Over the last few years, the focus of scientific activity has been towards investigating regenerative effects of both bone marrow concentrate (BMC) and bone marrow-derived MSCs (BMSCs) in cartilage repair. BMC, being a concentrate graft from bone marrow, aspirated usually from the iliac crest, contains a heterogenous cell population in which MSCs are present [52].

Heringou et al. [35] recently published a 15-year follow up to their original study, in which they treated OA patients who had to undertake total knee arthroplasty of both knees but chose not to have both surgeries done simultaneously. These patients, aged 65–90 years were offered an autologous BMC injection in the other knee (which was not undergoing TKA surgery), during the same anesthetic, on the same day. Patients were randomly assigned to two groups based on form of delivery of MSCs, either via intra-articular injection (IA group) or via implant in subchondral bone (SC group). Patients stopped taking inflammatory and/or analgesic drugs 3–4 weeks prior to the procedure; glucosamine was allowed for the patients who were previously using it. Post procedure, the patients were given analgesics (in relation with TKA), but no anti-inflammatory drugs. Follow-ups were performed at three and six months after surgery and then every year up until the recent 15-year follow up. In the next study, Heringou et al. [53] revealed several interesting findings from this study, mostly that in both SC group and IA group, injections of BMC led to significant pain relief. Conversion to TKA was postponed or avoided completely in the contra lateral joint of patients with bilateral osteoarthritis. Regarding the two cohorts, overall results have shown the subchondral injection to be more efficient in postponing TKA in the same grade of OA. The pain relief in case of the IA group generally did not last longer than 12 months, synovitis was not reduced, and lesions in subchondral bone were not decreased. Hence, many patients from the IA cohort eventually underwent TKA surgery for the particular knee. The most valuable finding is that the subchondral cell therapy treatment has the potential to become a primary treatment of OA, since this study presents long-term benefits (15 years) of the treatment and its ability to postpone, and even in some cases completely avoid TKA in some patients. This approach utilizes rather low concentrations of the MSC in BMC graft, and it is not possible to demonstrate the optimal MSC concentration, due to the heterogenous character of bone marrow graft. Consequently, we could consider this a factor of improvement of this technique, since the MSC number in BMC decreases relative to the age of the patient [54]. In order to move away from the heterogeneity of BMC grafts from the iliac crest, many studies are gravitating towards the route of cultured MSCs. By isolating, characterizing, and expanding the stem cell populations, we can overcome low stem cell yield and donor side morbidity limit [55].

Shadmanfar’s group applied these cultured BMSCs in their research into RA patients’ knees via intra-articular injections. In their study, they used approximately 40 million autologous MSCs. This study is the first triple-blind placebo-controlled clinical trial targeting the safety and tolerability of BMSCs therapy in RA-involved knees. The group was also able to achieve early clinical efficacy of the therapy. Patients enrolled in the study, who were 18–65 years old (mean age average was 50 years), were allowed continue taking DMARDS, but not allowed NSAIDs. Patients were randomly assigned to either receive MSC treatment or placebo (sterile saline), received the treatment once, and returned for check-up after 1, 3, 6, and 12 months in order to record the safety and efficacy of the treatment. Bone marrow was aspirated from iliac crest of each patient, and BMSCs were isolated and then cultured for 7–10 days, and then injected into the patient’s knee (if assigned to MSC group). In this study, no adverse effects were recorded during the injection, nor any of the follow ups. The MSC-treated group showed remarkable improvement in pain relief already 1 month post injection and was able to maintain this effect until the last follow up (at 12 months). There was a significant decrease in the Western Ontario and McMaster Universities Arthritis Index (WOMAC) scores for the MSC group (of −16.5 ± 13.5) vs. the placebo group (−6.7 ± 13.6), which indicates a major decrease in joint pain. Finally, importantly, unlike the placebo-treated subjects, the visual analogue scale (VAS) assessment proved the effectiveness of MSC treatment by decreasing knee pain by 50% at the last follow up. The study claims BMSCs’ intra-articular injection into patient RA-involved knee to be safe, well tolerated, and feasible at the employed dose and study design. The results also suggest a clinical benefit of the study but this is to be further researched with a larger number of RA patients with knee involvement [36].

