A Simple Urine Test May Catch Autism Early by Reading Gut Metabolites

Diagnosing autism currently depends on behavioral observation, a process that can take years and leaves families in agonizing uncertainty. A study published in Molecular Psychiatry by Christina K. Flynn and colleagues at Arizona State University suggests a faster, cheaper path: a non-invasive urine test that screens for 17 microbially-derived metabolites and correctly identifies children with autism with 90% sensitivity and 100% specificity.

Gut metabolites distinguish children with autism from typical peers

The core finding is stark. Children with autism spectrum disorder in the study had significantly elevated concentrations of microbially-derived metabolites (MDMs) compared to typically developing controls. These are small molecules that gut microorganisms produce when they break down amino acids like tyrosine, tryptophan, and phenylalanine, the same building blocks the body uses to synthesize dopamine and serotonin. Compounds linked to yeast and fungal activity were also elevated. Almost all children with autism had at least one MDM at levels above any typically developing child in the sample, and on average the ASD group had three metabolites exceeding that threshold, sometimes at concentrations 100 to 1,000 times higher.

The research team built a classification tool called the MDM System, which assigns a score based on how many of a child's metabolites fall outside a typical reference range. Across 52 children with ASD and 47 healthy controls aged 2 to 11, the system achieved that near-perfect diagnostic accuracy. The gut-brain axis research underpinning this work has been accumulating for years: ASD patients consistently show reduced gut microbiota diversity and imbalanced bacterial populations, with metabolite abnormalities now recognized as key mediators of neurobehavioral outcomes.

Why earlier biological screening could transform autism care

Autism affects an estimated 1 in 31 children in the United States, and average lifetime care costs reach $3.6 million per person. The waiting period before a formal behavioral diagnosis is one of the most distressing phases for families, often stretching past age 3 or 4 even when parents notice differences much earlier. A metabolite-based screen could change that calculus at two points: first, as a triage tool for younger children flagged as at-risk, and second, as a guide for clinicians managing children already diagnosed, helping them understand the biological factors driving a child's symptoms and monitor whether interventions are working.

The stakes of moving faster are well-established. Early autism diagnosis leads to more targeted interventions and measurably better outcomes in social development and adult independence. As Flynn put it: "For many families, one of the biggest challenges is the waiting, the not knowing. If this test shortens that gap, even by a little, that's meaningful because earlier intervention can really help."

How the MDM System was built and validated

The Arizona State team ran a multi-site study recruiting participants at clinical research centers in Tennessee and Texas. Urine samples were first analyzed using untargeted semiquantitative liquid chromatography and mass spectrometry (LC-MS) to identify the metabolite landscape broadly, then followed by targeted quantitative LC-MS to measure specific compounds precisely. The ASU team's approach produced a panel of 17 metabolites that most reliably separated the two groups, then wrapped those measurements in the multivariate MDM System scoring framework.

The results align with more than 40 prior studies that have linked elevated microbial metabolites to autism. Based on the data, Flynn and her co-authors propose a formal new subtype: "ASD associated with Microbially-Derived Metabolites," or ASD-MDM. They estimate this phenotype covers roughly 90% of autism spectrum disorder cases. The remaining 10% of children with ASD who showed no abnormal gut metabolites tended to have other significant metabolic issues, possibly genetic in origin.

Limitations that still need resolving before clinical rollout

The sample of 99 children total is large enough for a pilot study but modest by clinical-validation standards. The researchers are explicit that the study does not prove these metabolites cause autism, only that they track with it reliably. Before any test of this kind becomes standard-of-care, larger and more ethnically and geographically diverse cohorts need to replicate the findings. The specificity and phenotypic heterogeneity of autism biomarkers have long complicated the field: distinguishing ASD-specific signals from those shared with other neuropsychiatric conditions remains an open methodological challenge, and the MDM System has not yet been tested against children with ADHD, anxiety disorders, or other conditions that share some behavioral overlap with autism.

A deeper mechanistic question also lingers: how exactly do elevated levels of altered serotonin and dopamine analogs produced by gut bacteria influence neurodevelopment? The gut-brain axis is a plausible highway, but the specific routes and timing remain to be mapped.

What comes next for the test and ASD-MDM research

Analutos, a laboratory in the United Kingdom, is already offering the urine test internationally, giving the research an unusual head start on real-world deployment while validation trials continue. The team envisions the test serving a dual role: a low-barrier screening step for young children whose families or pediatricians have early concerns, and a monitoring tool for tracking biological response to treatment in children already in care.

That second application points toward a broader therapeutic agenda. Previous research cited by the authors showed that interventions specifically targeting the metabolite p-cresol sulfate improved both gut and behavioral symptoms. Microbiota-targeted interventions including fecal microbiota transplantation are already being studied in ASD, with variable but promising results. A validated metabolite screen that can tell clinicians which children have gut dysbiosis driving their symptoms, and measure whether an intervention is actually shifting that biology, would give those trials far sharper tools than they have today. Flynn's doctoral research, now published in a top-tier psychiatry journal and moving toward clinical deployment, has handed the field exactly that kind of instrument.