Summary:
Autism spectrum disorder (ASD) is a neurodevelopmental condition characterised by difficulties in social communication alongside restricted and repetitive behaviours. ASD is considered multifactorial, with both genetic and environmental contributors, including prenatal exposures, dietary influences, and alterations in gut microbiota and metabolism. Dietary strategies have increasingly been explored in ASD, with research focusing on effects on mitochondrial function, oxidative stress, and neurotransmitter-related pathways. Interventions such as gluten- and casein-free diets, the keto diet and nutrient supplementation have been studied, however the mechanism is not fully understood. This study is a case-control study with an intervention component, which investigated metabolic differences between children with ASD and neurotypical controls and evaluated changes following a gluten-free modified ketogenic diet. Ten children with ASD and ten matched controls underwent plasma metabolomic profiling. Children with ASD then completed a three-month dietary intervention, after which metabolic profiles and behavioural severity were reassessed. At baseline, children with ASD showed a distinct metabolic pattern compared with controls, including higher levels of L-leucine and coumarin and lower levels of betaine. After the dietary intervention, L-leucine and coumarin levels decreased, while betaine increased slightly. Improvements were also observed in behavioural severity scores, with no adverse effects reported. Overall, this study suggests that children with ASD may have measurable differences in how their bodies process certain nutrients and chemicals. These differences are not fixed and may change after dietary intervention. In this study, changes in these biological markers were seen alongside changes in behaviour. However, because the sample size was small, the results cannot be confidently applied to all individuals with ASD. Larger studies are needed to confirm these findings.
Abstract:
Background: Gluten-free modified ketogenic diets (GF-MKD) have gained interest as adjunct nutritional interventions in autism spectrum disorder (ASD). However, evidence regarding their systemic metabolic effects in children, particularly from non-Western populations, remains limited. Methods: An untargeted plasma metabolomics analysis was performed using liquid chromatography–tandem mass spectrometry (LC–MS/MS) in 10 Indian children with ASD and 10 age- and sex-matched neurotypical controls. Multivariate and machine learning–based approaches were applied to identify metabolites distinguishing ASD from controls. Children with ASD subsequently underwent a three-month GF-MKD intervention, after which plasma metabolomic profiles and autism severity, assessed using the Childhood Autism Rating Scale (CARS), were re-evaluated. Results: At baseline, children with ASD exhibited a distinct plasma metabolomic signature characterized by elevated L-leucine, a marked increase in coumarin (~ 6-fold), and reduced betaine levels. This metabolic profile differentiated ASD from controls with high discriminative accuracy (AUC = 0.93). Pathway enrichment analyses indicated alterations in branched-chain amino acid metabolism and one-carbon metabolic pathways. Following GF-MKD intervention, plasma levels of L-leucine and coumarin decreased by approximately 46% and 60%, respectively, while betaine levels showed a modest increase. Clinically, participants demonstrated a significant reduction in CARS scores (median decrease: 4.5 points; p < 0.05), indicating improvement in autism-related behavioural symptoms. No diet-related adverse effects were observed. Conclusions: Indian children with ASD display a modifiable plasma metabolomic profile involving key amino acid and methyl-donor pathways. Modulation of these metabolic disturbances following GF-MKD intervention was accompanied by behavioural improvement. These findings support the potential role of targeted dietary strategies in ASD and highlight the need for larger, randomized controlled trials to clarify underlying mechanisms and long-term clinical outcomes.
Article Publication Date: 16/03/2026
DOI: 10.1186/s12888-026-07917-1
