x-ray of knee with osteoarthritis

Watch the September 22, 2022, NIH VideoCast - Cartilage Preservation and Restoration for Knee Osteoarthritis.

See the NIH Record article: "NIAMS Hosts Roundtable on Regenerative Medicine in Knee OA"


Current medicinal treatments for knee osteoarthritis (OA) do not modify the disease course. A major barrier to the development of disease-modifying interventions is the nature of knee OA progression, where initial articular cartilage damage triggers a vicious cycle of molecular events and tissue interactions in a downward spiral and rapidly progresses to diffuse knee OA. Stopping this elaborate process is a challenge, making preservation and restoration of articular cartilage a critical topic of investigation.

Regenerative medicine is an emerging area of science that holds great promise for treating and even curing a variety of injuries and diseases. Over the last decade, a variety of regenerative medicine approaches that either preserve articular cartilage in people who started with a focal cartilage injury and thus are at a higher risk of developing knee OA or restore defective articular cartilage in patients with symptomatic knee OA have been investigated. The clinical use of biologics, including intra-articular injections of platelet-rich plasma (PRP) and bone marrow- or tissue-derived mesenchymal stem cells (MSC), are becoming increasingly prevalent as treatment strategies for patients with knee OA. Some molecules such as fibroblast growth factor (FGF) and exogenous cartilage matrixes that can be fixated over the perforated subchondral bone and act as scaffolds for released MSC and growth factors are in late stages of clinical development and show promise for knee OA. Moreover, extracellular vesicles [either microvesicles (MVs) or exosomes (Exo)] and gene therapy have improved repair in animal studies. Such progress notwithstanding, much work remains to be done toward the development of safe and effective regenerative medicine products.

In addition to the barrier noted above, therapeutic development for knee OA has been constrained by several factors including disease heterogeneity, the multitude of knee OA risk factors, the poor association between structural changes and clinically meaningful endpoints, discordances between preclinical and human models, and regulatory hurdles. Biologic therapies, for example, continue to be debated as a definitive treatment for symptomatic knee OA, although some clinical studies have demonstrated that they are relatively safe and lead to significant improvement of clinical and functional outcomes and patient’s quality of life. This debate is partially due to a lack of evidence from robust, well-designed clinical trials with reproducible methodology. Most level I evidence studies so far have issues such as small sample sizes, potentially inappropriate control cohorts, or relatively short-term follow-ups. Most importantly, there are also concerns regarding the heterogeneity and lack of standardization and consensus in the optimal preparations, source, delivery methods, and dosing of biologics for therapy. In July 2020, the U.S. Food and Drug Administration published guidance on the “Minimal Manipulation and Homologous Use” of cell and tissue products. With stricter interpretation of minimal manipulation and homologous use criteria for cell-based therapies from the FDA, many cell-based therapies that used to be self-designated as belonging to the 361 pathway (which is less regulated and not required to have clinical data) are likely to be reclassified as belonging to the 351 regulatory pathway (which is more regulated and required to have clinical data).

To better understand the regenerative medicine landscape of challenges, gaps, and opportunities, NIAMS held a roundtable to discuss expert perspectives on future directions of the field. The meeting brought together researchers from NIH intramural and extramural communities as well as a patient representative. This roundtable was videocast, and the recording of the meeting is available on the NIH videocast archive. In advance of the meeting, the Institute collected written input from the participants, who also canvassed their research communities, on the questions below.

  1. What evidence supports the safety and efficacy of the clinical use of state-of-the-art regenerative therapies for knee OA?
  2. What are the emerging approaches and areas of research offering new opportunities for cartilage preservation and restoration in knee OA?
  3. What are the main challenges and gaps impeding clinical translations of these emerging opportunities?
  4. Other than more funding, how can NIAMS facilitate progress in this research?

The community’s responses greatly enriched the discussion. Although not all responses submitted in advance were discussed at the roundtable, NIAMS staff read each comment provided. The Institute greatly appreciates the community’s input.

1Times are noted in approximate hours and minutes on the videocast. The link does not take the viewer directly to the timepoint in the videocast.

