Skepticism is fair here - when a paper claims it turned neutrophils into cancer-fighting CAR cells inside the body, your first reaction should be something like, "Sure, and my Wi-Fi only fails when I need it most." But this one is interesting for a real reason: it takes one of the immune system's most abundant, underappreciated cells and gives it a programmable anti-tumor job description.
Glioblastoma is the heavyweight troublemaker of brain cancer - aggressive, sneaky, and very good at turning its neighborhood into an immune-suppressive swamp. T cells often struggle to get in, stay functional, or keep recognizing the target. Meanwhile, neutrophils are everywhere in the blood and are perfectly built for rapid deployment. The catch is that they are short-lived, hard to genetically engineer, and in tumors they can act less like heroes and more like gullible security guards taking bribes from the villain.
That is the setup for this new paper by Chang and colleagues: instead of extracting cells, engineering them outside the body, and reinfusing them, they tried to create CAR-neutrophils in vivo using a modified RNA system called NeuSMRT (Chang et al., 2026).
A molecular group chat with very strict membership rules
The clever bit is not just "put CAR in neutrophil." The clever bit is how they restricted that message to neutrophils.
NeuSMRT uses engineered extracellular vesicles or lipid nanoparticles to deliver modified RNA, plus a microRNA-responsive switch that allows translation mainly in neutrophils. In plain English: the researchers built a package carrying instructions, then added a molecular bouncer at the door so the protein gets made in the right cell type. Gene regulation really is just biology inventing absurdly overcomplicated app permissions.
That matters because neutrophils are abundant and mobile, but they are also notoriously difficult to program directly. Earlier work from the same broader research track showed that CAR-neutrophils could be made from human pluripotent stem cells and could attack tumors, including glioblastoma models (Chang et al., 2022; Chang et al., 2023). This new paper skips the factory step and tries to do the engineering on site.
Why neutrophils are such weirdly good candidates
If cancer biology were a screenplay, neutrophils would be the side character everyone underestimated until season three. They are first responders from the innate immune system, and they naturally traffic into inflamed tissue. That makes them appealing for brain tumors, where delivery is half the battle and the blood-brain barrier acts like an extremely rude nightclub doorman.
The problem is that gliomas often reprogram neutrophils into tumor helpers. Prior work showed that glioma can skew neutrophils toward pro-tumor behavior and support angiogenesis and immune suppression (Magod et al., 2021). More recent reviews make the same point more broadly: neutrophils in cancer are plastic, complicated, and absolutely not a one-note cell type (Li et al., 2025; Li et al., 2025).
So the game here is not just "add weapon." It is "stop the cell from joining the wrong team, then hand it a better one."
What the paper actually found
In syngeneic glioma mice, the in vivo-generated CAR-neutrophils reduced tumor growth and extended survival. The effect did not seem limited to direct killing. The tumor microenvironment also shifted: more T cell recruitment and activation, less myeloid immunosuppression. That is a big deal because solid tumors, especially glioblastoma, often win by making the whole neighborhood hostile to immune attack.
The therapy also boosted the effects of chemotherapy and CAR-T therapy in preclinical models, which is exactly the sort of combination angle worth paying attention to. Solid tumors are rarely defeated by one magical cell type riding in on a white horse. Usually it is more like assembling an Avengers team where half the members are tired, one cannot cross the blood-brain barrier, and another has forgotten what antigen it is supposed to recognize.
The authors also reported anti-tumor activity in a humanized glioblastoma mouse model and feasibility and safety data in experimental dogs. That last part does not mean dog trials equal human success, obviously. But it does make this feel less like a beautiful cell-culture stunt and more like an early translational platform.
The part where we keep our lab goggles on
Before anyone starts printing "neutrophils save the day" T-shirts, this is still preclinical work. Glioblastoma has a long history of humbling elegant ideas. CAR-T therapy in recurrent high-grade glioma has shown feasibility and safety in humans, but efficacy remains uneven and the tumor microenvironment remains a bully with a chemistry degree (Brown et al., 2024, PMCID: PMC11031404).
Still, the concept is strong. Instead of treating neutrophils as collateral damage, this paper treats them as programmable infrastructure. From a genomics angle, that is the fun part: cancer is full of bad instructions, bad signaling, bad context. NeuSMRT is basically an attempt to slip a corrected line of code into the immune system and hope the runtime environment does not immediately catch fire.
If the approach proves reproducible and scalable, it could open a new lane in solid-tumor immunotherapy: not just engineering T cells, but engineering the myeloid ecosystem that decides whether T cells have any chance at all.
References
- Chang Y, Shao K, Li H, et al. CAR-neutrophils produced in vivo to treat glioma. Nature Biomedical Engineering. 2026. DOI: https://doi.org/10.1038/s41551-026-01656-0
- Chang Y, et al. Engineering chimeric antigen receptor neutrophils from human pluripotent stem cells for targeted cancer immunotherapy. Cell Reports. 2022;40(5):111128. DOI: https://doi.org/10.1016/j.celrep.2022.111128
- Chang Y, et al. CAR-neutrophil mediated delivery of tumor-microenvironment responsive nanodrugs for glioblastoma chemo-immunotherapy. Nature Communications. 2023;14:2266. DOI: https://doi.org/10.1038/s41467-023-37872-4
- Brown CE, et al. Locoregional delivery of IL-13Ralpha2-targeting CAR-T cells in recurrent high-grade glioma: a phase 1 trial. Nature Medicine. 2024;30:1001-1012. DOI: https://doi.org/10.1038/s41591-024-02875-1. PMCID: https://pmc.ncbi.nlm.nih.gov/articles/PMC11031404/
- Magod P, et al. Exploring the longitudinal glioma microenvironment landscape uncovers reprogrammed pro-tumorigenic neutrophils in the bone marrow. Cell Reports. 2021;36(3):109480. DOI: https://doi.org/10.1016/j.celrep.2021.109480
- Li Y-R, Zhu Y, Halladay T, Yang L. In vivo CAR engineering for immunotherapy. Nature Reviews Immunology. 2025;25:725-744. DOI: https://doi.org/10.1038/s41577-025-01174-1
- Li X, et al. Targeting neutrophils for cancer therapy. Nature Reviews Drug Discovery. 2025;24:666-684. DOI: https://doi.org/10.1038/s41573-025-01210-8
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.