Synovial Biomarker Modulation and Clinical Response to Umbilical Cord-Derived MSC Treatment in Knee Osteoarthritis: A Case Series Study

Abstract

Objective

Umbilical cord-derived mesenchymal stromal cells (UC-MSCs) are a promising treatment for knee osteoarthritis (KOA). This study aimed to (1) characterize early synovial fluid biomarker changes after intra-articular UC-MSCs therapy in advanced KOA and (2) explore their relationship with short-term clinical outcomes.

Methods

In this prospective, single-arm case series, 15 patients with advanced KOA (Kellgren–Lawrence III–IV) scheduled for total knee arthroplasty received a single intra-articular injection of 20 × 10⁶ allogeneic UC-MSCs. Synovial fluid was aspirated at baseline and approximately 6 weeks post-injection. A panel of 29 soluble biomarkers related to inflammation, matrix remodeling, immune signaling, angiogenesis, and metabolism was quantified using multiplex immunoassays. Clinical status was evaluated with Patient-Reported Outcome Measures (PROMs) before and at a mean of 48 ± 18 days after the injection. Paired differences were analyzed using the Wilcoxon signed-rank tests.

Results

Significant post-treatment differences were observed in biomarkers associated with extracellular matrix turnover (Matrix Metalloproteinase [MMP]-1, MMP-3, MMP-7, Cartilage Oligomeric Matrix Protein [COMP]), vascular remodeling (Vascular Endothelial Growth Factor [VEGF], Vascular Cell Adhesion Molecule [VCAM]-1), immune modulation (interleukin [IL]-8), and metabolic or structural regulation (Leptin, dickkopf [Dkk]-1). Most PROMs demonstrated significant improvements. These findings describe molecular changes in synovial fluid observed after intra-articular UC-MSC administration.

Conclusions

In this prospective single-arm cohort, intra-articular UC-MSC administration was feasible and well tolerated, and we observed pre–post changes in synovial fluid biomarkers and patient-reported outcomes. These findings do not establish causality and warrant confirmation in controlled studies. The study was registered at Clinicaltrials.gov: Mechanisms of Treatment Effects Using Cultured, Allogeneic Mesenchymal Stromal Stem Cells (MSCs) in Knee Osteoarthritis, NCT06078059, https://clinicaltrials.gov/study/NCT06078059?intr=NCT06078059.

Keywords

knee osteoarthritis biomarkers synovium umbilical cord stem cells

Introduction

Knee osteoarthritis (KOA) is one of the leading orthopedic diseases that affects millions of patients worldwide and represents a substantial socioeconomic burden.¹ The disease is characterized by progressive cartilage degradation, joint inflammation, and subchondral bone remodeling, which ultimately leads to pain, stiffness, and functional impairment, thereby significantly decreasing the patient’s quality of life. Although the prevalence of KOA is high, treatment options are largely palliative, focused on pharmacological as well as non-pharmacological interventions, with the main aim to achieve symptom relief rather than causal treatment.² Conventional therapies include systemic use of non–steroidal anti-inflammatory drugs (NSAIDs) or cyclooxygenase inhibitors and also corticoid or hyaluronic acid (HA) injections. These modalities can provide temporary symptom relief but cannot stop or reverse disease progression. As disease progresses to end-stage, total knee arthroplasty (TKA) remains the only definitive treatment, yet it carries surgical risks and is not suitable for all patients.³

The field of regenerative medicine has garnered much attention in the past largely due to merit of its potential in the treatment of KOA, particularly using mesenchymal stromal/stem cells (MSCs). Today, MSCs are widely recognized as potent cellular therapy tools due to their unique properties, such as anti-inflammatory, immunomodulatory, and trophic, which may provide disease-modifying effects in orthopedic conditions.⁴ The use of MSCs in KOA has been reported in a great number of various phase clinical trials. A number of early-phase trials report that intra-articular injection of MSCs can alleviate pain, improve joint function, and modulate synovial inflammation, thereby potentially delaying or even preventing the need for subsequent surgical intervention.⁵ Nevertheless, extensive knowledge gaps still exist regarding a comprehensive understanding of the mechanisms of action, which many believe to be fundamental in determining optimal cell dosage, treatment prognosis, and patient selection. For example, it is still not clear whether MSCs function solely by exerting immunomodulatory effects or whether there are also chondroprotective or regenerative mechanisms at play.

One of the ways to study MSC-associated pharmacological mechanisms is to look at changes of local microenvironment upon MSC administration, at the molecular level. To the best of our knowledge, only 1 study has examined biomarker response in synovial fluid (SF) after intra-articular MSC injection (using bone marrow-derived MSCs).⁶ Building upon this and other relevant literature, we designed our study with the aim to contribute valuable molecular and clinical insights into the early effects of MSCs in KOA.

Given the limitations of current palliative treatments and the invasive nature of TKA, MSC therapies may represent a promising intermediate strategy that could delay or modify the course of KOA progression. For orthopedic surgeons, especially those treating end-stage KOA, understanding the short-term molecular and symptomatic responses to MSC-based interventions may offer valuable insight into novel preoperative management options and patient stratification prior to surgery.

To our knowledge, this is the first report to evaluate both short-term clinical and molecular outcomes after intervention using UC-MSCs in KOA. Since many studies have evaluated and confirmed the safety of KOA therapy using MSCs from various sources (adipose tissue, bone marrow, umbilical cord),^7
-11 we designated safety as our secondary aim.

Material and Methods

Study Design

This prospective case series study was conducted at the Department of Orthopedic Surgery of the University Medical Center Ljubljana (Ortho-UMCL) in collaboration with the Slovenian Institute for Transfusion Medicine (SITM). The study protocol was approved by the National Medical Ethics Committee (Approval No. 0120–273/2022/3), and the study was registered at Clinicaltrials.gov (NCT06078059). Although the ClinicalTrials.gov registration initially described an arthrocentesis-only comparator arm, this arm was not implemented because performing arthrocentesis without therapeutic intent was considered ethically unjustified; therefore, the study was conducted as a single-arm case series. This study was designed as a prospective, exploratory case series focused on feasibility and within-subject (pre–post) molecular changes in SF rather than on confirmatory hypothesis testing. Therefore, no formal a priori power calculation was performed. The target enrollment of 15 patients was set pragmatically based on feasibility (availability of end-stage KOA patients scheduled for arthroplasty within the recruitment window). The resulting data set was intended to provide preliminary estimates of biomarker variability and effect sizes to inform the design of future controlled, adequately powered trials.

