Primary AML Cells Research Articles

TJC4, a Differentiated Anti-CD47 Antibody with Novel Epitope and RBC Sparing Properties

Introduction Tumor cells overexpress CD47 which engages signal-regulatory protein (SIRPa) on macrophages to deliver a "do not eat" signal to avoid being phagocytosed. Blocking CD47 using SIRPa-Fc or anti-CD47 antibodies (Ab) has emerged as a promising strategy to neutralize CD47 and promote tumor eradication. However, targeting CD47 led to significant anemia and thrombocytopenia in both pre-clinical studies and phase I trials as CD47 is also expressed on normal red blood cells (RBCs) and platelets. I-Mab has developed a novel CD47 antibody, TJC4 also known as TJ011133, which was endowed with an RBC sparing property and unique binding epitope, may have better safety profile based on the pre-clinical data. Methods A naïve human single chain variable fragment (ScFv) library was subjected for the binders to human CD47-extracellular domain (ECD). All the binders with unique sequences were converted to full antibodies and screened against human RBCs and tumor cells, leading to the discovery of TJC4. A series of head to head experiments have been performed with other CD47 antibodies to compare the in vitro RBC binding and hemagglutination, ability to block the CD47-SIRPa interaction and enhance the macrophage mediated phagocytosis of tumor cells. Different in vivo tumor models were employed to evaluate the anti-tumor efficacy of TJC4 either by mono or combination treatment. In addition, a comprehensive analysis of the hematological parameters was assessed in cynomolgus monkeys receiving a single intravenous infusion or weekly repeated injections. To explore the underlined mechanism of the RBC sparing properties of TJC4, the binding pose and epitope were identified by X-ray crystallography and the influence of CD47 glycosylation in RBCs were further examined. Results TJC4 is a fully human anti-CD47 IgG4 antibody that shares a similar binding affinity to human and cynomolgus monkey CD47. Like other anti-CD47 antibodies, TJC4 blocks the interaction of CD47 and SIRPa, leading to the enhanced macrophage phagocytosis of various CD47+ tumor cell lines and primary AML cells. Mono-treatment of TJC4 completely eradicated tumor cells in a Raji cell xenograft model and significantly extended the overall survival of treated mice in an AML model. When combined with Rituximab, TJC4 showed a superior efficacy in a DLBCL model over the mono-treatment group. TJC4 has unique RBC sparing properties as evident by the negligible binding to healthy human RBCs and platelets respectively. Single dose or repeat dose treatment of TJC4 minimally and transiently impacts RBCs in cynomolgus monkeys and no other safety findings were observed up to the highest dose (100 mg/kg). No impact was observed in platelets. The unique functional properties of TJC4 can be explained in part by its structure when in complex with CD47, which reveals an almost straight head-to-head binding and a novel conformational epitope that is distinct from other CD47 antibodies. Upon the structural analysis of the binding epitope, we identified a potential N-linked glycosylation site located nearby the two critical epitopes on the CD47 protein. Due to the nature of the high glycosylation degree of membrane proteins by RBCs, the N-linked glycan is hypothesized to function as a "shield" to block the exposure of the epitopes and prevent the TJC4 binding to human RBCs. This hypothesis is validated by the restoration of TJC4 binding to the deglycosylated RBCs after the PNGase treatment. Conclusion In summary, TJC4 is a next generation therapeutic anti-CD47 antibody that is devoid of the hematological liabilities while maintaining anti-tumor efficacy. These attributes of TJC4 differentiate it from other CD47 targeting agents currently in clinical evaluation. Disclosures Meng: I-Mab Biopharma: Employment, Equity Ownership. Wang:I-Mab Biopharma: Employment, Equity Ownership. Guo:I-Mab Biopharma: Employment, Equity Ownership. Cao:I-Mab Biopharma: Employment, Equity Ownership. Shen:I-Mab Biopharma: Employment, Equity Ownership.

TP53 Deficient/Mutant AMLs Are Resistant to Individual BH3 Mimetics: High Efficacy of Combined Inhibition of Bcl-2 and Mcl-1

In spite of recent progress in AML therapy, the outcomes of TP53 mutated AML remain extremely poor. The FDA-approved combination of Bcl-2 inhibition by venetoclax (VEN) with hypomethylating agents is resulting in CR/CRi rates of 70-95% and good tolerability in elderly AML patients. However, patients with TP53 mutations achieve lower response rates (CR/CRi 47%) (DiNardo CD et al., Lancet Oncol 2018; DiNardo CD et al., Blood 2019) and invariably relapse. Combined Bcl-2 inhibition and p53 activation is synthetic lethal in TP53 wild-type AML in part through targeting Mcl-1 (Pan R et al., Cancer Cell 2016). In TP53 deficient/mutant AML, direct targeting of Mcl-1 may partially compensate for the TP53 defect. We therefore postulate that combined inhibition of Mcl-1 and Bcl-2 effectively induces apoptosis in TP53 deficient/mutant AML cells. Reverse phase protein array analysis of a large cohort of newly diagnosed AML patients (n=511) enabled us to stratify patients into various prognostic groups based on p53 pathway protein expression, which included 8 core proteins: TP53, TP53pS15, MDM2, MDM4, TRIM24, SFN, IRS1.pS1101, and YWHAZ. The group with p53 pathway dysfunction, defined by high p53 protein levels, is characterized by poor outcomes (Quintas-Cardama A et al., Leukemia 2017). This group, encompassing both TP53 wild-type and TP53 mutations, had significantly lower expression of Bax (p=0.0007), the chief executioner of intrinsic apoptosis. A survey of Bcl-2 family proteins in TP53 wild-type and mutant AML showed that only Bax was significantly lower in patients with TP53 mutations (p=0.0498). We next investigated the roles of p53 in response to BH3 mimetics in TP53 wild-type and knockdown (KD) OCI-AML3 cells and in TP53 wild-type and mutant Molm13 cells, generated by long-term exposure of TP53 wild-type Molm13 cells to RG7388 (idasanutlin). Western blot analysis showed that Bax protein was consistently decreased in both TP53 KD and mutant cell lines compared to their respective controls. p53 KD OCI-AML3 and TP53 mutant Molm13 cells exhibited decreased sensitivity not only to VEN, but also to Mcl-1 inhibitor AZD5991 compared to OCI-AML3 vector control and Molm13 parental cells, respectively. To investigate if combined inhibition of Bcl-2 and Mcl-1 could counter-balance the loss of TP53 activity, p53 KD and TP53 mutated AML cells were treated with VEN and AZD5991: like in their respective control cells, the combination synergistically induced cell death in these cells (Fig. 1). The EC50 levels of VEN required for the described synergism in OCI-AML3 cells can easily be reached/exceeded clinically. The combination was also synergistic in leukemia cell lines lacking wild-type TP53 such as KG-1 and U937. Importantly, the combined inhibition was more effective than each single agent in primary AML cells and stem/progenitor cells lacking wild-type TP53 due to deletion of chromosome 17 or mutations in TP53 gene. Conclusion: AML cells deficient in functional TP53 have decreased Bax protein expression, increased apoptotic threshold and are more resistant to individual BH3 mimetics. Combined inhibition of Bcl-2 and Mcl-1 is highly synergistic in p53 deficient/mutant AML. Disclosures Carter: Amgen: Research Funding; AstraZeneca: Research Funding; Ascentage: Research Funding. Cidado:AstraZeneca: Employment. Drew:AstraZeneca: Employment. Andreeff:BiolineRx: Membership on an entity's Board of Directors or advisory committees; CLL Foundation: Membership on an entity's Board of Directors or advisory committees; NCI-RDCRN (Rare Disease Cliln Network): Membership on an entity's Board of Directors or advisory committees; Leukemia Lymphoma Society: Membership on an entity's Board of Directors or advisory committees; German Research Council: Membership on an entity's Board of Directors or advisory committees; NCI-CTEP: Membership on an entity's Board of Directors or advisory committees; Cancer UK: Membership on an entity's Board of Directors or advisory committees; Center for Drug Research & Development: Membership on an entity's Board of Directors or advisory committees; NIH/NCI: Research Funding; CPRIT: Research Funding; Breast Cancer Research Foundation: Research Funding; Oncolyze: Equity Ownership; Oncoceutics: Equity Ownership; Senti Bio: Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Eutropics: Equity Ownership; Aptose: Equity Ownership; Reata: Equity Ownership; 6 Dimensions Capital: Consultancy; AstaZeneca: Consultancy; Daiichi Sankyo, Inc.: Consultancy, Patents & Royalties: Patents licensed, royalty bearing, Research Funding; Jazz Pharmaceuticals: Consultancy; Celgene: Consultancy; Amgen: Consultancy.

