Cell Therapy Using Umbilical Cord-derived Mesenchymal Stromal Cells in SARS-CoV-2-related ARDS

General context: As of March 13, 2020, more than 145,000 cases of 2019-nCoV infection have been confirmed with 5,500 deaths worldwide. As of March 21, 14,469 cases have been confirmed in France, of which 562 have been fatal while 1,525 patients are currently hospitalized in intensive care units. Whereas the pandemic continues to spread, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes Severe Acute Respiratory Distress Syndrome in 30% of patients (Murthy et al., 2020) with a 30%-60% mortality rate. The main physio-pathological hallmark is an acute pulmonary inflammation. Currently, there is no treatment. The objective of this project is to treat intubated-ventilated patients presenting with a SARS-CoV2-related Acute Respiratory Distress Syndrome (ARDS) of less than 96 hours by three intravenous infusions of umbilical cord Wharton's jelly-derived mesenchymal stromal cells (UC-MSC) one every other day (duration of the treatment: one week). The primary endpoint is the PaO2/FiO2 variation from baseline at day 7. The evolution of several inflammatory markers, T regulatory lymphocytes and donor-specific antibodies will also be monitored. The trial will include 40 patients, of whom 20 will be randomized to cell-treatment administered via intravenous route while the remaining 20 patients will be randomized to receive a placebo solution in addition to the standard of care. Patients will be followed up to 12 months after treatment. State of the art: Mesenchymal stem cells feature several attractive characteristics: ease of procurement, high proliferation potential, capacity to home to inflammatory sites, anti-inflammatory, anti-fibrotic and immunomodulatory properties. Their therapeutic benefits have been demonstrated in \> 100 animal models, including sheep. Specifically, the therapeutic effects of MSC have been demonstrated in ARDS models induced by H1N1, H5N1, H9N2 influenza virus-associated pneumoniae and were also shown to reduce bacterial-induced acute lung injury in a human model of ex vivo perfused lung (Lee et al., 2013). In the clinics, MSC have demonstrated an excellent tolerance in over 3,000 patients (Thompson et al., 2020), regardless of the dosing and delivery route. Three phase I/II trials have included patients with an ARDS and in one (START II), MSC significantly reduced pulmonary endothelial injury (Matthay et al., 2019). If all MSC share several characteristics regardless of the tissue source, the highest productions of bioactive molecules and the strongest immunomodulatory properties are yielded by those from the Wharton's jelly of the umbilical cord (Romanov et al., 2019) which can be scaled-up to generate banks of cryofrozen and thus readily available products. So far, UC-MSC have been used in a wide variety of diseases (reviewed in Scarfe et al., 2018) and, in most cases, they have been delivered via the intravenous route which is clinically attractive because of its non invasive nature and the subsequent possibility of repeated administrations. Labeling techniques have shown that \>80% of intravenously injected MSCs are rapidly trapped in the lungs, followed by a rapid distribution of some of the injected MSCs to other tissues including liver, spleen, and inflammatory or injured sites (Brooks et al., 2018). Over all, these biodistribution patterns have been confirmed by human studies using magnetic resonance imaging (MRI), positron emission tomography (PET) and/or single-photon emission computed tomography (SPECT).2 A four-parameter model (injection rate, clearance rate, rate of extravasation and rate of intravasation) predicts that transplanted MSCs are only therapeutically active for a short period of time (probably less than 24 h) (Parekkadan and Milwid, 2010), a timescale consistent with that of the biological responses that they trigger. These assumptions are an incentive to repeated MSC administrations within a short period of time to induce a sustained therapeutic effect and has rationalized our protocol of injecting MSC one every other day over a one week-period, a design consistent with the earlier report of the benefits of delivering MSC at relatively small doses but in a repeated fashion in patients with graft-versus-host disease (Zhou et al., 2010). Once they have homed in the lungs, MSC have been reported to first induce an inflammatory response which is detectable at the tissue level and systemically (Hoogduijn et al., 2013) and is likely due to their interaction with resident lung cells once they have accumulated in the microvasculature. This initial response is then followed by a downstream phase of reduced immune reactivity (Hoogduijn et al., 2013), the mechanisms of which have been extensively investigated. Thus, 24 hours after their intravenous infusion, most of the UC-MSC that have accumulated in the lungs are dead after their phagocytosis by monocytes and neutrophils which then migrate through the blood stream, particularly in the liver (Leibacher et al., 2017; de Witte et al., 2018). Co-culture experiments have shown that the internalization of MSC fragments by monocytes triggers a phenotypic shift which translates into the upregulation of PD-L1 and CD90 along with an increased expression of mRNA levels for IL-1β, IL-6, IL-8 and IL-10 and a decreased expression of TNF-α. Of note, monocytes polarized towards an immune-regulatory phenotype increase the expression of Foxp3+ T regulatory lymphocytes while decreasing that of activated CD4+ cells (de Witte et al., 2018). That apoptosis of intravenously infused MSC is a requirement for their immunosuppressive function is further supported by the observation that the cytotoxic activity against MSC is predictive of clinical responses in patients treated by MSC for graft-versus-host disease, i.e., the best responders are those with high cytotoxicity; in this study, the postulated mechanistic link is that phagocytes that have engulfed apoptotic MSC then produce indoleamine 2,3-dioxygenase (IDO) and thus ultimately deliver MSC immunosuppressive activity (Galleu et al., 2017). Given the pathophysiology of SARS-CoV2, it is thus sound to hypothesize that the intravenous administration of UC-MSC during the initial phase of ARDS could control inflammation, accelerate its recovery with improved oxygenation, reduced mechanical ventilation and ventilation weaning time and therefore reduced length of stay in intensive care. This assumption is indeed supported by the recent results of a preliminary Chinese trial in which MSC (the source of which has not be specified) have been reported to improve pulmonary function and symptoms in 7 patients with COVID-19 pneumonia along with a rapid clearance of overactivated cytokine-secreting immune cells (CXCR3+CD4+ T cells, CXCR3+CD8+ T cells, and CXCR3+ NK cells), a decrease in TNF-α circulating levels and an increase in CD14+CD11c+CD11bmid regulatory DC cells and IL-10 levels, compared with a placebo group (Leng et al., 2020).