DART Molecule Research Articles

Abstract 1560: Selection of a bispecific trivalent HER2 x CD137 TRIDENT format providing optimal tumor-anchored immune co-stimulation

Abstract Introduction: CD137 (4-1BB) signaling provides co-stimulation of CD8 or NK cells following antigen or FcγR engagement, respectively. Efforts to leverage CD137 co-stimulation via agonistic monoclonal antibodies (mAbs) have been thwarted by limited clinical efficacy or unacceptable toxicity. Bispecific targeting strategies linking CD137 activation to a tumor-targeting moiety provides an approach to localize CD137 activation to the tumor microenvironment. Here we evaluate a panel of Fc-bearing HER2 x CD137 bispecific molecules incorporating different valency and geometry to define the format providing optimal CD137 co-stimulation in a tumor-cell anchor-dependent manner. Methods: An anti-HER2 mAb specificity that does not cross compete with margetuximab, trastuzumab or pertuzumab and a proprietary anti-CD137 mAb were utilized to assemble a set of HER2 x CD137 bispecific molecules in bivalent and tetravalent DART® or trivalent TRIDENTTM configurations. The resulting molecules were compared in binding, signaling and co-stimulation assays in the presence or absence of tumor cells expressing HER2. Combination studies were performed in vitro and in immune deficient mice reconstituted with human PBMCs. Results: TRIDENT molecules bearing bivalent CD137 and monovalent HER2 binding achieve optimal HER2-dependent tumor-cell anchored CD137 immune cell co-stimulation. CD137 co-stimulation increases proportionally with the level of HER2 expression as observed with HER2 1+ (MCF7 breast), HER2 2+ (JIMT1 breast) and 3+ (N87 gastric) tumor cells and was paralleled with increased HER2+/CD137+ cell association. No CD137 activation is observed in the absence of HER2-expressing tumor cells. HER2 x CD137 bispecifics enhance NK-cell proliferation and IFN-γ release induced by margetuximab, an Fc-optimized anti-HER2 mAb that up-regulates CD137 expression on NK-cells concomitant with enhanced ADCC against HER2-positive cells. Similarly, HER2 x CD137 bispecific molecules enhanced the in vitro activity of orlotamab, a B7-H3 x CD3 bispecific DART molecule that up-regulates CD137 during T-cell redirected killing. Finally, in vivo mouse studies demonstrate the ability of HER2 x CD137 molecules to expand tumor-associated CD8 cells when co-administered with a tumor targeted CD3 bispecific molecule and support enhance anti-tumor activity. Conclusion: An optimal HER2 x CD137 bispecific format providing maximal CD137 activation in a HER2- dependent manner was identified as a trivalent TRIDENT molecule bearing bivalent CD137 and monovalent HER2 binding. Combinatorial activity with HER2 x CD137 bispecifics was observed with both a HER2-directed therapeutic mAb and a CD3-engaging tumor-targeted bispecific molecule. HER2 x CD137 TRIDENT molecules have therapeutic potential and provide a structural template for incorporating alternate tumor- and/or co-stimulatory-targeting arms. Citation Format: Liqin Liu, Chia-Ying K. Lam, Ralph Alderson, Vatana Long, Yinhua Yang, Robert Burns, Lusiana Widjaja, Jonathan Li, Christina Wolf, Valentina Ciccarone, James Tamura, Gundo Diedrich, Ezio Bonvini, Syd Johnson, Paul A. Moore. Selection of a bispecific trivalent HER2 x CD137 TRIDENT format providing optimal tumor-anchored immune co-stimulation [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 1560.

PF155 COMBINATION OF FLOTETUZUMAB, A CD123 X CD3 BISPECIFIC DART® MOLECULE, AND BLINATUMOMAB, A CD19 X CD3 BITE MOLECULE, TO PREVENT ANTIGEN ESCAPE IN B‐CELL PRECURSOR ACUTE LYMPHOBLASTIC LEUKEMIA

