Erratum: A novel BCMA/CD3 bispecific T-cell engager for the treatment of multiple myeloma induces selective lysis in vitro and in vivo
Abstract B-cell maturation antigen (BCMA) is a highly plasma cell-selective protein that is expressed on malignant plasma cells of multiple myeloma (MM) patients and therefore is an ideal target for T-cell redirecting therapies. We developed a bispecific T-cell engager (BiTE) targeting BCMA and CD3ɛ (BI 836909) and studied its therapeutic impacts on MM. BI 836909 induced selective lysis of BCMA-positive MM cells, activation of T cells, release of cytokines and T-cell proliferation; whereas BCMA-negative cells were not affected. Activity of BI 836909 was not influenced by the presence of bone marrow stromal cells, soluble BCMA or a proliferation-inducing ligand (APRIL). In ex vivo assays, BI 836909 induced potent autologous MM cell lysis in both, newly diagnosed and relapsed/refractory patient samples. In mouse xenograft studies, BI 836909 induced tumor cell depletion in a subcutaneous NCI-H929 xenograft model and prolonged survival in an orthotopic L-363 xenograft model. In a cynomolgus monkey study, administration of BI 836909 led to depletion of BCMA-positive plasma cells in the bone marrow. Taken together, these results show that BI 836909 is a highly potent and efficacious approach to selectively deplete BCMA-positive MM cells and represents a novel immunotherapeutic for the treatment of MM.
Abstract Introduction Constitutive and immunoproteasome inhibitors (C&IPI) were thought to suppress nuclear factor-κB (NF-κB) pathway by preventing IκB degradation, which prevents NF-κB translocation into the nucleus. This mechanism of action has since been questioned by a number of studies. First, bortezomib promoted constitutive NF-κB activity in endothelial cell carcinoma. Second, NF-κB constitutive activity was resistant to bortezomib in multiple myeloma cell lines. Third, bortezomib increased IκB mRNA but post-transcriptionally downregulated IκB in normal cells and in multiple myeloma cell lines resulting in induced canonical NF-κB activation. Lastly, bortezomib increased nuclear levels of IκB as opposed to lowering cytoplasmic levels in cutaneous T cell lymphoma cell line suggesting that nuclear translocation of IκB was possibly responsible for NF-κB inhibition. The inhibitory activity of C&IPI on dendritic cells (DC) is of interest in the prevention of graft versus host disease (GvHD). It has been shown that different C&IPI impede DC maturation and T cell priming both in vitro and in vivo. Herein we sought to understand the mechanism of action of proteasome and immunoproteasome inhibitors on DC and to test their effect on IκB and NF-IκB expression. Materials and Methods We first performed RT PCR on lysates of DC obtained from the peripheral blood of 7 patients who received post-transplant cyclophosphamide and bortezomib as prevention of GvHD on a phase I clinical trial. Patients received allogeneic transplantation from matched-related or unrelated donors. Patients received no other immunosuppressive therapy except for rabbit anti-thymocyte globulin for those receiving graft from unrelated donor. Steroids were not allowed on the study. Samples were obtained on days +1, +4, and +7. The results were analyzed in comparison to samples obtained on day 0 before stem cell infusion. We then performed the same experiment on lysates of DC obtained from the peripheral blood of healthy volunteer donors. DC were untreated or incubated with bortezomib (10 nM for 4 h), carfilzomib (30 nM for 1 h), oprozomib (100 nM and 300 nM for 4 h), ONX 0914 (200 nM for 1 h), PR-825 (125 nM for 1 h), or PR-924 (1000 nM for 1 h). The drug concentration and duration of exposure were chosen based on the IC50 on proteasome activity and to reproduce in vivo conditions. We also performed IκB western blot on DC isolated from peripheral blood of healthy volunteers, untreated or incubated with bortezomib (10 nM for 4 h) or oprozomib (300 nM for 4 h). Each experiment was performed at least in triplicate. Results We found that the combination of cyclophosphamide and bortezomib significantly and progressively increased IκB mRNA while decreasing NF-κB mRNA in DC studied ex vivo. We also found that all studied C&IPI increased IκB mRNA to a variable degree while only oprozomib (300 nM) decreased NF-κB mRNA in DC in vitro. Finally, both bortezomib and oprozomib increased IκB protein level in DC in vitro (figure). Conclusion Our data suggest that C&IPI increase IκB expression in DC. As opposed to the previously reported data in other cell types, the effect is not associated with post-transcriptional downregulation. Cyclophosphamide and bortezomib also decrease NF-κB expression in DC in vivo while only oprozomib had the same effect in vitro. The effect of C&IPI on IκB and NF-κB expression may represent a new mechanism of action and suggests their effect may be cell-type dependent. Disclosures: Al-Homsi: Millennium Pharmaceuticals: Research Funding. Off Label Use: The use of cyclophosphamide and bortezomib for GvHD prevention. Lai:Millennium Pharmaceuticals: Research Funding.
