Introduction

Coronavirus disease 2019 (COVID-19) is a highly infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Based on available data, one of the hallmarks of SARS-CoV-2 infection pathogenesis is cytokine storm in the lung. Acute virus-induced cytokine release can lead to pulmonary edema, ventilatory dysfunction, and acute respiratory distress syndrome (ARDS) [1], especially in severe cases, which is characterized by the up-regulation of proinflammatory cytokines and chemokines, abnormal cellular immune response, respiratory and cardiovascular failure, end organ injury, and possibly death [2, 3]. To date, however, no specific antiviral drug has been proven effective in the treatment of COVID-19 and therefore is urgently required. A variety of strategies have been proposed to control cytokine storms and reduce mortality, such as selective cytokine blockade (e.g., anakinra or tocilizumab), JAK inhibition, intravenous immunoglobulin administration, mesenchymal stromal cell (MSC) therapy, and artificial blood purification [4, 5].

Based on the results of several models and stage-clinical trials of ARDS and sepsis, cell-based therapy, especially stem cell therapy, has emerged as a promising therapeutic area, especially for the treatment of COVID-19 [6,7,8]. MSCs are pluripotent cells obtained from different tissues, such as adipose and bone marrow. Due to their immunomodulatory properties, MSCs can alter immune cell function, modulate immune responses, and reduce inflammation-induced lung injury [9]. These cells can also prevent apoptosis and regenerate lung cells, especially type II alveolar cells, by producing growth factors such as keratinocyte growth factor (KGF), vascular endothelial growth factor (VEGF), and hepatocyte growth factor (HGF) [10]. Furthermore, by expressing indoleamine 2,3-dioxygenase (IDO), MSCs can stimulate interferon-γ (IFN-γ) production and IFN-γ-independent restriction of viral replication [11, 12]. Thus, the immunomodulatory, tissue repair, and antiviral properties of MSCs highlight their potential for the treatment of COVID-19 [10].

The safety and efficacy of MSCs have been demonstrated in multiple clinical trials related to the treatment of COVID-19 [6,7,8, 13]. As of October 2021, 51clinical trials for MSC-based COVID-19 therapy have been registered (http://clinicaltrials.gov/ and http://www.chictr.org.cn/) (Table 1). In this article, we summarize the findings of seven published clinical trials (Table 2) to help clarify the regulation of cytokines by MSCs in patients with COVID-19, and whether this therapy can improve patient symptoms, shorten hospital stays, and reduce mortality, in conjunction with a high safety and low adverse event profile. The underlying mechanisms by which MSCs may act are also explored, with the view to provide a reference for subsequent trials and research.

Table 1 Trial of registered MSCs in the treatment of COVID-19
Full size table
Table 2 Summary of published clinical trials of MSCs in the treatment of COVID-19
Full size table

Efficacy

Mortality

According to the World Health Organization (WHO), the cumulative number of deaths due to COVID-19 has surpassed 5.5 million globally. Multiple studies have shown that MSC therapy can significantly reduce the incidence and mortality of critical illness (Table 3) [14,15,16,17,18,19]. For example, Shu et al. [14] found the mortality in the MSC treatment group was zero. In addition, Xu et al. [17] reported significantly higher survival in patients treated with MSC (92.31%) than those in the routine treatment group (66.67%). Another study reported a 28-day survival rate of 91% in MSC-treated patients compared with 42% in non-treated controls, with the control group also showing a higher risk of death (hazard ratio (HR) 8.76; 95% CI 1.07–71.4) [18]. Furthermore, serious adverse event-free survival showed significant improvement with MSC treatment compared with the controls (HR 6.22; 95% CI 1.33–28.96). Thus, the above results confirm the safety of MSC therapy and its effectiveness at reducing mortality and improving survival. However, due to the small sample sizes in the above studies, data from large-scale clinical phase 3 trials are still required.