Development of Predictive Biomarker and Optimal Treatment Strategy with FLT3 Inhibitors in Acute Myeloid Leukemia

【Background】Recently, FLT3 inhibitors are approved for FLT3 mutated AML patients and other FLT3 inhibitors are also in clinical development. Since the target selectivity and the inhibitory activity of FLT3 inhibitors are varied, it is required to establish a therapeutic strategy in consideration of their characteristics. Most FLT3 mutated AML cells co-express wild type (Wt)-FLT3, and we previously demonstrated that FLT3 ligand (FL) stimulation attenuated the inhibitory effect of FLT3 inhibitors through the activation of Wt-FLT3 in Wt- and ITD-FLT3 co-expressing cells; however, little is known about this inhibitory mechanism in recently developed FLT3 inhibitors. In this study, we aimed to clarify the efficacy and characteristics of four FLT3 inhibitors including the impact of FL on AML cell lines and primary patients cells, and to identify biomarkers for drug selection and prediction of their efficacy. 【Methods】We examined the growth inhibitory effects of midostaurin, quizartinib, gilteritinib and FF-10101 at various concentrations of FL in ITD-FLT3 expressing 32D cells, Wt- and ITD-FLT3 co-expressing 32D cells,and FL and Wt-FLT3 co-expressing 32D cells. The inhibitory effects of these four FLT3 inhibitors were also examined in 87 primary AML cells with or without FLT3 mutationin vitro, and the correlation between their efficacy and clinical and molecular characteristics including genetic alterations and FLT3-ITD allelic ratio (ITD-AR) were investigated. Moreover, characteristics of residual AML cells after the treatment with FLT3 inhibitors were also examined in patient-derived xenografts (PDX) -AML model. 【Results】The inhibitory effect of FLT3 inhibitors was significantly impaired by FL stimulation dose dependently in all inhibitors except for midostaurin in Wt- and ITD-FLT3 co-expressing 32D cells. In FL and Wt-FLT3 co-expressing cells, GI50 value of gilteritinib was higher than that in ITD-FLT3 solo-expressing cells, and the similar tendency was observed with quizartinib. We examined the Growth inhibitory effects of these inhibitors in 33 FLT3 mutated and 54 FLT3 un-mutated primary AML cells. In primary cells, midostaurin showed lower selectivity to FLT3 mutation compared with other inhibitors. Subsequently, the correlation between the efficacy of FLT3 inhibitors and ITD-AR was examined. RNA or DNA based ITD-AR was not related to the growth inhibitory effect of FLT3 inhibitors except for gilteritinib in FLT3-ITD mutated patient cells; the GI50 value of gilteritinib in AML cells with RNA based ITD-AR-low were significantly higher than those in RNA based ITD-AR-high (P=0.036) (Figure).Moreover, FLT3-ITD cells with mutated NPM1 tend to have higher GI50values for all of FLT3 inhibitors, irrespective of ITD-AR. In AML-PDX treated with quizartinib or gilteritinib, FLT3-ITD-AR in residual AML cells was lower than that of non-treated cells, suggesting that co-expression level of Wt-FLT3 is related to the response to FLT3 inhibitors in vivo. 【Conclusions】FL affected the efficacy of FLT3 inhibitors in Wt- and ITD-FLT3 co-expressing cells, and the inhibitory effects on ITD-FLT3 and FL-Wt-FLT3 pathway were different among FLT3 inhibitors. Furthermore, NPM1 mutation and RNA based ITD-AR might be predictive biomarkers for the efficacy of FLT3 inhibitors in FLT3-ITD positive AML. The appropriate therapeutic strategy based on the characteristics of each inhibitor is necessary. Disclosures Ishikawa: Bristol-Myers Squibb: Honoraria; Abbvie GK.: Honoraria; Kyowa Hakko Kirin Co., Ltd.: Honoraria; Celgene Co., Ltd.: Honoraria. Goto:Celgene Co., Ltd.: Honoraria; Novartis Pharma Co., Ltd.: Honoraria; JCR Pharmaceuticals Co., Ltd.: Honoraria; Takeda Pharmaceutical Co., Ltd.: Honoraria. Ozawa:Pfizer Japan Inc.: Honoraria; Novartis: Honoraria; Kyowa-Hakko Kirin: Honoraria; Astellas Pharma Inc.: Honoraria. Kiyoi:Zenyaku Kogyo Co., Ltd.: Research Funding; Nippon Shinyaku Co., Ltd.: Research Funding; Eisai Co., Ltd.: Research Funding; Chugai Pharmaceutical Co., Ltd.: Research Funding; Sumitomo Dainippon Pharma Co., Ltd.: Research Funding; Takeda Pharmaceutical Co., Ltd.: Research Funding; Astellas Pharma Inc.: Honoraria, Research Funding; Otsuka Pharmaceutical Co.,Ltd.: Research Funding; FUJIFILM Corporation: Research Funding; Pfizer Japan Inc.: Honoraria; Kyowa Hakko Kirin Co., Ltd.: Research Funding; Daiichi Sankyo Co., Ltd: Research Funding; Bristol-Myers Squibb: Research Funding; Perseus Proteomics Inc.: Research Funding.

Hottip Lncrna Reinforces CTCF Defined Chromatin Boundaries and Drives Wnt Target Gene Expression in AML Leukemogenesis

We reported recently that HOXA locus associated lncRNA, HOTTIP, is highly expressed in AML patients carrying MLL rearrangement and NPM1c+ mutations. The expression of HOTTIP positively correlates with posterior HOXA gene expression and poor patient survival. We further demonstrated that HOTTIP acts as an epigenetic regulator to define oncogenic HOXA topologically associated domain (TAD) and drive HOXA associated leukemic transcription program. However, it remains unclear whether and how HOTTIP lncRNA is involved in remodeling leukemic genome to facilitate AML leukemogenesis. Here, we showed that HOTTIP regulates a fraction of CTCF binding sites (CBSs) in the AML genome by directly interacting with CTCF and its binding motifs. We carried out CTCF ChIP-seq and HOTTIP ChIRP (chromatin isolation by RNA purification)-seq comparing WT and HOTTIP knockout (KO) MOLM13 cells. KO of HOTTIP in MLL-rearranged MOLM13 AML cells specifically impaired CTCF binding sites that were co-occupied by HOTTIP lncRNA, whereas loss of HOTTIP did not affect global CTCF binding. These target genes include posterior HOXA genes and Wnt target genes such as C-MYC, EVI1, AXIN, and TWIST1. Furthermore, we found that HOTTIP interacts with its putative target sites by formation of DNA: RNA hybridization structure triple helix and R-loop in vivo and in vitro. We then carried out DRIP (DNA-RNA immunoprecipitation)-seq and DRIPc(DNA-RNA immunoprecipitation followed by cDNA conversion)-Seq, which utilize a sequence independent but structure-specific S9.6 antibody for DRIP to capture global R-loops, by comparing WT and HOTTIP KO MOLM13 cells. The obtained DRIP-seq and DRIPc-seq data were then incorporated and integrated with the HOTTIP ChIRP-seq and CTCF ChIP-seq data to explore global collaboration between R-loop and HOTTIP associated CTCF binding sites. We found that HOTTIP interacts with CTCF binding motif that defines the TADs and the promoters of the HOTTIP target genes by formation of R-loop or triple helix structure. Loss of HOTTIP disrupted the R-loop formation at promoters and enhancers of the HOTTIP target genes to inhibit their expression. In MLL-rearranged AML genome, in addition to the HOXA locus, CTCF forms leukemic specific TADs that protect aberrant Wnt target genes. Depletion of HOTTIP lncRNA impaired CTCF defined TADs in the Wnt target gene loci and reduced Wnt target gene expression. In contrast, overexpression of Hottip lncRNA (Hottip-Tg) in the mice bone marrow hematopoietic compartment perturbs hematopoietic stem cell (HSC) self-renewal and differentiation leading to AML like disease by reinforcing CTCF defined TADs, enhancing chromatin accessibility within TADs, and upregulating gene transcription in the Wnt target loci. Finally, when we treated HOTTIP expressed primary patient AML cells carrying MLL-rearrangement and their derived PDX mouse model with a canonical Wnt inhibitor, ICG-001, ICG-001 inhibited AML LSC self-renewal in in vitro by LTC-IC assays and in vivo leukemogenesis in the PDX mouse models with an aberrant HOTTIP lncRNA expression, but not in HOTTIP negative/low non-MLL AML samples. Thus, HOTTIP lncRNA and CTCF cooperate to specifically reinforce CTCF defined WNT target locus TADs and drive Wnt target gene expression in the HOTTIP expressed AML. Disclosures No relevant conflicts of interest to declare.

