Search

4–1BB (CD137) is an activation-induced costimulatory receptor that regulates immune responses of activated CD8 T and natural killer cells, by enhancing proliferation, survival, cytolytic activity, and IFNγ production. The ability to induce potent antitumor activity by stimulating 4–1BB on tumor-specific cytotoxic T cells makes 4–1BB an attractive target for designing novel immuno-oncology therapeutics. To minimize systemic immune toxicities and enhance activity at the tumor site, we have developed a novel bispecific antibody that stimulates 4–1BB function when co-engaged with the tumor-associated antigen 5T4. ALG.APV-527 was built on the basis of the ADAPTIR bispecific platform with optimized binding domains to 4–1BB and 5T4 originating from the ALLIGATOR-GOLD human single-chain variable fragment library. The epitope of ALG.APV-527 was determined to be located at domain 1 and 2 on 4–1BB using X-ray crystallography. As shown in reporter and primary cell assays in vitro, ALG.APV-527 triggers dose-dependent 4–1BB activity mediated only by 5T4 crosslinking. In vivo, ALG.APV-527 demonstrates robust antitumor responses, by inhibiting growth of established tumors expressing human 5T4 followed by a long-lasting memory immune response. ALG.APV-527 has an antibody-like half-life in cynomolgus macaques and was well tolerated at 50.5 mg/kg. ALG.APV-527 is uniquely designed for 5T4-conditional 4–1BB–mediated antitumor activity with potential to minimize systemic immune activation and hepatotoxicity while providing efficacious tumor-specific responses in a range of 5T4-expressing tumor indications as shown by robust activity in preclinical in vitro and in vivo models. On the basis of the combined preclinical dataset, ALG.APV-527 has potential as a promising anticancer therapeutic for the treatment of 5T4-expressing tumors.

Supplementary figures and methods

4–1BB (CD137) is an activation-induced costimulatory receptor that regulates immune responses of activated CD8 T and natural killer cells, by enhancing proliferation, survival, cytolytic activity, and IFNγ production. The ability to induce potent antitumor activity by stimulating 4–1BB on tumor-specific cytotoxic T cells makes 4–1BB an attractive target for designing novel immuno-oncology therapeutics. To minimize systemic immune toxicities and enhance activity at the tumor site, we have developed a novel bispecific antibody that stimulates 4–1BB function when co-engaged with the tumor-associated antigen 5T4. ALG.APV-527 was built on the basis of the ADAPTIR bispecific platform with optimized binding domains to 4–1BB and 5T4 originating from the ALLIGATOR-GOLD human single-chain variable fragment library. The epitope of ALG.APV-527 was determined to be located at domain 1 and 2 on 4–1BB using X-ray crystallography. As shown in reporter and primary cell assays in vitro, ALG.APV-527 triggers dose-dependent 4–1BB activity mediated only by 5T4 crosslinking. In vivo, ALG.APV-527 demonstrates robust antitumor responses, by inhibiting growth of established tumors expressing human 5T4 followed by a long-lasting memory immune response. ALG.APV-527 has an antibody-like half-life in cynomolgus macaques and was well tolerated at 50.5 mg/kg. ALG.APV-527 is uniquely designed for 5T4-conditional 4–1BB–mediated antitumor activity with potential to minimize systemic immune activation and hepatotoxicity while providing efficacious tumor-specific responses in a range of 5T4-expressing tumor indications as shown by robust activity in preclinical in vitro and in vivo models. On the basis of the combined preclinical dataset, ALG.APV-527 has potential as a promising anticancer therapeutic for the treatment of 5T4-expressing tumors.

