While conventional chimeric antigen -receptor (CAR) -T therapies have shown remarkable clinical activity in some settings, they can induce severe toxicities and are rarely curative. To address these challenges, we developed a controllable cell therapy where synthetic D -domain -containing proteins (soluble protein antigen -receptor X -linker [SparX]) bind one or more tumor antigens and mark those cells for elimination by genetically modi fi ed T cells (antigen -receptor complex [ARC] -T). The chimeric antigen receptor was engineered with a D -domain that speci fi cally binds to the SparX protein via a unique TAG, derived from human alpha-fetoprotein. The interaction is mediated through an epitope on the TAG that is occluded in the native alpha-fetoprotein molecule. In vitro and in vivo data demonstrate that the activation and cytolytic activity of ARC -T cells is dependent on the dose of SparX protein and only occurs when ARC -T cells are engaged with SparX proteins bound to antigen -positive cells. ARC -T cell speci fi city was also redirected in vivo by changing SparX proteins that recognized different tumor antigens to combat inherent or acquired tumor heterogeneity. The ARC-SparX platform is designed to expand patient and physician access to cell therapy by controlling potential toxicities through SparX dosing regimens and enhancing tumor elimination through sequential or simultaneous administration of SparX proteins engineered to bind different tumor antigens.

Supplementary Figure from Preclinical Efficacy of BCMA-Directed CAR T Cells Incorporating a Novel D Domain Antigen Recognition Domain

Abstract Chimeric antigen receptor (CAR) T-cell therapies directed against B-cell maturation antigen (BCMA) have shown compelling clinical activity and manageable safety in subjects with relapsed and refractory multiple myeloma (RRMM). Prior reported CAR T cells have mostly used antibody fragments such as humanized or murine single-chain variable fragments or camelid heavy-chain antibody fragments as the antigen recognition motif. Herein, we describe the generation and preclinical evaluation of ddBCMA CAR, which uses a novel BCMA binding domain discovered from our D domain phage display libraries and incorporates a 4-1BB costimulatory motif and CD3-zeta T-cell activation domain. Preclinical in vitro studies of ddBCMA CAR T cells cocultured with BCMA-positive cell lines showed highly potent, dose-dependent measures of cytotoxicity, cytokine production, T-cell degranulation, and T-cell proliferation. In each assay, ddBCMA CAR performed as well as the BCMA-directed scFv-based C11D5.3 CAR. Furthermore, ddBCMA CAR T cells demonstrated in vivo tumor suppression in three disseminated BCMA-expressing tumor models in NSG-immunocompromised mice. On the basis of these promising preclinical data, CART-ddBCMA is being studied in a first-in-human phase I clinical study to assess the safety, pharmacokinetics, immunogenicity, efficacy, and duration of effect for patients with RRMM (NCT04155749).

Abstract Genetically engineered T cells have demonstrated great promise in the treatment of hematologic malignancies including Acute Lymphocytic Leukemia, Non-Hodgkin's lymphoma, and Multiple Myeloma. However, those achievements are often associated with uncontrolled toxicities, compromised T cell persistence, and frequent tumor relapse. Herein, we report a novel T cell therapy comprising a soluble tumor targeting protein (SPRX001) that specifically binds BCMA-expressing multiple myeloma cells and “TAGS” those cells for destruction by TAG-specific, ex vivo transduced T cells known as Antigen Receptor Complex (ARC) T cells. Functional in vitro studies of ARC-T cells produced from healthy subjects in combination with SPRX001 after co-culture with multiple BCMA-expressing cancer cell lines demonstrated dose-dependent cytokine production, T cell proliferation, degranulation and cytotoxic activity. The cytolytic machinery of the ARC-T cells is only activated when the tri-complex of the ARC-T cell, SPRX001, and BCMA-expressing cell is fully formed. In the absence of any one component, ARC-T cells are not activated thereby preventing kill of the intended target cells. ACLX-001 demonstrated dose- and schedule-dependent in vivo efficacy in disseminated BCMA-expressing tumor models in NSG immunocompromised mice. SPRX001 doses of 0.3 mg/kg daily or 3 mg/kg twice a week were able to eliminate measurable tumor burden as early as 7 days post dosing that lasted for the duration of the study. ARC-T cell expansion and contraction tracked tumor burden in vivo and displayed a diverse memory phenotype. The data support the paradigm that ARC-T cells can be activated and silenced by controlling the dose and schedule of administered SPRX001, which may improve the safety and effectiveness of T cell therapy. Citation Format: Justin P. Edwards, Janine Buonato, David LaFleur, Jeffrey Swers, Jenny Mu, Liubov Zaritskaya, Sinnie Ng, Ankit Gupta, Hui Wang, Samantha McCullough, David Hilbert, David Tice. ACLX-001, a novel BCMA-targeted CAR-T cell therapy that can be activated, silenced, and reprogrammed in vivo with soluble protein adapters in a dose dependent manner [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 1549.

