Aug. 22, 2025
Before the Tangles
Biomarker tests developed at Pitt detect early Alzheimer’s pathology.
TOPICS: Aging & Lifespan | Alzheimers | Chronic Disease | Neuro & Cognitive Science | Research | School of Medicine
Look to the Stars
By Anastasia Gorelova
Years before tau tangles show up in brain scans of patients with Alzheimer’s disease, small amounts of the clumping-prone tau protein in its misfolded pathological forms litter the brain, cerebrospinal fluid and, potentially, blood.
A biomarker test developed at the University of Pittsburgh can detect those early stages of tau-tangle formation, research published in Nature Medicine in February 2025 suggests.
The cerebrospinal fluid biomarker test correlates with the severity of cognitive decline, independent of other factors, including brain amyloid deposition. That means it opens doors for early-stage disease diagnosis and intervention.
“Early detection of tangle-prone tau could identify the individuals who are likely to develop Alzheimer's-associated cognitive decline and could be helped with new generation therapies.”
Thomas Karikari, assistant professor of psychiatry
Most biomarker efforts have focused on early detection of amyloid-beta changes because they often precede tau abnormalities in Alzheimer’s disease. However, the clumping of tau protein into well-ordered structures referred to by pathologists as “neurofibrillary tangles” is a more defining event for Alzheimer’s disease as it is more strongly associated with cognitive changes.
“Early detection is key to more successful therapies for Alzheimer’s disease since trials show that patients with little-to-no quantifiable insoluble tau tangles are more likely to benefit from new treatments than those with a significant degree of tau brain deposits,” says Thomas Karikari, assistant professor of psychiatry at Pitt School of Medicine and senior author of the study.
Many older adults who have amyloid-beta plaques in their brains will never go on to develop cognitive symptoms of Alzheimer’s disease. Because of this, the diagnostics framework developed by the Alzheimer’s Association specifies the three neuropathological pillars necessary to diagnose the disease—combined presence of tau and amyloid-beta pathology and neurodegeneration.
In a quest for early and accessible biomarkers for Alzheimer’s disease, Karikari’s earlier work showed that a brain–specific form of tau, called BD-tau, can be measured in blood and reliably indicates the presence of Alzheimer’s disease-specific neurodegeneration. Karikari also showed that specific forms of phosphorylated tau in the blood can predict the presence of brain amyloid-beta without the need for costly and time-consuming brain imaging.
But these tools largely detect amyloid pathology, so the issue of early detection of tau still looms large. While tau-PET remains a reliable and accurate predictor of tau burden in the brain, the test’s utility is limited by availability, low resolution, high cost, labor and sensitivity. At present, tau-PET scans can pick up the signal from neurofibrillary tangles only when a large number are present in the brain, at which point the degree of brain pathology has become pronounced and is not easily reversible.
In this latest research, using the tools of biochemistry and molecular biology, Karikari and team identified a core region of the tau protein that is necessary for neurofibrillary tangle formation.
“Amyloid-beta is a kindling, and tau is a matchstick. A large percentage of people who have brain amyloid-beta deposits will never develop dementia. But once the tau tangles light up on a brain scan, it may be too late to put out the fire, and their cognitive health can quickly deteriorate,” says Karikari.
“Early detection of tangle-prone tau could identify the individuals who are likely to develop Alzheimer’s-associated cognitive decline and could be helped with new-generation therapies.”
In other recent work, Karikari was the senior author of a study that independently validated a new blood test platform that can simultaneously measure more than a hundred biomarkers of Alzheimer’s disease.
Published in the journal Molecular Neurodegeneration in October 2024, the study suggested the platform might improve the ability of clinicians to capture the multifaceted nature of Alzheimer’s pathology and streamline early disease diagnostics.
“Alzheimer’s disease should not be looked at through one single lens,” says Karikari.