AMPK agonism optimizes the in vivo persistence and anti-leukemia efficacy of chimeric antigen receptor T cells

Abstract

BACKGROUNDChimeric antigen receptor T cell (CART) therapy has seen great clinical success. However, up to 50% of leukemia patients relapse and long-term survivor data indicate that CART cell persistence is key to enforcing relapse-free survival. Unfortunately, ex vivo expansion protocols often drive metabolic and functional exhaustion, reducing in vivo efficacy. Preclinical models have demonstrated that redirecting metabolism ex vivo can improve in vivo T cell function and we hypothesized that exposure to an agonist targeting the metabolic regulator AMP-activated protein kinase (AMPK), would create CARTs capable of both efficient leukemia clearance and increased in vivo persistence. METHODSCART cells were generated from healthy human via lentiviral transduction. Following activation, cells were exposed to either Compound 991 or DMSO for 96 hours, followed by a 48-hour washout. During and after agonist treatment, T cells were harvested for metabolic and functional assessments. To test in vivo efficacy, immunodeficient mice were injected with luciferase+ NALM6 leukemia cells, followed one week later by either 991- or DMSO-expanded CARTs. Leukemia burden and anti-leukemia efficacy was assessed via radiance imaging and overall survival. RESULTSHuman T cells expanded in Compound 991 activated AMPK without limiting cellular expansion and gained both mitochondrial density and improved handling of reactive oxygen species (ROS). Importantly, receipt of 991-exposed CARTs significantly improved in vivo leukemia clearance, prolonged recipient survival, and increased CD4+ T cell yields at early times post-injection. Ex vivo, 991 agonist treatment mimicked nutrient starvation, increased autophagic flux, and promoted generation of mitochondrially-protective metabolites. DISCUSSIONEx vivo expansion processes are necessary to generate sufficient cell numbers, but often promote sustained activation and differentiation, negatively impacting in vivo persistence and function. Here, we demonstrate that promoting AMPK activity during CART expansion metabolically reprograms cells without limiting T cell yield, enhances in vivo anti-leukemia efficacy, and improves CD4+ in vivo persistence. Importantly, AMPK agonism achieves these results without further modifying the expansion media, changing the CART construct, or genetically altering the cells. Altogether, these data highlight AMPK agonism as a potent and readily translatable approach to improve the metabolic profile and overall efficacy of cancer-targeting T cells.