Abstract Despite recent therapeutic developments, multiple myeloma remains an incurable plasma cell malignancy. Poor prognosis for myeloma patients relapsing post-transplant calls for the need for novel treatment options. Immunotherapy with engineered T cells has proven highly efficacious against B-cell cancers, and early-phase clinical trials suggest that multiple myeloma is susceptible to this form of therapy. We designed a new chimeric T cell receptor, T cell antigen coupler (TAC), which relies upon activation through endogenous T cell receptor complex, thus allowing engineered T cells to auto-regulate their activity (Helsen et al, Nat. Comm., 2018). Using published single-chain antibody fragments (scFvs) C11D5.3 and J22.9-xi, we generated B cell maturation antigen (BCMA)-specific TAC receptors for targeting multiple myeloma. Primary human T cells were transduced with lentiviral vectors carrying different BCMA TAC constructs and assessed for in vitro functionality via cytokine production, cytotoxicity, and proliferation assays. In vivo efficacy and T cell tracking were performed in an established orthotopic xenograft mouse model based on a BCMA-positive KMS-11 cell line. C11D5.3 and J22.9-xi TAC T cells demonstrated comparable in vitro performance with both types of cultures efficiently killing BCMA-expressing targets, producing IFN-γ, TNF-α, and IL-2 cytokines, and undergoing multiple rounds of proliferation. In vivo, TAC T cells carrying either scFv were capable of curing mice bearing disseminated myeloma; however, the TAC T cells carrying J22.9-xi scFv were more potent on a per-cell basis (Figure 1A, top panel). Mice in remission 3 months post-treatment with a single dose of 106 TAC-positive T cells showed evidence of sustained anti-tumor protection upon rechallenge with a fresh dose of 106 KMS-11 tumor cells (Figure 1B). Mice treated with low-dose J22.9-xi T cells were more resistant to rechallenge than mice treated with a comparable dose of C11D5.3 TAC T cells. Tracking of the TAC T cells in vivo revealed that the J22.9-xi TAC T cells expanded to a much larger extent than the C11D5.3 TAC T cells (Figure 1A, bottom panel), indicating that there were likely more J22.9-xi TAC T cells present at the time of tumor rechallenge. To understand whether biological aspects of BCMA may influence the proliferative response of the TAC T cells, we explored the influence of APRIL, the soluble ligand for BCMA, on TAC T cell proliferation in vitro. Strikingly, despite comparable proliferation of both TAC T cell populations following stimulation with KMS-11 tumor cells in the absence of APRIL in vitro, the presence of APRIL had a strong inhibitory effect on proliferation of C11D5.3 TAC T cells and only a modest inhibitory effect on J22.9-xi TAC T cells. Our preclinical findings support further development of TAC T cells for the treatment of multiple myeloma and underscore the importance of T cell expansion in determining the therapeutic activity of engineered T cells. This work further reveals a novel observation that the natural ligand of BCMA can impair the therapeutic impact of T cells engineered with chimeric receptors directed against BCMA and provide a basis for advancing BCMA-specific TAC T cells into the clinic. Disclosures Denisova: Triumvira Immunologics: Patents & Royalties. Afsahi:Triumvira Immunologics: Patents & Royalties. Helsen:Triumvira Immunologics: Employment, Patents & Royalties. Bramson:Triumvira Immunologics: Employment, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties, Research Funding.
