Yes, stem cells can replace damaged cells in specific treatments, but they do not automatically rebuild any injured organ. Blood stem cell transplants restore blood-cell production Stem cell transplant. For burn-related damage to the corneal surface, a stem-cell-containing treatment produced a stable surface in 75 of 104 patients (72%) at one year Holoclar. Whether another treatment replaces cells depends on the cell type, the tissue, delivery, and human evidence—not simply the words “stem cell.”
Replacement means that new cells survive in the relevant tissue and take over a missing function. Symptom relief alone does not establish that this happened. The examples below separate established transplantation, experimental replacement, and treatments whose benefits do not demonstrate replacement.

What counts as replacement
Stem cells can renew themselves and develop into specialized cells Stemcells. That capacity is not proof that a particular injection rebuilds tissue in a patient. Some treatments transplant stem cells; others first grow specialized cells from them in a laboratory.
Autologous just means the cells come from you. Allogeneic means they come from a donor. Engraftment means transplanted cells have settled in and started working. In a blood transplant, the marrow is making new blood cells Stem cell transplantStem cell engraftment.
For a replacement claim you can trust, five things have to check out:
- Know exactly which cells are missing. “Damaged tissue” is not a diagnosis. A burn destroys the epidermis and sometimes the dermis. A heart attack kills heart muscle cells. Osteoarthritis degrades the cartilage that lines the joint. These are not interchangeable targets.
- The cells must reach the target. Giving cells into a joint, a vein, or a surgical graft does not establish the same distribution or engraftment.
- The cells must survive and do their job. Blood replacement means the new cells produce white cells, red cells, and platelets. In the cornea, success looks like a stable surface. For pancreatic islets, the cells must produce insulin.
- The treatment must address immune compatibility. Check whether matching or immune-suppressing medicines are needed and what risks they add.
- The benefit must be measured in a way that separates replacement from other effects. A pain score cannot show that cartilage regrew.
Replacing one cell population is not the same as curing a disease or rebuilding an entire organ. A stable corneal surface, for example, does not mean that all causes of impaired vision have been reversed.
Replacement versus repair and support
A scar can close a wound without recreating the original tissue. Likewise, a cell that changes nearby immune activity has not necessarily become the missing cell itself.
| Claim | What it means | A concrete example | What it does not prove |
|---|---|---|---|
| Replacement | New cells become the missing cell type and perform its job | Blood stem cells rebuild marrow production | That the same cells rebuild a knee or a brain |
| Repair | The tissue is patched, sometimes with scar, and function may improve | Scar formation after a heart attack | That new heart muscle has formed |
| Support | Cells may influence nearby tissue through signaling rather than becoming replacement cells | A proposed signaling mechanism that needs testing | That signaling produces a clinical benefit |
| Immune modulation | Cells change immune activity | Ryoncil for graft-versus-host disease, thought to work through an immune effect | That damaged tissue was replaced |
Where cell replacement is used or being tested
These examples have different evidence and regulatory pathways. An established blood transplant, an EU-authorized corneal product, a U.S. humanitarian-device skin graft, and an experimental islet infusion are not interchangeable treatments.
Blood and the immune system
High-dose chemotherapy or radiation can destroy blood-forming cells in your bone marrow. In a blood stem cell transplant, healthy cells are infused through a vein. The infusion takes 1 to 5 hours; engraftment and recovery take longer. The cells travel to the marrow and begin producing white cells, red cells, and platelets. Depending on the transplant, the cells come from the patient or a donor Stem cell transplant.
Doctors use it for selected patients with leukemia, lymphoma, multiple myeloma, myelodysplastic syndromes, aplastic anemia, sickle cell disease, and some autoimmune diseases Stem cell transplant. The process takes months. It starts with chemotherapy and sometimes radiation. The risks are real: infection, bleeding, infertility, cataracts, secondary cancers, organ damage, and graft-versus-host disease, a complication in which donor immune cells attack the recipient. Full immune recovery takes several months after an autologous transplant and one to two years after an allogeneic or identical-twin transplant Stem cell transplant.