BackgroundAPVO603 is a dual targeting bispecific antibody for 4-1BB (CD137) and OX40 (CD134), engineered with Aptevo's ADAPTIRTM technology. We have previously shown that the distinct characteristics of APVO603 may enable conditional agonism of 4-1BB and OX40 only when cross-linked through engagement of the other receptor via cis and/or trans cellular interactions. Thus, APVO603 is designed with the potential to overcome both the on-target toxicity and limited efficacy observed with 4-1BB and OX40 monoclonal antibody treatment in the clinic.MethodsGenevestigator Software was used to analyze curated transcriptomic data for the expression profiles of OX40 and 4-1BB across select human heme and solid cancer patient sample data sets, as well as, non diseased tissue. Primary inducible Treg (iTreg) cells were sub-optimally stimulated with an anti-CD3/CD28 antibody and cell proliferation was assessed using CFSE-labelled. Cytokines were measured using intracellular flow-based methods. For in vitro tumor lysis studies, activated T cells were co-cultured with Nuclight-labelled tumor cells expressing a tumor-associated antigen (TAA) and activated with TAA x CD3 bispecific protein. Live tumor cells were continually assessed using the Incucyte Live-Cell Analysis System and Cell-By-Cell Software Module.ResultsOX40 and 4-1BB displayed distinct tumor expression profiles, however, several tumor indications were identified with high co-expression and may aid in identifying indications for the clinical development of APVO603. In vitro, APVO603 favored activation of effector T cell subsets and had minimal impact in augmenting iTreg cells proliferation, cytokine production or expression of effector-related molecules, despite the fact that a portion of the iTreg cells expressed OX40 and 4-1BB. The mechanistic activity of APVO603 resulted in dose-dependent control of in vitro tumor growth when paired with a T-cell activating TAA x CD3 bispecific under standard conditions or those leading to T cell exhaustion. In preclinical assays using PBMCs sub-optimally stimulated with TAA x CD3, APVO603 enhanced TAA-expressing tumor cell lysis when compared to TAA x CD3 alone.ConclusionsAPVO603 is a dual-agonistic bispecific antibody that augments the effector function of activated CD4+ and CD8+ T cells and NK cells, but not iTreg cells, in a dose-dependent manner and reduces growth of tumors in vitro and in vivo. Further, mechanistic evaluation supports the ability of APVO603 to pair with T-cell modulating IO approaches to support a more fit T cell response and favorable TME. This preclinical data supports further development of APVO603, a promising immuno-oncology therapeutic with potential for benefit in hematologic and solid tumors.

APVO442 is a bispecific therapeutic candidate targeting PSMA and CD3 for redirection of T cell responses against PSMA-expressing solid tumors. The candidate bispecific was designed to have unique pharmacokinetic and safety properties to potentially maximize robust anti-tumor responses. APVO442 is designed with Aptevo’s ADAPTIR-FLEX࣪ technology to generate a molecule with low-affinity monovalent CD3 engagement, paired with high-affinity bivalent PSMA binding intended to be tumor - dependent while inducting optimal T-cell stimulation within the tumor microenvironment. The unique engineering of APVO442 reduces the potential peripheral CD3 T cell binding, minimizing the potential for on-target toxicity, such as peripheral cytokine release, while simultaneously increasing the likelihood to deliver an effective concentration of APVO442 locally within solid tumors. Preclinical data demonstrate that APVO442 is able to activate T cells and mediate directed cytotoxicity against PSMA-expressing tumor targets with comparable potency to α-PSMA x α-CD3 constructs with varying CD3 affinity or avidity both in vitro and in vivo. The CD3 activity is dependent on PSMA crosslinking as activity is not observed in the absence of tumor target, in line with the silenced Fc engineered into APVO442. Importantly, cytokine release was not observed in peripheral immune subsets in the absence of tumor targets, despite a beneficial cytokine profile observed in the presence of PSMA positive tumor cell lines. The unique biology elicited by APVO442 induces distinct kinetic activation of downstream transcriptional regulators following monovalent TCR engagement of APVO442 associated with reduced early signaling, but overall, a more sustained activation than higher affinity variants. In addition, APVO442 induces a distinct phenotypic profile on human T cells associated with optimal function and memory responses only in the presence of tumor targets. Further, combination of APVO442 with additional costimulatory agonists supports improved T cell activation and anti-tumor cytotoxicity in vitro. These data support the potential for unique clinical combination of APVO442 with immuno-oncology modalities that may support improved benefit in PSMA positive indications such as metastatic castration resistant prostate cancer. APVO442 is undergoing further preclinical in vitro and in vivo characterization in support of the IND-enabling package. Citation Format: Rebecca Gottschalk, Robert Miller, Brian Woodruff, Ashly Lucas, Elizabeth Haglin, Peter Pavlik, Michelle H. Nelson, Hilario J. Ramos. APVO442 is a distinct PSMA x CD3 targeted bispecific candidate designed to optimize T cell fitness and distribution to solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3434.