AbstractBispecific T cell engaging antibodies (BiTEs) address tumor associated antigens that are over-expressed on cancer but that can also be found on healthy tissues, causing substantial on-target/off-tumor toxicities. To overcome this hurdle, we recently introduced hemibodies, a pair of complementary antibody fragments that redirect T cells against cancer-defining antigen combinations. Here we show that hemibodies addressing CD38 and SLAMF7 recruit T cells for the exquisite elimination of dual antigen positive multiple myeloma cells while leaving single antigen positive bystanders unharmed. Moreover, CD38 and SLAMF7 targeting BiTEs, but not hemibodies induce massive cytokine release and T cell fratricide reactions, a major drawback of T cell recruiting strategies. Together, we provide evidence in vitro and in vivo that hemibodies can be developed for the effective and highly specific immunotherapy of multiple myeloma.
Dendritic cells (DCs) are potent antigen presenting cells that regulate the development of both innate and adaptive immune responses. According to their maturation status, DCs may ignite immune response or induce immune tolerance. Indeed, immature DCs (iDCs) present low levels of costimulatory molecules such as CD80 and CD86, and high levels of tolerogenic molecules such as B7H3. Upon exposure to maturation stimuli, DCs upregulate CD83 on their surface and gain the competence of stimulating T cell response. An efficient maturation is crucial for the generation of a specific cytotoxic T lymphocyte response, specially against cancer. However, recent reports have shown that Multiple Myeloma (MM) milieu can recruit DCs and reprogram them to sustain growth and survival of MM cells and protect them against immune response. Therapeutic approaches to restore DC functions rely on the identification of the pathways that are directly involved in induction of tolerance. Emerging evidence supports the role of microRNAs (miRNA) in the regulation of immune response and DC function. Among others, the miR-29 family seems to be involved in the modulation of NK activity, of Th1/Th2 phenotype switch and of DC differentiation from monocyte precursors. Besides, miR-29b targets and inhibits different and crucial immune-modulatory molecules such as B7H3, VEGF and IL-4. These findings suggest that miR-29 may play an important role in the multifaceted interplay between tumor cells and host’s immune system. To address this hypothesis, we generated iDCs from CD14+ monocytes of healthy donors and co-cultured them with: i) allogeneic (allo-) lymphocytes; ii) VEGF and IL-6 producing MM cells (RPMI8226 and U266); iii) allo-lymphocyte and MM cells. We found a consistent increase of miR-29b expression by RT-PCR during differentiation and maturation of DCs induced by allo-lymphocytes. However, when immature DCs were co-cultured with MM cells +/- allo-lymphocytes, a significant 3-fold reduction of miR-29b levels (p= 0.02) was observed (fig 1). This event occurred together with the absence of maturation markers, the persistence of high levels of B7H3 on the cell surface and with a raise in VEGF, IL-10, and IFN-gamma levels in the supernatant, confirming the MM-dependent impairment of the physiological DC maturation process. This latter concept was supported by the finding of increased number of CD4+CD25+Foxp3+T-regs in the DC/MM cell/allo-lymphocyte co-cultures as compared to the DC/allo-lymphocyte co-cultures (p= 0.05). To promote the recovery from the MM related immune-bias, we transiently transfected iDCs with miR-29b (29b-DCs) mimics or with a negative control (NC-DCs). We observed improved DC maturation (82.46% versus 39.89% of CD11c+/CD83+/CD86+ cells), reduced expression of B7H3 (33% reduction in MFI) and reduction of the T-reg number in 29b-DC/MM cell/allo-lymphocyte co-cultures as compared to NC-DC/MM cell/allo-lymphocyte co-culture. To investigate whether 29b-DCs were able to promote a specific CTL response against MM cells in vivo, we engrafted NOD/SCID γ chain-null mice with peripheral blood mononuclear cells (PBMCs) from HLA-A2+ healthy donors. DCs from the same donor were differentiated, transfected with either miR-29b or NC and then co-cultured with U266 for 48h. Mice were then vaccinated twice with either 29b-DCs or NC-DCs. Two weeks following the first injection, CD3+ human lymphocytes were recovered from mouse spleens (CD3 hu-splenocytes). We found an increased CD8/CD4 ratio in the CD3 hu-splenocytes collected from the 29b-DCs treated mouse as compared to control. To assess the capability of CD3+ hu-splenocytes to selectively kill U266 cells, we kept CD3 lymphocytes in culture in the presence of IL-15 for 48h. Then, we carried a cytotoxicity assay against U266 cell target. The highest specific lysis was attained with miR-29b DC primed CD3 hu-splenocytes (fig.2, p=0.03). Taken together, our data indicate that: a) miR-29b regulates DC differentiation/maturation and function; b) MM cells reduce the expression of miR-29b in DCs, thus contributing to the establishment of an immune-permissive microenvironment; c) replacement of miR-29b within DCs partially restores their differentiation and functions in vitro and their capability to induce antitumor specific T-cell response in vivo. On these findings, miR-29b mimics are attractive candidates to enhance immunotherapy approaches against MM. Disclosures: No relevant conflicts of interest to declare.
