The industry has a variance problem.
"Is stem cell therapy safe?" is a reasonable question with an unreasonable amount of noise around it. Part of the noise comes from the therapy's defenders, who sometimes talk as if any injection labeled "stem cells" carries the safety profile of the best-run clinical trials. Part comes from its critics, who sometimes talk as if every clinic outside the United States operates without oversight. Neither is accurate.
The underlying reality is variance. When researchers at the University of Minnesota and UC Davis surveyed the direct-to-consumer industry in 2016, they found hundreds of businesses in the United States alone marketing cell-based interventions, with enormous differences in what was actually being injected, by whom, and on what evidence.[1] A companion analysis mapped the same pattern worldwide.[2] Some providers use characterized, laboratory-tested mesenchymal stromal cells at defined doses. Others use unverified products of unknown cell count and identity. Both call it "stem cell therapy."
This is why the question cannot be answered with a yes or no about the field. It has to be answered in layers: the safety record of the cell type itself in controlled trials, the regulatory framework the clinic operates under, and the release standards of the specific laboratory supplying the cells. Each layer is checkable, and the rest of this article goes through them in order.
- The short version
- In controlled trials, mesenchymal stromal cell therapy has a consistently reassuring short-term safety record — no meta-analysis has linked it to tumor formation, and the most common attributable effect is transient fever. The real safety variable is not the cell type. It is the quality of the laboratory and the honesty of the clinic.
What the trial data actually shows.
The best evidence on MSC safety comes from systematic reviews that pool every eligible randomized controlled trial and count adverse events. The first major one, the SafeCell analysis published in 2012, examined prospective clinical trials of intravascular MSC delivery across conditions from heart disease to Crohn's disease. Its finding: a significant association with transient fever, and no association with acute infusional toxicity, organ system complications, infection, death, or malignancy.[3]
Because a 2012 snapshot could miss late-emerging harms, the same research group repeated the exercise in 2020 with roughly twice the number of trials and over 2,500 treated patients, including studies with years of follow-up. The conclusion held. MSC administration remained associated with fever and nothing more serious; malignancy rates in treated patients were no higher than in controls.[4] The paper's title states its finding plainly: cell therapy with MSCs "continues to appear safe."
Across two decades of randomized trials, the most consistent adverse event attributable to MSC infusion is a transient fever — and no meta-analysis has found an association with tumor formation.
Three caveats keep this honest. First, these analyses cover trial-grade cells — characterized, tested, and dosed under protocol. They say nothing about an unverified product from an unaudited supplier, which is exactly why the laboratory questions later in this article matter. Second, "safe" is not "effective"; safety data and efficacy data are separate questions, and efficacy varies by condition — the condition-specific evidence is reviewed in the companion articles on arthritis and neuropathy. Third, every medical procedure carries procedural risk — injection-site pain, bruising, the small infection risk of any needle — which is managed by clinical technique, not by the cells.
How cell therapy is regulated in Mexico.
A persistent myth holds that stem cell treatment in Mexico happens in a regulatory vacuum. It does not. Medical practice, cell banking, and the use of human tissue in Mexico are governed by the Ley General de Salud (General Health Law) and overseen by COFEPRIS — the Comisión Federal para la Protección contra Riesgos Sanitarios, Mexico's federal health-risk regulator and the counterpart to agencies like the FDA.[7]
Under this framework, several things require government licensing rather than mere registration. Establishments that procure, process, store, or dispose of human cells and tissues must hold a sanitary license and operate under a named, government-recognized responsible officer. Hospitals and clinics where procedures are performed carry their own establishment licenses. Physicians must hold a Mexican professional license (cédula profesional), with specialty credentials on top of it. These are documents, not claims — each one can be produced on request.
The practical difference between the Mexican and U.S. frameworks is not oversight versus none. It is that Mexican law gives licensed physicians wider latitude to use cell-based preparations in clinical practice, whereas the FDA restricts expanded (culture-grown) cells to registered trials. That latitude is why treatment is available in Mexico at all — and it is also precisely why the burden of vetting shifts toward the individual clinic and laboratory. The regulator sets the floor; the clinic's own standards determine everything above it. International bodies such as the ISSCR publish guidelines for what responsible clinical translation of stem cell science looks like anywhere in the world, and they are a fair yardstick to hold any provider against, in any country.[6]
What a serious laboratory tests before release.
The safety literature in Section 02 describes characterized cells, so the operative question for any patient is whether the cells they would receive are characterized the same way. That work happens in the laboratory, before a physician ever draws the dose, and it produces a specific document: the certificate of analysis (COA) for the batch.
A complete COA states the cell count, the viability percentage at release, the passage number (how many times the culture has been expanded — lower passages are preferred because extensively passaged cells decline in potency), the surface-marker results against the ISCT criteria,[5] and the sterility, mycoplasma, and endotoxin results for the lot. It is tied to a batch number that matches the vial. A clinic that works with a serious laboratory can show you this document without hesitation, because it already uses it internally to accept or reject incoming batches.
Where the cells come from matters too, and it is covered in depth in the article on cell sourcing: umbilical-cord tissue from screened, consenting donors after healthy full-term births has become the standard source for allogeneic therapy, in part because it allows exactly this kind of bank-level testing and characterization before any patient is involved.
How to vet a clinic — including this one.
Everything above reduces to a short list of questions that any legitimate provider can answer with documents and specifics, in writing, before you pay anything. Ask them of every clinic you consider, ours included:
- Which laboratory supplies your cells, and can I see a certificate of analysis for a recent batch? You are looking for a named, licensed lab and a COA showing count, viability, passage number, marker profile, and sterility results.
- What exact cell type and dose would I receive, and by what route? "Stem cells" is not an answer. "Umbilical cord–derived mesenchymal stromal cells, a stated number of cells, delivered intravenously or by image-guided injection" is.
- Who performs the procedure, and what is their license number? A Mexican physician's cédula profesional is verifiable through the federal registry. Board certifications have issuing bodies you can check.
- What licenses does the facility hold? Establishment and sanitary licenses exist on paper. A clinic operating inside an accredited hospital can also tell you which hospital and why.
- Will a physician review my records before quoting a protocol? A dose recommended before anyone has seen your history and imaging is a sales figure, not a clinical judgment.
- What should I not expect? This is the revealing one. An honest provider will name the limits — which outcomes the evidence supports, which it does not, and who is not a candidate.
The red flags are the mirror image of the list. Guaranteed results or the word "cure." One price and one dose for every patient and every condition. Reluctance to name the laboratory. Pressure to book before records review. Claims that a single treatment addresses a long list of unrelated diseases. None of these prove a clinic is dangerous; each one means the burden of proof has not been met, and in medicine that is reason enough to keep looking.
Safety, in the end, is not a property of a therapy in the abstract. It is a property of characterized cells, released by a tested process, delivered by licensed hands, to a patient whose candidacy was actually evaluated. Every one of those links can be verified before you commit — and a clinic worth choosing will be glad you asked.