22 February 2013. When it comes to vascular brain injury, the dementia field has been mired in uncertainty. What is it, exactly? How can scientists measure and treat it? How does it relate to Alzheimer's pathology or contribute to cognitive decline? Answers to these old questions are still lacking, but two recent studies published online February 11 and 18 in JAMA Neurology suggest that vascular damage and amyloid plaques occur independently in early stages of AD. William Jagust, University of California, Berkeley, and colleagues find that the pathologies are separate, and that only vascular injury tracks with cognitive decline. Likewise, scientists led by Adam Brickman, Columbia University, New York, report that white matter hyperintensities and amyloid are independent risk factors for Alzheimer’s disease (AD). These findings add to a growing body of literature suggesting that vascular causes of decline are distinct from amyloid pathology, and that they deserve their own spot at the dementia table.
“Vascular brain injury has typically taken on a secondary role in the hunt for what causes dementia in Alzheimer’s disease,” said Natalie Marchant, first author on the Jagust paper and now at the University of Sussex, U.K. “It is important to understand what Aβ does to cognition, but our study emphasizes that it is necessary to consider the impact of vascular brain injury, too.”
A number of studies suggest a link between vascular risk factors and Aβ aggregation (for a review, see Pluta et al., 2009). For instance, acute hypoxia, such as that resulting from a stroke, leads to more Aβ production in mouse neurons (Zhang et al., 2007). However, other studies suggest that AD pathology and blood vessel damage are quite autonomous risk factors for cognitive impairment (Hedden et al., 2012; Lee et al., 2011). Last year, Jagust’s group reported that cerebrovascular disease and amyloid deposition occur independently in cognitively normal people, and that only the former tracks with cognitive decline (see Marchant et al., 2012). Nevertheless, the researchers wondered if they might be missing a relationship between the two by studying healthy controls who, after all, are more likely to have only one or neither pathology.
To see if a link emerged in symptomatic people, Marchant and colleagues added people with mild cognitive impairment and early dementia to their sample. All participants came from the multisite Aging Brain Study, which enrolls adults 55 and older enriched for vascular risk factors including high cholesterol, elevated blood pressure, and a history of smoking. Researchers correlated magnetic resonance imaging (MRI) and positron emission tomography (PET) with Pittsburgh Compound B (PIB), with performance on cognitive tests. All participants had both types of scan. The researchers noted if and where brain infarcts (evidence of stroke) and white matter hyperintensities (lesions that indicate myelin and axon damage) occurred in the MRI scans and classified each individual as PIB positive or negative.
More than half of the 61 participants had an infarct, and slightly less than half were PIB positive. Still, the researchers found no relationship between the pathologies. Having a vascular brain injury did not hike a person’s odds that Aβ had aggregated, regardless of the number of infarcts or where they occurred in the brain. What’s more, neither white matter hyperintensities nor PIB status associated with cognition. Infarctions in the cortical and subcortical gray matter, however, did correlate with poorer cognitive performance.
“In this group, we find that some vascular injury seems to play a bigger role than amyloid in producing cognitive decline,” Jagust told Alzforum. “Vascular disease is operating in some other way—besides through amyloid—to harm the brain.”
Do these results contradict some of Jagust’s previous findings? He and others had reported that cerebrovascular risk factors heighten the chances of amyloid deposition (see Reed et al., 2012). Jagust explained that while the current data suggest that infarcts themselves do not promote amyloid buildup, other vascular factors could mediate that relationship. He will continue to probe other measures in their dataset, such as HDL and LDL cholesterol, as well as inflammation, to see how they relate to amyloid and cognitive function. In addition, the researchers will follow this cohort longitudinally and examine vascular and amyloid pathology at autopsy.
Amyloid plaques are poor predictors of cognitive decline to begin with. What about other Aβ species, such as soluble oligomers? They cannot be detected by the PIB-PET imaging used in this dataset, so might they still be linked to vascular disease? Possibly, Jagust said, but he is skeptical. “If a relationship exists between oligomeric Aβ and vascular injury, we ought to be seeing it when we look at fibrillary Aβ,” he told Alzforum. “It is hard to believe a link would be there for oligomeric Aβ, then go away when it becomes fibrillar, since there should be plenty of soluble Aβ around, too.”
Zoe Arvanitakis, Rush University Medical Hospital, cautioned that the sample size in this study is quite small, and is further divided into subgroups for analysis. Even so, she said, “This study underlines the importance of considering vascular factors in individuals with mild cognitive problems, even if AD pathology is suspected.”
Steven Greenberg, Massachusetts General Hospital, Boston, agreed. “The vascular contribution to dementia is not small—it’s really a substantial part of the overall level of cognitive impairment.” But these data cannot definitively say that no link exists between vascular and amyloid pathology, he argued. “The researchers focus only on people with no, or mild, cognitive impairment.” If they had included severely demented patients, a relationship may have arisen, he said.
Brickman did include people with more advanced disease in his sample of 100 people from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Unlike Jagust’s sample, this one was not enriched for vascular risk. However, first author Frank Provenzano and colleagues found a similar result. Both white matter hyperintensities and PIB uptake independently associated with an AD diagnosis and predicted who would progress from MCI to AD. “Even in this healthy population without a lot of vascular disease history, we are able to show the importance of cerebrovascular disease in the Alzheimer’s phenotype,” he told Alzforum.
The findings do not exclude the idea that one pathology may affect the other, said Brickman. In his sample—unlike Jagust’s—people with the most amyloid also had the most white matter hyperintensities. “It’s just too soon to say that they are truly independent processes,” Brickman told Alzforum. More work is required beyond these statistical correlation studies to determine how amyloid and vascular factors interact. Nevertheless, “It’s time we really start thinking about how vascular disease can be incorporated in our understanding of the pathogenesis of AD more formally,” Brickman said. “It’s not irrelevant, confounding, or a second pathological feature. It’s a true core aspect of the disease, and we really need to address it in studies and clinically.”
