使用基于仿生AeroCyte的人工抗原呈递细胞平台实现功能性CAR-T细胞的长期离体扩增

Strategies to sustain CAR-T cell persistence in vivo include enriching stem-like products, repeated infusions of engineered cells, and improving the tumor microenvironment. Peripheral T lymphocytes exhibit remarkable self-renewal, with studies showing up to 10 41 -fold expansion in syngeneic mice over 10 years (Nature, 2023; 614:762), and similar potential ex vivo, though technical barriers persist. Here, we assessed a biomimetic phospholipid microbubble platform (AeroCyte) designed as artificial antigen-presenting cells (aAPCs) to support long-term ex vivo CAR-T cell expansion. AeroCyte-J, coated with appropriate ligands, mimics natural cell-to-cell (juxtacrine) interactions for T cell stimulation and undergoes spontaneous dissolution in media. We hypothesized this system would enhance ex vivo production of stem-like T cells and enable repeated dosing to overcome hostile in vivo conditions. CD4 + T cells were isolated from human blood using anti-CD4-conjugated AeroCyte and activated with anti-CD3/CD28-conjugated AeroCyte (AeroCyte-J328) for two weeks, then restimulated biweekly for about 6 months, achieving up to 10 16 -fold expansion. Notably, expansion was absent without AeroCyte-J328. Next, three CAR-T cell lines targeting GD2, HLA-A2, or CD19, which were generated from a vial of frozen PBMCs from a different donor, were expanded over seven biweekly cycles using AeroCyte-J328. The expanded cells were CD3⁺ and included both CD4⁺ and CD8⁺ subsets. CAR-T cells from each stimulation cycle were evaluated for target-specific cytotoxicity by co-culturing them for three days without IL-2 with GFP-expressing Raji cells (CD19⁺, HLA-A2⁻, GD2⁻) and T98G cells (CD19⁻, HLA-A2⁺, GD2⁺). Using the number of target cells co-cultured with non-transduced PBMCs as the baseline control (set at 100%), all three CAR-T cell types demonstrated target specificity, particularly at lower effector-to-target (E:T) ratios, as determined by flow cytometry or fluorescence microscopy. Reproducibility was confirmed by expanding CAR-T cell lines derived from one of three additional donor PBMC samples across four biweekly cycles, which produced consistent and comparable results. Long-term expansion depended on continuous aAPC stimulation and is attributed to intrinsic T cell self-renewal rather than malignant transformation, given that expansion is AeroCyte-J328-dependent and the improbability of simultaneous malignant transformation across independent cultures. Additional preclinical and clinical validation of this simple and robust platform could make CAR-T therapy more accessible and affordable, as well as enable improved treatment of solid tumors, for example through repeated dosing. Supported by NIH grants R44CA265468 & R43CA298530