Evaluation of the safety and efficacy of using human menstrual blood‐derived mesenchymal stromal cells in treating severe and critically ill COVID‐19 patients: An exploratory clinical trial

Abstract The coronavirus disease 2019 (COVID‐19), caused by severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) was identified in December 2019 and has subsequently spread worldwide. Currently, there is no effective method to cure COVID‐19. Mesenchymal stromal cells (MSCs) may be able to effectively treat COVID‐19, especially for severe and critical patients. Menstrual blood‐derived MSCs have recently received much attention due to their superior proliferation ability and their lack of ethical problems. Forty‐four patients were enrolled from January to April 2020 in a multicenter, open‐label, nonrandomized, parallel‐controlled exploratory trial. Twenty‐six patients received allogeneic, menstrual blood‐derived MSC therapy, and concomitant medications (experimental group), and 18 patients received only concomitant medications (control group). The experimental group was treated with three infusions totaling 9 × 107 MSCs, one infusion every other day. Primary and secondary endpoints related to safety and efficacy were assessed at various time points during the 1‐month period following MSC infusion. Safety was measured using the frequency of treatment‐related adverse events (AEs). Patients in the MSC group showed significantly lower mortality (7.69% died in the experimental group vs 33.33% in the control group; P = .048). There was a significant improvement in dyspnea while undergoing MSC infusion on days 1, 3, and 5. Additionally, SpO2 was significantly improved following MSC infusion, and chest imaging results were improved in the experimental group in the first month after MSC infusion. The incidence of most AEs did not differ between the groups. MSC‐based therapy may serve as a promising alternative method for treating severe and critical COVID‐19.


INTRODUCTION
The coronavirus disease 2019 (COVID- 19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was initially identified in December 2019 as causing a cluster of respiratory infections. 1,2 COVID-19 quickly attracted global concern and panic since it is highly contagious. [3][4][5] Due to a lack of adequate awareness in the first few weeks of the outbreak, the number of infected patients increased swiftly, rapidly spreading to more and more countries. 6,7 As of January 7, 2021, there have been over 85 929 000 confirmed cases of COVID-19 worldwide, leading to 1 876 100 deaths (https://www.who.int/emer gencies/diseases/novel-coronavirus-2019). Currently, the number of infected patients is still increasing worldwide. COVID-19 has an incubation period which can range from 1 to 14 days but usually ranges from 3 to 7 days. 8 The main symptoms are fever, headache, dry cough, and chest tightness. [9][10][11] Many patients also experience a sore throat, diarrhea, nasal congestion, and rhinorrhea. 12,13 Severe patients often develop expiratory hyperextension and dyspnea 1 week after the onset of the disease. In the most severe cases, patients can quickly develop acute respiratory distress syndrome (ARDS), severe acute lung injury, septic shock, metabolic acidosis, and coagulopathy, as reported in a biopsy and autopsy study. 14 COVID-19 can easily cause expiratory dyspnea and ARDS. Of the COVID-19 patients, 13.8% of the cases were severe, 6.1% cases were critical, and about 2.3% cases had fatal outcomes. 15,16 Since no effective or authorized vaccines are available for preventing COVID-19 infections, a breakthrough in the therapeutic strategy is vital for the treatment of COVID-19 and especially for severe or critically ill patients who may develop ARDS and/or expiratory dyspnea. [17][18][19][20][21] Currently, a few drugs (such as remdesivir and dexamethasone) have shown positive preliminary results in randomized, controlled, open-label, clinical trials. 22,23 Even more excitingly, COVID-19 vaccines with acceptable safety, tolerability, and immunogenicity have been reported by Zhu et al 24 and Folegatti et al 25 as being effective in initial human clinical trials. Apart from these, many other groups are also developing available vaccines such as the BNT162 mRNA vaccine 26 sponsored by Pfizer Inc. and BioNTech SE; the chimpanzee adenovirus vectored vaccine ChA-dOx1 nCoV-19 (AZD1222) 27,28 sponsored by AstraZeneca; the mRNA-1273 vaccine 29 codeveloped by the Cambridge, Massachusetts-based biotechnology company Moderna, Inc., and the National Institute of Allergy and Infectious Diseases (NIAID); the adenovirus serotype 26 (Ad26) vaccine 30 developed by Janssen Vaccines & Prevention B.V.; the BBIBP-CorV vaccine 31 developed by the Beijing Institute of Biological Products; the CoronaVac vaccine 32 developed by Sinovac Life Sciences; and the NVX-CoV2373 vaccine 33 sponsored by Novavax, Inc. However, no specific drugs have been reported to be absolutely effective in treating COVID-19. Moreover, SARS-CoV-2-induced secondary infections have been reported to induce multiple organ dysfunction syndrome in severe or critically ill patients, and this issue remains a serious problem worldwide.
