Brain Age: A Promising Biomarker for Understanding Aging in the Context of Cognitive Reserve

INTRODUCTION

Cognitive decline is a major concern in aging populations. Detecting it before clinical symptoms emerge remains a significant challenge. A precise, reliable, and non-invasive biomarker for cognitive health could revolutionize how we monitor normal aging and lifestyle impacts. Such a tool would not only identify individuals at risk of cognitive decline years before symptoms manifest but also aid in early interventions and treatments.

METHODS

Longitudinal MRI data from 121 high cognitive reserve (HCR) individuals were compared to matched low cognitive reserve (LCR) individuals to evaluate four biomarkers for early cognitive decline and disease progression: brain age delta, cortical thickness, AD cortical signature, and hippocampal volume. Cross-sectional analyses were conducted at baseline, alongside longitudinal assessments spanning 1 to 12 years, to compare the performance and properties of these biomarkers.

RESULTS

The brain age metric emerged as the most reliable biomarker, demonstrating a significant ability to differentiate between groups at baseline ( β = 1.250, t = 3.521, p = 0.0009; linear regression model; AUC = 0.73). Furthermore, this biomarker maintained its robustness as the strongest predictor of group membership over a follow-up period of up to 12 years ( β = 0.409, p = 0.025; mixed-effects model), underscoring its potential for longitudinal monitoring of cognitive decline.

DISCUSSION

The brain age biomarker demonstrates potential as an effective indicator for early cognitive decline, capable of detecting changes years before clinical symptoms appear and tracking age-related brain and cognitive changes over time. These findings suggest that integrating MRI biomarkers with machine learning approaches could yield more accurate and reliable tools for assessing cognitive health, surpassing the limitations of relying solely on MRI biomarkers.

Key Points

Question: Which T1-weighted MRI biomarkers are most effective in predicting longitudinal cognitive deterioration in the aging population?

Highlights

• Four commonly used MRI biomarkers were assessed in the context of cognitive aging.

• Brain age is validated as a promising biomarker for aging and cognitive reserve.

• Machine learning boosts cognitive biomarker accuracy beyond neuroimaging alone.

• The findings underscore a direct association between structural brain reserve and cognitive reserve.

• Key factors in brain preservation may support high cognitive reserve in aging.