Presentation: Patients’ Perspectives in OA Management (Amye Leong, 29:331)

The meeting began with welcomes from the NIAMS Director, Dr. Lindsey Criswell, the NIAMS Deputy Director, Dr. Robert Carter, and the meeting co-chairs, Dr. Constance Chu and Dr. Scott Rodeo.

Ms. Amye Leong gave a patient perspective of the knee OA treatment and management landscape as well as her view of unmet needs. Ms. Leong urged participants to remember that the cells and tissues they study represent real people who could be helped by the ideas and research behind regenerative medicine. She addressed public misconceptions regarding knee OA, its treatment, and the commitment of researchers to understanding the condition. Studies have revealed racial disparities in knee OA prevalence, and clinical trials with participants from diverse backgrounds are still needed.

Discussion: Barriers for Patients Enrolling in Clinical Trials (43:581)

The group discussed patient education about clinical trials and the need for patient advocates who reflect the diversity of the patient population. Participants were reminded that prospective trial participants do not have the same scientific backgrounds as researchers. Patients rely on perceptions, who and what they know, their access, and cultural values to inform decisions about research participation. Investigators are encouraged to be transparent and to share with patients what works and what does not work. When patients choose not to continue participating, further follow-up is needed. Also, patient-centered groups could be effective intermediaries to build trust and share information that could increase enrollment.

Presentation: Blood Derived OrthoBiologic Therapies for OA (Constance Chu, 49:041)

Sports medicine cases have shown that platelet-rich plasma (PRP) treatment for knee injury can rapidly restore function. PRP is ill-defined, but in general, it is a blood product and has been regulated as such by the FDA. PRP has potential for treating a broad array of illnesses and associated symptoms, and public interest in this therapy is growing due to publicized treatment of high-profile athletes. Many knee OA patients are seeking PRP treatment as an alternative to knee replacement. However, PRP effectiveness in knee OA is challenging to determine because it depends on patients’ age and health status, as well as the quality of plasma preparations at the time of treatment.

A recent study examining patients in the early stages of OA showed no difference in outcomes between the treatment and placebo groups. Another study showed systematic and functional improvement with PRP treatment, and separate gait studies also demonstrated improvements. However, fewer than half of patients showed sustained improvement 6-to-24-months later. Identifying predictors of a successful response to PRP is challenging as no significant differences have been found between patients who have positive outcomes and those who do not. Laboratory studies show greater anti-inflammatory and regenerative factors in blood from young and healthy individuals than the blood of older individuals which showed pro-inflammatory and inhibitory effects. This suggests the need for greater caution in the use of blood products from older OA patients.

Dr. Chu concluded by noting that a systems-based approach emphasizing health and lifespan has the potential to deliver stronger and more sustainable results than PRP for early knee OA treatment. As an example, poor diet has been shown to promote inflammation so severe that it resembles a bacterial infection.

Discussion: Factors Potentially Influencing Clinical Utility of PRP (1:04:451)

Customized PRP interventions

Much of the evidence behind PRP does not support its use in knee OA, though it is relatively safe. However, PRP use paired with a more systems-based approach could improve the effectiveness of PRP treatment in OA. Measures of healthy lifestyle influencing response to PRP have not been included in clinical trials, but research teams are working to implement objective measures of these parameters. Some of these measures could be captured with wearable devices.

Some people, particularly older people with more severe disease, have factors in their blood that might limit the effectiveness of PRP and thus should not be concentrated and readministered with PRP. If these factors can be identified, they could be removed from patient samples. Similarly, any identifiable beneficial factors could be enhanced. With identification of these factors, patients could be tested for a regenerative phenotype, enabling use of their own blood; otherwise, they would need another PRP formulation. Combination therapies, such as using PRP with antifibrotic drugs or anti-inflammatories, could target systemic components related to aging and disease processes. This critically important approach has promise for effective and sustained treatment of knee OA.

Whether to recommend PRP treatment with limited evidence of efficacy remains an open question.

