A Pharmacologic Model Predicts that Tumor Debulking Improves CAR T-cell Efficacy in Large B-cell Lymphoma.

Chimeric antigen receptor (CAR) T cells produce durable remissions in some patients with large B-cell lymphoma, but outcomes are poor in patients with large tumor burdens or limited CAR T-cell expansion. To understand these relationships and explore potential interventions, we applied established population pharmacokinetic/pharmacodynamic principles to model kinetics of axicabtagene ciloleucel (axi-cel) concentrations and tumor responses to axi-cel, and validated model outputs using independent cohorts. This mechanistic model reproduces and explains poor outcomes associated with high tumor burden and low CAR T-cell expansion, finding that proliferation of large lymphoma populations can outpace the cytotoxic effect of CAR T cells. A high ratio of lymphoma cells to CAR T cells is effectively a mechanism of CAR T-cell resistance, which could be modified by tumor debulking before infusion. This model predicts that reducing tumor burden before CAR T-cell infusion may improve durable remission rate. Future clinical studies optimizing bridging therapy may therefore enhance the success of CAR T-cell therapies. SIGNIFICANCE: A population pharmacokinetic/pharmacodynamic model of axi-cel in large B-cell lymphoma explains the observation that high tumor burden and low CAR T-cell expansion predict poor outcomes. This model suggests tumor debulking before CAR T infusion or deploying CAR T therapy in measurable residual disease-positive patients after first-line treatment could improve CAR T success rates. See related commentary by Altrock, p. 11.