AAIC 2023 Poster: A novel automated multiplex immunoassay for the ultrasensitive detection of blood-based biomarkers for neurodegenerative diseases


Background: Efforts to identify accurate blood biomarkers for neurodegenerative disease hallmarks have been hampered by the lack of a proteomic tool that has the required sensitivity to detect very low concentrations of brain-derived proteins in plasma or serum and the ability to multiplex many analytes in a single assay. Here we evaluated NULISA™, a recently developed novel immunoassay with attomolar level sensitivity and high multiplex capability1, for its ability to detect serum biomarkers associated with Alzheimer’s disease.

Methods: Serum samples from 31 Alzheimer’s disease (AD) patients and 31 non-AD controls were analyzed using NULISA with a 200plex inflammation panel targeting a broad spectrum of inflammation-related cytokines, chemokines and other proteins. Five samples failed quality control and were excluded from further analysis, resulting in 28 AD and 29 control samples in the final dataset. Linear model analysis, including age and sex as covariates, was performed for each target for differential expression between AD patients and controls.

Results: Two significant differentially abundant proteins were identified. The level of glial fibrillary acidic protein (GFAP) was 1.84-fold higher in AD patients (adjusted p-value = 0.00014), whereas the level of S100 calcium-binding protein A12 (S100A12) was 2-fold higher in non-AD controls (adjusted p-value = 0.031). GFAP serum levels correlated strongly with total tau (Spearman rho = 0.69, p=1.7e-09), p-tau181 (rho=0.69, p=1.7e-09) and amyloid beta 42 (rho= -0.71, p=4.7e-10) levels in cerebrospinal fluid (CSF). S100A12 levels also correlated significantly with total tau (rho = -0.39, p=0.0031), p-tau181 (rho=-0.36, p=0.0053) and amyloid beta 42 (rho= 0.38, p=0.0067) CSF levels. Receiver operating characteristic analysis demonstrated that GFAP strongly discriminated AD from controls with an area under the curve (AUC) of 0.93 (95% CI 0.87-1.0), and S100A12 also exhibited good discrimination with an AUC of 0.72 (95% CI 0.59-0.86).

Conclusion:  This exploratory study identified serum GFAP levels as a strong discriminatory biomarker for AD, consistent with previous studies. S100A12 also demonstrated significant associations with AD and CSF biomarkers and thus warrants further study. NULISA holds great promise for the discovery and validation of blood-based biomarkers for the early detection and monitoring of neurodegenerative diseases.