Gut Feelings: What Fecal Transplant Trials Mean for Autism Treatment

For decades, autism research has focused on the brain — its wiring, its chemistry, its development. But some of the most surprising recent findings are coming from a very different organ: the gut. Trials of microbiota transfer therapy (MTT), a clinical refinement of the faecal transplant, have shown lasting reductions in both gastrointestinal and behavioural symptoms in autistic children. The results have re-energised a once-fringe idea — that the trillions of microbes living in our intestines may be quietly shaping how our brains behave.

For Australian families, where autism diagnoses have climbed steadily and the National Disability Insurance Scheme spends heavily on support, the appeal of a biological therapy is obvious. But so is the need for caution. Here is what the evidence actually shows, what it doesn't, and why microbiome medicine is at a delicate crossroads.

From experiment to two-year follow-up

The most discussed results come from a team at Arizona State University, whose follow-up study found that autism symptoms were reduced by nearly 50% two years after a course of microbiota transfer therapy. Children in the original open-label trial received an initial high dose of standardised gut microbiota from healthy donors, followed by lower maintenance doses over several weeks. The therapy targeted the well-documented overlap between autism and chronic gastrointestinal problems — constipation, diarrhoea, abdominal pain — that affect a large proportion of autistic children.

What surprised the researchers was not just the gut response, which they expected, but the durability of the behavioural changes. As reported by ASU News and published in Scientific Reports, parents and clinicians using standard rating scales recorded ongoing improvements in social skills, irritability and repetitive behaviours two years after treatment ended. The team's analysis of stool samples also showed that beneficial bacterial populations introduced during the therapy were still present — a sign that the transplant had achieved long-term engraftment rather than a fleeting bloom.

That detail matters. Many probiotic interventions show short bursts of benefit that vanish once the supplement stops. A microbiome that remains structurally altered years later is a different kind of evidence — closer to a true treatment effect than a placebo wash.

Why the gut-brain story is plausible

The idea that gut bacteria could influence neurodevelopment was once dismissed. It is now mainstream biology. The vagus nerve, the immune system and a long list of microbial metabolites form a constant signalling loop between the intestine and the brain. Short-chain fatty acids produced by gut bacteria can cross into circulation; some influence inflammation, others affect the production of neurotransmitters like serotonin, the vast majority of which is made in the gut, not the head.

Autistic children, on average, show measurably different gut microbial communities from neurotypical peers — lower diversity, different ratios of key species, and altered metabolite profiles. Whether those differences cause behavioural traits, result from them (through restricted diets, for example), or simply co-occur is still being untangled. The MTT trials offer a rare piece of causal evidence: change the microbiome substantially, and at least some symptoms appear to shift.

This is the foundation of what's now called gut-brain medicine — a field that includes work on depression, Parkinson's disease, anxiety and chronic fatigue. Autism is the most ethically and scientifically charged frontier, because the patients are children and the condition is lifelong.

The risk side of the ledger

Enthusiasm needs tempering with the warnings emerging from the same research community. A News-Medical report this year highlighted a study cautioning that faecal microbiota transplants can carry long-term risks that are only now becoming visible. Transferring an entire microbial ecosystem from one person to another is not like prescribing a single drug. Donors can pass on bacteria, viruses, fungi and metabolic tendencies that interact unpredictably with the recipient's biology. Cases of infection, and concerns about transmitting traits linked to obesity or autoimmune disease, have already prompted regulators to tighten donor screening.

For autism specifically, the population is paediatric, and the intervention is being considered for children whose immune and neurological systems are still developing. Any treatment that durably reshapes the gut ecosystem in a five-year-old is, by definition, a long-term experiment. The fact that the ASU follow-up found persistent engraftment is encouraging for efficacy but raises the stakes for safety: what is implanted may not be easily removed.

The trial design problem

This is where a sharp note from Science becomes essential reading. In a recent commentary, researchers argued that biological therapies — including microbiome-based ones — need definitive randomised controlled clinical trials before they enter routine practice. The published autism studies, while promising, have been small and largely open-label, meaning families and clinicians knew who was receiving treatment. That design is appropriate for early safety and signal-finding work, but it cannot rule out placebo effects, expectation bias in parent-reported scales, or the natural developmental gains that many autistic children make over a two-year period regardless of intervention.

A properly powered, blinded, placebo-controlled trial — ideally across multiple sites — is the threshold the field still needs to cross. Until then, MTT for autism sits in the same regulatory grey zone as many emerging biological therapies: scientifically interesting, plausibly effective, and not yet ready to be a standard of care.

What this means for Australian families

Faecal microbiota transplantation is approved in Australia for Clostridioides difficile infection that does not respond to antibiotics. The Therapeutic Goods Administration regulates it as a biological, with strict donor screening requirements. For autism, there is no approved protocol here, and the standardised microbial preparation used in the US trials is not commercially available.

That hasn't stopped some families from exploring overseas options or unregulated clinics — a pattern Australian clinicians have warned against. The combination of a vulnerable patient group, hopeful parents and an under-evidenced procedure is precisely where harm tends to occur. Families interested in the science have better, safer paths: enrolling in registered trials when they open, addressing diet and gastrointestinal symptoms through paediatric gastroenterology, and watching for the next wave of published data.

A broader shift in how we think about the brain

The deeper significance of these trials may be less about autism specifically and more about what they signal for medicine generally. If a two-week course aimed at the gut can produce measurable behavioural changes two years later, the implication is that some chronic neurological conditions are partly ecological — products of the relationship between a body and the microbial communities living in it. That reframes what "treatment" can mean. Instead of correcting a single molecule, we may increasingly be tending an ecosystem.

It also reframes what counts as evidence. Microbiome trials are messy: every donor is different, every recipient's existing microbiome is different, and outcomes may depend on diet, antibiotics taken years earlier, or even mode of birth. Designing rigorous studies in this space is genuinely hard, which is why the call from Science for definitive randomised trials matters so much. Without them, microbiome medicine risks repeating the cycle of hype and disappointment that has dogged stem-cell therapies.

For now, the most honest summary is this: faecal transplants for autism are not a cure, not yet a proven treatment, and not without risk — but they are a genuine scientific lead. They have moved the gut-brain hypothesis from speculation to something that can be tested, measured, and either confirmed or discarded. That, in a field where progress is measured in decades, is not a small thing.

Related on Bleen

Sources