Patient Selection

A total of 15 patients diagnosed with primary or post-traumatic KOA and scheduled for knee arthroplasty were recruited at the Ortho-UMCL. The patients were recruited during routine preoperative assessment in the anesthesiology outpatient clinic, typically 6 weeks before the planned surgery. After they were evaluated and approved as eligible for surgery by the anesthesiologist, the patients who fit basic inclusion criteria were presented with comprehensive information regarding the study and given informed consent to participate. The inclusion and exclusion criteria are shown in Table 1 . After patient consent was given, the clinicians recorded the patients’ baseline characteristics.

Table 1.

Inclusion and Exclusion Criteria.

Inclusion Criteria	Exclusion Criteria
- planned arthroplasty of the knee joint due to primary or post-traumatic KOA - age above 50 years - both sexes - eligibility for surgery, anesthesia, intervention - informed consent	- secondary KOA of another etiology - rheumatic diseases - uncontrolled metabolic diseases (e.g., insulin therapy) - crystallopathy - patients with blood clotting disorders - patients on anticoagulant therapy - heart failure (NHYA III/IV) - renal failure (grade 3 and above) - other diseases that can affect the immune response or pose an additional risk in the application of cells

Pre- and Post-Treatment Assessments

At baseline, we recorded patient demographic data, medical history, and their clinical status of the affected knee. The patients completed a set of Patient-Reported Outcome Measures (PROMs) to assess their perceived knee status, pain, functionality, and quality of life. For this purpose, we utilized the visual analog scale for pain assessment (VAS-P), Knee Injury and Osteoarthritis Outcome Score (KOOS), the European Quality of Life in Five Dimensions Three-Level (EQ-5D-3L), and EuroQoL self-rated health on a visual analog scale (EQ-VAS).^12
-14 The same set of PROMs was administered to patients on the day of admission for the knee arthroplasty procedure.

On day 0 (baseline), we performed neat SF aspiration without local anesthetics via supra-patellar arthrocentesis of the affected knee under aseptic conditions for determination of baseline synovial markers. Subsequently, the patients received an intra-articular injection of 20 × 10⁶ cultured, allogeneic UC-MSCs in a 4-ml dispersion through the same needle used for SF aspiration. After a brief observation period, the patients were discharged with detailed post-care instructions. The patients received a paper-based diary and were instructed to record VAS-P scores twice daily (morning and evening) on the same day, as well as any deviations in knee status, without reminder systems. The completed diaries were collected on the day of hospital admission for surgery. Only correctly and sufficiently completed diaries were included in the analysis. Recall bias was considered a study limitation. For close monitoring of any serious side events, we provided strict instructions and contact information.

On the day of surgery (on average 48 ± 18 days after injection), SF samples were collected again with the patient on the operating table prior to surgical exposure of the joint. The biological samples were immediately transported to SITM under controlled conditions (using transport media and cooling elements). To ensure confidentiality, patient identifiers were removed outside the clinical setting. At SITM, the SF samples were cryopreserved for later measurements using Luminex or Enzyme-Linked Immunosorbent Assay (ELISA) methods.

Objectives and Outcome Measures

This case series study aimed to contribute to deeper understanding of the mechanisms by which UC-MSCs contribute to KOA treatment. Primary and secondary outcome measures were defined. As primary outcome measures, key biomarkers were selected based on the strongest literature evidence supporting their association with KOA, guided by the BIPED (Burden of disease, Investigative, Prognostic, Efficacy of Intervention and Diagnostic) criteria.^15
-17 Secondary outcome measures focused on the clinical and safety analysis of this treatment approach.

Primary Outcome Measures

Analysis of soluble biomarkers in SF associated with:

Extracellular matrix (ECM) turnover and cartilage degradation:

Proteoglycans and glycoproteins: Hyaluronan, Aggrecan, COMP (Cartilage Oligomeric Matrix Protein), and PIIANP (Procollagen II N-terminal Propeptide).

Collagen breakdown markers: Matrix Metalloproteinase (MMP)-1, MMP-3, MMP-7, MMP-9, and MMP-13.

ECM remodeling inhibitors: Tissue Inhibitor of Metalloproteinases (TIMP)-1 and Syndecan-4.

Inflammation and immune system:

Cytokines and chemokines: tumor necrosis factor (TNF)-α, IL-6, IL-8, chemokine ligand (CCL)2, and CCL5.

Immune cell activation markers: CD163 and CD14.

Bone and cartilage metabolism:

Bone morphogenetic proteins (BMPs): BMP-2.

Wnt pathway inhibitors: Dickkopf (Dkk)-1 and Sclerostin.

Angiogenesis and Vascular markers:

Vascular Endothelial Growth Factor (VEGF), Vascular Cell Adhesion Molecule (VCAM)-1, and Intercellular Adhesion Molecule (ICAM)-1.

Adipokines and Metabolic Regulators:

Adiponectin and Leptin.

Other structural and Regulatory proteins:

Osteoprotegerin, Tenascin-C, and Fibroblast Growth Factor (FGF)-2.

Secondary Outcome Measures

7. Pain on Visual Analog Scale (VAS-P)

A 10-point scale where 0 represents “no pain” and 10 signifies worst possible pain.¹²

8. KOOS

KOOS is a knee-specific instrument that evaluates pain (9 items), symptoms (7 items), daily living function (17 items), sports and recreational activity (5 items), and knee-related QoL (4 items). Scores range from 0 to 100, with 0 indicating the worst knee condition and 100 indicating no symptoms.13

9. EQ-5D-3L questionnaire

EQ-5D-3L measures 5 domains, namely mobility, self-care, daily activities, pain/discomfort, and anxiety/depression, each with 3 levels (no issues, moderate issues, extreme issues). Patients selected the level that best describes their health state. Responses were converted into a single summary index using a validated scoring formula.14

10. EQ-VAS

EQ-VAS measures patient’s self-rated health using a visual analog scale from 0 to 100, where 0 represents the worst health state imaginable and 100 represents the best health state imaginable.¹⁴

11. Serious adverse event (SAE) monitoring

Safety was monitored from day 0 (intervention) up to 6 months after the surgery.