Abstract Chimeric-antigen receptor (CAR)-T cell immunotherapies have been remarkably effective in treating acute lymphoblastic leukemia. However, current strategies generally suffer from difficult, inefficient and costly manufacturing processes, significant patient side effects and poor durability of response in some patients. Here, we report for the first time a CAR-T cell therapeutic comprising a non-immunoglobulin alternative scaffold Centyrin molecule (a "CARTyrin") manufactured with a novel non-viral piggyBacTM (PB) transposon-based system. Our lead candidate, P-BCMA-101, encodes a CARTyrin that targets the B cell maturation antigen (BCMA) for the treatment of multiple myeloma (MM) and has several unique aspects that improve upon earlier CAR-T products. First, P-BCMA-101 is manufactured using only in vitro transcribed mRNA and plasmid DNA without the need for lentivirus or g-retrovirus, resulting in time and cost savings. Importantly, PB is also safer than viral systems due to a less mutagenic insertional profile and is non-oncogenic. Furthermore, PB can efficiently deliver transgenes as large as several hundred kilobases, and, once inserted, transgenes demonstrate more stable, prolonged and higher expression when compared to those delivered by virus. Second, a mutein of the dihydrofolate reductase (DHFR) gene is included in the P-BCMA-101 transgene that can be used in combination with the non-genotoxic drug methotrexate (MTX) to provide a simple and effective method of CARTyrin+ cell enrichment and reduces variability in patient product material. Third, P-BCMA-101 incorporates a safety switch for optional depletion in vivo in case of adverse events. Lastly, the CARTyrin is comprised of a BCMA-specific Centyrin, which are based on a human tenascin fibronectin type III (FN3) consensus sequence. Centyrins have similar binding affinities to the antibody-derived single chain variable fragments (scFv), but are smaller, more thermostable and predicted to be less immunogenic. Importantly, no signs of tonic signaling leading to T cell exhaustion have been observed with CARTyrins unlike scFv-based CAR molecules, which can interact with each other on the surface causing non-specific CAR signaling. The manufacture process of P-BCMA-101 from primary human T cells is straightforward, employs no cytokines, and easily produces enough CARTyrin+ cells to treat patients. Within 18 days of electroporation of purified T cells, we demonstrate > 95% of the cell product is positive for CARTyrin expression and ready to be administered. Notably, our manufacturing process results in > 60% of CARTyrin+ T cells exhibiting a stem-cell memory phenotype (i.e. CD45RA+ CD62L+). P-BCMA-101 cells exhibit specific and robust in vitro activity against BCMA+ tumor targets, ranging from high to very low levels of BCMA, as measured by target-cell killing and CARTyrin-T cell proliferation. Importantly, proliferating P-BCMA-101 cells were highly sensitive in vitro to activation of the safety switch. Finally, we have evaluated the anti-tumor efficacy of P-BCMA-101 in a model of human MM. NSG™ mice were injected IV with 1.5x106 luciferase+ MM.1S cells, an aggressive human MM-derived cell line. After the tumor cells were allowed to grow for 21 days, animals received a single IV administration of 5x106 P-BCMA-101 cells. All untreated control animals demonstrated a marked increase in serum M-protein levels, rapid growth of tumor cells demonstrated by bioluminescent imaging (BLI), and death within four weeks. In stark contrast, 100% of animals that received P-BCMA-101 rapidly eliminated tumors within 7 days as measured by BLI and serum M-protein levels and improved survival out to at least 60 days post-treatment. P-BCMA-101 is the first-in-class of Centyrin-based CAR therapeutics. The CARTyrin, combined with our advanced manufacturing processes, represents a significant improvement over first generation, immunoglobulin-based and virally-transduced CAR-T products. P-BCMA-101 exhibited an advantageous stem-cell memory phenotype