Each year, thousands of women in the UK learn they have ovarian cancer only after it has already spread — a delay that quietly narrows their chances of survival. Researchers have now developed a blood test that reads the molecular fingerprints cancer leaves in the bloodstream, using machine learning to detect patterns too complex for human analysis alone, achieving accuracy rates between 88 and 93 percent in early-stage detection. The test, if it clears regulatory review and enters NHS practice, could shift the moment of discovery from crisis to possibility — catching a disease that has long h
Blood test shows promise for early ovarian cancer detection with 90%+ accuracy
Ovarian cancer affects approximately 7,500 women annually in the UK, predominantly over age 50, with late diagnosis reducing treatment effectiveness.
A biological fingerprint unique to ovarian cancer
Mark
So this test finds cancer fragments in blood. How is that different from the blood tests doctors already use for ovarian cancer?
Mimi
The difference is what it's looking for and how it interprets it. Current blood tests look for specific proteins. This one looks for a combination of lipids and proteins together—a pattern. And it uses AI to recognize patterns that are too complex for a person to see.
Luke
But we should be careful here. The study tested stored samples, not living patients. Those are different things. We don't know yet if 88 to 93 percent accuracy in a lab translates to the same accuracy when you're actually screening women in a clinic.
Mark
Why would it be different?
Luke
Because real-world screening is messier. You have different populations, different equipment, different technicians. The samples in this study were carefully collected and preserved. That's not how NHS screening would work.
Mimi
That's exactly why they want to do prospective trials next. They want to follow women forward and see if the test works the way the lab results suggest.
Mark
And if it does work—if they get regulatory approval—what changes for patients?
Mimi
Women with symptoms like pelvic pain or bloating could get a blood test instead of waiting for scans or biopsies. If the test is positive, they'd know to pursue more investigation. If it's negative, they'd have more confidence it's not cancer.
Luke
Though we should note: the test was developed by the company selling it. That's not a reason to distrust the science, but it's worth knowing. Independent verification matters.
Mark
How long until this is available on the NHS?
Mimi
That depends on regulators. The company has to apply for approval. Then the NHS has to decide if it's worth the cost and where it fits into screening protocols.
Luke
And that's a real question. Early detection is great, but only if the healthcare system can actually handle the volume of positive results and follow-up testing. We don't know yet if the NHS is ready for that.
Mark
So it's promising, but not here yet.
Mimi
Exactly. Promising and worth pursuing, but still in the research phase.
Der Puls
- Ovarian cancer kills quietly — its early symptoms mimic everyday discomfort, and by the time most women are diagnosed, the disease has already spread beyond reach of easy treatment.
- A new blood test by AOA Dx identifies lipid and protein fragments shed by cancer cells, then deploys machine learning to find patterns across those biomarkers that no clinician could spot unaided.
- Tested on 832 samples across two universities in Colorado and Manchester, the tool achieved up to 93% accuracy overall and 88–91% accuracy specifically in early-stage cases — the window where treatment works best.
- Researchers and oncologists are cautiously optimistic but insist the next phase — prospective trials, regulatory approval, and NHS integration — will determine whether laboratory promise translates into real-world impact.
- If validated, the test could reduce the human and financial cost of late-stage ovarian cancer diagnoses, reshaping how a notoriously elusive disease is caught and treated across the UK.
Each year, thousands of women in the UK learn they have ovarian cancer only after it has already spread — a delay that quietly narrows their chances of survival. Researchers have now developed a blood test that reads the molecular fingerprints cancer leaves in the bloodstream, using machine learning to detect patterns too complex for human analysis alone, achieving accuracy rates between 88 and 93 percent in early-stage detection. The test, if it clears regulatory review and enters NHS practice, could shift the moment of discovery from crisis to possibility — catching a disease that has long hidden behind ordinary symptoms until it was too late.
Every year, roughly 7,500 women in the UK are diagnosed with ovarian cancer, most of them over fifty, and most of them too late. The disease's early symptoms — bloating, pelvic pain, a persistent sense of fullness — are easy to dismiss as ordinary. By the time they become urgent, the cancer has often already spread, and the window for effective treatment has narrowed.
