Abstract
The efficacy of CAR therapy is limited by T-cell exhaustion, which is mitigated by 4-1BB costimulation.
Major finding: The efficacy of CAR therapy is limited by T-cell exhaustion, which is mitigated by 4-1BB costimulation.
Mechanism: Physical interactions between clustered CARs induce tonic signaling that leads to T-cell exhaustion.
Impact: These data may explain the differential persistence of CD19 CAR T cells with 4-1BB versus those with CD28.
Chimeric antigen receptors (CAR) link antigen-binding domains with T-cell signaling domains and represent potential anticancer agents. However, only CARs targeting CD19 in hematologic malignancies have demonstrated significant clinical efficacy. To identify the mechanisms underlying the poor efficacy of CARs targeting other tumor antigens, Long and colleagues compared the in vivo antitumor activity of CD19 CAR T cells harboring the CD28 and CD3ζ signaling domains (CD19.28z) with that of GD2.28z CAR T cells designed to target GD2-expressing solid tumors. In contrast to CD19.28z CAR T cells, GD2.28z CAR T cells exhibited decreased persistence and failed to inhibit tumor growth in mice. In ex vivo expansion studies, GD2.28z CAR T cells showed indications of exhaustion, including increased rates of apoptosis, higher expression of inhibitory receptors, and reduced cytokine production. Unlike CD19.28z CARs, GD2.28z CARs exhibited basal levels of CD3ζ phosphorylation, indicative of tonic signaling during ex vivo expansion. Tonic GD2.28z CAR signaling induced early T-cell exhaustion independent of antigen and was mediated by physical interactions between single-chain variable fragments (scFv) that triggered CAR clustering at the cell surface. Ex vivo tonic activation and T-cell exhaustion occurred to varying degrees in all scFv-based CARs tested except the CD19 CAR. Furthermore, GD2 CARs incorporating the 4-1BB costimulatory endodomain (GD2.BBz) rather than CD28 showed lower levels of exhaustion markers, increased persistence, and greater antitumor efficacy in vivo. Expression of the GD2.BBz CAR resulted in reduced expression of exhaustion-associated genes as well as altered expression of genes involved in hypoxia, metabolism, and apoptosis compared with T cells expressing the GD2.28z CAR, suggesting that 4-1BB may mitigate exhaustion via multiple pathways. These studies provide a better understanding of how CD19 CARs achieve antitumor efficacy and may explain the increased persistence of CD19 CAR T cells that incorporate 4-1BB in clinical trials compared with those harboring CD28.
RSS Feeds