A Vascular Identity Crisis
These papers come at a time when the field is struggling to incorporate vascular injury into its overall concept of cognitive decline. “There is overwhelming epidemiological data to suggest that vascular factors contribute to late-life dementia,” said Charles DeCarli, University of California, Davis. Scientists have had trouble getting a handle on cerebrovascular damage, and some factions still see it as but a minor cause of cognitive decline, he said.
One problem is that vascular brain injury has been hard to define. It can take many forms—cerebrospinal fluid-filled brain cavities called lacunes, infarcts, microbleeds, white matter hyperintensities, small clots of circulating solid material called emboli—that occur in different areas of the brain and are associated with varying levels of cognitive decline. Without a precise definition of vascular dementia, scientists have had trouble taking any further steps toward clinical trial design or treatments.
To complicate things further, vascular injury rarely occurs alone. Although pure subcortical vascular dementia has been shown to exist, it is often accompanied by evidence of brain amyloid. That makes it hard to tease out which pathology causes what cognitive effects. Clinical trials for an anti-amyloid treatment may fail if many participants are demented because of vascular problems, said DeCarli.
Another limitation of the scanning technology is the resolution of MRI—it cannot detect microinfarcts, said Lenore Launer of the National Aging Institute in Bethesda, Maryland. She previously found through postmortem analysis that microinfarcts strongly correlate with cognitive decline (see Launer et al., 2011). “That is another vascular player that is underestimated in a lot of studies,” Launer told Alzforum. “There’s nothing to do about that until we come up with better imaging,” she added.
Researchers have been making some progress on definitions of vascular dementia and vascular cognitive impairment. The most recent attempt was published in a 2011 Statement for Healthcare Professionals by the American Heart Association/American Stroke Association (see Gorelick et al., 2011), on which DeCarli was an author. It defined vascular cognitive impairment (VCI) as “a syndrome with  evidence of clinical stroke or subclinical vascular brain injury and  cognitive impairment affecting at least one cognitive domain.” Vascular dementia is the most severe form of VCI. The paper went on to outline criteria for probable, possible, and unstable forms. The criteria depend on time between vascular events and cognitive symptoms, patterns in severity, and the presence of other pathologies. It may be the best definition so far, said Reinhold Schmidt, Medical University of Graz, Austria, but it will need further refinement. That is especially true for the diffuse subcortical cerebrovascular disease that is included in some of the subcategories.
Conducting longitudinal multimodal imaging studies may help improve the definition, said Schmidt. MRI and PET imaging could begin to define the type, extent, and location of small vessel lesions that lead to cognitive impairment, Schmidt said. Such studies would also reveal how vascular damage changes over time, how it relates to neurodegenerative pathologies, and how it tracks with clinical consequences. Monitoring people with one form of hereditary subcortical vascular disease—cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL)—may be especially helpful in informing researchers about the effects of pure cerebrovascular disease.
MRI is currently the best way to measure most types of vascular brain injury. Infarcts, white matter hyperintensities, and microbleeds all turn up on various types of MRI scans. Scientists also use transcranial Doppler ultrasonography to measure emboli (see ARF related news story). Further technological improvements could help, especially in measuring microinfarcts, said DeCarli. Better MRI resolution may allow researchers to see these in vivo, or perhaps a future PET tracer could pick them out, he said.
These attempts to define and understand just how cerebrovascular disease affects cognition are not academic; they are important to help physicians manage the disease and start working toward drug discovery. No approved VCI treatments exist. The hope is that genomewide association studies will reveal molecular pathways related to small vessel changes and cognitive impairment that could be targeted, Schmidt said, illustrating how far the field of VCI currently lags behind AD. Lack of targets is not the only impediment. Few, if any, clinical trials are underway to test drug candidates, in part because it is unclear how to measure clinical outcomes. The measurement scales used for AD are not appropriate for VCI because people’s memory may stay intact. That is one of the reasons Eisai withdrew its FDA application for Aricept® use in VaD in the early 2000s, Schmidt told Alzforum.
In the meantime, a slew of preventative measures can minimize risk, said DeCarli. For instance, reducing blood pressure has been found to cut dementia incidence by 50 percent (see Forette et al.,1998, and AlzRisk entry on blood pressure). Fighting obesity and diabetes also reduces vascular disease, he pointed out. The field should start designing trials to test such treatments on middle-aged people to look for effects on future dementia risk and include brain outcome measures in ongoing vascular intervention trials, DeCarli said. Going forward, scientists should work out how to incorporate risk factors into a public health message, said Launer.—Gwyneth Dickey Zakaib.
Marchant NL, Reed BR, Sanossian N, Madison CM, Kriger S, Dhada R, Mack WJ, Decarli C, Weiner MW, Mungas DM, Chui HC, Jagust WJ. The Aging Brain and Cognition: Contribution of Vascular Injury and Aβ to Mild Cognitive Dysfunction. JAMA Neurol. 2013 Apr 1:1-8. Abstract
Provenzano FA, Muraskin J, Tosto G, Narkhede A, Wasserman BT, Griffith EY, Guzman VA, Meier IB, Zimmerman ME, Brickman AM; for the Alzheimer's Disease Neuroimaging Initiative. White Matter Hyperintensities and Cerebral Amyloidosis: Necessary and Sufficient for Clinical Expression of Alzheimer Disease? JAMA Neurol. 2013 Feb 18:1-7. Abstract