Effects of Bacterial Metabolites on CAR-T Cell Therapy

Introduction

Cancer immunotherapy by chimeric antigen receptor modified T cells (CAR-T cells) has shown promising clinical efficacy in hematological malignancies. Nevertheless, the success of the therapy is limited, and patients suffer from relapse. Bacterial metabolites were described as important host intrinsic regulators of innate immune pathways like cGAS/STING and RIG-I/MAVS. However, their role on CAR-T cell therapies and mechanism of action is currently unknown. We investigated the CAR-T cell intrinsic and extrinsic role of intestinal derived interferon-linked bacterial metabolites on cytotoxicity and exhaustion.

Methods

Based on previously generated data we chose two bacterial derived metabolites showing beneficial effects on survival of patients after allogeneic stem cell transplantation. Anti-CD19-CD28z CAR-T cells were treated with metabolites for 72 hours and cytokine production as well as expression of extracellular activation markers were assessed by flow cytometry. To determine the killing capacity of metabolite treated CAR-T cells on CD19+ Nalm6 ALL blasts, in vitro bioluminescence-based cytotoxicity assays at a 1:8 E:T ratio over 24 hours were performed. Expression of extracellular and transcriptional exhaustion markers were analyzed by flow cytometry after anti-CD3/CD28 stimulation as well as after repetitive challenge with tumor cells.

Results

Our data suggest an immunosuppressive effect of one of the metabolites via direct modulation of CAR-T cells as well as innate immune cells. We could show that treatment with metabolite 1 leads to CAR-T cell activation, activation induced cell death (AICD) and increased expression of exhaustion markers (PD-1, LAG-3 and TIM3) in antiCD19-CD28z CAR-T cells, leading to impaired cancer cell cytotoxicity in vitro in a concentration dependent fashion.

To better investigate the complex interaction between innate immune cells and CAR-T cells we further set up an in vitro coculture assay of CAR-T cells and macrophages. Our data suggest effects of these metabolites on the polarization of macrophages modulating the CAR-T cell function and immune response. Additionally, we performed polyclonal knockouts in CAR-T cells of type I interferon related genes (Stimulator of interferon genes (STING), Retinoic acid-inducible gene I (RIG-I), Interferon-α/β receptor(IFNAR1), C/EBP Homologous Protein (CHOP)) and the metabolite sensor Aryl-hydrocarbon-receptor (AHR) in order to narrow down the mechanism of action of these metabolites on the T cell level, which revealed putative involvement of the AHR signaling pathway, but mostly independence from type I IFN pathways. Further, we could show distinct, metabolite specific cytokine secretion of monocyte derived macrophages and CAR-T cells after stimulation with the two metabolites.

Conclusions

Our data suggest a direct, CAR-T cell intrinsic and innate immune cell mediated immunosuppressive effect of metabolite 1. These findings will now be validated in vivo and the mechanism of action besides AHR-agonism will be further elucidated utilizing single cell RNAseq.

No relevant conflicts of interest to declare.