The accumulation of neurofibrillary tangles tracks with cognitive impairment in Alzheimer’s disease, but what about other forms of dementia? The second most common cause of dementia, cerebrovascular disease, shrinks the brain’s blood supply and damages white matter. Could it trigger tau pathology? Yes, according to a paper in the May 14 JAMA Neurology. Sang Won Seo, Sungkyunkwan School of Medicine, Seoul, South Korea, and colleagues scanned people with vascular cognitive impairment for neurofibrillary tau. They report that cerebrovascular disease boosts uptake of the tau tracer, and that tau accumulation in the inferior temporal cortex appears to account for cognitive impairment, independent of amyloid. The results suggest that tau could represent a common pathway for dementia triggered by both cerebrovascular and amyloid pathologies.

  • First look at tau PET in people with cerebrovascular disease.
  • AV1451 uptake is independent of amyloid, correlates with cognitive decline.
  • Tau may mediate effects of vascular dysfunction to cause dementia.

“The findings are of great interest and importance in that they add to the growing body of evidence that cerebrovascular pathology and tau can drive cognitive decline independent of Aβ,” said Mario Merlini at the Gladstone Institute of Neurological Disease in San Francisco.

The study is small and will need to be replicated, say the authors and several commentators, including Costantino Iadecola, Weill Cornell Medical School, New York. “Nonetheless, it’s one more piece of evidence that calls attention to the intriguing interaction between vascular factors leading to white-matter damage and pathology associated with neurodegenerative diseases,” Iadecola told Alzforum. The work raises possibilities for treating vascular dementia, he said. “In AD, people say amyloid is the gun and tau is the bullet. If tau is also the bullet in this context, then if we counteract tau we should have benefits. That will be the next idea to test.”

First author Hee Jin Kim focused on people with cognitive impairment or dementia due to cerebrovascular small vessel disease (CSVD). In this condition, exacerbated by age, high blood pressure, and other factors, blood vessels deep in the brain develop atherosclerosis, and clog up or start to leak. Damage shows up on MRI as cloudy splotches or small voids in the subcortical white matter, or small hemorrhages known as microbleeds. Symptoms include deterioration of executive function, memory loss, and other cognitive problems. CSVD also contributes to dementia in Alzheimer’s disease, where the two pathologies frequently co-exist.

For this multimodal imaging study, Kim recruited 61 people with subcortical vascular cognitive impairment. They were people who had MRI evidence of significant ischemic damage, indicated by multiple white-matter hyperintensities and voids, and who had trouble in at least one aspect of language, memory, visuospatial or executive function. The group was comprised of 72 percent women, 79 years old on average, and nearly half were amyloid-positive on florbetaben scans. About half had mild cognitive impairment, half dementia. They were scanned using the tau PET ligand AV1451, which binds to neurofibrillary tangles. The investigators also scanned two control groups: 19 small vessel disease-negative, amyloid-negative volunteers with normal cognition, and 27 small vessel disease-negative people with amnestic MCI or probable AD. 

Troubling Tau. Compared with normal controls (top row), uptake of AV1451 increases in people with subcortical vascular cognitive impairment (SVCI, second row). The increase is apparent in people who are amyloid-negative (third row) or -positive (fourth row), but in the latter, i.e. Alzheimer’s, tau explodes (bottom two rows). [Image from Kim et al., 2018.]

Kim and colleagues detected more AV1451 binding in the people with vascular cognitive impairment than in those with normal cognition. The greatest uptake occurred in the bilateral inferior temporal cortex. The investigators observed the increase whether or not the person was Aβ-positive. By comparison, people with AD had much higher tracer uptake with a wider distribution in the bilateral temporal, parietal, and frontal regions of the cortex.

The AV1451 signal in the inferior temporal regions correlated with CVSD severity, and was independent of amyloid load or age. In contrast, amyloid positivity in the people with vessel disease correlated with AV1451 binding in medial temporal regions, independent of CSVD status. In addition, the pattern of cortical spread in CVSD patients differed from that normally seen in AD, with vascular disease showing a relative sparing of the parahippocampal region. “Perhaps the vascular burden promotes tau in a somewhat different way than amyloid does,” Kim wrote in an email to Alzforum.

How did the tau signal relate to cognition? In the vascular disease group, uptake correlated with worse language function and diminished general cognitive status, as indicated by Mini Mental State Exam (MMSE) scores. There was no correlation between tau PET and other cognitive domains. Separate analysis of the Aβ-positive and -negative subgroups gave similar results.

