Chronic Pain: What Stem Cell Therapy Can and Cannot Do Today
Human studies exist for selected causes of chronic pain, especially knee osteoarthritis and disc-related low back pain. The results are mixed, and cell products are not interchangeable. The U.S. FDA states that regenerative medicine therapies are not approved for chronic pain or orthopedic conditions such as osteoarthritis, disc disease, back pain or knee pain.[1]
What the controlled data shows
Use the comparison group, not the headline.
Across 16 randomized knee-osteoarthritis trials with 807 participants, the 3–6 month average pain difference favored MSCs by 0.74 cm on a 10 cm VAS. The review used 1.5 cm as the minimally important difference.[2]
In RESPINE, 114 people with chronic low back pain were randomized to 20 million allogeneic bone-marrow MSCs or sham. At 12 months, the responder rate was 74% versus 69%; p=0.77, so the primary endpoint was not met.[4]
Does the research fit your diagnosis?
“Chronic pain” is too broad to predict a result.
The first question is not which cell product sounds strongest. It is whether the diagnosis matches a condition studied in controlled human trials.
Evidence is mixed. A 2024 review of 16 randomized trials and 807 participants found little to no clinically important average benefit at 3–6 months, while a 2026 meta-analysis of 11 randomized trials and 811 patients reported statistically significant improvement in several pooled outcomes. The reviews used different trial sets, comparators and outcome measures, so neither gives a personal expected result.[2][3] Review knee-specific evidence.
Sham-controlled studies exist, but results remain unsettled. RESPINE reported a 12-month responder rate of 74% with MSCs versus 69% with sham (p=0.77), and DREAM found similar clinical improvement in both groups through six months.[4][5]
Evidence from knee or disc studies cannot be assumed to apply to nerve injury, spinal cord injury or post-surgical neuropathic pain. For neurological injury, see the spinal cord injury evidence review.
Structural joint or disc trials do not establish benefit for widespread pain syndromes. A diagnosis-specific controlled evidence base is required before claims can be generalized.
Evidence map
What the data can—and cannot—support.
At 3–6 months, the 2024 review found a −0.74 cm pain difference on a 10 cm VAS versus a 1.5 cm minimally important difference, and +2.23 points in SF-36 physical function versus a 10-point minimally important difference. Both main estimates were rated moderate certainty.[2]
A newer meta-analysis of 11 RCTs and 811 patients reported a WOMAC mean difference of −11.05 at 12 months and IKDC +2.88 overall. The WOMAC estimate came from two studies, and the review pooled different controls and cell protocols.[3]
RESPINE did not meet its primary 12-month endpoint. DREAM found no significant clinical advantage over sham at any measured time point through six months, even though some MRI measures changed.[4][5]
In a 63-person intradiscal trial, at least 50% pain relief at 12 months occurred in 44% with BMC and 56% with sham; there was no significant difference. BMC is not the same product as culture-expanded MSCs.[6]
Already received a quote, protocol or cell-count claim?
Send the exact wording you were given. We can help you separate what is documented from what still needs a clear answer.
Know what is actually being offered
“Stem cell” can describe very different products.
Do not compare offers by cell count alone. Ask for the source, processing method, dose, route and regulatory status in writing.
A mixed concentrate from bone marrow. It is not the same product as purified or culture-expanded mesenchymal stromal cells.
Cells are processed and expanded before use. Manufacturing, characterization, sterility testing and regulatory requirements become central questions.
Allogeneic products add questions about donor screening, processing, immune response, traceability and batch quality.
Do not assume a marketed product contains living stem cells or is approved simply because it is described as “regenerative.” The FDA specifically warns about unapproved products including Wharton’s jelly, amniotic products and exosomes.[1]
For a plain-language explanation of how cells are processed and what can be verified in Japan, read the CPC Cell Processing Center guide.
What the clinical data shows
Four datasets that are useful when checking a “success rate.”
MSC injection versus placebo or usual care.
3–6 month pain: −0.74/10 versus a 1.5/10 minimally important difference. Physical function: +2.23/100 versus a 10-point minimally important difference.[2]
20 million allogeneic BM-MSCs versus sham.
Primary responder endpoint: 74% MSC versus 69% sham; p=0.77.[4]
Autologous BM-MSCs versus sham for multilevel disc degeneration.
Clinical VAS, ODI and SF-36 outcomes improved similarly in both groups; no significant between-group clinical advantage through six months.[5]
Bone marrow concentrate versus sham.
At 12 months, ≥50% pain relief: 44% BMC versus 56% sham. ≥30% ODI improvement: 56% versus 61%. No significant differences.[6]
Main 3–6 month pain and function effects were below prespecified minimally important differences; main evidence was moderate certainty.[2]
Reported significant pooled improvements including WOMAC −11.05 and IKDC +2.88. Different trial inclusion, comparators, endpoints and cell protocols can produce different pooled estimates.[3]
These figures describe study groups, not an individual forecast. Age, diagnosis, disease severity, product, dose, route, comparator and follow-up all affect how closely a study matches a person’s situation.
What published cost data shows
A published average is not a stem-cell quote.
A 2024 analysis identified 1,085 orthopedic regenerative-medicine provision plans in Japan. Price data were available for 1,014 plans. The mean was US$2,490 ±4,170 and the observed range was US$0–36,670; the zero-price plans were research funded.[7]
Why this cannot be used as a stem-cell price: among 1,085 orthopedic plans, 78% used PRP, 20% adipose-derived stem cells and 1% bone-marrow-derived stem cells. Class II plans averaged US$3,600 ±4,840. The dataset shows variation, not a standard fee for chronic pain.[7]
Cell source, processing, dose, procedure, guidance, testing, follow-up and rehabilitation should be itemized.