This is evidence of blood-cell replacement, not evidence that the same procedure regenerates cartilage, heart muscle, or spinal cord tissue.
The cornea
The cornea’s surface is constantly renewed by limbal stem cells living at its edge. After a chemical or physical burn those cells can be wiped out. Doctors call that loss limbal stem cell deficiency. The surface then turns unstable and vision drops.
Holoclar is a treatment approved in the European Union. It starts with your own limbal cells. They grow in a lab and are then implanted as a thin sheet. The European Medicines Agency describes it as a stem-cell treatment used “to replace damaged cells on the surface (epithelium) of the cornea” Holoclar. In a study that looked back at patients’ records, 75 of 104 patients (72%) had a successful implant at one year. Success meant a stable corneal surface with no defects and little or no ingrowth of blood vessels Holoclar. It calls for a biopsy from a patch of undamaged limbus (at least 1–2 mm²) and a qualified eye surgeon to implant it Holoclar.
The EU-approved use covers adults with moderate-to-severe limbal stem cell deficiency caused by burns. A stable surface is not a guarantee of normal vision. Reported side effects include eye pain, corneal surface defects, bleeding on the eye surface, and eyelid inflammation Holoclar.
Skin
Epicel is a cultured epidermal autograft authorized in the United States for deep dermal or full-thickness burns covering at least 30% of total body surface area. It can be used with split-thickness autografts or alone when those grafts are not an option Hde.cfm?id=H990002. It is authorized through a humanitarian device exemption (HDE). This pathway requires safety and probable benefit, but is exempt from the usual requirement to demonstrate a reasonable assurance of effectiveness Getting humanitarian use device market.
This is an epidermal, or outer-skin, graft—not a general-purpose stem cell injection. Its authorization should not be read as evidence that it recreates every layer and structure of normal skin.
Islets in a trial
A 2025 phase 1–2 study tested zimislecel, allogeneic islet cells grown from stem cells, in people with type 1 diabetes. The cells were infused into the portal vein, which carries blood to the liver. Fourteen participants completed at least 12 months of follow-up; 12 received a full single dose. All 14 had undetectable C-peptide before treatment and detectable C-peptide afterward, supporting engraftment and insulin-producing function. C-peptide is a marker of insulin production. All 12 full-dose participants met the reported severe-hypoglycemia and glycated hemoglobin (HbA1c, a measure of average blood glucose) outcome: no severe events and HbA1c below 7%. Ten of 12 (83%) were insulin-independent at day 365 40544428.
All participants received immune-suppressing drugs without glucocorticoids. Neutropenia, a low white blood cell count, was the most common serious adverse event and occurred in 3 participants. Two deaths occurred: one from cryptococcal meningitis and one from progression of pre-existing neurocognitive impairment. The study was small and uncontrolled; the reported analyses were interim and not prespecified. It was funded by Vertex Pharmaceuticals. The authors called for further clinical investigation 40544428.
The distinction is important: this report supports experimental functional cell replacement. It does not establish lifelong insulin independence, justify stopping prescribed insulin, or show that a different clinic infusion for diabetes has the same effect. The 2025 trial results are not themselves a marketing authorization.
Approval does not always mean replacement
Ryoncil illustrates why clinical benefit and replacement must be judged separately. It was the first mesenchymal stromal cell (MSC) therapy approved by the U.S. Food and Drug Administration (FDA), for steroid-refractory acute graft-versus-host disease in children aged 2 months and older. This complication can follow donor blood stem cell transplantation; “steroid-refractory” means that it has not responded to steroid treatment. In the single-arm study, 54 children received treatment and 38 responded (70%) at day 28: 16 complete and 22 partial responses. Median response duration was 54 days. These are disease-response outcomes, not measurements of tissue replacement Fda approves remestemcel l rknd steroid refractory acute graft versus host disease pediatricFdaDrugXsl.cfm?setid=a4e0918f 7444 4694 adeb d38d98345659&type=display.
