Back

Neuropilin-1 expression by solid tumors impairs CAR T cell function

Condon, L. E.; Bouillie, M.; Fuentealba, J.; Saitakis, M.; Hermine, O.

2025-09-18 immunology
10.1101/2025.09.16.676545 bioRxiv
Show abstract

Neuropilin-1 (NRP-1) is a versatile transmembrane protein expressed in numerous cell types and tissues, both in health and in disease. In particular, it is expressed by activated T cells, where it has been shown to play an inhibitory role, dampening their response to tumor cells. Chimeric antigen receptor (CAR) T cells have had considerable success in treating cancers such as B cell leukemias, but face several obstacles in the treatment of solid tumors. We hypothesized that NRP-1 may contribute to dampening CAR T cell responses to cancer. While NRP-1 blockade either through neutralizing antibodies or through CRISPR/Cas9-mediated deletion did not improve CAR T cell cytotoxic activity in an in vitro solid tumor model, NRP-1 was found to have a protective effect against CAR T cells when expressed by the tumor cells, in line with previous literature implicating NRP-1 as a pro-tumor factor involved in their growth and invasiveness. We found that NRP-1 was transferred from target cells to CAR T cells via trogocytosis, i.e. the "nibbling" and subsequent display of membrane proteins from one cell by another. CD19, the target antigen in this model, was also transferred to CAR T cells, raising interesting questions about trogocytosis as a marker for effective tumor cell killing through direct interaction. On the contrary, trogocytosis may impede effective CAR T function by promoting fratricide due to their display of target antigen. While no conclusive mechanisms for NRP-1 activity were found, several avenues of research have been opened up to further our understanding of the multifaceted roles of NRP-1 in CAR T cell activity and interactions with their targets.

Matching journals

The top 6 journals account 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.