Finally, the investigators tested the strength of association of each pathology—CVSD or amyloid—with AV1451 uptake and MMSE scores. A model of all these associations suggested that localized uptake of AV1451 in the inferior temporal cortex completely explained the effect of CSVD on cognition, while uptake into the medial temporal lobe mediated the effect of Aβ.

“It’s interesting that what correlates best with cognitive dysfunction due to CSVD is the tau and not the Aβ,” said Iadecola. “Studies before suggested that amyloid was triggered by these factors, which then increased tau. Their analysis suggests that tau is directly linked to the vascular risk factors, and not through Aβ.”

Julie Schneider, Rush University Medical Center, Chicago, was not convinced. She told Alzforum that dividing the vascular disease group into amyloid-positive or -negative does not take in to account the possible effects of subthreshold amyloid. “You have to make some kind of cut point, to determine negative or positive, but we know that there’s a continuum of amyloid levels,” she said. “Is it possible that the people who had tangles driving cognitive impairment in the vascular disease group had more amyloid and more AD pathology overall than the cognitively normal people? That’s the million-dollar question.” In addition, the association of vascular disease and tau might not be causative, but instead could reflect underlying risk factors that are shared with AD, like high blood pressure and ApoE, she said.

Merlini wondered exactly what AV1451 picks up in the patients with CSVD. “Although we cannot be sure, it may be that the tracer binding they see in the vascular cognitive impairment group is related to tau accumulated within the blood vessel wall, in line with our findings from postmortem studies in AD brain where this can occur in the absence of amyloid plaque burden,” he said (Merlini et al., 2016). However, this vessel-associated tau is different from neurofibrillary tangles seen in neurons, and it’s not clear how well AV1451 binds them (Bennett et al., 2018; Blair et al., 2015). “We can only be certain once we see a postmortem analysis of these brains,” he said.

In the same spirit, Merlini said he’d like to know the relationship between regional vascular dysfunction and AV1451 uptake. “We know the subjects have microbleeds and white-matter hyperintensities, but whether other and region-specific vascular deficits are contributing to the findings is not clear from this study,” he said. Knowing that could, for example, clarify whether the observed tracer uptake in the inferior temporal areas in vascular impairment reflects local vascular issues that may instigate tau pathology, or a pathway dysfunction that leads to tau accumulation.

In an email to Alzforum, Kim wrote that the group had not looked at the relationship between regional vascular disease and tau. “The common regions for vascular pathology are periventricular areas or deep white-matter regions in frontal or parietal areas. In AD, it is well known that the location of amyloid deposition does not match the location of tau. Likewise, the mismatch between vascular disease and tau did not surprise us. But this is an important point and we may look further into it.”

For next steps, Kim and colleagues are following up with the participants using neuropsychological tests and repeat amyloid and tau scans. “Our goal is to evaluate longitudinal changes of amyloid/tau/cerebral small vessel disease markers and their contribution to cognitive decline,” she wrote. Kim agreed that postmortem data is critical, adding, “Ultimately, we need autopsy data to confirm our findings through pathology.”

Schneider told Alzforum she thinks it’s unlikely that cerebrovascular disease is solely working through tangles to cause cognitive impairment. “We know that there is other damage associated with white-matter hyperintensities. But even if tau is not mediating or not completely mediating the effects of cardiovascular disease, it’s important to recognize the common co-occurrence of tau and cerebrovascular pathology. This work highlights that people with vascular disease have tau pathology and we need to think in terms of these co-pathologies.”—Pat McCaffrey


  1. This is an interesting study with well-characterized patient groups exploring the cross-sectional relationship between subcortical small vessel disease, amyloid, and tau deposits in patients with subcortical vascular cognitive impairment. Using flortaucipir- and florbetaben-PET, Kim and colleagues observed independent contributions of amyloid and small vessel disease with regional tau pathology. These findings are consistent with animal studies noting an association between increased cerebrovascular pathology and tau formation (Amtul et al., 2014; Bennett et al., 2018), and CSF studies have reported both independent and synergistic associations between amyloid pathology and small vessel disease on tau pathology or hippocampal volume (Freeze et al., 2017; Guzman et al., 2013; Kim et al., 2015). It may be that additive or synergistic relationships between pathologies can be detected at different stages of the disease.

