Two people can get the same score on a memory test today. Years later, one may start missing appointments and struggling with everyday tasks while the other still manages both with ease.
An early clue to that difference may lie in a brain scan, research suggests.
In a new study, researchers at Stanford Medicine and their colleagues wanted to know whether the way brain regions work together at rest could reveal who is more likely to have memory problems years later.
To find out what to look for, they started with two groups whose memory abilities were already quite different.
One group included 110 people with Alzheimer's disease. The other included 180 adults who were at least 75 and scored better on memory tests than is typical for middle-aged adults.
Researchers call people with this kind of exceptional memory in later life SuperAgers.
The team examined scans taken while participants lay still. The technique, called functional MRI, tracks changes in blood oxygen as an indirect measure of brain activity. The researchers looked for regions whose signals rose and fell together.
Some connections involving the front of the brain were stronger in SuperAgers. Some involving the insula, a region tucked deep in the brain, were stronger in people with Alzheimer's.
"This suggests that Alzheimer's vulnerability may be reflected in how the brain functions as an interconnected system," Kanhao (Alex) Zhao, a postdoctoral researcher specializing in computational neuroscience at Stanford Medicine and the study's lead author, told ScienceAlert.
The team used those patterns to develop the Alzheimer's Resilience–Vulnerability Connectome, or ARC.
"The ARC score summarizes whether an individual's brain connectivity more closely resembles that of SuperAgers, who maintain exceptional memory into old age, or individuals with Alzheimer's disease," Zhao said.
Lower scores were more SuperAger-like and higher scores more Alzheimer's-like. When the researchers tested the method on a separate set of scans from 143 people who already belonged to one of those two groups, it placed about 71 percent in the correct group.
The next step brought the team back to its original question. Could the same score tell them something about the future of people whose memory and thinking skills were still intact?
The researchers calculated ARC scores from the first scans of 1,315 adults with no cognitive impairment and compared them with test results and clinical assessments from later years. Most datasets covered about five to six years, and the longest extended to 12.
People with lower and higher ARC scores performed similarly on most tests at first. Over time, those with more Alzheimer's-like scans declined faster on assessments of memory and everyday functioning.
They were also more likely to develop Alzheimer's disease or mild cognitive impairment, which can make memory or thinking problems noticeable while people can still manage most daily activities on their own.
People with more SuperAger-like scans were younger on average. To see how much that age difference could explain, the researchers ran an additional analysis in one dataset. Differences in clinical ratings and test performance persisted after they accounted for age at the start.
Follow-up scans also showed that amyloid beta, a protein that can build up into plaques in the brain, accumulated faster in people with more Alzheimer's-like scans, even though the groups started with similar levels. The researchers found no clear difference in the buildup of tau, another protein linked to Alzheimer's.
Adding ARC improved predictions of who would develop mild cognitive impairment or Alzheimer's beyond what researchers could predict using age, scans measuring amyloid beta and tau, and genetic risk linked to the APOE gene. If the findings hold up, the score could help researchers recruit people at higher risk into Alzheimer's prevention trials before memory problems begin.
"We would need larger prospective studies following cognitively healthy individuals over several years, with validation across diverse populations and MRI scanners," Zhao said.
"These studies must establish reliable individual-level predictions, demonstrate added value beyond existing biomarkers, and determine whether ARC-informed assessments improve clinical decisions."
The research was published in Science Advances.
This article was fact-checked by Fiona MacDonald and edited by Fiona MacDonald. While we pride ourselves on our process, we are only human. If you spot a mistake, please let us know.