Split CAR-T Cells Targeting CD312 and TIM-3 for AML to Reduce the Risk of Antigen Escape

Background: Acute myeloid leukemia (AML) is a malignant disorder characterized by the presence of a limited number of leukemia-initiating cells or leukemic stem cells (LSCs), which form a clonal population. Traditional therapies often fail to achieve durable responses due to the disease's intrinsic resistance mechanisms and the presence of leukemic stem cells. CD312 Chimeric antigen receptor (CAR)-T cell therapy has emerged as a promising treatment modality. However, its efficacy is limited when targeting a single antigen, leading to antigen escape and subsequent relapse. TIM-3 CAR-T cell therapy may improve the clinical outcomes of patients with AML by eradication of the LSCs present in the minimal residual disease (MRD). Objective: In this study, we aim to evaluate the efficacy of modified split CD312/TIM-3 CAR-T cell against AML tumor cells with varying levels of antigen expression through in vitro experiments and immunodeficient mouse models, particularly focusing on their ability to eradicate LSCs and reduce the risk of relapse following CAR-T cell infusion. Methods: The expression levels of CD312 and TIM-3 were assessed in AML cell lines and primary blasts by flow cytometry. Subsequently, split CD312/TIM-3 CAR-T cell were developed. The cytotoxicity of CD312 CAR-T cell and split CD312/TIM-3 CAR-T cell were evaluated in vitro against AML cell lines, primary blasts. Additionally, the activation rates and cytotoxicity of split CD312/TIM-3 CAR-T cell against CD34+CD38- LSCs and normal monocyte were assessed in a coexisting milieu consisting of CAR-T cells, primary blasts, and CD64-positive monocyte. Then, the cellular characteristics and functionality were compared through in vitro experiments. Ultimately, the efficacy and safety of CD312 CAR-T cell and split CD312/TIM-3 CAR-T cell were evaluated in vivo. Results: Split CD312/TIM-3 CAR-T cell demonstrated significantly enhanced cytotoxicity against AML cell lines, primary blasts and CD34+CD38- LSCs compared to CD312 CAR-T cell. Meanwhile, split CD312/TIM-3 CAR-T cell revealed a remarkable cytotoxicity against CD34+CD38- LSCs and remained unaffected by monocyte influence derived from healthy donors PBMCs in a coexisting milieu consisting of CAR-T cells, LSCs and CD64-positive monocyte. In a relapsed AML animal transplantation model, split CD312/TIM-3 CAR-T exhibited remarkable anti-AML activity, leading to prolonged survival of the treated animals. Importantly, the split CAR-T cells could be safely eliminated in emergency situations through a natural safety switch mechanism, ensuring that hematopoietic stem cells remained unaffected. Conclusion The split CAR-T cells targeting CD312 and TIM-3 exhibit potential as a promising and secure therapeutic approach for patients with refractory and relapsed AML, particularly focusing on their ability to eradicate LSCs and reduce the risk of relapse following CAR-T cell infusion.