A Novel Combination Regimen of BET and FLT3 Inhibition for FLT3-ITD Acute Myeloid Leukemia

Background: Acute myeloid leukemia patients with FLT3-ITD mutations have a high risk of relapse and death. FLT3 tyrosine kinase inhibitors such as quizartinib and gilteritinib improve overall survival in relapsed patients, but their efficacy is limited and most such patients die of the disease. This is because even with potent FLT3 inhibition, the disease persists within the bone marrow microenvironment, mainly due to bone marrow stroma activating parallel signaling pathways that maintain pro-survival factors. BET inhibitors suppress pro-survival factors such as c-Myc and Bcl2, but these drugs thus far have shown only limited single-agent clinical potential. PLX51107 is a novel BET inhibitor designed to inhibit BET activity in intermittent daily fashion to allow for greater tolerability. We investigated whether the addition of PLX51107 to potent FLT3 inhibition with quizartinib could overcome the protective effect of the bone marrow stroma through inhibition of c-Myc expression. Methods: We developed a plasma inhibitory activity assay for c-Myc (c-Myc PIA) to assess in vivo efficacy of the PLX51107 in patients and to identify c-Myc-inhibitory doses of the drug. We tested PLX51107 alone and in combination with quizartinib in murine models of AML and against primary FLT3-ITD AML cells co-cultured with human bone marrow stroma. In addition, we analyzed gene expression patterns in the treatment models to explore the basis of any observed synergistic cytotoxic effect. Results: In a murine xenograft model of AML using MV4-11 cells, PLX51107 alone induced suppression of tumor growth in association with a 90% decrease in c-Myc gene expression. The combination of PLX51107 and quizartinib induced complete tumor regression in 5 out of 7 animals after 14 days of treatment (Figure 1). Animals treated with a 5-day course of quizartinib alone displayed tumor regression persisting until day 26 of treatment, while the addition of PLX51107 resulted in tumor regression until day 39. In patients treated with a dose of 120 mg/day of PLX51107, the c-Myc PIA demonstrated a robust suppression of c-Myc expression for roughly 6 hours, returning to baseline between 7 and 9 hours post-treatment (Figure 2). The mean plasma concentration to achieve this inhibition was 3.3 uM, which, accounting for the difference in protein drug binding between plasma and culture medium with 10% FBS, corresponded to a concentration of 250 nM PLX51107 in culture medium. With this same concentration and schedule (i.e., at concentrations and exposure times equivalent to what human patients would experience taking 120 mg daily of PLX51107 and 60 mg daily of quizartinib), the combination induced synergistic cytotoxicity in a series of 10 different FLT3-ITD AML patient blast samples co-cultured with bone marrow stroma (Figure 3). C-Myc RNA and protein were directly suppressed in these primary samples, and ingenuity pathway analysis of RNA expression confirmed that c-Myc associated genes displayed the highest level of down-regulation. Conclusions: These studies suggest that combination therapy with approximately 120 mg/day PLX51107 and 60 mg/day quizartinib will be a more effective therapy for relapsed FLT3-ITD AML than 60 mg/day of quizartinib alone. The combination of FLT3 inhibition and BET inhibition may represent an attractive therapeutic option for FLT3-ITD AML. Disclosures Hizukuri: Daiichi Sankyo Co, Ltd: Employment. Severson:Plexxikon Inc.: Employment. Powell:Plexxikon Inc.: Employment. Zhang:Plexxikon Inc.: Employment. Ma:Plexxikon Inc: Employment. Narahara:Daiichi Sankyo Co, Ltd: Employment. Sumi:Daiichi Sankyo, Inc.: Employment. Bollag:Plexxikon Inc.: Employment. Levis:FUJIFILM: Consultancy, Research Funding; Daiichi Sankyo Inc: Consultancy, Honoraria; Agios: Consultancy, Honoraria; Astellas: Consultancy, Research Funding; Novartis: Consultancy, Research Funding; Amgen: Consultancy, Honoraria; Menarini: Consultancy, Honoraria.

Role of Autophagy in Resistance of FLT3-ITD AML Stem Cells to FLT3 TKI Treatment

The FLT3-ITD mutation occurs in 25-30% of AML patients, and is associated with increased relapse rate and poor survival. FLT3 tyrosine kinase inhibitors (TKI) have shown encouraging results, but responses are not sustained. Therefore, there is considerable interest in understanding mechanisms of resistance of FLT3-ITD AML stem cells after TKI treatment. Autophagy is an important cellular response that maintains cell survival under stress. Here we investigated the regulation of autophagy in resistance of FLT3-ITD+ AML stem cells to FLT3 TKI treatment. MV4-11 and Molm13 FLT3-ITD AML cells were transduced with GFP-LC3-RFP-LC3 ΔG constructs to measure autophagic flux. Autophagyhigh AML cells showed slower growth(p<0.0001), enhanced sensitivity to the autophagy inhibitor Lys05 (p=0.01) and reduced sensitivity to the FLT3-ITD inhibitor AC220 (p=0.03) compared with autophagylow cells. Treatment with Lys05 in combination with AC220 resulted in significantly enhanced inhibition of AML cell growth (p=0.02) and increased induction of apoptosis (p=0.024) compared to AC220 alone. We confirmed that inhibition of autophagy by shRNA-mediated knockdown of the critical autophagy gene ATG5 also enhanced sensitivity of AML cells to AC220 treatment (p=0.004). Lys05 treatment also inhibited autophagic flux in primary human FLT3-ITD+ AML CD34+ stem/progenitor cells, and resulted in significantly increased apoptosis (78% vs 58%, p=0.0245) and reduced cell growth (47% vs 68%, p=0.003) when combined with AC220 compared to AC220 alone. In contrast Lys05 treatment had minimal effect on normal CD34+ stem cells, indicating that autophagy inhibition selectively enhances sensitivity of FLT3-ITD+ AML CD34+ cells to TKI treatment while sparing normal CD34+ cells. We tested the effect of Lys05 in vivo by treating NSG mice engrafted with primary FLT3-ITD AML cells with vehicle, AC220 (10mg/kg), Lys05 (20mg/kg) or combination for 3 weeks. Lys05 treatment reduced human CD45+ cell engraftment (8.7% vs 17%, p=0.03), and reduced CD45+CD33+ myeloid cells in BM and spleen (BM: 42% vs 81%, p<0.001, Spleen: 8% vs 14%, p=0.01), compared with vehicle. The combination of AC220 with Lys05 reduced CD45 engraftment (11% vs 37%, p=0.004) and absolute CD45+CD34+ stem cell numbers in BM compared with AC220 alone. The effect of Lys05 in vivo was also investigated in a transgenic FLT3-ITD AML mouse model (Mx1-cre-Tet2f/f, Flt3-ITDKi). The ability of in vivo Lys05 treatment to inhibit autophagy was confirmed by immunofluorescence. Lys05 inhibited autophagy flux to a greater extent in c-Kit+ stem/progenitor cells compared to differentiated cells. Leukemic mice were treated with PBS, Lys05(20mg/kg), AC220(10mg/kg) or the combination for 3 weeks. The combination of AC220 and Lys05 resulted in significantly greater reduction of WBC and neutrophil counts compared to the other groups(p=0.03). The AC220 and Lys05 combination reduced spleen weight and cellularity significantly more than AC220 alone (p=0.02), and splenic ST-HSC, MPP, GMP, MEP and BM GMP, MEP significantly more than other groups. However, BM ST-HSC and MPP (LT-HSC are absent in this AML model) were similar among the different treatment groups. The proportion of mature myeloid and T cells was increased in the combination group. Ki67 and Dapi labeling indicated that Lys05 reduced the frequency of ST-HSC and MPP in G0 and increased those in G1. The ability of Lys05 to promote quiescent LSC entry into cell cycle may contribute to enhanced sensitivity to AC220 when the two are used in combination. We show that induction of autophagy in MOLM13 cells by starvation and mTOR inhibitor was associated with reduced p53 expression, whereas autophagy inhibition by chloroquine or SAR405, a PIK3C3/Vps34 inhibitor, was associated with increased p53 expression by WB. Inhibition of growth or increased apoptosis in FLT3-ITD+ AML cells following autophagy inhibition were prevented by shRNA-mediated knockdown of p53, suggesting that effects of autophagy inhibition on FLT3-ITD+ AML cell growth were p53 dependent. We conclude that autophagy inhibition enhances sensitivity of FLT3-ITD+ AML stem cells to TKI treatment both in vitro and in vivo. We demonstrate important roles for autophagy in regulating p53 levels and cell cycle in FL3-ITD AML cells. These studies support further development of strategies to target autophagy and related pathways to enhance efficacy of TKI in eliminating AML stem cells. Disclosures No relevant conflicts of interest to declare.

