Neoadjuvant triplet immune checkpoint blockade in newly diagnosed glioblastoma
Nature Medicine. 2025;31(5):1557-1566. DOI 10.1038/s41591-025-03512-1. PMID 40016450. PMCID PMC12092302.
How to cite
AMA
Long GV, Shklovskaya E, Satgunaseelan L, Mao Y, da Silva IP, Perry KA, et al. Neoadjuvant triplet immune checkpoint blockade in newly diagnosed glioblastoma. Nat Med. 2025;31(5):1557-1566. doi:10.1038/s41591-025-03512-1APA
Long, G. V., Shklovskaya, E., Satgunaseelan, L., Mao, Y., da Silva, I. P., Perry, K. A., et al. (2025). Neoadjuvant triplet immune checkpoint blockade in newly diagnosed glioblastoma. Nature Medicine, 31(5), 1557-1566. https://doi.org/10.1038/s41591-025-03512-1BibTeX
@article{long2025neoadjuvant,
title = {Neoadjuvant triplet immune checkpoint blockade in newly diagnosed glioblastoma},
author = {Long, Georgina V. and Shklovskaya, Elena and Satgunaseelan, Laveniya and Mao, Yizhe and da Silva, Inês Pires and Perry, Kristen A. and others},
journal = {Nature Medicine},
volume = {31},
number = {5},
pages = {1557--1566},
year = {2025},
doi = {10.1038/s41591-025-03512-1},
pmid = {40016450}
}Glioblastoma is an aggressive brain tumor that almost always returns within two years, and the immune checkpoint drugs that reshaped other cancers have shown little benefit when given after the tumor is removed.
This report follows one newly diagnosed patient given a single dose of three checkpoint inhibitors together, before surgery rather than after, to test whether treating the tumor while still in place could provoke a stronger immune response. When the tumor was resected twelve days later, it carried heavy immune-cell infiltration, and circulating tumor cells in the blood fell to zero over the following months.
Seventeen months on, there was no definitive sign the cancer had come back. The authors offer the case as the rationale for a planned first-line trial, not as proof that the approach works.
Key findings
- A single dose of neoadjuvant triplet checkpoint blockade preceded marked tumor-infiltrating-lymphocyte infiltration and activation. One dose of nivolumab (480 mg), relatlimab (160 mg) and ipilimumab (80 mg) was given on day +1, and the tumor resected 12 days later showed the anti-PD-1 agent bound to tumor-infiltrating lymphocytes with marked infiltration and activation over the day −4 baseline biopsy.
- Circulating tumor cells zero-converted over the course of treatment. Using the AccuCyte–CyteFinder workflow, 16 glioblastoma circulating tumor cells were counted in 7.5 ml of blood before treatment (day −9), and none were detected after therapy (day +190).
- There was no definitive evidence of recurrence at 17 months. The most recent imaging, at day +526, showed grossly stable postsurgical changes and no features to suggest recurrent disease, against a median overall survival of 14.1 months reported for MGMT-unmethylated glioblastoma on standard care.
The AccuCyte–CyteFinder workflow in the methods
“A total of 7.5 ml whole-blood samples were processed using the AccuCyte-CyteFinder platform (RareCyte)”
— Long et al., Nature Medicine (2025), Methods, “CTC analysis”
Disclosure: RareCyte is named in the competing-interests statement of the publication cited above.
Why it matters for The AccuCyte–CyteFinder workflow users
If you track circulating tumor cells across a treatment course, look at what this case asked of the counting step. Glioblastoma tumor cells are rare in blood and lack the epithelial markers most enrichment methods select on, so here they were called on glial identity instead — positive for the nuclear dye DAPI, the astrocytic marker GFAP and EGFR, and negative for the leukocyte markers CD45 and CD66b. That readout only holds if the candidate cells are still on the slide to be scored. The AccuCyte–CyteFinder workflow isolates every nucleated cell from 7.5 ml of whole blood by density rather than selecting for a surface antigen first, then images the deposited population on the CyteFinder HT. That is what let this study count 16 tumor cells before treatment and confirm none afterward in the same patient. Treat the number as a research enumeration, not a clinical test — what the workflow supplies is a reliable count of rare cells you would otherwise lose to up-front selection. When the endpoint is a serial cell count, keeping every candidate in view is the property that carries it.