Background:Targeting CD19 antigen by BiTE molecule blinatumomab or CAR‐T cells has shown great efficacy and survival benefits in patients with relapsed or refractory B‐cell precursor acute lymphoblastic leukemia (BCP‐ALL). However, a significant number of patients will relapse and up to 60% of relapses are CD19 negative. CD123 is highly expressed in BCP‐ALL and as shown by our group (EHA 2018) can be targeted with flotetuzumab, a bispecific DART molecule that is currently being clinically evaluated for treatment of acute myeloid leukemia. For both blinatumomab and flotetuzumab, cytokine release syndrome has been observed, which is dose dependent for both bispecific constructs.Aims:1) To verify if blinatumomab and flotetuzumab act synergistically in primary BCP‐ALL, and2) to decipher if the release of cytokines can be reduced by using a suboptimal dose of both constructs while maintaining optimal activity.Methods:Bone marrow cells were isolated from BCP‐ALL patients and incubated either with flotetuzumab alone (0.01 to 1 ng/ml) or blinatumomab alone (0.1 to 10 ng/ml), or with their combination. T‐cell activation and cytokine production were measured after 24 h. The number of T cells and blasts was determined on days 4 and 6. Statistical significance was determined by unpaired Student's t‐test.Results:Flotetuzumab, blinatumomab and their combination induced dose‐dependent T‐cell activation, proliferation, cytokine production and cytotoxicity. After 6 days, maximal killing of primary BCP‐ALL blasts was induced either with 0.1 ng/ml flotetuzumab or 1 ng/ml blinatumomab (median 97.01% (93.88% ‐ 98.75%) and 98.39% (98.26% ‐ 99.34%) killed blasts respectively, vs 45.44% (39.58% ‐ 60.43%) in untreated control group, n = 3, Figure 1A). Although both 0.01 ng/ml flotetuzumab and 0.1 ng/ml blinatumomab resulted in suboptimal cytotoxicity, their combination induced the significant killing of BCP‐ALL blasts (median 85.78% (67.35% ‐ 90.68%) killed blasts, vs 45.44% in untreated control group, p = 0.025, n = 3, Figure 1A). In addition, we compared the cytokine release by T cells in primary BCP‐ALL samples treated with optimal concentrations of flotetuzumab or blinatumomab as single agents or blinatumomab + flotetuzumab combinations that resulted in significant blast killing (Figure 1B). Compared to the samples treated with 0.1 ng/ml flotetuzumab, secretion of cytokines was significantly lower in samples treated with the combination of 0.01 ng/ml flotetuzumab and 0.1 ng/ml blinatumomab (median 388.65 pg/ml vs 4.33 pg/ml IL‐2, p = 0.042; median 40.75 pg/ml vs 7.73 pg/ml IL10, p = 0.043; and median 769.5 pg/ml vs 24.37 pg/ml TNFα, p = 0.0074; n = 3). Furthermore, compared to optimal concentration of 0.1 ng/ml flotetuzumab TNFα production (median 769.5 pg/ml) was significantly lower in samples treated with combination of higher flotetuzumab and blinatumomab doses (0.025 ng/ml flotetuzumab + 0.1 ng/ml blinatumomab: median 40.31 pg/ml, p = 0.0088, n = 3; and 0.5 ng/ml flotetuzumab + 0.1 ng/ml blinatumomab: median 96.98 pg/ml p = 0.014; n = 3). Finally, TNFα concentration was significantly lower in the samples treated with 0.01 ng/ml flotetuzumab + 0.1 ng/ml blinatumomab combination than in those treated with 1 ng/ml blinatumomab (median 24.37 pg/ml vs 214.93 pg/ml, p = 0.039, n = 3).Summary/Conclusion:Our results indicate that dual targeting of CD123 and CD19 with combination of low doses of flotetuzumab and blinatumomab is an attractive approach to prevent antigen escape of BCP‐ALL, resulting in significant reduction of cytokine production by effector cells.image

A phase I, first-in-human, open label, dose-escalation and cohort expansion study of MGD019, a bispecific DART protein binding PD-1 and CTLA-4 in patients with unresectable or metastatic neoplasms.

TPS2661 Background: Immune checkpoint molecules, including CTLA-4 and PD-1, attenuate the duration and strength of adaptive immune responses to limit immune-mediated tissue damage. Tumors may inhibit cellular immune activation by expressing ligands that bind checkpoint molecules and inhibit T-cell function in the tumor microenvironment. Blockade of these inhibitory pathways is the primary mechanism of action of several novel cancer immunotherapy agents. Combined blockade of PD-1 and CTLA-4 with two checkpoint inhibitors, ipilimumab and nivolumab, increases antitumor activity beyond either single agent alone in patients with metastatic melanoma or other malignancies. MGD019, a novel bispecific molecule that co-engages and coordinately inhibits both PD-1 and CTLA-4 signaling, was developed to potentially improve antitumor activity and/or safety relative to the monoclonal antibody combination. MGD019 is an Fc-bearing tetravalent DART molecule (bivalent for each antigen) that can independently block either checkpoint molecule, with preferential co-blockade in cells co-expressing both molecules demonstrated in vitro. It is hypothesized that MGD019 might be clinically active in either checkpoint naïve or checkpoint experienced patients after prior PD-1/PD-L1 inhibitors. Methods: This Phase 1 study will characterize safety, dose limiting toxicities, and maximum tolerated dose (MTD)/maximum administered dose (MAD) of MGD019. Dose Escalation will enroll patients with advanced solid tumors of any histology in sequential escalating doses in cohorts of 3 to 9 patients in a 3+3+3 design. Once the MTD/MAD is reached, a Cohort Expansion phase will characterize safety and initial antitumor activity per RECIST v1.1 and irRECIST in patients with specific tumor types anticipated to be sensitive to dual checkpoint blockade. Additional endpoints include pharmacokinetics; immunogenicity; impact of MGD019 on various measures of immune-regulatory effects in peripheral blood and biopsy specimens; and relationship between antitumor activity and gene profiles, tumor mutational burden, and PD-1, PD-L1, and CTLA-4 expression on tumor cells and immune cell infiltrates within biopsy specimens. Patients will be followed for survival approximately every 3 months for 2 years. Clinical trial information: NCT03761017.

A Next-Generation Fc-Bearing CD3-Engaging Bispecific DART® Platform with Extended Pharmacokinetic and Expanded Pharmacologic Window: Characterization As CD123 x CD3 and CD19 x CD3 DART Molecules

