Twin Study Shows Gut Transplant Reshapes Microbiome in Autism, But Big Questions Remain

Autism spectrum disorder affects roughly one in 127 people worldwide, yet treatment options targeting its biological roots remain limited. A small but carefully designed study published July 2026 in Frontiers in Microbiology tested whether washed microbiota transplantation could shift the gut environment in autistic children toward the microbial profile of their neurotypical siblings, using identical twin pairs to strip away the confounding influence of genetics. The findings are preliminary, but the multi-omic signal is intriguing enough to justify a closer look.

Washed Microbiota Transplantation and What Sets It Apart From Standard FMT

Washed microbiota transplantation, or WMT, is a modified form of the better-known fecal microbiota transplantation (FMT). The key difference is in the preparation: WMT filters out undigested food residues, fungal material, parasite eggs, and pro-inflammatory metabolites before the transplant material is used. The goal is a cleaner, safer preparation that addresses some of the ethical and aesthetic concerns associated with raw stool transplants. Researchers led by S. Feng and colleagues administered WMT to the autistic child in each of three same-age twin pairs, then used metagenomic sequencing and targeted metabolomics to track what changed in the gut over time.

The broader scientific rationale for this approach sits in the microbiota-gut-brain axis, a signaling network linking gut microbial communities to brain function and behavior. As research on gut microbiota dysbiosis in ASD has shown, children with autism consistently show disrupted microbial communities that may affect emotional regulation and behavior through this axis. WMT is one of several interventions, alongside probiotics and prebiotics, being tested as a way to correct that disruption.

Gut Microbial Structure Shifted Toward Healthy Twin Profiles

The most striking result from the study was not in the clinical scores but in the microbiology itself. After WMT, the gut microbial structure and functional profiles of the treated autistic children showed a shifting trend toward the patterns seen in their typically developing co-twins. The number of differential metabolites between the autistic participants and their healthy siblings was markedly reduced post-intervention. Abnormal lipid and energy metabolism indicators also showed partial ameliorative trends.

Specific microbial taxa drew attention: Segatella, Negativibacillus, and Sangeribacter were flagged as potentially linked to incomplete phenotypic changes. Tyrosine and phenylalanine metabolic pathways, both relevant to neurotransmitter production, were among the metabolic signatures that partially resolved. Earlier long-term data on microbiota transfer therapy in autistic children from Arizona State University found that gut improvements and autism-related symptom gains were largely maintained two years after treatment, which lends some optimism to the idea that microbial remodeling can be durable. The Frontiers study adds a genetically controlled layer to that picture.

Twin-Pair Design as a Tool for Isolating Treatment Effects

The central methodological strength here is the twin-controlled design. Three pairs of age- and environment-matched twins were enrolled, one autistic and one typically developing sibling per pair. Only the autistic participant received WMT. Because twins share genetics and, in most cases, a household environment, the co-twin serves as a natural comparator that would be difficult to replicate with unrelated controls.

Twin studies offer a uniquely powerful way to disentangle genetic from environmental contributions to a trait, and applying that logic to a microbiome intervention is a smart move. Changes in microbial composition that appear after the transplant but not in the untreated co-twin are far more likely to reflect the intervention than they would be in a conventional cohort. That said, the sample is tiny: three pairs is enough to generate hypotheses, not to draw conclusions.

Clinical Rating Scales Did Not Reach Statistical Significance

The study measured behavioral and gastrointestinal outcomes using four validated instruments: CARS (Childhood Autism Rating Scale), ABC (Aberrant Behavior Checklist), SDSC (Sleep Disturbance Scale for Children), and BSFS (Bristol Stool Form Scale). All showed small numerical shifts toward healthier values after WMT, but none reached statistical significance.

This is an important caveat, and the authors are transparent about it. Non-significant numerical shifts cannot be interpreted as evidence of clinical symptom improvement. The gut is clearly responding to WMT; whether that microbial remodeling translates into measurable behavioral change in autism remains unresolved. This aligns with a broader caution noted in reviews of gut-brain axis research in ASD: many FMT and WMT studies come from open-label designs where placebo effects cannot be ruled out, and effect sizes on behavioral endpoints are difficult to interpret.

What Needs to Happen Before WMT Becomes a Realistic Option for Autism

The study authors call for large-scale randomized controlled trials, and that framing is appropriate. Three twin pairs is a proof-of-concept cohort, not a clinical foundation. Several open questions need addressing before WMT could be considered a realistic intervention for autism at scale.

First, autism affects an estimated 61.8 million people globally, roughly one in 127, according to a 2022 Lancet Psychiatry systematic review. Even a modestly effective gut-based intervention would have enormous reach if it held up in larger trials. Second, prior research noted that symptom improvements after FMT in autistic children could reverse within four weeks of stopping treatment, which raises questions about maintenance dosing. The Frontiers study adds the observation that multiple WMT courses may produce more significant improvements than a single round, but the optimal protocol is still undefined.

The metabolic and microbial signals from this twin cohort are genuinely interesting. Multi-omic convergence between treated autistic children and their healthy co-twins, across both metagenomic and metabolomic readouts, suggests WMT is doing something biologically real. Whether that something matters clinically for autism is what a properly powered trial will need to determine.

Evidence rating

INVESTIGATIONAL — 69/100

Caveats:

• Only three twin pairs were studied — far too small to draw reliable conclusions.
• Clinical rating scale improvements did not reach statistical significance and cannot be interpreted as evidence of symptom benefit.
• No control group received a sham or placebo transplant, so placebo effects cannot be ruled out.
• Funding was not disclosed in available materials.
• Results are exploratory and require replication in large, randomized controlled trials before any clinical inference is warranted.

Twin-paired pilot study (n=3 pairs) administering Washed Microbiota Transplantation to ASD-affected children and comparing outcomes to their typically developing co-twins. The design is exploratory; clinical symptom scales showed only non-significant numerical trends, and the sample size precludes any inferential statistics. Not yet replicated at scale.

References

1. S Feng, X Si, C Lu, Z Gao, J Wang, Q Yang, S Lu, T Su… (2026). Washed Microbiota Transplantation Improves Clinical Symptoms, Gut Microbiota, and Metabolic Profiles in Autism Spectrum Disorder in a Twin Cohort. Frontiers in. https://doi.org/10.3389/fmicb.2026.1885281
2. Ziji Fang, Yinxiong Zhou, Ke Chen, Juelan Wang, Xiaoli Liu, Ping Jia (2025). Gut microbiota and autism spectrum disorder: advances in dietary intervention strategies based on the microbiota-gut-brain axis mechanism. Frontiers in Neuroscience. https://doi.org/10.3389/fnins.2025.1587818
3. Study finds microbiota transfer therapy provides long-term improvement in gut health in children with autism | ASU News. news.asu.edu. https://news.asu.edu/20221214-study-finds-microbiota-transfer-therapy-provides-long-term-improvement-gut-health-children-autism
4. Monalisha Sahu, JosyulaG Prasuna (2016). Twin studies: A unique epidemiological tool. Indian Journal of Community Medicine. https://doi.org/10.4103/0970-0218.183593
5. The Gut-Microbiota-Brain Axis in Autism Spectrum Disorder – NCBI. NIH. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8900010/