Blood aging measures linked to cognitive scores
An observational study of 570 people found links to cognitive scores but no clear link to decline over an average of 1.7 years.
Limit: Follow-up averaged 1.7 years, limiting conclusions about longer-term cognitive changes
Edited by Boris Djordjevic · AgingResearch.News
An observational analysis followed 570 participants in the Parkinson’s Progression Markers Initiative, whose average age was 64. Women made up 59% of the sample; 96% were White. Participants averaged 16 years of education and generally had preserved cognitive function at entry.
Researchers calculated seven aging measures from DNA methylation, chemical marks on DNA, in blood samples. They compared these measures with repeated scores on a 30-point cognitive screening test over an average of 1.7 years. Higher aging measures were associated with lower overall scores across all seven measures. The links mainly reflected differences between people, rather than changes within the same person. After accounting for multiple comparisons, the analyses found no clear link between changes in the aging measures and cognitive scores, or between starting aging measures and the rate of cognitive decline.
Why it matters
Cognitive impairment is a major source of disability in Parkinsonian disorders. Distinguishing markers of current cognitive status from markers of change over time matters when evaluating biological aging measures as potential tools for assessing cognition.
Caveats
This observational study cannot establish cause and effect, and its short follow-up and screening test limit conclusions about longer-term or more detailed cognitive changes. The predominantly White, highly educated sample had generally preserved cognition; the preprint has not been peer-reviewed.
- DNA methylation
- Cognitive impairment
- Parkinsonian disorders
- Epigenetic aging measures
- Montreal Cognitive Assessment
- Humans
The paper
Hospital for Sick Children; Rush University Medical Center
medRxiv, May 13, 2026, Open access, Preprint, not peer-reviewed
