The tumor doesn't stay put. It reaches outward, eroding bone.
For decades, glioblastoma has been treated as a local catastrophe — a fire contained within the skull. Researchers at Montefiore Einstein and Albert Einstein College of Medicine have now shown that the tumor is something closer to a systemic architect, eroding bone, widening microscopic channels between skull and brain, and rewriting the immune landscape in ways that may explain why 15,000 Americans diagnosed each year so rarely survive beyond fifteen months. The disease, it turns out, does not respect the boundaries medicine drew around it.
- Glioblastoma actively eats away at skull bone along the seams where plates fuse, a pattern confirmed in both mouse models and human CT scans — and unique to this cancer among all brain injuries studied.
- As the skull erodes, the channels linking skull marrow to the brain multiply and widen, giving the tumor a highway to flood the immune system with its own instructions.
- The cancer nearly doubles inflammatory neutrophils while wiping out protective B cells in the skull marrow, converting the body's own immune architecture into a tool of its own aggression.
- Anti-osteoporosis drugs stopped the bone erosion but paradoxically accelerated tumor growth and blocked immunotherapy — a warning that intervening in one part of the system can destabilize another.
- Researchers now argue the only coherent path forward is to treat glioblastoma as a systemic disease, targeting the immune imbalance in the skull marrow itself rather than the tumor alone.
Glioblastoma kills faster than almost any cancer and resists treatment with grim consistency. About 15,000 people receive the diagnosis each year in the United States, and even those who undergo surgery, chemotherapy, and radiation live a median of just fifteen months. The disease has long been understood as a problem contained within the skull — but researchers at Montefiore Einstein Comprehensive Cancer Center and Albert Einstein College of Medicine have found evidence that fundamentally challenges that assumption.
The discovery began with a trail left by recent neuroscience: microscopic channels, thin enough to have gone unnoticed for years, that connect the skull's marrow directly to the brain beneath it. Molecules and cells move in both directions through these passages. Jinan Behnan and colleagues asked what happened to that connection when a glioblastoma took root nearby.
Using advanced imaging in mouse models, the team watched the tumors erode the skull bone itself — particularly along the seams where skull plates fuse. The effect was specific to glioblastoma and other malignant brain tumors. Strokes didn't cause it. Other brain injuries didn't. Even cancers that spread throughout the body didn't produce it. Human CT scans of glioblastoma patients showed the same pattern of bone thinning in the same locations.
As the bone eroded, the skull-to-brain channels grew wider and more numerous. Using single-cell RNA sequencing, the researchers confirmed what they suspected: the tumor was sending signals through those enlarged passages into the skull's marrow, rewriting the immune cells that live there. Inflammatory neutrophils nearly doubled. Protective B cells nearly vanished. The skull marrow had been converted into an environment that served the tumor's growth.
The effect was local to the skull in a specific way — marrow from the femur showed a different pattern entirely, with immune cell production suppressed rather than inflamed. Glioblastoma, the data suggested, was not simply a brain disease. It was systemic, reshaping the body's immune architecture differently depending on location.
Then came a troubling turn. Two FDA-approved anti-osteoporosis drugs successfully halted the skull erosion — but one accelerated tumor growth, and both blocked the effectiveness of anti-PD-L1 immunotherapy. Saving the bone made the cancer worse.
The research, published in Nature Neuroscience, points toward a different treatment philosophy: rather than targeting the tumor alone, restore the immune balance within the skull marrow — suppressing the inflammatory cells the tumor has amplified while rebuilding the protective populations it has erased. It is a shift from understanding glioblastoma as a local fire to recognizing it as a disease that rewrites the body's defenses from the inside out.
Glioblastoma kills faster than most cancers and resists treatment with stubborn efficiency. About 15,000 people receive this diagnosis each year in the United States. Those who undergo surgery, chemotherapy, and radiation—the standard arsenal—live a median of 15 months. The disease has long been understood as a problem contained within the skull. But researchers at Montefiore Einstein Comprehensive Cancer Center and Albert Einstein College of Medicine have discovered something that upends that assumption: the tumor doesn't stay put. It reaches outward, eroding bone, remaking the immune landscape in the skull itself, and in doing so, it may explain why so many treatments fail.
The finding emerged from work by Jinan Behnan and colleagues who were following a trail of recent neuroscience. Scientists had identified microscopic channels—thin enough that they had gone unnoticed for years—that connect the skull's marrow directly to the brain beneath it. These aren't one-way passages. Molecules and cells move both directions. Behnan's team wondered what happened when a glioblastoma took root on one side of that connection.