Prostate-Specific Membrane Antigen (PSMA), is a tumor-associated antigen (TAA) that is expressed on prostate cancers, including metastatic castration-resistant prostate cancer (mCRPC). Current chemotherapeutic approaches for mCRPC are challenged by development of resistance resulting in limited clinical benefit. Immuno-oncology therapeutic candidates such as bispecifics re-directing T-cell responses to eliminate tumors, are a promising strategy to overcome the limitations of current approaches and provide benefit to patients with aggressive cancers. Here we present preclinical data demonstrating a potential new approach using low affinity targeting of CD3 and high affinity targeting of PSMA for treatment of a solid tumor cancer.APVO442 is Aptevo's bispecific candidate targeting PSMA and CD3. This candidate was designed to have unique pharmacokinetic and safety properties to potentially maximize potent anti-tumor responses against mCRPC. The APVO442 bispecific T-cell engager uses Aptevo's ADAPTIR-FLEX technology to generate a molecule with low-affinity monovalent CD3 engagement, paired with high-affinity bivalent PSMA binding designed to deliver a highly selective T-cell response at the tumor. The unique engineering of APVO442 reduces the potential of binding to CD3 expressed on peripheral T cells, thus minimizing the potential for on-target toxicity, such as cytokine release, and increasing the potential to deliver the effective concentration of the molecule localized to solid tumors.Preclinical testing demonstrated that APVO442 exhibits optimal manufacturability and functional characteristics for lead candidate selection. Anti-PSMA x anti-CD3 constructs with varying binding strengths to CD3 were evaluated for specificity of CD3 binding, ability to enhance T-cell activation, and ability to elicit T-cell-mediated cytotoxicity against PSMA-expressing tumor targets with varying levels of PSMA expression. APVO442 demonstrated 10-fold reduced binding to CD3 and EC50 compared to the highest affinity constructs tested while retaining equivalent binding to tumor cells expressing various levels of PSMA. The differences in CD3 affinity were associated in a slightly lower EC50 of potency however, the maximal responses for in vitro activation of CD4+ and CD8+ T cells including upregulation of CD25/CD69 expression, proliferation and in vitro tumor lysis were comparable between low and high affinity CD3 constructs. Finally, APVO442 induced reduced levels of multiple cytokines in vitro when compared to high affinity competitor molecules.In vivo, APVO442 elicited robust anti-tumor responses of human PSMA-expressing tumors in a murine xenograft tumor model. The in vivo activity of APVO442 was comparable to high affinity CD3 engaging comparators with similarly measured PK profiles. Additional in vivo characterization of APVO442 is ongoing and continued pre-clinical studies are planned for 2021. Citation Format: Rebecca Gottschalk, Robert E. Miller, Lynda Misher, Michelle H. Nelson, Allison Chunyk, Brian Woodruff, Elizabeth Haglin, Gabriela Hernandez-Hoyos, Peter Pavlik, Catherine McMahan, Hilario J. Ramos, David Bienvenue. APVO442: A bispecific T cell-engaging candidate utilizing the ADAPTIR-FLEXTMplatform technology with unique properties designed for optimal tumor distribution and cytotoxic response against PSMA-expressing solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr LB172.

Regulatory T cell (Treg) depletion enhances the anti-tumor efficacy of CTLA-4 targeted mAbs in preclinical models, and is a mechanism supported by retrospective clinical data analysis. In response to these findings, many next-generation CTLA-4 targeting therapeutics have been designed to increase Fc-mediated Treg depletion in the tumor microenvironment. CTLA-4-targeted engineered toxin bodies (ETBs) are designed to deplete Tregs in the tumor microenvironment (TME) directly and in a manner independent of Fc-mediated effector functions, offering a unique approach to CTLA-4 targeted therapy. ETBs, or engineered toxin bodies, are large molecule proteins consisting of an antibody fragment genetically fused to a proprietary de-immunized (DI) form of the Shiga-like toxin A subunit (SLTA). Once engaged to the specific cell surface target of interest, in this case CTLA-4, ETBs internalize, route to the cytosol, and permanently inactivate ribosomes through an irreversible enzymatic process. This results in cell death via apoptotic mechanisms.Here we describe the preclinical characterization of a lead candidate CTLA-4-targeted ETB. CTLA-4-ETB-A directly binds and specifically kills CTLA-4 positive cells in vitro and induces apoptosis of ex-vivo expanded Tregs. CTLA-4-ETB-A is designed to bind CTLA-4 in a manner unique from classic blocking antibodies and is not expected to have sustained blocking ability in vivo due to the relatively short half-life of an ETB compared to a neutralizing mAb. We predict this will allow for focused Treg depletion in the TME based on target expression levels, while sparing autoreactive T cell activation in the periphery. Utilizing human CTLA-4 knock in mouse syngeneic tumor models, we show that CTLA-4-ETB-A can reduce the % Tregs and increase the CD8/Treg ratio in the TME. Preclinical assessment of safety and pharmacodynamic signals are underway in cynomolgus studies. Citation Format: Swati Khanna, Caleigh Howard, Lilia A. Rabia, Alvaro Aldana, Jay Zhao, Betty Chang, Hilario J. Ramos, Aimee Iberg. Preclinical characterization of a novel CTLA-4-targeted ETB for direct Treg depletion [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1627.

