Blood–CSF MiRNA Signatures in Alzheimer’s Disease Reveal Novel Biomarker Candidates: A Systematic Review

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline. MicroRNAs (miRNAs) have emerged as promising regulators of AD-related pathways and potential minimally invasive biomarkers. However, existing meta-analyses have yielded inconsistent results, and whether miRNA signatures are conserved across biological fluids remains unclear. The present work aimed to systematically review studies performing miRNA discovery or multi-candidate profiling in blood and CSF during the last 15 years, in order to identify miRNAs showing significant and consistent differential expression across independent studies. By systematically comparing the direction and consistency of miRNA dysregulation across independent studies and biological sources, we identified 80 miRNAs consistently dysregulated in both blood and CSF, revealing a reproducible cross-fluid signature of AD-associated molecular alterations. Notably, miR-451a and miR-16-5p emerged among the most consistently dysregulated miRNAs across fluids. In contrast, several miRNAs displayed divergent or compartment-specific patterns, highlighting that blood and CSF may capture distinct aspects of AD-related molecular changes. Among these, miR-484, miR-3200-3p and miR-598 were identified as novel CSF candidates, while miR-20a-5p, miR-363-3p and miR-5006-3p represented novel blood-specific candidates. Furthermore, a substantial proportion of detected miRNAs was associated with extracellular vesicle/exosomal fractions, supporting their potential stability and biological relevance. Overall, our findings demonstrate that AD-associated miRNA dysregulation comprises both conserved and biofluid-specific signatures. From a clinical perspective, the current evidence primarily supports further evaluation of miRNAs as diagnostic/discriminative biomarkers rather than validated tools for early detection, disease monitoring, or prognosis. Their clinical translation remains limited by biofluid and methodological heterogeneity and requires harmonized cross-cohort and longitudinal validation.