Autism and Stem Cell Therapy (MSCs): Literature Review Scope & Rationale

Differentiation

This is the first piece in the series that reviews actual clinical trial data rather than biological mechanisms, regulatory context, or logistics. It is genuinely non-duplicate. The source draft described specific trials (RCTs, a phase-II study, meta-analyses) without a single citation all underlying studies have now been identified and verified via literature search and are cited properly below.

Abstract

Research into mesenchymal stem cell (MSC) therapy for autism spectrum disorder (ASD) has grown over the past decade, driven by interest in the immunomodulatory, anti-inflammatory, and paracrine properties of these cells. This article reviews the current clinical trial evidence randomized controlled trials, open-label studies, and meta-analyses along with dosing routes, candidate cell types, safety data, and the significant limitations of the current evidence base. The findings are promising but heterogeneous: they do not yet establish a standardized protocol or definitive efficacy for core autism symptoms.

Randomized and Controlled Studies

A randomized controlled trial published in Asia-Pacific Psychiatry (Sharifzadeh et al., 2021) compared intrathecal autologous bone-marrow mesenchymal stem cells (BM-MSCs) plus rehabilitation therapy and risperidone against rehabilitation and risperidone alone in 32 children with ASD aged 5–15. The trial found the intrathecal BM-MSC injection to be safe and feasible but reported only limited additional clinical efficacy compared with standard care both groups improved, with the stem cell group showing no clear additional advantage on the outcome measures used.

A separate phase II, randomized, placebo-controlled trial (Dawson et al., 2020, published in The Journal of Pediatrics) evaluated intravenous umbilical cord blood infusion in children with ASD. This trial found no significant overall benefit on its primary endpoints. Exploratory subgroup analyses suggested possible signals in certain children, but subgroup findings of this kind are hypothesis-generating rather than confirmatory and should not be read as established efficacy.

Open-Label and Nonrandomized Studies

A 2025 open-label, randomized, controlled phase II trial (published in Stem Cell Research & Therapy) evaluated autologous bone marrow mononuclear cell (BMMNC) administration combined with educational intervention, assessing outcomes via DSM-5 criteria, the Childhood Autism Rating Scale (CARS), the Clinical Global Impression-Severity scale, and the Vineland Adaptive Behavior Scale at baseline, 2, 6, and 12 months. Programs of this kind combining cell-based treatment with structured, individualized developmental intervention have reported reductions in autism severity scores and gains in social communication, language, and daily living skills over extended follow-up. Because these studies are open-label (unblinded) and often lack a untreated control arm, their results carry a meaningfully higher risk of bias than the randomized, placebo-controlled trials above, and should be weighted accordingly.

Smaller case series and single-patient reports using Wharton’s jelly–derived MSCs, delivered via combined intrathecal and intravenous routes alongside neurorehabilitation, have described language and motor improvements in individual children. These reports support the rationale for continued investigation but do not constitute controlled evidence of efficacy.

Meta-Analyses and Systematic Reviews

The first meta-analysis specifically on stem cell therapy for ASD in children (Qu et al., 2022, Frontiers in Pediatrics) pooled data from 5 eligible studies. It found that Childhood Autism Rating Scale (CARS) scores were significantly lower in the stem cell group compared to controls (weighted mean difference: −5.96; 95% CI: −8.87 to −3.06; p < 0.0001). However, Clinical Global Impression (CGI) scores showed no significant difference between groups (RR = 1.01; 95% CI: 0.87 to 1.18; p = 0.89), and there was no significant difference in adverse event rates between stem cell and control groups (RR = 1.55; 95% CI: 0.60 to 3.98; p = 0.36). The authors concluded that stem cell therapy for children with autism “might be safe and effective,” while explicitly noting that the evidence was limited by small study sizes, non-standardized injection routes and doses, and a lack of long-term follow-up.

A separate systematic review and meta-analysis (Riva et al., 2021, Stem Cell Reviews and Reports) included 11 trials totaling 461 patients, encompassing both controlled and non-controlled, randomized and non-randomized studies. Consistent with the findings above, this broader review reported mixed results across outcome measures — some scales showed improvement, others did not reach statistical significance — reinforcing that the evidence base, while growing, remains inconsistent across different study designs and outcome tools.

Administration Routes and Dosing

Three delivery routes appear across the literature, each with a different risk-exposure tradeoff:

• Intrathecal (IT) delivery places cells more directly into the cerebrospinal fluid, theoretically increasing central nervous system exposure, but is more invasive than intravenous delivery.
• Intravenous (IV) infusion is the least invasive route and the most widely used in ASD studies, though the blood-brain barrier limits how directly infused cells can reach the central nervous system.
• Intraventricular or intracerebral delivery provides the most direct CNS access but carries meaningfully higher procedural risk and is far less commonly used or studied.

