-deoxyuridine incorporation, decreased apoptosis, an increase of malignant self-renewal, nor enhanced in vivo engraftment in xenotransplantations. Furthermore, we found that Eltrombopag was capable of increasing megakaryocytic differentiation and formation of normal megakaryocytic colonies in patients with AML and MDS. These results provide a preclinical rationale for further testing of Eltrombopag for treatment of thrombocytopenia in AML and MDS.

Mutations in the isocitrate dehydrogenase 1 (IDH1) gene are known driver mutations in acute myeloid leukemia (AML) and other cancer types. Patient outcomes in AML have remained poor, especially for patients above 60 years of age who typically do not tolerate high dose chemotherapy and stem cell transplantation, leading to cure rates below 20%. The development of novel targeted therapies for defined AML subtypes is urgently desired. Inhibitors of mutants of the closely related IDH2 gene as well as IDH1 have recently been described and show promising pre-clinical and early phase clinical activity. However, the specific molecular and functional effects of IDH1 inhibitors in AML, including in primary patients' cells, have not been reported yet. Here, we report the development of novel allosteric inhibitors of mutant IDH1 for differentiation therapy of acute myeloid leukemia. A high-throughput biochemical screen targeting an IDH1 heterodimer composed of R132H and WT IDH1 led to the identification of a tetrahydropyrazolopyridine series of inhibitors. Structural and biochemical analyses revealed that these novel compounds bind to an allosteric site that does not contact any of the mutant residues in the enzymes active site and inhibit enzymatic turnover. The enzyme complex locked in the catalytically inactive conformation inhibits the production of the oncometabolite 2-hydroxyglutarate (2-HG). In biochemical studies, we observed potent inhibition of several different clinically relevant R132 mutants in the presence or absence of the cofactor NADPH, accompanied by significant decrease in H3K9me2 levels. Allosteric inhibitor treatment of primary AML patients' cells with different clinically relevant R132 mutants of IDH1 ex vivo uniformly led to a decrease in intracellular 2-HG, abrogation of the myeloid differentiation block, increased cell death and induction of differentiation both at the level of leukemic blasts and immature stem-like cells. Allosteric inhibition of IDH1 also led to a decrease in blasts in an in vivo xenotransplantation model. At the molecular level, enhanced reduced representation bisulfite sequencing showed that treatment with allosteric IDH1 inhibitors led to a significant reversal of the DNA cytosine hypermethylation pattern induced by mutant IDH1, accompanied by gene expression changes of key sets of genes and pathways, including "Cell Cycle", "G1/S transition", "Cellular growth and proliferation", and "Cell death and survival". Taken together, our findings provide novel insight into the cellular and molecular effects of inhibition of mutant IDH1 in primary AML patients' cells. Furthermore, our study provides proof-of-concept for the molecular and biological activity of novel allosteric inhibitors for targeting of different mutant forms of IDH1 in leukemia, and opens new avenues for future investigations with these and other allosteric inhibitors for targeting mutant IDH1 in leukemia and other cancers. Disclosures Gao: GlaxoSmithKline: Employment. Pietrak:GlaxoSmithKline: Employment. Rendina:GlaxoSmithKline: Employment. Rominger:GlaxoSmithKline: Employment. Quinn:GlaxoSmithKline: Employment. Smallwood:GlaxoSmithKline: Employment. Wiggall:GlaxoSmithKline: Employment. Reif:GlaxoSmithKline: Employment. Schmidt:GlaxoSmithKline: Employment. Qi:GlaxoSmithKline: Employment. Zhao:GlaxoSmithKline: Employment. Joberty:GlaxoSmithKline: Employment. Faelth-Savitski:GlaxoSmithKline: Employment. Bantscheff:GlaxoSmithKline: Employment. Drewes:GlaxoSmithKline: Employment. Duraiswami:GlaxoSmithKline: Employment. Brady:GlaxoSmithKline: Employment. Concha:GlaxoSmithKline: Employment. Adams:GlaxoSmithKline: Employment. Schwartz:GlaxoSmithKline: Employment. McCabe:GlaxoSmithKline: Employment.