Background: CD3-engaging DART molecules direct T cells to tumor-expressed antigens. Flotetuzumab, a CD123 x CD3 bispecific DART molecule that targets the differential expression of the IL-3 receptor alpha chain (CD123) on acute myeloid leukemia (AML) blasts and leukemic stem cells, is currently being investigated in a phase 1 study in relapsed/refractory AML and myelodysplastic syndrome (MDS), with initial evidence of clinical activity1. A common adverse event in redirected T-cell killing interventions is cytokine release syndrome (CRS), a Cmax-associated event that may limit their therapeutic window. Owing to its short circulating half-life, flotetuzumab is administered as continuous infusion, which affords prolonged exposure with limited Cmax excursions. In contrast, long-acting Fc-bearing, CD3-engaging molecules would eliminate the need for continuous dosing, but would be especially impacted by CRS, given the high Cmax levels required to maintain adequate trough concentrations over several days or weeks. We previously reported the development of a CD3-engaging DART molecule with reduced affinity for CD3 that maintained maximal target cell killing and T-cell proliferation, albeit at higher concentrations than its wild-type (WT) counterpart, but with reduced cytokine release2. We report here further optimization and validation of the approach.Methods: Anti-CD3 variants spanning a range of affinities were formatted as Fc-bearing DART molecules with CD123- or CD19-targeting arms and evaluated for cell binding as well as for antigen-dependent induction of T-cell proliferation, cytolytic activity and cytokine release. A therapeutic index (TI) was determined from the ratio of in vitro cell killing activity to that of target cell-induced cytokine release. In vivo anti-tumor activity was assessed in human immune cell-reconstituted mouse tumor models or in mice expressing the epitope recognized by the CD3 arm of the DART proteins (human CD3 k/i mice). Safety, pharmacokinetic (PK) and biological activity of selected molecules was evaluated in cynomolgus monkeys.Results: Six Fc-bearing DART molecules (from a panel of 23 CD123 x CD3 DART variants) were selected for characterization. These molecules spanned a relatively small interval of low CD3 affinities, but varied greatly in activity, from inactive at all concentrations tested to being capable of lysing tumor cells, including primary leukemic and MDS blasts, to the same maximal level of killing observed with the parental WT DART molecule. Higher concentrations of the active variants were needed, owing to a deliberate design incorporating decreased CD3 binding. At levels capable of complete tumor cell lysis, active DART variants showed greatly reduced pro-inflammatory cytokine release (e.g., IL-2, IFN-γ and TNF-α) in vitro and in human PBMC-reconstituted mice compared to the WT version. Molecules were ranked based on their TI and a version with a highly favorable index was engineered as a B-cell targeting CD19 x CD3 DART molecule to conveniently ascertain its biological activity in cynomolgus monkeys. The CD19 x CD3 DART variant demonstrated good tolerability at single doses up to 30 mg/kg. B-cell depletion in the circulation and lymphoid tissues occurred at 1 mg/kg and achieved maximal depletion at ≥10 mg/kg, equivalent to the degree of reduction observed with 0.1 mg/kg of the WT molecule. However, only minimal circulating pro-inflammatory cytokines (including IL-6) were observed at all doses of the DART variant, while a substantial increase in IL-2, IL-6 and IFN-γ (among others) was observed with 0.1 mg/kg of the WT counterpart. The PK profile of the DART variant in cynomolgus monkeys appeared linear in the 1-30 mg/kg interval, with a half-life of 5 to 10 days, consistent with possible clinical dosing at weekly or longer intervals.Conclusions: A next-generation CD3-engaging Fc-bearing DART platform was generated through CD3 arm affinity modulation. These DART variants feature extended PK and an expanded TI suitable for potential clinical development in hematological malignancies.1. Uy et al, Blood 2017, vol. 130, Suppl 1 6372. Huang et al, Keystone Symposia: Antibodies as Drugs, 2/25-3/1/2018, Whistler, BC, Canada DisclosuresBonvini:MacroGenics: Employment, Equity Ownership. La Motte-Mohs:MacroGenics: Employment, Equity Ownership. Huang:MacroGenics, Inc.: Employment, Equity Ownership. Lam:MacroGenics, Inc.: Employment, Equity Ownership. Kaufman:MacroGenics, Inc.: Employment, Equity Ownership. Liu:MacroGenics, Inc.: Employment, Equity Ownership. Alderson:MacroGenics, Inc.: Employment, Equity Ownership. Stahl:MacroGenics, Inc.: Employment, Equity Ownership. Brown:MacroGenics, Inc.: Employment, Equity Ownership. Li:MacroGenics, Inc.: Employment, Equity Ownership. Sharma:MacroGenics, Inc.: Employment, Equity Ownership. Tamura:MacroGenics, Inc.: Employment, Equity Ownership. Johnson:MacroGenics, Inc.: Consultancy, Equity Ownership. Moore:MacroGenics, Inc.: Employment, Equity Ownership.

Phase 1 Cohort Expansion of Flotetuzumab, a CD123×CD3 Bispecific Dart® Protein in Patients with Relapsed/Refractory Acute Myeloid Leukemia (AML)

Phase 1 Cohort Expansion of Flotetuzumab, a CD123×CD3 Bispecific Dart® Protein in Patients with Relapsed/Refractory Acute Myeloid Leukemia (AML)

Integrated Pharmacokinetic/Pharmacodynamic Model of a Bispecific CD3xCD123 DART Molecule in Nonhuman Primates: Evaluation of Activity and Impact of Immunogenicity.

Purpose: Flotetuzumab (MGD006 or S80880) is a bispecific molecule that recognizes CD3 and CD123 membrane proteins, redirecting T cells to kill CD123-expressing cells for the treatment of acute myeloid leukemia. In this study, we developed a mathematical model to characterize MGD006 exposure-response relationships and to assess the impact of its immunogenicity in cynomolgus monkeys.Experimental Design: Thirty-two animals received multiple escalating doses (100-300-600-1,000 ng/kg/day) via intravenous infusion continuously 4 days a week. The model reflects sequential binding of MGD006 to CD3 and CD123 receptors. Formation of the MGD006/CD3 complex was connected to total T cells undergoing trafficking, whereas the formation of the trimolecular complex results in T-cell activation and clonal expansion. Activated T cells were used to drive the peripheral depletion of CD123-positive cells. Anti-drug antibody development was linked to MGD006 disposition as an elimination pathway. Model validation was tested by predicting the activity of MGD006 in eight monkeys receiving continuous 7-day infusions.Results: MGD006 disposition and total T-cell and CD123-positive cell profiles were well characterized. Anti-drug antibody development led to the suppression of T-cell trafficking but did not systematically abolish CD123-positive cell depletion. Target cell depletion could persist after drug elimination owing to the self-proliferation of activated T cells generated during the first cycles. The model was externally validated with the 7-day infusion dosing schedule.Conclusions: A translational model was developed for MGD006 that features T-cell activation and expansion as a key driver of pharmacologic activity and provides a mechanistic quantitative platform to inform dosing strategies in ongoing clinical studies. Clin Cancer Res; 24(11); 2631-41. ©2018 AACR.

