In Alzheimer’s disease research, the cerebral cortex has long occupied the spotlight, but white matter increasingly claims its share of attention. In the November 14 JAMA Neurology online, researchers led by Douglas Dean III at the University of Wisconsin-Madison report that myelin deterioration in normal individuals at risk for AD correlates with cerebrospinal fluid markers of plaques and tangles. “These new results emphasize the importance of adding a potentially key role for white matter very early in the course of AD pathogenesis,” wrote Gene Alexander at the University of Arizona, Tucson, in an accompanying editorial.

Myelin Omen.

Imaging uncovers correlations (colored) between myelin degradation in various regions of the brain and CSF Aβ42. [Courtesy of JAMA Neurology. Copyright 2016 American Medical Association. All rights reserved.]

AD is most often thought of as a disease of the cortex. Although white matter alterations have been observed in postmortem studies and MRI scans, overall, the AD research community has paid little attention to the role of myelin in AD (e.g., Bartzokis, 2011; Fletcher et al, 2013; Zhuang et al, 2013). 

Scientists have questioned whether white matter changes are important for AD, said Henrik Zetterberg, University of Gothenburg in Sweden, a co-author of the study. “I think many have felt they may be more coincidental than involved in the disease process,” Zetterberg told Alzforum.

Dean and colleagues looked at myelin health using a new computational model to analyze MRI data (Deoni et al., 2013). It goes by the mouthful “multicomponent-driven equilibrium single-pulse observation of T1 and T2,” thankfully shortened to mcDESPOT. Like other MRI approaches, mcDESPOT interprets signals that indicate the presence of water, but unlike other methods, it focuses on relaxometry measures that show how MRI signals decay over time. It also models the water in brain tissue as distributed between three compartments: myelin, intra- and extracellular locations, and free water (in the cerebrospinal fluid, for example). mcDESPOT allows researchers to calculate a myelin water fraction, which gives a quantitative measure of myelin content. Other imaging methods, such as the more commonly used diffusion tensor imaging, also provide views of myelin integrity, but mcDESPOT is uniquely sensitive and specific for myelin, and it provides quantitative data, said Dean.

The study builds on the work of the late George Bartzokis, who used relaxometric measures to study myelin in AD and other brain diseases while at the University of California, Los Angeles. In the new study, the researchers first used mcDESPOT to examine how myelin changes with age in a small group of normal people at risk for AD who were recruited from the Wisconsin Alzheimer’s Disease Research Center and the Wisconsin Registry for Alzheimer’s Prevention study run by co-authors Mark Sager and Sterling Johnson at the University of Wisconsin. Dean scanned 71 individuals between 48 and 72 years old, of whom 28 carried at least one APOE4 allele and 54 had a parental history of AD. Consistent with findings from other studies that used more traditional MRI techniques, Dean saw a steady, widespread decline in myelin with age, as reflected by the myelin water fraction (MWF), showing that aging takes a toll on myelin content and structure. “It is almost possible to tell a person’s age” by the MWF, said Paul Thompson at the University of Southern California in Los Angeles.

Next, the researchers looked at the relationship between the MWF and standard CSF biomarkers for AD plaques and tangles, including Aβ42, the soluble amyloid precursor protein fragment shed by β-secretase (sAPPβ), total tau (t-tau), and phosphorylated tau (p-tau181). The MWF negatively correlated with sAPPβ and the tau markers and with reduced CSF Aβ42 as measured by the t-tau/Aβ42, p-tau181/Aβ42, and sAPPβ/Aβ42 ratios. These correlations were evident across the brain, including frontal, temporal, parietal, and cerebellar white matter (see image above).

Intriguingly, the sAPPβ/Aβ42 ratio correlated best with the MWF decline in regions that are most likely to degenerate in AD, including temporal and frontal white matter, the body of the corpus callosum, and the cingulum. Evaluating how the biomarker levels dovetailed with the age-dependent decline of the MWF, the researchers found that higher ratios of p-tau181/Aβ42 and t-tau/Aβ42 predicted faster myelin degeneration.