APVO603 is Aptevo's bispecific candidate targeting 4-1BB and OX40. It was designed to have unique properties with the potential to overcome some of the clinical challenges observed with monoclonal antibody targeting these receptors. APVO603 is engineered as an FcγR-signaling deficient bispecific antibody that utilizes Aptevo's ADAPTIR technology for a distinct approach for dual targeting of 4-1BB and OX40 in the absence of additional effector activity. The distinct characteristics of APVO603 may enable conditional activation of 4-1BB and OX40 via agonism of these receptors only when cross-linked via engagement of the other receptor via cis and/or trans cellular interactions. Thus, APVO603 is designed with the potential to overcome both the on-target toxicity and limited efficacy observed with 4-1BB and OX40 monoclonal antibody treatment in the clinic. Anti-4-1BB and OX40 binding domains were optimized to increase affinity, function and stability, then incorporated into the ADAPTIR bispecific antibody platform to produce the APVO603 lead candidate. APVO603 was found to augment 4-1BB and OX40 activity in a dose-dependent manner and is strictly dependent on engagement of the reciprocal receptor to elicit 4-1BB or OX40 signaling in vitro. In preclinical assays using PBMCs sub-optimally stimulated with anti-CD3, APVO603 induces synergistic proliferation of CD4+, CD8+ T and NK cells when compared to anti-OX40 or 4-1BB antibodies with a wt Fc, included either individually or in combination. Additionally, APVO603 enhances proinflammatory cytokine production, granzyme B expression, and reduces the T cell exhaustion phenotype. The mechanistic activity of APVO603 resulted in dose-dependent control of in vivo tumor growth in a preclinical humanized murine xenograft model using established murine MB49 bladder tumors in human 4-1BB and OX40 double knock-in mice. APVO603 is a dual-agonistic bispecific antibody that augments the effector function of activated CD4+ and CD8+ T cells and NK cells in a dose-dependent manner and reduces growth of established tumors in vivo. This preclinical data demonstrates conditional dual stimulation of 4-1BB and OX40 and supports further development of APVO603, a promising immuno-oncology therapeutic with potential for benefit in solid tumors. This program is progressing into IND-enabling studies later this year. Citation Format: Michelle H. Nelson, Robert E. Miller, Secil Franke-Welch, Ruth Chenault, Hang Fang, Allison Chunyk, Gabriela Hernandez-Hoyos, Hilario J. Ramos, David Bienvenue, Catherine McMahan. APVO603: A dual 4-1BB and OX40 bispecific approach utilizing ADAPTIRTMtechnology designed to deliver a conditional T cell/NK response against solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr LB173.

Targeting PD-L1 has shown clinical efficacy in multiple solid tumor indications. Currently approved PD-L1 targeted approaches rely on monoclonal antibodies which sterically inhibit PD-L1 and prevent PD-1 mediated checkpoint activity. While these molecules have shown great activity in the clinic, the need for pre-existing tumor specific immunity and immune infiltration precludes responses in some patients and leads to resistance in others. Therefore, there remains a need for new modalities and treatment paradigms in these indications. Molecular Templates has developed MT-6402, an engineered toxin body (ETB) targeting PD-L1, as a single agent immunotoxin designed to overcome the challenges of current PD-L1 targeting approaches by 1) directly depleting PD-L1 positive tumor cells or immunosuppressive immune cells displaying PD-L1 in the tumor microenvironment and 2) delivery of an HLA: A*02 restricted viral peptide to alter the tumor immunophenotype for recruitment of CMV-restricted CTLs to target the tumor for depletion (antigen seeding). MT-6402 is slated for clinical development in 2021. Here we describe the preclinical characterization of several ETB candidates derived from MT-6402 delivering antigenic peptides restricted to the most prevalent MHC haplotypes in the U.S. population to broaden the patient population suitable for antigen seeding. ETBs were engineered with the ability to deliver viral peptides across a range of HLA restriction, including HLA: A*01, HLA: A*03, and HLA: A*24. ETBs were screened and benchmarked against MT-6402 and candidates were identified that retain comparable specificity, selectivity, and potency. Alteration of peptide antigen did not change the specificity or selectivity of ETBs which retained similar PD-L1 binding profiles to MT-6402. Binding profiles correlated to targeted potency and ETBs with varied HLA restricted peptides were found to target tumor and immune cells for depletion with similar potency to MT-6402. ETBs delivered an antigen seeding response in a PD-L1 dependent manner and only in conditions in which tumor cell and CTLs shared a matched HLA to the delivered antigenic peptide specificity. Preclinical assessment of the in vivo efficacy and safety profile of candidates is ongoing and further development is slated for 2021. Citation Format: Joseph D. Dekker, Swati Khanna, Elizabeth Saputra, Wenzhao Dong, Lindsey Aschenbach, Lilia A. Rabia, Garrett L. Cornelison, Michaela Sousares, Jay Zhao, Garrett L. Robinson, Betty Chang, Hilario J. Ramos. Engineered toxin bodies targeting PD-L1 to alter tumor immunophenotypes and deliver broad antigenic diversity and patient coverage [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1628.