Mesenchymal stem cells have been used for almost three decades, and have made great progress. 34 According to the recommendations of the International Society for Cell & Gene Therapy (ISCT) in 2019, mesenchymal stem cells should be named mesenchymal stromal cells (MSCs). 35 MSC-based cellular therapy has been the subject of an increasing number of studies due to the cells' self-renewing capacity, multipotent potential, low immunogenicity, anti-inflammatory activity, and ability to home to damaged tissue. [36][37][38] More importantly, MSCs have unique immunomodulatory functions of both innate and adaptive immune responses, making them an attractive cell therapy tool. 38,39 MSCs regulate adaptive immune responses mainly through targeting T lymphocytes, B lymphocytes, antigen-presenting cells (APCs), dendritic cells (DCs), natural killer (NK) cells, and regulatory T cells (Tregs). 38 MSCs also regulate innate immune responses mainly through targeting DCs, NK cells, innate TH17 cells, neutrophils, monocytes, macrophages, and mast cells. 39 Further, MSC-based therapies have shown promising results in several clinical studies across a variety of diseases. 40,41 With the development of stem cell research, researchers have found that after injecting MSCs, the human body activates the host's innate immune cascade system, such as complement and blood coagulation, which is defined as the instant blood-mediated inflammatory response (IBMIR). 42 IBMIR is of key importance considering the highly procoagulant state of many critically and severely ill patients in need of MSC therapy. 43 MSCs can be obtained from various sources, including bone marrow (BM), adipose (AD), umbilical cord (UC), placenta, menstrual blood, muscle, dental pulp, Wharton's jelly (WJ), fetal liver, amniotic membrane, amniotic fluid, urine etc. [44][45][46][47][48] Furthermore, MSC-based treatment has demonstrated promising results in studies on inflammatory lung disease, showing an ability to inhibit alveolar collapse, cell apoptosis, and collagen accumulation in lung tissues. 49 The angiotensin-converting enzyme 2 (ACE2) is identified as a receptor of SARS-CoV-2 entering into target cells. 50,51 In addition, researchers have demonstrated that MSCs do not express ACE2, and MSCs are resistant to SARS-CoV-2 infection as well as when exposed to SARS-CoV-2-infected cells. 52,53 Additionally, following infusion of allogeneic MSCs into nine patients with ARDS, Wilson et al 54 observed no prespecified side effects including cardiac arrhythmia, hypoxemia, and ventricular tachycardia. Further, our team reported that MSC transplantation significantly lowers the mortality of epidemic Influenza A (H7N9)-induced ARDS patients. 55 BM-MSCs have been shown to improve repair after ventilator-induced lung injury, to facilitate the resolution of inflammation, and to restore lung function and structure in ARDS patients. 56 With regard to the COVID-19 epidemic, MSCs from different sources (especially UC-MSCs) have been used in clinical trials. 53,57 A good proliferation rate plays an important role in clinical application, because stem cell-based treatment is dose-dependent, and usually human clinical research requires millions of stem cells. The doubling time for menstrual blood-derived MSCs is about 20 h, and the doubling time for BM-MSCs is about 40-45 h. Thus, MSCs from menstrual blood can obtain a better yield within a shorter time at early passages. 58,59 More importantly, menstrual blood-derived MSCs offer an alternative that is both painless and free of the ethical issues that may arise from BM-MSCs donations. 60 Thus, menstrual bloodderived MSC-based therapy may be a promising treatment for COVID-19, particularly to combat the inflammatory cytokine storms observed in severe and critical patients.
This study is an exploratory trial to assess the ability of menstrual blood-derived MSCs to treat severe and critically ill COVID-19 patients. To this end, we assessed the safety, therapeutic efficacy, and tolerability of transplanted MSCs with a 1 month follow-up after SARS-CoV-2 infection. In particular, we assessed any improvements in pulmonary function. Our results not only shed light on the ability of MSCs to treat COVID-19 patients, but also suggest that MSCs are a promising tool to treat acute or chronic pneumonia in future clinical applications.  53,62,63 Only severe and critically ill COVID-19 patients were included in the present study. Severe patients were defined as those with respiratory distress, respiratory rate ≥30 breaths/min; resting oxygen saturation ≤93%; or arterial blood partial pressure of oxygen (PaO 2 )/fraction of inspiration O 2 (FiO 2 ) ≤300 mmHg (1 mmHg = 0.133 kPa). Critical COVID-19 patients were defined as those with respiratory failure who required mechanical ventilation, those who had experienced shock, or those for whom a combination of organ failures necessitated monitoring and treatment in the intensive care unit (ICU). Exclusion criteria for this trial were as follows: (1) severe liver disease; (2) long-term hemodialysis and severe renal impairment or continuous renal replacement therapy; (3) comorbidities that might affect the ability of researchers to determine drug efficacy (mainly malignant tumors, active tuberculosis, interstitial pneumonia, and pulmonary heart disease); (4) treatment with glucocorticoid medications or other immunosuppressive drugs for longer than 2 weeks; (5) history of major surgery within 30 days of screening or presence of an unhealed surgical wound; (6) allergy to any active/inactive ingredients in the study drug; (7) pregnant or breastfeeding; (8) previous history of prothrombotic events (venous thromboembolism/stroke); (9) other circumstances judged by an investigator to preclude participation. These "other circumstances" leading to exclusion from the study included serious AEs for which the investigator judged that the risk of continuing to participate in the trial was too great, as well as the use of other treatments, without authorization and against medical advice, that could have affected the evaluation. These exclusion criteria followed the National Health Commission of China (Trial Version 5). If a patient met all the inclusion or exclusion criteria, he or she was then enrolled in the experimental group (MSC infusion + concomitant medications) or the control group (concomitant medications).

