Researchers at Stanford University have developed a brain activity score that, in a study published November 4 in Science Advances, predicted which cognitively healthy adults would later show memory decline, accumulate amyloid-beta plaques, or receive an Alzheimer's diagnosis. The measure, called the Alzheimer's resilience-vulnerability connectome, or ARC score, is based on resting-state functional MRI scans, which record patterns of activity across brain regions while a person lies still in a scanner.
The team built the score by comparing 180 super agers, people aged 75 and older whose memory performed at about the level of someone 45 years old, with 110 people who had Alzheimer's disease. The researchers chose Alzheimer's patients of similar age as the comparison group rather than typical older adults, arguing that typical older adults vary too much biologically to reveal which brain patterns actually protect against decline. The ARC score distinguished the two groups with about 85 percent accuracy.
When applied to more than 1,300 cognitively healthy adults aged 55 to 70, people with scores resembling the Alzheimer's pattern tended to decline faster on tests of memory, attention and executive function, while those with super-ager-like scores tended to decline most slowly. The study reports that the score predicted future decline more effectively than demographic factors, genetic risk, and amyloid-beta and tau biomarkers, and that combining it with those biomarkers improved prediction further. Follow-up ranged up to 12 years, though most participants were tracked for about five to six years.
Lead author Kanhao (Alex) Zhao and senior author Yu Zhang say the long-term aim is to use brain activity as a target for intervention. They mention cognitive training and transcranial magnetic stimulation as possible approaches, but neither has been tested for preventing cognitive decline in this setting, and the researchers do not yet know whether changing brain activity would alter the course of the disease. The team is preparing a trial that will measure ARC scores before and after cognitive training. The researchers also stress that ARC scores are not diagnostic tools, that scores can shift over time, and that the proportion of people with super-ager-like scores generally declines with age.