Table 3 Clinical efficacy of MSCs in the treatment of COVID-19
Full size table

Systemic changes and symptoms

MSC treatment can significantly shorten the time to clinical symptom improvement in COVID-19 patients. Lanzoni et al. [18] showed a significantly shorter COVID-19 recovery time following MSC treatment, with an HR for recovery of 0.29 (95% CI 0.09–0.95) in the control group versus the MSC-treated group. Several other studies have also demonstrated that MSC therapy can significantly expedite patient recovery. A recent experiment comparing pulmonary function recovery and comprehensive reserve capacity based on a 6-min walk test (6-MWT) found that walk distance was longer in MSC-treated patients compared with the controls, although maximal forced vital capacity (VCmax), diffusing lung capacity for carbon monoxide (DLCO), six category scale, oxygen therapy status, and mMRC dyspnea score did not differ significantly between the two groups [20]. In critically ill patients, the partial pressure of arterial oxygen: percentage of inspired oxygen (PaO2/FiO2) ratio showed improvement after MSC treatment [16]. Furthermore, MSC-treated patients demonstrated significant improvement in clinical symptoms and were discharged from the ICU within 2–7 days after MSC infusion, with significant relief of dyspnea, decrease in respiratory rate within 48–96 h, and improvement in oxygen saturation [15]. Shu et al. [14] also reported significant improvements in weakness, fatigue, shortness of breath, and low oxygen saturation in MSC-treated patients compared with the controls.

In addition to clinical symptoms, laboratory parameters have also been examined [15, 16], including C-reactive protein (CRP), alanine aminotransferase (ALT), creatinine, serum ferritin (SF), and platelet levels, which all returned to their normal range after MSC administration. These results suggest that MSCs not only improve pulmonary symptoms but also positively impact the functional recovery of multiple organs, such as the liver and kidney.

Computed tomography (CT) imaging

As clinical symptoms are influenced by multiple factors, and the evaluation of symptom relief is subjective, various researchers have assessed lung lesions in COVID-19 patients using imaging. Studies using CT imaging have reported that time of pulmonary lesions is significantly shortened in patients treated with MSCs. A single-center, open-label, randomized, standard treatment-controlled trial showed that CT score, number of lobes affected, ground glass opacity, and solid changes were significantly improved in MSC-treated patients compared with the controls [14]. In another study, the rate of chest imaging changes 1 month after MSC infusion was significantly improved in 85% of MSC-treated patients compared to 50% of control-group patients [17]. Previous research also reported that lung lesions were well controlled within 6 days and completely disappeared within 2 weeks after MSC infusion [16]. A phase 2 clinical trial also reported a significant decrease in total lesion proportion in the whole lung measured by CT from baseline to day 28 after MSC infusion [20].

Cytokines

Cytokine storms are considered one of the main characteristics of COVID-19. To date, however, no definitive therapy has been shown to completely control cytokine storm or restore organ damage caused by infection with SARS-CoV-2. MSC transplantation may act as an immunomodulator in the development of cytokine storm caused by inflammation. Therefore, many scholars believe that MSCs have a decided advantage in controlling cytokine storm induced by COVID-19.

Several published clinical trials indicate that MSCs can effectively control the expression of inflammatory cytokines in COVID-19 patients [14,15,16, 18, 19]. Several studies have reported that proinflammatory cytokines, such as interleukin (IL)-8, tumor necrosis factor (TNF)-α, CRP, IL-6, INF-γ, IL-2, IL-12, and IL-17A, decreased significantly after MSC infusion, while anti-inflammatory cytokines IL-4 and IL-10 increased significantly [14, 15, 18]. However, other research reported that IL-β and TNF-α levels were not significantly reduced after MSC administration [19]. In addition, a clinical trial reported no significant changes in cytokines, but a reduced trend in cytokine levels within 14 days (IFN-γ, TNF-α, monocyte chemokine-1 (MCP-1), interferon inducible protein-10 (IP-10), IL-22, IL-1RA, IL-18, IL-8, and MIP-1) [16]. Intragroup analysis found that granulocyte macrophage colony stimulating factor (GM-CSF), IFN-γ, IL-5, IL-6, IL-7, TNF-α, TNF-β, platelet derived growth factor-BB (PDGF-BB), and RANTES decreased significantly from days 0 to 6 in the UC-MSC treatment group [18]. These data, to some extent, demonstrate the effect of MSC treatment in patients with COVID-19, and the differences in outcomes between trials may be due to differences in disease degree and age. Thus, the results of large-scale phase 3 trials should help further define the role of MSCs in treatment.

Safety

Drug and treatment safety must be considered in clinical application. Recently, the clinical outcome of a 65-year-old female patient with COVID-19 treated with allogeneic human umbilical cord blood-derived MSCs (UCB-MSCs) was reported, with no adverse events noted during treatment and was well tolerated [21]. Hashemian et al. also reported that the liver and kidney function of patients were not affected during MSC infusion [15]. Another study found that MSCs can cure or significantly improve the functional results of patients without obvious adverse events [22]. These results illustrate, to some extent, the safety of MSCs for the treatment of COVID-19, but the number of included experimenters is small and lack of randomized controlled trials. In view of this, many clinical trials have begun to evaluate the safety, feasibility, and tolerability of MSCs.