Its label says the mechanism of action “is not clear but may be related to immunomodulatory effects.” Laboratory data show inhibition of T-cell activation. The label also states that no studies have evaluated Ryoncil’s pharmacokinetics, or its behavior in the body. That is a product-specific evidence gap, not proof that all infused MSCs behave identically FdaDrugXsl.cfm?setid=a4e0918f 7444 4694 adeb d38d98345659&type=display.
The label warns about infusion reactions, transmission of infectious agents, and ectopic tissue formation—tissue forming where it should not. MSCs can differentiate into bone, cartilage, and fat cells, but this capacity does not establish that they replace the desired tissue safely FdaDrugXsl.cfm?setid=a4e0918f 7444 4694 adeb d38d98345659&type=display.
Does an IV infusion reach damaged tissue?
Blood stem cell transplantation shows that an IV route can work when the cells reach and engraft in the marrow Stem cell transplant. It does not establish that a different cell product reaches damaged cartilage, heart muscle, or brain tissue. Delivery and survival must be demonstrated for the treatment being offered.
In one mouse experiment using 500,000 labeled MSCs per animal, viable cells were detected in the lungs for up to 24 hours, but not in the liver, spleen, kidneys, or bone marrow at the sampled times; cell debris was later detected in the liver. This illustrates a possible delivery barrier in that animal model. It does not establish human cell distribution or clinical efficacy Full.
If a clinic says its cells “home to damage,” ask for human evidence showing where that specific product goes, how long the cells survive, and whether they perform the missing function. A general account of stem cell biology does not answer those questions.
Where replacement was not shown
Knee osteoarthritis
A 2025 Cochrane review included 25 trials and 1,341 participants. Eight trials (459 participants) compared a stem cell shot in the knee with a placebo injection. At six months, pain on a 0–10 scale was 1.2 points better with stem cells: 3.3 versus 4.5 (7 trials, 445 participants). Function on a 0–100 scale, where lower is better, was 14.2 points better: 32.1 versus 46.3 (7 trials, 432 participants). Certainty was low. No included study assessed radiographic disease progression; that is missing structural evidence, not a finding that cartilage regrowth was measured and ruled out. The review searched through September 15, 2023 CD013342_what are benefits and risks stem cell injections knee osteoarthritis.
A 2024 meta-analysis of 16 trials (807 people) concluded that MSC therapy probably produces little to no important improvement in pain or function at 3–6 months. The average pain difference was 0.74 cm on a 10-cm scale, below the review’s 1.5-cm threshold for an important difference. That is an average-effect threshold, not a rule that no individual can notice improvement. This main conclusion had moderate certainty. There was low-certainty evidence of more adverse events (risk ratio 2.67, 95% confidence interval 1.19–5.99) and knee pain or swelling (risk ratio 1.58, 95% confidence interval 1.04–2.38) 38777213.
The reviews reach different conclusions about the size and importance of symptom improvement. Neither establishes that injected cells replaced lost cartilage.
Heart failure
A 2016 Cochrane review examined bone-marrow-derived cell therapy for chronic ischemic heart disease and heart failure: 38 randomized trials and 1,907 participants. Low-quality evidence suggested lower mortality at 12 months or longer (risk ratio 0.42, 95% confidence interval 0.21–0.87; 491 participants in 9 trials), with signals for fewer non-fatal heart attacks and arrhythmias. There was no clear evidence of reduced heart-failure rehospitalization or the combined outcome of death, heart attack, or rehospitalization CD007888_stem cell treatment chronic ischaemic heart disease and congestive heart failure.
Results for pump function varied by measurement and analysis. The authors reported some improvements in ejection fraction overall, but the long-term MRI comparison showed no clear difference (mean difference −1.60 percentage points, 95% confidence interval −8.70 to 5.50; only 25 participants in 1 trial). The search ended in December 2015, so this is historical evidence, not a complete account of current heart-cell research. Neither a mortality signal nor a change in pumping performance alone establishes that new cells became heart muscle CD007888_stem cell treatment chronic ischaemic heart disease and congestive heart failure.
Do clinic filings prove replacement?
