Article In Brief
In blood samples from participants in the longitudinal Framingham Heart Study, researchers applied a highly sensitive test for total tau and found that those with higher levels had an increased risk for stroke.
Elevation of tau in the bloodstream is an independent risk factor for stroke, according to a study in the July 4 online issue of Annals of Neurology.
The finding provides a new biomarker for assessing stroke risk and supports the idea that subclinical vascular damage causes neuronal injury.
“This is a very nice paper that expands our knowledge about the overlap of markers of neurodegenerative and vascular mechanisms of brain disease,” commented Mitchell S. Elkind, MD, FAAN, professor of neurology and epidemiology at Columbia University, who was not involved in the study.
The new finding is the most recent result to come from the Framingham Heart Study, the pioneering longitudinal study of health and disease in a community west of Boston. Previous work from the study helped lead to an understanding of many of the major stroke risk factors, including hypertension, diabetes, smoking, and cholesterol, and the to the development of the Framingham Stroke Risk Profile, which combines these many risk factors to provide a 10-year stroke probability.
As part of the ongoing study, participants donate blood specimens, which can be unfrozen and analyzed anew as new tests become available. Recently, a highly sensitive test for total tau (t-tau) in plasma has been developed, which Sudha Seshadri, MD, FAAN, and colleagues used to show earlier this year that elevated plasma t-tau was a biomarker for dementia risk. [The study finding was reported in the April 18 issue of Neurology Today.]
In the new study, working again with colleagues from the Florey Institute for Neuroscience and Mental Health in Melbourne, Australia, and Boston University, Dr. Seshadri used the test to ask whether that same elevation also predicted an increased risk of stroke.
Dr. Seshadri is a senior investigator for the Framingham Heart Study, founding director of the Glenn Biggs Institute for Alzheimer's and Neurodegenerative Diseases at UT Health San Antonio, and is professor of neurology at Boston University School of Medicine.
In her previous study, Dr. Seshadri found that elevated plasma t-tau was associated with microinfarcts at autopsy, “so we suspected that tau might be increased in those most at risk for stroke,” she said.
Study Design, Findings
To test that association, the researchers thawed blood samples that had been taken from 2794 individuals in the Framingham Study between 2005 and 2008. They then analyzed them for t-tau using a “single-molecule molecular array,” a highly sensitive immunoassay with a detection limit far below an ELISA test, the usual standard for protein detection. The test detects total tau, including normal and phosphorylated isoforms.
Over the following 8.3 years, participants had 101 incident strokes, 82 of which were ischemic. Those participants who were in the top quintile for t-tau had a 2.28-fold increased risk of a stroke compared with those in the bottom four quintiles. The risk remained significantly elevated after adjusting for age, sex, systolic blood pressure, treatment for hypertension, current smoking status, total cholesterol, HDL cholesterol, atrial fibrillation, diabetes, and cardiovascular disease (hazard ratio, 2.01; 95% confidence interval, 1.32-3.08).
Dr. Seshadri suspects that the elevation in tau reflects neuronal injury from microvascular injury. “Vascular injury could account for both the increase in stroke risk and increase in tau in the blood,” she said. “This is an observational study in humans, so some of those steps will need proving in animal studies, but that is our thinking.”
But if tau elevation reflects risk from underlying vascular damage, why isn't that risk already captured by the Framingham Stroke Risk Profile? “I don't think we completely know why tau elevation is adding information” to the Profile, Dr. Seshadri said. Tau elevation may be tracking the aggregate burden of neuronal injury due to vascular factors, she suggested. By getting their blood pressure under control and stopping smoking, a person might reduce their risk as determined by the Profile, but the cumulative effect of vascular damage, reflected in tau elevation, might nonetheless put them at higher risk for stroke.
“Another possibility is that individuals have different degrees of resilience to these vascular risk factors,” Dr. Seshadri added. “It is possible that tau is identifying people who, at the same level of these other risk factors, have a greater risk of neuronal injury. The Framingham Stroke Risk Profile is a great risk stratification tool, but it is not perfect.”
Whatever the explanation, Dr. Seshadri said, “tau appears to add more information about stroke risk, over and above other risk factors. This is something novel.”
“Although this observational study is unable to provide a mechanism for the association of plasma t-tau with stroke risk, the most likely reason for the association is that elevated t-tau is a measure of subclinical, or more minor, vascular injury to the brain,” said Dr. Elkind of Columbia University. “This would be consistent with previous findings that so-called silent brain infarcts or ischemic white matter damage predict future clinical strokes. It could be that these types of subclinical vascular brain injury are a measure of cumulative damage to blood vessels in the brain, beyond that detected by the burden of known risk factors, which ultimately leads to stroke.”
“What remains unknown, however, is whether t-tau represents a different pathway towards neuronal injury that could be exploited for a novel treatment approach,” he added. “At present, that doesn't seem to be the case. Instead it appears to be a reflection of injury from other, perhaps more traditional, causes like hypertension and diabetes, even if it provides a more sensitive measure of those causes.”
Ralph L. Sacco, MD, MS, FAHA, FAAN, professor and Olemberg Chair of Neurology at the Miller School of Medicine at the University of Miami, and executive director of the McKnight Brain Institute, agreed that the finding is interesting and novel, and that it adds to the finding that plasma t-tau is a biomarker for dementia.
“It is still too early to recommend this test to help predict stroke risks,” he said. “We already know with reasonable accuracy other modifiable, conventional risk factors that accurately predict stroke,” including hypercholesterolemia, diabetes, high blood pressure, and smoking, which cumulatively predict 10-year risk with a high degree of accuracy. “Even when we know there is an increased risk of stroke, it is hard to get people to modify their lifestyle and control risk factors. Even if total tau increases risk, we don't know yet how to modify it.”
The single-molecule molecular array test for t-tau is not yet clinically available, but given its ability to add to an understanding of stroke risk, and perhaps more importantly to predict dementia risk, it may become available in the future. That would require replication in larger cohorts of varying ethnic groups, and determination of clinically relevant thresholds, Dr. Elkind noted.
“If the plasma t-tau marker does make it to clinical practice, like markers of cardiovascular risk such as cholesterol and C reactive protein concentrations, then the next step would be to show that choosing a given therapy, such as statins, based on a risk prediction paradigm using t-tau, actually improves clinical outcomes and reduces strokes,” he said. “That would probably require a very large sample size, but it could be done.”
Drs. Seshadri, Dr. Sacco, and Elkind report no conflicts.