Abstract
Background
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder requiring early and accurate diagnosis. Conventional diagnostic methods such as cerebrospinal fluid (CSF) analysis and neuroimaging are invasive, costly, and not easily accessible. Saliva has emerged as a promising non-invasive alternative for biomarker detection.
Objective
To systematically review current evidence regarding salivary biomarkers for the diagnosis of Alzheimer’s disease.
Methods
A systematic review was conducted in accordance with PRISMA guidelines. Electronic databases including PubMed, Scopus, Web of Science, and Google Scholar were searched for studies published up to March 2025. Studies investigating salivary biomarkers in human participants with Alzheimer’s disease were included. Study selection and data extraction were independently performed by two reviewers. Due to heterogeneity in biomarker types, analytical methods, and study designs, a qualitative synthesis was conducted instead of meta-analysis.
Results
A total of 12 studies met the inclusion criteria. Identified salivary biomarkers included amyloid-beta (Aβ), tau proteins, microRNAs, lactoferrin, neurotransmitter metabolites, oxidative stress markers, and multi-omics signatures. Lactoferrin demonstrated particularly high diagnostic accuracy (AUC ~0.95), while salivary microRNA-485-3p and multi-omics approaches achieved AUC values approaching 0.92–0.96. Advanced Raman spectroscopy combined with machine learning demonstrated diagnostic accuracy approaching 99%. However, some biomarkers such as salivary total tau and acetylcholinesterase showed limited or inconsistent associations with Alzheimer’s disease.
Conclusion
Salivary biomarkers demonstrate significant potential as non-invasive tools for Alzheimer’s disease diagnosis. However, variability across studies and lack of methodological standardization currently limit clinical applicability. Further large-scale longitudinal studies are required to validate biomarker performance and establish standardized protocols.