Blood‐based markers of neurodegeneration linked with brain atrophy and cognition in aging
Grégoria Kalpouzos, Filip Magnusson, Jonatan Gustavsson, Farshad Falahati, Goran Papenberg, Francesca Mangialasche, Davide Liborio Vetrano- Psychiatry and Mental health
- Cellular and Molecular Neuroscience
- Geriatrics and Gerontology
- Neurology (clinical)
- Developmental Neuroscience
- Health Policy
- Epidemiology
Abstract
Background
Neurofilament light chain (NfL) and glial fibrillary acidic protein (GFAP) have been linked to aging‐related brain tissue loss and cognitive decline, both related to neurodegeneration. Recently, NfL and GFAP have become quantifiable in blood; little is known about their association with brain structure in aging adults.
Method
Single‐Molecule Array (SIMOA) technology was used for ultra‐sensitive detection and quantification of NfL and GFAP in blood plasma at baseline in healthy volunteers (N = 102, 52 female, 52‐79 years old). These individuals also underwent magnetic resonance imaging twice, 3 years apart (N = 72 at follow‐up), and cognitive testing. We examined the relationships among NfL, GFAP, age, brain atrophy parameters (gray‐matter volume ‐ GMV, cortical thickness ‐ CT) and cognitive performance (composite score of episodic and working memory, executive functions and perceptual speed). The relationships between blood markers, GMV and CT were assessed using voxel‐ and surface‐based morphometry, cross‐sectionally and longitudinally, adjusting for age, sex and education.
Result
Older age was related to higher levels of NfL and GFAP (ps < .001; Figure 1). Cross‐sectionally, older age was related to lower GMV in the hippocampi and lateral middle temporal regions, and lower CT in lateral frontoparietal areas. Longitudinally, atrophy was located in fronto‐parieto‐temporal regions, more strongly in the left hemisphere (all ps < .05 Family‐Wise Error corrected). Controlling for age, sex, education and total intracranial volume, higher NfL levels were associated with lower volume in inferior temporal regions (cross‐sectionally), and with atrophy in right hippocampus (longitudinally) (p < .001 uncorrected; Figure 2). Higher GFAP levels were associated with lower volume and thinner cortex in frontoparietal regions, and atrophy in right frontal cortex (p < .001; Figure 2). After splitting the individuals as decliners and non‐decliners based on their cognitive‐performance changes, it was found that the negative associations between NfL and inferior‐temporal volume, and between GFAP and fronto‐temporal volume, were driven by the decliners (Figure 3).
Conclusion
Higher NfL or GFAP levels are related to atrophy, beyond age, in key regions known to be affected in neurodegenerative disorders, possibly indicating early neuropathological changes.