TL;DR
This study identifies blood-based protein biomarkers that can predict the progression of cognitive decline, neuropsychiatric symptoms, and brain changes over four years in an Asian memory clinic cohort. Key proteins like GFAP and NEFL show strong prognostic value, offering potential for early detection and targeted therapies for neurodegenerative diseases.
What Problem Does This Paper Address?
Diagnosing and predicting the progression of neurodegenerative diseases like Alzheimer's is challenging. Current methods often rely on expensive and invasive techniques like PET scans or lumbar punctures. While blood-based biomarkers are highly desirable for their accessibility and cost-effectiveness, their prognostic power for various aspects of disease progression (cognition, behavior, and brain structure) in diverse populations, particularly in regions with a high prevalence of cerebrovascular disease, remains underexplored. This paper aims to fill this gap by investigating a comprehensive panel of plasma proteins to predict longitudinal changes in cognitive function, neuropsychiatric symptoms, and neuroimaging markers in a Singaporean cohort.
Methods
The researchers employed a prospective, longitudinal study design, following 528 attendees of a Singaporean memory clinic for four years. They collected baseline plasma samples and performed annual neurocognitive and neuropsychiatric assessments. Brain MRI scans were conducted every two years to monitor structural changes.
The core of their analysis involved profiling 1536 plasma proteins using the Olink Explore 1536 platform. This platform allows for the simultaneous measurement of a broad range of proteins covering major biological pathways, including neurological, immune, cardiovascular, and inflammatory processes.
To identify prognostic biomarkers, they used multivariable regression models to assess the association between baseline protein levels and subsequent changes in:
- Cognitive decline: Defined by a significant increase in the Clinical Dementia Rating Scale Sum-of-Boxes (CDR-SB).
- Neuropsychiatric symptomatology progression: Defined by a significant increase in the Neuropsychiatric Inventory score.
- Neuroimaging outcomes: Specifically, the progression of cerebral small vessel disease (measured by white matter hyperintensity volume) and hippocampal atrophy.
All statistical analyses were corrected for the false discovery rate to ensure robust findings.
Datasets
The study utilized a biobank of plasma samples from 528 participants in a Singaporean memory clinic. This cohort was followed prospectively for four years, with longitudinal data collected on cognitive function, neuropsychiatric symptoms, and brain MRI. No new benchmark datasets were introduced; the study leverages existing clinical data and a well-characterized cohort.
Key Insights and Ideas
- Blood is a rich source for predicting disease trajectory: The study powerfully demonstrates that a broad panel of blood-based proteins can predict not just cognitive decline but also behavioral changes and structural brain alterations over time. This moves beyond single-biomarker approaches to a more holistic view.
- Common proteins link diverse outcomes: Proteins like GFAP and NEFL emerge as significant predictors across multiple domains – cognitive decline, hippocampal atrophy, and even domain-specific cognitive decline. This suggests these proteins might be involved in fundamental pathological processes affecting the brain.
- Neuropsychiatric symptoms have distinct protein signatures: While NEFL is a common player, proteins like DCBLD2 are specifically highlighted for their association with the progression of neuropsychiatric symptoms, indicating that behavioral changes might have unique underlying molecular mechanisms.
- Cerebrovascular disease context matters: The study's setting in Southeast Asia, with a high prevalence of cerebrovascular disease, adds crucial context. The identified biomarkers are relevant even in this complex clinical landscape, suggesting their utility in a broader population.
- Cell adhesion and signal transduction are key pathways: The top biological pathways implicated in both cognitive and neuropsychiatric progression are cell adhesion and signal transduction. This points to disruptions in how cells interact and communicate as central to neurodegeneration.
Key Results
- Cognitive Decline: Out of 528 subjects, 31.6% experienced incident cognitive decline. Among 245 proteins associated with this decline, GFAP and NEFL were the most potent predictors.
- Neuropsychiatric Symptomatology Progression: 55.68% of participants showed significant progression. NEFL and DCBLD2 were identified as the most prognostic proteins for this outcome, with 310 proteins showing significant associations.
- Neuroimaging Outcomes: NEFL and TNFRSF11B were strongly linked to cerebral small vessel disease progression. GFAP and NEFL were associated with hippocampal atrophy rates.
- Domain-Specific Decline: The top proteins (GFAP, NEFL, DCBLD2, TNFRSF11B) also showed significant associations with decline in multiple cognitive domains (global, executive function, language, memory, attention, visuomotor, visuoconstruction) over four years.
- Biological Pathways: Cell adhesion and signal transduction were identified as the top two biological pathways associated with both cognitive decline and neuropsychiatric symptom progression.
Limitations and Future Work
The authors acknowledge that these findings identify "late-stage protein biomarkers." This implies that while prognostic, they might not be suitable for very early detection. Further validation in larger, independent cohorts, particularly those with diverse ethnic backgrounds and varying degrees of cerebrovascular disease, is warranted.
Future work should focus on the mechanistic elucidation of these identified biomarkers. Understanding how GFAP, NEFL, DCBLD2, and TNFRSF11B contribute to disease progression is crucial for developing them into effective therapeutic targets. Investigating their interplay with other biological processes and their potential as diagnostic or monitoring tools in clinical practice is also a key next step.
Summary updated