Hematopoietic stem cell transplantation (HSCT) is one of the primary efficient cell-based therapies utilized to treat autoimmune diseases (20, 21). They operate based on eliminating autoreactive immune cells and reconstituting the self-tolerant immune system. Although HSCT is an intensive one-off procedure, it brings a treatment-free remission for the patients upon success. With more than 3300 auto-immune patients registered for HSCT, the European Society for Blood and Marrow Transplantation (EBMT) Autoimmune Diseases Working Party (ADWP) has made significant progress in the field (22, 23). HSCT is currently the therapeutic modality for multiple sclerosis, systemic sclerosis, and Crohn’s disease by the professional stem cell transplant societies of America, Europe, and Brazil (24). The rationale and mode of action of HSCT were reviewed by another group (22). Given the prerequisites of this type of treatment, allogeneic HSCT is still considered a high-risk strategy that is only prescribed for patients in the very severe stages of autoimmune diseases (19, 25–27).

Dendritic cells (DCs) are antigen-presenting cells (APCs) that regulate innate and acquired immune systems, balancing antigen-specific immunity and promoting immune tolerance. Based on their subtypes, DCs can play different and sometimes opposite roles. DCs act as pro- or anti-inflammation inducers upon external application. For instance, in the context of autoimmune diseases, they can contribute to the worsening of the disease through the production of pro-inflammatory cytokines and the activation of autologous antigen-reactive T cells (28). In the context of organ transplantation, donor DC-derived EVs can promote undesirable allograft-targeting immune responses (29). At the same time, tolerogenic DCs comprise a specific type of APCs with immunoregulatory function (30, 31).

The function of DCs depends on several factors, such as pattern recognition receptors (PPRs), specific cytokines and metabolites, some TLRs agonists, dexamethasone, and vitamin A or vitamin D3, which could be applied to maintain tolerogenic DCs stable and to enhance their lymph node-homing capacity (32, 33). The situation seems to be fine-tuned by their maturation status. Studies show that immature and semi-mature DCs demonstrate tolerogenic features (34) through different mechanisms (35). In experimental models, immature DC-derived EVs that carry tolerogenic content were applied successfully in the context of rheumatoid arthritis (36–38). DC-derived EVs maintain their producer cells’ features. Hence they could be considered proper cell replacements (39). These EVs may be enriched with either immunoregulatory or immunostimulatory content to target a particular subset of the immune cells, resulting in their suppression, reprogramming or activation. As well-known anti-inflammatory molecules, EVs with tailored anti-inflammatory content are shown to be produced by DCs, modified for the benefit of IDO1, TGF-β1, IL-10, IL-4, Fas-L, and CTLA-4 (40).

Regulatory T cells (Tregs) actively maintain peripheral immune tolerance (41). Two major types of Treg cells (polyclonal and antigen-specific Tregs) have been investigated based on their immunomodulatory capacities. Pre-clinical studies indicate that antigen-specific Tregs have superior tolerogenic properties compared to polyclonal Tregs as they have more defined specificity (42). Potential applications, large-scale production, safety and efficacy of Tregs produced via good manufacturing practice (GMP) were recently described (43, 44).

An alternative approach to deploying tailored Tregs is via their genetic engineering, known as chimeric antigen receptors (CAR)-Tregs. Tregs are engineered as an improved version of the cells, which facilitates the design of advanced therapeutic procedures to induce immune-tolerogenic responses (43, 45, 46). The biology of Treg cells, advantages, challenges, and some technical details were discussed by Raffin and Bluestone (47). The possible role of these cells in effectively managing autoimmune diseases remains to be elucidated.

MSCs possess high regenerative capacities and special immunoregulatory functions affecting all types of innate and adaptive immune cells (48). MSCs, trans-differentiated to a macrophage-like profile in vivo, were shown to carry secretory capabilities that resembled phagosomes. This functional shift in the injury site is accompanied by the overexpression of CD45 and major histocompatibility complex (MHC) class II while maintaining their MSC-specific surface marker expression (49).

