Abstract:Introduction: Adoptive cell therapies using CAR T cells achieve great success in treating hematological malignances. Despite high complete response rates (70%-90%) in relapsed/refractory (r/r) B-cell acute lymphoblastic leukemia with current CD19 CAR T cell therapies, 40%-60% of responders experience relapse with up to 60% due to CD19 antigen loss. Similarly, despite high initial response rates in CAR T cell treated r/r multiple myeloma patients, most eventually experience disease progression with relapses due to antigen loss. Dual targeting approaches combining two tumor antigens could overcome these limitations and increase treatment effectiveness. CAR T cell therapies commonly rely on lentiviral vectors for CAR T cell engineering. While these viral vectors can provide high transduction efficiencies, they have limited genetic cargo capacity, oncogenicity safety concerns, and high manufacturing cost and complexity. To address these shortcomings, we developed a non-viral delivery approach using a bioengineered mammalian transposase derived from the mammal, Myotis lucifugus (bMLT) that efficiently integrates large DNA cargo (>100 kb). Unlike gene editing, this system inserts new DNA into human T cells without causing double-strand DNA breaks or needing host DNA repair. Here, we show the ex vivo and in vivo efficacy of this platform to generate dual CAR T cells to overcome antigen loss in hematological malignancies. Methods: Dual CAR T cells targeting CD19/CD22 or BCMA/GPRC5D antigens were generated by delivering a DNA dual CAR cargo (>7.4 kb) and bMLT mRNA to T cells by an optimized electroporation protocol. CAR-expressing T cells were enriched and expanded in the presence of methotrexate. CAR expression, CD4 and CD8 T cell frequencies, and memory and exhaustion phenotypes were analyzed by flow cytometry. CAR T cell potency was assessed using a luciferase-based cytotoxicity assay at different E:T ratios and by measuring the release of IFNγ, TNFα, and GZMB in tumor co-culture assays. In vivo efficacy was evaluated in a Raji lymphoma xenograft mouse model, where tumor antigen loss was simulated by co-injecting a small number of CD19-KO Raji cells. Results: Our transposon technology achieved over 70% stable dual CAR T expression with a high frequency of Tscm/Tcm cells (>65%), low exhaustion markers expression (<2% co-expressing PD-1, TIM-3, and LAG-3), and efficient killing and IFNγ release in tumor co-cultures. Importantly, the dual CD19/CD22 CAR T cells, unlike single-targeting CD19 CAR T cells, were highly efficacious in vivo against dual antigen-expressing Raji cells but completely abrogated tumor growth against a mix of wild-type and CD19-KO Raji cells. Conclusions: Our bioengineered mammalian transposon technology efficiently generates potent dual-targeting CAR T cells with an optimal immunophenotype profile that overcomes tumor antigen loss in vivo. This technology can surpass the cargo capacity and other limitations of viral vectors, enabling more complex CAR T cell designs for cancer immunotherapy. Citation Format: Francisco Navarro, Amanda Vandeveer, Joshua LaMora, Javier García-Vilas, Priya Viswanathan, Joseph J Higgins. A bioengineered mammalian transposon for dual targeting CAR T cells: Addressing antigen escape in hematological cancers [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Tumor Immunology and Immunotherapy; 2024 Oct 18-21; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2024;12(10 Suppl) nr B042.

Abstract B042: A bioengineered mammalian transposon for dual targeting CAR T cells: Addressing antigen escape in hematological cancers