Introduction

The classical acute respiratory distress syndrome (ARDS) was described for the first time in 1967 by Ashbaugh et al. as a destructive lung injury with an uncontrolled inflammatory process [1]. This acute inflammatory process causes severe alveolar damage and capillary basement membrane leakage leading to a progressive respiratory failure with high morbidity and mortality burden (Reviewed in [2, 3]). Classical ARDS can result from different causes including sepsis, pneumonia, and trauma (Reviewed in [2]). During the last years, more and more studies point towards that classical ARDS is an umbrella term that includes several different ARDS phenotypes [2, 4, 5••, 6, 7]. For example, Calfee et al. described in 2014 the two subgroups hyper- and hypo-inflammatory ARDS. The hyper-inflammatory group, with hallmarks such as high plasma levels of inflammatory markers including interleukin (IL)-6, IL-8, and plasminogen activator inhibitor-1 (PAI-1), was associated with more severe disease progression and lower survival rate [4]. These two phenotypes have also been identified in other cohorts and clinical trials including for example the SAILS trial and the HARP-2 trial [8, 9] with similar findings.

During the last decades, much effort has been put into understanding the pathogenesis and pathophysiology of ARDS, and many clinical trials have been completed in the search for an effective treatment. In particular, recent clinical trials have investigated mesenchymal stromal cell (MSC)-based therapies, based on the results from very successful pre-clinical studies utilizing bacteria, endotoxin, smoke inhalation, and other models of acute lung injury. These clinical studies have all demonstrated safety but unfortunately failed to uniformly prove significantly increased clinical outcomes [10,11,12,13].

In 2019, the coronavirus disease 2019 (COVID-19)-associated ARDS was described for the first time [14, 15]. Although there are similarities between the classical ARDS and the COVID-19-associated ARDS pathology, an increasing number of studies demonstrate that there are also differences between the syndromes [2, 16,17,18], which will be summarized in the section below (Fig. 1). The first clinical investigation using MSC-based therapy to treat COVID-19-associated ARDS was initiated very early in the pandemic outbreak, and the numbers of completed studies are currently increasing (Table 1) [19••]. Similar to the results from the MSC trials on patients with classical ARDS, MSC infusions were shown to be safe for the patients. Interestingly, a pooled analysis of the clinical trials using MSCs to treat COVID-19-associated ARDS completed between January 2020 and the end of July 2022 demonstrated a relative risk reduction for all-cause COVID-19 mortality (RR = 0.63) [19••]. However, the number of studies is still fairly small, and results derived from the different studies are difficult to compare to each other since the standard clinical treatment strategies changed during the pandemic, and sometimes also during an ongoing study [19••, 20, 21].

Fig. 1
figure 1

Important pathological similarities and differences between classical ARDS and COVID-19-associated ARDS. Classical ARDS and COVID-19-associated ARDS share several similarities in their pathology including significant lung inflammation with fluid accumulation in the alveoli, respiratory failure, and excessive immune response, but there are also important differences between the two syndromes which have been summarized in this figure. Understanding these differences is important for the clinical management and the development of therapeutic strategies for both classical ARDS and COVID-19-associated ARDS. Abbreviations: ARDS, acute respiratory distress syndrome; COVID-19, coronavirus disease 2019; SARS-CoV-2 virus, severe acute respiratory distress syndrome coronavirus 2. This figure was illustrated using Biorender.com

Full size image
Table 1 COVID-19 clinical trials
Full size table

In the first part of this review, we will discuss the differences and similarities between ARDS and COVID-19-associated ARDS pathology. In the second part, we will summarize, discuss, and compare the results from the clinical trials using MSC-based treatment for ARDS and COVID-19-associated ARDS.

