Published: October 2026 | Category: Musculoskeletal and Orthopedic Research
Reviewed Evidence
Reviewed Study 1: Cell-Based Versus Corticosteroid Injections for Knee Pain in Osteoarthritis: A Randomized Phase 3 Trial
Authors: Mautner K, Gottschalk M, Boden SD, et al.
Publication and Date: Nature Medicine. 2023;29(12):3120–3126.
DOI/Link: https://doi.org/10.1038/s41591-023-02632-w | PubMed
Reviewed Study 2: Comparison of Biological Characteristics of Human Adipose- and Umbilical Cord-Derived Mesenchymal Stem Cells and Their Effects on Delaying the Progression of Osteoarthritis in a Rat Model
Authors: Ju Y, Yi L, Li C, et al.
Publication and Date: Acta Histochemica. 2022;124(6):151911.
DOI/Link: https://doi.org/10.1016/j.acthis.2022.151911 | PubMed
Reviewed Study 3: Clinical Efficacy and Safety of the Intra-Articular Injection of Autologous Adipose-Derived Mesenchymal Stem Cells for Knee Osteoarthritis
Authors: Kim KI, Lee MC, Lee JH, et al.
Publication and Date: American Journal of Sports Medicine. 2023;51(9):2243–2253.
DOI/Link: https://doi.org/10.1177/03635465231179223 | PubMed
Reviewed Study 4: Umbilical Cord-Derived Mesenchymal Stromal Cells for Knee Osteoarthritis
Authors: Matas J, Orrego M, Amenabar D, et al.
Publication and Date: Stem Cells Translational Medicine. 2019;8(3):215–224.
DOI/Link: https://doi.org/10.1002/sctm.18-0053 | PubMed
Supporting Publications
Supporting Publication 1: Autologous Cell Injections for Knee Osteoarthritis Display Greater Responsiveness Than Allogenic Cellular Products and Corticosteroids in a Sex-Dependent Manner
Authors: Mautner K, Kaiser JM, Boggess B, et al.
Publication and Date: American Journal of Sports Medicine. 2025;53(12):2889–2897.
DOI/Link: https://doi.org/10.1177/03635465251365521 | PubMed
Supporting Publication 2: Cartilage Regeneration and Inflammation Modulation in Knee Osteoarthritis Following Injection of Allogeneic Adipose-Derived Mesenchymal Stromal Cells: A Phase II, Triple-Blinded, Placebo Controlled, Randomized Trial
Authors: Sadri B, Hassanzadeh M, Bagherifard A, et al.
Publication and Date: Stem Cell Research & Therapy. 2023;14(1):162.
DOI/Link: https://doi.org/10.1186/s13287-023-03359-8 | PubMed
Supporting Publication 3: Comparison of Mesenchymal Stem Cells From Bone Marrow, Umbilical Cord Blood, and Umbilical Cord Tissue in Regeneration of a Full-Thickness Tendon Defect In Vitro and In Vivo
Authors: Yea JH, Kim Y, Jo CH.
Publication and Date: Biochemistry and Biophysics Reports. 2023;34:101486.
DOI/Link: https://doi.org/10.1016/j.bbrep.2023.101486 | PubMed
Supporting Publication 4: Intervertebral Disc Repair by Allogeneic Mesenchymal Bone Marrow Cells
Authors: Noriega DC, Ardura F, Hernandez-Ramajo R, et al.
Publication and Date: Transplantation. 2017;101(8):1945–1951.
DOI/Link: https://doi.org/10.1097/TP.0000000000001484 | PubMed
Scientific Review By: Michael Healey, M.Ed., C.A.S.
Executive Summary and Key Takeaways
Primary Objective: Compare autologous MSC approaches with allogeneic donor-derived MSCs, with attention to cell expansion, biological activity, cartilage formation, paracrine signaling, inflammatory pathways, and outcomes in cartilage, disc, and tendon research.
Most Direct Human Evidence: A 480-patient randomized knee osteoarthritis trial directly compared autologous bone-marrow aspirate concentrate, autologous adipose-derived stromal vascular fraction, allogeneic umbilical-cord-tissue MSCs, and corticosteroid injection. At one year, no cellular treatment showed a significant pain advantage over corticosteroid injection.
Key Finding: Umbilical-cord MSCs often expand rapidly in laboratory studies, but faster growth has not established clinical superiority. Across the reviewed literature, symptom improvement is more consistent than objective evidence of cartilage restoration.
Clinical Caution: Cell therapies differ substantially. Bone-marrow aspirate concentrate and stromal vascular fraction are mixed point-of-care preparations, while culture-expanded MSC products are more selected and extensively manufactured. Results should not be treated as interchangeable.
