Subtle changes in how different parts of the brain communicate can predict the likelihood of developing Alzheimer's disease years before memory lapses appear, according to a study published in Science Advances. Stanford Medicine researchers utilized brain scans and a computer model to identify early indicators of cognitive decline.
The study compared resting brain scans from patients diagnosed with Alzheimer's disease and "super agers"—older individuals who maintain memory capabilities comparable to much younger people. Instead of focusing on physical markers like amyloid plaques or brain shrinkage, the researchers examined the communication patterns between various brain regions.
By comparing these activity patterns, they developed a measurement called the Alzheimer's resilience-vulnerability connectome score, or ARC score. This score can differentiate between super agers and individuals with Alzheimer's with approximately 85% accuracy.
"I'm excited we established a signature that can help us better understand the difference between people who can maintain excellent cognitive performance and people who show cognitive decline," said Yu Zhang, the study's senior author and an assistant professor of psychiatry and behavioral sciences at Stanford Medicine, in a press release.
The research team then applied this scoring model to over 1,300 cognitively unimpaired adults aged 55 to 70. While all participants scored normally on memory tests, their brain scans revealed differing activity patterns. Some resembled those of super agers, while others were more similar to patterns observed in individuals with Alzheimer's.
Over follow-up periods of up to 12 years, participants whose brain activity patterns more closely aligned with Alzheimer's profiles exhibited faster declines in memory and thinking. These individuals were also more likely to develop amyloid plaques and eventually receive an Alzheimer's diagnosis, the researchers found.
The ARC score demonstrated comparable or superior predictive ability compared to existing risk measures, including amyloid and tau biomarkers, as well as genetic testing. "Our model shows comparable or better prediction ability, and it can do so at an earlier time point. And if we combine our score with the existing biomarkers, the prediction performance improves," stated lead author Kanhao Zhao, a postdoctoral scholar in psychiatry at Stanford Medicine.
Alzheimer's disease affects millions worldwide, with a significant challenge being that neurological damage often begins decades before symptoms manifest, limiting the effectiveness of treatments at later stages. The researchers emphasized that the brain scan method is not yet ready for standalone clinical diagnosis and requires further studies to confirm its accuracy and compare it with other emerging tests.
However, understanding these differences in brain activity could be crucial for developing and testing future treatments, including non-invasive brain stimulation techniques. "There are two main components: One is to know which part of the brain to target, the other is how we should change it," Zhang explained. "That's why identifying the super-ager brain resilience pattern is a critical component — so we know what a good pattern should be."