Abstract
Background:
Coronavirus disease 2019 (COVID-19) continues to spread quickly throughout the world, mainly due to the lack of effective drug therapies and vaccines. The effectiveness of the antiviral drug umifenovir needs to be further clarified.
Methods:
This retrospective cohort study included 1254 patients who were diagnosed with COVID-19 between February 19 and April 5, 2020 in Hubei Maternity and Child Health Hospital. They were divided into umifenovir group (n = 760, 60.60%) and control group (n = 496) without using umifenovir. The primary endpoint was a composite of intubation or death in a time-to-event analysis. The clinical outcomes were compared between the two groups using multivariable Cox analysis with inverse probability weighting according to the propensity score.
Results:
A total of 760 (60.60%) patients received umifenovir, and 496 patients did not do so. Of the enrolled patients, 1049 (83.65%) had mild or moderate COVID-19, and the remaining 205 had severe or critical COVID-19. The mortality rate in the umifenovir group was 2.76% (21/760) versus 2.02% (10/494) in the control group. In terms of treatment outcomes, the discharge status of the patients in the umifenovir group was no better than that in the control group after propensity score matching (n = 485 in each group). In addition, the respiratory rate, a severe condition, or critical condition of the disease were the three main risk factors affecting the endpoint of death (p = 0.0028, p = 0.0009 and p < 0.0001, respectively).
Conclusion:
This retrospective cohort study showed that oral administration of umifenovir alone did not improve outcomes for patients with COVID-19.
Introduction
The coronavirus disease 2019 (COVID-19) pandemic has spread to more than 188 countries with more than 267,184,623 confirmed cases and more than 5,277,327 confirmed deaths worldwide by December 9, 2021. 1 The consequences of this global pandemic are not only to potentially overwhelm national and international health care and economic systems but it may also carry long-term health risks in addition to primary COVID-19. As COVID-19 affects the entire body and has multiple impacts on the immune system, the long-term complications are currently difficult to predict.2,3 The most critical side effect of COVID-19 is severe bilateral pneumonia which may progress to multi-organ failure and death. 4 It is possible that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) may linger in communities for a long time as there is an average 2-week latency period in asymptomatic populations. 5
In terms of mitigating COVID-19, prevention, in the form of appropriate vaccines, is currently the best option, and the first vaccine was created by the Chenwei’s team in China. 6 However, the number of available vaccines is still limited and far from the number required, given that the rate of diagnosis of new cases of COVID-19 currently exceeds 260,000,000. 1
Hydroxychloroquine and chloroquine have attracted the most interest as a potential treatment for COVID-19 and was conclusively assessed in a report by Mehra et al., 7 which analyzed data from 671 hospitals across six continents. The study was unable to confirm any benefits of hydroxychloroquine, when used alone or with a macrolide, on COVID-19 in-hospital outcomes. However, a correlation between hydroxychloroquine use was found between decreased in-hospital survival and an increased frequency of ventricular arrhythmias.
Umifenovir (sold under the brand name Arbidol) is a broad-spectrum antiviral drug which is licensed in Russia and China for the prevention and treatment of human influenza A and B infections, in addition to post-influenza complications, and can also offer protection against other enveloped and non-enveloped viruses.8 –10 Molecular dynamics and structural studies have shown that the spike glycoprotein of SARS-CoV-2-19 is a target of umifenovir. 11 However, the clinical effect of umifenovir on COVID-19 is not conclusive as the few clinical studies conducted have not included many patients. In January 2020 in Wuhan, China, a retrospective study including an umifenovir treatment group (n = 16) and a lopinavir/ritonavir (LPV/r) treatment group (n = 34) in patients with COVID-19 reported that umifenovir was superior to LPV/r in reducing the duration of time that patients returned positive RNA test results. 12
Clinical trials focusing on the combination therapies with other drugs have also been conducted. A clinical trial using umifenovir in combination with LPV/r or LPV/r alone in 16 patients with COVID-19 found that the clinical outcomes of umifenovir plus LPV/r were more favorable than LPV/r alone. 13 Another study, which included three hospitals in the Wuhan and Guangdong Provinces of China included 114 patients with COVID-19 treated with umifenovir/interferon (IFN)-α2b therapy versus IFN-α2b alone, showed no significant difference in terms of the clinical outcomes between the two groups. 14 A study which included 81 COVID-19 patients reported no significant difference in the clinical outcome between patients who received umifenovir (n = 45) and those who did not (n = 36). 15 Studies with larger sample sizes are required to verify the clinical effects of umifenovir to treat COVID-19.
The aim of this large-scale retrospective cohort study was to evaluate the clinical outcomes of umifenovir as a treatment for COVID-19 at a large center with a substantial number of COVID-19 patients. The primary endpoint of the study was to assess the number of deaths, according to propensity score, in patients who received umifenovir treatment for COVID-19 and those who did not.