Genomic studies have shown that human cancer is rarely associated with a complete loss of transcripts; instead, acquired DNA alterations often occur within the non-coding part of the genome, are enriched in gene-regulatory regions, and cause only moderate transcriptional changes. It is currently not well understood how such moderate gene expression changes impact normal tissue function and how they contribute to malignant transformation. Acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) develop through a multi-step transformation process originating in hematopoietic stem cells (HSCs) and mainly present in the elderly (median age of >65 years at diagnosis). Although loss or near-complete loss of the hematopoietic transcription factor PU.1 induces AML in mice, a similar degree of PU.1 impairment is exceedingly rare in human AML; yet moderate PU.1 inhibition is common in AML patients. At the example of the Ets-family transcription factor PU.1, which is indispensable for HSC function and the differentiation of cells within the myeloid as well as lymphoid lineages, we tested the hypothesis that even moderate gene expression alterations of key regulators can drive malignant transformation. We assessed the effects of minimal PU.1 inhibition on hematopoiesis in a novel mouse model that co-models the genomic context found in aging human individuals and patients with MDS/AML. Mice lacking Msh2, the key component of the MutSα and MutSβ complexes mediating DNA mismatch repair, accumulate elevated numbers of point mutations, in particular C/G>T/A transitions and small insertions/deletions resembling the mutation spectrum acquired in HSCs in aging human individuals and patients with MDS and AML. We crossed Msh2-/- mice with animals carrying a heterozygous deletion of an upstream regulatory element of PU.1 (UREΔ/+). UREΔ/+Msh2-/- mice exhibited a significant, but very modest reduction of PU.1 expression on average by 26-37% in fractionated hematopoietic multipotent stem and myeloid progenitor cells. Strikingly, this minimal reduction of PU.1 led to the emergence of an aggressive, transplantable AML in more than two thirds of UREΔ/+Msh2-/- mice which was never observed in URE+/+Msh2-/- mice. Overt leukemia was preceded by a preleukemic phase hallmarked by an expanded population of multipotent murine hematopoietic stem cell enriched cells (HSPCs) that was myeloid-biased and less quiescent than their wild type counterpart. Longitudinal monitoring of preleukemic UREΔ/+Msh2-/- mice revealed a progressive increase in immature myeloid cells along with a gradual decrease in mature myeloid cells, as well as expansion of phenotypic HSPC compartments and multi-lineage dysplasia resembling human MDS. AML progression was accompanied by additional inhibition of a PU.1-cooperating factor, interferon responsive factor 8 (Irf8). Irf8 expression restoration rescued impaired expression of genes harboring PU.1/IRF consensus binding sites, led to the loss of aberrant self-renewal, promoted myeloid differentiation, and induced apoptosis in leukemic UREhetMsh2-/- cells demonstrating that Irf8 impairment functionally cooperates with minimally reduced PU.1 expression in our model. We also found evidence of disease-relevant joint PU.1/IRF8 inhibition in human myeloid leukemogenesis: (1) patients with MDS with a higher risk for the progression to AML had lower IRF8 levels; (2) lower IRF8 expression was detected specifically in AML patients with reduced PU.1 levels; (3) restoration of IRF8 expression induced differentiation in IRF8 low expressing AML cells, and (4) a positive correlation of PU.1 and IRF8 expression was found in human leukemia stem cells, but not in healthy HSCs. Strikingly, comparative pathway analysis revealed a genome-wide molecular resemblance of preleukemic and leukemic UREΔ/+Msh2-/- mice with gene expression profiles from human MDS and AML patients, respectively. Our study demonstrates that minimal reduction of a key lineage-specific transcription factor that commonly occurs in human disease is sufficient to initiate cancer development and provides mechanistic insight into the formation and progression of preleukemic stem cells in MDS and AML. Disclosures No relevant conflicts of interest to declare.

