ALBERT

Description: The FLT3 receptor is a potential target in AML due to its role in leukemogenesis and its high degree of expression on blasts from approximately 90% of acute myeloid leukemia (AML) patients. In addition, mutant forms of FLT3, including internal tandem duplications (ITD) in the juxtamembrane region and point mutations in the kinase domain, constitutively activate FLT3 signaling. ITD mutations in particular are also associated with poor prognosis. A number of small molecule tyrosine kinase inhibitors (TKI) against FLT3 are currently in clinical trials and have shown some clinical activity. However, TKIs have various limitations, including their lack of specificity, which may produce toxicities, and can select for drug resistant cells. In an attempt to overcome some of these limitations and to generate new agents which might cooperate in targeting FLT3, we generated a fully humanized phage display monoclonal antibody (EB10). This antibody is capable of inhibiting both ligand-activated wild-type and, to a lesser degree, ligand-independent mutant FLT3 signaling. When EB10 is used to treat cells expressing activated FLT3, inhibition of downstream pathways including STAT5, AKT and MAPK are also frequently seen. EB10 treatment of cells expressing FLT3 in the presence of NK cells leads to antibody-dependent cell-mediated cytotoxicity (ADCC). EB10 treatment of NOD/SCID mice injected with FLT3 expressing AML cell lines or with primary AML blasts significantly prolongs survival and/or reduces engraftment of leukemic cells. EB10 proved efficacious in vivo against cells even when in vitro EB10 treatment did not significantly reduce FLT3 signaling. This indicates that ADCC may be the primary mechanism mediating cytotoxicity as opposed to direct FLT3 inhibition. In contrast to the effects on AML cell lines and primary samples, EB10 treatment did not significantly reduce NOD/SCID engraftment of normal human CD34+ hematopoietic stem cells. Anti-FLT3 antibodies, like EB10, may be a promising therapeutic agent that can specifically target malignant cells with limited toxicities against normal hematopoietic stem cells and should be considered for clinical trials.

Description: The class III receptor tyrosine kinase, FLT3, is expressed by 〉90% of B-lineage acute lymphoblastic leukemias (ALL) blasts. In addition, it is expressed at extremely high levels in ALL patients with MLL-rearrangements or hyperdiploidy and sometimes mutated in these same patients. In this report, we investigated the effects of EB10, an anti-human FLT3 monoclonal antibody capable of preventing binding of FLT3 ligand (FL), on ALL cell lines and primary cells. In vitro studies, examining the ability of EB10 to inhibit FLT3 activation and downstream signaling in ALL cell lines and primary blasts, yielded variable results. In some cell lines FLT3 phosphorylation was inhibited and with it, downstream activation of pathways involving MAPK, AKT, and STAT5 phosphorylation. However, several cell lines actually exhibited FLT3 activation upon antibody treatment, possibly because of antibody-mediated receptor dimerization, and subsequent activation of downstream pathways. Nevertheless, through antibody-mediated cellular cytotoxicity (ADCC) such an antibody could still prove efficacious against leukemia cells in vivo. In fact, EB10 treatment significantly prolongs survival and/or reduces engraftment of ALL cell lines and primary ALL blasts in NOD/SCID mice. This effect might be even more pronounced in a host that was less immune compromised than are NOD/SCID mice. The leukemic cells surviving EB10 treatment in the mice were characterized by FACS analysis and found to express low levels or no FLT3. In contrast to the reduction in engraftment of human ALL primary blasts, EB10 treatment of NOD/SCID mice did not reduce engraftment of human hematopoietic CD34+ cells. Taken together, these data demonstrate that EB10 is selectively cytotoxic to ALL blasts while having little effect on normal hematopoiesis. Such an antibody, either naked or conjugated to radioactive isotopes or cytotoxic agents, may prove useful in the therapy of infant ALL as well as childhood and adult ALL patients whose blasts typically express FLT3.

