Exosomes are extracellular vesicles (EVs) surrounded by a phospholipid membrane, either a cell membrane- or an endosomal-derived membrane, that contains different cell-specific receptors, which are important in cell-to-cell communication [133]. MSCs-derived exosomes isolated from different tissue sources are free of cells and have shown therapeutic potential to treat many diseases [134,135]. Pre-clinical in vivo studies have demonstrated positive effects in joints and have confirmed the effectiveness of EVs injections as a minimally invasive therapy [136]. Particularly, MSCs-derived exosomes have cartilage repair properties and can delay OA progression through a variety of mechanisms; for example, stimulating ECM secretion, promoting chondrocyte proliferation, inhibiting chondrocyte apoptosis, and maintaining chondrocyte homeostasis [137]. Recently, it has been demonstrated that IA injection of BMSCs-derived exosomes reduces inflammation and cartilage damage, and inhibits OA progression. These effects are achieved thanks to the phenotypic transformation of macrophages from M1 to M2, together with a decrease in inflammatory cytokines and the release of anti-inflammatory cytokines [138]. A research study carried out by Wang et al. reported that embryonic MSCs-derived exosomes stimulate the synthesis of type II collagen and decrease the production and expression of MMPs with thrombospondin-like motifs-5 (ADAMTS-5), providing a stable ECM in a surgically induced OA model in mice [139]. On the other hand, Tofiño-Vian et al. revealed that EVs isolated from human ADSCs promote chondroprotective functions due to a decrease in MMP activity, decrease the secretion of inflammatory mediators and stimulate anti-inflammatory cytokine IL-10 production [140]. Furthermore, Cosenza et al. reported that both exosomes and BMMSCs microparticles decrease the expression of catabolic and inflammatory markers [141]. Moreover, in a recent study in which MSCs-derived exosomes were used in cartilage repair, an increased matrix deposition and cellular proliferation and better histological scores were observed in animals treated with MSCs-derived exosomes [142].

In an in vitro study carried out by He et al. in rats with OA, significant reconstituted collagen type II and impaired MMP-13 protein expression in knee joint cartilage were observed after exosome treatment, demonstrating a significant chondrocyte proliferation and migration capacity [143]. Additionally, human articular chondrocytes and fibroblast-like synoviocytes isolated from the same OA patients were cocultured in 2D as well as in 3D conditions with fluorescently labeled ADSC-EVs, and analyzed by flow cytometry or confocal microscopy. In both 2D and 3D conditions, fibroblast-like synoviocytes were more efficient in interacting with ADSC-EVs, and 3D imaging showed a faster uptake process. The removal of the HA component from the ECM of both cell types reduced their interaction with ADSC-EVs only in the 2D system, showing that 2D and 3D conditions can yield different outcomes when investigating events wherein ECM plays a key role. These results indicate that studying EVs’ binding and uptake both in 2D and 3D guarantees a more precise and complementary characterization of the molecular mechanisms involved in the process [144].

MSCs-derived exosomes are a promising new cell-free approach to treat OA and other joint conditions. However, further investigations in humans are required to assess the effectiveness and feasibility of this therapy.

Stem cell therapy could become a first-line therapy to treat OA. The efficacy of MSC-based therapies has been related to the paracrine secretion of trophic factors, and exosomes that play a fundamental role in tissue repair.

PRP is a cell-free and autologous product of fractionated plasma derived from the patient’s own blood, obtained after a specific centrifugation process. The platelets contain alpha granules that are rich in several GFs with chemoattractant and mitogenic functions, which help to attract surrounding cells to injured areas. This product can also contain a high number of leukocytes, which may have a negative effect on tissue regeneration. In fact, there is a wide variation in the reported protocols for the standardization and preparation of PRP, with variable reported efficacy. PRGF could be an alternative approach; it is a 100% autologous and biocompatible preparation, with a moderate platelet concentration and no leukocytes or erythrocytes, and it is elaborated by a single centrifugation process [145,146]. It is considered a safe product due to its autologous origin [134].

PRP regulates cartilage repair by stimulating the synthesis of proteoglycans, aggrecans or type II collagen by chondrocytes; the proliferation of synoviocytes; and the chondrogenic differentiation of MSCs. Additionally, it decreases the catabolic effects of cytokines, such as IL-1, or proteolytic enzymes, such as matrix metalloproteinases [18,19,20,29,147,148].

Despite the variability in PRP preparations, several published research and clinical studies have shown excellent outcomes. Cuervo et al. showed significant pain relief and an improvement in limb function after a single IA PRGF infiltration in canine patients with OA secondary to hip dysplasia, and this effect was maintained for more than 6 months [148]. Other studies have addressed a maximum effect 180 days after a single IA PRP injection in patients with knee OA [149]. Furthermore, in a study where patients were treated with PRP combined with HA, a significantly greater pain relief was observed in patients receiving the combination of both treatments compared with those only treated with PRP, suggesting that the association of PRP and HA could be a better long-term option [150]. Huang et al. evaluated PRP IA injections once, twice, or three times per month, and demonstrated the positive effects of PRP up to a 12-month follow-up in patients receiving one or two IA injections, whereas these effects were maintained in those patients receiving three IA injections [151]. With regard to variability in PRP preparations, a systematic review to investigate the superiority of PRP over HA concluded that it is necessary to standardize the centrifugation protocol, the activating agents, the administration frequency, and the injected total volume [152].