Quantification of Synovial Fluid Biomarkers

Aspirated SFs were transported at 4 °C and processed within 2 hours. SFs were centrifuged at 2000 × g for 15 minutes to pellet debris, the supernatant was aliquoted in cryovials and stored at −80 °C until further analysis. Commercial immunoassays ( Table 2 ) were used according to manufacturers’ instructions. Prior to analysis thawed SF samples were treated with 2 mg/ml bovine testicular hyaluronidase (type I-S, Sigma-Aldrich, St. Louis, MO) at 37 °C for 15 minutes (except samples for Hyaluronan ELISA which were analyzed untreated). As reported before,¹⁸ intra-assay precision can be improved by hyaluronidase treatment of SF. After treatment, the samples were appropriately diluted for each assay. All samples were analyzed in duplicates. Analytes are shown in Table 2 .

Table 2.

Immunoassays Used in the Study.

Platform	Assay	Manufacturer	Cat. Number
Luminex	Adiponectin, CCL2, CD14, Dkk-1, IL-6, Leptin, MMP-3, MMP-9, Sclerostin, Tenascin C, VCAM-1, BMP-2, CCL5, CD163, ICAM-1, IL-8, MMP-1, MMP-7, MMP-13, Syndecan-4, TNF-α, VEGF	R&D Systems, MN	LXSAHM-22
Luminex	FGF-2, Osteoprotegerin, TIMP-1	R&D Systems, MN	LXSAHM-03
ELISA	Hyaluronan	R&D Systems, MN	DY3614-05
ELISA	COMP	R&D Systems, MN	DY3134
ELISA	Aggrecan	BioVendor, Czech Republic	RIS001R
ELISA	PIIANP	EMD Millipore	EZPIIANP-53K

Intervention

The investigational medicinal product (IMP) consisted of allogeneic UC-MSCs, manufactured at the SITM following validated standard operating procedures. These procedures have been approved by the National Medical Ethics Committee (No. 0120–60/2018/7, approved on April 24, 2018). The IMP was derived from single-donor umbilical cord tissue as the starting material for manufacture. Collection, transport, and tissue processing followed validated standard operating procedures at the SITM. Briefly, cords were cleaned and dissected to isolate stromal tissue, which served as the starting material for UC-MSC outgrowth and expansion under good manufacturing practice (GMP) conditions. The cells were cultured in MEM α culture medium (Gibco), supplemented with 10% human AB serum (produced at SITM) and 50 μg/ml gentamicin (Gibco), at 37 °C and 5% CO₂. After an initial culture period of 7 to 10 days in passage 0 (p0), adherent cells proliferated exponentially and reached adequate confluence within the next 7 days. The cells were then harvested at 80% confluence and reseeded for further expansion in p1, with complete medium exchange every 3 to 5 days. At the end of p1, the MSCs were harvested (using trypsin) and cryopreserved in uniform aliquots as “off-the-shelf” units in the SITM cryostorage facility, making them available on a per-patient demand.

Before application, an aliquot was thawed and cultured for an additional 3 to 5 days (p2) to ensure sufficient cell quantity and preconditioning (viability, functionality). The UC-MSCs were manufactured in a GMP environment at the SITM using validated SOPs and donor screening, as approved by the National Medical Ethics Committee. For this study, 2 single-donor GMP lots were used (i.e., cells from 2 different donors); lots were expanded at low passage, cryopreserved as off-the-shelf aliquots, and briefly recovered prior to injection as described above. Batch release required viability ≥85%, identity/phenotype (≥95% CD73/CD90/CD105; ≤2% CD14/CD34/CD45/HLA-DR), negative sterility and mycoplasma, and endotoxin <0.5 EU/ml when measured. Product characterization included trilineage differentiation and an immunosuppressive potency assay (dose-dependent inhibition of T-cell proliferation). We have reported full manufacturing and QC methodology previously.¹⁹ The final product consisted of 5 × 10⁶ MSCs/ml in a physiological solution containing 0.5% human albumin. The product was formulated in syringes in a total volume of 4 ml (20 × 10⁶ UC-MSCs). Administration was performed using an 18G, 1.2 × 50 mm hypodermic needle. Each participant received a single intra-articular injection of 20 × 10⁶ UC-MSCs formulated in 4 ml of saline containing 0.5% human albumin, without anesthesia. The injection was scheduled 6 weeks prior to planned knee arthroplasty.

Study Termination Criteria

Participant safety was the highest priority. The study was planned for termination if serious adverse events (SAEs) were to be identified during the intervention period or up to the date of surgery. We additionally monitored SAEs for up to 6 months post-surgery. Discontinuation criteria included allergic reactions, localized or systemic infections related to the injection, any other adverse events classified as grade 3 or higher based on CTCAE v5.0 criteria (serious adverse reactions, SARs). In the event of SAEs, the principal investigator (PI) was to inform the study coordinator within 24 hours and notify the National Center for Pharmacovigilance at the Agency for Medicinal Products and Medical Devices of the Republic of Slovenia, followed by an additional report submitted to the National Medical Ethics Committee for further review.

Confidentiality Measures

Participant’s identities were protected through an anonymization system in which each subject was assigned a unique code for all documentation. Only the PI was authorized to maintain a record linking patient identities to their respective study codes. No document leaving the research facility contained patient names or identifying details, ensuring that all research data remained confidential. Participants maintained the right to access their personal information collected during the study, to request corrections or deletions where necessary, and file complaints with regulatory bodies such as the Information Commissioner of the Republic of Slovenia.

Availability of Data and Materials

The study records were securely stored on the hospital’s servers, and access was granted only through password-protected network computers. The data will be preserved for 10 years, after which, these will be permanently deleted.