Description: FMS-like tyrosine kinase 3 (FLT3), a class III receptor tyrosine kinase, is expressed at high levels in the blasts of approximately 90% of patients with acute myelogenous leukemia (AML). Internal tandem duplications (ITDs) in the juxtamembrane domain and point mutations in the kinase domain of FLT3 are found in approximately 37% of AML patients and are associated with a poor prognosis. We report here the development and characterization of a fully human anti-FLT3 neutralizing antibody (IMC-EB10) isolated from a human Fab phage display library. IMCEB10 (immunoglobulin G1 [IgG1], κ) binds with high affinity (KD = 158 pM) to soluble FLT3 in enzyme-linked immunosorbent assay (ELISA) and to FLT3 receptor expressed on the surfaces of human leukemia cell lines. IMC-EB10 blocks the binding of FLT3 ligand (FL) to soluble FLT3 in ELISA and competes with FL for binding to cell-surface FLT3 receptor. IMC-EB10 treatment inhibits FL-induced phosphorylation of FLT3 in EOL-1 and EM3 leukemia cells and FL-independent constitutive activation of ITD-mutant FLT3 in BaF3-ITD and MV4;11 cells. Activation of the downstream signaling proteins mitogen-activated protein kinase (MAPK) and AKT is also inhibited in these cell lines by antibody treatment. The antibody inhibits FL-stimulated proliferation of EOL-1 cells and ligand-independent proliferation of BaF3-ITD cells. In both EOL-1 xenograft and BaF3-ITD leukemia models, treatment with IMC-EB10 significantly prolongs the survival of leukemia-bearing mice. No overt toxicity is observed with IMC-EB10 treatment. Taken together, these data demonstrate that IMC-EB10 is a specific and potent inhibitor of wild-type and ITD-mutant FLT3 and that it deserves further study for targeted therapy of human AML. (Blood. 2004;104:1137-1144)

Description: The FLT3 receptor tyrosine kinase is highly expressed in most acute leukemias and frequently mutated in acute myeloid leukemia (AML). The mutated form of the receptor is constitutively activated and known to play an important role in AML, but the activation state of the overexpressed wild-type (wt) receptor is, at present, unknown. In this study, we examined the activation state of the wild-type receptor in AML. We found that the wild-type receptor was constitutively phosphorylated/activated in 8 of 12 primary AML samples and 4 of 13 leukemia cell lines. To explain why wtFLT3 is often activated, we investigated the expression of its ligand, FL, by these same cells. Coexpression of FL with FLT3 was a universal finding in both primary AML samples and leukemic-derived cell lines. To further prove that autocrine signaling was accounting for the activation, we showed that conditioned media but not fresh media was able to activate FLT3. In addition, an antibody that blocks binding of ligand to the receptor blocks FLT3 activation. Finally, depletion of FL from conditioned media is able to block the activation of FLT3. Taken together, these findings represent strong evidence that wtFLT3 is often constitutively activated in AML and thus, like its mutated form, might contribute to the altered signaling that characterizes leukemogenesis.

Description: FLT3 is a class III receptor tyrosine kinase overexpressed by blast cells in approximately 90% of acute myeloid leukemia and acute lymphoid leukemia (ALL) patients. Recent findings testing a small molecule FLT3 inhibitor (PKC412) in combination with eight conventional antileukemic agents, reported that methotrexate in particular did not combine well with PKC412 against leukemia cells (Furukawa, Y. et al., Leukemia 2007, 21: 1005–1014). IMC-EB10 is a fully human neutralizing antibody that binds FLT3 with high affinity (Kd 158 pM) and blocks FLT3 ligand binding to FLT3 (IC50 ~10 nM), leading to inhibition of MAPK, STAT5, and PI3K/Akt signaling in leukemia cells. Here we test for benefits of adding methotrexate to IMC-EB10. In an SEMK2 model of ALL we evaluated survival effects of treatments administered intraperitoneally, beginning 1 day after after intravenous injection of 5 × 10e5 cells/mouse. In this model the maximum tolerated dose (MTD) of methotrexate administered once per week for 3 weeks, was determined to be 100 mg/kg. At this MTD, Methotrexate increased median survival to 44 days, compared to 30 days in the control group (p

Description: The receptor tyrosine kinase FLT3 is overexpressed in blasts of ~90% of acute myelogenous leukemia (AML) and the majority of B-lymphoid leukemia patients. Internal tandem duplications (ITDs) in the juxtamembrane region and point mutations in the kinase domain of FLT3 are found in ~37% of AML patients and are associated with a poor prognosis. We have recently developed a fully human monoclonal antibody (IMC-EB10) which binds with high affinity to FLT3 receptor on human leukemia cells. In the present study, a novel auristatin conjugate of the anti-FLT3 antibody (EB10-MMAF) was prepared using a dipeptide linker that allows for drug release inside the lysosomes of antigen-positive cells. The MMAF conjugates were stable in buffers and plasma. EB10-MMAF (drug/antibody raito = 8) was highly potent, and selectively inhibited the growth of FLT3-expressing leukemia cells with an IC50 of 0.19 nM and 0.08 nM for MV4;11 and BaF3-ITD cells (both positive for FLT3-ITD), 1.11 nM, 6.18 nM and 1.82 nM for REH , EOL-1, EM3 cells (all three positive for wild-type FLT3), and 135 nM for JM1 (negative for FLT3). An MMAF conjugate with a control antibody was not active in these cell lines (IC50s 〉 5.9 uM). Flow cytometric analysis with annexin V indicated that EB10-MMAF treatment induced apoptosis of leukemia cells in vitro. In vivo treatment with EB10-MMAF strongly inhibited leukemia growth and prolonged survival of mice in both EOL-1 and BaF3-ITD leukemia models. In summary, immunoconjugates composed of a fully human anti-FLT3 antibody and a potent auristatin drug may provide a valuable therapeutic approach for AML and other FLT3-positive leukemias.