Stratification of disease treatment and concomitant medications
Patients were enrolled and admitted between January and April 2020 at either Renmin Hospital of Wuhan Univer-sity or the First Affiliated Hospital, College of Medicine, Zhejiang University. A clinical plan was designed for each patient based on their clinical needs. Since the patients in this clinical study were either severe or critical, concomitant medications were allowed, and the details of all additional treatments were recorded. Drugs that were required to treat other diseases were allowed if an investigator judged that the safety and efficacy of the study medications would not be affected. However, the type and dose of the medications were kept as consistent as possible while prioritizing patient safety. For any concomitant medications or other treatments, detailed medication information was recorded on the original medical record and the eCRF.
Of the 44 patients enrolled in this study, 36 were from Wuhan University People's Hospital, including 20 severe patients and 16 critically ill patients. The remaining eight patients were enrolled from the First Affiliated Hospital, College of Medicine, Zhejiang University including six severe patients and two critically ill patients. Due to the high risk of mortality, patients were given the choice as to which group they were placed in, and 26 patients chose to receive the experimental treatment along with the comprehensive treatment, while 18 patients chose to be included in the control group and receive only the comprehensive treatment. The standard of care was consistent between the two hospitals.

MSC preparation, cell transplantation, and subsequent observation
Allogeneic, menstrual blood-derived MSCs (no. SC0100919001, no. SC0100919004, and no. SC0100919005 provided by Innovative Precision Medicine (IPM) Group, Hangzhou, China) were obtained from three healthy female donors (age range, 20-45 years), and the volunteers were educated and provided signed, informed consent before donation, as described in previous studies. 55,64 The donation protocol was authorized by the Ethics Committee of Zhejiang University. The mononuclear cells within the menstrual blood were collected and purified, and cell viability was measured prior to seeding for cell culture according to the staining with trypan blue solution (Thermo Fisher Scientific, No. 15250061). MSCs were passaged at 70-80% confluence. Surface-labeled molecules (including CD29, CD34, CD45, CD73, CD90, CD105, CD117, and HLA-DR) were measured using flow cytometry (CytoFLEX LX Flow Cytometer, Beckman) and detailed procedures were described in a previous study. 65 Supporting information Table S1 includes detailed information on the antibodies used for surface marker analysis. Detection of the differentiation potential of MSCs into the osteogenic differentiation medium, chondrogenic differentiation medium, and adipogenic differentiation medium A and B (Cyagen Biosciences) and their detailed information has been reported in a previous study. 66 Table S2 includes the primer sequences for PCR analysis. The resulting cryopreserved MSCs were shipped frozen to the hospitals in a validated liquid nitrogen (≤−135 • C) dry shipper. Before their use, MSCs were resuspended in Plasma-Lyte 148 at room temperature by a local laboratory with a specialized cellular therapy center, and the control group was also administered the same volume of Plasma-Lyte 148. The viability of MSCs for the three donors should be >90%, which was a criterion for use in the clinical study guided by the Innovative Precision Medicine (IPM) Group. MSCs were used for treatment at the fifth passage, as described in our previous report. 66 Twenty-six patients received MSC transplantation, and 18 patients received all treatments except MSC transplantation. Complete case enrollment details and disease severity frequencies of the 44 COVID-19 patients included in the present study can be viewed in Table 1. Each COVID-19 patient in the experimental group used the MSC sample from one donor for all three treatment injections. Doctors observed hemodynamic and respiratory parameters at the bedside for at least 1 h to ensure that each patient was stable before MSC transplantation. Then, the MSC infusion was initiated using a standard blood filter tube. A researcher remained at each patient's bedside to continuously monitor the patient for any adverse reactions during the 24 h following treatment. Based on data from studies on the use of MSCs to treat H7N9-induced ARDS, 55 MSCs were administered as three infusions totaling 9 × 10 7 MSCs every other day (day 1, day 3, and day 5). Each infusion contained 3 × 10 7 cells resuspended in 500 mL saline solution and was performed at a speed of 30-40 drops/min for about 15 min, followed by a speed of 100-120 drops/min for 2 h to retain MSC vitality.