Various trials suggest that MSC therapy is safe for COVID-19. A prospective double-controlled trial reported no adverse or serious adverse events related to MSC treatment [19]. Another phase 1 clinical trial evaluated the safety of MSC infusion in patients with moderate to severe COVID-19 and reported that three patients experienced adverse effects, although these events were considered to be caused by disease progression based on pre-existing symptoms [16]. Similarly, two clinical randomized controlled trials [18, 20] showed that the overall incidence of adverse events was similar in the MSC and control groups, but these events were largely unrelated to treatment, and significantly more subjects experienced serious adverse events in the control group than in the MSC group. Thus, the above findings reflect the safety of MSC therapy and its potential benefits at reducing COVID-19-related adverse events.

In general, although the safety of MSCs is acceptable, treatment may not be applicable to some COVID-19 patients with serious complications. For example, a phase 1 trial [15] showed that multiple infusions of high-dose allogeneic MSCs were safe and rapidly improved respiratory distress and reduced inflammatory biomarkers in some cases of critically ill COVID-19-induced ARDS, but four patients with multiple organ failure or sepsis died within 5–19 days (mean 10 days) after the first MSC infusion. These findings suggest that contraindications need to be strictly considered when selecting patients for MSC therapy.

Mechanisms of COVID-19 treatment with MSCs

Several studies [23,24,25,26,27] have shown that MSCs can be safely infused intravenously or via the endobronchial route in humans, thus allowing MSCs to accumulate in the lungs to improve the lung microenvironment, protect alveolar epithelial cells, protect against pulmonary fibrosis, and improve lung function [28,29,30]. Although the specific molecular mechanisms underlying the effects of MSCs on COVID-19 treatment require further research, several studies have investigated possible processes, as summarized below.

Immunomodulation

MSCs exhibit strong immunomodulatory potential. Their immunomodulatory functions are mainly exerted through cell-to-cell contact, paracrine secretion, endocrine action, and immune cell interactions (e.g., T cells, B cells, natural killer (NK) cells, macrophages, monocytes, dendritic cells (DCs), and neutrophils). Thus, MSCs participate in both innate and adaptive immunity [31,32,33], with regulatory T cell (Treg) and monocyte interactions appearing to play a key role [34]. Of course, it may vary according to the pathological mechanism of the disease, source of MSCs, and route of administration.

MSCs show immunosuppressive effects when exposed to sufficiently high levels of pro-inflammatory cytokines [35], but can promote an inflammatory response under low levels of TNF-α and IFN-γ [35]. Thus, MSCs may need to be triggered by inflammatory cytokines to become immunosuppressive, and the inflammatory environment may be a key factor affecting immunoregulation of MSCs. One of the characteristics of COVID-19 is the formation of an inflammatory cytokine storm, which may provide an inflammatory environment for the immunomodulation of MSCs.

After infusion of MSCs in COVID-19 patients, IL-1RA, IL-6, HGF, prostaglandin E2 (PGE2) secreted by MSCs promoted monocyte/macrophage differentiation into anti-inflammatory/immunomodulatory (type 2) phenotype, and directly inhibited differentiation into type 1 phenotype and DCs [36, 37]. On the one hand, type 2 monocytes/macrophages secrete high levels of IL-10, reduce expression levels of IL-12p70, TNF-α, and IL-17, prevent monocytes from differentiating into DCs, and convert monocytes into an anti-inflammatory, IL-10-secreting subtype via a positive feedback loop [36], while high levels of IL-10 inhibit T cell activity [38]. On the other hand, macrophages release more CCL-18 and transforming growth factor-β1 (TGF-β1) during differentiation into type 2 macrophages, which helps induce Treg formation, while CCL-18 converts memory CD4+T cells into CD4+CD25+Foxp3+Treg cells and increases IL-10 and TGF-β1 generation [39,40,41]. Furthermore, it was observed in the asthma model that the infused MSCs was engulfed by pulmonary macrophages, resulting in a shift of monocytes to a type 2 immunosuppressive phenotype, polarization of CD14++CD16− classical monocytes to a CD14++CD16+CD206+ immunoregulatory intermediate subset with anti-inflammatory properties, and increased expression of IL-10 and programmed death ligand-1 (PD-L1) [42,43,44], thereby driving the immune response toward an anti-inflammatory response.