Study Overview and Clinical Objectives
Current research does not establish that autologous or umbilical-cord mesenchymal stromal cells are superior for osteoarthritis. This review compares cell expansion, biological activity, manufacturing, dosing, and outcomes across these approaches.
The strongest head-to-head evidence comes from Mautner and colleagues. Although the paper is titled a phase 3 trial, its methods describe a phase 2/3 design. The four-arm, multicenter, single-blind trial assigned 480 patients to autologous bone-marrow aspirate concentrate, autologous adipose stromal vascular fraction, allogeneic umbilical-cord-tissue MSCs, or corticosteroid injection. Pain improved in every group, without a significant between-group difference in the co-primary pain outcomes at one year. The autologous treatments were mixed point-of-care preparations, not culture-expanded MSCs.
A 2025 post hoc analysis of 381 completers applied a 25% improvement threshold and found that patients receiving bone-marrow aspirate concentrate were more likely to achieve meaningful pain improvement than those receiving a corticosteroid injection, with treatment response varying by sex. This exploratory result qualifies but does not reverse the primary null comparison.
The remaining studies were weighted by design and clinical relevance. Kim, Matas, and Sadri provide controlled human knee evidence; Ju provides a direct preclinical source comparison; and Noriega and Yea extend the discussion to disc and tendon research.
Biological Mechanism & Science

Cell expansion affects how quickly and consistently a research dose can be produced. Umbilical-cord tissue is collected after birth and contains relatively young cells that can be expanded without a harvesting procedure for the recipient. Autologous adipose and bone-marrow products avoid donor-recipient mismatch, but their characteristics may vary with the patient’s age, health, inflammation, harvest method, and processing.
Expansion speed is a manufacturing advantage, not proof of greater clinical effectiveness.
Ju and colleagues found that umbilical-cord MSCs proliferated faster and formed larger pellets in chondrogenic culture than adipose-derived MSCs. Overall chondrogenic potential was comparable. In rats with surgically induced osteoarthritis, both sources reduced cartilage degeneration, preserved aggrecan, and reduced chondrocyte apoptosis. Neither source clearly outperformed the other, and two injections were not significantly better than one during the 12-week study.
MSC effects may involve more than direct differentiation into cartilage-like cells. Proposed paracrine mechanisms include the release of cytokines, growth factors, and extracellular vesicles that influence chondrocyte survival, synovial inflammation, immune-cell activity, angiogenic signaling, and extracellular matrix turnover.
Sadri and colleagues reported that IL-6 decreased significantly 12 weeks after allogeneic adipose-derived MSC treatment, while IL-10 increased at one week before decreasing at 12 weeks. These findings suggest short-term immunomodulatory activity, but the 40-patient study was too small to establish a durable mechanism or comparative advantage.
Subchondral bone remodeling and synovial-fluid inflammation are relevant to osteoarthritis but were not measured consistently across the reviewed trials. Clinical pain improvement may reflect changes in inflammation or joint signaling even when MRI does not demonstrate cartilage repair.
Key Findings and Patient Outcomes

In the original Mautner trial, none of the three cellular approaches produced a significant pain advantage over corticosteroid injection at one year. The finding weighs against claims that autologous or allogeneic cell source alone determines clinical benefit. It also demonstrates why direct comparative trials are more informative than comparisons across separate studies.
The investigators subsequently conducted a post hoc responder analysis of 381 participants who completed 12 months of follow-up. Responders were defined as patients achieving at least a 25% improvement in VAS or KOOS pain scores. Bone-marrow aspirate concentrate produced higher adjusted odds of a VAS response than corticosteroid injection (aOR, 2.02; 95% CI, 1.09–3.76). Response patterns also differed by sex. Because this was an exploratory analysis that included only patients who completed the study, the findings should be considered preliminary and do not change the trial’s primary result.
Kim and colleagues randomized 261 patients with Kellgren–Lawrence grade 3 knee osteoarthritis to a single injection of culture-expanded autologous adipose-derived MSCs or placebo. At six months, the MSC group showed greater improvement in pain and WOMAC function scores. Radiologic outcomes and MRI cartilage-defect changes did not significantly differ between groups. The trial supports short-term symptom improvement, not confirmed cartilage regeneration.
Matas and colleagues randomized 26 patients to hyaluronic acid, one 20-million-cell umbilical-cord MSC injection, or two 20-million-cell injections administered six months apart. At 12 months, the repeated-dose group showed better pain and function outcomes than the hyaluronic-acid group and generally better outcomes than the single-dose group. The small sample limits certainty, and the study did not compare umbilical-cord MSCs with a culture-expanded autologous product.