The pumped hydropower energy storage (PHES) assessments carried out so far have been focused on large water bodies obtained from global or restricted-use databases, or, on the other hand, on the application of methodologies to specific areas focusing on the detection of dams. In addition, many assessments do not include data optimization, or include it at the end of the process and are subject to the prior application of restrictions, often stipulated with subjective criteria. The aim of this article is to design a universal and easily applicable methodology for the assessment of viable PHES potential, which provides immediate and reliable results to assist in the energy planning of a given territory. It is classified in ravine basins, including an optimization before using the restrictions. The island of Gran Canaria is taken as the territory of application, whose density of dams is the highest in the world and whose share of hydroelectric energy is, at present, null, besides, no PHES studies have been carried out. The results show that the PHES potential in Gran Canaria is 5996 MWh after applying the optimization and all technical constraints. If all environmental constraints were rigorously applied, the island would have no possible pairing. The results demonstrate the importance of tailoring the restrictions to each particular territory.

CD45 is highly expressed on the cell surface of all nucleated hematopoietic cells, including malignant cells of B, T and myeloid lineage. CD45, a receptor tyrosine phosphatase that has important roles in antigen receptor signaling, has multiple isoforms that are differentially expressed on immune cell subsets. Targeting of CD45 with antibody-based therapies, tested in conjunction with reduced dose chemotherapy and radiation regimens, has shown promise in preclinical models and clinical trials as a conditioning therapy for bone marrow transplant (BMT), although these approaches are associated with safety limitations. A single agent, targeted conditioning method could increase patient safety and eliminate genotoxic effects associated with the conditioning regimens. Conditioning therapy by depletion of CD45 positive cells specifically has potential applications in oncology as well as non-oncologic, hematopoietic disease settings where engraftment of donor or modified stem cells could provide benefit. Engineered toxin bodies (ETBs) are comprised of a proprietarily engineered form of Shiga-like Toxin A subunit (SLT-A) genetically fused to antibody-like binding domains. ETBs work through novel mechanisms of action and are capable of forcing internalization, self-routing through intracellular compartments to the cytosol, and inducing potent cell-kill via the enzymatic and permanent inactivation of ribosomes. Targeted ETBs are being developed to specifically target and destroy CD45 expressing cells. ETBs targeting CD45 specifically bind to, internalize into and then trigger cell death of the target cells in vitro. ETBs have been designed using antibody fragments that recognize all isoforms of CD45. The CD45 targeted ETB has shown high potency in immortalized blood cancer cell lines of T cell, B-cell and eosinophil origin, including leukemia and lymphoma cell lines. Additionally, the ability of the CD45 targeted ETBs to deplete primary cells from healthy donors has been demonstrated. The direct cell kill activity of the CD45 targeted ETBs have the potential to deplete immune cells in the periphery and the bone marrow, as well as to deplete malignant hematological cells expressing CD45. In vivo depletion of human immune cells in murine models by CD45 targeted ETBs will be evaluated. Citation Format: Aimee Iberg, Garrett L. Robinson, Sara LeMar, Joseph D. Dekker, Jay Zhao, Hilario J. Ramos, Melissa M. Singh, Erin K. Willert. CD45 targeted engineered toxin bodies deplete hematopoietic and malignant cells [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 521.