MSCs act on the adaptive immune system, particularly T cells [32, 33], in various ways, e.g., inhibition of the proliferation, cytokine secretion, and cytotoxicity of T cells and regulation of T helper 1 (Th1)/T helper 2 (Th2) balance and Treg function. MSCs can induce IL-10 and PGE2 production and inhibit IL-17, IL-22, and IFN-γ levels to limit Th17 differentiation [45] and suppress Th17 responses by modulating the IL-25/STAT3/PD-L1 axis [46]. Upon interaction with DCs, MSCs can cause a shift from pro-inflammatory Th1 cells to anti-inflammatory Th2 cells [47]. Induction of CD4+CD25+Foxp3+ Tregs is one of the main features of MSC-mediated immune regulation, and MSCs can secrete TGF-β1 and IDO to induce the formation of Tregs [39, 47]. MSCs can directly interact with B cells and promote the generation of regulatory B cells (Bregs), which secrete IL-10 to convert effector CD4+T cells into Foxp3+Tregs [48, 49]. MSCs can also exert direct immunosuppressive effects on T cell behavior by suppressing CD4+T cell activation via the secretion of PD-L, including PD-L1 and PD-L2 [50, 51].

Promotion of tissue repair and regeneration

In addition to their immunomodulatory properties, the multilineage differentiation ability of MSCs also makes them ideal candidates for cell therapy. Numerous studies have demonstrated the regenerative capacity of MSCs in musculoskeletal system, nervous system, cardiac muscle, liver, cornea, trachea, and skin tissue repair [52].

COVID-19 is characterized by lung tissue damage, which can cause systemic multi organ damage, especially in patients who develop ARDS. After injection, MSCs can differentiate into lung tissue or secrete factors (e.g., angiopoietin-1 (ANGPT1), EGF, VEGF, PGE2, HGF, VEGFA, KGF, and IL-10) that can induce host repair/regenerative mechanisms [53], promote epithelial and endothelial repair, increase alveolar fluid clearance, regulate lung epithelial and endothelial permeability, and reduce inflammation in patients with ARDS lung injury [52, 54, 55]. In addition, these factors can promote tissue repair by supporting the growth and differentiation of local stem/progenitor cells, regulating the deposition of extracellular matrix molecules, stimulating anti-scarring pathways, and inducing neovascularization [56, 57]. Induction factors secreted by MSCs, such as VEGF, brain-derived neurotrophic factor (BDNF), and TGF-β1, can promote the development of self-repair. Activin A, EGF, KGF, HGF, and IGF-2 play important roles in MSC differentiation into epithelial cells by triggering appropriate signaling pathways [58, 59]. MSCs also show anti-apoptotic effects [53], and the release of KGF and HGF by MSCs can protect alveolar epithelial cells from apoptosis by increasing B-cell lymphoma-2 (BCL-2) expression and inhibiting hypoxia-inducible factor-1α (HIF-1α) protein expression [60]. Furthermore, the expression of various factors, such as VEGF, HGF, and TGF-β, can reverse endothelial cell apoptosis [61].

The immunomodulatory properties and tissue repair/regeneration abilities of MSCs highlight their potential in the treatment of COVID-19. MSCs can also exert antibacterial effects by secreting soluble mediators to reduce the number of bacteria, improve the antibacterial response of immune cells, and inhibit the migration of proinflammatory cells into infected tissues [53]. Thus, MSCs can participate in the inhibition of viral replication via different mechanisms [11, 12].

Conclusions

As cells with multilineage differentiation ability, MSCs are potential candidates for cell-based therapy to treat COVID-19. Various clinical trials have demonstrated the efficacy and safety of MSCs for the treatment of COVID-19 patients, especially critically ill patients, not only improving clinical symptoms, hospital stay, cytokine release, and mortality, but also showing a high safety profile with limited adverse events. However, further evidence from large-scale and long-term phase 3 clinical trials is still required. Although the therapeutic action of MSCs in COVID-19 patients appears to involve immunomodulation, promotion of repair/regeneration, and inhibition of viral replication through cell-to-cell contact and paracrine activity, the specific mechanism needs further investigation. However, we expect more data from currently progressing trials should support the role of MSCs in COVID-19.