Materials and methods
Study design and participants
A total of 1492 patients who were admitted from February 19 to April 5, 2020 were enrolled in the study. The inclusion criteria were a diagnosis with COVID-19 and the availability of complete patient information. The exclusion criteria were incomplete diagnostic information, incomplete treatment information, and death within 24 h of admission. Each patient RNA tested positive for SARS-CoV-2 by oropharyngeal or nasopharyngeal specimen testing and was diagnosed with COVID-19.
Data were obtained from patient electronic medical records, including age, sex, vital signs, symptoms, details of underlying illnesses, laboratory findings, patient classification, historical and current medication lists, diagnoses, and clinical notes.
Grouping and treatment
The patients were divided into two groups based on whether they received umifenovir or not (umifenovir group and control group). All of the included patients received symptomatic treatment for COVID-19, including regular laboratory monitoring and standard care. Budesonide with the nebulizer as inhalation treatment was given for 5 min/day. Besides, patients were given standard treatment according to the ‘Diagnosis and treatment plan for the New Coronavirus Pneumonia’, including electrolyte balance, nutritional support oxygen therapy, and traditional Chinese medicine compounds, including Xuebijing (Tianjin HongRi Pharmaceutical Co., Ltd., 50 ml + 0.9% normal saline 100 ml, twice a day, the infusion was completed within 30–40 min) and Lianhua Qingwen (Shijiazhuang Yiling Pharmaceutical Co., Ltd, four pills each time, three times a day). Patients in the umifenovir group received 0.2 g umifenovir tid (three times a day). For the purpose of further clarifying the impact of umifenovir based on the severity of COVID-19, we divided the patients into a mild or moderate subgroup and a critical or severe subgroup according to the criteria of the diagnosis and treatment program of COVID-19. 16 To assess the clinical safety profile, white blood cell (WBC) count, platelet count, parameters that reflect the liver and kidney function, such as aspartate aminotransferase, alanine aminotransferase, bilirubin, and creatinine were also analyzed.
Endpoint
The main endpoint was the time from the date of admission to death or intubation. For patients who died after intubation, the primary endpoint was defined as the time of intubation.
Umifenovir tablet exposure
Patients were defined as receiving umifenovir at study baseline or during the follow-up period before intubation or death. The study baseline was defined as 24 h after admission.
Statistical analysis
The demographic and clinical characteristics of patients are expressed as median and interquartile range (IQR) for continuous variables, and as frequencies and percentages for categorical variables. The primary endpoint was the time from study baseline to death or discharge, and patients without a primary endpoint by April 5, 2020 had their data removed from analysis. Both univariable and multivariable Cox proportional hazards regression models, adjusted for all the baseline demographic and clinical covariates, were used to compare the overall survival (OS) between umifenovir and control groups.
To reduce the effects of confounding factors, we conducted analyses with four propensity score methods: propensity score matching, inverse probability weighting, overlap weighting, and covariate adjustment using the propensity score. The individual propensity score of receiving umifenovir was estimated via a multivariable logistic regression model which included all covariates of the multivariable Cox proportional hazards regression model. The patient ratio in umifenovir and control groups was 1:1 based on the nearest available matching method with calipers of width equal to 0.1 of the standard deviation of the logit of the propensity score (R package ‘MatchIt’). 17 The weights of inverse probability weighting and overlap weighting were calculated using R package ‘PSW’ with the covariates in the previous final propensity score (PS) model. 18 We included the absolute mean differences to evaluate the balance of covariates between two groups, and the criterion for covariate unbalance was set to 0.1. For the PS models, Kaplan–Meier curves and Cox models were also conducted.
The chi-square test was applied to compare the outcome of hospitalization between umifenovir and control groups. All analyses and figures were performed using the R software package (version 3.5.2). All reported p-values were two-sided, and a p-value less than 0.05 was considered statistically significant.
Results
Characteristics of the cohort
Of the initially enrolled 1429 COVID-19 patients admitted to Guanggu District of Hubei Maternity and Child Health Hospital between February 19 and April 5, 2020, 175 were excluded from the study because the available clinical information was incomplete. Finally, 1254 patients were included for analysis (Figure 1). Of the 1254 COVID-19 patients, 760 (60.61%) received umifenovir. The mortality rate in the umifenovir group was 2.76% versus 2.02% in the control group. The median age was 60.50 years. Of them, 736 (58.69%) were female; 1049 (83.65%) had mild or moderate COVID-19, and 205 had severe or critical COVID-19.
Flowchart of patients’ selection for the study.
The distribution of the patient characteristics according to the group is shown in Table 1. In the samples before PS matching, the baseline of the two groups were relatively balanced, except for the lymphocyte count and respiratory rate. Patients in the umifenovir group had a lower lymphocyte count than patients in the control group (median, 1.54 versus 1.66) (Supplementary Table 1). However, after propensity score matching and weighting, the patient characteristics were balanced between the two groups (Table 1).
