Back

IL-18 metabolically reprograms CAR-expressing natural killer T cells and enhances their antitumor activity against neuroblastoma

Barragan, G. A.; de la Cerda, D. A.; Landoni, E.; Dholakia, K.; Humeniuk, P.; Caraballo, L. D.; Wang, Y.; Tian, G.; Yang, B.; Guo, L.; Wood, M.; Rios, X.; Xu, X.; Courtney, A. N.; Dotti, G.; Metelitsa, L. S.

2025-04-24 immunology
10.1101/2025.04.18.649551 bioRxiv
Show abstract

Invariant natural killer T cells (NKTs) have intrinsic anti-tumor properties that make them promising candidates for chimeric antigen receptor (CAR)-based immunotherapies. Transgenic cytokine expression has been shown to enhance the potency of cellular immunotherapies, and we hypothesized that co-expressing IL-18 alone or with IL-15 would boost CAR-NKT therapeutic potential. To test this hypothesis, we generated retroviral constructs expressing IL-15 and/or IL-18 with the inducible caspase 9 (iC9) safety switch and co-transduced them with a GD2-specific CAR into human NKTs. Co-expression of IL-18 or IL-15/IL-18 increased GD2.CAR-NKT in vitro cytotoxicity, proliferation, and cytokine secretion compared to IL-15 alone. In a metastatic neuroblastoma model, GD2.CAR-NKTs expressing constructs with IL-18 controlled tumor growth better than cells expressing IL-15 only, but mice in the IL-15/IL-18 group developed severe toxicities not observed in the IL-18-only group. Mechanistically, we found that IL-18 drives a distinct transcriptional profile from IL-15 in CAR-NKTs marked by lower expression of exhaustion gene signatures and enrichment of metabolism-related processes. Finally, targeted metabolomics revealed that IL-18 induces broad metabolic reprogramming in CAR-NKTs including enhancement of oxidative phosphorylation, glycolysis, glutaminolysis and purine metabolism. These results support the use of IL-18 in developing the next generation of cytokine-armed CAR-NKT cancer immunotherapy.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.