Engineered toxin bodies (ETBs) comprised of a proprietarily engineered Shiga-like Toxin A subunit (SLTA) genetically fused to antibody-like binding domains work through novel mechanisms of action and can force internalization, self-route through intracellular compartments to the cytosol, and induce potent cell-kill via the enzymatic and permanent inactivation of ribosomes. Our PD-L1 targeted ETB, MT-6402, includes antigen seeding technology (AST), and additionally delivers a viral antigen for presentation in complex with MHC-I to resident viral-specific cytotoxic T lymphocytes (CTLs). The fusion protein provides a powerful, dual mechanism of action to specifically target and destroy PD-L1 positive tumor and inhibitory immune cells. In vitro testing of human tumor cell lines has demonstrated that cell surface PD-L1 expression is required for activity of the PD-L1 targeted ETBs. MT-6402 demonstrated effective depletion of PD-L1 expressing cells across a panel of immortalized cancer cell lines from multiple origins, including lung, skin, breast, and ovary. In a co-culture assay of PD-L1 target cells and antigen specific T-cells, PD-L1 targeted MT-6402 delivers a viral antigen to the target cells and can activate cytokine secretion and T-cell mediated killing. Additionally, MT-6402 treatment of human peripheral blood mononuclear cells (PBMCs) leads to selective depletion of PD-L1 positive cells (IFN-γ stimulated monocytes) in the absence of depletion of PD-L1 low/negative lymphocytes. Murine models, including patient derived xenografts, have demonstrated that the PD-L1 targeted ETBs are efficacious in vivo. Additional in vivo tumor models to further describe the effect of MT-6402 are being explored. Exploratory studies in a primate model have shown that MT-6402 can be tolerated at levels shown to induce T cell activation (type 1 cytokines). MT-6402 delivers a powerful and unique dual mechanism of action. The combination of a PD-L1 specific direct cell kill and redirection of a robust effector T cell response to the tumor has potential benefit in solid tumor indications, including in the relapsed setting, when disease has progressed after checkpoint and/or other therapies. Citation Format: Hilario J. Ramos, Brigitte Brieschke, Sara LeMar, Joseph D. Dekker, Aimee Iberg, Garrett L. Robinson, Asis Sarkar, Banmeet Anand, Melissa M. Singh, Jay Zhao, Jack P. Higgins, Erin K. Willert. In vivo efficacy of a PD-L1 targeted, antigen seeding engineered toxin body [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3366.

Tumor resident regulatory T cells (Tregs) are important mediators of an immunosuppressive tumor microenvironment (TME) promoting tumor immune evasion. The presence of Tregs, and a higher ratio of Tregs to effector T cells in the TME, are associated with poor prognosis. The depletion of Tregs in the TME is expected to re-expose the tumor to the immune system to allow for tumor control. CLTA-4 is expressed on the cell surface of Tregs and is a clinically validated target. Better responses to CTLA-4 monoclonal antibody (mAb) treatment are correlated with stronger ADCC-mediated Treg depletion in preclinical models, and patient FcγR polymorphism has been reported to correlate with response to CTLA-4 mAb therapy. Towards improving on current therapies, many efforts to increase effective depletion of Tregs (such as by Fc effector modification) are being pursued. Engineered toxin bodies (ETBs) are comprised of a proprietarily engineered form of Shiga-like Toxin A subunit (SLT-A) genetically fused to antibody-like binding domains. ETBs work through novel mechanisms of action and are capable of forcing internalization, self-routing through intracellular compartments to the cytosol, and inducing potent cell-kill via the enzymatic and permanent inactivation of ribosomes. Targeted ETBs are being developed to specifically target and destroy CTLA-4 expressing cells. In vitro, cells expressing CTLA-4 are effectively and specifically killed by targeted ETBs. This direct cell kill activity of the ETB has the potential to deplete Tregs in the TME without a requirement for ADCC mechanisms. ETBs have been designed to bind various CTLA-4 epitopes, and to comprise of different targeting domains formats, including monomeric and diabody single chain variable fragments (scFvs) as well as single domain and biparatopic antibody fragments. The development of a CTLA-4 targeted ETB is ongoing. The entry of a protein with direct cell kill properties into the therapeutic space represents a differentiated mechanism of action to deplete Tregs for ultimate re-invigoration of the anti-tumor immune response, and has the potential to provide benefit to patients, including in the relapsed or refractory setting. Citation Format: Aimee Iberg, Edith Acquaye-Seedah, Lilia A. Rabia, Garrett L. Robinson, Hilario J. Ramos, Joseph D. Dekker, Jay Zhao, Erin K. Willert. CTLA-4 targeted engineered toxin bodies designed to deplete regulatory T cells (Tregs) [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2278.