Statistical Methods

We compared the values of the presence of selected molecular markers in the SF within the group of 13 out of 15 patients, who received UC-MSCs therapy, due to sample contamination in 2 patients. We compared baseline values with values at the time of surgery (6 weeks after the application of UC-MSCs). The obtained numeric variables were compared using paired-samples Wilcoxon signed-ranked test. Because of the exploratory nature of our study, we did not adjust for multiple comparisons.

In the translational clinical part of the study, the PROMs scores (VAS-P, KOOS, EQ-5D-3L-EQ-VAS) between baseline (prior to intervention) and at the time of surgery (post-intervention) were compared with paired-samples t-test for 15 patients. To assess the normality of data, we used the Shapiro–Wilk test. The level of statistical significance was set at P < 0.05.

Data were analyzed using databases with the Excel spreadsheets and the Jamovi Project software (version 2.5).

Results

Subjects

Out of 15 enrolled patients, 10 (67%) were women and 5 (33%) were men. None of the patients underwent previous surgeries of the index knee. Age ranged between 55 and 83 years, with the average being 71.0 (8.4) years. Body mass index (BMI) was between 22.5 and 36.6 with the mean value of 30.7 ± 5.4 kg/m². Six (40%) knees were left, and 9 (60%) were right. In 14 (93%) cases, the lower limb alignment was in varus, and in 1 (7%) case, it was in valgus. The mean hip-knee-ankle (HKA) angle was 4.8 ± 3.4 of varus. Six (40%) knees were classified as Kellgren–Lawrence grade III and 9 (60%) as grade IV.

Clinical Results

The clinical results are presented in Table 3 . Significant improvements were observed in 7 out of 9 PROMs.

Table 3.

Knee-Specific Patients’ Outcomes for Pain, Functioning, and Quality of Life for 15 Subjects With Terminal KOA.

PROMs	Subjects (N = 15)
PROMs	Pre-injection	Post-injection	Difference (post-injection − pre-injection)	P-value
KOOS Symptoms	50 ± 23	60 ± 27	10 ± 14 *	0.03
KOOS Pain	37 ± 16	52 ± 21	15 ± 12 *	< 0.001
KOOS ADL	43 ± 15	58 ± 24	15 ± 14 *	0.001
KOOS Sport	15 ± 19	31 ± 28	16 ± 2 *	0.009
KOOS QoL	19 ± 11	31 ± 21	12 ± 17 *	0.02
KOOS Total	33 ± 14	46 ± 23	13 ± 13 *	0.002
EQ-VAS	54 ± 16	60 ± 24	6 ± 27	0.4
EQ-5D-3L TTO	0.37 ± 0.16	0.45 ± 0.24	0.08 ± 0.20	0.2
VAS-P	6.9 ± 1.6	4.4 ± 2.6	– 2.5 ± 2.4 *	0.002

Data was presented as means with standard deviations. Significant improvements (differences between pre- and post-injection scores) are marked with *. The exact P-values are reported.

The KOOS subscale Pain was determined as the primary clinical outcome measure. The mean improvement met the minimal important change (MIC) defined at 12.4 for non-surgical interventions in knee OA treatment.²⁰ MIC in the KOOS subscale pain was achieved in 11 out of 15 patients (73%), who were categorized as responders.

In total, 12 patients sufficiently and regularly completed the provided diaries to report daily VAS-P scores. All of them completed at least 4 weeks of entries. Therefore, the results for the first 4 weeks are presented in Figure 1 , together with the mean values recorded before the injection, in the afternoon after the injection, and the final value on the day of hospital admission for the planned surgery.

Figure 1.

Mean daily VAS-P score after UC-MSC injection. Daily VAS-P dynamics during the first 4 weeks post-injection were shown. Mean values recorded before the injection, in the afternoon after the injection, and on the day of hospital admission for the planned surgery were also presented.

In 10 subjects (67%), some kind of transient reaction to the injection was observed. One patient had just 1 day of pain, 1 had 1 day of swelling and pain, 4 had 2 days of swelling and pain, 3 had 3 days of swelling and pain, and 1 remained symptomatic for 1 week. Among them, 2 patients also developed a subfebrile state with body temperature up to 37.5 °C, which normalized the day after the injection. No SAEs were observed up to 6 months post-surgery.

Biomarker Assessment in the Synovial Fluid

We analyzed a selected set of biomarkers associated with ECM turnover, cartilage degradation, inflammation, immune response (shed receptors), bone and cartilage metabolism, angiogenesis, monocyte/macrophage-recruiting chemokines, adipokines, and metabolic regulators, as well as other structural and regulatory proteins in SF of patients at baseline, and at approximately 6 weeks after UCMSC injection. Our selection was based on markers with the strongest literature support. Analyses were performed in samples of 13 out of 15 patients, since SF samples from 2 patients presented obvious blood contamination. Summarized in Table 4.

Table 4.

Summary of Key Synovial Biomarker Changes Following UC-MSC Therapy.

Biological Process	Biomarker	Change after therapy	P-value	Interpretation
ECM Remodeling/Degradation	COMP	↑ significant	0.005	Increased cartilage matrix turnover
	MMP-1	↑ highly significant	< 0.001	Active extracellular matrix remodeling
	MMP-3	↑ highly significant	< 0.001	Supports clearance of damaged matrix
	MMP-7	↑ significant	0.021	Involved in matrix remodeling
	TIMP-1	↑ (not significant)	0.273	Potential regulatory balance to MMP activity
Inflammation/Immune Response	IL-8	↑ slight but significant	0.04	May support angiogenesis and repair
	CD163	↑ trend (M2 macrophage)	0.068	Anti-inflammatory and pro-resolving phenotype
	CCL2	↓ trend	0.11	Reduced monocyte recruitment
Angiogenesis/Vascular Remodeling	VEGF	↑ significant	0.017	Supports synovial revascularization
	VCAM-1	↑ significant	0.01	Endothelial activation and cell migration
Metabolic/Adipokine Response	Leptin	↑ significant	0.017	Involved in metabolism and immune modulation
	Adiponectin	↑ (not significant)	0.146	Potential protective metabolic effect
Repair/Structural Molecules	Dkk-1	↑ slight but significant	0.01	Wnt pathway regulation; bone homeostasis
	Tenascin-C	↑ trend	0.08	Supports tissue repair and remodeling