Hematopoietic stem cells (HSCs) maintain lifelong blood homeostasis. While many of the cell-intrinsic mechanisms regulating HSC function at steady state have been well characterized, the role of inflammatory cytokines and other environmental factors in tailoring blood production following physiological insults has become a topic of emerging interest. The cytokine interleukin-1 (IL-1) is a prototypical pro-inflammatory cytokine that plays a key role in host inflammatory responses to injury and infection, and is associated with elevated myeloid cell production. Importantly, IL-1 also drives a wide range of chronic inflammatory conditions such as diabetes, obesity, and arthritis that are often characterized by deregulated blood homeostasis. Here, we show at single-cell resolution using continuous tracking technology that IL-1 drives accelerated HSC cell division kinetics and myeloid differentiation via the rapid activation of a precocious PU.1-dependent myeloid gene program. Activation of this program requires direct IL-1R signaling and subsequent activation of IKK kinases, and instructively primes HSCs to adopt a myeloid fate. Strikingly, we demonstrate that IL-1 produced by myeloid cells and endothelial cells of the bone marrow (BM) niche exerts similar effects in vivo, and is required for efficient myeloid recovery following acute challenges such as transplantation or myeloablation. On the other hand, we find that chronic IL-1 exposure substantially remodels HSC blood output, resulting in myeloid overproduction and expansion of myeloid-biased multipotent progenitor (MPP) compartments at the expense of lymphoid and erythroid lineages. Critically, chronic IL-1 erodes HSC self-renewal, significantly impairing their regenerative capacity following transplantation. On the other hand, chronically exposed HSCs recover their function upon IL-1 withdrawal. Collectively, these findings identify IL-1 as a critical regulator of HSC fate and lineage specification via activation of a PU.1 circuit. They also demonstrate a role for IL-1 as a double-edged sword in HSC biology, promoting HSC regeneration in response to acute insults while severely disrupting HSC self-renewal and lineage output during chronic exposure, hence identifying IL-1 as an important and therapeutically targetable factor underwriting myeloid overproduction and other deregulations that contribute to the pathogenesis of a variety of chronic inflammatory diseases and blood disorders. Disclosures No relevant conflicts of interest to declare.

The marrow microenvironment contributes to the pathogenesis of ineffective hematopoiesis in Myelodysplastic Syndromes (MDS). Since mutations and cytogenetic alterations seen in the hematopoietic compartment are generally not present in marrow stromal cells, we hypothesized that epigenetic alterations may be responsible for altered stroma function in MDS. To elucidate the epigenome of MDS microenvironment, we used the HELP (HpaII tiny fragment Enrichment by Ligation-mediated PCR) assay to study cytosine methylation patterns of 50,000 CpGs loci covering 14,000 promoters in primary stromal cells. Global DNA methylation of 6 MDS marrow-derived stroma was analyzed by HELP assay and compared to 3 healthy controls. MDS stroma showed aberrant hypermethylation compared with controls (3626 hypermethylated vs 306 hypomethylated loci in untreated MDS stromal cells) that preferentially occurred outside of CpG islands (Test of Proportions, P value After demonstrating aberrant hypermethylation in MDS stroma, we next examined the functional role of these alterations by treatment with DNMT inhibitors. 5-Aza treatment of MDS stroma enhanced hematopoietic activity and erythroid differentiation from co-cultured healthy CD34+ cells. Significantly increased erythroid differentiation was seen in CD34+ cells cocultured with aza pretreated MDS stroma as shown by their glycophorin A positivity. Conclusions: These results reveal widespread aberrant epigenetic changes in the MDS marrow microenvironmentthat affect many important pathways. A recent report had shown that activation of beta-catenin in the bone marrow niche can trigger MDS/AML in vivo. Our results demonstrate that epigenetic silencing of WNT antagonists can potentially lead to activation of the WNT pathway in MDS stroma. Finally, we demonstrate that DNA methyl transferase inhibitors alter the epigenomic profiles of stromal cells and it raises the possibility that epigenetic modulation of the stroma by these drugs contributing to their therapeutic efficacy. Disclosures No relevant conflicts of interest to declare.