Some observational data suggest that longer follow-up periods and/or repeated infusions may correlate with better reported outcomes, though this observation is not established as definitive and could reflect other factors, including continued concurrent developmental therapy during longer follow-up windows.

Candidate Cell Types and Potential Moderators

Mesenchymal stem cells, including Wharton’s jelly–derived MSCs, currently appear the most studied and generally favorably tolerated candidate for ASD research, largely due to their immune-modulating properties and safety profile across trials. Cord-blood stem cells (autologous or allogeneic) are more readily accessible but have produced less consistent outcomes across the trials reviewed above, including the negative primary-endpoint result in the Dawson et al. (2020) trial.

Some studies have explored potential moderators of treatment response factors including younger age (approximately 3–7 years), lower body weight, and lower baseline inflammatory markers have been proposed as possibly associated with more favorable responses. These associations come from exploratory or subgroup analyses and should be treated as hypotheses for further study, not as established selection criteria.

Safety Profile

Across the studies reviewed, short-term safety has generally been reported as acceptable. The most reported adverse events are mild and transient fever, fatigue, and infusion-related reactions. Serious adverse events have been reported as rare across these trials. The meta-analysis by Qu et al. (2022) found no statistically significant difference in adverse event rates between stem cell and control groups. That said, careful candidate screening, aseptic technique, structured monitoring during and after treatment, and longer-term safety surveillance remain warranted, particularly given the still-limited long-term follow-up data across this body of research.

Limitations of the Current Evidence Base

High protocol heterogeneity: cell source (autologous vs. allogeneic; cord blood vs. MSC subtype), dose, delivery route, treatment intervals, and outcome measurement tools differ substantially across studies, making direct comparison difficult.

Small sample sizes and relatively short follow-up periods limit statistical power and the ability to draw firm efficacy conclusions.

Limited use of biological markers (biomarkers) restricts the field’s ability to identify which specific children are most likely to respond, or why.

A meaningful share of the most encouraging results come from open-label or non-randomized designs, which carry a higher risk of bias than the randomized, placebo-controlled trials, which have shown more modest or null results on primary endpoints.

Practical Takeaways for Families and Clinicians

Current evidence supports a generally acceptable short-term safety profile, and some studies report improvements in select outcome domains for some children particularly in the context of careful candidate selection, structured follow-up, and concurrent developmental therapy.

The evidence remains insufficient to support universal recommendations. The field’s own meta-analyses call explicitly for larger, blinded, randomized trials with standardized protocols, validated biomarkers, and longer-term outcome tracking.

Regulatory status: stem cell therapy for autism remains investigational in many jurisdictions and does not have regulatory approval (e.g., FDA approval in the United States) for ASD indications specifically. For the regulatory picture in Thailand specifically, see our companion article on the regulatory landscape.

Conclusion

The clinical trial evidence for stem cell therapy in autism is real, growing, and worth taking seriously but it is also genuinely mixed. Randomized, placebo-controlled trials have generally shown safety without clear efficacy on primary endpoints, while open-label studies combining cell therapy with structured developmental intervention have reported more encouraging results on secondary or exploratory measures. Meta-analyses pooling this evidence describe the overall picture as promising but not yet conclusive, with substantial methodological limitations that the field’s own researchers have flagged. Families and clinicians considering this option should weigh this evidence honestly — neither dismissing it nor overstating what it currently shows.

This article is for educational purposes and does not constitute medical advice. It summarizes published research and does not represent a guarantee of outcome for any individual. Please consult a qualified physician to discuss whether any specific treatment approach is appropriate for your child.

References

Sharifzadeh, N., Ghasemi, A., Tavakol Afshari, J., Moharari, F., Soltanifar, A., & Talaei, A. (2021). Intrathecal autologous bone marrow stem cell therapy in children with autism: A randomized controlled trial. Asia-Pacific Psychiatry, 13(2), e12445.

Dawson, G., Sun, J. M., Baker, J., Carpenter, K., Compton, S., Deaver, M., et al. (2020). A phase II randomized clinical trial of the safety and efficacy of intravenous umbilical cord blood infusion for treatment of children with autism spectrum disorder. The Journal of Pediatrics, 222, 164–173.e5.

Qu, J., Liu, Z., Li, L., Zou, Z., He, Z., Zhou, L., Luo, Y., Zhang, M., & Ye, J. (2022). Efficacy and safety of stem cell therapy in children with autism spectrum disorders: A systematic review and meta-analysis. Frontiers in Pediatrics, 10, 897398.

Riva, G., et al. (2021). Stem cell therapy in the treatment of patients with autism spectrum disorder: A systematic review and meta-analysis. Stem Cell Reviews and Reports.

Outcomes of autologous bone marrow mononuclear cell administration combined with educational intervention in the treatment of autism spectrum disorder: a randomized, open-label, controlled phase II clinical trial. (2025). Stem Cell Research & Therapy.