Skip to main content
Premium Trial:

Request an Annual Quote

Gothenburg Team Identifies Brain-Specific Tau Protein as Potential Marker for Stroke Outcomes


NEW YORK – A team led by researchers at the University of Gothenburg has identified a form of tau protein as a potential blood-based marker for projecting patients' functional outcomes following stroke.

In a study published last month in Neurology, the scientists observed that higher blood levels of brain-derived tau (BD-tau) following a stroke were linked to poor outcomes at three months follow-up, indicating that the marker could be useful for quickly assessing stroke severity.

The researchers are also exploring whether the marker could be useful in Alzheimer's patients for detecting edema and hemorrhages caused by treatment with anti-amyloid drugs.

Tara Stanne, a researcher at the Institute of Biomedicine at the University of Gothenburg and one of the two first authors on the paper, said that currently the best predictors of functional outcome following an ischemic stroke are the patient's age and stroke severity, the latter of which is assessed by a clinical rating score called the NIH Stroke Scale (NIHSS).

Stanne said there are currently no clinically implemented blood-based biomarkers that can accurately reflect acute neuronal damage after a stroke or that can help improve outcome predictions.

The marker emerged from efforts by Stanne's Gothenburg colleagues to develop an assay specific to tau produced in the brain, as opposed to peripheral tissues. Tau is a known marker of brain injury, and hyper-phosphorylated forms of the protein and tau aggregates have also been linked to various neurodegenerative conditions. One form of phosphorylated tau, phosphorylated-tau 217 (p-tau 217), is a leading blood-based marker for Alzheimer's disease.

Tau isn't specific to the brain, however, noted Fernando Gonzalez-Ortiz, a physician and researcher at Gothenburg and author on the Neurology study. Forms of the protein are also produced by peripheral tissues. These peripherally produced forms contain a region produced by the transcription of an exon that is not included in BD-tau. In a paper published last year in Brain, Gonzalez-Ortiz and colleagues detailed an anti-tau antibody specific to the BD-tau form.

In the Brain study, the researchers focused primarily on the use of BD-tau as a blood-based Alzheimer's marker, observing that while total tau levels in serum did not correspond to total tau levels in cerebrospinal fluid (CSF) due to the product of tau in peripheral tissues, blood-based BD-tau did correspond to CSF tau measurements and was able to identify Alzheimer's positive individuals as effectively as CSF total tau.

Gonzalez-Ortiz and his collaborators then reached out to Stanne and her colleagues, thinking that because BD-tau reflects neuronal damage in the brain specifically, it might prove useful as a stroke marker, as well.

In the Neurology study, the researchers looked at 454 subjects from the Sahlgrenska Academy Study on Ischemic Stroke (SAHLSIS) cohort, a group of adults under the age of 70 who have suffered ischemic stroke and controls, and 364 subjects from the SAHLSIS2 cohort, which included adults of all ages. They measured BD-tau levels in these participants using Quanterix's HD-X system and found that after adjusting for age, sex, day of blood draw, and NIH stroke scale score, higher BD-tau levels were "significantly associated with increased odds of unfavorable outcome" at three months.

Stanne said she and her colleagues are now working to assess how well BD-tau levels mirror the size of the ischemic stroke, so-called infarct volumes.

They are also looking to see if the marker may hold more promise in particular patient subgroups. "We are trying to dig into that, looking into different stratified analyses to see where BD-tau may add the most predictive information over known clinical parameters," Stanne said.

Another key question is the ideal time point for measuring the marker. In the Neurology study, the researchers analyzed samples collected several days after the subjects' strokes (median of four days for the SAHLSIS cohort and two days for the SAHLSIS2 cohort). Stanne said it will be important to explore how close to the time of the stroke sampling for the marker can be done.

In terms of future clinical implementations, Stanne said that "BD-tau may prove useful as a noninvasive, cost-effective biomarker to complement the widely used NIHSS score and other clinical assessments." For example, the marker could help to improve clinical confidence and help to refine patient inclusion criteria in future clinical trials for stroke therapies.

She added, however, that "it is still early days, and the results clearly need to be validated in other larger clinical cohorts."

Stanne said that in the near term, the marker would likely prove most useful for refining patient inclusion in clinical trials for stroke therapies.

Gonzalez-Ortiz said that he and his colleagues also hope to link the Alzheimer's and stroke portions of their research by exploring whether BD-tau might be a useful marker for early detection of side effects from treatment with anti-amyloid drugs like Eisai's Leqembi (lecanemab).

"We know that a main side effect of [anti-amyloid] drugs is ARIA — amyloid-related imaging abnormalities," he said. "And among them, we have edema and hemorrhagic abnormalities. [BD-tau] seems to be very sensitive at capturing CNS changes and CNS insults. So, what we are trying to do at the moment is find the right collaboration to show that the marker can work in the context of patients receiving anti-amyloid therapies to detect early ARIA."

Gonzalez-Ortiz said that BD-tau could also lead to the development of more accurate blood-based Alzheimer's diagnostics. He noted that while in a neurodegenerative context, phosphorylated tau forms like p-tau 217 appear highly specific to Alzheimer's, hyper-phosphorylated tau is also produced outside the brain by conditions including certain cancers. Being able to distinguish between phosphorylated tau forms in blood produced by the brain as opposed to peripheral tissues could become important if Alzheimer's testing moves into a more general, asymptomatic population.

Gonzalez-Ortiz said that he and his colleagues do not themselves have plans to commercialize a BD-tau assay. The rights to the anti-BD tau antibody used in the Brain and Neurology papers are owned by UK firm Bioventix, which produced the antibody.