Description: Introduction: Donor plasmacytoid dendritic cells (pDCs) have important roles in the pathogenesis of graft versus host disease (GvHD) after allogeneic hematopoietic stem cell transplantation (allo-HSCT). Vasoactive intestinal peptide (VIP), a potent anti-inflammatory neuropeptide with pleiotropic effects, has shown promise in managing GvHD through induction of regulatory dendritic cells (Chorny, Blood 2006) and regulatory T cells (Delgado, JLB 2005). Our previous data showed that transplantation of hematopoietic stem cells (HSC) lacking VIP expression in enhanced Th1 polarization and antiviral immunity (Li, JI 2011). Whether VIP produced by pDCs plays a key role in preventing GvHD is the focus of the current study. Method: A reductive allo-HSCT model was established by transplanting purified populations of 5 x 103 hematopoietic stem cells (HSC), 5 x 104 pDCs and 106 T cells from B6 (H-2Kb) to MHC mismatched B10.BR (H-2Kk) recipients. Mice received either VIP knockout pDCs (experimental group) or wild-type pDCs (control). Survival and clinical GvHD scores were monitored for at least 60 days. Serum pro-inflammatory cytokines were measured by luminex on day 3 and day 8. Additional recipients were transplanted for interim analyses of GvHD pathology in histological sections of small intestine and colon on day 8 and day 15. To investigate pDC migration and proliferation status, 5 x 104 luciferase-expressing pDCs, together with 5 x 103 WT B6 HSC, were transplanted to MHC mismatched Balb/c (H2kd) recipients. Bioluminescent imaging was performed on day 14. We used mice carrying an eGFP reporter linked to a VIP promoter to measure VIP production by donor cells in vivo. The construct has the eGFP reporter gene, followed by a polyadenylation sequence, inserted into BAC clone RP23-25A8 at the initiating ATG codon of the first coding exon of the VIP gene, so that eGFP expression is driven by the regulatory sequences of the BAC gene. The BAC is inserted randomly in the genome. We used serial breeding to establish the eGFP-VIP reporter in B6 mice. Bone marrow pDCs were sorted from transgenic eGFP-VIP B6 mice and used in transplant experiments, allowing VIP gene activity from donor-derived pDCs to be detected as GFP expression using confocal microscopy. Results: Compared to recipients of WT pDCs, recipients of VIP-KO pDCs have a significantly lower survival rate with increased GvHD clinical score (Figure 1; *p

Description: Immunoglobulin secreted by myeloma cells contains a unique antigenic determinant (idiotype [Id]) that may serve as a tumor-specific antigen. Although Id-protein–specific T-cell responses have been reported in patients with myeloma, it is not known whether primary myeloma tumor cells can present naturally processed Id peptides on their surface as a target. We immunized 2 healthy human stem-cell donors with Id proteins from their recipients. T cells from the immunized donors released high levels of T-helper 1–type cytokines in response to stimulation with myeloma cells from their recipients. The T-cell–mediated cytokine response to tumor cells was blocked by a major histocompatibility complex (MHC) class I monoclonal antibody, whereas the response to soluble Id protein was dependent on MHC class II. To investigate whether Id-specific CD8+ T cells can recognize and kill autologous myeloma cells, we generated T cells from peripheral blood mononuclear cells from a third patient with myeloma by means of in vitro stimulation with autologous dendritic cells pulsed with Id protein. Tumor-specific lysis of myeloma cells was demonstrated by the lack of killing of autologous nonmalignant B cells or natural killer–sensitive K562 cells. Lysis of autologous myeloma targets was restricted by MHC class I molecules. These data represent the first report of class I–restricted T-cell recognition of fresh autologous myeloma targets and formally demonstrate that human myeloma cells can serve as targets of an Id-specific T-cell response.

