CTE in 2026: what the new evidence established, and what it still cannot

Two developments in 2026 moved the science on chronic traumatic encephalopathy, and they moved it in different directions — one toward firmer ground on what CTE does, the other toward an early and very preliminary answer on how it might one day be detected in a living person.

The dementia finding

On 27 January 2026, Alzheimer's & Dementia published an analysis of 614 brain donors, led by Rachael M. Layden of the Boston University Alzheimer's Disease Research Center [s1][s2]. The design isolated 366 donors who had CTE and no other progressive brain disease, and compared them with 248 donors without CTE [s1][s2]. That isolation is the point of the study. Earlier work had struggled to separate CTE's contribution to dementia from the Alzheimer's pathology that frequently accompanies it.

The result: CTE stage IV, the most advanced of four stages, was associated with 4.48 times the odds of dementia (95% CI 1.97–10.90) compared with no CTE [s1]. Stage III carried an odds ratio of 2.12 [s1]. Stages I and II showed no significant association with dementia status [s1].

Higher CTE stage was also associated with greater informant-reported cognitive symptoms, with stage IV donors showing roughly double the symptom scores of those without CTE [s1]. On mood and behavioural measures, the study found nothing: "CTE stage of any severity was not associated with any of the mood scales" or the behavioural measures examined [s1]. The authors note this contradicts assumptions built into existing clinical criteria for the condition [s1].

Michael Alosco, associate professor of neurology at Boston University Chobanian & Avedisian School of Medicine and co-director of clinical research at the BU CTE Center, framed the implication as a diagnostic problem: "CTE has a significant impact on people's lives, and now we need to accelerate efforts to distinguish CTE from Alzheimer's disease" [s2]. The work was funded by the National Institute of Neurological Disorders and Stroke, the National Institute on Aging, the Department of Veterans Affairs and the Nick and Lynn Buoniconti Foundation [s2].

The limits the authors state

This is a brain bank study, and the authors are explicit about what that means. Donors to a CTE-focused brain bank "are more likely to be symptomatic and have CTE neuropathology", which can inflate the associations found [s1]. The sample was 97% male, 81.4% white, and 80.3% had played American football at some level from youth to professional, with a mean age at death of 51.99 years (SD 20) [s1]. Generalising to women, to other sports, or to non-sport head injury exposure is not supported by this dataset [s1].

Dementia status and symptoms were established retrospectively through next-of-kin report, introducing recall bias [s1]. Age differed significantly across CTE stages despite statistical adjustment [s1]. The authors conclude that prospective clinical-pathological studies with objective assessments and age-matched controls are needed to validate the findings [s1][s2].

What the study does not establish is how common any of this is among people who played contact sports. A brain bank cannot produce a denominator. The odds ratios describe the relationship between advanced CTE pathology and dementia within this donor sample; they say nothing about an individual player's risk of developing either.

The imaging development

The second development is much earlier stage and should be read as such. At the Society of Nuclear Medicine and Molecular Imaging 2026 annual meeting, researchers reported results from a novel tau PET radiotracer, 18F-OXD-2314, in three retired collision-sport athletes with suspected CTE and seven healthy controls [s3].

Compared with controls, the athletes' scans showed elevated tracer uptake at the grey-white matter junction and in white matter — the distribution pattern characteristic of CTE pathology at autopsy [s3]. Separately, a tritiated form of the compound bound to tissue in all post-mortem CTE cases examined, which the researchers describe as early biological confirmation that the tracer attaches to tau pathology in human CTE tissue [s3].

Isabelle Boileau, senior scientist and associate director of the Brain Health Imaging Centre at the Centre for Addiction and Mental Health in Toronto, is among the researchers, working with colleagues at the University Health Network, the University of Toronto and McMaster University [s3]. The team's own framing is conditional: "If validated, 18F-OXD-2314 could help provide the first accurate in-life diagnostic biomarker for CTE" [s3]. The work is described as at the early clinical research stage [s3].

Three subjects and seven controls is a feasibility result, not a diagnostic test. It cannot establish sensitivity, specificity, or whether the tracer distinguishes CTE from other tauopathies in people whose diagnosis is genuinely uncertain — which is the only situation where such a test would be needed. Confirmation would require imaging in living patients who subsequently come to autopsy, which by definition takes years.

Why the two findings connect

Alosco's stated priority — distinguishing CTE from Alzheimer's disease — is the bridge [s2]. The January study's contribution is to show that advanced CTE pathology is associated with dementia even when Alzheimer's and other progressive diseases are absent, with an effect size the researchers describe as comparable in strength to advanced Alzheimer's pathology [s2]. That strengthens the case for treating CTE as a distinct clinical entity rather than a variant presentation.

But a distinct clinical entity is only useful if it can be identified in a living person, and at present CTE can be confirmed only at autopsy [s3]. That is the gap the tracer work is aimed at, and it is nowhere near closed.

What to watch

Whether prospective cohorts — following living former athletes with objective cognitive testing and imaging, then obtaining autopsy confirmation — begin reporting, since that is the design the January authors identify as necessary [s1]. Whether 18F-OXD-2314 progresses to a larger imaging series with a comparison group that includes Alzheimer's patients, rather than only healthy controls [s3]. And whether the negative finding on mood and behaviour is replicated, because if it is, the clinical criteria currently used to describe suspected CTE in living people will need revising [s1].

Sources

• [s1] CTE neuropathology alone is associated with dementia and cognitive symptoms — Alzheimer's & Dementia, 27 January 2026 (DOI 10.1002/alz.71032)
• [s2] CTE should be recognized as new cause of dementia, study suggests — News-Medical, 27 January 2026
• [s3] PET suggests potential CTE biomarker in living humans — Society of Nuclear Medicine and Molecular Imaging via EurekAlert, 31 May 2026