OP7.6 - Preclinical development of a bispecific HIV x CD3 DART molecule that redirects T cells to kill HIV envelope (env)-expressing cells

OP7.6 - Preclinical development of a bispecific HIV x CD3 DART molecule that redirects T cells to kill HIV envelope (env)-expressing cells

995O - Interim results from a phase 1 first-in-human study of flotetuzumab, a CD123 x CD3 bispecific DART molecule, in AML/MDS

995O - Interim results from a phase 1 first-in-human study of flotetuzumab, a CD123 x CD3 bispecific DART molecule, in AML/MDS

Abstract 3642: Tumor-antigen expression-dependent activation of the CD137 costimulatory pathway by bispecific DART® proteins

Abstract Introduction: CD137 (4-1BB) is a co-stimulatory molecule expressed by activated T and NK cells that, upon interaction with its CD137 ligand, further supports cell activation, proliferation and survival. Activation via CD137 holds great promise for cancer immunotherapy; however, current CD137 agonistic interventions are associated with systemic safety concerns. To develop a therapeutic modality that reduces the potential for systemic CD137 effects, we applied the DART® bispecific platform to generate proteins that can induce tumor-antigen dependent T-cell activation. Methods: DART molecules were constructed containing anti-CD137 variable regions together with either anti-HER2 or anti-EphA2 variable regions. DART binding properties were evaluated by ELISA or flow cytometry; signaling responses assessed using a NF-κB luciferase reporter cell line expressing CD137. Co-stimulatory activity was characterized with primary human T cells in the presence or absence of tumor target antigen-expressing cells. Results: ELISA and flow cytometry analysis demonstrated that both HER2 x CD137 and EphA2 x CD137 DART molecules bind their respective target antigens. Co-culturing of a CD137/NF-κB reporter cell line with tumor lines expressing HER2 or EphA2 revealed tumor antigen-dependent CD137 pathway activation by HER2 x CD137 and EphA2 x CD137 DART molecules, respectively. To evaluate the effects of HER2 x CD137 and EphA2 x CD137 DART molecules on T-cell responses, co-stimulation T-cell assays were performed. In the presence of the relevant antigen-positive cell line, each respective DART molecule was able to promote T-cell proliferation and cytokine release in a HER2 or EphA2-dependent manner. No T-cell co-stimulation was observed by either DART molecule in the absence of antigen-expressing tumor cells. Furthermore, the level of tumor antigen-dependent co-stimulation supported by the DART molecules correlated with the level of tumor target expression. Consistent with the preferential induction of CD137 by the CD8 T cell subset, CD137-based DART proteins induced a substantial increase in the fraction of CD8+ central memory and effector memory T cells in the presence of the proper tumor antigen expressing cells. Conclusions: HER2 x CD137 and EphA2 x CD137 DART proteins promote T-cell co-stimulation in a tumor antigen-dependent manner and may provide an opportunity to target the CD137 co-stimulatory pathway for cancer immunotherapy, while limiting systemic T-cell activation and related side effects. Citation Format: Liqin Liu, Chia-Ying K. Lam, Vatana Long, Lusiana Widjaja, Yinhua Yang, Kalpana Shah, Doug Smith, Joanna Pan, Syd Johnson, Ezio Bonvini, Paul Moore. Tumor-antigen expression-dependent activation of the CD137 costimulatory pathway by bispecific DART® proteins [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3642. doi:10.1158/1538-7445.AM2017-3642

Abstract 3637: Co-targeting of PD-1 and CTLA-4 inhibitory pathways with bispecific DART® and TRIDENT™ molecules

Abstract Introduction: Immunotherapy with the combination of monoclonal antibodies (mAbs) that block PD-1 (nivolumab) and CTLA-4 (ipilimumab) has shown clinical benefit beyond that observed with either mAb alone. We have developed PD-1 x CTLA-4 bispecific proteins aimed at inducing antitumor immunity through simultaneous blockade of both checkpoint molecules. Two proteins, a tetravalent PD-1 x CTLA-4 bispecific DART® protein (bivalent for both PD-1 and CTLA-4) and a trivalent PD-1 x CTLA-4 TRIDENT™ protein (bivalent for PD-1 but monovalent for CTLA-4) were engineered; the TRIDENT protein was designed to promote CTLA-4 blockade through avidity when bound to PD-1 on cells co-expressing both receptors, thus favoring PD-1 over CTLA-4 blockade in cells that do not co-express both molecules. Methods: Proteins were engineered from humanized PD-1 and CTLA-4 mAb sequences and demonstrated favorable biophysical properties. Binding assays were performed by ELISA or flow cytometry. Co-ligation of PD-1 and CTLA-4 was assessed in an enzyme-fragment complementation assay (DiscoverX). T-cell activation was tested in reporter cells, staphylococcus enterotoxin B-stimulated PBMCs or MLR assays. Results: Both DART and TRIDENT proteins showed equivalent potency in binding immobilized recombinant PD-1 or PD-1-expressing cell lines, inhibition of PD-1 interaction with PD-L1 or PD-L2 as well as reversal of PD-1/PD-L1 mediated T-cell signal inhibition. In all assays, both formats showed activities that were comparable to those of the precursor PD-1 mAb. With respect to CTLA-4, the DART protein showed a minor potency loss in binding to CTLA-4 expressing cells, inhibition of CTLA-4/B7 interaction and reversal of T-cell signal inhibition compared to the precursor mAb. The TRIDENT protein, however, showed substantial lower potency than the DART protein in all CTLA-4 assays, consistent with the monovalent nature of the CTLA-4 arm. Importantly, in cells that co-express both receptors, DART and TRIDENT proteins show comparable co-engagement of PD-1 and CTLA-4, as shown by enzyme-fragment complementation, suggesting that anchoring through PD-1 compensates for the decreased CTLA-4 avidity of the TRIDENT molecule when both target receptors are expressed. Similarly, in T-cell co-activation assays, both DART and TRIDENT proteins showed a comparable increase in IFN-gamma response that recapitulated that observed with the combination of the individual parental mAbs. Conclusion: Both PD-1 x CTLA-4 DART and TRIDENT bispecific molecules block PD-1 and CTLA-4 pathways, with the TRIDENT protein demonstrating a PD-1-biased binding preference, consistent with its design intended to reduce CTLA-4 blockade in the absence of PD-1 co-expression. Both molecules showed comparable T-cell activation activity. Further development of bispecific PD-1 x CTLA-4 molecules for cancer treatment is warranted. Citation Format: Gurunadh R. Chichili, Kalpana Shah, Ross La Motte-Mohs, Alex Berezhnoy, Daorong Liu, Jessica Hill, Leilei He, Christine Shoemaker, Jennifer Brown, Kurt Stahl, Hua Li, Jonathan Li, Sergey Gorlatov, Jim Tamura, Valentina Ciccarone, Ralph Alderson, Ezio Bonvini, Paul A. Moore, Syd Johnson. Co-targeting of PD-1 and CTLA-4 inhibitory pathways with bispecific DART® and TRIDENT™ molecules [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3637. doi:10.1158/1538-7445.AM2017-3637