Inhibition of Mcl-1 Enhances the Efficacy of Tyrosine Kinase Inhibition in FLT3 Mutated AML and Synergizes with Venetoclax Targeting AML Stem Cells

Inhibition of Mcl-1 Enhances the Efficacy of Tyrosine Kinase Inhibition in FLT3 Mutated AML and Synergizes with Venetoclax Targeting AML Stem Cells

CD155-Tigit Pathway Modulation in Dendritic Cell/Acute Myeloid Leukemia Fusion Vaccine Model

Background: We developed a vaccine consisting of primary AML cells fused with autologous dendritic cells (DC) such that a broad array of antigens are presented in the context of DC-mediated costimulation, thereby expanding the anti-AML T cell repertoire. The reestablishment of tumor tolerance by the microenvironment (TME) can limit vaccine efficacy and reversal of TME-mediated immunosuppression has the potential to enhance vaccine immunogenicity. To this end, using an immunocompetent syngeneic AML murine model, we have shown that hypomethylating agents (HMAs) reverse critical aspects of the immunosuppressive TME, and that HMA + fusion vaccination enhances AML-specific immune responses and survival. The TIGIT-CD155 inhibitory checkpoint pathway is a novel pathway fostering tolerance in the innate and adaptive immune arms. Blockade of this pathway reverses tumor-mediated T cell quiescence and promotes DC and NK cell immune surveillance. We aimed to examine if TIGIT-CD155 pathway inhibition +/- HMA enhances vaccine potency in a murine model. Methods/Results: Using a syngeneic immunocompetent murine AML model, we first investigated the impact of functional TIGIT monoclonal (mAb) blockade +/- decitabine (DAC) +/- vaccine on immune modulation and murine survival. C57BL/6J mice were retro-orbitally inoculated with 5x104 TIB-49 murine AML cells. Animals treated with TIGIT mAb monotherapy had inferior survival as compared to those treated with DAC or DC/AML vaccine monotherapies. Mice treated with vaccine alone, vaccine + TIGIT mAb, and vaccine + DAC + TIGIT mAb had similar survival but each cohort had improved survival compared to untreated animals, suggesting a vaccine-mediated effect not further enhanced by TIGIT mAb. To this end, we sought to investigate if deleting CD155, the native receptor to TIGIT, would impact AML immunogenicity and murine survival. We determined that baseline expression of CD155 in the murine AML cell line, TIB-49, is 72% by Ab staining and flow cytometric analysis. Using CRISPR/cas9 technology, we generated a CD155 knockout TIB-49 cell line. Black mice were retro-orbitally inoculated with 5x104 CD155-deleted or control TIB-49 AML cells. Animals were treated with vaccine, DAC, or both therapies. BLI imaging was performed 1-2/week. Animals inoculated with CD155-deleted AML cells that were treated with both DAC and vaccine show no signs of disease and remain alive 80 days following inoculation as compared to mice treated with vaccine or DAC alone (Fig 1,2). Conclusions: We demonstrate that CD155-specific deletion from a murine AML cell line confers a survival advantage when mice are exposed to HMA + vaccine as compared to either agent alone. Studies including TCR repertoire analyses are underway to better understand the immunologic underpinnings of this findings as targeting this pathway has promising translational potential. Disclosures Rosenblatt: Parexel: Consultancy; Imaging Endpoint: Consultancy; Celgene: Research Funding; BMS: Research Funding; Amgen: Other: Advisory Board; Merck: Other: Advisory Board; BMS: Other: Advisory Board ; Dava Oncology: Other: Education; Partner Tx: Other: Advisory Board. Kufe:Genus Oncology: Equity Ownership; Nanogen Therapeutics: Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Canbas: Consultancy, Honoraria; Hillstream BioPharma: Equity Ownership; Victa BioTherapeutics: Consultancy, Equity Ownership, Honoraria, Membership on an entity's Board of Directors or advisory committees; Reata Pharmaceuticals: Consultancy, Equity Ownership, Honoraria. Avigan:Takeda: Consultancy; Parexel: Consultancy; Janssen: Consultancy; Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Pharmacyclics: Research Funding; Juno: Membership on an entity's Board of Directors or advisory committees; Partners Tx: Membership on an entity's Board of Directors or advisory committees; Partner Tx: Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Membership on an entity's Board of Directors or advisory committees.

In vivo efficacy of mutant IDH1 inhibitor HMS-101 and structural resolution of distinct binding site.

Mutations in isocitrate dehydrogenase 1 (IDH1) are found in 6% of AML patients. Mutant IDH produces R-2-hydroxyglutarate (R2HG), which induces histone- and DNA-hypermethylation through inhibition of epigenetic regulators, thus linking metabolism to tumorigenesis. Here we report the biochemical characterization, in vivo antileukemic effects, structural binding and molecular mechanism of the inhibitor HMS-101, which inhibits the enzymatic activity of mutant IDH1 (IDH1mut). Treatment of IDH1mut primary AML cells reduced 2-hydroxyglutarate levels (2HG) and induced myeloid differentiation in vitro. Co-crystallization of HMS-101 and mutant IDH1 revealed that HMS-101 binds to the active site of IDH1mut in close proximity to the regulatory segment of the enzyme in contrast to other IDH1 inhibitors. HMS-101 also suppressed 2HG production, induced cellular differentiation and prolonged survival in a syngeneic mutant IDH1 mouse model and a patient-derived human AML xenograft model in vivo. Cells treated with HMS-101 showed a marked upregulation of the differentiation-associated transcription factors CEBPA and PU.1, and a decrease in cell cycle regulator cyclin A2. In addition, the compound attenuated histone hypermethylation. Together, HMS-101 is a unique inhibitor that binds to the active site of IDH1mut directly and is active in IDH1mut preclinical models.

Suicide gene‑armed measles vaccine virus for the treatment of AML.

Virotherapy comprises a novel therapeutic approach to selectively eliminate cancer cells. Preclinical, as well as clinical data have demonstrated the efficacy of tumor-selective (oncolytic) viruses in hematological malignancies. In this study, we infected AML cell lines and primary AML cells from patients with measles vaccine virus either expressing GFP or armed with super cytosine deaminase, which converts the prodrug, 5-fluorocytosine, into the chemotherapeutic compound, 5-fluorouracil. Target cell density of the measles entry receptor, CD46, infection rates of targeted leukemic cells, tumor cell viability, and apoptotic rates were determined. We found that measles vaccine virus infected the leukemic blasts and profoundly diminished the number and viability of leukemic cells via the induction of apoptosis. The conversion of 5-fluorocytosine to 5-fluorouracil exerted a potent additive tumoricidal effect. This was also observed in cases when leukemic cells displayed only moderate susceptibility to the oncolytic virus and hence direct oncolysis. Taken together, in this study, we provide a first characterization of the combinatorial use of measles vaccine virus and 5-fluorouracil for treatment of AML. Our approach to site-specifically produce the active drug and combine this agent with the direct lytic effect of virotherapy may overcome present limitations and constitutes a feasible method with which to introduce 5-fluorouracil in the treatment of AML.