Biological measurements and clinical evaluation indices
Laboratory measurements of blood test results, liver function markers, and inflammatory indicators were carried out at the Medical Inspection Center of the First Affiliated Hospital, College of Medicine, Zhejiang University and Renmin Hospital of Wuhan University. Factors which were investigated for having an association with therapeutic features or outcomes were as following: (1) baseline characteristics including age, underlying conditions, and clinical symptoms; (2) laboratory data and chest computed tomography (CT) data; and (3)  The objective of the current study was to evaluate the safety and efficacy of MSC transplantation as a treatment for COVID-19. The primary endpoint of the analysis was survival rate from January to April 2020, and this was based both on the survival of a full analysis set (FAS) and a per protocol set (PPS). The FAS, which was also the effectiveness analysis set, consisted of all 44 subjects who were initially enrolled in this study, which included 26 patients in the experimental group and 18 patients in the control group. Since two cases were initially included in the study who violated the exclusion criteria, the remaining 42 subjects were included in the PPS, which included 24 patients in the experimental group and 18 patients in the control group. The secondary endpoints for this study included measures of effectiveness and tolerability, primarily including negative viral test results, time taken to recover from all symptoms, change in chest CT results, change in indicators of inflammation, change in oxygenation index, occurrence of shock, incidence of multiple organ failure, length of hospital stay, number of days in the ICU, and respiratory support status. Before and after the MSC infusion, patients were tested using several laboratory indices, including those related to hematuria routine, liver and kidney function, coagulation function, vital signs, physical examination, oxygenation (FiO 2 , peripheral oxygen saturation [SpO 2 ], oxygen saturation [SaO 2 ], and PaO 2 ), and inflammatory factors (including interleukin [IL]-6 and C-reactive protein [CRP]). Figure 1 shows the CONSORT diagram ( Figure 1A) and detailed infusions within 1 month ( Figure 1B) for this clinical study. Safety was measured by the frequency of treatment-related adverse events (AEs) and through careful surveillance of laboratory indices. Clinical data were obtained on each day of MSC infusion (days 1, 3, and 5) as well as on days 7, 14, and 30 of the posttreatment period.

Statistical analysis
To compare the experimental and control groups, χš or Fisher's exact tests were used, as appropriate, for both the FAS analysis and the PPS analysis. The FAS analysis included 44 patients, while the PPS analysis included a subset of 42 patients. The Kaplan-Meier method was utilized to analyze the survival time of discharged patients, and two-sided 95% exact confidence intervals (CIs) were calculated using the log-rank test. We also calculated a Cox proportional hazards model to assess factors affecting survival, adjusting for gender, and age as covariates.
For the length of hospitalization and ICU stay, a Wilcoxon rank sum test was used to compare the differences between groups. For analysis of inflammatory indices, (including CRP and IL-6) and oxygenation indices (FiO 2 , SpO 2 , SaO 2 , and PaO 2 ), Wilcoxon rank sum tests were used comparing the experimental group before and after the MSC infusion. Statistical analysis was carried out using SAS9.4. P value < .05 was considered as statistically significant.

Patient characteristics and MSC treatment
The viability of MSCs for the three donors was >90% (91% for SC0100919001 and SC0100919004, and 92% for SC0100919005). MSCs strongly expressed CD29, CD73, CD90, and CD105; and MSCs were negative for CD34, CD45, CD117, and HLA-DR (Supporting information Figure S1). Furthermore, MSCs can be successfully induced into osteogenic, adipogenic, and chondrogenic cells through the specific medium, and the representative picture for each kind of differentiated cells is shown in Supporting information Figure S2. A representative electrophoretogram with two pairs of primers is shown in Supporting information Figure S3. MSC ACE2 expression was negative according to PCR analysis, with ddH 2 O as a negative control and 293T cells as a positive control.