Abstract 4608: A 5T4 x CD3 bispecific DART® molecule with extended half-life for T-cell immunotherapy of cancers

Abstract Introduction: 5T4 (trophoblast glycoprotein), an oncofetal antigen involved in embryo development, is expressed on the cell surface of multiple cancers. Several 5T4-directed interventions have been reported, including a cancer vaccine and an antibody-drug conjugate. A superantigen-driven redirected cell killing modality was also developed, but obstacles, such as immunogenicity, remain to a successful 5T4 therapeutic molecule. We have developed an Fc-bearing 5T4 x CD3 DART® bispecific protein designed to redirect T cells to target 5T4-expressing tumors. Methods: 5T4 x CD3, a humanized Fc-bearing DART molecule, was stably expressed in CHO cells and purified to homogeneity via a standard antibody-purification platform. In vitro characterization and functional studies were performed with 5T4-positive tumor cell lines and human T cells. In vivo studies were performed in immune-deficient tumor-bearing mice co-implanted with activated human T cells or reconstituted with human peripheral blood mononuclear cells (PBMCs). Pharmacokinetic studies were performed in both human FcRn transgenic mice and cynomolgus monkeys. Results: 5T4 x CD3 demonstrated bispecific binding properties to both human and cynomolgus monkey antigens. In redirected cytolysis studies, 5T4 x CD3 mediated lysis of 5T4-positive pancreatic, lung, renal, triple-negative breast, and ovarian cancer cell lines with EC50 values ranging between 0.03 and 0.08 ng/mL. 5T4 x CD3 displayed favorable pharmacokinetics with a prolonged circulating half-life in human FcRn transgenic mice and cynomolgus monkeys. Tumor clearance studies in NOD/SCID mice implanted subcutaneously with activated human T cells and tumor cells demonstrated robust inhibition of tumor growth upon intravenous administration of 5T4 x CD3 at doses as low as 0.8 μg/kg, but not with a CD3-binding control DART protein. In addition, human PBMC-reconstituted NOD/SCID/IL2 gamma-chain null mice were implanted intradermally with renal and pancreatic tumor cell lines or orthotopically with a triple-negative breast cancer line and, after tumor establishment, were treated with 5T4 x CD3. Anti-tumor activity was observed at doses as low as 4 μg/kg. Immunohistochemical analysis of tumor xenografts confirmed tumor clearance was associated with T-cell recruitment into the tumor mass. Conclusions: In summary, 5T4 x CD3 displays robust antitumor activity against several cancer cell lines in vitro and in vivo together with a favorable pharmacokinetic profile and merits further consideration as a potential treatment for 5T4-positive cancers. Citation Format: Ling Huang, Gurunadh Chichili, Ralph Alderson, Francine Chen, Jennifer Brown, Hua Li, Valentina Ciccarone, Jim Tamura, Daorong Liu, Liqin Liu, Syd Johnson, Ezio Bonvini, Paul Moore. A 5T4 x CD3 bispecific DART® molecule with extended half-life for T-cell immunotherapy of cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4608. doi:10.1158/1538-7445.AM2017-4608

A phase I, first-in-human study of MGD006/S80880 (CD123 x CD3 DART) in AML/MDS.

TPS7070 Background: MGD006/S80880 is a novel CD123 x CD3 DART molecule designed to target CD123-positive cells for recognition and elimination by CD3-expressing T lymphocytes as effector cells. CD123, the alpha chain of the interleukin 3 receptor (IL-3Ra) is known to be highly expressed in > 90% of AML patients and at least 50% of MDS patients. Based on these observations, targeting CD123 could be a promising strategy in the preferential ablation of AML and MDS cells. Methods: MGD006 is currently being evaluated in a Phase 1 dose-escalation and cohort expansion study in relapsed/refractory (R/R) AML or intermediate-2/high risk MDS. The objectives of the study are to determine the MTDS and safety profile, and describe the pharmacokinetics and preliminary activity of MGD006. Patients are dosed in 28-day cycles. All patients start with a lead-in continuous IV infusion of 30ng/kg/day for 3 days followed by 100ng/kg/day for 4 days. Subsequent weeks (2-4) are dosed in two different schedules. One arm receives MGD006 for 4 days on/3 days off and the second arm receives MGD006 for 7 days continuously at the maximal dose assigned to each cohort (up to 1000ng/kg/day). Beginning with the second cycle, all patients are administered MGD006 for 4 days on/3 days off at the maximal dose/cohort. Patients can continue on treatment until 2 cycles (8 weeks) after the attainment of a CR, maximum of 12 cycles of MGD006, DLT or treatment failure. Once the MTDS is identified, two cohorts of 24 patients each, one in AML and one in MDS, will be enrolled. Response is assessed by IWG or IPSS criteria for AML and MDS, respectively. Signs and symptoms of cytokine release syndrome (CRS), a common AE, are graded according to Lee criteria. The study continues to enroll patients with open sites in the US, France, Germany, Italy, and The Netherlands. ClinicalTrials.gov #NCT02152956. EudraCT #2015-003813-11.