Engineered toxin bodies (ETBs) are comprised of a proprietarily engineered form of Shiga-like Toxin A subunit (SLT-A) genetically fused to antibody-like binding domains. ETBs work through novel mechanisms of action and are capable of forcing internalization, self-routing through intracellular compartments to the cytosol, and inducing potent cell-kill via the enzymatic and permanent inactivation of ribosomes. Three ETBs are in clinical development (MT-3724 targeting CD20, MT-5111 targeting HER2, and TAK-169 targeting CD38). The novel mechanism of action has potential benefit in different indications including in the relapsed setting, when disease has progressed after chemotherapies and other targeted therapies, and additionally may be able to be combined with standard of care. Additional ETBs are being developed that target other cell surface receptors for solid and hematological malignancies, including SLAMF7 (CS1). SLAMF7 is a clinically validated target of monoclonal antibody therapy. Although clinical responses to the monotherapy were not observed, patients treated with combination therapy using standard of care agents that are known to stimulate immune effector cell activity have improved responses. Thus, a SLAMF7 targeted agent that can act independent of effector cells by directly killing tumor cells has the potential to provide broader benefit to patients. ETBs targeting SLAMF7 harbor the catalytic activity of the SLTA subunit and can effectively and specifically kill multiple myeloma cell lines expressing the target protein in vitro. The coupling of specific targeting to myeloma cells with potent and direct cell kill activity has the potential to deplete malignant cells without reliance on effector functions of the patient immune system, such as antibody or complement dependent cellular cytotoxicity or phagocytosis, and could be beneficial when disease progresses post antibody therapy. Additionally, the novel mechanism of action of an ETB also has potential to be combined with standard of care therapies and/ or to be effective in the relapsed or refractory setting. Citation Format: Aimee Iberg, Garrett L. Cornelison, Caleigh Howard, Garrett L. Robinson, Jay Zhao, Hilario J. Ramos, Erin K. Willert. Novel engineered toxin bodies targeting SLAMF7 (CS1) [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 539.

12 Background: Engineered toxin bodies (ETBs) comprised of a proprietarily engineered Shiga-like Toxin A subunit (SLT-A) genetically fused to antibody-like binding domains work through novel mechanisms of action and can force internalization, self-route through intracellular compartments to the cytosol, and induce potent cell-kill via the enzymatic and permanent inactivation of ribosomes. Our PD-L1 targeted ETB includes antigen seeding technology (AST), and additionally delivers a viral antigen for presentation in complex with MHC-I to resident viral-specific cytotoxic T lymphocytes (CTLs). The fusion protein provides a powerful, dual mechanism of action to specifically target and destroy PD-L1+ tumor and inhibitory immune cells. Methods: In vitro activity of ETBs was measured by cell viability and kinetic imaging of labeled target cells in an effector T-cell co-culture assay. Repeat dosing in non-human primates (NHPs) and murine models harboring human PD-L1+ patient derived tumors were used to evaluate tolerability and efficacy. Results: ETBs targeting PD-L1 have potent direct and specific activity across human tumor cell lines and human donor derived monocytes expressing PD-L1 in vitro. In co-culture assays, AST capable ETBs had a deeper response when target cells express both PD-L1 and the appropriate MHC-I allele in the presence of antigen specific CTLs. In vivo, PD-L1 targeted ETBs induced tumor growth delay, regression and survival benefit in multiple PD-L1+ NSCLC PDX models. These ETBs were also tolerated in repeat dosing NHP studies. Conclusions: The potent targeted activity of ETB mediated tumor cell destruction is demonstrated across tumor cell lines from multiple indications, on PD-L1+ immune cells ex-vivo, and by tumor growth and survival benefits in PDX models in vivo. The AST capable, PD-L1 targeted ETBs provide dual novel mechanisms to deplete PD-L1+ cells and control malignancies. Both the SLT-A mediated direct cell kill and activation of viral-specific CTLs circumvents a need for high mutational burden or anti-tumor immune-cell infiltrate at the tumor site, providing a potential treatment option for patients whose disease has progressed after checkpoint therapy.