Abstract 3612: Decreased vitamin C uptake mediated by SLC2A3 promotes leukemia progression and impedes Tet2 restoration

Abstract Background: Two new studies in Nature and Cell, from Agathocleous et al. (2017) and Cimmino et al. (2017), respectively, showed that vitamin C regulates HSC function and suppresses leukemogenesis by modulating Tet2 activity. However, little is known about the relationship between vitamin C transporter and hematological malignancies. Purpose: In this study, we investigated the role of SLC2A3, a gene that encode GLUT3 which mainly transport glucose and oxidized form vitamin C, in hematological malignancies. Through this research we identified a set of biomarkers which can predict the effect of vitamin C treatment and develop stagey to improve this putative innovative treatment. Results: We analyzed gene expression patterns of the major GLUT family genes in primary AML blast cells using previously published microarray datasets (GSE37307) and (GSE9476) and determined that SLC2A3 gene expression was significantly decreased in blast cells compared with normal hematopoietic cells. By contrast, expression of SLC2A1, SLC2A2 and SLC23A2 was not significant different in between primary AML cells and normal hematopoietic cells. Next, we analyzed the relevance of SLC2A3 expression to the survival of AML patients from TCGA and TARGET database and found that below-median SLC2A3 expression was associated with inferior overall survival. Subsequently, we detected the SLC2A3 mRNA expression level on 4 AML cell lines and 3 DLBCL cell lines, there were very low expression level on OCI-AML-3 and OCI-LY1. To check whether decreased expression of SLC2A3 has an impact on the Vitamin C effect, we treated the 7 cell lines with dose dependent vitamin C, up to 500 μM, 48 hours. The cell viability assay showed that OCI-AML-3 and OCI-LY1 have no response to Vitamin C, however the proliferation of NB4, HEL, HL60, OCI-LY19 and Toledo was suppressed by 250 μM Vitamin C. Conclusions: SLC2A3 expression was decreased in AML blast cell and down regulation of SLC2A3 was closely associated with the poor outcomes of AML patients. In vitro study showed that SLC2A3 was essential for the effect of Vitamin C. Therefore, SLC2A3 could be a potential biomarker to predict the effect of Vitamin C treatment. In addition, through targeting SLC2A3, efficacy of Vitamin C treatment could be improved. Citation Format: Jun Liu, Junshik Hong, Kwangsung Ahn, Sungsoo Yoon. Decreased vitamin C uptake mediated by SLC2A3 promotes leukemia progression and impedes Tet2 restoration [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3612.

Abstract 2964: Pharmacological inhibition of WIP1 by GSK2830371 sensitizes AML cells to MDM2 inhibitor Nutlin-3a

Abstract Introduction: PPM1D (wild-type p53-inducible protein phosphatase WIP1) is a member of the PP2C family serine/threonine phosphatase involved in negative regulation of cell stress response pathways, leading to suppression of p53 after stress. To restore p53 function through MDM2 inhibition (Nutlin-3a) is a promising approach in AML. Promising results of the combination of MDM2 inhibitors and WIP1 inhibitor (WIP1i), obtained in many preclinical studies of solid tumors, paved the way for their application in AML. We investigated whether the inhibition of WIP1 (GSK2830371) could sensitize AML cell lines and primary cells to Nutlin-3a (Nut-3a) in order to obtain a novel therapeutic strategy for AML patients, based on restoration of p53 activity. Methods: In vitro viability (by WST-1 reagent) and Annexin-V/PI apoptosis assay were performed on a panel of TP53-wt AML cells (MOLM-13, MV-4-11 and OCI-AML3), on TP53-mutated NOMO-1 AML cells and on primary AML samples. Gene expression profile (GEP) and Western Blot (WB) analyses were performed on MV-4-11 and NOMO-1 after 16h. Results: Combined inhibition of increasing dosage of Nut-3a (0.5 to 5 uM) and WIP1i (5 to 20 uM) synergistically reduces TP53-wt AML cells viability, while NOMO-1 resulted to be insensitive to the combination. The combination index analyses showed a synergistically effect of the combination of both compounds on TP53-wt AML cells. Annexin V/PI staining showed that WIP1i sensitizes TP53-wt AML cell lines and primary samples to Nut-3a-induced apoptosis, when compared with single treatment. No effect was seen in NOMO-1. GEP demonstrated that MV-4-11 cells exhibits a major response to drug combination with an upregulation of cell cycle control genes, a downregulation of DNA repair machinery genes, upregulation of MDM2 and of TP53-downstream genes (eg.CDKN1A), confirming the activation of p53 pathway. NOMO-1 cells showed upregulation of resiliency-mechanism and confirmed insensitivity to both treatment. WB analysis confirmed GEP data showing an increased expression of p53 and p21 in wt-TP53 cell line after 16h of combined-treatment. Conclusions: We identified a novel synergistic drug combination between Nut-3a and WIP1i that induces apoptosis in AML cell lines and primary samples. GEP and WB of TP53-wt MV-4-11 and TP53-mutated NOMO-1 cells showed mechanisms underlying drug sensitivity and resistance giving novel insights on potential markers of response and novel drugable targets. In vivo studies are needed to confirm these preclinical data. Supported by: AIRC, FP7-NGS-PTL, Fondazione del Monte, HARMONY. Citation Format: Maria Chiara Fontana, Jacopo Nanni, Giovanni Marconi, Martina Pazzaglia, Matteo Bocconcelli, Antonella Padella, Simona Soverini, Ilaria Iacobucci, Cristina Papayannidis, Anna Ferrari, Maria Teresa Bochicchio, Enrica Imbrogno, Michele Cavo, Andrea Ghelli Luserna di Rora, Giorgia Simonetti, Giovanni Martinelli. Pharmacological inhibition of WIP1 by GSK2830371 sensitizes AML cells to MDM2 inhibitor Nutlin-3a [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2964.

Abstract LB-274: Identification and characterization of LY3410738, a novel covalent inhibitor of cancer-associated mutant Isocitrate Dehydrogenase 1 (IDH1)

Abstract Somatic gain-of-function mutations in IDH1 have been identified in multiple tumor types including AML, glioma, cholangiocarcinoma and chondrosarcoma. The neo-enzymatic activity of mutant IDH1 results in accumulation of the oncometabolite 2-hydroxyglutarate (2-HG), leading to a hyper-methylation phenotype, a block in cell differentiation, and tumor growth. Using a structure-based drug design approach, we have developed a highly potent covalent inhibitor of mutant IDH1. LY3410738 modifies a single cysteine (Cys269) in the allosteric binding pocket and rapidly inactivates the enzyme with a KI/Kinact = 84,257 M-1sec-1. The compound selectively inhibits the 2-HG in IDH1 mutant tumor cells without depleting the levels of a-ketoglutarate. Using patient-derived primary AML cells, we demonstrated that LY3410738 was more potent than AG120 in reversing the block in differentiation associated with IDH1 mutant activity. In vivo, LY3410738 displayed prolonged pharmacodynamic activity, depleting 2-HG levels in tumors at low circulating exposures and for an extended time after clearance of compound. Importantly, LY3410738 has the ability to cross the blood-brain barrier and can achieve concentrations in the brain that exceed those needed to engage the target. Consistent with this, LY3410738 effectively inhibits 2HG in orthotopic glioma models. Using patient-derived IDH1 mutant orthotopic AML models, we demonstrated that LY3410738 effectively inhibited 2-HG, induced differentiation, and cleared AML from mice. Collectively, LY3410738 represents the first covalent brain-penetrant mutant IDH1 inhibitor with potential for Best-in-Class activity. Citation Format: Nathan Brooks, Robin DeWalt, Serge Boulet, ZhaoHai Lu, Lisa Kays, Rachel cavitt, Sandra Gomez, John Strelow, Paul Milligan, Kenneth Roth, Renato Bauer, Stephen Antonysamy, Patric Hahn, Zoran Rankovic, Denis McCann, Gary Mo, Ramon Tiu, Timothy Burkholder, Sandaruwan Geeganage, Raymond Gilmour. Identification and characterization of LY3410738, a novel covalent inhibitor of cancer-associated mutant Isocitrate Dehydrogenase 1 (IDH1) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-274.