Abstract 1489: Development of a humanized ROR1 x CD3 bispecific DART® molecule for the treatment of solid and liquid tumors

Abstract Introduction: The receptor tyrosine kinase-like orphan receptor 1 (ROR1) is overexpressed in chronic lymphocytic leukemia and a subset of solid tumors, including lung, breast, ovarian, colon, and pancreatic cancers, as well as sarcoma. Limited adult tissue expression and its absence in normal leukocytes makes ROR1 a promising cancer therapeutic target. We have developed a Dual-Affinity Re-Targeting (DART®) protein for redirecting T lymphocytes to lyse tumor cells via monovalent recognition of ROR1 on tumors and CD3 on T cells. ROR1 x CD3 DART protein was engineered for improved half-life with the incorporation of a modified Fc domain, lacking effector function. Methods: The ROR1 x CD3 DART protein was stably expressed in CHO cells and purified to homogeneity by a standard antibody platform. Bispecific binding was evaluated by ELISA and SPR analysis. In vitro functional studies were performed with lymphoma and solid tumor cell lines in the presence of primary human T cells. Tumor growth inhibition was evaluated in NOD/SCID/IL-2 gamma chain KO (NOG) mice coimplanted with human T cells and either mantle cell lymphoma (MCL) or lung cancer cell lines (5:1 effector : target cell ratio) followed by treatment with ROR1 x CD3 DART protein by intravenous (IV) administration. In vivo activity was also evaluated in human PBMC-reconstituted NOG/B2m deficient mice bearing established intradermal tumor xenografts following IV treatment with ROR1 x CD3 DART molecule. Pharmacokinetic analysis of the DART molecule was performed in human neonatal Fc receptor (hFcRn) transgenic mice. Results: The ROR1 x CD3 DART protein displayed the expected bispecific binding for ROR1 and CD3 antigens and retained the affinity and specificity of the parent mAbs. The DART molecule mediated dose-dependent lysis of ROR1-positive MCL and solid tumor (breast, lung, and osteosarcoma) cell lines through recruitment of human T cells. DART molecule-mediated killing of ROR1-expressing target cells was accompanied by target-dependent T-cell activation and cytokine release; however, no activity was observed in the absence of target cells and no cytokine release was observed with human PBMCs alone. The ROR1 x CD3 DART protein displayed extended circulating half-life after administration to hFcRn-transgenic mice. In mouse efficacy studies, the growth of HBL-2 (MCL), HOP-92 (lung cancer), or NIH-1975 (a lung cancer line resistant to erlotinib) cells co-implanted with human T cells in NOG mice was inhibited by treatment with the ROR1 x CD3 DART protein at doses in the mcg/kg range. The ROR1 x CD3 DART molecule also demonstrated antitumor activity with high complete response rates in human PBMC-reconstituted mice bearing established HBL-2 cell xenografts. Conclusion: The promising in vitro and in vivo activity of the Fc-bearing ROR1 x CD3 DART molecule supports further investigation as a potential candidate for the cancer treatment. Citation Format: Bhaswati Barat, Gurunadh Chichili, Valentina Ciccarone, James Tamura, Sergey Gorlatov, Michael Spliedt, Francine Chen, Scott Koenig, Paul Moore, Ezio Bonvini, Ralph Alderson, Syd Johnson. Development of a humanized ROR1 x CD3 bispecific DART® molecule for the treatment of solid and liquid tumors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1489.

Abstract 1498: Development of an IL13Ralpha2 x CD3 bispecific DART® protein for redirected T-cell killing of solid tumors