Recently, a very unique study concerning immune-related genes in autoimmune diseases and their expression was published by Ghoryani and his colleagues. In this clinical trial, refractory RA patients (of 44 ± 7:50 years), who were receiving the maximum approved dose of conventional DMARDs, were recruited. The patients intravenously received autologous MSCs isolated from the bone marrow. Cells used in this study were harvested and cultured, developing more homogenous culture of BMSCs. Each patient received one dose of MSCs, in which the number of cells varied based on subject’s body weight (BMSCs, 1 × 10⁶ cells per kg). Immunological factors were obtained from patients’ peripheral blood mononuclear cells (PBMCs) and evaluated at 1, 6, and 12 months. The results presented by Ghorvani’s group showed substantial increase in IL-10 and transforming growth factor-beta 1 (TGF-β1) levels, both of which are two major cytokines of Tregs. Additionally, the study showed increased levels of forkhead box P3 (FOXP3), a unique Treg transcription factor, at the end of the intervention (at 12-month check-up). This actively demonstrates that the BMSCs had immunoregulatory effect on regulatory T cells, possibly differentiating T lymphocytes from Tregs. The scientific group discovered via correlation analysis a negative relationship between levels of IL-4 and the Disease Activity Score-28 for Rheumatoid Arthritis (DAS28-ESR). All the previously stated data were supported by demographic data, showing a considerably lower DAS28-ESR compared to before the treatment, throughout the whole study. None of the patients reported any adverse events at any of the follow-ups during the study. In future research, increase of MSCs and/or repetitive treatment with MSCs will be looked into [37].

Human umbilical cord became a prominent source of MSC in regenerative medicine. Human umbilical cord blood-derived MSCs (hUCB-MSCs) possess several advantageous properties compared to BMSCs, including accessibility, increased proliferation, and decreased immunogenicity, which makes them an interesting option for RA treatment. Their immunoregulatory traits make them promising allogenic source of MSCs to be used in therapy. Although their repair mechanisms in RA therapy have not been fully exposed, they are recognized for their self-restorative properties and multipotential differentiation ability [56]. To this day, there is a lack of comprehensive reports concerning clinical benefits of hUCB-MSCs in the treatment of RA, as well as safety evaluation, and possible cartilage repair mechanism.

Wang and colleagues conducted a cohort recruiting active RA patients with inadequate response to DMARDs. They divided these patients into two groups; the first group received DMARDs with hUCB-MSCs, and the second one DMARDs with medium without hUCB-MSCs. The DMARDs + hUCB-MSCs group intravenously received a small dose of DMARDs along with 4 × 10⁷ of hUCB-MSCs in stem cell solvent. The DMARDs + medium without hUCB-MSCs group received a small dose of DMARDs with stem cell solvent without hUCB-MSCs, also via intravenous infusion. Additionally, subjects in the DMARDs + hUCB-MSCs cohort received more than one treatment and therefore were divided into 3 groups according to different intervals after the first treatment. Group 1 received the second treatment after 3 months, group 2 after 6 months, and group 3 after 8 months. In order to assess the safety of the treatment, patients’ physical health, liver, and kidney function was evaluated along with hematological and biochemical testing, urine analysis, chest radiography, and electrocardiograph (ECG) being performed before and after the treatment (for both treated and control group). No major abnormal findings related to adverse effects were observed in the study. Clinical effects suggest that DMARDs + hUCB-MSCs administration resulted in long-term mitigation of disease activity of refractory RA [38]. It is believed the clinical benefits were linked to the decreased expression levels of inflammatory cytokines and chemokines, the increase in regulatory T cells in peripheral blood, and the upregulation of IL-4-producing Th2 cells [57]. Wang and his group suggest the major potential mechanism behind positive effects of hUCB-MSCs are anti-inflammatory affects along with the improved immune-modulation and the induced immune-tolerance. Rapid joint pain relief and alleviation of joint swelling within 12 h post-treatment was reported, and maintained throughout the whole study. The second cycle delivered even better clinical benefits, and over all resulted in better quality of life for the treated patients [38]. The data suggest that patients infused with MSCs that are HLA haploidentical or completely HLA mismatched with the stem cell donor and recipient show no difference in clinical effects and have no immunological memory to the infused MSCs [58]. The core message of this study is that RA patients who were previously nonresponsive to traditional medication treatment received significant improvements, including symptom alleviation and cytokines decrease, when treated with hUCB-MSCs, and these positive effect of the MSC therapy were prolonged and stabilized via repetitive treatment [38].

