J Transl Med. 2026 Jul 15. doi: 10.1186/s12967-026-08562-8. Online ahead of print.
ABSTRACT
Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative syndrome diagnosed clinically using standardized criteria, with neuropathological confirmation of motor neuron loss and TDP-43 aggregates in postmortem brain tissue. Extracellular vesicles (EVs) have emerged as potential minimally invasive biomarkers for ALS, but studies vary widely in methodology and reproducibility. We conducted a systematic review and meta-analysis to evaluate the diagnostic potential of EV-associated proteins and RNAs in ALS. Following PRISMA guidelines, we searched PubMed and EMBASE from inception to May 21st, 2026. Forty-one studies met inclusion criteria. Where published summary statistics were available, these were used directly; where they were not, data were reconstructed from figures or obtained from authors and re-analyzed to derive standardized effect sizes and exploratory diagnostic accuracy estimates. Random-effects models were used for continuous outcomes, and diagnostic accuracy was assessed using hierarchical summary ROC and bivariate random-effects models. Publication bias was evaluated using Begg, Egger, and funnel plots. EV-associated TDP-43 was the most frequently studied protein. Meta-analysis of five studies showed a moderate but non-significant increase in EVs from ALS vs. controls (SMD = 1.30) with high heterogeneity (I = 97.8%). Sixteen studies assessing EV-RNA biomarkers showed minimal overlap and limited independent replication. Diagnostic accuracy meta-analysis across 11 studies yielded moderate performance (AUC = 0.839). No publication bias was found across both meta-analyses. EV biomarkers for ALS show biological promise but are limited by methodological variability and insufficient replication. This work highlights the need for standardized protocols, transparent data sharing, and independent validation.
PMID:42458453 | DOI:10.1186/s12967-026-08562-8
