How the brain behaves at rest may help flag who is likely to develop Alzheimer's disease long before memory problems appear, according to research led by Stanford Medicine and published November 4 in Science Advances. The same measure also flagged people whose brains appear unusually resistant to decline and who may be on track to become so-called super agers.
How the score was built
The researchers started with adults aged 75 and older at opposite ends of cognitive aging: people with Alzheimer's and super agers. In this work, a super ager was someone whose memory test results matched those of a typical 45-year-old, with executive function normal for their age. Stanford says roughly 10% of older adults may fit the label, with the figure depending on the definition used.
Senior author Yu Zhang, an assistant professor of psychiatry and behavioral sciences, said earlier work mostly set super agers against ordinary older adults, a group whose hidden brain biology varies widely. His team instead used Alzheimer's patients of similar age as the contrast.
The team analysed resting-state functional MRI, which captures activity while someone rests motionless in a scanner, for 180 super agers and 110 people with Alzheimer's. The scans were drawn from the Alzheimer's Disease Neuroimaging Initiative, the Harvard Aging Brain Study and the Open Access Series of Imaging Studies, among other large projects. Differences in functional connectivity between the two groups produced what the researchers call the ARC score, short for Alzheimer's resilience-vulnerability connectome. It told the groups apart with roughly 85% accuracy.
What it predicted
Next, the team scored more than 1,300 participants whose ages ranged from 55 to 70 and who showed no cognitive impairment. At the time of scanning, higher and lower scorers did not clearly differ in cognition. Over later years, though, those with more Alzheimer's-like patterns were more likely to see their thinking skills slip, to build up amyloid-beta plaques and to receive an Alzheimer's diagnosis, over periods of as long as 12 years. Lead author Kanhao (Alex) Zhao, a postdoctoral scholar in psychiatry, said most follow-up lasted around five to six years.
People whose patterns sat closest to the super-ager end tended to lose ground most slowly. According to Stanford, the score outperformed genetics, demographics and established amyloid-beta and tau markers as a predictor of decline. Zhao said the model shows "comparable or better prediction ability" than existing biomarkers, at an earlier time point, and does better still when paired with them.
Limits and next steps
The researchers say the score is not a diagnostic test but a way to keep track of cognitive health. Zhang called any individual result a snapshot rather than destiny, noting that fewer people show resilient patterns as they get older, for instance after brain injuries or the onset of neuropsychiatric conditions.
Zhang hopes cognitive training or transcranial magnetic stimulation might one day push brain activity toward resilience. Neither has been tested against cognitive decline, and it is unknown whether altering these patterns would change how the disease progresses. His group is planning a trial that will compare people's scores ahead of and following a course of cognitive training.
Scientists from UCLA, NYU Grossman School of Medicine, Penn State College of Medicine, Lehigh University, the University of Texas at Austin and Children's National Hospital also took part. Funding came from the National Institutes of Health, the Rosenkranz Foundation-Knight Initiative, the Alzheimer's Association and Lehigh University.