Twenty-six patients were included in the experimental group and treated with MSC transplantation and combination therapy. Among these, 16 (61.5%) were classified as severe and 10 (38.5%) were classified as critical. Eighteen patients were included in the control group and received only combination therapy. Among these, 10 (55.6%) were classified as severe and 8 (44.4%) were classified as critical ( Table 2). The experimental group contained 17 males (65.38%) and nine females (34.62%), while the control group contained 13 males (72.22%) and five females (27.78%). The mean and median age of the experimental group was 58.31 ± 12.49 and 57.50 years, respectively, while the mean and median age of the control group was 61.11 ± 11.03 and 64.00 years, respectively. There were no statistically significant differences with regards to gender or age (P > .05). There were no significant differences (P > .05) in clinical symptoms (including fever, expiratory dyspnea, sore throat, diarrhea, and chest tightness) between the experimental group and the control group during the baseline period ( Table 2). The symptom of cough showed a significant improvement following MSC infusion on day 1 (P = .037) compared to that of the control group, but no differences were found at other times points. During the period of MSC infusion (days 1, 3, and 5) and the post-treatment period (days 7, 14, and 30), the symptoms of fever, cough, sore throat, diarrhea, and   Note. *Represents significant difference in experimental group and control group (P < .05); "n" represents number; "N/A" represents not applicable; "#" represents Fisher's exact test to obtain the P value, and others use χš test without label. Note. *Represents significant difference (P < .05); P a value is the experimental group of the difference between severe and critically ill patients in each type of combination medication in the experimental group; P b is the difference test between severe and critically ill patients in each type of combination medication in the control group; P c is the test group in each type of combination medication test the difference with the control group (do not distinguish the severity of the disease); "n" represents number; "N/A" represents not applicable. chest tightness were not significantly different (Table 2). There was a significant improvement in expiratory dyspnea while undergoing MSC infusion on day 1 (P = .016), day 3 (P = .040), and day 5 (P = .031) compared to the control group (Table 2), but there were no significant differences on day 7 (P = .631), day 14 (P = .635), or day 30 (P = 1.000).

Analysis of the use of concomitant medications
As per the principles of treatment for severe and critically ill COVID-19 patients, combination drugs were allowed to treat the patients as effectively as possible. Concomitant medications were mainly used for symptomatic treatment, antiviral treatment, antibacterial treatment, hormone therapy, intestinal microbial state regulation, extracorporeal blood purification, traditional Chinese medicinal treatment, and basic disease treatment, among others. As shown in Table 3, there were no statistical dif-ferences in the types of combined medications between the two groups (P > .05). However, there were intragroup differences with regard to concomitant treatment using extracorporeal blood purification both in the experimental group (P < .001) and in the control group (P = .015). Specifically, concomitant treatment with an extracorporeal blood purification system was employed more often for critically ill patients than for severe patients in both the experimental and the control groups.

Assessment of the efficacy of MSC infusion after 1-month follow-up
The primary endpoint for this study was the survival rate of severe and critical COVID-19 patients with or without MSC infusion. Treatment efficacy was assessed throughout the study period both in the FAS and the PPS. As shown in Table 4, the survival rate for the experimental group was 92.31% (24/26) for the FAS, while the survival rate for the F I G U R E 2 Kaplan-Meier Curve of survival probability for the FAS (A) and the PPS (B) analysis. The experimental group showed better survival than the control group in both analysis sets, suggesting that MSC transplantation improved the survival rate of COVID-19 patients control group was 66.67% (12/18). This difference in survival rate was statistically significant (P = .048). Similar results were obtained when the PPS was analyzed separately. The survival curves for the FAS and the PPS are presented in Figure 2A, B, respectively. Our results sug-gest that MSC infusion exerts a positive therapeutic effect on the survival rate of severe and critically ill COVID-19 patients.
For severe patients, a Cox proportional hazards model of the FAS showed that the hazards ratio (HR) for the TA B L E 5 Secondary endpoints used as efficacy indicators for the 44 COVID-19 patients in the experimental group and the control group experimental group relative to the control group was .10 (95% CI, 0.00-1284.41), which was not statistically significant (P = .437; Supporting information Table S3). Using gender and age as covariates, the adjusted HR (95% CI) was .00 (0.00-∞), which was also not statistically significant (P = .970). For critically ill patients, the HR (95% CI) for the experimental group relative to the control group was .34 (.06-1.88; P = .218), and the adjusted HR (95% CI) with gender and age as covariates was .11 (.01-.89), which was statistically significant (P = .039). Similar results were obtained when the PPS was analyzed (Supporting information Table S4): the adjusted HR (95% CI) with gender and age as covariates was .12 (0.00-.95), which was statistically significant (P = .047). Thus, our results suggest that MSC transplantation increases survival more for critically ill patients than for severe patients when adjusting for gender and age. The secondary endpoints for this study mainly included negative viral test results, time taken to recover from all symptoms, length of hospital stay, number of days in the ICU, occurrence of shock, incidence of multiple organ failure, change in chest CT results, and respiratory support status. Detailed results are presented in Table 5. The average time taken to recover from viral infection, as indicated by a negative viral test, was 15.75 ± 13.71 and 18.31 ± 9.86 days for the experimental and the control groups, respectively. Statistical analysis showed that these differences were not significant (P = .251). The average time to improvement for the experimental group and the control group was 3.00 ± 3.05 and 8.80 ± 10.77 days, respectively, meaning that the average time taken to improve for the experimental group was 5.8 days shorter than that for the control group, and this difference was statistically significant (P = .049), indicating that MSC infusion was able to shorten the time required for treatment. There was no significant difference in either the length of hospital stay or in the number of days in the ICU (both P > .05), and there was also no significant difference in the occurrence of shock or multiple organ failure between the groups (both P > .05). Further, there was no statistical difference in the use of respiratory support between the two groups ( Table 6). One month after MSC infusion, 20 patients in the experimental group and 12 patients in the control group underwent chest CT by three respiratory physicians who were blinded to the treatment group; 17 (85.00%) patients in the experimental group had improved, while 3 (15.00%) patients showed no significant change. In contrast to the experimental group, six (50.00%) patients in the control group had improved, while the other six (50.00%) showed no significant change. Figure 3 shows representative CT scans documenting lung improvement at various time points for both the control and the experimental group. The difference in the improvement of chest imaging results in the first month after MSC infusion was significant. Representative CT images of both groups at post-treatment days 7, 14, and 30 are shown in Figure 3. Together, these results suggest that the relative improvement rate was higher for the experimental group during the 1 month after MSC infusion than it was for the control group.

