What Is Stem Cell Therapy for Autism?
"Stem cell therapy for autism" usually refers to the infusion of mesenchymal stem/stromal cells (MSCs) into a child or adult with autism spectrum disorder (ASD), with the stated goal of influencing biological processes — most often neuroinflammation — that some researchers hypothesize contribute to certain autism presentations. MSCs are cells that can be isolated from several tissue sources, including umbilical cord tissue, bone marrow, and adipose (fat) tissue. Most current research and most U.S. clinical access programs use donor (allogeneic) umbilical-cord-derived MSCs rather than a patient's own cells.
It's important to be precise about what this treatment is and is not. It is not a genetic therapy, it does not alter a person's DNA, and it is not positioned in the research literature as a cure for autism. Autism spectrum disorder is a lifelong neurodevelopmental difference, and BetterStem does not frame it as a condition that must be eliminated — the stated goal of MSC research in this space is to explore whether cell therapy may support certain aspects of daily functioning or quality of life, not to change who a person is.
Because "stem cell therapy" is sometimes used loosely to describe several different products and delivery methods — MSCs, cord blood mononuclear cells, exosomes, different infusion routes — this guide uses precise terminology throughout.
Mesenchymal Stem Cells vs. Exosomes — Two Different Things Often Confused
MSCs are living cells. Exosomes (a type of extracellular vesicle) are tiny particles that cells — including MSCs — release; they carry proteins, lipids, and genetic material that may influence nearby cells, but they are not cells themselves and cannot divide or engraft. Some clinics market exosome products for autism based on the idea that they might deliver the same signaling benefits as MSCs without using live cells.
Important
As of this writing, there is no completed, published human clinical trial evidence for exosome therapy in autism — only early-stage, unpublished trial registrations and preclinical animal research.
What the Evidence Shows
The Duke IMPACT Study — Largest Trial to Date
137 children ages 4–11, randomized, double-blind, placebo-controlled, testing a single IV infusion of umbilical-cord-tissue-derived MSCs. Results posted in 2026 did not show a statistically significant difference between the MSC group and placebo on the primary outcome measure.
- Earlier, smaller studies — including a 2019 open-label safety study (Riordan et al., 20 participants) and a 2013 Chinese cohort study (Lv et al., 37 participants) — reported some improvement in behavioral rating scales, but lacked placebo control, used small samples, and cannot establish that the cells caused the changes observed.
- Duke- and Sutter Health-sponsored randomized trials of autologous (the child's own) umbilical cord blood — a related but distinct cell product — also did not find a statistically significant benefit over placebo.
- Systematic reviews (Villarreal-Martínez et al. 2021/2022; Qu et al. 2022) describe the evidence as preliminary, note generally favorable short-term safety signals, and consistently call for larger, controlled trials before drawing conclusions about effectiveness.
What Is Still Unknown
- Why the largest controlled trial (Duke IMPACT) did not replicate the more promising signals seen in smaller, uncontrolled earlier studies.
- Whether any particular cell dose, source, or administration route is more effective than another — no study to date has directly compared these variables in autism.
- Which children, if any, are more likely to respond, and by what measurable mechanism.
- Whether findings from a study using one specific cell product can be generalized to a different clinic's product — dose, processing, and donor source all vary.
- Whether exosome therapy provides any measurable clinical benefit for autism in humans — this has not yet been studied in a published human trial.