No. In its U.S. consumer alert, updated in April 2024, the FDA warned about regenerative products marketed without approval and without adequate evidence of safety or effectiveness Consumer alert regenerative medicine products including stem cells and exosomes. Its patient information reports blindness, tumor formation, and serious infections after unapproved products, and warns about unwanted immune reactions and unintended tissue growth. These reports do not establish how often a complication occurs Important patient and consumer information about regenerative medicine therapies.
Japan separates provision of regenerative medical care from marketing authorization of regenerative medical products. A provider may file a provision plan under the Act on the Safety of Regenerative Medicine; that is not the same route as product authorization under the Pharmaceuticals and Medical Devices Act. The frameworks took effect in November 2014. Filing alone is not the entire compliance process or a finding of efficacy 0003. The Ministry of Health, Labour and Welfare publishes provider and plan details for patients to check 0000186471.
A published plan is not a product approval, so “registered clinic” and “approved therapy” are not interchangeable.
If a clinic says it’s approved or registered, ask which product, under which approval or plan number, and for which condition and country. Then ask separately what human evidence supports replacement. The FDA also cautions that a ClinicalTrials.gov listing or FDA registration does not mean that a product is legally marketed Important patient and consumer information about regenerative medicine therapies.
What proof would change the answer
Evidence needs to connect the delivered cells to the restored tissue or function. No single measurement fits every organ, and imaging or better function alone may not establish where new cells came from.
- A biopsy should show that the new cells are the right type and are in the right place.
- Imaging can assess durable structural change, but does not necessarily identify the cells responsible.
- A function test should support a role for the transplanted cells. Blood production after engraftment and the appearance of C-peptide after the islet infusion are examples; interpretation depends on the pretreatment findings and other treatments.
- Controlled trials help distinguish clinical benefit from placebo, natural recovery, and rehabilitation. Mechanistic measurements are still needed to establish replacement.
- Safety data should cover serious harms along with mild side effects.
Animal experiments can investigate survival and distribution, but cannot establish human tissue regeneration. Trial registration records a study, not its success. PubMed indexing is not a quality endorsement: the National Library of Medicine states that it does not review, evaluate, or judge individual articles Disclaimer.
A useful rule: symptom improvement can matter to a patient without proving that transplanted cells replaced damaged tissue.
Questions that test a replacement claim
- Which specific cells are missing, and how was that established?
- What exactly will be given—cell type, source, dose, route, and processing?
- Is this a licensed product, a clinical trial, or a provision plan? What is the product, approval, or plan number?
- Has anyone shown that these cells survive and become the missing cell type in people like me? Was that measured by biopsy, imaging, or a function test?
- How many patients were studied, compared with what, for how long, and what happened to them?
- What are the known harms, and how will you monitor me?
- What proven treatment would I delay or give up?
Get the answers in writing before you pay.
If a knee injection is offered
Ask to see the study behind the word “regeneration.” If its result is a pain-score improvement, check whether cartilage structure or the identity of new cells was measured too. In the Cochrane review above, the 1.2-point average pain advantage did not come with an assessment of radiographic progression. A symptom result cannot fill that evidence gap CD013342_what are benefits and risks stem cell injections knee osteoarthritis.
Next, compare the study’s cell source, processing, dose, and injection route with the proposed treatment. If those details differ or are missing, ask why its results should apply. Take the written answers to a clinician who knows your diagnosis and can discuss established alternatives; do not stop or delay prescribed care on the strength of a replacement claim.
Considering cell-based care in Japan?
Questions about process, eligibility, or required documents can be sent through the contact form.
The bottom line on replacement
Stem cell replacement is real, but tissue-specific. Holoclar’s 75-of-104 one-year surface outcome concerns burn-related corneal damage—not whole-eye regeneration Holoclar. The islet study’s 10-of-12 insulin-independence result concerns an experimental treatment with immune suppression, not a general diabetes infusion 40544428.
For any other claim, the decisive question is not whether stem cells can specialize in a laboratory. It is whether this product, delivered this way, has been shown in comparable people to survive and perform the missing job—and at what risk.
This is general information, not an assessment of your eligibility for treatment. A licensed clinician who has reviewed your records should help you weigh the evidence, risks, and alternatives.