Presence of AEs following treatment
The severity of each AE related to COVID-19 infection was graded from 1 to 5 for both the experimental group and the control group, and a summary of all AEs experienced by the 44 COVID-19 patients included in the current study are presented in Supporting information Table  S5. According to the statistical analysis, 76.92% (20/26) of the patients experienced a total of 56 AEs in the experimental group, including 40 AEs of grade 1, 6 AEs of grade 2, 3 AEs of grade 3, 4 AEs of grade 4, and 3 AEs of grade 5. In contrast, 100.00% (18/18) of the patients experienced a total of 59 AEs in the control group, including 39 AEs of grade 1, 5 AEs of grade 2, 3 AEs of grade 3, 7 AEs of grade 4, and 5 AEs of grade 5 (Supporting information  Table S5). During the study period, no patients in either the experimental group or the control group showed AEs that necessitated withdrawal from the study. A detailed analysis of the AEs observed in the experimental group and the control group is presented in Table 8. As shown in Table 8, 10 of the 56 AEs observed in the experimental group were grade 3 or higher, while 15 of the 59 AEs observed in the control group were grade 3 or higher. Except for a difference in the incidence of high blood pressure between the two groups, there were no significant differences (P > .05) between the groups for any AEs related to the measured clinical indices including the blood test results, liver function markers, blood lipid levels, renal function, myocardial enzymes, electrolyte disturbances, and clinical symptoms. High blood pressure was observed in two (3.57%) patients in the experimental group and six (10.18%) patients in the control group (P = .048). In summary, the frequency of each AE was statistically similar between the two groups, except for the AE related to high blood pressure, which was more common in the control group. Further, the experimental group showed a lower incidence of AEs (76.92%) than the experimental group (100.00%), but the difference was not statistically significant. Together, these results suggest that the MSC infusion protocol used in this study showed good safety outcomes.

DISCUSSION
The initial symptoms of COVID-19 are often fever, cough, sputum, and shortness of breath. These in turn can lead to dyspnea, ARDS, lung injury, shock, and eventual multiple organ failure. 68 Note. *Represents significant difference (P < .05) by the Wilcoxon rank sum test. "N" represents the number of valid observation data. CRP, C-reactive protein; FiO2, fraction of inspiration O2; IL-6, interleukin 6; Min, minimum; Max, maximum; PaO2, partial pressure of oxygen; SaO2, oxygen saturation; .SpO2, peripheral oxygen saturation.
that a deceased COVID-19 patient showed a large amount of sputum, presumably causing severe ARDS, as well as a large number of inflammatory factors in the lung tissue. 13 This suggests that effective treatment of ARDS and prevention of multiple organ failure is a key strategy in preventing mortality in COVID-19 patients. Severe ARDS causes breathing difficulties, and it has been suggested that resolving breathing difficulties in COVID-19 patients in a timely manner may make it is possible to inhibit COVID-19 progression. Interestingly, we found that patients treated with MSCs experienced immediate and dramatic relief from breathing difficulties associated with COVID-19 on day 1 (P = .016), day 3 (P = .040), and day 5 (P = .031) compared with the control group. Further, our results show that both SpO 2 and PaO 2 , indices associated with oxygenation levels, were significantly improved after MSC infusion. These results further support the use of MSC infusion as a method to combat ARDS and expiratory dyspnea, particularly for critically ill COVID-19 patients.
Multiple complications of the COVID-19 epidemic make it difficult to fully treat patients with this disease. Many interventions in China and around the world have proven effective to reduce the epidemic and prevent the virus from continuing to spread. 5,[70][71][72][73][74] Although effective social distancing can mitigate the virus's persistence, treating COVID-19 is also an important strategy toward ending the current pandemic.
X-ray and chest CT imaging results of COVID-19 patients in the ICU have been shown to reveal the presence of pneumonia, known as novel coronavirus pneumonia. These imaging results reveal bilateral, multilobular involvement as well as subsegmental consolidation. 75 In the present study, we observed a statistically significant difference in the rate of improvement of chest CT results in the first month after MSC infusion. A total of 85.00% of patients with MSC treatment showed improved chest CT results, compared with only 50.00% of patients in the control group. These results provide further evidence for the efficacy of MSC infusion.