Due to their unique features, MSCs are currently widely applied in the late phases of clinical trials. Previous clinical studies regarding the application of MSCs for immune-related diseases are classified based on criteria, including disease type, donors, and tissue sources (50). A brief timeline for significant events in studying the immunosuppressive effects of MSCs and the progress in their clinical applications was discussed in detail by Wu et al., confirming their safety and efficacy (51).

The consequences of autologous/allogeneic MSC therapy are heavily investigated in clinical trials to treat autoimmune and inflammatory diseases along with transplant rejection, including graft versus host disease (GvHD), acute respiratory distress syndrome (ARDS), acute lung injury (ALI), ankylosing spondylitis, autoimmune hepatitis, type 1 and 2 diabetes, refractory autoimmune thrombocytopenia, skin diseases, inflammatory bowel disease (IBD), MS, CD, Stevens-Johnson syndrome (SJS), systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), osteoarthritis (OA), and even in coronavirus disease of 2019 (COVID-19) as an immune-dysregulating infectious disease. As dysfunctional MSCs were proposed to be involved in the pathogenesis of some autoimmune diseases, such as SLE or immune thrombocytopenia (ITP), the utilization of MSC-based therapeutics could be considered more seriously as an effective treatment strategy (52–55).

In addition to the intrinsic capacity for migration towards inflammatory tissues, MSCs possess different immunomodulatory properties, including the inhibition of apoptosis, induction/maintenance of immune tolerance, and immunosuppression (10). The beneficial properties of MSCs, which play a role in modulating inflammatory responses and infiltration processes, have been attributed to a synergy by i ) MSC-released signaling molecules, ii ) the reaction of immune cells and other target cells to these signaling molecules, and iii ) feedback in the MSC-molecule-target cell axis (56). However, the precise way MSCs act is still under investigation.

MSCs produce and release extracellular vesicles (EVs) enriched with immunomodulatory factors as long as their cellular resources are kept in proper and defined conditions (57). Furthermore, they can induce tolerogenic Treg cells and even tolerogenic DCs following their administration (58, 59). The immunomodulatory, pro-angiogenic, and tissue-tropic activities of MSCs-secretome and -EVs are comparable to the producer cells, and they were proposed as valuable sources for treating inflammatory disease (60, 61). Hence, based on their paracrine effects, the potential to alter the function of immune cells, and considerable potential for EV production, we focus on these cells and their EV counterparts. The potential of MSCs for large-scale production of EVs with immunomodulatory content was compared with HEK293, ESCs, iPSCs, and ESC/iPSC-derived MSCs from different aspects in a previous publication (62). We highlight some open issues regarding the differences between cells and EVs to compare their applications.

EVs of 30 nm to 1 µm in diameter are described as “heterogeneous bilayer membranous vesicles lacking a functional nucleus” and as “multi-signal messengers” which are enriched with a variety of biomolecules (68, 69). The biogenesis, intracellular trafficking, and secretion of EVs are tightly regulated by specific proteins, including the RAB family of small GTPases (70) and lipids, and need the molecular motors- (myosin, kinases, dynein, and GTPases) mediated cytoskeleton rearrangement (68, 71). These vesicles share characteristics regarding their membrane organizers, lipid content, cell adhesion molecules, intracellular trafficking mediators, enzymes, signal transduction, biogenesis factors, chaperones, and nucleic acids, while they are different in some others (72). Identifying these commonalities and differences between EV types will help to better understand their biology and function.

Different classification criteria and nomenclature have been applied to define various types of vesicles (73), including exosomes, microvesicles/ectosomes, apoptotic bodies, migrasomes, and oncosomes that are released by the cells under normal physiological and pathological conditions (74–76), amongst exosomes are the most well-known cellular nano-vesicles (77)( Figure 2 ). The most recently introduced member of the secreted nano-particles family includes exomeres [with a dot-shaped morphology], which are smaller than 50 nm and are enriched with metabolic pathways [glycolysis and mTORC1 signaling regulatory proteins such as β-galactoside α2, 6-sialyltransferase 1 (ST6Gal-I) and the EGFR ligand, amphiregulin (AREG) (76, 78, 79). Supermeres are the other new class of extracellular particles, indicating a different protein and RNA cargo than EVs and exomeres. Although similar to EVs and exomeres, they have the potential to carry extracellular RNA in a protected mode and play a role during chronic disease conditions (80–82). It was proposed that supermeres may be functionally involved in the immune supervision of cell-death-derived ribonucleoprotein complexes (83).