Acute Myeloid Leukemia (AML) and Myelodysplastic syndrome (MDS) arise from accumulation of multiple stepwise genetic and epigenetic changes in hematopoietic stem cells (HSC) and/or committed progenitors. A series of transforming events can initially give rise to pre-leukemia stem cells (pre-LSC) as well as fully transformed leukemia stem cells (LSC), both of which need to be targeted in strategies aimed at curing these diseases. We conducted parallel transcriptional and epigenetic analysis of highly fractionated stem and progenitor populations in individual patients of MDS and identified STAT3 upregulation in MDS HSCs. qRTPCR in an independent set of sorted MDS/AML HSCs (Lineage-negative, CD34+, CD38-) confirmed the significant increase in STAT3 expression in 60% of cases when compared to healthy controls. We further analyzed gene expression profiles of CD34+ cells from 183 MDS patients and found significant increased expression of STAT3 in MDS when compared to healthy controls (FDR Next, we studied the functional role of STAT3 by using Pyrimethamine as a clinically relevant inhibitor. Pyrimethamine is an FDA approved antifolate compound that was found to be a specific inhibitor of STAT3 activity in an initial screen of 1120 compounds. 3D modeling studies reveal that Pyrimethamine can occupy the SH2 domain of STAT3 protein with high avidity. It shows half-maximal activity (EC50) for STAT3 inhibition at approximately 1.5 μM, well within the plasma concentrations found with clinical use. We assessed the effect of Pyrimethamine on cellular proliferation of multiple leukemic cell lines (KG-1, KT-1 and CMK), and observed that these were significantly inhibited at 120 hours of exposure in a dose dependent fashion (t test, p value In summary, we have found significant demethylation and increased expression of STAT3 in sorted HSCs from AML and MDS patients. High STAT3 expression is a marker of adverse prognosis in a large cohort of MDS patients. In vitro studies show that Pyrimethamine can inhibit STAT3 activation and is able to inhibit proliferation by inducing apoptosis in leukemic cells. A phase II study evaluating Pyrimethamine for the treatment of intermediate/high-risk MDS patients that have relapsed or are refractory to azanucleoside therapy has been initiated. Disclosures Brown: Sanofi, Onyx, Vertex, Novartis, Boehringer, GSK, Roche/Genentech, Emergent, Morphosys, Celgene, Janssen, Pharmacyclics, Gilead: Consultancy.

Cell cycling is a tightly regulated process involving the structured expression modulation of various regulatory genes. This process is crucial for the maintenance of cell survival/proliferation in both normal and malignant hematopoietic cells. We have previously described the highly expressed CIITA-BX648577 gene fusion (Steidl C. et al., Nature 2011), involving the novel gene locus BX648577/FLJ27352 /hypothetical LOC 145788/C15orf65 and the Class II Transactivator (CIITA) in the Hodgkin’s lymphoma cell line KM-H2. While CIITA is well known to be involved in the regulation of immune responses, specifically through regulation of the Major Histocompatibility Complex (MHC)-II, nothing is known about the expression and function of the BX648577 locus. The objective of the current study was to (I) study RNA and protein expression of the putative full length gene encoded by the BX648577 (TIHL) gene locus, and (II) study its biological function in normal and malignant hematopoietic cells, including its effects on cell proliferation, clonogenicity and cell death. We detected robust endogenous TIHL RNA and protein expression in a variety of healthy and malignant hematopoietic cell types using quantitative real-time polymerase chain reaction (qRT-PCR) and western blot analysis using a TIHL-specific antibody. At the functional level, we found that ectopic expression of the highly conserved full length TIHL protein in human NB4 leukemia cells and murine hematopoietic progenitor HPC-7 cells leads to enhanced clonogenicity and increased proliferative capacity with significant increases in the percentage of cells in S-phase of the cell cycle. Furthermore, we observed more aggressive leukemia and decreased survival of NSG mice following retro-orbital transplantation of TIHL-expressing compared to empty control-expressing NB4 cells. Interestingly, although we did not observe a change in the rate of cell proliferation or colony forming ability following TIHL overexpression in the ATRA-resistant cell line NB4.306, there was a significant alteration in its cell cycle distribution, with an increase in the fraction of cells in S-phase. The increase in S-phase cells was confirmed by 5-ethynyl-2’-deoxyuridine (EdU) incorporation assays and flow cytometry. Similarly, knockdown of TIHL using 2 independent lentiviral shRNAs, led to a significant decrease in the growth of both NB4 and acute myeloid leukemia KG1a cells in both suspension cultures and semi-solid media. Although we observed slightly increased apoptosis upon TIHL downregulation, the changes could be more significantly attributed to a decrease in the percentage of cells in S-phase within 2-3 days after transduction with the lentiviral shRNAs. Finally, in silico analysis of the TIHL promoter identified various predicted transcriptional regulators of TIHL, the majority of which are cell cycle specific transcription factors including Nuclear Receptor Subfamily 5 Group A Member 1 (NR5A1), the ets domain transcription factors ELF5 and ELF1, and Glioma-Associated Oncogene Homolog 1 (GLI-1). Our findings thus far strongly support a novel role for TIHL in cell cycle regulation/modulation in both normal and malignant hematopoiesis. Future directions include gene expression studies to identify downstream targets of TIHL following overexpression and knockdown and co-immunoprecipitation coupled to mass spectrometry analysis will be used to identify direct interacting protein partners of this novel gene. Disclosures: No relevant conflicts of interest to declare.

