CSF and Plasma Analytes Reveal Early Clues to Alzheimer’s Development

Specific inflammatory proteins — detectable in cerebrospinal fluid (CSF) and, to a lesser extent, in plasma — may be indicative of future cognitive decline in people who are at risk for Alzheimer’s disease (AD) but lack symptoms. So found a large proteomic analysis that quantified nearly 3,000 analytes from plasma and CSF over a two-year period from 118 older adults.

The results identified potential biomarkers of future cognitive decline as well as important clues to early mechanisms of AD progression. The study was recently published online in Annals of Clinical and Translational Neurology.

“This study provides a window into central nervous system changes that often occur in individuals who have the APOE Ɛ4variant or who are positive for the presence of amyloid,” says first and corresponding author Jagan Pillai, MD, PhD, a behavioral neurologist who directs Cleveland Clinic’s Center for Brain Health in Cleveland. “Accumulating AD pathology appears to shape inflammatory responses, which are detectable in plasma and CSF before the onset of cognitive impairment.”

The quest for markers for preclinical intervention

Starting treatment for AD before symptoms arise requires accurate biomarkers that reliably indicate those at high risk for future cognitive decline. Increasing evidence indicates that pathophysiological processes beyond amyloid and tau pathology are involved in driving the rate of cognitive decline.

Considerable interest has arisen in the role of neuroinflammatory processes, with investigations focusing on possible related markers in plasma and CSF. However, studies to date provide only limited information on the complex neuroinflammatory changes activated concurrently in CSF and plasma in preclinical AD.

This study was designed to further characterize AD development by longitudinally following a cohort at high risk for developing AD and by integrating CSF and plasma proteomics, cognitive evaluations and imaging studies that quantify amyloid burden. Study goals were to identify predictive markers of cognitive decline in blood and CSF and to gain a better understanding of underlying processes.

Study design and findings

The cohort consisted of 118 cognitively normal participants aged 65 to 80 years. Individuals at high risk of developing cognitive decline were included (54% had the APOE Ɛ4 variant, and 26% were diagnosed with preclinical AD based on amyloid positivity).

Proteomic analyses were performed at baseline and at two-year follow-up on 1,331 analytes in CSF and 1,501 analytes in plasma.

Data analysis was conducted on participants who had datasets for both time points (CSF, n = 76 participants; plasma, n = 82). Key results included the following:

• 48 follow-up-to-baseline analyte ratios in CSF (3.6%) and one in plasma were significantly associated with amyloid status (positive or negative).
• 350 follow-up-to-baseline analyte ratios in CSF (26.3%) and 91 in plasma (6.1%) were significantly associated with the APOE Ɛ4 allele.
• Among the 350 significantly associated analytes in CSF, 44 were related to both APOE Ɛ4 and amyloid positivity.
• 12 analytes in CSF and 25 analytes in plasma were significantly associated with cognitive decline over two-year follow-up.

Potential biomarkers and mechanisms of cognitive decline

Dr. Pillai highlights several key implications of the study results.

APOE Ɛ4 and amyloid pathologic pathways likely converge

The finding that participants with the APOE Ɛ4 variant and those with amyloid positivity shared 44 upregulated CSF analytes suggests a convergence of pathologic molecular pathways. Multivariable modeling revealed that some were central hub analytes, indicating the presence of coordinated biological pathways rather than isolated protein changes. Hub analyte changes over the two years likely reflected remodeling of these pathways driven by APOE Ɛ4 and amyloid pathology. Implicated pathways were related to axon guidance, synaptic organization and extracellular matrix signaling.

Inflammatory analytes are linked to cognitive decline

Several analytes related to longitudinal cognitive decline converged on inflammatory signaling pathways. Eotaxin-1 — a chemokine that attracts eosinophils and is a signal of CSF inflammation — was most predictive of longitudinal cognitive decline, especially in women. It was present in both CSF and plasma and proved to be the most consistent marker among different assay platforms and analytic methods. “This makes eotaxin-1 a particularly good candidate for a biomarker of future cognitive decline in individuals with normal cognition who are at risk for AD,” Dr. Pillai notes.

CSF provides more information than plasma

Overall, CSF proteomics yielded stronger APOE Ɛ4 and amyloid signatures than did plasma. The authors speculate that plasma is less reflective of pathways that lead to clinical AD because it is farther removed from central APOE Ɛ4 and amyloid processes and is more prone to incidental inflammatory responses generated peripherally.

A basis for future research

Dr. Pillai emphasizes that the findings of this research should be viewed as hypothesis-generating. He adds that the research design, integrating a variety of assessment modalities and analytic methods over a two-year period, makes the results particularly robust.

“Our findings warrant further investigation with larger cohorts and longer longitudinal follow-up,” he concludes. “A better understanding of early AD processes and the ability to identify AD before it is clinically apparent may be keys to more effectively tackling this disease.”