Description: Immunoglobulin secreted by myeloma cells contains a unique antigenic determinant (idiotype [Id]) that may serve as a tumor-specific antigen. Although Id-protein–specific T-cell responses have been reported in patients with myeloma, it is not known whether primary myeloma tumor cells can present naturally processed Id peptides on their surface as a target. We immunized 2 healthy human stem-cell donors with Id proteins from their recipients. T cells from the immunized donors released high levels of T-helper 1–type cytokines in response to stimulation with myeloma cells from their recipients. The T-cell–mediated cytokine response to tumor cells was blocked by a major histocompatibility complex (MHC) class I monoclonal antibody, whereas the response to soluble Id protein was dependent on MHC class II. To investigate whether Id-specific CD8+ T cells can recognize and kill autologous myeloma cells, we generated T cells from peripheral blood mononuclear cells from a third patient with myeloma by means of in vitro stimulation with autologous dendritic cells pulsed with Id protein. Tumor-specific lysis of myeloma cells was demonstrated by the lack of killing of autologous nonmalignant B cells or natural killer–sensitive K562 cells. Lysis of autologous myeloma targets was restricted by MHC class I molecules. These data represent the first report of class I–restricted T-cell recognition of fresh autologous myeloma targets and formally demonstrate that human myeloma cells can serve as targets of an Id-specific T-cell response.

Description: Introduction: In allogeneic hematopoietic stem cell transplant (AHSCT), donor T cells promote hematopoietic engraftment, reconstitute T-cell immunity and mediate potent beneficial antitumor effects, such as graft versus leukemia (GVL) as well as detrimental graft-versus-host disease (GVHD). We have shown that purified bone marrow pDC to a donor graft composed of purified HSC and T cells significantly improved long-term leukemia-free survival without increasing the risk of GVHD (Lu BLOOD 2012). Vasoactive intestinal peptide (VIP) is a neuropeptide/neurotransmitter, which acts as a major anti-inflammatory factor in animal models of inflammatory and autoimmune diseases. VIP is produced by T cells and dendritic cells under conditions of inflammation (Li Cancer Research 2016). VIP-signaling, thus represents a newly appreciated co-inhibitory pathway involved in T cell activation and expansion and persistence of antigen-specific T cells, but the role of VIP produced by donor dendritic cells in allo-BMT is unclear. Our previous data has shown that production of VIP in pDC improves survival in a murine allo-BMT model. Since the thymus plays a critical role in regenerating naive T cells in which allo-reactive donor T cells undergoes central deletion. We hypothesized that donor pDCs that home to thymus modulate the negative selection of allo-reactive T cells and iTreg production through VIP signaling. Methods and Results: Our previous data has shown that the mice receiving HSC, T cells and WT pDC had a significantly higher survival (71%) compared to those receiving VIP-KO pDC (31%). On day 15 after transplant, recipients of WT pDC, VIP-KO pDC and no pDC developed ~98% chimerism, without significant differences among the three groups. Local production of VIP in pDC inhibited activation and Th1 immune polarization of donor T cells. Recipient spleens were harvested on day 15 after transplant for analysis of cytokine production by donor T cells. The percentage of CD8+ donor T cells producing IL17 was significantly higher in recipients of VIP-KO pDC compared to recipients of WT-pDC (Fig 2.A-B). The ratio of foxp3+ CD4+ donor T-reg to IL17+ CD4+ T cells from recipients of WT pDC was almost 3 times higher than the recipients of VIP-KO pDC (Fig 2.C-D). To visualize production of VIP by donor pDC, B10.BR mice were transplanted with 5,000 stem cells, 1M T cells and 50,000 pDCs from VIP-GFP (VIP promoter and GFP reporter) or GFP mice. On day 7 post-transplant, the thymus was examined with confocal microscopy with GFP (green), anti-PDCA-1-Alexa Fluor 568 (red) and DAPI (blue). A superimposed profile of the thymus showed that donor GFP pDCs homed to thymus (Fig.2-A), and that donor pDC in the recipient thymus produced VIP (Fig.2-B). Conclusion: Expression of VIP in donor pDCs inhibited activation and Th17 immune polarization of donor CD8+ T cells after allo-BMT. Foxp3+ expression tended to be higher among CD4+ donor T cells from recipients of WT pDCs compared with recipients of VIP-KO pDCs. Thus, VIP-producing donor pDC in thymus could be very critical to contribute to negative selection of allo-reactive donor T cell or facilitate the generation of Foxp3+ nTreg. This data supported a new mechanism by which GvHD maybe regulated and central tolerance maintained. Ongoing experiments aim at defining the role of VIP production in the thymus by donor pDC in positive and negative T cell selection. Disclosures Waller: Celldex: Research Funding; Novartis Pharmaceuticals Corporation: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Cambium Medical Technologies: Consultancy, Equity Ownership; Kalytera: Consultancy; Pharmacyclics: Other: Travel Expenses, EHA, Research Funding.