and demonstrated specific and potent anti-tumor efficacy against BCMA+ myeloma cells both in vitro and in vivo. Based on these results, we plan to initiate a phase I clinical trial of P-BCMA-101 for the treatment of patients with relapsed and/or refractory MM. Disclosures Hermanson: Poseida Therapeutics: Employment. Barnett:Poseida Therapeutics: Employment. Rengarajan:Poseida Therapeutics: Employment. Codde:Poseida Therapeutics: Employment. Wang:Poseida Therapeutics: Employment. Tan:Poseida Therapeutics: Employment. Martin:Poseida Therapeutics: Employment. Smith:Poseida Therapeutics: Employment. Osertag:Poseida Therapeutics: Employment, Equity Ownership. Shedlock:Poseida Therapeutics: Employment.
Abstract Background: We recently described the T cell antigen coupler (TAC) technology (Helsen et. al. Nature Communications) which is a chimeric receptor that targets antigens in an MHC-independent fashion and activates T cells by co-opting the natural TCR receptor. In vitro and in vivo assessments of TAC T cells in solid tumor models have revealed that TACs mediate biological effects that are distinct from conventional chimeric antigen receptors (CARs) and offer safety advantages, including greater target selectivity and reduced off-target toxicity. Here, we present in vitro and in vivo data showing that TAC-engineered T cells directed against CD19 and BCMA demonstrate robust anti-tumor efficacy in haematological malignancies with no detectable side effects. Materials and Methods: T cells from health donors were engineered with TAC receptors directed against CD19 or BCMA using lentivirus vectors. Flow cytometry was employed to measure surface expression of TAC receptors, cytokine production and proliferation of TAC T cells following stimulation with relevant target cells. Antigen-specific toxicity was measured using a luciferase-based killing assay. Anti-tumor activity was measured against acute lymphoblast leukemia for CD19 and multiple myeloma for BCMA xenografts in immunodeficient NRG mice. Results: Engineering T cells with TAC receptors targeted against either CD19 or BCMA revealed antigen-specific activation of cytokine production, cytotoxic function and proliferation. TAC T cells, but not CAR engineered T cells, show significant selectivity towards the context of antigen presentation. This is reflected by the differential proliferative response to a diverse framework of antigen surface arrangement, potentially indicating that TAC T cells are less susceptible to off target activation and the resulting toxicities. Treatment of established NALM-6 xenografts (acute lymphoblastic leukemia) and KMS-11 xenografts (multiple myeloma) with CD19 TAC T cells and BCMA TAC T cells, respectively, resulted in clearance of tumors within a few weeks of T cell infusion. Mice that cleared tumors following TAC T cell treatment were resistant to subsequent challenge with fresh tumor cells demonstrating persistence of TAC T cells. Treatment with control TAC T cells that carry no binding domain had no impact on tumor growth. Monitoring of TAC T cells post-infusion revealed robust expansion that peaked in the peripheral blood 1-2 weeks post-infusion. A clinical manufacturing protocol has been developed for the CD19 TAC T cells in anticipation of human trials. Conclusion: Our pre-clinical evaluation suggests that TAC therapy has the potential to becoming a safer and more effective alternative to conventional CAR therapy. A first in human Phase I/II trial with CD19 TAC T cells is expected to start in the first half of 2019. Disclosures Helsen: Triumvira Immunologics: Employment, Patents & Royalties. Hammill:Triumvira Immunologics: Other: Holding shares, Patents & Royalties. Mwawasi:Triumvira Immunologics: Other: Holding shares, Patents & Royalties. Hayes:Triumvira Immunologics: Employment. Afsahi:Triumvira Immunologics: Patents & Royalties. Denisova:Triumvira Immunologics: Patents & Royalties. Bramson:Triumvira Immunologics: Employment, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties, Research Funding.