Presentation: Stem-Cell Based Cell Therapies for OA (Scott Rodeo, 1:17:271)

Usable stem cells include exogenous cells, such as those from bone marrow, adipose and perinatal tissues, and the intrinsic “stem cell niche.” Exogenous cells, especially mesenchymal stromal cells (MSCs) and connective tissue progenitors, are symptom-modifying in humans and are thought to act on synovial cells by producing both immune-modulating and anti-inflammatory factors. The anti-inflammatory and immune-modulating effects of exogenous cells appear to be dependent on the patient’s immune system environment. Whether the cells are structure-modifying remains unclear.

MSCs directly affect synovial inflammation, as shown in mouse models of OA with synovitis. Clinical studies support positive effects of cell therapy on knee OA symptoms. They reduce symptoms, but only limited in vitro, animal, and clinical studies have shown early evidence of structure-modifying potential. Data support a predictive donor selection criterion based on anti-inflammatory biomarkers that could be useful for studies of MSC therapy for OA.

Discussion: Factors Influencing Study Design and Applicability (1:39:221)

Differences between allogeneic and autologous treatments

Autologous cell preparations used in clinical settings tend to be poorly characterized and heterogenous. Allogeneic preparations would allow characterization of cells, cell sorting, and culture expansion of desired cells. Tests of purity, potency, and practical biomarkers in cell preparations and the correlation of those cell characteristics with patient outcomes would be informative.

Failure of intrinsic responses to resolve inflammation

Timing of the intervention, as well as the cellular environment in the patient’s joints, can affect the effectiveness of stem cell therapies, including whether a patient may need to receive a series of administrations. This is because the intrinsic responses for healing are stimulated with acute injury or inflammation, but the response turns toward fibrosis rather than continued healing with chronic, unresolved inflammation.

Quality of clinical trials

In general, the quality of study methodology of stem cell therapies for knee OA has been moderate-to-poor. Researchers can improve quality by characterizing the cells being used and documenting how they were obtained and prepared. The influence of patient phenotype, including age, sex, and clinical comorbidities, also could be better characterized to understand additional factors contributing to therapeutic efficacy.

Future directions for stem cell therapies

Characterization of cell populations will likely uncover optimal cell formulations for therapy. Improved characterization also will lead to knowledge about interactions between intrinsic stem cell and progenitor cell populations within tissues and important roles of immune cells. Although federal regulations can be perceived as a barrier to allogeneic studies, this is a common hurdle for many types of clinical trials.

Presentation: Cartilage Regeneration by Chondrogenic Small Molecules (Marc Hochberg, 1:54:291)

Both inflammatory and anti-inflammatory cytokines play a role in cartilage repair and regeneration by targeting the bone-cartilage unit. Repair strategies using small molecules include promoting chondrogenesis by using compounds such as resveratrol, statins, melatonin, and glucocorticoids. Other small molecules with structure- and disease-modifying potential include glucosamine, estrogen, and activated vitamin D, among others. These have been efficacious for cartilage repair in murine models. Researchers also have identified novel small molecule compounds that target cartilage regeneration through known mechanistic pathways and high throughput screening. Those compounds also have potential as disease-modifying therapies for OA.

Discussion: Clinical Considerations (2:15:401)

Early detection to block disease progression

Identification of patients with knee OA symptoms who lack joint degeneration would provide opportunities to intervene prior to the development of severe disease with discernable structural changes. The first step of such treatment could be to invoke an anti-catabolic response to prevent further cartilage degradation. The second step would be the application of an agent to promote cartilage regeneration. However, this two-step approach is unlikely with a single small molecule.

Administration routes for clinical application

Small molecules in clinical studies have mostly been orally administered. Intra-articular administration is associated with limited long-term effectiveness, as formulations tend to have shorter half-lives. Furthermore, intercalation of a compound into the cartilaginous matrix is necessary for activating chondrocytes, but intercalation cannot be guaranteed through an injectable administration.