Currently, an effective vaccine would be the best way to combat SARS-CoV-2 infection, and many groups have presented preliminary basic and clinical data related to vaccine development. 24,25,[76][77][78] However, assessing the safety and efficacy of any vaccine will take a relatively long period of time. Apart from targeted vaccine development, other therapeutic strategies are being developed in the race against time to end the global pandemic. Remdesivir, an inhibitor of the viral RNA-dependent, targets nascent viral RNA chains, resulting in premature termination of the viral life cycle. 23 can significantly shorten the recovery time of COVID-19 patients with lower respiratory tract infections in a doubleblind, randomized controlled clinical study. 80 Corticosteroids are important immunomodulators for the clinical treatment of SARS. 81 Studies have shown that compared with patients with mild to moderate disease, patients with severe SARS-CoV-2 have higher levels of proinflammatory cytokines in serum samples. 82 A meta-analysis indicated that the mortality rate was reduced in severe COVID-19 patients treated with corticosteroids. 83 More recently, the recovery collaborative group reported that dexamethasone (a type of corticosteroid) significantly decreased 28day mortality in patients hospitalized with COVID-19. 22 Corticosteroids have therefore become a potential treatment for COVID-19 patients. Compared with remdesivir or corticosteroids, MSC infusion has the potential to significantly reduce dyspnea in a relatively short period of time, as MSCs act by targeting secretory factors and MSCreleased extracellular vesicles deliver microRNA, mRNA, or DNA. 84,85 The main methods currently employed to treat COVID-19 patients (especially severe and critically ill patients) include the following: (1) convalescent plasma therapy; (2) antiviral drug therapy; (3) traditional Chinese and western medicine; (4) MSC-based therapy; and (5) immune-mediated therapy. [86][87][88][89][90][91] Studies on these methods have accelerated the screening of effective drugs, explored new treatment methods, and attempted to prevent mild cases from progressing in severity. The current clinical study reports that MSC infusion enhanced the survival rate for severe or critically ill COVID-19 patients in both the FAS (92.31% survival in the experimental group vs 66.67% in the control group; P = .048) and the PPS (95.83% survival in the experimental group vs. 66.67% in the control group; P = .031). Our results agree with a previous report detailing the potential for MSC infusion to treat critically ill COVID-19 patients by alleviating acute respiratory dysfunction and pulmonary fibrosis. 91 There are some other sources of MSCs used in clinical studies for treating COVID-19, and these are included in a systematic review and metaanalysis. [92][93][94] Leng et al 53 used UC-MSCs to treat seven COVID-19 patients, and had three control patients who were infected with COVID-19; there was 100% survival in the experimental group versus 66.67% survival in the control group. Shu et al 95 investigated 12 severe COVID-19 patients using UC-MSCs as the experimental group, and 29 severe COVID-19 patients using a placebo as the control group. Their results showed a 100% 28-day survival rate in the experimental group versus 89.66% survival in the control group. More recently, Meng et al 57 performed a study in 18 hospitalized patients with moderate to severe COVID-19 pulmonary disease, 9 of whom were treated with UC-MSC infusions and with 9 control patients. All patients sur-vived both in the experimental group and in the control group, but the degree of severity of COVID-19 (moderate to severe) might have affected the outcome. Leng et al, 53 Shu et al, 95 and our study enrolled more severe or critically ill COVID-19 patients. Recently, Sánchez-Guijo et al 96 investigated 13 severe COVID-19 patients using an AD-MSCs infusion for a clinical study that was nonrandomized and without a control; the results showed 84.62% survival using an AD-MSCs infusion. Therefore, MSC transplantation from different sources appears to be a candidate method to improve outcomes for critical cases. Coagulopathy and thromboprophylaxis are very common after MSC infusion. 15 Current clinical data indicate that MSCs from human menstrual blood does not clot in patients, which is very favorable for intravenous infusions. One possible reason could be that MSCs express a low level of procoagulant tissue factor TF/CD142, which needs to be systematically investigated and verified in future preclinical studies. Although different sources of MSCs have been investigated for effectiveness in treating COVID-19, more optimized treatment strategies are critical to evaluate and control blood compatibility, optimize cell transfusion, and monitor the real-time dynamics of cells in the body to develop safer and more effective MSC treatments. 43 Several concomitant treatments have been shown to exert a synergistic role with MSC transplantation. Of note, Peng et al 97 reported that intravenous infusion of convalescent plasma as a treatment for severe COVID-19 may have synergistic characteristics with MSC transplantation in inhibiting cytokine storms, promoting lung injury repair, and recovery of pulmonary function. Although further study is required to establish safety and efficacy, the current body of evidence suggests that MSC transplantation might be an effective treatment for severe and critical COVID-19.