The first phase Ia clinical trial extending their preclinical research on hUCB-MSCs safety and tolerability in patients with RA was published by E. Park et al. Treatment was delivered via single intravenous infusion of cultured hUCB-MSCs. This was an open-label, dose-escalation study and the study subjects had moderate RA and were on a stable dose of methotrexate (MTX) for at least 12 weeks before being enrolled into the study [39]. Throughout the study, patients maintained their regimen of corticosteroids; coincidentally none of the patients previously received DMARDs. Patients were divided into 3 groups, of which the first cluster received 30 min infusion of lowest dose, with cell number of 2.5 × 10⁷ hUCB-MSCs. When no dose-limited adverse events arose, the next group received a dose of 5 × 10⁷ cells. Safety and tolerability were assessed for this cohort before moving to the final patient cluster, who received the highest dose of 1 × 10⁸ of hUCB-MSCs. Safety and tolerability were measured (hematological and biochemical tests, urine analysis, chest radiography, and ECG) after 24 h, 72 h, 1 week, and 4 weeks following the infusion. Preliminary efficacy assessment (via DAS28, a pain visual analog scale (VAS), and health assessment questionnaire (HAQ)) was obtained after 4 weeks. In this study, no adverse events nor dose-limiting toxicity (DLT) was reported. Although assessing the clinical benefits of hUCB-MSC treatment was not the primary objective in this trial, Park and colleagues reported a decrease in DAS28 at week 4 after the treatment, and a decrease in IL-1β, IL-6, IL-8, and TNF-α levels were measured at 24 h in the highest dose cohort (1 × 10⁸ cells). Additionally, this study reported a significant increase in IL-10 levels, in a cluster treated with 5 × 10⁷ cells after 24 h. For the future trials, this research group plans to investigate long-term DLT and safety, including a placebo group, as well as evaluate the safety of repetitive infusions, including patients previously treated with DMARDs, and the assessment of clinical outcomes utilizing imaging methods [39].

A recent study by Wang et al. [40] utilized umbilical cord tissue-derived MSCs (UC-MSCs) and demonstrated its long-term safety and efficacy in RA therapy. The data presented in this study proved stable clinical outcomes for 3 years post single dose treatment. The data presented showed that combination of DMARDs with cultured UC-MSCs therapy was safe, and drastically improved quality of life of RA patients included in this trial. They had previously reported safety and efficacy of this treatment for up to 8 months, and now have proven that the DMARDs + UC-MSCs treatment alleviated RA symptoms, reduced HAQ, and DAS28 scores long-term [40]. Furthermore, the levels of inflammatory and/or RA serological makers significantly decreased, and were maintained for 3 years in comparison to pre-treatment. All patients given DMARDs with UC-MSCs reported rapid remission in disease activity and had improvements in diet, sleep, and physical strength compared to control group, who experienced no such improvements. In this study, they presented two particularly extraordinary clinical outcomes of two patients, who experienced remission nearly 6 months post-treatment, and have since maintained these positive effects, which enables them free movement and a life free of pain, joint swelling, and deformity [40]. The same group, provide evidence that UC-MSCs based therapy seem to be efficient and safe to be used in clinical practice [59].

Lately, an alternative source of MSCs has been employed in clinical research of RA; adipose-derived mesenchymal stem cells (ADMSCs) propose an easy access with minimally invasive procedure with multiple collection sites [60,61]. ADMSCs have proven to be capable of multilineage differentiation, with high proliferative potential and surface proteins, that make them a suitable candidate for cell-based therapeutic strategies [62].