However, the isolation steps or the characterization methods for a definite type of vesicles are not convincing enough. For instance, essential controls for the different subtypes are not considered while comparing the functional immune-related properties. On the other hand, it was shown by experimental studies that their immunomodulatory characteristic is prominently diminished upon removing EVs from MSC-secretome. Although several subtypes of nano-vesicles are currently recognized, due to the recent bibliographic studies (2002-2021), two terms, “exosomes” and “extracellular vesicles”, are yet the most frequent keywords applied to describe the role of these cell-derived membranous nano-vesicles in autoimmune diseases (84, 85). Despite the initial trend, it is now evidenced that the impact of EV subtypes and their versatility is so critical compared to what it inferred at first (86, 87) and should not be underestimated during designing and performing future pre-clinical experiments and clinical trials. Here, following the International Society for Extracellular Vesicles (ISEV) consensus recommendation (88), we apply the term “extracellular vesicles” as the representative for the nano-sized natural and non-replicative particles focusing on membranous exosomes and ectosomes, which are produced by almost all cell types (68, 76, 89–92).

Isolation (93), characterization (94), labeling, and quantification strategies of different EV subtypes are constantly evolving (95). It should be noted that the various isolation methods of EVs end with different characteristics and potencies in terms of application (96, 97). It was shown that EVs from 3D cultures are functionally different and are more similar to the patient-derived EVs compared to those obtained from traditional 2D-culture conditions (98–100). These imply that spending time for optimization of isolation method/s (101, 102) to reach pure EVs with higher efficiency (103) in a reproducible manner (104) is of great importance. This way, it would be possible to compare the results more accurately (102).

EVs’ biogenesis, biological properties, and contents have been comprehensively discussed over the last few years and shown that they are essential for their function and clinical applications (68). The quality of endogenous EV preparations (105), from body fluids, tissues, or intracellular spaces (40), is affected by the circadian clock (106). In addition, isolation method, purity, storage protocol (107), time and temperature (108), presence or absence of anti-coagulants, isotonic buffers, other reagents such as cryoprotectants (109), pH, and ionic strength (110–112) are the other influencing factors. The concentration and content of EVs are influenced by demographic factors, including sex, age and race to some extent (113) and more than that, the EV cargo could be deliberately engineered through microenvironment modifications (114).

EVs are considered a powerful means for short- and long-distance communications (juxtacrine, autocrine, paracrine, and endocrine modes (65)) of the cells (115–117). These cross-talks have remained conserved throughout evolution and play a role in “inter-kingdom” mutual communications (118). This way, EVs can mediate (attenuate or promote) cell death, inflammation, and immune responses in a context-dependent mode (119, 120). Extracellular matrix-bound EVs are a recently-introduced subset of EVs with predicted roles in tissue repair and regeneration, which may be mediated by their active or passive tissue traversing or extracellular matrix (ECM) remodeling capacity (121). While cells are changing entities, EVs do not change following their release. Their biological function could, however, be impaired upon exposure to hostile microenvironments (122). In other words, EVs are membranous packages that carry various DNA-, RNA-, protein-, lipid cargoes and metabolites to their target cells (69, 123–125). They transfer active components to their target cells via different mechanisms, including receptors (integrins, tetraspanins, lectins, proteoglycans)-mediated binding, plasma membrane fusion, macropinocytosis, and lipid raft-mediated endocytosis, clathrin-mediated endocytosis, phagocytosis, and caveolae-dependent endocytosis, and trigger certain downstream events.