Using advanced imaging on mice with two different types of glioblastoma, the researchers watched the tumors do something unexpected: they ate away at the skull bone itself, particularly along the seams where the skull plates fuse together. The erosion was specific to glioblastoma and other malignant brain tumors. Strokes didn't cause it. Other types of brain injury didn't. Even systemic cancers—the kind that spread throughout the body—didn't produce this effect. When the team examined CT scans of glioblastoma patients, they found the same pattern of bone thinning in the same anatomical locations.
As the skull eroded, something else happened: the channels connecting skull to brain grew wider and more numerous. The researchers suspected these enlarged passages might become a highway for the tumor to send signals into the skull's marrow, reshaping the immune cells that live there. Using single-cell RNA sequencing, they confirmed it. The glioblastoma had tilted the immune balance dramatically toward inflammation. Neutrophils—inflammatory cells that normally help fight infection—nearly doubled in number. Meanwhile, B cells that produce antibodies and other protective B cells nearly vanished. The skull marrow had been converted into an environment that fed the tumor's aggression.
The effect was distinctly local to the skull. When the researchers examined marrow from the femur—the thighbone—they found a different pattern entirely. The cancer suppressed genes needed to produce immune cells there, rather than activating them. This suggested glioblastoma wasn't simply a brain disease at all. It was systemic, reaching into the body's immune architecture and rewriting it in ways that varied by location.
Then came a troubling discovery. The team tested whether anti-osteoporosis drugs—medications already approved by the FDA to prevent bone loss—might halt the skull erosion. Two drugs, zoledronic acid and denosumab, did stop the bone from eroding. But zoledronic acid also accelerated tumor growth in one type of glioblastoma. Both drugs blocked the benefits of anti-PD-L1, an immunotherapy designed to boost tumor-fighting T cells. In trying to save the bone, the researchers had made the cancer worse.
The implications are stark. Current treatments for glioblastoma treat it as a localized problem—remove the tumor, poison the remaining cells, bombard them with radiation. But if the disease is remaking the immune system across the skull and beyond, those approaches will always be fighting half the battle. Behnan and colleagues suggest a different strategy: restore the normal balance of immune cells in the skull marrow itself. Suppress the inflammatory neutrophils and monocytes while rebuilding the T and B cell populations that the tumor has erased. It's a shift from treating glioblastoma as a brain problem to treating it as a systemic one. The research, published in Nature Neuroscience, offers a map for where to look next.
Notable Quotes
This notoriously hard-to-treat brain cancer interacts with the body's immune system, which may explain why current therapies treating glioblastoma as a local disease have failed.— Jinan Behnan, Ph.D., assistant professor at Albert Einstein College of Medicine
The skull-to-brain channels allow an influx of pro-inflammatory cells from the skull marrow to the tumor, rendering the glioblastoma increasingly aggressive and untreatable.— E. Richard Stanley, Ph.D., professor of developmental and molecular biology at Einstein
The Hearth Conversation Another angle on the story
Why does glioblastoma specifically erode the skull when other brain cancers don't?
That's the question the researchers are still working through. The erosion seems tied to how aggressively glioblastoma signals through those skull-brain channels. It's not just that the tumor is there—it's actively reaching out and reshaping the bone around it.
So the tumor is essentially hacking the immune system through the marrow?
In a way, yes. It's sending signals that flip the marrow's immune cells from a protective state to an inflammatory one. The tumor thrives in inflammation. It's like the cancer has found a way to recruit the body's own defenses to work against itself.
That explains why the anti-osteoporosis drugs backfired. They stopped the erosion but didn't address the immune hijacking.
Exactly. You can shore up the bone, but if the marrow is still pumping out inflammatory cells that feed the tumor, you've only solved half the problem. You might have even made it worse by blocking the channels that could have let immune cells reach the tumor.
Is there a way to reverse the immune shift once it's happened?
That's what they're exploring now. The idea is to suppress the inflammatory cells the tumor is promoting while simultaneously restoring the B and T cells it's eliminated. It's not about fighting the tumor directly—it's about restoring the immune landscape so the body can fight it naturally.
Why has this systemic aspect been missed until now?
Because we've been looking at glioblastoma as a brain problem. The skull-brain channels are only recently understood. Once you see that connection, everything changes. The tumor isn't contained. It's reaching into the body's immune system and rewriting the rules.