Acute myeloid leukemia (AML) is an aggressive disease associated with poor clinical outcome. Less than one third of patients achieve durable remission with current treatment regimens, and prognostication and risk stratification are challenging. We have recently reported that the non-clustered homeobox gene, H2.0-like homeobox (HLX), is 2 to 16 fold overexpressed in more than 80% of patients with AML, across all major disease subtypes, and higher levels of HLX are associated with poor overall survival in AML. Inhibition of HLX in both murine and human AML cells has a significant anti-leukemic and differentiation-inducing effect suggesting HLX and its downstream targets as novel therapeutic targets in AML. In order to better understand the role of Hlx at the stem cell level and in myeloid differentiation in vivo, we generated knock-in mice conditionally overexpressing Hlx from the Rosa26 locus and bred them to mice that bear Cre recombinase under the control of the pIpC-inducible, hematopoietic specific promoter, Mx1. Animals overexpressing HLX exhibit elevated WBC counts and abnormal myeloid cells in the peripheral blood. Analysis of the bone marrow reveals expansion of the granulocyte-macrophage progenitor population (lin- ckit+ cd34+ CD16/32high) and expansion of immature myelocytes (ckit+ cd34+ CD16/32high Gr1int). Hlx knock-in bone marrow cells, and specifically immature granulocyte precursors, exhibit enhanced serial clonogenicity in methylcellulose colony assays, and a differentiation block and maintenance of immaturity in response to GM-CSF. Internal tandem duplications of FLT3 (FLT3-ITD) are seen in approximately 30% of all AML patients, and frequently co-occur with elevated HLX levels. Correlative analyses showed that AML patients with mutant FLT3 and low HLX have overall survival similar to WT FLT3 patients, and survive significantly longer than patients with mutant FLT3 and high HLX (p=0.005), demonstrating that FLT3 mutations confer poor prognosis only if HLX is highly expressed, and suggesting that HLX and mutant FLT3 functionally cooperate. We retrovirally co-expressed HLX and FLT3-ITD, or FLT3-ITD alone (plus an empty control), in primary Lin-Kit+cells and transplanted them into congenic recipient animals. Four weeks after transplantation, donor chimerism was 4-fold increased on average in the peripheral blood (PB) and bone marrow (BM), and by 12 weeks post-transplantation mice expressing FLT3-ITD and HLX developed AML with large numbers of leukemic blasts in the peripheral blood and bone marrow. We then crossed our new Hlx knock-in mouse model with previously generated FLT3-ITD knock-in mice. Strikingly, heterozygous double-transgenic mice expressing both the knock-in FLT3-ITD mutation and HLX develop acute myeloid leukemia after a latency of 2 months. Morphological and flow cytometric analysis revealed large numbers of blasts circulating in the peripheral blood and replacing the marrow, as well as substantial leukemic infiltrates in the spleen and liver. Our studies reveal a critical role for HLX in conferring a differentiation block and increased clonogenicity at the pre-leukemic stem and progenitor cell level in a genetic in vivo model. Furthermore, a novel compound knock-in mouse model of Hlx overexpression and FLT3-ITD demonstrates that Hlx can initiate AML in cooperation with FLT3-ITD in vivo. Disclosures: No relevant conflicts of interest to declare.

