A diagnosis that took two years could arrive in weeks
For decades, patients with a rare cancer of the eye and brain have endured a diagnostic odyssey lasting years, their disease advancing while physicians chased misleading symptoms. Researchers at Fudan University have now identified a protein — HAVCR1 — that distinguishes these lymphomas from benign inflammation with striking precision, offering the prospect of diagnosis measured in weeks rather than years. The discovery, published in Clinical Chemistry, reminds us that in medicine, the difference between a test and a treatment is often only time.
- Patients with primary vitreoretinal lymphoma wait an average of two years for diagnosis while a cancer that mimics common eye inflammation quietly advances toward the brain.
- In 60 to 90 percent of untreated cases, the disease spreads intracranially, raising the stakes of every month lost to misdiagnosis or inconclusive biopsy.
- The protein HAVCR1, identified by a Shanghai research team, achieved 92–100% diagnostic accuracy in eye fluid and 97–99% in spinal fluid — outperforming every current standard method.
- Crucially, the marker rises and falls with disease activity, meaning it could replace not just the initial biopsy but the repeated procedures used to monitor whether treatment is working.
- Before it reaches clinics, HAVCR1 must be validated across multiple laboratories and patient populations — a necessary threshold that researchers say is the next clear step.
A patient arrives with blurred vision and eye inflammation. The ophthalmologist suspects uveitis. Months pass, symptoms persist, and inconclusive tests accumulate. Two years later — sometimes longer — a diagnosis finally arrives: primary vitreoretinal lymphoma, a rare cancer of the eye. By then, the disease may already be moving toward the brain.
Researchers at Fudan University in Shanghai believe this diagnostic odyssey may soon be avoidable. They have identified a protein called HAVCR1 that could transform how these aggressive malignancies are detected. Their findings, published in Clinical Chemistry, suggest the marker could enable diagnosis weeks or months earlier, with far fewer invasive procedures, while also helping doctors track whether treatment is working.
Primary central nervous system lymphoma — encompassing cancers of the brain, spinal cord, and eyes — ranks among the deadliest blood cancers, and cases have risen steadily over fifty years. The current diagnostic standard measures interleukin levels in fluid samples, achieving accuracy of only 80 to 90 percent. For the eye variant, the challenge is steeper still: the cancer mimics uveitis so closely that doctors frequently treat the wrong condition, while the tumor cells themselves are too scarce and fragile for reliable biopsy. The disease progresses to the brain in 60 to 90 percent of untreated cases.
To find a better marker, the team screened proteins from 199 lymphoma patients and 179 controls. HAVCR1 emerged as consistently elevated in cancer patients — achieving 92 to 100 percent accuracy in eye fluid and 97 to 99 percent in spinal fluid, while distinguishing cancer from uveitis far more reliably than existing methods. The marker also responded dynamically to treatment: levels fell when patients improved and remained elevated when disease persisted, suggesting it could serve both as a diagnostic tool and a monitor of therapeutic response.
The path to clinical use still requires validation across laboratories and broader patient populations. But for patients currently waiting years for answers while a rare cancer advances, the prospect of a faster, more reliable test represents a meaningful change in what medicine can offer them.
A patient with blurred vision and eye inflammation arrives at a clinic. The ophthalmologist suspects uveitis, a common inflammatory condition. Months pass. The symptoms persist or worsen. Another doctor, another test, another inconclusive result. Two years later—sometimes longer—a diagnosis finally arrives: primary vitreoretinal lymphoma, a rare cancer of the eye. By then, the disease may have already begun its march toward the brain.
This diagnostic odyssey, familiar to patients with primary vitreoretinal lymphoma and related cancers of the central nervous system, may soon become avoidable. Researchers at Fudan University in Shanghai have identified a protein marker called HAVCR1 that could transform how these aggressive but uncommon malignancies are detected. The findings, published in Clinical Chemistry, suggest the biomarker could enable diagnosis weeks or months earlier, with far fewer invasive procedures, and could help doctors track whether treatment is actually working.
