Three retired collision-sport athletes with suspected chronic traumatic encephalopathy showed elevated uptake of the experimental tau tracer [18F]OXD-2314 at the junction between gray and white matter, and in white matter, compared with seven healthy controls, according to a preliminary abstract presented at the Society of Nuclear Medicine and Molecular Imaging annual meeting in Los Angeles, held from May 30 to June 2, 2026. The work, abstract 262070 in the Journal of Nuclear Medicine supplement, comes from Isabelle Boileau, Lucas Narciso, Emily Murrell, Neil Vasdev and colleagues at the Brain Health Imaging Centre, Centre for Addiction and Mental Health, in Toronto. All ten participants had a 120 minute dynamic brain PET scan after injection of about 190 MBq of the tracer, with MRI used to define regions of interest. Distribution volume ratio was somewhat raised in cortical regions in the athletes and, on the researchers' reading, significantly higher at the gray-white junction. In separate laboratory work on donated brain tissue, the tritiated version of the same molecule, [3H]OXD-2314, produced a signal in every post-mortem CTE case examined.
The distribution matters because it is the one CTE is known for. The disease is defined at autopsy by tau deposits clustered around small blood vessels at the depths of cortical sulci, near the gray-white boundary, a pattern quite unlike the Alzheimer's distribution that existing tau tracers were built to detect. CTE cannot at present be diagnosed in a living person; diagnosis is made after death. A tracer optimised for non-Alzheimer's tau, and reading out in the right place, is the missing instrument that would let the disease be studied, staged and eventually treated in life. Boileau said that if validated, the tracer could provide the first accurate in-life diagnostic biomarker for CTE, and described the radiopharmaceutical as being at the early clinical research stage.
[18F]OXD-2314 emerged from ligand-based design work published in Nature Communications in June 2024, in which more than 150 analogues of an earlier 4R-tau lead were screened against post-mortem tissue from progressive supranuclear palsy, corticobasal degeneration and Pick's disease. Its first-in-human evaluation, in four healthy adults aged 49 to 65, was published in the European Journal of Nuclear Medicine and Molecular Imaging in June 2025; no adverse events were reported and the tracer crossed the blood brain barrier with fast washout from all brain regions. A companion abstract at the same 2026 meeting reported raised uptake in four people with early progressive supranuclear palsy against the same seven controls.
The caveats are large and the researchers do not hide them. This is a conference abstract, not a peer-reviewed paper: three patients and seven controls, no autopsy confirmation in the living participants, and no published test of how well the scan separates people with CTE from people without it. The athletes were described as having suspected CTE, a clinical impression, not a proven one. Boileau herself said the signal was not exactly what the team expected. Earlier tau tracers repeatedly showed off-target binding that confounded interpretation, and nothing yet published rules that out here. Whether the gray-white junction signal reflects CTE tau, some other feature of injured white matter, or noise in a very small sample is not settled by this work.