Tissue-derived EVs (Ti-EVs) reflect the tissue of origin microenvironment and cell-to-cell intercommunications and are critically important in studying and monitoring some developmental events. Like cell-culture-derived EVs, optimized and standardized protocols should be applied for efficient isolation and enrichment of Ti-EVs (126). Examples of the importance and usability of Ti-EVs have been reported in the pathogenesis, diagnosis, and treatment of Alzheimer’s disease (127, 128). EVs from individuals diagnosed with AD possess a specific protein content and miRNAs profiles which could be applied for early detection of the disease. In a recent experiment, NMDAR2A was identified as a CNS-specific EV surface marker, and it was shown in a multicenter study, including both discovery and validation cohorts, that the number of plasma EVs with NMDAR2A and L1CAM expression is lower in patients with AD in comparison to the healthy controls (129). On the other hand, as EVs are responsible for the “spread of tau pathology”, their secretion inhibition could be proposed as a novel targeted therapeutic strategy for patients with Alzheimer’s disease (126, 130, 131). Bone marrow MSC-EVs can decrease BACE-1 and Aβ, increase sphingosine-1-phosphate and exert amenable results in the context of Alzheimer’s disease (66, 132, 133).

Preparing qualified endogenous EVs from bio-fluids as the primary source is more complex, as several factors interfere with efficient EV isolation and purification (134). The quality of the final sample will be determined as a consequence of donor-related factors in addition to the biophysical and chemical properties of the fluid. The material of the sample collection tube, the application of coagulant or anticoagulants, the initial volume of the sample, mixing or agitation steps, the duration of pre-processing or pre-storage steps, and hemolysis degree or other depletion strategies are among the essential determinants of the quality of EVs. It is also essential to manage possible contaminations (111, 135).

Among their broad applications, EVs are sometimes utilized based on their nano-sized features and ability to pass through blood barriers (136). In such cases, EVs could be applied as a new paradigm with amenable priorities (137, 138). EVs can achieve most features of the synthetic nano-carriers (i.e., liposomes and nano-particles). Moreover, they have higher biocompatibility, chemical stability, longer-distance intercellular communications, and cell fusion capacity (139). As non-toxic nano-carriers, EVs demonstrated better pharmacokinetics and pharmacodynamics features (absorption, distribution, metabolism, and excretion) than synthetic nano-carriers (140, 141). However, there are controversies regarding their application as drug delivery systems (DDSs) like nano-shuttles (142). Some technical limitations have been raised regarding their flexibility in applying different reagents, preparation procedures, and surface functionalization strategies (139). Also, they have widely been recognized as natural and valuable means for diagnosing and monitoring various disease conditions and novel therapeutic agents (111, 143).

Therapeutic applications of EVs in managing different inflammatory and autoimmune diseases and their immunomodulatory features and composition were discussed in recently published articles (144–146). Extensive parallel research during the last decade has revealed new aspects in the biology of MSCs, leading to changes in our orientation regarding their applications. These changes highlighted the importance of the paracrine effects of MSCs and proposed the possibility of replacing the cells with their cell-free products, especially in the allogeneic context (147). “Secretome” is the common term that represents all substances released by MSCs, including free-soluble factors (various growth factors, chemokines, and cytokines) and insoluble EVs (148–150). Different aspects of the pleiotropic therapeutic effects of MSCs and their paracrine action were discussed in a recent publication, emphasizing the active components of the MSCs’ secretome (116). The paracrine strength of these cells is so significant that it has been suggested that the name of these cells would be changed to the medicinal signaling cells (151, 152). Munoz-Perez et al. reviewed the latest trends regarding the potential of MSC-secretome to treat immune-encountered inflammatory diseases (153).

There is a growing tendency toward local administration of EVs employing diverse scaffolds from acellular tissues to solid patches and hydrogels (154–158). One good reason favoring the local administration of EVs is the lower amount required, making their utilization much more cost-effective. Furthermore, their gradual and sustained release following the administration is a positive point (87). Moreover, the application of scaffolds/biomaterials increases the bioavailability, homing, and/or biocompatibility of EVs at the site of injection, while they are also effective in avoiding the side effects of the injection procedure (159). This preference may be pronounced when dealing with local inflammatory conditions or tissue-specific autoimmune diseases.

