Multimodal quantitative MRI revealed distinct cross-sectional age-related biophysical alteration patterns (early, late, and inverted-U) across the adult lifespan, with subcortical and thalamic regions as key contributors to age estimation, and transcriptomic integration linking these patterns to neurodevelopmental and neurodegenerative gene expression signatures.
Key Findings
Results
Three distinct patterns of age-related biophysical alterations were identified in the human brain using quantitative MRI metrics.
The three patterns identified were: early, late, and inverted-U patterns of cross-sectional age-related change.
Three qMRI metrics were employed: quantitative susceptibility mapping (QSM), longitudinal relaxation rate (R1), and effective transverse relaxation rate (R2*).
Both linear and nonlinear modeling approaches were applied to characterize these patterns across the adult lifespan.
The patterns were observed across the adult lifespan, indicating different brain regions or tissue types follow distinct aging trajectories.
Results
Subcortical and thalamic regions were identified as key contributors to brain age estimation in predictive modeling.
Predictive modeling was used to identify age-sensitive imaging features from the multimodal qMRI data.
Subcortical and thalamic regions showed particular importance as contributors to accurate age estimation.
These regions were identified as candidate age-sensitive imaging features that warrant further validation.
The predictive modeling leveraged all three qMRI metrics (QSM, R1, and R2*) to estimate brain age.
Results
qMRI-derived age-related patterns spatially co-localize with gene expression signatures enriched in neurodevelopmental and neurodegenerative pathways.
Transcriptomic data were integrated with imaging-derived age-related patterns to identify molecular underpinnings.
The imaging-derived patterns showed spatial co-localization with specific gene expression signatures.
Gene expression signatures were enriched in both neurodevelopmental and neurodegenerative pathways.
This transcriptomic integration provides potential molecular bases for the observed biophysical alterations detected by qMRI.
Methods
Multimodal qMRI was used to non-invasively investigate brain tissue biophysical properties across the adult lifespan.
QSM, R1, and R2* were chosen as complementary quantitative metrics reflecting different tissue properties.
The study employed a cross-sectional design to characterize age-related changes.
The approach provides non-invasive insight into brain tissue properties that may reflect early markers of neurodegenerative disease.
The study advances understanding of biophysical alterations and their molecular bases, which were described as 'poorly understood' prior to this work.