Considering ECM turnover and markers of cartilage degradation, we observed a significant increase in COMP levels post-treatment (P = 0.005) with an approximate 1.5-fold increase, suggesting a surprising increase in cartilage-related matrix proteins. Regarding the levels of hyaluronan, aggrecan, and PIIANP, we observed no such differences ( Fig. 2A ). Interestingly, differences were also seen regarding some MMPs. Compared to baseline values, there was a significant increase in MMP-1 (P < 0.001) and MMP-3 (P < 0.001), as well as a slight but significant increase in MMP-7 (P = 0.021). No differences were seen for MMP-9 and MMP-13 (Fig. 2B). In this context, the inhibitor of MMPs, TIMP-1, increased numerically post-treatment; however, the difference was not statistically significant (P = 0.273) (Fig. 2C).

Figure 2.

Synovial fluid changes in extracellular matrix turnover and cartilage degradation markers following UC-MSC treatment. (A) Levels of hyaluronan, aggrecan, COMP, and PIIANP were measured at baseline and post-treatment (6 weeks). A significant increase was observed in COMP (P = 0.005). (B) Levels of MMP-1, MMP-3, MMP-7, MMP-9, and MMP-13. Significant increases were seen in MMP-1 (P < 0.001), MMP-3 (P < 0.001), and MMP-7 (P = 0.021); MMP-9 and MMP-13 remained unchanged. (C) Levels of TIMP-1 and syndecan-4. No significant differences were observed. Statistical analysis was performed using the Wilcoxon signed-rank test. Data are presented as matched pairs (n = 13). Data are presented as mean ± SD.

In terms of inflammation-related markers, we generally observed that all of them, including IL-8, IL-6, and TNF-α, were present at very low levels, with individual outliers in some cases (Fig. 3A). Nevertheless, we observed a minor increase in IL-8 (P = 0.04) post-treatment. On the contrary, levels of IL-6 and TNF-α remained largely unchanged. Similarly, we also did not see changes in levels of chemokine CCL5, which, with the exception of individual outliers, was detected at very low concentrations. The levels of monocyte chemoattractant protein 1, CCL2, were, on the contrary, readily detectable and showed a median decrease post-treatment, although not quite statistically significant (P = 0.11) (Fig. 3B).

Figure 3.

Effects of UC-MSC treatment on synovial inflammation and immune response markers. (A) Cytokines TNF-α, IL-6, and IL-8 were measured at baseline and post-treatment. A minor but significant increase in IL-8 was observed (P = 0.04), while TNF-α and IL-6 remained unchanged. (B) Chemokines CCL2 and CCL5 were assessed. CCL2 showed a trend toward decrease (P = 0.11); CCL5 levels were stable. (C) Soluble monocyte/macrophage markers CD14 and CD163. A trend toward increased CD163 was detected (P = 0.068), while CD14 levels remained unchanged. The Wilcoxon signed-rank test was used for comparisons (n = 13). Data are presented as mean ± SD.

The levels of shed receptor CD14, reflecting monocyte/macrophage presence, were more or less identical at baseline and post-treatment. The median levels of CD163, a marker of type 2, anti-inflammatory macrophage polarization, increased numerically post-treatment with a P-value of 0.068; therefore, the change did not reach statistical significance (Fig. 3C).

Regarding bone and cartilage metabolism, Dkk-1 changed significantly, although very slightly (P = 0.01), while the levels of BMP-2 and sclerostin remained unchanged (Fig. 4). The markers of angiogenesis revealed a significant increase between baseline and post-treatment in the levels of VEGF (P = 0.017) (Fig. 5). Similarly, the levels of adhesion molecules showed significant changes; however, they only appeared meaningful in the case of VCAM-1 (P = 0.01) due to negligible differences in the median for ICAM-1, with the exception of 2 outliers (P < 0.001).

Figure 4.

Bone and cartilage metabolism markers in synovial fluid pre- and post-UC-MSC treatment. Levels of BMP-2, Dkk-1, and sclerostin were measured. A slight but statistically significant increase was observed for Dkk-1 (P = 0.01), whereas BMP-2 and sclerostin levels did not significantly change. Comparisons were made using the Wilcoxon signed-rank test (n = 13). Data are presented as mean ± SD.

Figure 5.

Changes in angiogenesis and vascular adhesion molecules following UC-MSC treatment. VEGF, ICAM-1, and VCAM-1 levels were analyzed. VEGF (P = 0.017) and VCAM-1 (P = 0.01) significantly increased after treatment; ICAM-1 showed a statistical difference mainly due to outliers (P < 0.001), but median values remained comparable. Wilcoxon signed-rank test (n = 13). Data are presented as mean ± SD.

Some adipokine markers showed significant responses, with a post-treatment increase observed for leptin (P = 0.017) (Fig. 6). Measurement of adiponectin showed a trend for increase after treatment; however, the difference was not significant (P = 0.146). Among other structural and regulatory proteins, tenascin-C increased numerically post-treatment; however, this change did not reach statistical significance (P = 0.08). The levels of FGF-2 (P = 0.839) and osteoprotegerin (P = 0.685) remained unchanged (Fig. 7).

Figure 6.

Effects of UC-MSC treatment on adipokine profiles in synovial fluid. Leptin and adiponectin levels were measured at baseline and post-treatment. A significant increase was observed for leptin (P = 0.017), while adiponectin showed a non-significant trend toward higher levels (P = 0.146). Statistical analysis was performed using the Wilcoxon signed-rank test (n = 13). Data are presented as mean ± SD.

Figure 7.

Structural and regulatory proteins in synovial fluid before and after UC-MSC therapy. Levels of osteoprotegerin, tenascin-C, and FGF-2 were measured. Tenascin-C showed a trend toward increase post-treatment (P = 0.08), while FGF-2 (P = 0.839) and osteoprotegerin (P = 0.685) remained stable. Wilcoxon signed-rank test (n = 13).

