Characterization of the endophenotypes and identification of new biomarkers in the Autism Spectrum Disorder

Background: Autism Spectrum Disorder (ASD) is a highly heterogeneous neurodevelopmental condition, characterized by wide variability in clinical presentation, developmental trajectories, and underlying biological mechanisms. This heterogeneity poses major challenges for diagnosis and stratification and suggests the involvement of complex interactions among genetic, epigenetic, metabolic, immune, and neurobiological factors. Rationale: The identification of biologically meaningful endophenotypes may provide a refined framework for characterizing ASD beyond categorical diagnostic definitions. Integrative approaches combining molecular, peripheral, neurophysiological, and clinical data may support the identification of coherent biological patterns associated with specific developmental and behavioural profiles. Methods: This thesis comprises four experimental studies. Study I examines genetic variants involved in oxidative stress regulation and xenobiotic metabolism. Study II investigates genome- wide DNA methylation profiles. Study III integrates clinical and neuropsychological characterization with peripheral biochemical, metabolic, inflammatory, and neurotrophic biomarkers. Study IV provides an exploratory analysis of structural brain MRI and electroencephalographic (EEG) features in relation to clinical and neurodevelopmental characteristics. Results: Genetic variants related to redox regulation are associated with ASD, alongside alterations in oxidative stress markers. Epigenetic analyses reveal diMerential methylation patterns involving immune-related and neurodevelopmental pathways. Peripheral biomarker analysis highlights inter-individual variability and identifies associations with developmental and behavioural features. Neuroimaging and EEG findings are described in relation to clinical and neurodevelopmental profiles. Conclusions: Overall, this work supports a multidimensional view of ASD, in which the integration of genetic, epigenetic, peripheral biological, neuroimaging, neurophysiological, and clinical data contributes to the identification of endophenotypic dimensions and improved stratification of the disorder.