Primary central nervous system lymphoma—a category that includes cancers affecting the brain, spinal cord, and eyes—ranks among the deadliest of blood cancers. Cases have climbed steadily over the past fifty years despite improvements in imaging and treatment. The current gold standard for diagnosis relies on measuring interleukin levels in fluid samples, but this method is accurate only 80 to 90 percent of the time. For patients with the eye variant, primary vitreoretinal lymphoma, the diagnostic challenge is even steeper. The cancer mimics uveitis so closely that doctors often treat the wrong condition. The tumor cells themselves are scarce and fragile, making them difficult to biopsy reliably. The result: diagnosis typically takes up to two years and requires multiple procedures. Meanwhile, the disease progresses to the brain in 60 to 90 percent of untreated cases, raising the risk of permanent neurological damage or death.
To find a better marker, the Shanghai team screened proteins in fluid samples from 199 patients with these lymphomas and 179 controls without cancer or with benign inflammatory conditions. Using advanced protein-detection technology, they identified HAVCR1 as consistently elevated in cancer patients. In eye fluid from patients with vitreoretinal lymphoma, the marker achieved diagnostic accuracy between 92 and 100 percent. In spinal fluid from those with brain lymphoma, accuracy reached 97 to 99 percent. Crucially, HAVCR1 distinguished cancer from uveitis far more reliably than the current markers, which can produce ambiguous results when inflammation is present.
The biomarker offered another advantage: it responded to treatment. When patients responded well to therapy, HAVCR1 levels dropped. When disease persisted despite treatment, levels remained high. This dynamic behavior suggests the marker could serve double duty—not just diagnosing cancer, but monitoring whether a patient is actually getting better. The researchers determined that the cancer cells themselves produce HAVCR1, strengthening confidence that the marker reflects true disease rather than incidental inflammation.
The implications are substantial. A simple fluid test could replace or reduce the need for tissue biopsies, which are invasive, risky, and often inconclusive in eye cancers. Diagnosis could shift from years to weeks. Treatment could begin sooner, when the disease is less advanced. Doctors could watch for relapse without repeated procedures. Yet the path to clinical use is not automatic. The researchers acknowledge that further validation across different laboratories and patient populations is necessary before HAVCR1 becomes standard practice. The marker is promising, they emphasize, but not yet proven at scale. Still, for patients currently waiting months or years for answers while a rare cancer advances, the prospect of a faster, more reliable test represents a genuine shift in what's possible.
Notable Quotes
HAVCR1 is a robust fluid-based biomarker for PCNSL and PVRL with superior diagnostic accuracy compared with existing cytokine assays and dynamic responsiveness to treatment— Fudan University research team
The Hearth Conversation Another angle on the story
Why does it take so long to diagnose this particular cancer when others can be caught relatively quickly?
The eye variant especially is a mimic. It looks and acts like a common inflammatory disease, so doctors treat inflammation first. And you can't easily biopsy the eye without damaging it. The cells are fragile. So you end up doing multiple procedures, getting unclear results, and the patient waits.
And during that waiting period, what's happening to the disease?
It's progressing. In most cases, it spreads from the eye to the brain. That's when it becomes much harder to treat and the neurological damage sets in. Early diagnosis isn't just about convenience—it's about preventing irreversible harm.
So this biomarker is essentially a blood test that could replace all that?
Not quite a blood test—it's in the fluid around the brain and eye. But yes, a simple fluid sample instead of repeated biopsies. And it's accurate 92 to 100 percent of the time, compared to 80 to 90 percent with current methods.
What makes HAVCR1 different from the markers doctors use now?
The current markers are cytokines—general inflammation signals. They light up when there's inflammation, period. HAVCR1 is produced by the cancer cells themselves. It's specific. And it changes with treatment, so you can actually see if therapy is working.
Is this ready to use in hospitals tomorrow?
Not yet. It's been validated in one research cohort. Before it becomes standard, it needs to work the same way in different labs, different patient populations. But the researchers are confident about the direction.
What happens to a patient once this is available?
Instead of two years and multiple procedures, they get a diagnosis in weeks. Treatment starts before the cancer spreads to the brain. That's the difference between manageable disease and permanent neurological damage.