It was confirmed that MSC-EVs are therapeutically efficient in 30 animal models mimicking various human diseases (111). While different roles were described for EVs during the pathogenesis/immunopathology of Rheumatoid arthritis and joint impairment of RA patients, the therapeutic impact of MSCs- and non-MSCs-EVs (dendritic cells, polymorphonuclear neutrophils, myeloid-derived suppressor cells) were confirmed in animal studies (160). The first case report regarding the benefits of the MSC-EVs was related to a patient with GvHD (161). The high significance of MSC-EVs, as a novel therapeutic for managing Alzheimer’s disease, has been thoroughly discussed recently in mice and human cases, emphasizing the mechanisms of the disease progression (66).

Similar to their producer cells, MSC-EVs eventually found their place in pre-clinical and clinical studies (162). However, each step of scalable EV manufacturing under GMP compliance is still exposed to several challenges (163, 164). These challenges exist not only in upstream and downstream stages (165, 166) of EV production and isolation, such as cell culture and purification steps, respectively, but also during EV storage (108), quality control (167), and functional evaluation steps (148, 168).

EVs were previously enriched from the conditioned media of different immune cells, including dendritic cells, T regulatory cells, M1/M2 macrophages, and CD4⁺ T cells (169–171), and their potential was evaluated in the context of autoimmune diseases (44). However, despite the initial trend, it is now evident that these vesicles, regardless of their producer cell type, are produced with different biological content under different circumstances (172), and their application is accompanied by several challenges and complications (173–175). One remaining question could be: How can we learn from the long-term and extensive research on their producer cells to make EVs-based therapeutics safe and efficient in the frame of translational medicine? Finally, the main questions considered to be addressed by this review are the level of superiority of EVs to intact cells in reducing patients’ symptoms of autoimmune disorders and the maturity of techniques for preparing clinical-grade EVs (176).

MSCs have immunosuppressive and immune privilege properties; hence, their interactions with the immune system are expected to benefit immune homeostasis and restore immune tolerance. However, there is concern regarding their allogeneic and repeated administration (177). MSC-EVs, like their original cell sources (MSCs), are believed to have low immunogenicity as well as lower carcinogenic risk (178). They do not release certain inflammatory factors (IL-1, IL-6, and IL-8), which may be vastly released from dead cells following cell injection procedures, especially in tissues lacking adequate blood/nutrients (179). In addition, EVs demonstrate enhanced circulation stability due to the immune system’s evasion (39). Although similar to allogeneic cells, these particles may potentially carry some immunogenic proteins (e.g., MHC molecules), it is open to question whether these proteins are transferred from the vesicles to other cells in communication and if they can trigger alloimmune responses ( Table 1 ) (29, 177, 180).

MSCs‐EVs exert their immunomodulation through a similar mechanism known for MSCs‐associated tolerance. This process is mediated by programmed death ligand‐1, galectin‐1, and TGF-β (206). They impose immunomodulation instead of immunosuppression, leading to reduced immune reactions and homeostasis (188, 189). MSC-EVs contain many parental biomolecules, including chemokines, cytokines, and growth factors (207). Suh et al. described the mechanisms responsible for the therapeutic effects of EVs in the context of different inflammatory diseases. Similar to the cells, the mechanisms responsible for the therapeutic effects of the EVs in the context of different inflammatory diseases are reducing the microglia/macrophage activation, oxidative stress, pro-inflammatory cytokine and chemokine release, T-cell activation, tissue fibrosis, viral infectivity, immune cell infiltration, and apoptosis/necrosis (146). Moreover, EVs regulate M1/M2 macrophage polarization for the benefit of M2 cells, induce collagen regeneration, and prevent scar generation (181).

Despite similar immunomodulatory properties, differences in the two entities’ level and mechanism of action have been reported. The way EVs communicate with the effector cells of the immune cells is not as simple as it sounds and the mediators of MSC-EV-based immune modulation are yet remained incompletely described (182). In comparison to MSCs, EVs have less cell-dependent and independent mechanisms to exert their immunomodulatory properties upon exposure to various immune system cells (183–187). On the other hand, EVs can penetrate deeper into the tissues (208, 209), so they may have a higher chance of regulating the immune cells (183). In addition, the therapeutic anti-inflammatory properties of EVs could be more pronounced due to their capacity for passing through blood barriers (210). Also, the immune rejection rate of EVs is considerably low compared to the producer cells, as they have a low-level expression of membrane histocompatibility molecules (187). EVs have been shown to carry, mediate, and regulate cytokine transport from cells to members of the immune system (190).