Differences in Classical ARDS and COVID-19-Associated ARDS Pathophysiology

Classical ARDS and COVID-19-associated ARDS share several similarities in their pathology including significant lung inflammation with fluid accumulation in the alveoli, respiratory failure, and excessive immune response, but there are also important differences between the two syndromes (Fig. 1) [2, 16,17,18, 22]. The most obvious difference is that COVID-19-associated ARDS is exclusively caused by the SARS-CoV-2 virus, while the classical ARDS can have different etiologies including for example trauma, sepsis, and aspiration [2, 3, 14, 15, 22]. Moreover, other significant differences include differences in respiratory mechanics where higher respiratory system compliance and increased dead space fractions have been reported in patients with COVID-19-associated ARDS compared to patients with the classical ARDS [16, 23••, 24]. Moreover, increased levels of thrombotic mediators and lower expression of interferons have been reported in COVID-19-associated ARDS compared to the classical ARDS [25, 26]. A reduction of neutrophil-to-lymphocyte ratio with an impaired or delayed lymphocyte activation has also been observed in COVID-19-associated ARDS compared to classical ARDS, where the reduced neutrophil-to-lymphocyte ratio has been correlated to an increased disease severity [25, 26]. There is also evidence that patients with COVID-19-associated ARDS have elevated levels of circulating D-dimer [23••]. Understanding these differences is important for the clinical management and the development of therapeutic strategies for both classical ARDS and COVID-19-associated ARDS.

MSC-Based Therapies for Classical ARDS and COVID-19-Associated ARDS

What Is Known from Pre-clinical Experiments on MSC Treatment for Virus-Induced Acute Lung Injury?

There is a large body of literature demonstrating the efficacy of MSC administration in pre-clinical models of acute lung injury; however, most of them have focused on endotoxin- or bacterial-induced lung injury [27, 28] and not so much focus, so far, has been on virus-induced lung disease. Nevertheless, a few papers on MSC-based treatment for virus-induced acute lung injury have been published with contradictory results. For example, in a recent paper, Tan et al. investigated the effect of MSC treatment in H1N1 influenza virus-induced acute lung injury. Here, the authors reported that MSC treatment decreased the total cell count in bronchoalveolar lavage fluid and increased the number of infiltrating CD4+, CD8+, B-cells, T-cells, and monocyte in the alveolar space, but did not result in an improved survival rate or reduced viral load compared to untreated control cells [29••]. Similar results, i.e., no improved survival rate and no reduction in viral load, have also been reported by other groups [30, 31]. In contrast, Qin et al. demonstrated that MSC treatment reduced herpesvirus-68-induced pneumonia with decreased lung damage, decreased levels of inflammatory markers, and inhibition of viral replication compared to untreated control mice [32]. Similar results were reported by Chan et al., where they reported that mice infected with influenza A/H5N1 treated with MSCs had an increased survival rate compared to controls treated with control fibroblasts [33]. As such, there is no general conclusion regarding the therapeutic effect of MSCs on virus-induced acute lung injury. One potential explanation for this could be that MSCs are effective against specific viruses. For example, Tan et al. demonstrated that the majority of MSCs in their study expressed α-2,6-linked SA (influenza A/H1N1 virus binding receptors) and were highly susceptible to infection of the virus. Interestingly, it has been reported that human MSCs do not express the angiotensin-converting enzyme 2 (ACE2) receptor [34••], suggesting that MSCs might not be susceptible to infection of the COVID-19 virus.

What Is Known from Clinical Trials?

MSC-Based Clinical Trials in Patients with Classical ARDS

Since the two first phase I trials on MSC-based treatments for classical ARDS in 2014–2015 [10, 11], several other trials have been completed which all demonstrated that MSCs were well-tolerated in this patient group, but no significant improved lung function or other clinical relevant outcomes were consistently observed [12, 35, 36••, 37]. The data obtained in the clinical trials mentioned above, except the two latest publications, have been extensively summarized by us and others [38,39,40] and will therefore not be covered here. In the more recent study by Wick et al., the authors measured potential biomarkers in the airspace and in circulation in ARDS patients included in the START trial study 48 h after treatment with MSC or placebo. Here, they found that there was a decrease in the airspace proteins Ang-2, IL-6, and sTNFR1 in patients treated with MSC compared to patients in the placebo group. Interestingly, the levels of measured biomarkers in the circulation differed very much from those measured in the mini-bronchoalveolar lavage fluid samples. This is important information to consider when collecting samples for treatment evaluation and biological understanding, as biomarkers isolated from the plasma versus the airspace most likely reflect different biological processes [37]. The most recent completed clinical trial in classical ARDS was published in 2022; here, the authors used multipotent adult progenitor cells to treat classical ARDS (moderate-to-severe) in a multicenter, randomized, double-blind, dose-escalation, placebo-controlled phase 1/2 trial. The patients were given either 300 × 106 or 900 × 106 cells diluted in 300 ml PlasmaLyte-A or placebo through a 200-µm blood filter tubing set as a single peripheral or central venous infusion. Similar to the other completed trials, the cells were demonstrated to be well-tolerated, and no acute safety concerns were observed. There was one death that occurred in the cell-treated group; however, it was determined by the data and safety monitoring board (DSMB) to be unrelated to the cell therapy. At day 28 after treatment, there was an increased number of treatment-emergent adverse events in the group that received the cells compared to the placebo group (91.3% in patients receiving 900 × 106 cells vs. 60% for placebo); however, there was a lower mortality rate in the cell-treated group both at day 28 (25% in patients receiving 900 × 106 cells vs. 40% for placebo) and at day 365 (40% in patients receiving 900 × 106 cells vs. 50% for placebo) compared to placebo [36••].

