Trial listings go stale within months, and a listing you cannot verify is worth nothing. This page teaches you to query the registries yourself — and explains why joining a trial is not the same as buying a treatment.
Verification note: every figure and trial record below was pulled directly from ClinicalTrials.gov on 29 July 2026. Check them yourself — the instructions are in the next section, and any claim on this page should be re-checked before you rely on it.
First: How to Search the Registries Yourself
This is the part worth learning, because it does not go out of date.
ClinicalTrials.gov
- Go to clinicaltrials.gov
- In "Condition/disease" enter your diagnosis — knee osteoarthritis, multiple sclerosis
- In "Intervention/treatment" enter mesenchymal stem cell (or exosome, or the specific product)
- Set "Study status" to Recruiting and Not yet recruiting
- Add your country or city under Location
Each result links to a full record. The contact details of the study coordinator are on that record — you can email or call them directly. You do not need an intermediary, and you should not pay anyone to do this for you.
The other registries
- WHO ICTRP — who.int/clinical-trials-registry-platform<sup>2</sup> — aggregates registries worldwide, useful outside the US
- EU Clinical Trials Information System (CTIS) — the EU register, which replaced EudraCT
- ISRCTN — isrctn.com — UK and international
- ANZCTR — anzctr.org.au — Australia and New Zealand
- jRCT / UMIN-CTR — Japan
⚠️ A note on reading published trial reports too: papers can be corrected or retracted after publication, and the correction is not always visible on the copy you are reading. An example in this field is an umbilical-cord MSC multiple-sclerosis feasibility study that has since had a correction published<sup>8</sup>.
How to Read a Registry Record
Four fields tell you most of what you need. They are the fields that summaries usually omit.
Two more worth checking:
- Sponsor and sponsor class. Industry, Other (usually academic), NIH. An industry sponsor is not disqualifying — most drug development is industry-run — but it is context
- Design: allocation and masking. Randomised and double-blind are what make a result interpretable. Non-randomised, open-label, single-arm means there is no comparison group, and no comparison group means the study cannot show that a treatment works, however many people enrol
⚠️ How to spot a study that cannot produce efficacy evidence
Some registered studies are structured so that they can never demonstrate that anything works. A real, currently recruiting example, which you can open and read:
NCT04684602 — "Evaluation of the Safety, Tolerability and Efficacy of Regenerative Therapy for the Treatment of Various Chronic and Acute Conditions"
Sponsor: an industry LLC · 5,000 patients (estimated) · non-randomised · no masking ·
no data monitoring committee · eleven unrelated conditions in one protocol — autoimmune,
cardiovascular, diabetic, skin, musculoskeletal, neurodegenerative, neurological, pulmonary, sexual,
urological and viral · completion 2030 ·
**entire inclusion criteria: "Age 18 and older · Ability to provide informed consent · Availability
for follow up visits."**
There is no disease-specific eligibility requirement and no control group. A study like this can collect safety information and patient-reported outcomes. It cannot tell anyone whether the treatment works, because there is nothing to compare against.
It is a registered clinical trial with a valid NCT number. That is exactly why the number alone is not reassurance.
What the Landscape Actually Looks Like
Figures pulled from ClinicalTrials.gov on 29 July 2026. Re-run these queries before relying on them.
⚠️ That last number is the important one. Phase 3 is where a treatment is tested at scale against a proper comparator, and it is what regulators require. Fourteen worldwide, across every indication, against 372 active studies overall, tells you the field is still overwhelmingly in early-phase, exploratory work — a picture consistent with long-standing assessments of translational progress in this field<sup>3</sup>.
⚠️ You will see "500+ active trials" quoted widely across this field. ClinicalTrials.gov returns 372. A larger worldwide figure is plausible once other registries are included, but anyone quoting a number should be able to tell you which registry, which filters, and which date — and if they cannot, treat the number as decoration.
Examples of Genuinely Recruiting Phase 3 Trials
Verified against the registry on 29 July 2026. Every one of these is checkable in one click, and you should check it — status changes.
