Paper Details

ABSTRACT

Introduction: BCMA CAR-T therapy produces high response rates in relapsed/refractory multiple myeloma (MM). However, durable disease control remains limited, as most patients ultimately relapse, partly due to insufficient CAR-T cell persistence. Major drivers of CAR-T dysfunction include chronic antigen exposure and the suppressive tumor microenvironment, which together push CAR-T cells toward terminal differentiation and exhaustion. Because CAR-T differentiation and metabolic states are shaped during ex vivo culture, we sought manufacturing conditions that favor a persistent, exhaustion-resistant phenotype. We compared two cytokine conditions (IL-2 vs IL-7/IL-15) and screened three small molecules, each acting through a distinct mechanism: vitamin C (VC; epigenetic modifier promoting stem-like/memory programs), creatine (Cr; phosphocreatine energy buffer supporting CD8 effector function), and nicotinamide (NAM; NAD-pathway modulator sustaining mitochondrial fitness and lowering inhibitory-receptor expression). We evaluated how these conditions affect memory phenotype, exhaustion, and metabolic resilience under repeated stimulation.

Methods: Human T cells engineered to express a BCMA CAR were expanded in IL-2 or IL-7/IL-15, each with vehicle control (PBS), VC, Cr, NAM, VC plus Cr, or VC plus NAM. To model chronic activation stress, cells were restimulated every other day and compared with cells under standard culture. Viability was tracked from days 0 to 11; memory differentiation, including stem-cell memory, central memory, effector memory, and terminal-effector subsets, CD4/CD8 composition, and PD-1 expression were quantified by flow cytometry on CAR⁺ cells; and cellular bioenergetics, including basal and maximal respiration, ATP-linked respiration, glycolysis, and glycolytic capacity, were measured by Seahorse extracellular flux analysis. To assess antitumor function, conditioned cells from each arm were co-cultured with BCMA⁺ MM H929 cells at a 1:1 effector-to-target ratio. Cytotoxicity and H929-induced secretion of IFN-γ, TNF-α, and granzyme B were measured at 24 and 48 hours in both cytokine backgrounds, with target-free cultures as the reference.

Results: Among the tested conditions, VC+NAM most consistently protected BCMA CAR-T cells from repeated-stimulation stress. Under restimulation, VC+NAM maintained day-11 viability at 85% in IL-2 and 81% in IL-7/IL-15, versus 45% and 54% in PBS controls. Without restimulation, viability remained high at 93% and 94%, respectively. Even after repeated stimulation, VC+NAM-conditioned cells retained an early-memory profile, with stem-cell memory plus central-memory subsets comprising 96.6% and 88.6% under IL-2 and IL-7/IL-15, significantly higher than in PBS controls. VC+NAM also maintained a predominantly CD8 product (approximately 76% under IL-2 and 69% under IL-7/IL-15), while limiting PD-1 expression to 0.4%, significantly lower than in controls. VC+NAM attenuated activation-associated bioenergetic activity in both cytokine backgrounds, reducing basal and maximal respiration, ATP-linked respiration, glycolysis, and glycolytic capacity relative to controls. Importantly, this restrained metabolic state did not compromise antitumor function. At 48 hours, VC+NAM-conditioned cells achieved cytotoxicity of approximately 45% in IL-2 and 31% in IL-7/IL-15, significantly higher than in controls. VC+NAM-conditioned cells also retained H929-induced production of IFN-γ (26 ng/mL), TNF-α (0.5 ng/mL), and granzyme B (10 ng/mL), demonstrating preserved antimyeloma effector function.