Biomarkers in Autism Spectrum Disorder: From Traditional Constraints to AI-Driven Horizons in Precision Medicine

Objective: The aim of this review is to comprehensively evaluate the potential of biomarkers in the diagnosis, prognosis, and individualized treatment of Autism Spectrum Disorder (ASD), considering its complex nature and current diagnostic limitations. Methods: The article provides an updated literature review focusing on blood- and urine-based biomarkers (oxidative stress, inflammation, neurotransmitters, microbiome), neuroimaging, genetic, and physiological markers. It also examines key challenges, ethical considerations, and promising future directions such as artificial intelligence (AI)-assisted multi-omics data integration. Results: Biomarkers measured in blood and urine (e.g., isoprostanes, 8-OHdG, and inflammatory cytokines) highlight the role of oxidative stress and chronic inflammation in ASD pathophysiology. Neuroimaging and genetic markers show promise for early risk identification and biological subtyping. However, most current studies suffer from small sample sizes, replication issues, and lack of standardization, limiting their clinical applicability. Conclusion: Biomarker-based approaches hold promise for making ASD diagnosis more objective and for facilitating earlier intervention by reducing diagnostic delays. In the future, AI-driven multi-omics integration may provide deeper insights into ASD heterogeneity and support the development of personalized treatment strategies. Managing ethical concerns such as privacy and discrimination through a neurodiversity-oriented perspective will be crucial in this process.