Abstract 4529: Mapping the protein interactome of mitochondrial intermembrane space proteases identifies a novel function for HTRA2

Abstract Mitochondria possess unique proteases that localize to specific sub-compartments of the organelle. However, the functions of these proteases are largely ill-defined. Here, we used proximity-dependent biotinylation (BioID) to map the interactomes of seven proteases located in the intermembrane space of the mitochondria. The mitochondrial intermembrane space proteases HTRA2, OMA1, YME1L1, LACTB, IMMP1L, IMMP2L and PARL were cloned in-frame with the abortive E. coli biotin ligase BirA*, and expressed in 293 T-REx cells. Cell culture media was spiked with biotin for 24 hrs, the cells lysed, and biotinylated proteins were isolated and identified by mass spectrometry. In total, we identified 342 different proteins as high confidence interactors of the seven mitochondrial proteases. Of these, 272 are assigned a GO mitochondrial annotation, and 230 proteins interacted with only 1 or 2 proteases in our dataset. Validation efforts were focused on high temperature requirement peptidase A 2 (HTRA2). HTRA2 is a serine protease that is released into the cytoplasm during apoptosis where it binds Inhibitor of Apoptosis Proteins (IAPs). However, little is known about the function of HTRA2 in the mitochondria. HTRA2 interacted with 60 mitochondrial, 11 nuclear and 4 cytoplasmic proteins, including its known interactor XIAP, and consistent with its known localization to these cellular compartments. HTRA2 interacted with 8 out of 13 components of the MIB complex, a multiprotein assembly that is essential for proper mitochondrial cristae formation. Knockdown of HTRA2 with shRNA in 293T-REx cells disrupted cristae formation and this phenotype was rescued by expression of an shRNA-resistant HTRA2 cDNA. Compared to normal hematopoietic cells, HTRA2 mRNA expression levels are increased in a subgroup of primary AML cells. HTRA2 knockdown in OCI-AML2 leukemia cells led to a similar disruption of mitochondrial cristae. Knockdown of HTRA2 in OCI-AML2 cells led to increased levels of the MIB subunit IMMT, but not two other MIB complex subunits, SAMM50 and CHCHD3. Finally, in cell-free assays, we demonstrate that recombinant HTRA2 can degrade recombinant IMMT, but not SAMM50 or CHCHD3.Thus, we have mapped the interactomes of the proteases of the mitochondrial intermembrane space. Through this effort, we discovered that HTRA2 regulates protein levels of the MIB complex subunit IMMT and that disruption of this process affects mitochondrial cristae formation. Citation Format: Aaron D. Botham, Etienne Coyaud, Sanjit Nirmalanandhan, Marcela Gronda, Rose Hurren, Neil Maclean, Jonathan St. Germain, Sara Mirali, Estelle Laurent, Brian Raught, Aaron Schimmer. Mapping the protein interactome of mitochondrial intermembrane space proteases identifies a novel function for HTRA2 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4529.

PF239 COMBINED TARGETING OF BCL-2 AND SET-PP2A INTERACTION INDUCES POTENT AND SYNERGISTIC ANTI-LEUKEMIC EFFECTS ON ACUTE MYELOID LEUKEMIA

Background: Acute myeloid leukemia (AML) is an aggressive malignancy that arises from the transformation of hematopoietic stem cells. Most AML patients are older than 60 years, and in this group only 5–15% of cases are cured (Döhner et al., 2015). Therefore, it is necessary to explore new therapeutic options. Overexpression of the anti-apoptotic BCL-2 protein is a common finding in AML. ABT-199 (Venetoclax), a potent and selective inhibitor of BCL-2, has been recently developed. However, clinical activity of ABT-199 monotherapy is modest in AML (Konopleva et al., 2016), which could be explained by the genetic and epigenetic heterogeneity of AML. Our group has shown that inactivation of PP2A is present in 78% of AML patients, and that FTY720 treatment restores PP2A activity by disrupting the SET-PP2A interaction, thereby inhibiting proliferation of AML cells (Pippa et al., 2014). Since PP2A promotes apoptosis by affecting the phosphorylation state of a variety of pro- and anti-apoptotic factors, we hypothesized that BCL-2 targeting could synergize with PP2A re-activation and induce synergistic anti-leukemic effects. Aims: To investigate whether the combination of ABT-199 with FTY720 could induce synergistic anti-leukemic effects in vitro and in vivo. Methods: A panel of six human AML cell lines was used to perform cell culture experiments. To generate zebrafish AML xenograft models, AML cell lines were treated for 24 h with ABT-199, FTY720 or their combination. Subsequently, cells were stained with a fluorescent cell tracker and injected into the yolk sac of 48 hours-post-fecundation wild type zebrafish embryos. After injection, cell proliferation and colonization index were measured. Colony-Forming-Unit (CFU) assays in semi-solid medium were performed on patient-derived AML cells isolated from three patients with AML at diagnosis. Results: GI50 values for ABT-199 and FTY720 showed that HL-60, MV4–11 and MOLM-13 were the most sensitive cell lines to ABT-199, while OCIAML-3 and HEL cells were resistant to BCL-2 specific inhibition (GI50 value greater than 10 μM). Regarding FTY720 treatment, we observed that OCIAML-3 and MV4–11 were the most sensitive cell lines, while HEL cells were FTY720-resistant. By western-blot analyses, we ascertained that sensitivity to ABT-199 was dependent on BCL-2, MCL-1 and BCL-XL expression levels, and sensitivity to FTY720 was dependent on PP2A and SET expression. Next, survival of AML cells was analyzed after exposure to ABT-199, FTY720, or their combination at different concentrations (accordingly on the above GI50 calculated values). Analyses of the combination index values showed that combination of ABT-199 and FTY720 had a potent and synergistic anti-leukemic effect, even in OCI-AML3 cells, which were ABT-199-resistant cells. Mechanistically, combination of ABT-199 and FTY720 induced a synergistic pro-apoptotic activity by increasing significantly the number of Annexin/IP positive cells, the activity of Caspases 3/7, and PARP cleavage. Compared to single treatments, treatment with ABT-199 plus FTY720 also reduced significantly cell proliferation and/or cell invasive potential of HL-60, MOLM-13, and MV4–11 cells in zebrafish xenograft models. Furthermore, we demonstrated in primary-AML cells that the combined treatment significantly decreased the number as well as the size of CFUs. Summary/Conclusion: Combined targeting of BCL-2 and the SET-PP2A interaction has potent and synergistic anti-leukemic effects in vitro and in vivo. Our findings provide a rational basis for further testing of novel combined therapies with ABT-199 and FTY720 in AML patients.

Interim results from a phase Ib/II clinical study of the glutaminase inhibitor telaglenastat (CB-839) in combination with azacitidine in patients with advanced myelodysplastic syndrome (MDS).

7037 Background: Glutaminase (GLS) is an enzyme that catalyzes conversion of glutamine to glutamate, providing key metabolic fuel for tumor cells. GLS is highly expressed in AML and high-risk MDS, particularly in the setting of complex cytogenetics. Preclinical studies of primary AML cells demonstrate dependence on glutamine; glutaminase inhibition led to reduced cell growth and induced apoptosis. This study was designed to evaluate the safety and efficacy of the oral glutaminase inhibitor CB-839 in combination with azacitidine for MDS. Methods: This is a single arm Phase Ib/II trial of CB-839, evaluating the dose of 600 mg BID orally daily in combination with azacitidine (AZA) for intermediate and high-risk MDS in 28-day cycles. The primary outcome for the Phase 1 portion was to confirm the safety and recommended Phase 2 dose of CB-839 in combination with AZA. Secondary endpoints evaluate efficacy and clinical activity using IWG response criteria for MDS including hematological improvement (HI), complete response (CR), marrow CR (mCR), stable disease (SD) and no response (NR). Results: A total of 10 pts with MDS were enrolled; the Phase I portion is now complete, confirming CB-839 600 mg BID with standard AZA. Median age was 71 [47-76], 90% were men. 7 pts were treatment naïve, 3 pts had prior HMA exposure. 6 pts were intermediate-2, and 4 pts intermediate-1 by IPSS with high-risk mutations. 3 pts had complex cytogenetics. The most frequent mutations were ASXL1 (n = 5), TET2 (n = 4) and TP53 (n = 4). Median number of cycles was 3 [1-9]. The most common non-hematological AEs were grade 1-3 nausea (60%) and constipation (50%). No DLTs were identified; one pt experienced grade 3 transaminitis requiring dose reduction of CB-839 (to 400 mg BID) with resolution. Seven pts (70%) achieved mCR/CR with HI, 2 pts had SD, and 1 pt had NR. The median time to mCR was 3 months [1-3]. 4 pts proceed to hematologic stem cell transplant (HSCT) after 2-4 cycles. 1 pt progressed to AML after 3 cycles of therapy, and 3 pts remain on study. Conclusions: 600 mg BID CB-839 is safe and well tolerated in combination with azacitidine in pts with MDS. The Phase II portion is ongoing. Clinical trial information: NCT03047993.

