Scientists achieve 90% accuracy with breakthrough blood test for brain tumours

Glioblastoma patients currently face delayed diagnosis and invasive procedures; earlier detection through blood test could improve treatment outcomes and reduce brain damage from late-stage tumours.
Follow the tumour's behaviour through a simple blood sample
Professor Hamerlik describes the test's potential to track cancer progression in real time without invasive procedures.

For generations, the shadow of a brain tumour has fallen slowly — announced only after repeated visits, delayed scans, and the quiet accumulation of lost time. Researchers at the University of Manchester, driven in part by personal grief and scientific conviction, have developed a blood test capable of detecting aggressive glioblastomas with over 90% accuracy, potentially compressing a months-long diagnostic odyssey into a single clinical moment. The test does not merely identify disease at the outset; it watches how that disease responds to treatment, offering doctors a living window into the tumour's behaviour. If clinical trials succeed and regulatory approval follows, what was once an ordeal of invasive procedures may one day begin with something as unremarkable as a routine blood draw.

  • Glioblastoma patients currently endure six to eight GP visits before anyone orders an MRI — a delay that allows tumours to grow and prognoses to worsen.
  • A dual-protein blood test developed in Manchester achieves over 90% accuracy in detecting these tumours and tracking their response to chemotherapy in near real time.
  • The test's signal remained stable even as tumours evolved genetically, suggesting it could catch recurrence earlier than conventional imaging allows.
  • Clinical trials are now running across ten sites in the UK and internationally, with researchers targeting NHS approval within a decade.
  • The lead researcher, whose father died of glioblastoma, envisions the test functioning like a rapid Covid test — fast, accessible, and deployable by a GP before a specialist is ever seen.

A patient arrives at their GP with headaches. They return again. And again. Under current practice, this cycle may repeat six to eight times before an MRI is ordered — and by then, if a tumour is present, it has likely grown. Researchers at the University of Manchester, working with collaborators in Denmark, believe they have found a way to collapse that timeline: a blood test detecting aggressive brain tumours with over 90% accuracy.

The test identifies two protein markers that appear in the bloodstream when certain brain tumours are active. Validated across glioblastoma patients — one of the most lethal brain cancers — it tracked them through surgery, radiation, and chemotherapy. Despite the genetic complexity of these tumours, the blood signal remained stable and informative throughout. The findings were published in Neuro-oncology Advances.

Professor Petra Hamerlik, who leads the research and whose father died of glioblastoma, speaks about the disease with the weight of both personal loss and scientific purpose. What makes the test especially significant, she explains, is its dual function: it can detect a tumour's presence and monitor how a patient responds to treatment in real time — telling a doctor within weeks whether chemotherapy is working, rather than waiting months for imaging to confirm it.

Hamerlik envisions a straightforward clinical pathway: a GP suspects a brain tumour, orders the blood test, receives a risk score, and refers the patient for MRI only if the result warrants it. She imagines it eventually resembling a rapid Covid test — quick, accessible, and available in primary care. The goal is NHS approval within a decade.

Ten trial sites across the UK and internationally are now testing the tool's real-world performance. The stakes are high. Earlier detection means smaller tumours, less damage to surrounding brain tissue, and meaningfully better outcomes. For patients facing a disease that kills most within two years of diagnosis, the difference between early and late detection is not a matter of convenience — it is a matter of time itself.

A patient walks into their GP's office complaining of headaches. They return a week later with the same complaint. They come back a third time. Under current practice, this cycle might repeat six to eight times before anyone orders an MRI scan to check for a brain tumour. By then, if a tumour is present, it may have grown considerably larger, and the prognosis has worsened. Researchers at the University of Manchester, working with collaborators in Denmark, believe they have found a way to collapse that timeline into something far more immediate: a simple blood test that can detect aggressive brain tumours with over 90% accuracy.

The breakthrough centres on two protein markers that appear in the bloodstream when certain brain tumours are present. The test was validated rigorously across glioblastoma patients—one of the most lethal forms of brain cancer—tracking them through surgery, radiation, and chemotherapy. What emerged from the data was striking: despite the genetic diversity and constant evolution of these tumours, the signal in the blood remained stable, robust, and highly informative. The findings were published in Neuro-oncology Advances and represent what researchers describe as a significant step toward transforming how brain tumours are diagnosed and monitored.