Discussion

The analysis of SF biomarkers revealed dynamic changes indicative of active matrix remodeling and immunomodulation. We demonstrated significant post-treatment increases in cartilage turnover markers such as COMP, along with elevations of MMP-1, MMP-3, and MMP-7. Although elevations in MMPs are often associated with matrix degradation, prior work indicates that controlled MMP activity can participate in ECM turnover and repair processes.^21,22 In the context of MSC therapy, such upregulation likely reflects an orchestrated, early remodeling phase rather than pathological cartilage destruction. Importantly, MMP-13/MMP-9 and aggrecan, hyaluronan, and PIIANP remained stable, arguing against a generalized catabolic response. These observations should be interpreted as consistent with early remodeling of the joint microenvironment, while acknowledging the exploratory nature of our study. The TIMP-1 did not differ significantly between time points.

The inflammatory component within the SF exhibited selective modulation. Pro-inflammatory cytokines TNF-α and IL-6 remained stable, suggesting that UC-MSC therapy did not elicit a generalized inflammatory response. In addition, both cytokines were detected at very low levels, particularly TNF-α. We observed a slight but significant increase in the levels of IL-8. Beyond its role as a pro-inflammatory cytokine, IL-8 is also known to promote angiogenesis and can support tissue repair processes.²³ Other markers associated with activity of immune cells indicated a shift toward a pro-resolving joint environment. The soluble macrophage scavenger receptor CD163 showed a trend toward increased levels post-treatment, consistent with M2-like macrophage polarization and resolution of inflammation. We also observed a trend toward a decrease in CCL2 concentrations, a known monocyte chemoattractant factor, although the change was not statistically significant. Given the established contribution of low-grade synovitis to osteoarthritis symptoms,^24,25 the immunological findings reported here may be consistent with the clinical improvements observed, but causal links cannot be inferred from this case series.

Biomarkers of vascular remodeling also changed significantly. After UC-MSC administration, the levels of VEGF and ICAM-1 were elevated, suggesting enhanced synovial vascularization. A potential improvement in tissue perfusion and nutrient delivery within the joint may not only support matrix remodeling but may also mitigate ischemia-related nociceptive signaling, thereby contributing to early pain relief.^26,27 Furthermore, we observed modest changes in metabolic mediators, including a significant increase in leptin and a non-significant rise in adiponectin. While leptin has been positively correlated with disease progression previously,²⁸ its role following MSC therapy remains complex and may involve the modulation of cellular metabolism and immune responses within the joint microenvironment.^29
-31 These findings are therefore presented as hypothesis-generating. Tenascin-C, a matricellular protein involved in tissue repair,³² showed no significant difference and is reported descriptively.

Clinically, patients reported improvements across several PROMs, including pain and function. Although we do not claim causality, the pattern of biomarker changes alongside symptom gains is compatible with an early remodeling response. Such interpretation could potentially explain instances where markers such as COMP increased, while symptoms improved.³³ In our view, early symptomatic benefits are most plausibly linked to initial microenvironmental reprogramming, notably matrix remodeling and attenuation of inflammatory status, rather than to direct, immediate structural regeneration. These dynamic biomarker changes provide hypothesis-generating mechanistic insight and align with a broader shift in OA therapeutics toward microenvironmental modulation. They also underscore the multifactorial nature of OA as a complex, systemic disease. Confirmation of mechanism and durability, however, will require controlled, adequately powered trials with longer follow-up.

Transient self-limiting local reactions occurred in 67% of subjects, without any SAE. Although high, this rate is comparable to previous UC-MSC studies, where reported frequencies range from 35% to 100% depending on dose. In the trial by Matas et al.,¹¹ the reaction rate in the equivalent dose group was almost identical (68.7%). The observed pattern of reactions in our cohort also corresponds to the transient increase in VAS-P on the first day after the application, while the rise in the final score may, to some extent, reflect preoperative stress (Fig. 1). Across MSC therapies in general, transient reaction rates vary broadly (2-60%), with higher frequencies typically seen in allogeneic products.³⁴ In the systematic review by Riggle et al.,³⁵ the overall incidence was 12.3%, with the highest rate for UC-MSCs (51.7%). Compared with other intra-articular injectables, the incidence is higher than for corticosteroids and hyaluronic acid (5-10%) and approaches the upper platelet-rich plasma (PRP) ranges (18-30%).^36
-38 Notably, these reactions are usually mild and self-limiting, while SAE remains rare across all injection types.

This study has several limitations that should be acknowledged. First, the design was a prospective case series without a control group or randomization, which limits the ability to establish causality or differentiate treatment effects from natural disease fluctuations or placebo response. In this manner, it should be noted that a portion of the observed molecular shifts may reflect a nonspecific response to intra-articular manipulation, rather than UC-MSC-specific activity. Previous studies have reported on short-term SF biomarker modulation after non-MSC injections, including HA, with decreases in IL-6 and changes in SF rheology and cartilage-related constituents,^39
-41 and corticosteroids, which were shown to alter biomarkers of cartilage/bone turnover.⁴² Randomized comparisons of PRP versus HA further show reductions in SF inflammatory cytokines over similar time windows,⁴³ while meta-analyses indicate that saline/placebo injections can yield clinically meaningful symptom improvements.^44,45 Collectively, these data support a cautious interpretation of our results in the absence of a control arm and motivate future controlled designs linking biomarker trajectories to clinical outcomes.

Second, all patients included were scheduled for TKA, which may not fully represent the broader KOA population that seeks MSC therapy to avoid surgery. Third, we did not perform imaging-based assessments, such as magnetic resonance imaging (MRI), to evaluate potential structural cartilage changes or synovitis. Finally, the short follow-up period precludes conclusions about long-term outcomes or disease-modifying effects. A key limitation is the variability in the timing of the post-treatment assessment (mean 48 ± 18 days). Because synovial biomarkers may show time-dependent kinetics after an intra-articular intervention, sampling over a relatively broad early window may increase inter-individual variability and potentially mask or exaggerate changes that peak earlier or later. Similarly, PROM improvements may evolve over time and may not align temporally with molecular changes. Finally, given the exploratory nature of the biomarker panel, P-values are reported descriptively and were not adjusted for multiple comparisons. However, emphasis was placed on effect direction, magnitude, and biological plausibility rather than on statistical significance alone.