Abstract A076: DSP107—a novel SIRPα-4-1BBL dual signaling protein (DSP) for cancer immunotherapy

Abstract Background: The (re)activation of anticancer innate and adaptive immunity is at the forefront of developments in cancer therapy. Here, we report on a new immunotherapeutic fusion protein, termed Dual Signaling Protein 107 (DSP107). DSP107 was designed to combine activation of innate and adaptive immunity, by both blocking CD47/SIRPα interaction and activating 4-1BB. CD47 is overexpressed on cancer cells and upon binding to SIRPα on phagocytes transmits a “don’t eat me” signal, thereby suppressing innate immunity. 4-1BB is a costimulatory receptor that is transiently upregulated on tumor-infiltrating T-cells and is considered a surrogate marker for the tumor-reactive T-cell population. Activation of 4-1BB using its ligand or by agonistic antibodies reactivates anti-cancer T-cell immunity. In DSP107, the extracellular domains of SIRPα and 41BBL have been fused, yielding a dual function protein. DSP107 is produced as a homotrimer due to the trimerization property of 41BBL, an essential element for activating the 41BB receptor, a member of the TNF super-family of receptors. DSP107 was designed to bind to CD47 on tumor cells, thereby removing the inhibitory signal delivered to phagocytes. Simultaneously, CD47-mediated surface immobilization of DSP107 enables delivery of the 41BBL-4-1BB costimulatory signal to tumor localized T-cells. This dual immunomodulatory effect of DSP107 is designed to unleash both innate and adaptive immune responses targeted to the tumor site. Methods and Results: Trimeric DSP107 was successfully produced in a mammalian expression system. Both sides of DSP107 bound their cognate counterparts in kinetic Blitz binding assays and on human tumor and immune cell surfaces. The binding affinity of DSP107 was 1.6 nM for human CD47 and 0.69 nM for human 4-1BB as determined using BIAcore analysis. DSP107 blocked the interaction of SIRPα with CD47 in an ELISA-based competition assay (EC50 of 0.03 nM). DSP107 induced granulocyte- and macrophage-mediated phagocytosis of several lymphoma, leukemia and carcinoma cell lines in vitro. Further, DSP107 treatment triggered phagocytosis of primary AML cells by autologous macrophages. Co-treatment with DSP107 and therapeutic tumor-targeting antibodies, i.e., rituximab or cetuximab, resulted in enhanced phagocytosis of lymphoma or carcinoma cells, respectively. In a reporter assay measuring IL-8 secretion upon binding to/activation of 4-1BB, DSP107 activated 4-1BB signaling only in the presence of CD47-expressing cells. Further, DSP107 augmented the activation of purified T-cells activated by suboptimal concentrations of αCD3 + IL2 or αCD3/αCD28 Dynabeads in CD47 coated plates, as measured by percentage of CD25 expressing cells (up to 3-fold). When PBMCs were co-cultured with or without CD47-expressing cancer cells and stimulated with suboptimal concentrations of αCD3 + IL2, DSP107 treatment resulted in increased secretion of IFNg up to 2-fold), and increased T-cell proliferation (up to 2-fold). Conclusions: Here we demonstrate the feasibility and functional activity of DSP107, a novel therapeutic protein that combines activation of both the innate and adaptive immune system. Dual targeting, by the two functional sides of DSP107, offers multiple functionalities that act simultaneously and may result in a synergistic effect. DSP107 is now in IND-enabling studies and CMC development. The DSP platform can be designed for selective tumor site or microenvironment targeting and is adaptable to most checkpoint targets. Citation Format: Yosi M. Gozlan, Susan Hilgendorf, Alexandra Aronin, Yehudith Sagiv, Liat Ben-gigi-Tamir, Shira Amsili, Ami Tamir, Iris Pecker, Shirley Greenwald, Ayelet Chajut, Adam Foley-Comer, Yaron Pereg, Amnon Peled, Michal Dranitzki-Elhalel, Edwin Bremer. DSP107—a novel SIRPα-4-1BBL dual signaling protein (DSP) for cancer immunotherapy [abstract]. In: Proceedings of the Fourth CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; Sept 30-Oct 3, 2018; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2019;7(2 Suppl):Abstract nr A076.

The Novel Small Molecule Inhibitor KS99 Targets AML and Inhibits Stemness Markers STAT3 and ALDH

Introduction: Signal transducer and activator of transcription 3 (STAT3) has been shown to be activated in leukemic stem cells (LSCs), especially in Pre-LSCs, and activation has been associated with poor prognosis in multiple cancers including AML. In AML, sorted CD34+/CD38- and aldehyde dehydrogenase-positive (ALDH+) cells have significantly phosphorylated STAT3 when compared to healthy controls. The Isatin analog, KS99 has shown selective anti-cancer activity against Multiple Myeloma (MM) which may, in part, be mediated by inhibition of BTK activation. Here we demonstrate that KS99 selectively targets AML CD34+/CD38- and ALDH+ cells and inhibits STAT3 phosphorylation while sparing normal hematopoietic stem and progenitor cells (HSPCs).Experimental Design: The pro-apoptotic activity of KS99 for primary human AML cells (n=21) and human AML cell lines (n=9) was evaluated using Annexin V and cell proliferation (MTS) assays. Colony forming assays were carried out to demonstrate the specific LSC-targeted effect of KS99. Flow cytometry was used to detect STAT3 phosphorylation and apoptosis in CD34+/CD38- LSCs and ALDH activity in a series of human AMLs. Next, to assess the preclinical efficacy, the human AML cell line U937 was injected intravenously (IV) into NSG mice. Mice with an established leukemic burden were randomized into treatment groups and treated through an intraperitoneal (IP) injection with either vehicle control or KS99 (2.5mg/kg) every other day x8. Two days after the final treatment, leukemic burden (hCD45+) was quantitated in the bone marrow.Results: AML cell lines exhibited dose-dependent sensitivity to KS99 in the nanomolar range (IC50: 150-400nM) as did most primary AML cells (Figure: A). Interestingly, poor prognosis AML subsets, AML with MDS related changes (MDS-RCs), were more sensitive than De Novo AML cases (p= 0.0077) with or without a NMP1 mutation (p= 0.012, p=0.045, respectively). Within De Novo AML, NMP1 wild-type cases were more sensitive than NMP1 mutant cases (p= 0.02) (Figure: B). Furthermore, KS99 treatment selectively reduced the clonogenicity of human AML patient cells vs. normal cord blood mononuclear cells (p<0.001) (Figure: C). KS99 induced apoptosis in the CD34+/CD38- gated cell population of primary AML, leading to reduced blast-like and stem-like cells, but not other cells. It also showed a reduction in ALDH+ cells in an Aldefluor assay. Likewise, KS99 treatment showed reduced pSTAT3 in viable CD34+/CD38- cells (Figure: D). Inhibition of pBTK, BCL-2, and pSTAT3 were seen in Western blot experiments. Animal studies showed 76 % reduction of hCD45+ U937 cells in bone marrow of KS99-treated mice when compared to bone marrow isolated from vehicle-treated animals (p<0.05) (Figure: E).Conclusion: KS99 selectively targets LSCs from poor prognosis AML patients while sparing normal HPSCs. This effect may be mediated via targeting of STAT3 signaling and ALDH, which are critical for proliferating LSCs, but not normal HPSCs. Further studies will aim to demonstrate mechanisms of activity and preclinical efficacy in patient-derived AML xenograft mouse models. [Display omitted] DisclosuresDovat:Elf Zone, Inc.: Membership on an entity's Board of Directors or advisory committees.

Identification of TIM-3/Gal-9 Autocrine Loop As a Novel Wnt-Ligands Independent Machinery for the Constitutive Activation of Canonical Wnt Pathway in AML-LSCs