MSC-Based Clinical Trials in Patients with COVID-19-Associated ARDS

After the COVID-19 outbreak, the enthusiasm for using MSCs as cell-based therapy was once again raised leading to a dramatic increase in clinical trials using MSCs as therapy for COVID-19-associated ARDS. Searching on the PubMed database for published clinical trials through October 2, 2023, using the keywords “COVID-19” and “mesenchymal stromal cells,” we identified 24 published studies (summarized in Table 1). In this section, we will briefly summarize and discuss the more recent papers and highlight some interesting lessons that can be learned from these trials as well as discuss differences and similarities with completed MSC-based trials on patients with classical ARDS.

In 2022, Kirkham et al. published a meta-analysis of controlled trials of MSC-based treatment for patients with COVID-19-associated ARDS. This was a systematic search of the literature conducted on studies published until November 15, 2021. Based on the results from the reviewed studies, the authors concluded that MSCs likely can reduce mortality in patients with critical or severe COVID-19 because they found evidence that MSC-based treatment reduced the relative and absolute risk of death at the study endpoint [41•]. All studies included in this meta-analysis were however very small and different investigational protocols were used [41•]. Earlier this year, 2023, Soetjahjo et al. published a double-blind, randomized, placebo-controlled, multicenter trial (NCT04333368) involving severe COVID-19 patients in which they gave three injections of umbilical cord-derived MSCs with 1 × 106 cells/kg body weight per time point. The study enrolled 42 patients who were randomly assigned into two equal groups and aimed to investigate the safety and effectiveness of MSC-based treatment. No decrease in the length of hospitalization was seen in the MSC-treated group compared to the control group. However, the MSC-treated group had a significant increase in oxygenation index and a smaller increase in procalcitonin values compared to the control group [42•]. In the trial by Zarrabi and colleagues (IRCT20200217046526N2), MSC treatment was combined with a dose of extracellular vesicles (EVs) derived from MSCs. In this randomized, multicentric, phase II clinical trial, 43 patients with severe COVID-19 were enrolled (MSC alone, n = 11; MSC combined with EVs, n = 8; control group, n = 24), and the study aimed to assess safety and efficacy of two doses of perinatal tissue-derived MSC or one dose of MSCs followed by a dose of MSC-derived EVs. The authors reported the treatments to be safe with minimal adverse events, and a decreased serum level of inflammatory markers was seen in all study groups; however, there was a more prominent change in the MSC alone and MSC combined with EVs compared to controls [43•]. Li et al. published their 2-year follow-up results from a randomized, double-blind, placebo-controlled trial (NCT04288102) [44], in which 100 patients with severe COVID-19 were included. The patients received either 3 MSC infusions (n = 65, 4 × 107 cells per infusion) or placebo (n = 35) on days 0, 3, and 6 in combination with standard of care. The authors observed that MSC administration was safe 2 years after treatment; however, the efficacy of MSC treatment reported at the 1-year follow-up [45] was not significantly sustained at the 2-year follow-up according to 6-min walking distance data, quality of life, and extent of lung damage. There were no significant differences in pulmonary fibrosis based on the CT images between the MSC group and the placebo group at 24-month follow-up [44]. Taken together, it is very difficult to draw any conclusions from these different studies on COVID-19-associated ARDS because since they were performed during the pandemic, they are small studies, the standard of care changed between different trials, and sometimes within one trial, different MSC sources, doses, and criteria were used. However, there are several clinical trials that report at least some beneficial effects, and similar to the MSC trials on classical ARDS, they demonstrate that MSC-based therapy is safe also for COVID-19-associated ARDS. As suggested by Kirkham et al., one option would be to develop a “master protocol” to ensure consistency of cell product production and manufacturing and dosing strategies to simplify the ability to compare results between different clinical trials [41•]. However, creating a “master protocol” for ensuring consistent product manufacturing among the clinical trials would entail several difficulties including for example intellectual patent rights associated with each pharmaceutical industry and differences in regional laws and regulations.