Unlike the cells, MSC-EVs cannot impact the production of CD3+ T cells. Allogeneic EVs could bind to T cells, but they do not stimulate these cells during in vitro experiments unless in the presence of APCs in the culture media. In addition, EVs do not change the ratio of B cells and plasma cells following their co-culture with human peripheral blood mononuclear cells (PBMCs) in vitro (182, 191–193). Allogeneic EVs would activate T cells following their in vivo administration in mice only if delivered into an inflammatory microenvironment (196). MSC-EVs can interfere with the feed-forward loop between complements and neutrophils via the inhibition of the complement activation in a CD59-mediated manner. Hence, MSC-EVs are influential players able to modulate inflammatory status spatially and temporally (194). While MSCs and their cell-free counterparts can increase the ratio of FOXP3+CD25+CD4+ T regulatory cells, EVs are less functional against DCs (185, 195).

The therapeutic effects of extracellular vesicles as new biologicals for immune regulation have been supported by pre-clinical studies. Native or modified MSC-EVs were used to regulate the immune-associated cells while focusing on their mechanism of action (211, 212). There is almost no doubt that MSC-EVs can trigger anti-inflammatory responses based on restoring the balance action in the milieu or immune cells. This could be attributed to their miRNAs, other non-coding RNAs, or protein cargo; otherwise, surface EV-associated proteins may pave the role by activating the downstream intracellular pathways in recipient cells (211, 213). The heterogeneous nature of producer cells, diverse EV isolation, quantification and standardization strategies and the critical role of disease-specific pathways are considered the main challenges to introducing unique regulatory mechanisms (212, 214). Moreover, the clinical practice of EVs is challenged by their heterogeneity to some extent. Optimization of current digital assays has been proposed to do single-vesicle-resolution studies, especially during the clinical analysis of liquid biopsies (215). Immune biocompatibility of EVs is the other criterion to be investigated in autoimmune diseases. It was shown in a previous study that although both autologous and allogeneic small EVs may have efficient therapeutic effects, autologous EVs are more viable and impactful in damaged tissues (216, 217).

Recent updates unravel the crucial role of EVs and vesicular trafficking in the immune system (65) and during the pathogenesis of immune-based diseases, in addition to their therapeutic perspectives (218). An accurate understanding of the mechanisms by which tumors or other pathological statuses induce immune cell dysfunction via their extracellular particles will allow a better understanding of how vesicles interact with the cells of the immune system. These data would potentially facilitate developing novel methods to face the overactive immune system and the associated diseases.

The migratory capacities of MSCs towards ischemic and damaged sites have received substantial consideration for treating diffuse and localized inflammatory and degenerative conditions (219). However, low homing efficiency is a significant drawback associated with temporary therapeutic benefits (220). Hence, diverse approaches have been employed to enhance the entrapment rate of systemically infused MSCs (221). EVs must also follow the route to the affected areas as membrane-bound cell-derived structures. Various cell-to-cell communication functions are ascribed to EVs, including autocrine and paracrine missions (222). Sung et al. discovered that EVs could even contribute to directional cell migration. They confirmed that malignant cells could find their path toward EV deposits, resulting in EVs’ endocytosis (223). In specific migration paths, cells use almost the same delivery process as depicted for growth factors and peptides, in addition to their active migration (224). Here, we raise some open issues to compare MSCs and EVs concerning their homing capabilities.