Presentation: Matrix-Based Therapies and Tissue-Engineered Cartilage (Daniel Grande, 2:23:331)

Scaffolds are particularly useful in cartilage tissue engineering due to the ease with which surgeons can insert them as 3-dimensional spatial fillings. Autologous chondrocyte-scaffold products are currently in phase 3 clinical trials. The body of clinical evidence for scaffold treatment in reducing knee pain and restoring knee function is growing. A recent review found that autologous matrix-induced chondrogenesis (AMIC) performs better than the more established matrix-assisted autologous chondrocyte implantation (MACI) in repairing defects in knee cartilage, and AMIC also demonstrated a lower rate of complications. AMIC and its therapeutic framework have many applications in the context of developing adjunct materials for microfractures; these new applications are modernizing treatment options that have remained static for more than two decades.

Discussion: Evidence Base for Advancing Engineered Cartilage Therapies (2:40:361)

Outcomes of studies with arthritic knees

Cartilage defects can be repaired effectively in early stages. However, the field of matrix-based therapies and engineered cartilage is unlikely to reach the point of slowing disease progression without reliable biomarkers to identify OA at earlier stages.

Formation of hyaline cartilage rather than fibrocartilage

Hyaline cartilage resists fibrosis, but the cellular architecture is unique and has not been successfully replicated in a way that maintains a durable structure.

Presentation: Extracellular Vesicles for Treating Osteoarthritis (Daniël Saris, 2:53:001)

The lack of standardized characterization of exosomes limits understanding of the role of exosomes in modulating joint homeostasis. Isolating functional exosomes from the environments in which they are naturally active is challenging. Current evidence shows both chondroprotective and chondro-supportive effects of exosome therapies, though challenges persist in determining the best route of delivery to cartilage and in identifying ideal sources from which to isolate usable exosomes. Before moving to clinical trials in OA, researchers are standardizing reproducible exosome protocols to establish a more reliable evidence base.

Discussion: Gaps in Understanding of Exosomes Limit Utility as a Therapeutic Approach (3:08:061)

Pathway for characterizing exosomes

Research in this field is still in its early stages, but the field accepts that researchers must look beyond surface markers to move forward. Identification of exosome contents, which includes determining why certain factors are present and why they are secreted by cells in a particular environment and at a particular time, are ongoing. Researchers are revisiting their understanding of underlying mechanisms and the precise function of exosomes as they are secreted into a target environment. Studies suggest that future therapies may forgo use of cells and concentrate more on using extracellular vesicles.

Presentation: Gene Therapy and Gene-editing Approaches for Arthritis (Farshid Guilak, 3:12:201)

The primary goal of gene-editing therapy is to employ synthetic biology for development of self-regulating cells. Gene therapy approaches can be used to design “smart cells” that can autoregulate delivery of biologic drugs through synthetic gene circuits. To keep these cells in place, researchers have fixed them into small disks of cartilage in a rheumatoid arthritis model, which showed that on-demand drug delivery reduces pain. Currently, “on/off” manipulation of drug delivery relies on exogenous drug administration. Researchers have developed mechanically sensitive synthetic gene circuits, in which mechanical loading can activate drug production. These circuits have pre-defined responses, drug delivery durations, and recovery times. Chronotherapy could use cellular circadian clock-based circuits for drug delivery to coincide with diurnal disease flares. Murine models have shown that chondrocytes have stable ‘internal clocks’ with predictable cycles that allow synchronization of drug delivery with endogenous timing systems.

Discussion: Progress Toward a Therapeutic Approach (3:31:101)

Longevity of implants

Cartilage implantation, where engineered cells are embedded within cartilage matrix, lasts for months and even years in joints. Other procedures, such as the model of hormone-secreting birth control implants, provide strong evidence for long-term implantation of biomaterials in the body.

Plasticity of genetic programming effects on surrounding cells

Current studies use induced pluripotent stem cells (iPSCs) in a relatively uniform cell population. Cell edits cannot transfer to nearby cells, but rather edited cells protect neighboring cells from cytokine effects. So far, the lifetime of edited cells has been long, with implanted cells showing undiminished activatability at 6 months; however, if edited cells are injected instead of being implanted, they only survive for about two days.

Phenotypic strategy

Laboratories are currently using a stepwise differentiation protocol that mimics embryogenesis of cartilage. However, the specific strategy is not that important as long as it forms stable cartilage.