Introduction Bispecific T-cell engager (BiTE) antibodies represent a novel therapeutic option for patients with multiple myeloma (MM). BiTE antibodies lack Fc region, and have variable domain only, they can simultaneously bind to two different epitopes i.e. cluster of differentiation 3 (CD3) molecules on tumor-specific T cells, and a specific antigen on myeloma cells, which leads to T-cell dependent destruction of myeloma cells. Currently, blinatumomab, specific for CD3 and CD19 is the only Food and Drug Administration FDA approved BiTE antibody for clinical use in patients with relapsed/refractory (RR) B-cell acute lymphoblastic leukemia, several similar BiTE antibodies are under development. Methods Following PRISMA guidelines, we performed comprehensive literature on 4/15/19 cross-referencing the terms "bispecific antibodies" and "multiple myeloma" using PubMed, Embase, Web of Science, Cochrane Library, Clinicaltrials.gov and review of international medical meeting abstracts. Initially, 256 articles were identified and after detailed scrutiny, one phase 1 clinical trial with prelim results, 4 preclinical and 4 ongoing clinical trials were included. Results Preclinical trials: Anti-BCMA x Anti-CD3 Bispecific Antibody: BiTE antibodies are still in early development in MM, and most of the published data is about the pre-clinical phase. In preclinical trials, Hipp et al. 2017 and Cho et al. 2018 reported that AMG 420 (BI 836909) and AMG 701, which are anti CD3 and B-cell maturation antigen (BCMA), are highly efficacious in vitro in the killing of myeloma cells and potently induces autologous tumor cell lysis in patients with both newly diagnosed and RRMM regardless of their disease status. In mouse xenograft models reconstituted with human T cells, in vivo efficacy of AMG 420 was reported with an overall response in 6 of 10 animals, with all 6 responders became tumor-free at the end of the study. In an orthotopic L-363 xenograft model, treatment with AMG 420 resulted in prolonged median survival of 43-43.5 days. Dilillo et al. 2018 and Ji Li et al. 2017 reported similar in vivo results for REGN5458 and BFCR4350A respectively. Clinical trials: Currently, there are 5 phase 1 ongoing clinical trials (Table 1). Updated results of only first in human phase I AMG 420 are available. Forty-two MM patients with a high tumor burden and four prior lines of therapy were given 2.5 treatment cycles with AMG 420. Overall thirteen (31%) patients responded to AMG 420 therapy, with complete response (CR) in 6 (14.2%) patients, very good partial response (VGPR) in 2 (5%) patients and partial response (PR) in 2(5%) patients. Eleven of these patients responded in the first treatment cycle, with a median response time of 1 month. Twenty-five (57.1%) patients discontinued treatment due to progressive disease. Four deaths were reported; 2 from disease progression and 2 due to adverse events; neither of them was treatment-related. Serious adverse events were reported in twenty-one (50%) patients, the infection was reported in twelve (29%) and polyneuropathy in three (7%), eighteen (43%) required hospitalization. Treatment-related serious adverse events included three (7%) patients with grade 2-3 cytokine release syndrome, three (7%) with polyneuropathy and one (2.3%) with edema. Conclusion After the success of naked antibodies like daratumumab and elotuzumab for MM, there is a need to develop immunotherapy using conjugated antibodies and BiTE antibodies to overcome the challenge of MM resistance and relapse to prior therapies. Preclinical data with BiTE antibodies are promising; AMG 420 anti-CD3/BCMA BiTE has already been granted fast track status by the FDA. We anticipate that drug will enter phase 2 clinical trials for drug development against RRMM Other BiTE antibodies with strong preclinical efficacy are under development and data from larger prospective clinical trials is needed to explore their efficacy in the treatment of multiple myeloma. Table 1 Disclosures Anwer: In-Cyte: Speakers Bureau; Seattle Genetics: Membership on an entity's Board of Directors or advisory committees.
