Vitality Watch

Dossier pubmed observational 11-dec-2025
Human cohort, case-control, or epidemiological association

Your brain doesn’t look a day past 70! Cross-sectional associations with brain-predicted age in the cognitively-intact oldest-old.

Key claim: Cognitively intact adults aged 85–99 show substantially younger MRI brain-predicted ages on average (mean brainPAD ≈ −8 years), with only female sex among common risk factors linked to delayed brain aging and no association between brainPAD and cognitive performance.

Abstract

The cognitively-intact oldest-old (85 +) may be the most-resilient members of their birth cohort; due to survivorship effects (e.g., depletion of susceptibles), risk factors associated with brain aging biomarkers in younger samples may not generalize to the cognitively-intact oldest-old. We evaluated associations between established aging-related risk factors and brain-predicted age difference (brainPAD) in a cross-sectional cognitively-intact oldest-old sample. Additionally, we evaluated brainPAD-cognition associations to characterize brain maintenance vs. cognitive reserve in our sample. Oldest-old adults (N = 206; 85-99 years; Montreal Cognitive Assessment > 22 or neurologist evaluation) underwent T1-weighted MRI; brainPAD was generated with brainageR, such that more-positive brainPAD reflected more-advanced brain aging. Sex, education, alcohol and smoking history, exercise history, BMI, cardiovascular and metabolic disease history, and anticholinergic medication burden were self-reported. Global cognitive z-score and coefficient of variation were derived from the UDS 3.0 cognitive battery; crystallized-fluid discrepancy was derived from the NIH Toolbox Cognitive Battery. Mean brainPAD was -7.99 (SD: 5.37; range: -24.50, 6.03). Women showed more-delayed brain aging than men (B = -2.9, 95% CI = -4.6, -1.1, p = 0.002). No other exposures were significantly associated with brainPAD. BrainPAD was not associated with any cognitive variable. These findings suggest that cognitively-intact oldest-old adults may be atypically-resistant to risk factors associated with aging in younger samples, consistent with survivorship effects in aging. Furthermore, brainPAD may have limited explanatory value for cognitive performance in cognitively-intact oldest-old adults, potentially due to high cognitive reserve. Overall, our findings highlight the impact of survivorship effects on brain aging research.

Why this matters

Brain-age clocks are increasingly used as mechanistic readouts for aging interventions, but this cohort shows the clock’s risk-factor and cognition associations may not hold in the cognitively intact oldest-old — a group heavily shaped by survivorship. For the mechanisms area, it flags that ‘brain maintenance’ at extreme ages may be a qualitatively different state from midlife brain aging, not just a slower version of it.

Source

Link

Briefing card

Your brain doesn't look a day past 70! Cross-sectional associations with brain-predicted age in the cognitively-intact oldest-old.
Stage: observational · pubmed · 11-dec-2025

If you're over 85 and cognitively intact, a 'younger' MRI brain age in this group did not track with better cognitive performance.

Cognitively intact adults aged 85–99 show substantially younger MRI brain-predicted ages on average (mean brainPAD ≈ −8 years), with only female sex among common risk factors linked to delayed brain aging and no association between brainPAD and cognitive performance.

Cross-references (0)

None recorded — doctrine links and citations appear here as scans and citation sweeps find them.

External: doi:10.1007/s11357-025-02027-4

build build 880 · 2bcef25 · 2026-11-18