Translating to clinical research

The immense challenges in scaling up smaller studies to more sophisticated animal models to determine dosing and to control outputs will need to be overcome as researchers aim to translate these studies into clinical trials. A sustainable method for using load-bearing capacity as a signal for delivering analgesic compounds remains to be determined.

Presentation: MSC Therapy for Osteoarthritis: Status (Frank Barry, 3:41:461)

Very few studies with MSCs advance to phase 3 clinical trials. This may be due in part to phenotypic differences between cells isolated from different sites or to variation in cell differentiation potential caused by differences in the method for cell expansion. These challenges underscore the need to develop engineering approaches that can standardize manufacturing of MSCs. To elucidate a mechanism of action for MSC activity in OA, researchers must first understand how the in vivo disease environment elicits phenotypic changes in the implanted cells. Current evidence demonstrates that adipose-derived cells are more therapeutically useful than bone marrow-derived cells, which may guide further development of MSC-related therapies.

Discussion: State-of-the-Science for Advancing MSC Utility (4:05:221)

Gene expression in licensed cells and those retrieved in vivo

Two sets of genes are dominantly upregulated when cells are implanted regardless of whether the cells are commercially available or autogenic – genes that produce anti-inflammatory proteins and those that code for secreted proteins.

Differences in patient responses

Differences in patient responses have not been fully explored, but researchers are looking at the relationships between biological characteristics of cell preparations and documented clinical outcomes.

Moving from autologous to allogeneic cells

Because large scale trials of autologous stem cells are impractical, studies of allogeneic stem cells are a next logical step. Most studies of allogeneic cells are being done outside of the United States, and U.S. researchers may be able to leverage European experiences. Classification of stem cells as biologic products may present a hurdle.

Opportunities to improve guidelines

International Society for Cell and Gene Therapy (ISCT) guidelines identified biomarkers that are reliable for basic experiments, but they were not viewed as practical for more nuanced studies. Original frameworks, such as for use of flow cytometry, are still useful and are critical for scaling up cell-based therapies, but surface phenotypic characterizations have been slow to expand. Single-cell sequencing of preparations would enable standardized delivery to patients, and biological assays relating to potency could determine dosing effectiveness and safety.

Mechanism of early and rapid response to treatment

In pain studies, rapid improvement can sometimes come from a placebo effect. Biologically, any true and measurable pain relief due to MSC administration is likely related to the upregulation of anti-inflammatory cytokines. In the longer term, sustained improvement might be related to differentiation of delivered cells into connective tissue and chondrocyte phenotypes but is more likely to be due to an interaction between delivered cells and host immune cells that produces inflammatory or immune-mediated effects.

Presentation: Regulatory Considerations for Development of Biological Products to Treat Osteoarthritis (Larissa Lapteva, 4:19:091)

The 21st Century Cures Act defines regenerative therapies as cellular therapies and therapeutic tissue engineering products, human cell and tissue products, and combinations of these, which have sustained effects on human cells and tissues. Regulations for drug and biologic products by the FDA are based on both the Food, Drug, and Cosmetic Act and the Public Health Service Act. For the development of advanced cell therapies, the FDA considers factors such as biological activity, product potency, and dosing, which present a challenge given the difficulty of defining a specific mechanism of action for these products. In the early stages of clinical development, researchers should closely monitor inflammatory and immunogenic reactions, contamination, and infections. In the longer term, care should be taken in trials to monitor for ectopic tissue formation, tumorigenicity, and acceleration of cartilage degradation. Regulatory requirements for biologic, drug, and device approval include evidence of effectiveness, adequate and well-controlled studies, and reasonable assurance of safety.

Discussion: PRP as a Regulated Biologic Example (4:52:151)

Categorizing PRP in FDA regulation

Applications for FDA approval of PRP use include those for investigational devices and investigational new drugs. PRP has a vast array of clinical applications, so the classification under FDA rules depends on the intended use and administration.