8044 Background: ISB 1342 is a bispecific antibody heterodimer based on the Ichnos proprietary Bispecific Engagement by Antibodies based on T cell receptor (BEAT) platform. ISB 1342 is a first-in-class CD38 T cell engager under investigation in subjects with relapsed multiple myeloma refractory to proteasome inhibitors (PIs), immunomodulators (IMiDs) and daratumumab (study ISB 1342-101). Methods: ISB 1342 was engineered with a single chain variable fragment (scFv) arm that specifically recognizes a cluster of differentiation (CD)3-epsilon (CD3ε) and a fragment antigen binding (Fab) arm which specifically recognizes CD38 and does not compete with daratumumab. By co-engaging CD3ε on T cells and CD38 on tumor cells, ISB 1342 redirects T cells to kill CD38-expressing tumor cells. This mechanism of action is differentiated from existing monospecific CD38 targeting therapies and was designed to overcome resistance to daratumumab in multiple myeloma. Results: In vitro, ISB 1342 killed a large range of CD38-expressing tumor cell lines (EC50:12 to 90 pM) with 8 to 239-fold superior efficacy than daratumumab. ISB 1342 was also able to efficiently kill CD38 low-intermediate-expressing tumor cells that were poorly killed by daratumumab. ISB 1342 retained the potency to kill CD38 low-intermediate-expressing tumor cells when used in sequential or concomitant combination with daratumumab. In addition, the presence of soluble CD38 or glucocorticoid did not impact ISB 1342 killing potency. ISB 1342 was constructed with a double LALA mutation that dampens the binding to Fcγ receptors and C1q. Consistently, ISB 1342 showed only residual Fc-mediated effector functions and its mechanism of tumor cell killing critically relies on the engagement and the activation of T lymphocytes. ISB 1342 showed a favorable on target specificity profile in vitro and was unable to activate T cells in the absence of CD38 positive target cells. Further, ISB 1342-induced tumor cell killing was not associated with a detectable T cell fratricide in vitro. Finally, the potency of ISB 1342 was assessed in vivo in a therapeutic model of a subcutaneously established Daudi tumor co-xenografted with human PBMCs. In marked contrast to daratumumab, which induced only a partial tumor control, ISB 1342 induced complete tumor eradication when injected intravenously weekly at 0.5 mg/kg. As anticipated, the ISB 1342 control molecule (ISB 1342_13DU) made of an irrelevant CD38 binder failed to control tumor growth. The release of the Granzyme A and B, TNF-alpha and CXCL-10 in the tumor micro-environment one week post-treatment was strongly and significantly increased by ISB 1342 but not by daratumumab and ISB 1342_13DU; this represents a correlate of anti-tumor immunity associated with ISB 1342 efficacy in vivo. Conclusions: Hence the higher potency of ISB 1342 relative to daratumumab supports the ongoing clinical development in multiple myeloma patients.
Abstract Background. T-cell bispecific antibodies (TCBs) simultaneously binding CD3 on T-cells and individual tumor antigens, activate T-cells and destroy tumor antigen carrying cells. B-cell maturation antigen (BCMA), a surface antigen reported to be expressed on normal and malignant plasma cells (PCs), could represent a potentially promising target for TCBs in multiple myeloma (MM). The Aim of our study was to: i) assess expression of BCMA in normal and malignant PCs as well as cells of the bone marrow (BM) microenvironment by gene expression profiling and flow cytometry to validate it as potential clinical target for TCBs; ii) to evaluate activity of EM801 as member of a novel class of BCMA-TCBs in vitro on primary myeloma cells and in vivo in the H929-xenograft reconstituted NOG mouse model; and iii) to delineate its mechanism of action. Results. Expression. We investigated the expression of BCMA in CD138-purified PCs from BM aspirates obtained from 726 patients including MGUS (n=62), asymptomatic (n=59) and symptomatic MM (605), as well as different BM cellular subsets from healthy donors (n=10 PCs; plasmablasts, memory B-cells, T-cells, CD34+, CD14+, CD15+, n=5 each; n=8 mesenchymal stromal cells) using Affymetrix