We originally identified T-cell immunoglobulin mucin-3 (TIM-3) as a leukemic stem cells (LSCs)-specific surface molecule and a useful marker for discriminating LSCs from hematopoietic stem cells (HSCs). Furthermore, we recently identified an unique autocrine loop composed of TIM-3 and its ligand galectin-9 (Gal-9). This TIM-3/Gal-9 autocrine loop enhances self-renewal capacity of AML-LSCs and contributes to leukemia progression(Kikushige et al., Cell Stem Cell 2015).To clarify the molecular mechanism how TIM-3 signaling enhances stem cell properties of AML-LSCs, we performed shRNA-mediated knock-down(KD) of TIM-3 in KASUMI-3 (TIM-3+ AML cell line). As a result, TIM-3-KD significantly reduce the proliferation of KASUMI-3. To investigate how TIM-3-KD attenuated leukemia propagation, we compared gene expression profile of scramble-infected control and TIM-3-KD KASUMI-3 cells. Gene Set Enrichment Analysis (GSEA) revealed TIM-3-KD resulted in the significaltly attenuated expression of HSCs and LSCs-related genes (Eppert et al., Nature Med 2011) and canonical Wnt pathway-related genes (BioCarta). Since canonical Wnt pathway is known to regulate stem cell properties via inducing the nucleus accumulation ofβ-catenin in many types of stem cells including HSCs, we next examined the β-catenin level in TIM-3-KD-KASUMI-3 cells. Array scan analysis revealed that KD of TIM-3 significantly impaired the the nucleus accumulation of β-catenin in KASUMI-3 cells. In various types of cancers, the constitutive activation of β-catenin is one of the most common oncogenic signatures, and it is mainly driven by somatic mutations in the genes related to the canonical Wnt pathway. However, such mutations are rarely detected in AML.We, therefore, hypothesized that TIM-3/Gal-9 autocrine loop signaling might be a novel mutation-independent machinery for the constitutive activation of canonical Wnt pathway in primary AML.To test this hypothethis, we first evaluated the phosphorylation status of LDL receptor-related protein 6 (LRP6), because LRP6 phosphorylation is a crucial step for canonical Wnt pathway activation. Surprisingly, Gal-9 stimulation induced phosphorylation of LRP6 and subsequent β-catenin accumulation in primary TIM-3+ AML cells even in the absence of Wnt ligand. On the other hand, LRP6 phosphorylation induced by Gal-9 stimulation was canceled by the blocking antibody to TIM-3 (F38-2E2). These results indicate that Gal-9 stimulation should induce canonical Wnt pathway activation via TIM-3 dependent manner in primary AML cells. Since Gal-9 ligation to TIM-3 is known to induce phosphorylation of tyrosine residues of TIM-3 cytoplasmic tail, leading to the recruitment and activation of several Src family kinases (SFKs) for downstream signal transduction. We found that TIM-3/Gal-9 interaction recruited hematopoietic cell kinase (HCK), a member of SFKs, to TIM-3 and activated HCK by phosphorylation of the SH1 domain (Y410 in HCK) in TIM-3+ AML cells. Furthermore, we confirmed a specific HCK inhibitor A-419259 efficiently blocked the LRP6 phosphorylation and subsequent β-catenin accumulation induced by TIM-3/Gal-9 interaction in primary AML cells. These results suggest that the canonical Wnt pathway activation induced by Gal-9 ligation to TIM-3 is mainly mediated by HCK in primary AML cells.We next focused on p120-catenin as a bridging molecule between HCK activation and LRP6 phosphorylation because, as previously reported, this molecule can be directly activated by SFKs and the activation of this molecule is an initial step for canonical Wnt pathway activation preceding LRP6 phosphorylation. We found that Gal-9 ligation to TIM-3 immediately induced the direct association of HCK and p120-catenin, leading to the activation of p120-catenin in TIM-3+ AML cells. We also confirmed that specific inhibition of HCK completely abrogated the activation of p120-catenin and subsequent canonical Wnt pathway activation in primary TIM-3+ AML cells, indicating that the activation of p120-catenin mediated by HCK is a bridging moleculer machinery for connecting TIM-3 signaling and previously identified canonical Wnt pathway. Since TIM-3 as well as HCK is specifically expressed in LSCs but not in HSCs, we conclude that TIM-3/Gal-9 autocrine loop is a novel Wnt-ligands independent machinery for the constitutive activation of canonical Wnt pathway specific for AML-LSCs. (summarized in figure) [Display omitted] DisclosuresAkashi:Celgene: Research Funding, Speakers Bureau; Pfizer: Research Funding; Bristol-Myers Squibb: Research Funding, Speakers Bureau; Kyowa Hakko Kirin: Research Funding, Speakers Bureau; Taiho Pharmaceutical: Research Funding; MSD: Research Funding; Asahi-kasei: Research Funding; Chugai Pharma: Research Funding; Ono Pharmaceutical: Research Funding; Otsuka Pharmaceutical: Research Funding; Eli Lilly Japan: Research Funding; Novartis pharma: Research Funding; Eisai: Research Funding; sanofi: Research Funding; Astellas Pharma: Research Funding.

Long Term Culture of Primary AML and CML Cells: Effects of Growth Factor Composition on Cell Expansion and Differentiation

Introduction:In vitro experiments with primary acute and chronic myeloid leukemia (AML, CML) cells are widely applied for testing of anti-tumor drug activity. Several studies also documented long-term maintenance of primary cells with the support of stromal cells and growth factors. However, to our knowledge, none of these studies sufficiently analyzed the differentiation and phenotype changes occurring during the long-term culture.Aim:To evaluate the expansion and phenotype changes of primary AML and CML cells in long-term culture using different cytokine cocktails.Methods:Primary leukocytes from 6 de novo AML samples, 5×10^5 per well, (various French-American-British classification) and CD34+ cells isolated from 4 de novo CML samples, 15-50×10^3 per well, were seeded onto a confluent layer of human bone-marrow mesenchymal stromal cells (MSCs) in 24-well plates. Four culture conditions were tested: a) “stem cell” supporting condition, designated as C4, with 4 early acting cytokines SCF, TPO, FLT3-ligand and IL-6, b) myeloid cell supporting condition C6, with the C4 cytokines plus two additional late acting growth factors IL-1 and IL-3, c) myeloid cell supporting condition C7 with the same cytokines as C6 plus terminal differentiation factor GM-CSF, d) condition adopted from previous studies, designated as C3, containing cytokines TPO, IL-3, G-SCF. Additional negative control condition C-neg did not include MSCs nor any cytokines. Experiments were analyzed after 2-weeks of culture. Phenotypization was performed both pre- and post-culture by flow cytometry.Results:For all AML samples, the tested culture conditions showed successful maintenance of cells in vitro, although with highly heterogeneous expansion rates for the individual samples (0-14 fold expansion, yielding up to 3.5×10^6 cells per well). No AML cells survived in the negative control condition C-neg, for 4/6 samples. The maximum expansion (representing relative yield of 100%) was obtained with the myelo-condition C7 in 5/6 patients. The mean relative yield from all patients for C7 was 97%. The other conditions provided mean relative yields of: C6 - 79%, C4 - 43%, C3 - 44%, and C-neg - 3.5%.Comparison of pre- and post-culture phenotypes revealed differentiation of AML cells in vitro, demonstrated by the decrease of primitive CD34+ fraction and shifting of less differentiated CD33dim cells into mature CD33bright fraction (mean log2 reduction in CD34+ cells: C4 - 0.5×, p=0.4, C7 - 1.4×, p=0.01, C6 - 1.3×, p=0.01, C3 - 1.0×, p=0.03; mean log2 reduction in CD33dim cells: C4 - 0.8×, p=0.2, C7 - 2.0×, p<0.001, C6 - 2.0×, p<0.001, C3 - 1.1×, p=0.07). Expression of CD14 and also CD117 showed large intra-patient differences (both increase and decrease), without statistically significant differences between the culture conditions. CD123 expression remained stable.The CML samples were cultured only with C4, C6, C3 conditions, without the terminal differentiation C7 condition. Expansion reached 33-260 fold; yielding up to 3.3×10^6 cells per well. C3 condition provided the highest yields (mean relative yields: C4 - 29%, C6 - 68%, C3 - 93%, and C-neg - 0%). All three tested conditions produced dominant fraction of more mature myeloid CD33+ cells (mean, 86-94%). Nevertheless, the C4 condition provided the highest proportion of CD34+ cells and produced the least mature CD14+ monocytes, compared to C6 and C3 (mean, CD34+ cell fraction of leukocytes: C4 - 31.7%, C6 - 6.8%, C3 - 4.5%; mean, CD14+ cell fraction of leukocytes: C4 - 6%, C6 - 21%, C3 - 9.6%).Conclusions:All tested culture conditions were capable of maintaining both primary AML and CML cells from all samples, during the 2-week culture. The highest in vitro expansion was achieved with conditions using the late-acting cytokines which however induced the highest cell differentiation as well. The C3 condition reported in previous studies also provided insufficient conservation of the original phenotypes. In this setup, the best preservation of original phenotypes and more primitive cells was achieved using the C4 “stem cell” condition containing early acting cytokines.Supported by Ministry of Health of the Czech Republic, grant nr. 17-30397A. All rights reserved. This report was written with the support of the Specific University Research (nr. MUNI/A/0968/2017) provided by MEYS. DisclosuresMayer:Affimed: Research Funding; Eisai: Research Funding; Johnson & Johnson: Research Funding; Roche: Research Funding; Novartis: Research Funding.