In conclusion, this is the first report to characterize early joint microenvironmental changes after UC-MSC administration in end-stage KOA. The observed biomarker pattern, together with concomitant improvements in PROMs, is compatible with a short-term microenvironmental reprogramming response (matrix turnover and modulation of inflammatory status) rather than immediate structural regeneration. These findings are hypothesis-generating and outline a testable framework in which early molecular shifts may precede and potentially forecast later clinical or structural outcomes. Future studies should include controlled, adequately powered longitudinal studies to link early biomarkers to delayed imaging/structural endpoints and clinical outcomes and to evaluate patient phenotyping strategies that enable therapeutic matching. If validated, this approach could inform biomarker-guided UC-MSC therapy aimed at maximizing clinical benefit.

Footnotes

Authors’ Note

This work was performed at the University Medical Centre Ljubljana and the Slovenian Institute for Transfusion Medicine.

ORCID iDs

Matic Kolar

Matej Drobnič

Ethical Considerations

The study protocol was approved by the National Medical Ethics Committee (Approval No. 0120–273/2022/3) on July 27, 2022.

Consent to Participate

All subjects provided written informed consent prior to enrollment in the study. The research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

Consent for Publication

Not applicable.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was founded by the internal funding of the Slovenian Institute for Transfusion Medicine (SM-2922 to U.Š.) and by the University Medical Centre Ljubljana – Institutional Research Grant (20240118 to M.D.).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data used and/or analyzed in the present study are available from the corresponding author on a reasonable request.

References

Sharma

Osteoarthritis of the knee. N Engl J Med. 2021;384(1):51-9.

Mora

Przkora

Cruz-Almeida

Knee osteoarthritis: pathophysiology and current treatment modalities. J Pain Res. 2018;11:2189-96.

Van Manen

Nace

Mont

MA.

Management of primary knee osteoarthritis and indications for total knee arthroplasty for general practitioners. J Am Osteopath Assoc. 2012;112(11):709-15.

Berebichez-Fridman

Gómez-García

Granados-Montiel

Berebichez-Fastlicht

Olivos-Meza

Granados

, et al. The holy grail of orthopedic surgery: mesenchymal stem cells—their current uses and potential applications. Stem Cells Int. 2017;2017:2638305.

Pal

Das

Roy

Mukherjee

Halder

Ghosh

, et al. Recent trends of stem cell therapies in the management of orthopedic surgical challenges. Int J Surg. 2024;110(10):6330-44.

Chahal

Gómez-Aristizábal

Shestopaloff

Bhatt

Chaboureau

Fazio

, et al. Bone marrow mesenchymal stromal cell treatment in patients with osteoarthritis results in overall improvement in pain and symptoms and reduces synovial inflammation. Stem Cells Transl Med. 2019;8(8):746-57.

Cao

Sheng

Wang

Chen

Zhang

, et al. Efficacy and safety of mesenchymal stem cells in knee osteoarthritis: a systematic review and meta-analysis of randomized controlled trials. Stem Cell Res Ther. 2025;16(1):122.

Tian

Cao

Zhang

Relative efficacy and safety of mesenchymal stem cells for osteoarthritis: a systematic review and meta-analysis of randomized controlled trials. Front Endocrinol. 2024;15:1366297.

Long

Zhang

Zeng

Yang

, et al. The effectiveness and safety of mesenchymal stem cells in the treatment of osteoarthritis: a systematic review and meta-analysis of 28 randomized controlled trials. Stem Cells Int. 2022;2022:6151866.

10.

Xiao

Wang

Luo

Effects of the umbilical cord mesenchymal stem cells in the treatment of knee osteoarthritis: a systematic review and meta-analysis. Medicine. 2024;103(46):e40490.

11.

Matas

García

Poblete

Vernal

Ortloff

Luque-Campos

, et al. A phase I dose-escalation clinical trial to assess the safety and efficacy of umbilical cord-derived mesenchymal stromal cells in knee osteoarthritis. Stem Cells Transl Med. 2024;13(3):193-203.

12.

Carlsson

. Assessment of chronic pain. I. Aspects of the reliability and validity of the visual analogue scale. Pain. 1983;16(1):87-101.

13.

Roos

Lohmander

LS.

The Knee injury and Osteoarthritis Outcome Score (KOOS): from joint injury to osteoarthritis. Health Qual Life Outcomes. 2003;1:64.

14.

EuroQol Group. EuroQol—a new facility for the measurement of health-related quality of life. Health Policy. 1990;16(3):199-208.

15.

Boffa

Merli

Andriolo

Lattermann

Salzmann

Filardo

Synovial fluid biomarkers in knee osteoarthritis: a systematic review and quantitative evaluation using BIPEDs criteria. Cartilage. 2021;13(1 Suppl):82S-103S.

16.

Kraus

Karsdal

MA.

Osteoarthritis: current molecular biomarkers and the way forward. Calcif Tissue Int. 2021;109(3):329-38.

17.

Convill

Tawy

Freemont

Biant

LC.

Clinically relevant molecular biomarkers for use in human knee osteoarthritis: a systematic review. Cartilage. 2021;13(1 Suppl):1511S-1531S.

18.

Jayadev

Rout

Price

Hulley

Mahoney

Hyaluronidase treatment of synovial fluid to improve assay precision for biomarker research using multiplex immunoassay platforms. J Immunol Methods. 2012;386(1-2):22-30.

19.

Švajger

Zajc

Česnik

Hartman

Pooled umbilical cord-derived MSCs retain key phenotypic and immunomodulatory properties—A path toward consistent manufacturing. Tissue Cell. 2025;96:103036.

20.

Silva

MDC

Perriman

Fearon

Couldrick

Scarvell

. Minimal important change and difference for knee osteoarthritis outcome measurement tools after non-surgical interventions: a systematic review. BMJ Open. 2023;13(5):e063026.