Do these correlations mean that MWF measured by mcDESPOT could be used as a diagnostic test for AD? Probably not, said Zetterberg, adding, “Because they happen in many other conditions, white matter changes are too unspecific to become an AD test.” That said, together with other measurements, including more specific biomarkers for AD, the approach may help better predict who will decline to dementia, suggested author Barbara Bendlin at the University of Wisconsin-Madison.

What do these new findings reveal about early AD? “It is too early to say,” said William van Nostrand at Stony Brook University in New York. “This study shows a really nice correlation, but at this point it’s still just correlative,” he said. Using experimental models to tease out causes and effects is one way to move forward, he suggested. Tracking people’s MWF and AD biomarkers as they age could help discern whether myelin degradation or Aβ pathology come first. Bendlin has already started such a study. Thompson agreed that will be valuable. He added that correlating biomarker measurements with early cognitive changes may yield a fuller clinical picture of the disease process, and help inform future therapeutic studies. “It’s very important to know how early you can see divergence between patients and healthy controls,” he said.—Marina Chicurel

Marina Chicurel is a writer based in Santa Cruz, California.


  1. Age is the greatest risk factor for late-onset Alzheimer’s disease, which accounts for at least 95 percent of all incident cases. Given that white matter microstructure is strongly correlated with age, it stands to reason that age-related changes in this non-neuronal tissue may contribute to the pathogenesis of Alzheimer’s disease.

    In their paper, Dr. Dean and colleagues tested this hypothesis by correlating brain MRI estimates of myelin content (Myelin Water Fraction [MWF], R1, R2) and CSF measurements of amyloid and tau in a sample of cognitively asymptomatic adults enriched for AD risk. In addition to corroborating the known correlation between myelin and age, they reported that diminished myelin was associated with lower Aβ42 and higher tau (Ttau, Ptau181), tau-Aβ ratios (Ttau/Aβ42, Ptau181/Aβ42), and soluble APP (sAPPβ, sAPPβ/Aβ42), and that elevated tau-Aβ ratios mediate accelerated age-related myelin loss. These voxelwise correlations were localized to many white matter regions implicated in previous work on white matter changes in aging and Alzheimer’s disease.

    Their work demonstrates how myelin is not simply a bystander in this disease about which most research has been woefully over-focused on a handful of proteins and one cell type. These meaningful associations, albeit correlative, exemplify the clinical translation of our growing knowledge of neuron-glia interactions in the basic sciences. As a field we can no longer afford to be wedded to such circumscribed hypotheses of how Alzheimer’s disease begins and progresses, especially in the face of such data. Their work reminds us of what we in the field know all too well—that Alzheimer’s disease, much like the person it afflicts, is complex, multifaceted, and changing with age.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Alzheimer's disease as homeostatic responses to age-related myelin breakdown. Neurobiol Aging. 2011 Aug;32(8):1341-71. PubMed.
  2. . Loss of Fornix White Matter Volume as a Predictor of Cognitive Impairment in Cognitively Normal Elderly Individuals. JAMA Neurol. 2013 Nov 1;70(11):1389-95. PubMed.
  3. . Microstructural white matter changes, not hippocampal atrophy, detect early amnestic mild cognitive impairment. PLoS One. 2013;8(3):e58887. PubMed.
  4. . One component? Two components? Three? The effect of including a nonexchanging "free" water component in multicomponent driven equilibrium single pulse observation of T1 and T2. Magn Reson Med. 2013 Jul;70(1):147-54. Epub 2012 Aug 22 PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Association of Amyloid Pathology With Myelin Alteration in Preclinical Alzheimer Disease. JAMA Neurol. 2017 Jan 1;74(1):41-49. PubMed.
  2. . An Emerging Role for Imaging White Matter in the Preclinical Risk for Alzheimer Disease: Linking β-Amyloid to Myelin. JAMA Neurol. 2017 Jan 1;74(1):17-19. PubMed.