6th Edition of Neurology World Conference 2026

Abstract

Introduction:
Alzheimer’s disease (AD), the most common form of dementia, is a complex polygenic disease characterized by the accumulation of Amyloid-β (Aβ42) and hyperphosphorylated Tau 181 (p-tau181) proteins in the brain. An increased level of these proteins, among others, in the AD brains and cerebrospinal fluid (CSF) have been detected years before the symptoms of AD appear. Therefore, studying AD proteome in CSF can reflect its diverse underlying pathophysiology and pave the way for reliable diagnostic and therapeutic advancements. However, a relatively small sample size of existing studies and use of less throughput protein measurement assays that can only detect a limited range of protein analytes have been significantly hindering the diagnostic and prognostic potential of AD CSF proteome. 

Methods:
Here, we present one of the largest AD proteomic profiles, based on 7,029 protein analytes measured in CSF of a total 3,065 individuals in a three-stage study. Firstly, discovery was performed in 836 samples from the Knight Alzheimer Disease Research Center (Knight ADRC) and 618 samples from the Fundació ACE Alzheimer Center Barcelona (FACE) discovery cohorts using the ATN framework (AT- = 680 and AT+ = 490). Secondly, the proteins that passed multiple test correction in these dataset were further tested in 832 individuals (AT- = 235 and AT+ = 358) from the Alzheimer disease neuroimaging initiative (ADNI) and Barcelona-1 replication cohorts. Lastly, the proteins that passed multiple Bonferroni correction on the meta-analysis were further utilized for creating AD prediction models and conducting pathway enrichment analysis to gain mechanistic insights into AD pathophysiology.

Results:
The differential abundance analysis in discovery cohorts identified 3,565 proteins to be significantly (FDR < 0.05) altered. Of these, 2,543 also passed FDR in the replication datasets and had the same direction of effect. Finally, by meta-analyzing discovery and replication findings, we identified 2,233 proteins to be significantly (P-Bonferroni < 0.05) altered in AD CSF proteome. Some of the highly significant proteins included YWHAG (P-Bonf < 4.5×10-219), SMOC1 (P-Bonf < 5.8×10-208), PPP3R1 (P-Bonf < 9.2×10-164), NRGN (P-Bonf < 3.2×10-119), and NEFL (P-Bonf < 1.8×10-37). By using lasso regression, we identified a 39 proteins signature with high predictor power (Discovery AUC = 1.0; Replication AUC = 0.99) representing a robust and precise AD diagnostic biomarker. Disease and pathway enrichment analysis highlighted several neurological disorders (e.g. AD, tauopathy, synucleinopathy, and motor neuron disease) and neuronal functions (neuron projection morphogenesis, synapse assembly and organization, axonogenesis, and neuron differentiation) to be significantly enriched in the altered AD CSF proteome.
 
Conclusion:
Our findings show the promising potential of AD CSF proteomics in the development of reliable and robust AD prediction model. The employed systematic analysis of aptamer-based proteomic data revealed differentially abundant proteins and various biological pathways that are compromised in AD, thereby, increasing our understanding of the AD biology. To conclude, our findings may accelerate the development of effective intervention therapies that target the earliest molecular triggers of AD.