High-Plex CSF proteomics reveals a distinct vascular- immune signature in β-amyloid-negative corticobasal syndrome

Background

The clinical phenotype of corticobasal syndrome can be caused by aggregation of various proteins in the brain, including 4-repeat tau in a substantial proportion of the patients. Reliable biomarkers for differentiating corticobasal syndrome with aggregation of 4-repeat tau from related parkinsonian disorders remain limited. We therefore aimed to (i) map syndrome-specific CSF proteomic signatures across various parkinsonian disorders, (ii) identify associated molecular pathways and (iii) explore protein biomarkers capable of differentiating β-amyloid negative corticobasal syndrome.

Methods

We quantified 127 CSF proteins using the Nucleic acid Linked Immuno-Sandwich Assay (NULISA) on a biomarker-enhanced cohort of clinically characterized patients with Parkinson’s disease, multiple system atrophy, progressive supranuclear palsy and corticobasal syndrome. Cohort inclusion required fulfillment of established clinical diagnostic criteria supported by complementary biomarkers indicative of the presumed underlying proteinopathy. Differential protein abundance was assessed using age- and sex-adjusted generalized linear models. Pathway-level alterations were examined using panel-aware gene set enrichment analysis. Biomarker candidates for distinguishing β-amyloid-negative corticobasal syndrome from other parkinsonian disorders were identified using elastic-net feature selection and validated by receiver operating characteristic analysis.

Results

β-amyloid-negative corticobasal syndrome exhibited the most pronounced proteomic alterations, characterized by widespread increases in protein abundance and enrichment of inflammatory and vascular pathways, including IL-10 associated signaling when compared to controls. A core set of 37 proteins was consistently altered relative to other parrkinsonian syndromes, with a subset of proteins converging across multiple analytical approaches, including inflammatory mediators (CXCL8, IL-18, CX3CL1) and vascular/metabolic proteins (VEGFA, PGK1). These proteins also demonstrated high discriminatory potential, with an area under the curve above 0.85.

Conclusions

High-plex CSF proteomics suggests that β-amyloid-negative corticobasal syndrome is associated with a distinct vascular-immune molecular profile, indicating molecular differences from clinically similar parkinsonian disorders. These findings provide insight into biological processes linked to corticobasal syndrome and highlight candidate protein signatures that to our knowledge have not yet been systematically associated with β-amyloid-negative CBS and may support molecular stratification in parkinsonian disorders.