⭐ The CARTISTEM trial is worth singling out. CARTISTEM is already an approved product in South Korea, and this trial compares it against a surgical comparator — the design that actually answers a patient's question, which is not "does it do something?" but "is it better than what I would otherwise be offered?"
⚠️ *All enrolment figures above are estimated*** — these trials are still recruiting, so no actual figure exists yet.
⛔ Participation Is Not Purchase
This is the most important section on this page.
If you pay a clinic for a stem cell treatment, you are not taking part in research. Specifically:
A treatment sold commercially is not "the same as what is being trialled", even when the same cell type is used. The trial is the comparison. Without it, an individual outcome — yours or anyone's — cannot distinguish treatment effect from natural history, regression to the mean, or expectation.
⚖️ The honest trade-off, stated fairly. A trial may randomise you to placebo, will impose visit schedules, and may not accept you at all. Commercial treatment gives you certainty of receiving the product, immediately, on your schedule. Those are real advantages and some people reasonably choose them. What you should not believe is that paying makes you part of the research.
Questions worth asking before enrolling in any trial
- What phase is this, and what did the earlier phases show?
- What is my probability of receiving placebo?
- How many visits, over how long, and where?
- What happens to my access to the treatment when the trial ends?
- Are travel and accommodation covered?
- Who is the sponsor, and is there an independent data monitoring committee?
- Where will the results be published, and by when?
Approved Cell Therapy Products
Unlike the trial landscape, this list is short and verifiable — these products have completed regulatory review.
⚠️ Note what is not on this list. There is no approved MSC product for osteoarthritis, back pain, COPD, autoimmune disease generally, anti-ageing, or any of the wellness indications. Approval for one product in one indication in one country is not approval for cell therapy as a category.
Open Questions the Field Is Still Working On
These are genuine, active areas of uncertainty — and the fact that they are still open is itself informative.
Dosing. Optimal cell count is unresolved. ⚠️ In one randomised trial of allogeneic MSCs in ageing frailty (NCT02065245, n=65)<sup>4</sup>, physical performance improved in the 100-million-cell group and not in the 200-million-cell group<sup>5</sup> — one trial, not to be over-read, but it argues against assuming more is better.
Allogeneic vs autologous. The field is trending toward allogeneic "off-the-shelf" products: manufacturing scale, immediate availability, standardised quality control. ⚠️ Open questions remain about donor selection, long-term persistence, and sensitisation risk on repeat dosing.
Fresh vs cryopreserved. Fresh cells avoid cryopreservation stress; cryopreserved cells enable centralised manufacturing and quality control. ⚠️ Direct comparative trials are limited, and claims in either direction should be treated as preliminary.
Route of administration. Intra-articular, intravenous, intrathecal and topical routes are all under study. ⚠️ Evidence for one route does not transfer to another — this is one of the commonest errors in marketing material.
Product identity. ⚠️ Several widely cited "MSC" trials use different products. MultiStem is a multipotent adult progenitor cell (MAPC); PISCES used neural stem cells; some studies use extracellular vesicles rather than cells. Check what was actually administered, not what the summary calls it.
Safety: What Is and Is Not Established
Reasonably established: systematic reviews of intravascular MSC administration have found no consistent association with serious adverse events, with transient fever the most reproducible finding<sup>6</sup>.
⚠️ Less settled than usually presented:
- Product-dependent risk: MSC preparations vary in procoagulant tissue factor content and can trigger an instant blood-mediated inflammatory reaction. Safety data from bone-marrow-derived products does not automatically extend to adipose or perinatal (umbilical cord) products<sup>7</sup>
- Long-term follow-up remains limited. Statements about tumorigenicity are supported for short to medium-term follow-up, which is what has been measured
- Adverse-event reporting in this literature is inconsistent, and reviewers have repeatedly called for more rigorous reporting<sup>6</sup>
Stay Current
Trial status changes constantly. Rather than relying on any static list:
- Set up a ClinicalTrials.gov saved search with email alerts for your condition and intervention
- Re-check the status of any trial before contacting a site
- Ask any clinic quoting trial evidence for the NCT number, then open it yourself