Scientific evidence exists regarding PRP injections in all stages of knee OA [30,153]. In a study with 214 patients with knee OA, 155 patients were classified as having Kellgren–Lawrence (KL) grade 1 OA, while 59 were grade 2 OA. Patients were treated with three PRP IA injections in knee joints with 4-weeks intervals, and the WOMAC was evaluated at the time of induction and with 6-month intervals. The mean WOMAC score before treatment was 83.05, and after 6 months, it was significantly reduced to 38.84 [154]. Patients with knee OA graded stage 1 or 2 on the KL scale showed improved WOMAC scores at 1, 2, and 6 months after PRP IA injection [155]. A significant pain relief and functional improvement was reported 3 months after PRP treatment, mainly in lower OA grades [151] compared to control groups [156]. Additionally, Cook and Smith compared PRP with traditional knee OA treatments. Patients reported good pain relief for at least 6 months, and commonly for a year or longer following the three PRP injection regimens [157]. In addition, the effect of IA PRP injection was evaluated with VAS pain scores in patients with knee OA by Taniguchi et al., and the results showed that the average VAS pain scores improved at 6-month follow-up from 71.6 to 18.5 (p < 0.05), with 80% of patients experiencing a 50% decrease, or higher, in VAS pain scores [158]. Similar results were obtained in a different study wherein a reduction in pain and lameness was observed in human patients treated with PRP with knee OA after 1 year follow up [159]. MRI image studies were carried out by Ornetti et al., showing no OA progression in 73% of the patients treated with PRP [160].

This therapeutic option has also reported beneficial effects in animals. In a recent study, the effects of a single PRP IA injection were evaluated in dogs with knee OA secondary to cranial cruciate ligament rupture. Improvements in kinetics were seen in dogs 3 months after treatment [18]. Another similar study was carried out by Vilar et al., where the PRP effect was evaluated in dogs with knee OA secondary to cranial cruciate ligament rupture. Lameness and gait improvements 3 months after IA injection were observed [161]. In horses with moderate to severe forelimb OA, PRP treatment did not cause statistically significant differences in gait, although the authors suggested a possible improvement in lameness despite the lack of statistical difference [162]. Another study in 16 rats with OA induced by monoiodoacetate injection showed promising results. Rats were randomly divided into treatment group A (n = 8 rats) or non-treatment group B (n = 8 rats). The treated group (A) received a single 0.5 mL injection of activated PRP 18 days after the monoiodoacetate injection, and histological assessment of treated joints revealed a higher chondrocyte population and greater cartilage thickness than in untreated animals [163]. Moreover, a study in rabbits in which the perichondral sheaths were dissected and the costal cartilages were removed concluded that the administration of PRP is helpful in improving chondrocyte density, resulting in increased ECM [164].

A combination of IA and intraosseous (IO) PRP infiltration has also been proposed with promising results. Better outcomes at 6 and 12 months were obtained in patients when PRP was applied both IO and IA, compared to the IA application alone [169,170].

The therapeutic potential of PRP products to treat OA is not fully understood. Further studies focused on the combination of other regenerative therapies with PRP are necessary, as well as more investigations in the field of IO infiltration.

In OA, bioregenerative therapies have been demonstrated to be a great option over other treatments due to their therapeutic potential. In recent years, OA cell therapies have been developed as an alternative or additional therapy to traditional methods, with the aim of creating a new tissue displaying the most similar characteristics to native cartilage.

The optimal OA therapy should halt disease progression through the repopulation of the injured tissue with chondrocytes able to produce a hyaline matrix, restoring cartilage structural and functional properties. Different new approaches for cartilage regeneration have been proposed, for example, anti TNF-α therapies, as TNF-α plays a considerable role in OA pathogenesis [171]. Moreover, it has been demonstrated that other biologic agents that inhibit nerve growth factor improve function and reduce pain in OA patients [172]. Future perspectives are focused on gene therapy with encoding genes for chondrogenic GFs and anti-inflammatory cytokines; particularly, this therapy exerts its effects through intracellular nucleic acid transfection and translation into protein [173]. Recently, therapeutic strategies that combine cell and gene therapy based in the production of protein platforms have been proposed as good strategies to treat OA [174,175]. The main limitation of gene therapy is its carcinogenic potential as well as the elevated costs [176].

Other important considerations include the time that the therapeutics stay in the joint tissue. Different natural and synthetic scaffolds (amphiphilic polymeric micelles and hydrogels) are being evaluated to achieve increased articular dwelling of the drugs [177]. Moreover, several studies have reported the beneficial results of nanoparticles used for targeted drug delivery and sustained release in OA joints [178].

Nanotechnological strategies combined with cell-based therapy, biological and gene treatments could be a future perspective in the clinical management of the pathology.