Acute myeloid leukemia (AML) presents significant challenges for CAR T-cell therapy due to its pronounced heterogeneity and the lack of leukemia-specific surface antigens. Frequently targeted antigens, such as CD33, CD123, and CLL-1, are also present on normal hematopoietic progenitors, resulting in on-target, off-tumor toxicity and restricting clinical translation. To address these challenges, logic-gated CAR T-cell strategies have been developed to enable combinatorial antigen recognition. These approaches incorporate engineered circuits, including AND, OR, and NOT gates, as well as synNotch receptors, split-CAR configurations, and inhibitory platforms (iCARs and Tmod), to improve discrimination between leukemic and normal cells. In AML, CAR T-cell efficacy and persistence are further affected by the immunosuppressive bone marrow and lymphoid microenvironment, which involves immune cell trafficking, cytokine signaling, and lymphatic immune regulation. Preclinical studies employing dual-target strategies, such as CD33/CD123 and CLL-1/CD123, have shown improved antileukemic efficacy with reduced hematopoietic toxicity. This review summarizes the molecular principles underlying logic-gated CAR-T systems and examines their translational application in AML. Additionally, it highlights emerging evidence connecting the regulation of lymphatic and immune microenvironments to CAR T-cell persistence, trafficking, and toxicity and discusses future strategies, such as single-cell antigen mapping, computational circuit engineering, and synthetic immune programming, to enhance the precision and clinical feasibility of next-generation AML immunotherapies.
Neeli et al. (Fri,) studied this question.