Researchers said the test is not available for clinical use and cannot diagnose Alzheimer’s alone
A Montreal cohort was one of four used to validate a blood test that identified advanced Alzheimer’s tau tangle pathology in 728 participants.
The cross-sectional study of plasma eMTBR-tau across four cohorts appeared online in JAMA on Monday, and the findings were reported at the Clinical Trials on Alzheimer’s Disease conference in Boston, according to Reuters.
The test measures endogenously cleaved microtubule-binding region tau, or eMTBR-tau, a fragment that reflects insoluble tau tangles inside brain cells.
Wagner Brum of Banner Sun Health Research Institute in Sun City, Arizona, and co-authors used a single-plex nucleic acid-linked immunosandwich assay from Alamar Biosciences that requires 35 microlitres of plasma per sample, against the 1.5 mL used by the mass spectrometry method available in only a few laboratories.
Two cohorts were American autopsy groups: 281 brain donors in the Arizona Study of Aging and Neurodegenerative Disorders and Brain and Body Donation Program, and 240 at the University of California, Irvine, Alzheimer Disease Research Center.
The imaging cohorts were the Canadian Translational Biomarkers in Aging and Dementia study at the Montreal Neurological Institute, with 99 cognitively impaired participants, and the Swedish BioFINDER-2 study, with 108.
In the Arizona group, the age- and sex-adjusted geometric mean was 0.0021 pg/mL at Braak stages I and II and 0.107 pg/mL at Braak stage VI, a 50.5-fold difference.
The Irvine group showed 0.0040 pg/mL and 0.101 pg/mL at the same stages, a 25.2-fold difference.
Both carried a P value below .001.
Niklas Mattsson-Carlgren of Lund University in Lund, Sweden, who led the study, told Reuters in a statement that one of the hardest tasks in diagnosis is establishing that “Alzheimer’s pathology is not just present, but that is present in a sufficiently advanced biological stage to be a likely explanation for symptoms.”
The Arizona fold change for plasma phosphorylated tau 217 between Braak stages I-II and stage VI was 4.6, against 50.5 for eMTBR-tau.
Separating Braak stages I to IV from stages V and VI, the area under the curve was 0.93 for eMTBR-tau and 0.84 for p-tau217.
In the Montreal cohort, eMTBR-tau concentrations were 59.9 times higher at tau PET Braak stages V and VI than at stage 0, against 4.9 times for p-tau217.
Between 22.9 percent and 59.1 percent of eMTBR-tau values fell below the lower limit of quantification of 0.007 pg/mL, depending on the cohort.
Nicholas Ashton of Banner Sun Health Research Institute in Sun City, Arizona, who co-led the study, described a scalable way to measure tau tangle buildup as “a major step forward in our understanding of Alzheimer’s” in a statement reported by Reuters.
The test is not available for clinical use and is not meant to diagnose Alzheimer’s disease on its own, the researchers said in the JAMA report.
A second JAMA paper applying the same assay to chronic traumatic encephalopathy, published 19 November 2026, examined 21 male brain donors who had played American-style football.
Eighteen had a neuropathologic diagnosis of CTE. Median plasma eMTBR-tau was 0.019 pg/mL in CTE-positive donors and 0.001 pg/mL in the three who were CTE-negative.
Carla Palleis and Günter Höglinger of LMU University Hospital in Munich, Germany, writing in an accompanying JAMA editorial on tau biomarkers, said the CTE work should be regarded as preliminary, citing the 21-person cohort and the three-donor comparison group.


