Autism and Diagnosis

Metabolites and Phenotypes

In a brand new paper entitled. “Elevated Microbially-Derived Metabolites in Autism: A Possible Diagnostic Screening Test for a Distinct ASD Phenotype,” Dr. Flynn and colleagues investigated whether specific urinary metabolites produced by intestinal microbes could serve as predictive biomarkers for autism spectrum disorder. The study proposes that a large subset of children with autism may represent a biologically distinct phenotype characterized by markedly elevated microbial metabolites, which the authors term ASD-MDM (Autism Spectrum Disorder associated with Microbially-Derived Metabolites).

The investigators analyzed urine samples from 52 children with ASD and 47 typically developing (TD) controls, ages 2–11 years. They measured microbial metabolites derived primarily from phenylalanine metabolism, tryptophan metabolism, and yeast metabolism. These were chosen because decades of autism, microbiome, and metabolic research have repeatedly implicated abnormalities in microbial metabolism of aromatic amino acids, especially phenylalanine and tryptophan, along with evidence of elevated fungal/yeast metabolites in subsets of children with ASD.

The findings were striking. Children with ASD demonstrated significantly higher concentrations of many microbial metabolites than controls. Some metabolites were elevated by hundreds to thousands of percent, and in certain children concentrations were reported to be 100–1000 times higher than the highest levels observed in any control participant. Twenty-three of twenty-four measured microbial metabolites were higher in ASD participants. Ninety percent of the ASD participants (45 of 50) had one or more extremely elevated MDMs. Particularly notable were elevations in compounds related to p-cresol, p-cresol sulfate, phenylacetylglutamine, indole derivatives, and arabinitol, suggesting abnormalities in microbial metabolism involving aromatic amino acids and yeast overgrowth.

Approximately 40% of individuals with ASD experience chronic gastrointestinal symptoms, such as constipation, diarrhea, bloating, abdominal pain, and reflux, and the severity of these symptoms often parallels the severity of autism traits. Research suggests that early-life gut dysbiosis is common in ASD and may influence neurodevelopment through the gut-brain axis by altering short-chain fatty acids, immune signaling, detoxification pathways, and neurotransmitter metabolism. "Many microbially-derived metabolites have biological effects that are context-dependent, influenced by concentration, developmental timing, host physiology, and co-occurring metabolic perturbations. In the context of early neurodevelopment, abnormal elevation of certain microbial metabolites—particularly those derived from tryptophan and phenylalanine metabolism—has been associated with adverse effects on gut barrier function, immune signaling, neurotransmitter pathways, and neurobehavioral outcomes" (Flynn et al 2026) One of the best-studied examples is p-cresol, a microbial metabolite consistently elevated in children with ASD that has been linked to adverse effects on gut function, mitochondria, immune regulation, and brain development, with levels correlating with autism severity. Additional microbial and yeast-associated metabolites, including indoxyl sulfate and arabinitol, further support the hypothesis that dysbiosis-associated metabolic disturbances may contribute to symptoms in a substantial subset of children with ASD.

Rather than focusing on a single biomarker, the authors created an MDM System™ Total Score, representing the number of metabolites exceeding the highest value observed in the control group. Children with ASD averaged approximately three elevated metabolites, while all controls scored zero by definition. Using the presence of one or more highly elevated metabolites as a screening criterion, the test achieved 90% sensitivity and 100% specificity within this cohort. Approximately 80–90% of ASD participants demonstrated this metabolite signature.

The authors argue that these findings support the existence of a biologically meaningful ASD subgroup linked to gut microbial metabolism and the gut-brain axis. They propose that microbial metabolites may contribute to neurodevelopmental dysfunction through effects on immune signaling, mitochondrial function, neurotransmitter pathways, intestinal permeability, and neuroinflammation. However, the study was not designed to prove causation, only association.

Overall, this study provides some of the strongest evidence to date that abnormal gut microbial metabolism is associated with autism in a substantial subset of children. If independently validated, urinary microbial metabolite testing could become an inexpensive, noninvasive screening tool and further support the concept that autism includes biologically distinct subtypes with potentially different therapeutic targets.

The take home: the most important implication is not the screening test itself, it is the possibility that microbiome derived biochemical signaling may identify a specific, biologically coherent autism phenotype rather than autism being a single disorder, which I believe to be true. This aligns closely with emerging models of ASD as a collection of overlapping pathophysiologic endotypes and not a single disorder or condition.

More data more understanding,

Dr. M

Flynn Mol Psych

From the paper: "ASD prevalence has dramatically increased from approximately 1 in 10,000 in 1980 to 323 per 10,000 in 2022 in the United States [2, 3]. Average lifetime care costs are estimated at $3.6 million per individual; however, individuals severely affected by the disorder can require much more expensive care [4].

Early behavioral intervention can reduce the core symptoms of ASD and improve outcomes in later life [5, 6]. Behavioral interventions are most effective beginning in early infancy, preferably in the first 24 months of life [7]; however, the average age of first diagnostic assessment interview in the United States (US) is 47 months [8]. Thus, there exists an urgent need to develop a non-invasive screening tool to identify those at high risk for ASD to facilitate earlier diagnosis and treatment [9, 10]. Identifying children with ASD as young as possible and treating them can improve long-term outcomes, enhance quality of life, and potentially save hundreds of billions of dollars per year." (Flynn et al 2026)