Sadri and colleagues randomized 40 patients to 100 million allogeneic adipose-derived MSCs or saline. The study reported clinical improvements, temporary changes in inflammatory biomarkers, and slight increases in cartilage thickness on MRI. Of the 12 measured femoral and tibial regions, statistically significant changes were limited to the medial-posterior and medial-anterior tibial regions. These exploratory imaging and biomarker findings require confirmation in larger trials.
Noriega and colleagues randomly assigned 24 patients with chronic lumbar disc degeneration to receive either donor-derived bone-marrow MSCs or a placebo procedure. Pain and disability improvement appeared concentrated in a responder subgroup representing approximately 40% of the cohort. Degeneration measured using Pfirrmann grading improved in the MSC group and worsened in controls over one year. The small trial supports further degenerative disc research but does not predict results for umbilical-cord or adipose MSCs in osteoarthritis.
The tendon evidence remains preclinical. Yea and colleagues found that umbilical-cord-tissue MSCs produced stronger tendon-related gene expression, greater tendon-like matrix formation, and better short-term histologic repair than bone-marrow or umbilical-cord-blood MSCs in laboratory and rat studies. These findings may help guide future research but do not establish benefit for human tendon disease.
Clinical Data Summary
|
Clinical Metric |
Study Specification |
|---|---|
|
Evidence Base |
Four controlled human knee osteoarthritis trials, one post hoc responder reanalysis, one randomized degenerative disc trial, and direct preclinical cartilage and tendon comparisons |
|
Largest Direct Comparison |
N=480; autologous bone-marrow aspirate concentrate, autologous adipose stromal vascular fraction, allogeneic umbilical-cord-tissue MSCs, and corticosteroid injection |
|
Responder Reanalysis |
N=381 completers; ≥25% improvement threshold; BMAC showed higher adjusted VAS responder odds than corticosteroid injection, with sex-related heterogeneity |
|
Culture-Expanded Trials |
Autologous adipose MSCs: N=261; allogeneic umbilical-cord MSCs: N=26; allogeneic adipose MSCs: N=40 |
|
Autologous Preparations |
Culture-expanded adipose MSCs, bone-marrow aspirate concentrate, and adipose stromal vascular fraction |
|
Allogeneic Products |
Culture-expanded umbilical-cord-tissue, adipose-derived, and bone-marrow-derived MSCs |
|
Expansion Finding |
Umbilical-cord MSCs proliferated faster than adipose MSCs in the direct laboratory comparison |
|
Chondrogenesis Finding |
Adipose and umbilical-cord MSCs showed comparable overall chondrogenic potential and similar effects in the rat osteoarthritis model |
|
Anti-Inflammatory Finding |
Allogeneic adipose MSCs were associated with short-term IL-6 and IL-10 changes in a small human trial |
|
Clinical Conclusion |
No clinical trial has established overall superiority of autologous or allogeneic MSC approaches |
|
Structural Evidence |
Human MRI and radiologic findings remain mixed and less consistent than improvements in pain and function |
Stemedix Clinical Context
Autologous and allogeneic MSC approaches involve different tradeoffs, not a proven winner. Autologous products avoid donor-recipient mismatch but require harvesting and individualized processing, and their properties may reflect the patient’s age, health, and inflammatory environment. Allogeneic products can be screened, expanded, characterized, stored, and delivered without a patient-specific manufacturing cycle, but donor selection, culture conditions, potency, immune response, and batch consistency remain important.
The evidence does not support treating laboratory viability as a clinical outcome. A product that expands rapidly and remains viable before injection may be easier to manufacture, but effectiveness also depends on product composition, dose, passage number, potency, delivery route, disease stage, joint environment, rehabilitation, and outcome selection.
Improved pain does not necessarily mean that damaged cartilage or bone is being repaired. Imaging and other objective measures are needed to determine whether a treatment actually changes the structure of the joint or slows the disease, rather than simply improving symptoms.
Overall, the literature supports continued controlled research. It does not establish that autologous cells are superior because they come from the patient or that umbilical-cord MSCs are superior because they are younger and expand more rapidly. Larger direct trials using standardized manufacturing and the same clinical, biomarker, and imaging endpoints are needed.
Frequently Asked Questions
Are autologous or umbilical-cord stem cells better supported for osteoarthritis research?
Neither has demonstrated overall superiority. In the largest direct trial, autologous bone-marrow aspirate concentrate, autologous adipose stromal vascular fraction, and allogeneic umbilical-cord-tissue MSCs did not produce significantly different pain outcomes from corticosteroid injection at one year.
A later post hoc analysis found higher VAS responder odds with bone-marrow aspirate concentrate than corticosteroid injection, but that exploratory finding does not establish overall superiority.
Are bone-marrow aspirate concentrate and stromal vascular fraction the same as culture-expanded MSCs?
No. They are mixed cell preparations produced with less manipulation and contain multiple cell types and biological components. Culture-expanded MSC products undergo additional selection and expansion. Results from one product category should not automatically be applied to another.