So far, only first-generation MSC products have been used in all MSC therapy for classical and COVID-19-associated ARDS; however, a large body of literature indicates that pre-activating MSCs with appropriate cues prior to infusion could enhance their therapeutic potency [46,47,48]. For example, IFN-γ pre-treated MSCs have been demonstrated to inhibit T-cell proliferation as well as inhibit T-cell production of IFN-γ, TNF-α, and IL-2 in vitro [46]. However, contradictory results have been published on the actual in vivo effect of IFN-γ pre-treated MSCs in experimental graft versus host disease models [48, 49]. In another study, Bustos et al. pre-treated MSCs with serum obtained from ARDS patients and found that pre-treated MSCs produced increased levels of anti-inflammatory cytokines such as IL-10 and IL-11RN and decreased levels of pro-inflammatory cytokines such as IL-6, IL-8, IFN-γ, and IL-1β [47]. An altered secretome profile has also been demonstrated by MSCs treated with bronchoalveolar lavage fluid samples obtained from ARDS patients [50]. An increasing number of publications suggest that the MSC therapeutic function depends on the microenvironment they encounter [47, 50,51,52,53]. Therefore, it is essential to understand how MSC function is altered after entering a COVID-19 infectious environment containing large concentrations of pro-inflammatory cytokines (cytokine storm) and neutralizing antibodies and B-cell responses. In a recently published study, it was demonstrated that MSCs inhibit B-cell differentiation and block pan-antibody secretion, findings that may have implications for B-cell-mediated anti-viral responses [54]. Another important factor is the increased levels of D-Dimer observed in COVID-19 patients [23••]. In a few case reports, elevation levels of D-dimer have been observed after MSC treatment and linked with serious side effects such as pulmonary embolisms and venous clots [55, 56]. However, the pooled analysis on the clinical trials using MSCs to treat COVID-19-associated ARDS found that MSC-based treatment was safe for patients with COVID-19-associated ARDS [19••]. However, further studies are warranted before we can understand the exact impact of the COVID-19-associated ARDS environment on infused MSCs.

Summary and Final Remarks

The completed clinical trials have all demonstrated that MSC-based treatment is safe to be used as treatment for patients with classical ARDS and COVID-19-associated ARDS, despite the different etiologies and differences in pathophysiology. Some of the clinical trials published during the last years and recent meta-analyses suggest that MSCs could potentially reduce mortality in patients with severe COVID-19-associated ARDS.

After several decades of progression in the field of MSC-based therapies for respiratory diseases with good pre-clinical outcomes and very stimulating results, we have now reached a plateau phase without a well-defined track forward. After several years with many completed clinical trials reporting no significant improved outcomes, it is easy to be critical and question if MSC-based therapies would be a likely future treatment option for patients with respiratory failure or severe acute lung disorders. However, we strongly believe that MSC-based therapy will be a future therapeutic option for at least subgroups of patients within specific inflammatory lung disorders such as ARDS and COVID-19-associated ARDS. But to advance to the next step, it is important to take a step back. We need to return to do some bench work and to repeat many of the in vitro and pre-clinical experiments with all the advanced techniques and instruments that are now available to us, because we believe that it is crucial that we understand (i) the MSC biology, (ii) the MSC–host environment interaction, (iii) the plasticity of in vivo MSCs, and (iv) which subgroups of patients that truly have a chance of benefit from this type of treatment before we can obtain significantly improved outcomes in future MSC-based clinical trials for acute inflammatory lung disorders.