Glioblastoma kills within 12 to 18 months. Now researchers at McMaster and SickKids say an old HIV drug could interrupt the cellular conversation that keeps the tumor alive.
image from Gemini Imagen 4
Researchers at McMaster University and the Hospital for Sick Children discovered that glioblastoma tumors recruit oligodendrocytes—brain cells that normally insulate nerve fibers—to support tumor growth through CCR5 receptor signaling. The HIV drug Maraviroc, already FDA-approved in 2007, was shown to interrupt this cross-cell communication in lab models, significantly slowing tumor growth and suggesting a potential repurposing path that could bypass the typical drug development timeline. This represents a new vulnerability in glioblastoma, an aggressive brain cancer with a median survival of 12-18 months and virtually unchanged standard of care for two decades.
An HIV drug and a brain tumor: the repurpose angle that glioblastoma researchers have been waiting for
Glioblastoma kills most patients within 12 to 18 months of diagnosis. The standard of care has barely changed in two decades. Now a team at McMaster University and the Hospital for Sick Children in Toronto has found something different: a reason the tumor keeps winning, and a drug that might already be in the pharmacy.
The study, published in Neuron (DOI: 10.1016/j.neuron.2025.09.003), describes a previously unrecognized collaboration between glioblastoma cells and oligodendrocytes — brain cells whose day job is insulating nerve fibers and keeping neural signaling running smoothly. The researchers found that oligodendrocytes can be recruited by the tumor and repurposed to support its growth. When they blocked this cross-cell signaling in lab models, tumor growth slowed significantly.
The mechanism centers on a receptor called CCR5. That's the same receptor HIV uses to enter immune cells, and it's already targeted by an HIV drug called Maraviroc (Selzentry, Pfizer, 2007) — approved by the FDA in August 2007. The team showed that Maraviroc can interrupt the oligodendrocyte-cancer cell chatter in their models — suggesting a potential repurposing path that bypasses the usual decade-plus drug development timeline.
"Glioblastoma isn't just a mass of cancer cells, it's an ecosystem," said Sheila Singh, co-senior author and professor of surgery at McMaster. "By decoding how these cells talk to each other, we've found a vulnerability that could be targeted with a drug that's already on the market."
Singh and Jason Moffat, co-senior author and head of the Genetics and Genome Biology program at SickKids, have been building toward this. Their 2024 study in Nature Medicine showed that a migration path used by cells during brain development can be hijacked for cancer cell invasion. The new paper adds a layer: the tumor doesn't work alone. It enlists support from ordinary brain cells that have no business helping it.
The CCR5 angle connects this work to a small but persistent thread in oncology. CytoDyn presented data at the AACR Brain Cancer conference showing that CCR5 expression is significantly elevated in primary glioblastoma tumors compared to normal brain tissue, and that higher CCR5 correlates with poor prognosis. Their CCR5 antibody leronlimab showed synergy with standard chemotherapy in preclinical models. That work is separate from the Singh/Moffat findings, but it adds context: the CCR5-glioblastoma connection isn't invented here.
The key caveat is the one that applies to most lab-model oncology findings: mice are not people, and dishes are not tumors. Blocking a signaling pathway in a controlled model is a meaningful step toward understanding the disease, but the distance to a clinical trial is measured in years and failures. The research team is pointing at a target, not a cure.
What's worth sitting with is the framing. For decades, glioblastoma research focused on the cancer cell itself. The emerging picture — including this new paper — treats the tumor as a tissue ecosystem problem. Stop listening to the cancer cell and start listening to everything around it. That's a useful reorientation for a field that has run out of obvious moves.
The paper's co-first authors are Kui Zhai of the Singh Lab at McMaster and Nick Mikolajewicz, formerly a postdoctoral fellow in the Moffat Lab at SickKids.
Story entered the newsroom
Research completed — 4 sources registered. McMaster/SickKids Neuron paper (2026): oligodendrocytes (normally myelinate neurons) are co-opted by glioblastoma to support tumor growth via CCR5 sig
Draft (569 words)
Reporter revised draft (569 words)
Reporter revised draft (541 words)
Reporter revised draft (547 words)
Reporter revised draft based on fact-check feedback (547 words)
Approved for publication
Published (545 words)
Get the best frontier systems analysis delivered weekly. No spam, no fluff.
Biotech & Life Sciences · 23h 39m ago · 4 min read
Biotech & Life Sciences · 1d ago · 4 min read