A new blood test developed by AOA Dx may offer a way to open that window earlier. The test works by detecting fragments that cancer cells shed into the bloodstream as they grow — a biological fingerprint made up of lipids and proteins unique to ovarian cancer. What makes it distinctive is not just what it looks for, but how it reads what it finds: a machine learning algorithm, trained on thousands of patient samples, identifies patterns in those biomarkers too subtle and interwoven for human analysis to catch.
A study published in Cancer Research Communications tested the approach on 832 samples from two universities — Colorado and Manchester. In Colorado, the test correctly identified ovarian cancer 93 percent of the time across all stages, and 91 percent in early-stage cases. Manchester results were similarly strong: 92 percent overall, 88 percent in early stages. The company's chief science officer described the tool as capable of navigating the molecular complexity of the disease across its different subtypes and stages.
Professor Emma Crosbie of the University of Manchester, a consultant in gynecological oncology, sees genuine potential — but calls for prospective trials that follow women forward in time rather than analyzing stored samples. The real questions now are practical ones: how the test performs in clinical settings, how it fits within NHS infrastructure, and whether those accuracy rates hold under real-world conditions. The science exists. The harder work of earning a place in the clinic is only beginning.
Every year, roughly 7,500 women in the UK receive an ovarian cancer diagnosis. Most are over fifty. Many of them are told too late—after the disease has already spread beyond the ovaries, when treatment becomes harder and outcomes grow grimmer. The standard approach to finding it hasn't changed much: doctors order scans, run blood tests, sometimes perform biopsies. It works, but not always in time.
Now researchers say a simple blood test might change that calculus. The test, developed by a company called AOA Dx, hunts for something cancer leaves behind: fragments that cancer cells shed into the bloodstream as they grow. These fragments carry a particular signature—a combination of lipids, the fat-like molecules that coat cells, and certain proteins. Together, they form what amounts to a biological fingerprint unique to ovarian cancer.
The innovation isn't just in what the test looks for. It's in how it reads what it finds. The test uses machine learning, an algorithm trained on thousands of patient samples, to spot patterns in those lipids and proteins that no human eye could detect. The patterns are too subtle, too numerous, too interwoven. A computer can hold them all at once.
A study published in Cancer Research Communications tested the approach on 832 samples collected at two universities: Colorado and Manchester. The results were striking. In the Colorado samples, the test correctly identified ovarian cancer 93 percent of the time across all stages of disease, and 91 percent of the time in early stages. In Manchester samples, accuracy was 92 percent overall and 88 percent in early stages. Those numbers matter because early detection is where the real difference lives. Ovarian cancer caught before it spreads responds far better to treatment.
The symptoms that might prompt a woman to seek testing—pelvic pain, bloating, feeling full quickly after eating, frequent urination—are easy to miss or dismiss. They're common. They're vague. A blood test that could confirm or rule out cancer before those symptoms become urgent could reshape how the disease is caught and treated. Alex Fisher, the company's chief operating officer, says the test can detect ovarian cancer "at early stages and with greater accuracy than current tools." Dr. Abigail McElhinny, the company's chief science officer, frames it differently: by combining multiple types of biomarkers through machine learning, they've built a tool that can navigate the molecular complexity of the disease across its different subtypes and stages.
Emma Crosbie, a professor at the University of Manchester and a consultant in gynecological oncology at Manchester University NHS Foundation Trust, sees the potential to reshape patient care. She and her colleagues want to run more trials—prospective studies that would follow women forward in time rather than backward through stored samples—to understand how this test might actually work inside the NHS, how it would fit into existing systems, what it would cost, and whether those 88-to-93 percent accuracy rates hold up in the real world.
That's the next step. Regulatory approval. Integration into the health service. The test exists. The science is solid. What comes now is the slower, harder work of proving it belongs in the clinic.
Bemerkenswerte Zitate
By using machine learning to combine multiple biomarker types, we've developed a diagnostic tool that detects ovarian cancer across the molecular complexity of the disease in sub-types and stages.— Dr. Abigail McElhinny, chief science officer of AOA Dx
AOA Dx's platform has the potential to significantly improve patient care and outcomes for women diagnosed with ovarian cancer.— Emma Crosbie, professor at University of Manchester and consultant in gynecological oncology
Möchten Sie die ganze Geschichte?
Das Original lesen bei The Independent ↗
Technology