Professor Petra Hamerlik, who leads the research at Manchester and chairs the Brain Tumour Charity's translational neuro-oncology programme, has personal stakes in this work. Her father died of glioblastoma. She speaks about the disease with the clarity of someone who has watched it destroy someone she loved and who has spent her career trying to prevent that destruction in others. "Glioblastoma is one of the most devastating cancers we face," she said. "The lack of reliable tests has been a major barrier to earlier diagnosis and treatment response monitoring." The current standard of care—surgical biopsies and extensive MRI imaging—is invasive, time-consuming, and expensive. A blood test would be none of those things.

What makes this particular test remarkable, Hamerlik explained, is its dual function. It can identify whether a tumour is present at the outset, but it can also track how a patient responds to treatment in real time. When patients began chemotherapy after surgery, the biomarkers rose as tumours regrew and fell again as the drugs took effect. This means a doctor could know within days or weeks whether a treatment was working, rather than waiting months for imaging to reveal the answer. The test maintained its 90% accuracy even when disease returned, suggesting it could catch recurrence earlier than conventional monitoring allows.

The practical application Hamerlik envisions is straightforward. A patient presents to their GP with headaches. If the doctor suspects a brain tumour, they order the blood test. A risk score emerges. If it suggests malignancy, the patient is referred for MRI confirmation. No more cycling through multiple visits. No more delay. She imagines the test eventually taking the form of something like a rapid Covid test—quick, accessible, deployable in primary care settings. The goal is to bring it into the NHS within a decade, pending regulatory approval.

Clinical trials are now underway across six sites in the United Kingdom and four internationally, testing the test's real-world performance and refining its application. The timing matters. New glioblastoma treatments are in development, and Hamerlik is keen for this diagnostic tool to be ready when those medicines arrive. Early detection fundamentally changes outcomes. A smaller tumour caught before it has time to invade surrounding brain tissue means less damage to healthy neural structures, better surgical outcomes, and improved survival prospects. Late diagnosis, by contrast, means larger lesions, more extensive damage, and a steeper climb toward recovery.

Dr Simon Newman, chief scientific officer at the Brain Tumour Charity, called the research "a significant step towards a simple blood test that could help clinicians detect glioblastoma and monitor how patients are responding to treatment in real time." The language is measured, but the implication is profound. For patients and families facing glioblastoma—a disease that kills most of those diagnosed within two years—the difference between early detection and late detection can be the difference between years of life and months. A blood test that collapses the diagnostic timeline and tracks treatment response with precision is not merely a convenience. It is a lifeline.

Glioblastoma is one of the most devastating cancers we face. The lack of reliable tests has been a major barrier to earlier diagnosis and treatment response monitoring.
— Professor Petra Hamerlik, University of Manchester
Early and accurate diagnosis is absolutely critical for people with brain tumours, yet current tools are limited and often invasive.
— Dr Simon Newman, Brain Tumour Charity
The Hearth Conversation Another angle on the story
Inventor

Why does it matter that this test works on blood rather than requiring a biopsy?

Model

A biopsy means surgery. You're opening the skull, taking tissue, sending it to a lab. It's invasive, it carries risk, and it takes time. A blood test is something you can do in a GP's office in five minutes. You can repeat it easily, as often as you need, without risk.

Inventor

The 90% accuracy—is that good enough for something this serious?

Model

For a screening tool that flags which patients need an MRI, yes. It's not meant to replace imaging. It's meant to get you to imaging faster, with more confidence. It catches the signal early, before a patient has cycled through six visits to their doctor.

Inventor

You mentioned the test tracks treatment response. How does that change what a doctor can do?

Model

Instead of waiting three months for a scan to see if chemotherapy is working, you can see the biomarkers shift in weeks. If they're not moving the right direction, you know to try something else. You're not wasting time on a treatment that isn't working.

Inventor

Professor Hamerlik mentioned her father had glioblastoma. Does that shape how she talks about this work?

Model

It does. She's not speaking from theory. She knows what it means to watch someone decline from this disease, to wish there had been earlier warning, better tools. That's why she's so precise about the practical details—the GP visits, the timeline, the moment when intervention becomes possible.

Inventor

What's the biggest hurdle now?

Model

Regulatory approval and scaling. The science works. The trials are running. But getting it into NHS clinics, training GPs to use it, integrating it into standard care—that takes time and resources. She's realistic: a decade before it's routine.

Inventor

If this works as hoped, what changes for a patient?

Model

Everything. Instead of months of uncertainty and delay, you get answers in weeks. Instead of invasive procedures, you get a blood draw. Instead of guessing whether treatment is working, you know. For a disease as aggressive as glioblastoma, that acceleration of knowledge and response could mean the difference between years and months.

Contact Us FAQ