Stanford scientists find brain activity patterns that may predict Alzheimer's years before symptoms
Stanford Medicine researchers have developed a brain-activity scoring system that can signal Alzheimer's risk years in advance. It predicted cognitive decline better than several existing indicators.
Subtle patterns of brain activity can point to Alzheimer's disease risk years before memory problems appear or harmful proteins build up, according to a new study led by Stanford Medicine in the United States. The research was published on 4 November in Science Advances. It matters because, unlike conventional biomarkers, such activity patterns might one day be altered through targeted interventions.
How the study worked
The team compared brain activity in adults aged 75 and older at opposite ends of cognitive aging. One group had Alzheimer's disease; the other were "super agers", older people whose memory resembles that of people decades younger. In this work, that meant people at least 75 who scored on memory tests like a typical 45-year-old while keeping age-normal executive function. The researchers used resting-state functional MRI scans from 180 super agers and 110 people with Alzheimer's.
From the differences, they built the Alzheimer's resilience-vulnerability connectome score, or ARC score. It distinguished the two groups with roughly 85% accuracy.
Predicting decline in healthy people
The score was then applied to more than 1,300 cognitively healthy adults aged 55 to 70. At the start, people with high and low scores showed no clear differences in cognitive performance. In follow-up data, however, those whose brain activity looked more like that of Alzheimer's patients were more likely to experience cognitive decline, accumulate amyloid-beta plaques and be diagnosed with Alzheimer's. Follow-up reached up to 12 years, though most participants were tracked for five to six. People with super-ager-like patterns generally kept their abilities better.
The score predicted decline more effectively than several established indicators, including demographic factors, genetic risk and amyloid-beta and tau biomarkers. Combining it with those biomarkers improved prediction further.
Limitations
The researchers stress that the ARC score is not meant to diagnose Alzheimer's and is not a permanent forecast, since brain activity changes over time. It is not known whether changing brain activity would alter the course of the disease. Senior author Yu Zhang plans a trial measuring ARC scores before and after cognitive training.