The research team of Álvaro-Gracia published results of the first multicenter, dose escalation, randomized, single-blind, placebo-controlled, phase Ib/IIa clinical trial with active refractory RA patients, intravenously treated with allogenic ADMSCs. The primary goal of this study was to assess the safety of allogenic ADMSCs intravenous infusions, and find DLT [41,63]. The secondary purpose was double-blinded preliminary efficacy evaluation. Subjects enrolled in this study were previously unsuccessfully treated with at least one to two standard anti-RA non-biological treatments. The study included a wash-out period without treatment before the trial started. Patients were randomly divided into 4 cohorts, according to the number of cells they received—cohort A: 1 million/kg, cohort B: 2 million/kg, cohort C: 4 million/kg, or placebo: receiving Ringer’s lactate solution. The treatment was administered via three intravenous infusions at days 1, 8, and 15, and follow-up visits were conducted at weeks 1, 2, and 3, and at months 1, 2, 3, 4, 5, and 6 [41]. In this study, the researchers decided to utilize allogeneic ADMSCs, due to their availability in the cell bank, which makes them readily available for usage. Cells were originally obtained from lipoaspirates and were tested for viability, population doublings, morphology, potency, identity, purity, sterility, and genetic stability, among other quality controls. The presented data suggest that no treatment-related toxicity occurred and the therapy resulted in an overall favorable safety profile. No venous thrombosis or pulmonary thromboembolism arose, nor did any life-threatening events or deaths occur. One DLT was reported in a subject from cohort A at day 8 after the 2nd infusion, who suffered lacunar stroke, which was deemed as likely related since no other apparent causes were found. The purpose of this study was mainly to investigate toxicity and safety of this therapy, although the efficacy data provided in this trial showed a better response in comparison to the placebo group. Nonetheless, assessment of clinical efficacy outcomes must be cautiously interpreted, since the study was not designed towards efficacy evaluation [41].

Mallinson’s research group has published data from a distinctive study with the main objective to stratify highly refractory RA patients receiving MSC therapy, in order to investigate biomarkers associated with ADMSCs therapy response in RA patients [42,64]. For this trial, they selected RA patients who were previously intravenously treated with MSC therapy, and either responded to the therapy (responders), or did not respond to the MSC treatment (non-responders). RNA from pre-treatment plasma samples from both responders and non-responders were analyzed via circulating miRNA microarrays. These miRNA biomarkers were further investigated so as to accurately evaluate relative expression between two patient groups and 10 most significantly differentiated miRNA biomarkers were selected [42]. Based on statistical significance, 3 final candidates were established—miRNA biomarkers miR- 26b-5p and miR-495-3p were recognized to be significantly upregulated in the responder group, and miR-487b-3p came very close to being significantly upregulated. The study claimed that the three targeted miRNA biomarkers present potential in discrimination between cell-based therapy responders vs. non-responders, which makes them a crucial element in the success of these pioneering therapeutic approaches. Although the presented results seem promising, there is a need for a larger sample size and further investigation of the three miRNA biomarker candidates [42].

Lee et al. published results from their most recent phase IIb, double-blinded, and placebo-controlled clinical trial, assessing the safety and efficacy of intra-articular autologous ADMSC injections in patients with knee OA [43,65]. Patients were blindly assigned to ADMSCs injection (MSC group) or normal saline injection (control group). The MSC cohort received a high dose of cells (1 × 10⁸ cells) and patients’ activity was not restricted, allowed full weight-bearing post-treatment. Enrolled subjects, between 18 to 75 years old, had OA of the knee joint. All anti-OA medication was discontinued for 2 weeks before the treatment. Cells were harvested via lipoaspiration from the participants and culture-expanded. Cell number, viability, purity (CD31, CD34, CD45), identity (CD 73, CD 90), sterility (bacterial and fungal), and endotoxin and mycoplasma contamination was evaluated before intra-articular administration [43,66]. Efficacy and safety were inspected at 1, 3, and 6 month follow ups. In this report, Lee and his colleagues showcased significant changes in cartilage defect after single treatment of autologous ADMSCs. This one-step treatment resulted in 55% reduction in the WOMAC total score, 59% in the WOMAC pain score, 54% in the WOMAC stiffness score, and 54% in the WOMAC physical function score at 6 months post-treatment. This study was lacking in sample size, hence the group suggests a larger group and longer follow-up is needed in the next trial [43].

It is difficult to fairly assess differences between particular MSC sources and discuss whether UC-MSCs, ADMSCs, or BMSCs are superior to the others. The variances such as study design, cell type, rehabilitation protocols, and adjunct therapy hinder the possibility to statistically evaluate the clinical studies [67,68].

Furthermore, the utilization of allogeneic vs. autologous MSCs in treatment is to be investigated more deeply in order to compare their therapeutic potential. Autologous MSCs are believed to be safer, since they do not initiate an immune reaction. Nevertheless, the donor site morbidity might be a downfall of the autologous MSCs compared to the allogeneic MSCs. However, tumorigenesis, disease transmission, and possible host immune rejection still remain a concern when dealing with allogeneic MSCs and must be studied in the future [69,70].