Abstract Unlike the tumor targets associated with initial CAR T cell successes, the vast majority of tumor associated antigens are not unique, but rather expressed at elevated levels compared to normal tissue. For example, while CD123 is overexpressed on many AML tumor cells, concerns regarding on-target, off-tumor toxicities - often associated low-level CD123 expression on normal endothelial cells - has hindered clinical success. Therefore, future generations of CAR T therapies will require novel strategies that have the ability to discriminate between normal and elevated antigen levels - potentially through mechanisms that involve 1) the simultaneous or sequential targeting of multiple tumor associated antigens and 2) agents that contain affinity-tuned targeting domains. Due to the large inventory of validated antibodies against a variety of therapeutic targets, scFv naturally emerged as the obvious and justifiable targeting domain for chimeric antigen receptors. However, many of the characteristics that have made antibodies versatile and effective recombinant therapeutics (e.g., high affinity, bivalency, antibody-dependent cytotoxicity, complement-dependent cytotoxicity, FcRn recycling, and low renal filtration rates) are not advantageous for membrane associated chimeric receptors. Furthermore, scFv are not native protein structures and their development, particularly as it pertains to solubility and aggregation, can be challenging. Therefore, we sought to develop a simple, highly selective targeting domain that could be engineered into complex, potentially multispecific therapeutics. We describe the design and development of non-scFv-derived binding domains. Using phage display and targeted mutagenesis, we identify a series of binding domains that target CD123 with high specificity - as characterized by functional and tissue binding studies. These domains exhibit affinities that range over 2 logs and demonstrate in vitro potencies that correlate with their affinities. CARs comprised of higher affinity binding domains mediate potent T-cell activation and cytolysis of CD123-expressing target cells and induce complete durable remission in two AML xenograft models. We also describe a strategy of engineering less immunogenic binding domains through the identification and removal of putative T cell epitopes and demonstrate that the resultant variants retain biological activity. Finally, we demonstrate further potential of our binding domains by generating functional, bi-specific CARs comprised of a CD123-specific binding domain and a CD19-specific scFv. The ability to incorporate our target-specific binding domains with a range of affinities into complex chimeric receptors, affords a viable alternative to scFv as targeting domains in CAR T cell therapeutics. Citation Format: David W. LaFleur, Haiying Qin, Justin P. Edwards, Liubov Zaritskaya, Ankit Gupta, C. Jenny Mu, Laura K. Richman, Terry J. Fry, David M. Hilbert. Chimeric antigen receptors incorporating novel binding domains targeting CD123 direct potent antitumor activity of T cells: Correlation between affinity and activity [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 601.

3041 Background: Conventional chimeric antigen receptor T cell (CAR-T) therapies have achieved limited clinical success in the treatment of solid tumors, in part due to the challenges of identifying tumor antigen(s) that are uniquely expressed on tumor cells. The dearth of such targets requires that current CAR-T therapies be re-engineered to preferentially target tumor cells thereby mitigating potential on-target off-tumor toxicity to normal cells. Herein we describe a novel cell therapy platform comprising Antigen Receptor Complex T (ARC-T) cells that are readily activated, silenced, and reprogrammed in vivo by administration of a novel tumor-targeting soluble protein antigen-receptor X-linker (sparX). The formation of the ARC-T, sparX, and tumor complex is required for the ARC-T to kill the tumor. Because ARC-T activity is entirely dependent on the dose of sparX administered, therapeutic doses of sparX may be defined that preferentially target cells over-expressing a target antigen and thus limit coincident kill of normal cells expressing lower levels of target antigen. Methods: We have created a library of sparX that bind different cell surface antigens, including HER2. The HER2 sparX was tested as both monovalent and bivalent constructs in vitro by assessing ARC-T cell activation, cytokine release and target cell cytotoxicity. In vivo efficacy models utilized NSG mice and incorporated tumor volume measurements and histopathologic assessments to evaluate tumor clearance. Results: In vitro studies demonstrate that co-culture of ARC-T cells, sparX-HER2 and HER2-expressing target cells drives T cell activation, expansion, cytokine secretion and cytotoxicity of target cells in a dose-dependent manner. Furthermore, by affinity tuning the HER2 binding domain and bivalent formatting of sparX-HER2, we achieved selective killing of HER2-overexpressing breast cancer cells with minimal effect on cells expressing HER2 levels representative of normal tissues. In vivo proof-of-principal studies with ARC-T/sparX-HER2 similarly demonstrate complete eradication of HER2-overexpressing solid tumor cells. Conclusions: These results demonstrate that a single intravenous dose of ARC-T cells can traffic to a solid tumor site and induce tumor eradication upon systemic administration and co-localization of tumor-targeting sparX in a mouse model. Bivalent formatting of sparX-HER2 further enabled ARC-T sensitivity to target antigen density to avoid the on-target off-tumor toxicity that has hindered conventional monovalent CAR-T treatments.

Chimeric antigen receptor (CAR) T cell therapies have demonstrated impressive initial response rates in hematologic malignancies. However, relapse rates are significant, and robust efficacies in other indications, such as solid tumors, will likely require novel therapeutic strategies and CAR designs. To that end, we sought to develop simple, highly selective targeting domains (D domains) that could be incorporated into complex, multifunctional therapeutics. Herein, we describe the identification and characterization of D domains specific for CD123, a therapeutic target for hematologic malignancies, including acute myelogenous leukemia (AML). CARs comprised of these D domains mediate potent T cell activation and cytolysis of CD123-expressing target cells and induce complete durable remission in two AML xenograft models. We describe a strategy of engineering less immunogenic D domains through the identification and removal of putative T cell epitopes and investigate the binding kinetics and affinity requirements of the resultant D domain CARs. Finally, we extended the utility of D domains by generating functional, bi-specific CARs comprised of a CD123 specific Ddomain and a CD19-specific scFv. The properties of D domains suggest that this class of targeting domain may facilitate the development of multi-functional CARs where conventional, scFv-based designs may be suboptimal.