21.

Werb

Matrix metalloproteinases: effectors of development and normal physiology. Genes Dev. 2000;14(17):2123-33.

22.

Chen

Development of matrix metalloproteinases-mediated extracellular matrix remodeling in regenerative medicine: a mini review. Tissue Eng Regen Med. 2023;20(5):661-70.

23.

Rennekampff

Hansbrough

Kiessig

Doré

Sticherling

Schröder

JM.

Bioactive interleukin-8 is expressed in wounds and enhances wound healing. J Surg Res. 2000;93(1):41-54.

24.

Atukorala

Kwoh

Guermazi

Roemer

Boudreau

Hannon

, et al. Synovitis in knee osteoarthritis: a precursor of disease. Ann Rheum Dis. 2016;75(2):390-5.

25.

Primorac

Molnar

Matišić

Hudetz

Jeleč

Rod

, et al. Comprehensive review of knee osteoarthritis pharmacological treatment and the latest professional societies’ guidelines. Pharmaceuticals. 2021;14(3):205.

26.

Maeda

Kawamura

Sasaki

Shimamura

Shibuya

Harada

, et al. Intravenous infusion of bone marrow-derived mesenchymal stem cells improves tissue perfusion in a rat hindlimb ischemia model. Sci Rep. 2022;12(1):16986.

27.

Bandak

Boesen

Bliddal

Riis

Nielsen

Klokker

, et al. Exercise-induced pain changes associate with changes in muscle perfusion in knee osteoarthritis: exploratory outcome analyses of a randomised controlled trial. BMC Musculoskelet Disord. 2019;20:1-12.

28.

Min

Shi

Han

Chen

Shi

, et al. Serum levels of leptin, osteopontin, and sclerostin in patients with and without knee osteoarthritis. Clin Rheumatol. 2021;40(1):287-94.

29.

Astudillo

Ríos

Pastenes

Pino

Rodríguez

JP.

Increased adipogenesis of osteoporotic human-mesenchymal stem cells (MSCs) characterizes by impaired leptin action. J Cell Biochem. 2008;103(4):1054-65.

30.

Jiang

Wang

Chen

Zhang

, et al. Leptin signaling is required for augmented therapeutic properties of mesenchymal stem cells conferred by hypoxia preconditioning. Stem Cells. 2014;32(10):2702-13.

31.

Yang

Jiang

Chen

Nan

, et al. Leptin increases mitochondrial OPA1 via GSK3-mediated OMA1 ubiquitination to enhance therapeutic effects of mesenchymal stem cell transplantation. Cell Death Dis. 2018;9(5):556.

32.

Unno

Hasegawa

Suzuki

Iino

Imanaka-Yoshida

Yoshida

, et al. Tenascin-C promotes the repair of cartilage defects in mice. J Orthop Sci. 2020;25(2):324-30.

33.

Yamamoto

Wilkinson

Bou-Gharios

Targeting dysregulation of metalloproteinase activity in osteoarthritis. Calcif Tissue Int. 2021;109(3):277-90.

34.

Iijima

Isho

Kuroki

Takahashi

Aoyama

Effectiveness of mesenchymal stem cells for treating patients with knee osteoarthritis: a meta-analysis toward the establishment of effective regenerative rehabilitation. NPJ Regen Med. 2018;3:15.

35.

Riggle

McLellan

Bohlen

Wang

Complications of stem cell-based injections for knee osteoarthritis: a systematic review. HSS J. 2025;21(4):476-84.

36.

Bannuru

Osani

Vaysbrot

McAlindon

TE.

Comparative safety profile of hyaluronic acid products for knee osteoarthritis: a systematic review and network meta-analysis. Osteoarthritis Cartilage. 2016;24(12):2022-41.

37.

Bennell

Paterson

Metcalf

Duong

Eyles

Kasza

, et al. Effect of intra-articular platelet-rich plasma vs placebo injection on pain and medial tibial cartilage volume in patients with knee osteoarthritis: the RESTORE randomized clinical trial. JAMA. 2021;326(20):2021-30.

38.

Singh

Knapik

Polce

Eikani

Bjornstad

Gursoy

, et al. Relative efficacy of intra-articular injections in the treatment of knee osteoarthritis: a systematic review and network meta-analysis. Am J Sports Med. 2022;50(11):3140-8.

39.

Kusayama

Akamatsu

Kumagai

Kobayashi

Aratake

Saito

Changes in synovial fluid biomarkers and clinical efficacy of intra-articular injections of hyaluronic acid for patients with knee osteoarthritis. J Exp Orthop. 2014;1(1):16.

40.

Vincent

Percival

Conrad

Seay

Montero

Vincent

KR.

Hyaluronic acid (HA) viscosupplementation on synovial fluid inflammation in knee osteoarthritis: a pilot study. Open Orthop J. 2013;7:378-84.

41.

Bagga

Burkhardt

Sambrook

March L. Longterm effects of intraarticular hyaluronan on synovial fluid in osteoarthritis of the knee. J Rheumatol. 2006;33(5):946-50.

42.

Klocke

Levasseur

Kitas

Smith

Hirsch

Cartilage turnover and intra-articular corticosteroid injections in knee osteoarthritis. Rheumatol Int. 2018;38(3):455-9.

43.

Wang

Ding

Gao

, et al. Impact of autologous platelet-rich plasma therapy vs. hyaluronic acid on synovial fluid biomarkers in knee osteoarthritis: a randomized controlled clinical trial. Front Med. 2023;10:1258727.

44.

Saltzman

Leroux

Meyer

Basques

Chahal

Bach

Jr , et al. The therapeutic effect of intra-articular normal saline injections for knee osteoarthritis: a meta-analysis of evidence level 1 studies. Am J Sports Med. 2017;45(11):2647-53.

45.

Altman

Devji

Bhandari

Fierlinger

Niazi

Christensen

Clinical benefit of intra-articular saline as a comparator in clinical trials of knee osteoarthritis treatments: a systematic review and meta-analysis of randomized trials. Semin Arthritis Rheum. 2016;46(2):151-9.