Objective: To evaluate the performance of a multiplexed proteomic assay in a large multi-center Lewy Body dementia (LBD) cohort.

Background: LBD, which comprises Parkinson’s disease dementia (PDD) and Dementia with Lewy bodies (DLB), lacks established biofluid markers reflecting its complex molecular pathophysiology. The NULISAseq central nervous system (CNS) disease panel is a recently released commercial multiplexed proteomic assay that uses proximity extension technology to measure 120 pre-selected analytes associated with a variety of neurodegenerative mechanisms.

Method: We applied the NULISAseq CNS disease panel to 685 CSF samples from 476 unique individuals with clinical diagnoses of control, PD without cognitive impairment, PD with cognitive impairment, and DLB. A subset of individuals had alpha-synuclein seed amplification assay (αSyn-SAA) results and/or paired longitudinal CSF samples. Differential expression, correlation analyses, and machine learning were used to identify markers highly associated with disease diagnosis, SAA positivity, and cognitive progression. We also compared LBD abundance trends to those observed in a separate NULISAseq analysis of Alzheimer’s disease (AD) CSF samples.

Results: Of the 120 panel markers, nearly half (n=53) were significantly altered across one or more Lewy body diseases. DLB featured the most robust differential expression of the three disease groups compared to controls. Alpha-synuclein (SNCA) was not altered across disease, though it was among the 17 proteins significantly altered between SAA+ and SAA- DLB. Longitudinal analysis revealed four proteins (NEFL, NRGN, CCL26, CRH) featuring baseline levels that distinguished cases with stable versus declining cognition over the next two years. Machine learning identified a series of 10-analyte panels capable of discriminating Lewy body diagnoses and SAA positivity with high accuracy (AUC > 0.9). Synaptic dysfunction (NRGN, SNAP25, NPTX2), inflammation (CCL17, CCL3, CCL26), and cell metabolism (ENO2, KLK6, UCHL1) were highly represented in these diagnostic panels. Several of these markers were also among the 12 proteins that featured divergent abundance trends in DLB and AD.

Conclusion: Our results highlight the utility of NULISAseq multiplex proteomic analysis in the identification of diagnostic and prognostic CSF biomarkers of LBD that reflect its diverse molecular pathophysiology.

« Back to 2025 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/multiplex-proteomic-analysis-of-lewy-body-dementia-reveals-cerebrospinal-fluid-biomarkers-of-disease-pathophysiology-and-progression/