Transcriptional and functional profiling reveals unique activation of Adapter CAR T cells in acute myeloid leukemia

Chimeric antigen receptor (CAR) T cell therapy has shown great promise in treating malignant diseases, yet its efficacy is often limited by challenges such as controllability and the lack of tumor-specific antigens. Adapter CAR (AdCAR) T cells address these limitations by redirecting T cell activity through adapter molecules rather than direct interaction with tumor-associated antigens, offering an on-/off- switch mechanism and the potential for multitargeting. While functionality of AdCAR T cells has been demonstrated in various models, the underlying AdCAR T cell response to differing adapter concentration and its comparison to conventional CAR T cells remain poorly characterized. In this study, we systematically examined adapter-dependent AdCAR T cell responses using functional assays and single-cell RNA sequencing in an acute myeloid leukemia model. Adapter concentration was found to determine AdCAR T cell activation dynamics, transcriptional states and metabolic reprogramming. At low adapter concentrations, an interferon-responsive state was observed, while high concentrations induced strong cytolytic activity, cytokine secretion, and metabolic shifts toward glycolysis and oxidative phosphorylation. Importantly, increasing adapter concentrations enhanced tumor control of sub-optimally activated AdCAR T cells, demonstrating the platform’s tunability. A hallmark gene signature was identified across different adapters and target cells, comprising the interferon-responsive gene MX1 and activation-associated genes including ENO1, HSP90AB1 , and RRM2 . Our findings provide a mechanistic framework for tuning AdCAR T cell responses and offer critical insights for optimizing adapter dosing to enhance the safety and efficacy of tunable CAR T cell therapies.