The Urinary Metabolomic Profile Correlates with Clinical Severity of Young Autistic Children.

Background: The identification of sensitive and specific biomarkers of Autism Spectrum Disorder (ASD) represents one of the main challenges. In this regard, metabolomics has the potential to identify a characteristic ASD metabolic fingerprint which may support the clinician for an earlier diagnosis. New evidence shows a different urinary metabolic profile in autistic children, characterized by altered levels of gut-microbial cometabolites, tryptophan, homocitrulline, quinic, xanthurenic citric and puric acid. However, to our knowledge only few studies have correlated the urinary metabolomics findings with clinical features and the severity of autism.

Objectives: The aim of our study was to characterize the urinary metabolic signature of ASD children compared to age matched healthy controls and to investigate the correlation between the clinical phenotype severity and the urinary metabolomic profile.

Methods: 57 children (age range 2–11 years) were enrolled in the study: 31 ASD patients and 26 typically developing children without familiarity for ASD as healthy controls. Standardized clinical assessment (Autism Diagnostic Observation Schedule, Second Edition- ADOS-2; Aberrant Behaviour Checklist - ABC) was performed for ASD children. Morning urine samples were collected for all participants and urinary metabolites were analysed and quantified by Gas chromatography–mass spectrometry (GC-MS). Correlations between the ASD metabolic fingerprint and psychological tests (ADOS-2; ABC) were investigated using the Projection to Latent Structures (PLS) method.

Results: A distinct urinary metabolomics profiles emerged in the ASD children compared to healthy controls (OPLS-DA R2 = 0.766; Q2 = 0.479 P < 0.001). The purine, phenylalanine and tyrosine metabolism were the most significantly altered pathways.Different levels of gut microbial metabolites, such as hippurate, phenylalanine, tyrosine, indoleacetic, aminomalonic acids, N-acetylglucosamine and a dysregulation of metabolites expression of oxidative stress (tryptophan, cystine, uric acid) were found in ASD population. Moreover, higher levels of para-cresol and adipic acid, possible indirect results of environmental exposure, were detected in children with ASD compared to healthy controls. Finally, a positive correlation emerged between the urinary levels of para-cresol, adipic acid and the clinical severity in terms of both core symptoms (ADOS: R2=0,890273) and aberrant behaviours (ABC: R2=0,970057).

Conclusions: We report a significant difference in the urinary metabolomic profile of young idiopathic autistic children compared to healthy subjects. Our findings confirm a possible associative relationship between the autistic urinary metabolomic profile, gastrointestinal dysbiosis and oxidative stress. We also found a positive clinical correlation between core symptoms, aberrant behaviours and high concentrations of p-cresol and adipic acid whose alterations can be expression of environmental exposure. Thus, a distinct metabolomic profile related with a more severe clinical outcome emerged from our study.