Abstract Introduction. IL13Rα2 is a membrane-bound protein expressed on a number of malignant tumors, but not at significant levels in most normal tissues. Clinically tested therapies to treat cancer by targeting IL13Rα2 to date have included a ligand toxin conjugate (IL13-PE) and IL13Rα2 directed CAR-T cells. Both strategies appear to be well tolerated, although with mixed results in tumor eradication. An alternative strategy for targeting IL13Rα2 is an IL13Rα2 x CD3 Dual-Affinity Re-Targeting (DART®) bispecific molecule designed to co-engage IL13Rα2 on tumor cells and cytotoxic T cells through CD3, resulting in killing of tumor cells. Methods. Mouse monoclonal antibodies (mAbs) were generated using standard immunization protocols followed by binding and epitope binning analyses. Cell binding was performed by flow cytometry and frozen normal and tumor tissues analyses by immunohistochemistry (IHC). In vitro studies were performed with cancer cell lines and primary human T cells or peripheral blood mononuclear cells (PBMCs). In vivo studies were performed in immune-deficient tumor-bearing mice co-implanted with activated human T cells or reconstituted with human PBMCs. Results. A panel of mAbs that bind human and cynomolgus monkey IL13Rα2 proteins were selected, representing a range of binding characteristics and epitope diversity. IHC showed favorable normal vs tumor tissue reactivity, with high levels of IL13Rα2 expression in glioblastoma and melanoma. The mAb panel was converted to DART molecules with an anti-CD3 arm, assessed for the ability to mediate cytotoxic T lymphocyte (CTL) activity and cytokine release and a lead selected for humanization and engineering into an IL13Rα2 x CD3 DART molecule incorporating an Fc domain to prolong circulating half-life. The IL13Rα2 x CD3 DART protein bound human and cynomolgus IL13Rα2 (KD = 0.64 nM and 0.5 nM, respectively) and mediated CTL activity against LOX-IMVI (melanoma) and SK-MES-1 (lung adenocarcinoma) target cells. Cytokines (IFN-γ, IL-10, and TNF-α) were released in the presence of IL13Rα2-expressing target cells and human PBMC, but not with PBMCs alone. Administration of the IL13Rα2 x CD3 DART molecule (≤50 μg/kg IV for 4 consecutive days) prevented tumor growth when tumor cells (A375 and LOX-IMVI melanoma, U87 glioblastoma, or DAOY medulloblastoma cell lines) were co-mixed with activated human T cells and implanted subcutaneously in NOD/SCID/IL2gamma-chain null mice. Tumor eradication was also observed in NOD/SCID/IL2gamma-chain/MHCI KO mice reconstituted with human PBMCs and bearing established LOX-IMVI melanoma tumors following 2 weekly IV doses of 50 μg/kg IL13Rα2 x CD3 DART molecule. Conclusions: Robust in vitro and in vivo antitumor activity for an IL13Rα2 x CD3 DART protein was demonstrated. Additional studies are underway to further characterize the molecule as a development candidate for treatment of IL13Rα2-positive cancers. Citation Format: Jill Rillema, Monica Licea, Shereen Saini-Lal, Doug Smith, Francine Chen, Annie Lam, Yinhua Yang, Liqin Liu, James Tamura, Ralph Alderson, Ezio Bonvini, Syd Johnson, Paul Moore. Development of an IL13Ralpha2 x CD3 bispecific DART® protein for redirected T-cell killing of solid tumors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1498.

Abstract 583: Evaluation of EphA2 as a therapeutic target for redirected T-cell killing by DART® bispecific molecules

Abstract Introduction: EphA2 is a receptor tyrosine kinase that plays a critical role in cancer progression through both ligand-dependent and independent mechanisms. The broad overexpression in tumors but limited normal tissue expression of EphA2 makes it an attractive therapeutic target amenable for redirected T-cell killing via an EphA2 x CD3 Dual-Affinity Re-Targeting (DART®) molecules designed to co-engage cytotoxic T cells with EphA2 expressing tumor cells. Methods: Anti-EphA2 monoclonal antibodies (mAbs) were identified by target-specific screening of a library generated by whole-cell immunization with proprietary cancer cell lines, including models of cancer stem cells. The binding and signaling properties of the antibodies were characterized by ELISA, SPR, flow cytometry and phosphorylation assays. Receptor binding regions were determined by ELISA based competition assays and by utilizing human-mouse chimeric EphA2 molecules. Immunohistochemistry (IHC) was performed on frozen normal and tumor tissues. In vitro functional studies were performed with various cancer as well as transfected cell lines, and primary human T cells or PBMCs. In vivo activity was evaluated in xenograft models in immune deficient mice. Results: EphA2 mAbs encompassing diversity in binding kinetics and effects on receptor phosphorylation were classified in 5 discrete binding groups. The majority interacted with the N-terminal ligand-binding domain of EphA2 and most mAbs within that group interfered with ligand binding. The majority of mAbs displayed little IHC reactivity with normal tissue, while strong staining of cancer tissues was observed, including colon, lung, pancreas, ovary, bladder and breast cancers. Seven mAbs recognizing independent epitopes were selected for conversion into EphA2 x CD3 DART molecules that showed a range of potency in redirecting T cells to kill EphA2-expressing target cells. A lead EphA2 x CD3 DART molecule was selected based on potency and cross-reactivity with the cynomolgus monkey ortholog; this lead was engineered with a human Fc domain to confer an extended circulating half-life. The resulting Fc-bearing EphA2 x CD3 DART molecule demonstrated in vivo anti-tumor activity at doses as low as 20 mcg/kg in NOD/SCID/IL-2 gamma chain null mice co-implanted with activated human T cells and MDA-MB231 breast cancer cells. Conclusion: EphA2 is a potential cancer target for redirected T-cell killing applications independent of ligand mediated mechanisms. Further preclinical assessment of EphA2 x CD3 DART molecules as a strategy for targeting EphA2-expressing malignancies is warranted. Citation Format: Claudia B. Fieger, Kalpana Shah, Gurunadh Chichili, Jonathan Li, Thomas Son, Jeffrey Hooley, Francine Chen, Sergey Gorlatov, Steve Burke, Valentina Ciccarone, Ralph Alderson, Deryk Loo, Syd Johnson, Ezio Bonvini, Paul Moore. Evaluation of EphA2 as a therapeutic target for redirected T-cell killing by DART® bispecific molecules. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 583.

Development of PF-06671008, a Highly Potent Anti-P-cadherin/Anti-CD3 Bispecific DART Molecule with Extended Half-Life for the Treatment of Cancer.

Bispecific antibodies offer a promising approach for the treatment of cancer but can be challenging to engineer and manufacture. Here we report the development of PF-06671008, an extended-half-life dual-affinity re-targeting (DART®) bispecific molecule against P-cadherin and CD3 that demonstrates antibody-like properties. Using phage display, we identified anti-P-cadherin single chain Fv (scFv) that were subsequently affinity-optimized to picomolar affinity using stringent phage selection strategies, resulting in low picomolar potency in cytotoxic T lymphocyte (CTL) killing assays in the DART format. The crystal structure of this disulfide-constrained diabody shows that it forms a novel compact structure with the two antigen binding sites separated from each other by approximately 30 Å and facing approximately 90° apart. We show here that introduction of the human Fc domain in PF-06671008 has produced a molecule with an extended half-life (~4.4 days in human FcRn knock-in mice), high stability (Tm1 > 68 °C), high expression (>1 g/L), and robust purification properties (highly pure heterodimer), all with minimal impact on potency. Finally, we demonstrate in vivo anti-tumor efficacy in a human colorectal/human peripheral blood mononuclear cell (PBMC) co-mix xenograft mouse model. These results suggest PF-06671008 is a promising new bispecific for the treatment of patients with solid tumors expressing P-cadherin.