When systemically administered, exogenous MSCs imitate the delivery route via which endogenous MSCs reach their destination (225). As tissue injuries often coincide with inflammation or ischemia, inflammatory factors could provide cues to mobilize cells towards the damaged tissues. The expression of homing receptors on the cell surface is proven to mediate MSCs’ migration towards their ligands at the ischemic target microenvironment (226). Upon arriving at the damaged site, circulating MSCs or their bioactive components need to leave the vasculature and transmigrate across the endothelium to get into the stromal region, where they exert their primary function both in passive and active manners (220, 227). Paracrine interactions of culture-expanded cells and the target milieu necessitate reciprocating various bioactive determinants (228). Furthermore, the bio-distribution of exogenous MSCs has always been challenging, as many infused MSCs are entrapped in other organs, particularly the lungs, liver, and spleen (220).

Differences between cells and their extracellular descendants have put a debate regarding their potential applications (see Table 2 for a detailed comparison). The preparation of characterized EVs and their standardization needs more interdisciplinary knowledge, as they are nano-sized entities (273). In the case of cells, there is less concern regarding the preparation of a determined number, while in the case of EVs, it is not as easy due to the differences in the availability and sensitivity of the quantification methods (274). Different criteria have been applied to quantify EVs, including the number of producer cells in the culture, the protein content of EV preparations, and the number of EVs/ml of the final preparation. This puts a big hurdle in comparing the results of different experiments (185). Unlike cells that need vascular structures to receive essential nutrients and survive, EVs could be administered in tissues with a lower capillary network, such as intervertebral discs. Also, as EVs do not release toxic or harmful metabolites, there is no concern regarding their utilization in such tissues (179).

Having mentioned these differences, EVs generally have some clinically relevant superiorities to cell therapy (185, 264, 265). The immune rejection rate of EVs is considerably low compared to the producer cells (181), so they have an increased half-life and are more stable in circulation (39). EVs demonstrate an innate tropism to specific tissues (76) and exert cell-targeting properties (111). Unlike cells, there is no concern regarding the possibility of necrosis or their abnormal differentiation (181). They do not show self-replicative and tumor-formation properties (139) and are not seriously affected by the surrounding inflammatory microenvironment (186). They reach deeper into injured tissue layers (183). Moreover, EVs have a more straightforward pre-banking capacity, are less sensitive to low temperatures, and are more stable during freeze-thaw cycles. EVs can also be combined with existing compositions or drug delivery methods (66). These vesicles would provide the scientists with a suitable manipulative platform (269).

Both cells and EVs can be contaminated with bioactive components, including but not limited to viruses, bacteria, and endotoxins, that should be checked before any administration. Considering similarities between viruses and exosomes (306–309), viral contaminations for EV preparations are one of the main concerns. Usually, the size of bacteria and fungi are larger than EVs; however, infections specifically by intracellular contaminations such as mycoplasma can change both cell and EVs properties (310–313).

The risk of uncontrolled proliferation of cells is associated with cell-based therapies. Unlike cells, EVs cannot proliferate, and multiple administrations are usually required to reach comparable results. However, more studies are required to show if EVs have machinery like viruses to propagate.

Similar to cells, EVs carry DNA (314–316) and mitochondria content (317–319). Recently, it has been shown that genetic materials can also be incorporated into the EV bio-corona (63, 319–321). Therefore, safety concerns are associated with administering both cells and EVs regarding their genetic materials.

Considering the danger of cross-species contaminations, cells are cultured using human-derived components in xeno-free culture systems. The same issues for EVs should be considered even more rigorously. Firstly, cells can uptake materials from the medium during the culturing process and re-packing them into their EVs derivatives. Second, EVs or particles originating from bovine-derived serum supplements cannot be efficiently removed based on current methods (321–324). Therefore, the risk of cross-species contaminations should be assessed in both cells and EV administration.

Unlike cells, EVs can cross barriers, including but not limited to the blood-brain barrier (325–327). The risk of passing through the barriers is another issue that should be considered explicitly in the EV administration.

The issues raised about the enclosed risk of EV administration belong to unmodified EVs. In cases where there is a need to expose the cells to different physical or chemical agents or to engineer the producer cells or EVs, more details should be taken into account (328). Maybe that is why a debate is being opened about whether exosomes could be considered a medicinal product rather than a biological product similar to their producer cells. The point here remains to be done is the release of internationally accepted guidelines, designed explicitly for EVs and their clinical applications.