LAUSR

Dear Editor, Acute respiratory distress syndrome (ARDS) is a highly heterogeneous and devastating condition, and there are no effective pharmacological treatments. A considerable proportion of patients initially diagnosed with ARDS experience rapid resolution of their symptoms in the following 24 h. Unless those patients are excluded from therapeutic trials, there is a risk of trial failure [1]. Confirmed ARDS (cARDS), for which prognostic and predictive factors are well known, should be further investigated in interventional trials [2]. Many preclinical studies have shown that mesenchymal stromal cells (MSCs) could be a promising treatment for inflammation modulation and tissue repair in ARDS. Previous safety studies reported that MSCs originating from different tissues were well tolerated in ARDS patients [3–5]. Nevertheless, the safety and benefits of MSC treatment for cARDS should be further investigated. Here, by collecting and analyzing the preliminary data from a prospective single-arm study (NCT03608592) conducted in a forty-bed multidisciplinary ICU of a tertiary comprehensive hospital, we aimed to investigate the salvage effect of umbilical cord-derived MSCs (UCMSCs) in patients with cARDS. Confirmed ARDS was defined as an initial diagnosis with moderate to severe ARDS according to the Berlin definition and sustained PaO2/FiO2 (P/F ratio) less than 200 after 24 h of conventional therapy. A single dose (1 million/kg) of fresh manufactured UCMSCs was intravenously administered. The primary outcome was infusion safety. Survival at 60 days, organ function, lung injury parameters and biomarkers were also observed (see online ESM—Study protocol). In this cohort, 50 patients met the moderate to severe ARDS criteria at baseline. Of those patients, 20 (40%) improved to mild ARDS or no ARDS in the first 24 h, and 4 were excluded due to the discontinuation of active treatment. Finally, 22 of 26 (85%) patients with cARDS were enrolled, treated with infusions of freshly cultured UCMSCs (viability > 90%) and completed the followup. These patients were severely ill, as indicated by the APACHE II score (22 ± 7) and nonpulmonary SOFA score (9 ± 4). Even with 24 h of protective ventilation [mean tidal volume 6.8 ml/kg, PEEP 11 cmH2O and median plateau pressure (Pplat) 28.5 cmH2O], the preinfusion P/F ratio was still low (101 ± 35), and the Murray lung injury score was still extremely high (3.1 ± 0.6) in these patients before infusion (see Table 1). All the subjects well-tolerated the cell infusion. Their vital signs were stable, and no infusion-associated event occurred within the first 24 post-infusion hours. Ten patients died within 60 days, with a mortality rate of 45% (10/22). Among the nonsurviving patients, 4 died *Correspondence: [email protected] 4 Surgical and Transplant Intensive Care Unit, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, Guangdong, People’s Republic of China Full author information is available at the end of the article