Targeting HIV reservoir by DART molecules that recruit T cells to HIV Env expressing cells: comparison of HIV arms derived from broadly reactive neutralizing or non-neutralizing anti-Env antibodies

Targeting HIV reservoir by DART molecules that recruit T cells to HIV Env expressing cells: comparison of HIV arms derived from broadly reactive neutralizing or non-neutralizing anti-Env antibodies

Abstract 2464: DART® molecules with enhanced DR5 agonistic activity for improved cancer cell cytotoxicity

Abstract Introduction: Death Receptor 5 (DR5) belongs to the tumor necrosis factor receptor superfamily. The target and its associated signaling pathway are preferentially active on cancer cells. DR5 elicits pro-apoptotic signal through receptor oligomerization upon binding of TRAIL or an agonistic mAb. Chemotherapy and radiation also can synergize with the DR5 pathway. Various therapeutics targeting DR5 have been generated but clinical outcomes have been generally disappointing. Here we describe optimization of the agonistic activity of anti-DR5 mAbs by enabling multivalent target engagement via engineering into DARTs; covalently-linked Fv-based diabodies optimized for manufacturability and stability (1) whose modular nature facilitates tailoring avidity and PK properties to the specific needs of the application. Methods: Anti-DR5 mAbs were selected from a panel of mAbs generated from cancer cell immunizations and characterized for binding properties. DARTs with varying anti-DR5 specificities, valency and incorporation of either a wild type IgG1 Fc or an Fc mutated to eliminate FcγR binding, were expressed in CHO cells and purified to homogeneity. DR5 DARTs and mAbs were characterized across a panel of cell lines including those derived from colorectal, lung, pancreatic, breast and prostate cancer and by IHC on normal and tumor tissue specimens. Results: mAbs from whole cancer cell immunizations displaying differential expression on both normal and cancer tissues were subjected to antigen identification and identified a subset with reactivity to DR5. Binding analyses revealed DR5 mAbs recognizing non-overlapping epitopes that do not block the TRAIL-DR5 interaction. Upon secondary cross-linking or when combined, the DR5 mAbs induced apoptosis across multiple cancer cell lines. Further potency enhancement independent of secondary cross-linking was obtained by the engineered multivalent DR5-targeting DARTs. Dose-dependent growth inhibition assay demonstrated that the DR5-targeting DARTs delivered significantly greater potency than TRAIL (≥100-fold) across a broad range of cancer cell lines. Furthermore, the DR5-targeting DARTs maintained the exquisite tumor/normal differential reactivity displayed by the parental DR5 mAbs from which they were derived. Combination of an HDAC inhibitor with DR5-targeting DARTs revealed maintenance of synergistic activity and potential to overcome pathway resistance. Conclusion: Whole cancer cell immunizations yielded DR5 mAb candidates with desirable binding and functional properties. Through incorporation of the anti-DR5 mAb specificities into multivalent DART molecules, we have generated a new class of therapeutics that may overcome the limitations of existing DR5-based therapeutics. The data support the use of DR5-targeting DARTs to target this apoptotic pathway in multiple cancer cells. Reference: (1) Johnson et al. 2010. J Mol Bio. 399:436-49 Citation Format: Jonathan C. Li, Kalpana Shah, Jill Rillema, Francine Chen, Doug Smith, Steve Burke, Valentina Ciccarone, Sergey Gorlatov, Ralph Alderson, James Tamura, Ezio Bonvini, Syd Johnson, Paul A. Moore. DART® molecules with enhanced DR5 agonistic activity for improved cancer cell cytotoxicity. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2464. doi:10.1158/1538-7445.AM2015-2464

A CD3xCD123 bispecific DART for redirecting host T cells to myelogenous leukemia: preclinical activity and safety in nonhuman primates.

Current therapies for acute myeloid leukemia (AML) are largely ineffective, and AML patients may benefit from targeted immunotherapy approaches. MGD006 is a bispecific CD3xCD123 dual-affinity re-targeting (DART) molecule that binds T lymphocytes and cells expressing CD123, an antigen up-regulated in several hematological malignancies including AML. MGD006 mediates blast killing in AML samples, together with concomitant activation and expansion of residual T cells. MGD006 is designed to be rapidly cleared, and therefore requires continuous delivery. In a mouse model of continuous administration, MGD006 eliminated engrafted KG-1a cells (an AML-M0 line) in human PBMC (peripheral blood mononuclear cell)-reconstituted NSG/β2m(-/-) mice at doses as low as 0.5 μg/kg per day for ~7 days. MGD006 binds to human and cynomolgus monkey antigens with similar affinities and redirects T cells from either species to kill CD123-expressing target cells. MGD006 was well tolerated in monkeys continuously infused with 0.1 μg/kg per day escalated weekly to up to 1 μg/kg per day during a 4-week period. Depletion of circulating CD123-positive cells was observed as early as 72 hours after treatment initiation and persisted throughout the infusion period. Cytokine release, observed after the first infusion, was reduced after subsequent administrations, even when the dose was escalated. T cells from animals with prolonged in vivo exposure exhibited unperturbed target cell lysis ex vivo, indicating no exhaustion. A transient decrease in red cell mass was observed, with no neutropenia or thrombocytopenia. These studies support clinical testing of MGD006 in hematological malignancies, including AML.