In the current study, it was observed that many clinical symptoms were ameliorated in the 1-month period following MSC transplantation with combined therapy. No cases of pulmonary embolism were observed in the patients who underwent MSC infusion, although this side-effect is considered to be the main concern regarding MSC safety. Rather, our results indicate that MSC therapy is a safe and effective therapeutic strategy to rescue severe and critical lung problems induced by SARS-CoV-2. Thus, in the present study, preliminary clinical data were provided with regard to the short-term safety (1-month follow-up) and therapeutic effect of MSC transplantation to treat severe and critically ill patients.
The current study does not provide long-term evidence related to MSC-induced AEs. In a previous report from our group, MSCs were used to treat 17 H7N9 patients and four of those patients were followed up for 5 years without observing any AEs. 55 In the current study of severe and critical COVID-19 patients, almost no significant differences in the occurrence of AEs in the short term were observed between the control group and the group receiving MSC transplantation. Hence, this study found that MSC infusion is associated with good safety outcomes. In addition, although the potential application value of MSC treatment in COVID-19 is obvious, the innate and adaptive immune compatibility test of MSC is incorporated into the current cell detection system, and the establishment of strict monitoring standards for biosafety and effectiveness with regard to the coagulopathy and thromboprophylaxis is crucial. 15 Researchers should also pay strict attention to obtaining a suitable source, the product quality, monitoring of the various physiological and biochemical indicators of patients throughout the process, and strict prevention of unnecessary safety hazards. 98 Recently, Leng et al 53 published a clinical study on using MSCs to treat COVID-19 patients. This study investigated inflammatory and immune functioning as well as adverse effects in seven patients for 14 days post-MSC transplantation. The authors reported that MSCs appeared to significantly improve the functional outcomes of all seven patients without any observed AEs. However, more clinical data are still needed to identify any short-term adverse reactions following MSC administration. Zheng et al 99 recently reported that 12 patients with moderate to severe ARDS did not experience any infusion-related reactions or serious treatment-related AEs following MSC transplantation. Additionally, Meng et al 57 performed a study that included 18 hospitalized patients with moderate to severe COVID-19 pulmonary disease, nine of whom were treated with UC-MSC infusions. According to their results, intravenous UC-MSC infusion was safe and well-tolerated throughout the 1-month follow-up period. Although longterm follow-up data regarding the tolerability and safety of MSC transplantation are lacking, MSC transplantation may still be an effective method treatment for COVID-19, especially for severe and critical cases.
There are several limitations to this exploratory trial. First and foremost, one patient experienced an AE over grade 3 associated with severe abnormal liver function. More data from a larger study are needed to determine whether MSC infusion can lead to an infusion reaction in certain patients, especially as infusion reactions can cause shortness of breath. Since only 26 patients were treated with MSCs in the current study and because we used a multicenter, open, and parallel-controlled study design, our study may not have detected important AEs associated with MSC treatment. Secondly, we should stress that this clinical trial did not use a standard design owing to the unique nature of the COVID-19 outbreak and the ethical limitations associated with treating severe COVID-19 patients. Thirdly, we reported that the experimental group showed greater improvements in CT results than the control group during the 1-month study period. However, we only observed significant improvements in expiratory dyspnea for the experimental group versus the control group on days 1, 3, and 5, and no differences were observed between the groups at other time points with regard to lung function. We speculate that MSCs exert a short-term effect to improve expiratory dyspnea, while the long-term improvements were more due to the actions of concomitant medications. Moreover, we did not observe significant changes in the levels of inflammatory factors following MSC infusion, even though mortality was significantly decreased in the experimental group. Therefore, the mechanism by which MSCs reduce mortality should be investigated in future more comprehensive clinical trials. Moll et al investigated that the differences between fresh and cryopreserved MSCs were limited but significant. 100 Fresh MSCs are the best choices, however, considering the COVID-19 outbreak, the lack of sufficient donors for providing menstrual blood in a short time, and the consistent standard of care between the two hospitals situated in different cities, freshly thawed MSCs were used in this study.
Finally, the small sample size of the current study limited our ability to obtain enough clinical data to draw strong conclusions.

CONCLUSIONS
This prospective and systematic study assessed the ability of menstrual blood-derived MSCs to treat both severe and critically ill COVID-19 patients. The results of this multicenter, open, and parallel-controlled clinical study suggest that menstrual blood-derived MSC transplantation significantly lowers the mortality of severe and critical SARS-CoV-2-induced COVID-19. Menstrual blood-derived MSCs may act by alleviating the breathing difficulties caused by COVID-19 and reducing the symptoms of ARDS or expiratory dyspnea. Although the body of research on MSCs is still in its infancy and lacks important long-term safety information, MSC-based therapy may serve in future clinical applications as an alternative method for the treatment of COVID-19.