Brain structure changes appear seven years before Alzheimer’s plaques are detected

Structural changes in the brain’s outer layer can be detected years before amyloid-beta proteins reach high levels on brain scans. This discovery suggests that physical alterations to brain tissue might be an earlier sign of Alzheimer’s disease than previously thought. The findings were published in Nature Neuroscience.

Alzheimer’s disease is a progressive neurological condition that destroys memory and other important mental functions. One of the main biological hallmarks of the disease is the accumulation of amyloid-beta, a protein that clumps together to form plaques between nerve cells in the brain. Historically, the standard model of Alzheimer’s assumed that these amyloid plaques build up first and trigger brain shrinkage later. A 2010 study proposed exactly this timeline, arguing that amyloid buildup is the very first event in the disease process, with brain volume loss occurring only as a later consequence.

Earlier research showed that elevated amyloid-beta levels accelerate brain volume loss. For instance, a study covered by PsyPost in 2024 found that people with high amyloid levels exhibited faster brain volume loss even before cognitive symptoms appeared. However, hints emerged that brain tissue might paradoxically swell and change well before amyloid scans turn positive. A 2020 review noted that the outer layer of the brain temporarily thickens or swells in the earliest stages of genetic Alzheimer’s before it begins to waste away.

The new study directly tracks people over time to confirm that these cortical changes can be detected years before amyloid buildup crosses the threshold of detection on brain scans. The cerebral cortex is the wrinkled outer layer of the brain, playing a major role in memory, thinking, and reasoning. The thickness of this layer typically decreases as people age, a process known as cortical thinning. The research, led by James M. Roe and Yunpeng Wang of the University of Oslo, aimed to see if changes in this thinning process happen before a person registers as positive for amyloid-beta.

To investigate this timeline, the researchers analyzed data from three large, long-term studies of cognitive aging. They focused on individuals who were cognitively healthy, meaning they did not show signs of dementia or cognitive impairment. The researchers used magnetic resonance imaging, or MRI, which is a technique that uses magnetic fields to create detailed images of brain structure. This allowed them to measure the thickness of the cerebral cortex over time. They also used positron emission tomography, or PET scans, which use a mildly radioactive tracer to reveal the presence and quantity of amyloid-beta plaques in the brain.

The final sample included 4,570 MRI scans from 1,051 individuals. Among these participants, 691 also had amyloid-beta PET scans available. The research team used the PET scan data to divide the participants into two distinct groups. The first group consisted of 77 converters, which were people who initially had low levels of amyloid-beta but later crossed the threshold into high, positive levels during the study. The second group consisted of 412 individuals who remained negative for high amyloid-beta across all of their available PET scans.

The scientists then looked back at the MRI data. For the converters, they artificially cut off the timeline, excluding any MRI scans taken near or after the time the person tested positive for amyloid-beta. Instead, they focused purely on brain scans taken between one and ten years before the individual’s first positive PET scan. By doing this, they could compare the early brain structures of people who would eventually develop high amyloid-beta to those who never did. In their analyses, the researchers controlled for variables like biological sex, average age, overall brain size, and the specific type of MRI scanner used.

The results provided evidence that individuals who later developed high amyloid levels showed a thicker cerebral cortex compared to those who remained amyloid-negative. These individuals also exhibited a slower rate of age-related cortical thinning over time. Essentially, the expected rate of brain shrinkage was reduced in the years leading up to a positive amyloid scan. These physical differences were detectable in MRI scans taken at least seven years before the participants reached amyloid levels high enough to be classified as positive.

These structural differences were especially prominent in the frontal regions of the brain. The researchers wanted to know if these differences were simply a byproduct of slowly rising amyloid levels that had not yet crossed the positive threshold. When they adjusted their statistical models to account for the exact, continuous levels of amyloid buildup measured in the early PET scans, many of the cortical thickness differences persisted. This indicates that the structural changes to the cortex are at least partly independent of early amyloid accumulation.

The researchers also found a spatial relationship between the brain changes and amyloid buildup. The geographic pattern of cortical thickening closely mapped onto the regions of the brain where amyloid-beta tends to accumulate. In addition, the timing of these structural changes paralleled the spread of the plaques. Brain regions that experienced earlier structural changes tended to be the exact regions where amyloid began to deposit first.

As with all research, there are a few things to keep in mind. The time cutoffs used in the study were based on when an individual was first observed to have high amyloid-beta, not the exact biological moment they crossed the threshold. Because PET scans were taken at intervals, some people might have developed undetected amyloid buildup earlier than their first positive scan recorded. To address this, the researchers used statistical predictions to estimate the exact age of positivity, and the results remained consistent, though this approach relies on mathematical assumptions.

Another detail to consider is that the study only included cognitively healthy older adults. Participants who volunteer for long-term aging studies often represent a particularly healthy and high-performing demographic, which might not completely reflect the general population. It is also important to note that the study did not include measures of tau, another key protein involved in Alzheimer’s disease that strongly affects brain shrinkage. The researchers pointed out that because tau typically accumulates after amyloid-beta, it is unlikely to be driving the structural changes seen seven years prior to amyloid positivity.

Future research will need to explore what exactly causes this early increase in cortical thickness. An apparent increase in thickness on an MRI scan could reflect several different biological processes, such as an inflammatory response to early amyloid deposits, or changes in the supporting cells of the brain, rather than the addition of new neurons. Tracking these changes alongside other biological markers will help clarify whether this early swelling is a protective mechanism or the first step in the neurodegenerative process.

The study, “Cortical thickness changes precede high levels of amyloid by at least 7 years,” was authored by James M. Roe, William J. Jagust, Susan M. Landau, Theresa M. Harrison, Håkon Grydeland, Maksim Slivka, José-Luis Alatorre-Warren, Pablo F. Garrido, Øystein Sørensen, Edvard O. S. Grødem, Tyler J. Ward, Esten H. Leonardsen, Alice Murphy, JiaQie Lee, Tormod Fladby, Atle Bjørnerud, Kristine B. Walhovd, Anders M. Fjell, Didac Vidal-Piñeiro, and Yunpeng Wang.