Do umbilical-cord MSCs grow better than adipose MSCs?
In the laboratory comparison reviewed here, umbilical-cord MSCs grew and multiplied faster. However, both cell sources showed a similar ability to form cartilage and produced similar results in the rat osteoarthritis model. This means that faster cell growth did not translate into better treatment results.
Does pain relief mean MSCs regrew cartilage?
No. Pain and function can improve without measurable cartilage restoration. In the large culture-expanded autologous adipose trial, clinical outcomes improved at six months, but MRI cartilage-defect changes did not significantly differ from placebo.
What is the role of paracrine signaling in cartilage repair?
MSCs may release cytokines, growth factors, and extracellular vesicles that affect inflammation, immune activity, chondrocyte survival, and extracellular matrix turnover. These signals may help create a more favorable joint environment without the injected cells becoming permanent replacement cartilage.
Is there clinical evidence for stem cell therapy in degenerative disc disease?
There is limited early evidence. A randomized 24-patient trial of allogeneic bone-marrow MSCs reported improvement in a responder subgroup representing approximately 40% of the cohort. Pfirrmann grading improved in the MSC group and worsened in controls over one year. The study was small, and its findings cannot be assumed to apply to other cell sources or products.
Regulatory Disclaimer
This review is provided by Stemedix for educational and informational purposes only and represents an academic commentary on published literature. The featured studies evaluate experimental procedures. Regenerative cell therapies discussed are not FDA approved for osteoarthritis, degenerative disc disease, or tendon repair unless explicitly stated otherwise. Consult a board-certified physician before making medical decisions.
References
- Mautner K, Gottschalk M, Boden SD, et al. Cell-based versus corticosteroid injections for knee pain in osteoarthritis: a randomized phase 3 trial. Nature Medicine. 2023;29(12):3120–3126. doi:10.1038/s41591-023-02632-w. https://pubmed.ncbi.nlm.nih.gov/37919438/
- Mautner K, Kaiser JM, Boggess B, et al. Autologous cell injections for knee osteoarthritis display greater responsiveness than allogenic cellular products and corticosteroids in a sex-dependent manner. American Journal of Sports Medicine. 2025;53(12):2889–2897. doi:10.1177/03635465251365521. https://pubmed.ncbi.nlm.nih.gov/41014273/
- Ju Y, Yi L, Li C, et al. Comparison of biological characteristics of human adipose- and umbilical cord-derived mesenchymal stem cells and their effects on delaying the progression of osteoarthritis in a rat model. Acta Histochemica. 2022;124(6):151911. doi:10.1016/j.acthis.2022.151911. https://pubmed.ncbi.nlm.nih.gov/35764040/
- Kim KI, Lee MC, Lee JH, et al. Clinical efficacy and safety of the intra-articular injection of autologous adipose-derived mesenchymal stem cells for knee osteoarthritis: a Phase III, randomized, double-blind, placebo-controlled trial. American Journal of Sports Medicine. 2023;51(9):2243–2253. doi:10.1177/03635465231179223. https://pubmed.ncbi.nlm.nih.gov/37345256/
- Matas J, Orrego M, Amenabar D, et al. Umbilical cord-derived mesenchymal stromal cells for knee osteoarthritis: repeated MSC dosing is superior to a single MSC dose and to hyaluronic acid in a controlled randomized Phase I/II trial. Stem Cells Translational Medicine. 2019;8(3):215–224. doi:10.1002/sctm.18-0053. https://pubmed.ncbi.nlm.nih.gov/30592390/
- Sadri B, Hassanzadeh M, Bagherifard A, et al. Cartilage regeneration and inflammation modulation in knee osteoarthritis following injection of allogeneic adipose-derived mesenchymal stromal cells: a phase II, triple-blinded, placebo controlled, randomized trial. Stem Cell Research & Therapy. 2023;14(1):162. doi:10.1186/s13287-023-03359-8. https://pubmed.ncbi.nlm.nih.gov/37316949/
- Yea JH, Kim Y, Jo CH. Comparison of mesenchymal stem cells from bone marrow, umbilical cord blood, and umbilical cord tissue in regeneration of a full-thickness tendon defect in vitro and in vivo. Biochemistry and Biophysics Reports. 2023;34:101486. doi:10.1016/j.bbrep.2023.101486. https://pubmed.ncbi.nlm.nih.gov/37234487/
- Noriega DC, Ardura F, Hernandez-Ramajo R, et al. Intervertebral disc repair by allogeneic mesenchymal bone-marrow cells: a randomized controlled trial. Transplantation. 2017;101(8):1945–1951. doi:10.1097/TP.0000000000001484. https://pubmed.ncbi.nlm.nih.gov/27661661/