Engineered Toxin Bodies (ETBs) are comprised of a deimmunized Shiga-like toxin subunit A (SLTA) genetically fused to an antibody-like targeting domain. The antibody targeting domain allows for specific targeting of cancer cells while the SLTA component promotes self-internalization of ETBs, an activity that allows for the delivery of an enzymatic and permanent ribosomal destruction against targeted cells even in the context of non-or-poorly internalizing receptor targets. Molecular Templates has developed PD-L1-targeting ETBs as an approach to directly target tumor cells and overcome resistance mechanisms against PD-1 and PD-L1 antibodies. The cytotoxicity delivered by PD-L1-specific ETBs is engineered to be independent of a requirement for tumor infiltrating lymphocytes (TILs), high tumor mutational burden, or modulatory effects of the tumor microenvironment. Further, the activity is not dependent on blockade of the PD-1/PD-L1 checkpoint axis. Thus, PD-L1 targeting ETBs represent a distinct class of therapeutics with direct cell-kill mechanism of action and ability for activity in patients who have progressed on current standard of care or checkpoint therapy. In this study, we highlight the efficacy and safety profile of MT-6020, a human and cynomolgus cross-reactive, PD-L1 targeted, ETB. MT-6020 retains potent catalytic activity and mediates enzymatic destruction of ribosomes at comparable levels to wild-type SLTA in a cell free model. In addition, MT-6020 binds to human NSCLC, Melanoma, and TNBC tumor cell lines with nM affinity and mediates cellular cytotoxicity via ribosomal destruction at low nM to sub-nM potency. MT-6020 binds to cell lines expressing non-human primate (NHP)-PD-L1 and elicits cytotoxic responses comparable to those observed on human tumor target cells. MT-6020 demonstrated pharmacodynamic and pharmacokinetic effects and displayed a favorable tolerability profile in a repeat dose NHP study at doses that are above the presumed therapeutically active concentration. Further our lead PD-L1 ETB, MT-6035, is built upon the MT-6020 scaffold and can deliver a viral peptide for cell surface presentation to and targeting by a specific antiviral CTL population (antigen seeding technology (AST)) for a second and complementary mechanism for tumor cell destruction. MT-6020 and MT-6035 represent a novel approach to targeting and destroying tumors expressing PD-L1 that is unlikely to be inhibited by resistance mechanisms to current checkpoint inhibitors, is well tolerated in relevant toxicity models, and has the capacity for activity in indications where standard of care has failed. Molecular Templates is poised to initiate clinical development of the PD-L1 targeted-ETB (AST), MT-6035, in 2H - 2019. Citation Format: Hilario J. Ramos, Asis K. Sarkar, Sara Le Mar, Brigitte Brieschke, Joseph D. Dekker, Veronica R. Partridge, Pablo A. Maceda, Michaela M. Sousares, Garrett L. Robinson, Aimee Iberg, Shaoyou Chu, Jensing Liu, Jack P. Higgins, Erin K. Willert. The Safety and efficacy profile of a PD-L1 directed, Engineered Toxin Body, as a novel targeted direct-cell kill approach for the treatment of PD-L1 expressing cancers [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 3900.

Immunotherapeutic approaches such as CAR-T and T cell engaging bispecific antibodies have shown positive clinical outcomes and demonstrate the value of harnessing cytotoxic T lymphocyte (CTL) activity for tumor intervention. Here we describe Engineered Toxin Bodies (ETBs) with Antigen Seeding Technology (AST) as a unique tumor targeted therapeutic approach to recruit endogenous polyfunctional memory CTL responses against cancers. Molecular Templates’ ETB platform combines antibody domains with a proprietary form of the Shiga like Toxin A (SLTA) subunit to specifically target cells and exploit SLTA’s intrinsic ability to self-internalize and destroy ribosomes. With the addition of a foreign class I antigenic peptide fused to an ETB (AST), antigens present in complex with endogenous MHC I to initiate a high avidity T cell response to target cells. Utilizing this approach, we demonstrate ETBs targeted to clinically relevant markers can seed antigenic peptides to hematological tumor, solid tumor, and primary cells. Expanded CTLs from donors with preexisting Human Cytomegalovirus (CMV) reactivity phenotypically resembled memory expansion from natural infection. We next delivered class I CMV peptide to tumor cells by AST with ETBs targeted to PD-L1 and co-cultured tumor cells with expanded CTLs. This led to activation of CMV specific T cells, direct tumor-T cell engagement, and potent cell kill activity above that of parental ETBs. CTL phenotyping and cytokine profiling revealed polyfunctional T cell activation driven by AST enabled ETBs. Thus, ETBs delivering peptide via AST provide additive mechanisms of action: direct cell kill via ribosomal destruction, and antigen specific polyfunctional memory CTL activation.