Document resource
Background Adoptive cell therapy (ACT), in which T cells are isolated from a patient, expanded to high numbers, and reinfused as a ‘living’ anticancer drug, has shown remarkable success in some cancers. However, relapses are common, and some cancers have no response to ACT. While we have made major advances in engineering better receptors, the cell culture and expansion steps use media formulations that contain hyperphysiologic concentrations of many nutrients, most notably glucose, which in some media can be 10-12x higher than typical serum levels. We hypothesize hyperglycemic conditions used to expand T cells may favor expansion but at the cost of cellular function and longevity.Methods We generated therapeutic T cells from mice or humans, expanding them in different concentrations of glucose representative of commonly used culture media. We assessed the cells functionally and metabolically throughout and at the end of T cell expansion, as well as interrogated their therapeutic efficacy in multiple in vivo models.Results Therapeutic murine TCR-Tg and human CAR-T cells cultured in hyperglycemic media show reduced efficacy when used to treat tumor-bearing animals. Hyperglycemic culture conditions do not promote enhanced T cell expansion, and in fact render T cells poorly responsive to TCR stimulation. Mechanistically, we find exposure to supraphysiologic glucose drives elevated O-linked glycosylation, most notably O-GlcNAc, across the proteome. O-GlcNAc is generated via the hexosamine biosynthesis pathway, a branching point of glycolysis, suggesting that exposure to high levels of glucose may drive enhanced glycolytic flux or shunting through branch pathways. Indeed, we have found elevated expression of OGT, the enzyme that covalently adds O-GlcNAc to proteins, in high glucose T cells. Secondarily, high glucose conditions also drive tonic calcium fluxes which may prematurely induce a hypofunctional phenotype. After T cell infusion, T cells cultured in hyperglycemic conditions do not compete effectively in the tumor microenvironment and fail to produce inflammatory cytokines, even days after reinfusion. Therapeutic T cells expanded in high levels of glucose and isolated from the tumor bear an increased exhaustion profile suggesting that excess glucose may drive T cell terminal dysfunction.Conclusions Our data suggest T cells cultured in hyperglycemic conditions have decreased sensitivity to stimulation, particularly early in activation which leads to poorer T cell polyfunctionality and anti-tumor efficacy. Our data suggest that in T cell manufacturing ‘more is not better’, and that a more regulated and physiologic approach to cell expansion may bring the promise of T cell therapies to more patients.Ethics Approval All animal studies approved under IACUC protocol number 23073380 (Greg M Delgoffe) by the University of Pittsburgh.