Magnetic resonance imaging (MRI)–derived brain age estimates vary substantially between individuals. The Brain Age Gap Estimation (BrainAGE) metric quantifies the difference between an individual’s MRI-predicted brain age and chronological age. The authors investigated whether early deviations in structural brain ageing, measured as elevated BrainAGE, precede future cognitive impairment and whether BrainAGE adds predictive value beyond conventional demographic, genetic, cognitive, and MRI measures.
The study addresses two linked questions: do cognitively normal people who later develop mild cognitive impairment (MCI) or dementia show higher BrainAGE before clinical onset, and does BrainAGE improve prediction of future cognitive decline when combined with established predictors?
The primary analyses used longitudinal structural MRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Key findings were replicated in the Kuopio Osteoporosis Risk Factor and Prevention Study (OSTPRE), a population-based cohort of women in Eastern Finland. The OSTPRE study has ethical approval from the Ethics Committee of Kuopio University Hospital and obtained required Finnish data permissions; participant consent and adherence to ethical guidelines were reported in the source. The article is presented as a medRxiv preprint and has not been peer reviewed.
The authors applied BrainAGE estimation to structural MRI scans to quantify deviations from expected brain ageing. Longitudinal MRI allowed assessment of both cross-sectional BrainAGE differences and within-person changes over time. Analyses compared individuals who remained cognitively stable with those who later converted to MCI or dementia, examining BrainAGE values several years prior to diagnosis and trajectories of BrainAGE change.
Predictive models incorporated demographic variables, APOE4 status, cognitive measures, and conventional MRI-derived structural measures alongside BrainAGE to determine whether BrainAGE adds complementary prognostic information.
In the ADNI sample, participants who later converted to MCI or dementia had higher BrainAGE values several years before their clinical diagnosis compared with those who remained stable. Longitudinal analyses within ADNI showed that converters not only had elevated BrainAGE at earlier time points but also exhibited steeper increases in BrainAGE over time, indicating accelerating structural brain ageing before clinical onset.
These observations suggest that detectable structural divergence begins years in advance of symptomatic MCI and that BrainAGE captures progressive change in brain structure preceding diagnosis.
The principal cross-sectional differences in BrainAGE between future converters and stable participants, and the association between higher BrainAGE and increased risk of future conversion, were replicated in the OSTPRE cohort. Replication in a population-based sample of women supports the generalizability of the core findings beyond the ADNI research cohort.
The source did not report additional cohort-specific methodological details beyond stating that OSTPRE served as a replication sample and that appropriate ethical approvals and data permissions were in place.
Across cohorts, elevated BrainAGE values were associated with an increased risk of future conversion from cognitive normality to MCI, and higher BrainAGE predicted faster subsequent memory decline. These associations indicate that BrainAGE relates both to diagnostic progression and to domain-specific cognitive trajectories.
The source emphasizes that BrainAGE identifies early structural aging differences in individuals who are still cognitively normal at the time of MRI, highlighting its potential for early risk stratification.
When added to predictive models that already included demographic factors, APOE4 genotype, baseline cognitive performance, and conventional MRI-derived structural measures, BrainAGE consistently improved prediction of future cognitive outcomes. The improvement was observed across analyses, with the largest incremental benefit seen for individuals who converted after longer follow-up intervals, implying that BrainAGE is particularly informative for identifying risk years before clinical onset.
The authors interpret these results as evidence that BrainAGE captures facets of early structural brain ageing that are not fully accounted for by standard volumetric or morphometric MRI measures, thereby providing complementary prognostic information.
The findings indicate that structural brain ageing, as measured by BrainAGE, begins to diverge years before MCI onset and that BrainAGE can improve early identification of cognitively normal individuals at elevated risk of future cognitive decline when integrated with other biomarkers.
Limitations and caveats noted in the source include that the report is a preprint and has not been peer reviewed; therefore, the results should not yet guide clinical practice. The source provided no additional numerical effect sizes, confidence intervals, or detailed model performance metrics in the abstract. Specific methodological parameters, such as the BrainAGE algorithm implementation details, exact follow-up intervals, and quantitative improvement in prediction metrics, were not reported in the abstract and would require consultation of the full preprint.
Overall, the study presents MRI-derived BrainAGE as a promising complementary marker for early structural brain ageing that precedes MCI and may enhance prediction of future cognitive decline when used alongside demographic, genetic, cognitive, and conventional MRI measures.