Evidence review
Am I a candidate? What the trials actually enrolled
Most people assume they are either too old, too far along, or not far enough along. The published trials enrolled a much wider range of patients than that assumption suggests — which is a good reason not to rule yourself out before anyone has looked.
Scope. This article describes the populations studied in published research. It is educational, it is not medical advice, and nothing here means that you personally are or are not a candidate. That determination is made by a physician after reviewing your records. These therapies are not approved by the U.S. Food and Drug Administration for this use.
Who was actually in these studies
A 2025 systematic review in Frontiers in Medicine catalogued the eight placebo-controlled trials of cell therapy for knee osteoarthritis. The population they enrolled is more useful to a prospective patient than most people realise:
| Characteristic | Range across the trials |
|---|---|
| Disease severity (Kellgren-Lawrence) | Grades 1 to 4 — most trials, grades 2–4 |
| Mean age | 47.6 to 67.2 years |
| Mean BMI | 25.0 to 30.9 kg/m² |
| Eligibility | Adults 18+, knee osteoarthritis confirmed clinically and/or radiologically |
Two things stand out. First, the severity range is wide: these were not only early, mild cases. Trials enrolled patients across grades 2 to 4, and one included grade 1. Second, the mean BMI in several trials sat in the overweight-to-obese range — a population often assumed to be excluded from this kind of study.1
The practical read: the people studied were, broadly, people with real knee osteoarthritis in middle and later life who had not yet had a knee replacement.
That is a much larger group than the “perfect candidate” most clinics describe. It is also not a promise that any individual within it will benefit.
The clearest exclusion in the literature
The exclusion criterion that appears consistently is prior knee replacement or other major surgery on that same knee. It is worth knowing this early: if you have already had the joint replaced, the published evidence does not speak to your situation at all, and any clinic implying otherwise is going beyond what exists.
Beyond that, individual trials applied their own criteria — active infection, certain cancers, uncontrolled systemic disease, pregnancy, and specific medications are common exclusions in this kind of research. This is exactly the territory a proper assessment covers, and exactly why it cannot be done from a web page.
What is still being argued about
The honest caveat is that which patients respond best remains genuinely unsettled. There is not yet a validated way to predict, from your file alone, how much benefit you would get.
What the pooled analyses have started to separate is the treatment side rather than the patient side. A 2024 review of 18 trials and 1,174 patients found in subgroup analysis that adipose-derived cells outperformed bone marrow and umbilical cord sources, and that allogeneic preparations outperformed autologous ones.2 The authors were explicit that heterogeneity was high and that these are signals, not settled conclusions.
That matters for a practical reason: it means the differences between clinics are not marketing differences. Cell source and preparation are among the variables the literature is actively trying to pin down, so asking a clinic what it uses and why is a substantive question, not a formality.
What a real assessment looks at
A candidacy review is mostly a records exercise. What a physician is trying to establish:
- What is actually happening in the joint — from imaging, not from symptoms alone. Pain and structural damage correlate imperfectly.
- Where you are in the disease and how quickly it has moved.
- What has already been tried, and what happened. Prior injections and surgery change both the picture and the options.
- What else is going on medically — other conditions, medications, anything that changes the risk side of the equation.
- Whether something else should be addressed first. Sometimes the right answer is a different treatment, or the same treatment later.
A meaningful share of people who ask are told no, or “not yet.” That is not a failure of the process — it is the process working. An assessment that never says no is not an assessment.
How to find out, in the order that saves time
- Get your imaging from the last twelve months — the images themselves, not just the radiologist's report.
- Write down the history: how long, what makes it worse, what you have tried, what helped and for how long.
- List medications and supplements, with doses.
- Ask your own physician for their assessment first. A second opinion is more useful when the first is on record.
- Have your records reviewed and ask for the answer in writing, with the reasoning — including the reasoning for a no.
This costs you the time it takes to gather documents. It is the only way to replace a guess about yourself with an answer about yourself.
References
- Yin F, Wu H, Tong D, Luo G, Deng Z, Yan Q, Zhang Y. Contextual effects of mesenchymal stem cell injections for knee osteoarthritis: systematic review and meta-analysis of randomized controlled trials. Frontiers in Medicine, vol. 12, 17 September 2025. DOI: 10.3389/fmed.2025.1636181. Read the full text (open access).
- Tian Q, Qu, Cao, Zhang. Relative efficacy and safety of mesenchymal stem cells for osteoarthritis: a systematic review and meta-analysis of randomized controlled trials. Frontiers in Endocrinology, 10 June 2024. DOI: 10.3389/fendo.2024.1366297. Read the full text (open access).
Every figure quoted above is taken directly from the cited publications. Where a study reports uncertainty, we report it too.