Abstract 651 Acute myeloid leukemia (AML) is an aggressive disease which is associated with poor clinical outcome. Less than one third of patients achieve durable remission with current treatment regimens, and prognostication and risk stratification are challenging. Identification and functional studies of genes and pathways which regulate leukemic transformation and maintenance is instrumental to understanding the pathogenesis of AML and for development of novel therapeutic strategies. Several members of the Hox (class I homeobox) family of transcription factors have been implicated in the regulation of normal hematopoiesis and leukemogenesis. Less is known about the role of non-clustered (class II) homeobox genes. We found that a new non-clustered homeobox gene, H2.0-like homeobox (HLX), regulates early hematopoiesis and promotes AML in mice and humans. HLX is 2 to 16 fold overexpressed in more than 80% of patients with AML, across all major disease subtypes. Higher levels of HLX are associated with poor overall survival in 3 different, large cohorts of AML patients (N=601, p=2.3×10−6), and HLX holds up as an independent prognostic factor in a multivariate analysis. ShRNA-mediated inhibition of HLX in both murine and human AML cells significantly inhibits leukemic growth and clonogenic capacity, and overcomes the differentiation block of AML cells. When we analyzed pre-leukemic hematopoietic stem and progenitor cells (HSPC) in a PU.1 URED/D AML mouse model, we found a 4-fold elevation of Hlx, suggesting that Hlx is involved in malignant transformation. Overexpression of HLX in wildtype HSPC in a competitive, congenic transplantation model led to near complete depletion of long-term HSC and 16-fold enrichment of myeloid progenitors with a surface phenotype slightly past the GMP stage (CD45+Kit−CD34−CD44highCD49bhighCD11bmid). Overexpression of HLX in HSPC in vitro led to a myeloid differentiation block and to formation of aberrant, CD34−Kit− progenitors with unlimited serial clonogenicity. The mechanism of action of Hlx is so far unknown. The presence of a C-terminal homeobox domain suggests Hlx may directly interact with DNA, however, no studies have shown DNA binding by Hlx or identified direct Hlx target genes. We find that mutation of only two residues of the Hlx homeodomain is sufficient to completely abrogate the differentiation block induced by HLX overexpression in HSPC, indicating Hlx is acting through the DNA-binding ability of its homeodomain. Furthermore, we have now identified direct HLX target genes in both HSPC and AML cells using a combination of expression microarrays and chromatin-immunoprecipitation (chIP). We find that HLX regulates a set of genes which mediate its leukemia-promoting functions, such as BTG1, and we have used chIP to identify a subset of these genes, including PAK1, that are direct targets of HLX. Internal tandem duplications of FLT3 (FLT3-ITD) are seen in approximately 25% of all AML patients, and confer a poor prognosis. Correlative analyses showed that AML patients with mutant FLT3 and low HLX have overall survival similar to WT FLT3 patients, and survive significantly longer than patients with mutant FLT3 and high HLX (p=0.005), demonstrating that FLT3 mutations confer poor prognosis only if HLX is highly expressed, and suggesting that HLX and mutant FLT3 functionally cooperate. We find that co-expression of HLX and FLT3-ITD leads to dramatically enhanced cytokine independent growth and clonogenicity of 32D cells as well as primary murine HSPC in vitro. When we retrovirally co-expressed HLX and FLT3-ITD, or FLT3-ITD alone (plus an empty control), in primary Lin−Kit+ cells and transplanted them into congenic recipient animals, we found that four weeks after transplantation donor chimerism was 4-fold increased on average in the peripheral blood (PB) and bone marrow (BM), and by 12 weeks post-transplantation mice expressing FLT3-ITD and HLX developed AML with large numbers of leukemic blasts in the PB and BM. We have generated knock-in mice conditionally overexpressing Hlx from the Rosa26 locus and ongoing studies include crossing these mice into FLT3-ITD knock-in animals. In summary, our studies have identified HLX as a novel key transcription factor involved in the regulation of early hematopoiesis and AML pathogenesis, and suggest HLX and downstream pathways as promising new therapeutic targets in AML. Disclosures: No relevant conflicts of interest to declare.

Expression of the transcription factor PU.1 is dependent on a highly conserved upstream regulatory element (URE) located 14 kilobases upstream of the transcriptional start site. The proximal promoter of PU.1 alone, without this enhancer, displays 100-fold reduced promoter activity and is not capable of driving PU.1 expression in transgenic mice. Targeted disruption of the URE reduces PU.1 expression by 80% and leads to AML in mice. However, the mechanisms mediating the long-range regulatory function of the URE are largely unknown. Here, we identified a binding site of the chromatin-remodeling special AT-rich sequence-binding protein 1 (SATB1). Utilizing EMSA and ChIP, we found that SATB1 binds to the URE in myeloid U937 cells. Luciferase assays demonstrated 7-fold reduced reporter activity upon disruption of the SATB1 site. Lentiviral overexpression of SATB1 in primary murine Lin−, Kit+ progenitor cells led to an upregulation of PU.1 expression. We did not observe this effect in URE−/− progenitors indicating that the PU.1 regulatory function of SATB1 is mediated by the URE. Inhibition of SATB1 by siRNA in U937 cells led to a decrease in PU.1 expression. This inhibitory effect was not seen in myeloid URE−/− cells. To address the question at which stages during myeloid development SATB1 regulates PU.1, we sorted KSL-HSC, CMP, GMP, and MEP of SATB1 knockout mice and determined PU.1 expression levels. Interestingly, PU.1 expression was 88% and 80% reduced in SATB1−/− GMP and MEP, while KSL-HSC and CMP did not show significantly changed PU.1 levels. This finding indicates a stage-specific regulatory function of SATB1 during myelopoiesis. When we sequenced the URE in human individuals with AML we identified a SNP that abates binding of SATB1 and leads to decreased PU.1 expression in GMP and MEP. While the overall frequency of the homozygous SNP was not significantly changed in AML patients compared with healthy controls, this SNP was 2.3-fold more frequent in patients with AML with complex karyotype than in AML with normal karyotype (p

The transcription factor Nuclear-Factor-Erythroid2 (NF-E2) is overexpressed in the vast majority of patients with polycythemia vera (PV). PV affects precisely those lineages expressing NF-E2: hematopoietic precursors, erythroid, megakaryocytic and granulocytic cells. In murine models, NF-E2 overexpression increases proliferation and promotes cellular viability in the absence of EPO. EPO-independent growth is a hallmark of PV. We therefore hypothesized that NF-E2 overexpression contributes to erythrocytosis, the pathognomonic symptom of PV. Consequently, we investigated the effect of NF-E2 overexpression in healthy CD34+ cells. NF-E2 overexpression lead to a delay in erythroid differentiation, manifested by a belated appearance of GlycophorinA-positive mature erythroid cells. Differentiation delay was similarly observed in primary PV patient erythroid cultures compared to healthy controls. Protracted differentiation lead to a significant increase in the accumulated number of erythroid cells both in PV cultures and in CD34+ cells overexpressing NF-E2. Similarly, NF-E2 overexpression altered erythroid colony formation, leading to an increase in BFU-E formation. These data argue that NF-E2 overexpression delays the early phase of erythroid differentiation, resulting in an expansion of erythroid progenitors, thereby increasing the number of erythrocytes derived from one CD34+ cell. These data propose a role for NF-E2 in mediating the erythrocytosis of PV.

Despite recent advances in characterizing molecular markers for the diagnosis of polcycythemia vera (PV), the aberrations leading to disease development remain unknown. We therefore used expression profiling to identify candidate genes involved in the pathophysiology of PV. RNA from purified granulocytes of 40 PV patients was analyzed by hybridizing individual samples to a pool of 50 healthy controls. Of the 7,496 genes represented in the cDNA array, 253 were upregulated more than 1.5 fold in PV compared to healthy controls (p< 0.01, FDR corrected). Promoters for 26 of the 253 genes overexpressed in PV are regulated by members of the Sp1 family of transcription factors. We have therefore hypothesized that altered activity of one or several Sp1-like transcription factors may contribute to the molecular etiology of PV. Here we report that one of the Sp1 target genes identified, the transcription factor NF-E2, is overexpressed in 37 of the 40 PV patients (92.5%) assayed by microarray. NF-E2 overexpression was confirmed by Northern Blot and quantitative RT-PCR analysis. Transcription factor overexpression varies from 2.3 to 40 fold, with a median increase of 7 fold in PV patients compared to healthy controls. The NF-E2 protein is readily detected in PV granulocytes by Western Blot whereas it is undetectable in healthy control cells. Immunohistochemistry revealed that in PV bone marrow, NF-E2 is overexpressed in megakaryocytes as well as erythroid and granulocytic precursors. Several published observations suggest that NF-E2 is an exceptionally promising candidate in the molecular etiology of PV. Firstly, the transcription factor is expressed in hematopoietic precursors as well as in erythroid, megakaryocytic and granulocytic cells, those lineages affected in PV. Secondly, Sayer et al. have shown that overexpression of NF-E2 in fetal liver cells leads to the development of Epo-independent erythroid colonies, analogous to the endogenous erythroid colonies (EECs) observed in PV patients. Furthermore, ectopic expression of NF-E2 results in the spontaneous emergence of morphologically mature erythroid cells in the absence of Epo and can reprogram monocytic cells towards erythroid and megakaryocytic differentiation. These data support the hypothesis that the concentration of an individual transcription factor can control lineage commitment. We thus propose that in PV patients elevated concentrations of NF-E2 alter the physiological transcription factor balance leading to an overproduction of erythroid and, in select patients, megakaryocytic cells/platelets. In this model the level of NF-E2 overexpression determines both the severity of erythrocytosis and the concurrent presence or absence of thrombocytosis.