DNA-microarrays. BCMA expression was observed in malignant PC from 723/726 (99.5%) MGUS and MM patients, 10/10 normal PCs and 5/5 plasmablasts; gene expression of BCMA was undetectable in all other normal BM subsets. Using multiparameter flow cytometry, BCMA surface expression on malignant PCs was confirmed in 40/40 patients while being absent on normal BM cells. BCMA is thus a potential target in virtually all myeloma patients. Activity. In vitro, EM801 induced concentration dependent significant cell death in malignant plasma cells in BM-samples of 21/28 (75%) previously untreated and 8/10 (80%) relapsed/refractory MM patients in concentrations ranging from 10pM to 30nM. No or only minor unspecific toxicity on cells of the BM microenvironment was observed. In vivo efficacy of EM801 was studied in a subcutaneous H929 myeloma cell line xenograft model in NOG (NOD/Shi-scid/IL-2Rγnull) mice reconstituted with human PBMCs. Three doses of EM801, i.e. 0.026, 0.26 and 2.6 nM/kg, the same doses of a BCMAxCD3-(scFv)2 and two control groups were investigated (n=9 mice/group). Three weekly intravenous doses were given, starting on day 19 after tumor cell injection when tumor volumes were 293±135 mm3. On day 47, all mice from control groups had their tumors grown beyond 2000 mm3 and were euthanized for ethical reasons. In contrast, at 2.6 nM/kg (0.5 mg/kg) EM801 tumor regression was already observed after the second i.v. injection in 6/9 animals and the tumor regressed to 16±3 mm3 on day 47. BCMAxCD3-(scFv)2 bispecific antibody without Fc did not show any efficacy at all doses studied. Regarding the mechanism of action, we first demonstrated that EM801 effectively binds myeloma cells and T-cells with a strength of 1622±410 pN (5-10 fold of control) as measured by atomic force microscopy. Secondly, increasing concentrations (0.03-30nM) of EM801 led to progressive T-cell activation in primary BM samples, with significantly increased levels of CD69 (P<0.001), CD25 (P<0.001) and HLADR (P=0.001) expression in both CD4 and CD8 T-cells as compared to an unspecific TCB. Thirdly, EM801 induced significant secretion of interferon-γ (19-3000 pg/ml), granzyme B (68-2986 pg/ml), and perforin (145-3712 pg/ml) as measured by ELISA, together explaining the strong in vitro and in vivo activity of EM801. Conclusions. BCMA is selectively expressed at the RNA (723/726) and protein (40/40) levels on malignant PCs from virtually all MM patients, and thus represents a promising TCB-target. The novel BCMA-TCB EM801 was effective in vitro in 29/38 (76%) primary MM patients' BM samples at picomolar to low nanomolar concentrations, easily achievable in vivo in patients, as well as in the H929-xenograft reconstituted NOG mouse model at 0.5 mg/kg once a week. Neither in vitro (the BM microenvironment) nor in vivo the compound shows significant toxicity or side effects. EM801 confers cytotoxicity by effectively coupling T-cells with malignant PCs, inducing T-cell activation, secretion of interferon-γ, granzyme B and perforin, and thereby effectively killing malignant PCs. EM801 is thus a promising new compound for the treatment of multiple myeloma to be investigated in clinical phase I/II trials. Disclosures Seckinger: EngMab AG: Research Funding; Takeda: Other: Travel grant. Neuber:EngMab AG: Research Funding. Vu:EngMab AG: Employment, Equity Ownership, Membership on an entity's Board of Directors or advisory committees. Strein:BB Biotech AG: Membership on an entity's Board of Directors or advisory committees; Novimmune SA: Membership on an entity's Board of Directors or advisory committees; EngMab AG: Employment, Equity Ownership, Membership on an entity's Board of Directors or advisory committees. Hundemer:EngMab AG: Research Funding. San Miguel:Bristol-Myers Squibb: Honoraria; Celgene: Honoraria; Janssen-Cilag: Honoraria; Millennium: Honoraria; Novartis: Honoraria; Sanofi-Aventis: Honoraria; Onyx: Honoraria. Hose:Takeda: Other: Travel grant; EngMab AG: Research Funding. Paiva:Celgene: Consultancy; Janssen: Consultancy; Binding Site: Consultancy; BD Bioscience: Consultancy; EngMab AG: Research Funding; Onyx: Consultancy; Millenium: Consultancy; Sanofi: Consultancy.
