- Published in
- European Psychiatry (European Psychiatric Association congress abstract, Vol. 69 Suppl. 1, p. S534)
- Authors of report
- S. M. M. Fekrat, V. Sharma, S. Prasad, and S. Gunturu.
- Date of report
- Medical conditions
- Autism
Major Points and Findings:
Unlike the other summaries in this section, this one is based on a congress abstract, because no full research paper exists. The abstract was published in a supplement to European Psychiatry (volume 69, supplement 1, page S534) as part of the European Psychiatric Association’s meeting. It has no methods section, no forest plot, no table of included studies and no peer-reviewed full text. The summary below covers only what the abstract reports and should not be read as an account of a completed paper. The abstract is included because it is the most recent attempt to pool this literature, and because the earlier meta-analysis it supersedes is already cited on our autism page.
Aim:
To pool the clinical studies of stem cell therapy in children with autism and estimate an overall effect on symptoms, and to identify optimal protocols and patient selection criteria.
Methods (as reported in the abstract):
A PRISMA-guided search of PubMed, Scopus and PsycINFO on 17 September 2025 returned 1,698 records. After removing 399 duplicates, 1,299 were screened by title, abstract and full text. Randomized controlled trials, controlled clinical trials and open-label studies were eligible. Diagnosis had to rest on standardised criteria (DSM-5, ADOS or ADI-R). Interventions included mesenchymal stem cells, umbilical cord blood and neural stem cells. Preclinical studies, case reports and non-English publications were excluded. Two reviewers extracted data independently. Twenty-two studies were included. Outcome instruments across them included CARS, ATEC, the Vineland scales, the Aberrant Behavior Checklist and the PDDBI.
Results (as reported):
- Pooled across all included studies, symptoms improved after stem cell treatment, with a standardised mean difference of −1.03 (95% CI −1.23 to −0.83; Z = 10.03; p < 0.00001).
- Heterogeneity for that figure was I² = 95%.
- In the subset of studies with a placebo comparison, the effect was larger and heterogeneity was absent, with an SMD of −4.99 (95% CI −8.11 to −1.88; Z = 3.14; p = 0.002; I² = 0%).
Conclusions:
The authors state that the existing literature demonstrates efficacy for stem cell therapy in autism, and call for further trials in varied geographical settings and among minority populations.
This conclusion has to be weighed against the heterogeneity figure. An I² of 95% means that almost all the variation between the pooled studies comes from real differences between them (in design, population, cell type, dose and route) and very little from chance. A pooled estimate from studies that differ this much indicates the direction of an effect. It cannot be quoted as the size of a treatment effect. The placebo-controlled subset shows no heterogeneity, but it is a small part of an evidence base that the comprehensive reviews describe as overwhelmingly open-label.
Background Information:
The earlier meta-analysis in this area (Qu and colleagues, 2022, also summarised in this section) pooled five studies and reached a similar conclusion while flagging small samples, no standardised route or dose, and short follow-up. Four years later, with seventeen more studies, the direction of the result is unchanged and most of the caveats still apply.
A pooled effect size from this field is an average across studies that used different cells, at different doses, by different routes, in different countries, mostly without blinding. The summary figure should therefore be treated with caution, and the caution should be stated wherever the number is quoted. None of this is a reason to dismiss the work.
This is a summary of independent research published elsewhere. It is not a report of Beike treatment outcomes.