    With respect to cognition, it is interesting to note that while the tau patterns in the subcortical vascular cognitive impairment group are similar to what is typically observed in patients with prodromal Alzheimer’s disease (Johnson et al., 2016), the relationships with cognition are different, with tau pathology in subcortical vascular cognitive impairment being primarily related to language and not memory. Given the small sample size, the findings need to be replicated in larger cohorts. The mediation model showing that the relationship between either amyloid burden or small vessel disease on cognition is mediated by distinct patterns of tau pathology is suggestive that this may be the case, though more regions should be explored. However, longitudinal investigations and drug interventions will be necessary to ascertain the temporal and potentially causal relationships.

    This work represents an important piece in the larger discussion on whether cognitive impairment due to Alzheimer’s disease and cerebrovascular pathology share common endpoints with respect to tau, neurodegeneration, and neuronal dysfunction. A rich neuropathological literature that informs us that the co-occurrence of cerebrovascular and AD pathology is quite common (e.g., Schneider et al., 2007), emphasizing that a better understanding of how vascular and AD pathologies interact in vivo is a critically important question going forward. Given that overall lower levels of tau PET signal were seen in SVCI as compared to ADCI, the observations here are consistent with the idea that cerebrovascular disease reduces the threshold at which Alzheimer’s disease pathology (especially tau pathology) leads to cognitive impairment, at least in impaired individuals with some level of increased amyloid burden. The authors in the current article used a visual assessment to define presence of amyloid, but it would be interesting for future studies that use quantitative measurements of amyloid burden to investigate whether there is a dose-response relationship between the magnitude of small vessel disease and the threshold at which amyloid and tau pathologies impact cognition. Ultimately, understanding the link between vascular pathology and Alzheimer’s disease pathology will be important for the implementation of vascular biomarkers in the diagnostic research criteria framework (Vemuri and Knopman, 2015Jack et al., 2018) and, more broadly, to the success of clinical trials in late-onset AD.  


    . Comorbid Aβ toxicity and stroke: hippocampal atrophy, pathology, and cognitive deficit. Neurobiol Aging. 2014 Jul;35(7):1605-14. Epub 2014 Jan 8 PubMed.

    . Tau induces blood vessel abnormalities and angiogenesis-related gene expression in P301L transgenic mice and human Alzheimer's disease. Proc Natl Acad Sci U S A. 2018 Feb 6;115(6):E1289-E1298. Epub 2018 Jan 22 PubMed.

    . White Matter Hyperintensities Potentiate Hippocampal Volume Reduction in Non-Demented Older Individuals with Abnormal Amyloid-β. J Alzheimers Dis. 2017;55(1):333-342. PubMed.

    . White matter hyperintensities and amyloid are independently associated with entorhinal cortex volume among individuals with mild cognitive impairment. Alzheimers Dement. 2013 Oct;9(5 Suppl):S124-31. PubMed.

    . NIA-AA Research Framework: Toward a biological definition of Alzheimer's disease. Alzheimers Dement. 2018 Apr;14(4):535-562. PubMed.

    . Tau positron emission tomographic imaging in aging and early Alzheimer disease. Ann Neurol. 2016 Jan;79(1):110-9. Epub 2015 Dec 15 PubMed.

    . Hippocampal volume and shape in pure subcortical vascular dementia. Neurobiol Aging. 2015 Jan;36(1):485-91. Epub 2014 Sep 28 PubMed.

    . Mixed brain pathologies account for most dementia cases in community-dwelling older persons. Neurology. 2007 Dec 11;69(24):2197-204. PubMed.

    . The role of cerebrovascular disease when there is concomitant Alzheimer disease. Biochim Biophys Acta. 2015 Sep 25; PubMed.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Tau pathology-dependent remodelling of cerebral arteries precedes Alzheimer's disease-related microvascular cerebral amyloid angiopathy. Acta Neuropathol. 2016 May;131(5):737-52. Epub 2016 Mar 17 PubMed.
  2. . Tau induces blood vessel abnormalities and angiogenesis-related gene expression in P301L transgenic mice and human Alzheimer's disease. Proc Natl Acad Sci U S A. 2018 Feb 6;115(6):E1289-E1298. Epub 2018 Jan 22 PubMed.
  3. . Tau depletion prevents progressive blood-brain barrier damage in a mouse model of tauopathy. Acta Neuropathol Commun. 2015 Jan 31;3:8. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Assessment of Extent and Role of Tau in Subcortical Vascular Cognitive Impairment Using 18F-AV1451 Positron Emission Tomography Imaging. JAMA Neurol. 2018 Aug 1;75(8):999-1007. PubMed.