Concentrated marrow, culture-expanded MSCs and donor-derived products are not interchangeable.
Have a study, brochure or quote in front of you?
Send the document as written. We can map its numbers, product description and source references into a shorter comparison.
What “better” actually looks like
Define “better” before judging the result.
There is no single evidence-based recovery timeline because products, injection routes and diagnoses differ. Controlled studies measure change over months and use predefined pain or function thresholds.
Trials commonly record pain, function and adverse events at several points such as 3, 6 and 12 months.
The 2024 review used 1.5 cm on a 10 cm VAS as the minimally important pain difference.[2]
Defined as VAS improvement of at least 20% and 20 mm, or ODI improvement of 20%, from baseline to month 12.[4]
Walking, sitting, lifting, sleep, work and disability can matter more than a small change in one pain score.
Survival rate is not a standard effectiveness endpoint for chronic pain. The useful outcomes are pain, function, disability, quality of life, medication use, adverse events and how long any change lasts. If another disease affects prognosis, survival belongs to that disease—not to a generic stem-cell success rate.
Don’t skip the safety questions
“Your own cells” does not automatically mean risk-free.
The FDA has reported serious harms from unapproved regenerative products, including infections, unwanted inflammatory or immune reactions, neurological events and tumor formation. These reports cover different products and uses and are not a chronic-pain-specific event rate.[1]
Exact diagnosis. What anatomical pain source is being targeted and how was it confirmed?
Exact product. What is the cell source, processing method, dose and route?
Matching evidence. Are there controlled human studies using the same product for the same diagnosis?
Regulatory status. Is it approved for this indication, investigational, or provided under another legal framework?
Complication plan. Who manages infection, inflammatory reactions or unexpected neurological symptoms?
Written follow-up. Which pain, function and safety outcomes will be recorded, and when?
Japan: what “regulated” actually means
A filed regenerative-medicine plan is not the same thing as product approval.
Japan uses two regulatory frameworks: the Act on the Safety of Regenerative Medicine covers research and medical practice involving cell therapy, while the PMD Act regulates commercialization of regenerative medical products.[8]
How we can help
Turn reports, quotes and technical claims into one clear brief.
We provide information support, record organization and medical-language assistance. We do not diagnose, prescribe, administer cell products or guarantee eligibility or outcomes.
Your stated diagnosis is compared with relevant controlled human studies so you can see where evidence fits, conflicts or is missing.
Imaging reports, previous procedures, medication history and current function are organized into a concise chronology.
Cell source, dose, route, evidence, regulatory wording, cost and follow-up are placed side by side instead of buried in sales material.
You receive a practical list of questions to take to a licensed medical professional before making a decision.
What to read next
Go deeper only where it changes your decision.
These topics connect directly to diagnosis, cell processing, regulation or record preparation.
Quick answers
Clear limits are as useful as positive findings.
Can stem cells cure chronic pain?
No cure can be promised. Evidence differs by diagnosis and product. Controlled studies in knee osteoarthritis and disc-related low back pain are mixed, and symptom improvement does not prove permanent disease modification.[2][4]
Does less pain mean cartilage or a disc regenerated?
No. Symptoms and structural imaging are separate outcomes. In DREAM, some MRI measures changed after BM-MSC injection, but VAS, ODI and SF-36 outcomes did not show a significant clinical advantage over sham through six months.[5]
How do I know whether a study applies to me?
Match the exact diagnosis, target tissue, cell source, processing method, dose, route, comparator and follow-up. A knee osteoarthritis study cannot establish benefit for fibromyalgia, neuropathic pain or unexplained widespread pain.
Is there a reliable success rate?
No single percentage represents “stem cell treatment for chronic pain.” RESPINE reported 74% responders with MSCs and 69% with sham at 12 months (p=0.77). In the BMC disc trial, ≥50% pain relief at 12 months was 44% with BMC and 56% with sham. The comparator changes how a percentage should be interpreted.[4][6]
How long does recovery take?
There is no single evidence-based timetable because the procedure, target and product differ. Trials judge outcomes over several months. A useful follow-up plan defines activity guidance, pain and function measures, adverse-event checks and the dates they will be reviewed.
Can these studies tell me whether I can avoid surgery or stop pain medication?
Not reliably. The controlled studies cited here do not establish a universal rate for avoiding surgery or stopping medication. If either is an important goal, it should be stated before any procedure and tracked as a separate outcome with a licensed clinician.
Is there a survival rate?
Survival is not a standard effectiveness endpoint for chronic pain. Pain, function, disability, quality of life, medication use, adverse events and durability are more relevant. Survival questions belong to any separate underlying disease that affects prognosis.
Does availability in Japan mean a product is approved?
No. Japan’s Act on the Safety of Regenerative Medicine covers research and medical practice using cell therapy, while the PMD Act regulates commercialization of regenerative medical products. A reviewed or filed provision plan is not the same as marketing approval for a product.[8]
What should I send for an information review?
Your current diagnosis, symptom duration, imaging or test reports, previous procedures, medication history, current limitations and any quote, brochure, protocol or cell-count claim you already have. Personal identifiers can be removed when they are not needed.
Check the sources
Read the evidence behind the claims.
Regulators, government sources, randomized trials and peer-reviewed systematic reviews are prioritized. Numbers are shown with the limits needed to interpret them.
Now what?
Send your diagnosis, key reports and any offer you are considering.
We’ll organize your information, match the product and diagnosis to published human evidence, flag unanswered questions and prepare a shorter brief for discussion with licensed medical professionals.
Educational and information-support content only. This service does not diagnose, prescribe, provide medical care or guarantee suitability or outcomes. New weakness, loss of bladder or bowel control, major trauma, fever with severe pain, rapidly worsening neurological symptoms or other urgent concerns require direct medical evaluation.