Discussion: Responses to Canvassing Questions (5:16:211)

Supporting evidence for state-of-the-art therapies

Autologous therapies tend to be safe, but there is mixed evidence for efficacy. Gaps in understanding of these therapies and a lack of structured clinical trials present challenges to the establishment of a conclusive and reliable evidence base for using them. Researchers must take systems-based approaches, especially toward understanding and employing combination therapies.

Emerging approaches

Current thinking about orthobiologics tends to focus on PRP, bone marrow aspirate concentrate (BMAC), and stem cells, but there should also be an added focus on senolytic drugs and supplements. These approaches are not mutually exclusive, however; they can very much be used in combination. iPSC-based therapies are still in the early stages of development, but they may be more efficacious than other methods. A suggested combination approach was to examine these emerging therapies alongside trusted metrics for physical activity to gain an understanding of overall health and loading impacts on disease state.

Challenges and gaps impeding clinical translation

Persistent efforts are underway to gain more control over variation and to develop more well-defined standards for therapies like PRP. Without this control and standardization, the potential for meaningful prospective randomized trials is limited. Controlling for variability of patients is perhaps the greatest challenge – some patients present with comorbidities whereas others do not. The information available about study participants has typically not included information about comorbidities, limiting the usefulness of retrospective analyses for determining which patients would respond well. Recruitment for clinical trials also is potentially limited by a lack of effective communication to patients about these therapies.

Conclusion: Priority areas for research progress

Roundtable participants urged NIAMS to support quality investigations of orthobiologics and related studies, with a focus on development of therapeutic standards. They indicated a great scientific need to build an evidence base for characterization and standardization of therapeutics. They also noted the need for appropriate dosing and delivery, quality control of the studies, better sub-phenotyping of OA patients, and systemic approaches such as a combination of multiple treatments.


BAJPAYEE, Ambika, Ph.D., M.Eng., Northwestern University
BARRY, Frank, Ph.D., Colorado State University
BRYANT, Stephanie, Ph.D., University of Colorado, Boulder
BUXTON, Denis, Ph.D., National Heart, Lung, and Blood Institute
CHU, Constance, M.D., Stanford University (co-chair)
ELISSEEFF, Jennifer, Ph.D., Johns Hopkins School of Medicine
GOMOLL, Andreas, M.D., Hospital for Special Surgery
GRANDE, Daniel, Ph.D., Feinstein Institutes for Medical Research
GUILAK, Farshid, Ph.D., Washington University
HOCHBERG, Marc, M.D., M.P.H., University of Maryland
HUARD, Johnny, Ph.D., Steadman Philippon Research Institute
LAPTEVA, Larissa, M.D., M.H.S., M.B.A., U.S. Food and Drug Administration
LATTERMANN, Christian, M.D., Brigham and Women’s Hospital
LEONG, Amye, M.B.A., Healthy Motivation
LUTZKO, Carolyn, Ph.D., Cincinnati Children’s Hospital
MONT, Michael, M.D., Sinai Hospital of Baltimore, Inc.
MUSCHLER, George, M.D., Cleveland Clinic Foundation
ROBEY, Pamela, Ph.D., National Institute of Dental and Craniofacial Research
RODEO, Scott, M.D., Hospital for Special Surgery (co-chair)
SARIS, Daniël, M.D., Ph.D., Mayo Clinic College of Medicine
SHAPIRO, Shane, M.D., Mayo Clinic
WHITE, Daniel, P.T., Sc.D., M.Sc., University of Delaware


BELKIN, Alexey, Ph.D.
BURROWS, Stephanie, Ph.D.
CARTER, Robert, M.D.
CRISWELL, Lindsey, M.D., M.P.H., D.Sc.
DRUGAN, Jonelle, Ph.D., M.P.H.
JACKSON, Thomas, M.S., M.P.H.
KIRILUSHA, Anthony, Ph.D.
LESTER, Gayle, Ph.D.
NGUYEN, Van, Ph.D.
PARK, Heiyoung, Ph.D.
PEI, Wuhong, Ph.D.
ROBINSON, Daphné, Ph.D.
TAYLOR, James, Ph.D.
WASHABAUGH, Charles (Chuck), Ph.D.
ZHENG, Xincheng (Ted), M.D., Ph.D.