Demographic and clinical characteristics of patients receiving or not receiving umifenovir tablets.
IQR, interquartile range; SaO2, oxygen saturation.
Respiratory rate of COVID-19 patients on admission.
Refer to China’s National Health Commission. Diagnosis and Treatment Protocol for COVID-19 Pneumonia (fifth edition).
Treatment outcomes
Among the 1254 patients included in this analysis, 990 patients (78.95%) were cured of COVID-19, 31 (2.47%) patients died, and 11 patients required intubation (Supplementary Table 2). There was a statistically significant difference between deaths in the umifenovir group compared to the control group before propensity matching (p = 0.0359).
A rough analysis showed no significant difference in the risk of death in patients who received umifenovir compared with the control group (hazard ratio, 1.033; 95% CI, 0.482–2.213) (Table 2).
Associations between umifenovir tablets use and endpoint of death.
CI, confidence interval.
Shown is the hazard ratio from the multivariable Cox proportional hazards model, with stratification according to age, sex, basic disease, and body mass index, and with additional adjustment for age, sex, current medications, vital statistics, and laboratory tests on presentation. The analysis included all 1254 patients.
Shown is the hazard ratio from a multivariable Cox proportional hazards model with the same strata and covariates with matching according to the propensity score. The analysis included 970 patients (485 who received umifenovir tablets and 485 who did not).
Shown is the primary analysis with a hazard ratio from the multivariable Cox proportional hazards model with the same strata and covariates with inverse probability weighting according to the propensity score. The analysis included all the patients.
Shown is the primary analysis with a hazard ratio from the multivariable Cox proportional hazards model with the same strata and covariates with overlap weighting according to the propensity score. The analysis included all the patients.
Shown is the hazard ratio from a multivariable Cox proportional hazards model with the same strata and covariates, with additional adjustment for the propensity score. The analysis included all the patients.
Multivariate regression analysis showed that umifenovir use, hormone use, sex, age, SaO2, neutrophil count, lymphocyte count, hs-CRP, hypertension, diabetes, and tumors were not predictive factors of death.
However, the respiratory rate, the severity of disease, and the criticality of the disease were risk factors for death (p = 0.0028, p = 0.0009, p < 0.0001, respectively) (Supplementary Table 3).
Matching analysis based on propensity score showed no significant association between use of umifenovir and the endpoint of death (hazard ratio, 0.574; 95% CI, 0.181–1.814). The results of the four methods (propensity score analysis, matching, inverse probability weighting, and overlap weighting) were similar (Table 2, Figure 2).
Freedom from primary endpoint of death. (a) The unmatched analysis. (b) The propensity scores matching analysis. (c) The inverse probability weighting analysis. (d) The overlap weighting analysis.
In the control group, there were 424 mildly and moderately ill patients, of whom 1 patient died; and 70 severely and critically ill patients, of whom 9 patients died. In the umifenovir group, there were 625 mildly and 135 moderately ill patients, in whom 20 patients died. There was no statistically significant difference between the control and umifenovir groups.
Discussion
In the current study, which included a large number of consecutive COVID-19 patients, there were no significant differences in the mortality rate between the umifenovir and control treatment groups, regardless of the severity of COVID-19. In terms of clinical characteristics, the only difference was a significantly lower lymphocyte count in the umifenovir group than in the control group, but this was no longer significant after propensity matching.
As a broad-spectrum antiviral drug, umifenovir has a direct antiviral effect on the early phase of SARS-CoV-2 replication within each cell. Further molecular dynamics studies demonstrated that the target of umifenovir action is the SARS-CoV-2 spike glycoprotein which also plays a key role in allowing the virus to enter host cells. 11 Therefore, it is logical to postulate that umifenovir could be a promising treatment for COVID-19. Earlier pilot studies in China showed the benefit of umifenovir monotherapy to treat COVID-19. 12 The trial included an LPV/r group (n = 34) and an umifenovir group (n = 16). LPV/r was administered at a dose of 400 mg/100 mg, bid, po, and umifenovir was administered at a dose of 0.2 g, tid, po, and both were administered for 1 week. They found that umifenovir was more effective at reducing the viral load than LPV/r, as there was no detectable viral load in the umifenovir group, but only 15 (44.1%) patients who were treated with LPV/r alone had no detectable viral load 14 days after admission. 12 However, this study had a limited number of patients in each group and information bias between the two groups. LPV/r, which was an effective antiviral treatment for severe acute respiratory syndrome virus, 19 did not improve clinical outcomes in patients with severe COVID-19 in a clinical trial containing 99 patients treated with LPV/r and 100 patients receiving standard care. 20 An in vitro study using six licensed anti-influenza drugs against SARS-CoV-2 virus reported that umifenovir was the only drug that inhibited the viral infection during and after cell entry. 21
Combination therapy using umifenovir and IFN-α2b was carried out in three hospitals in Hubei and Guangdong provinces and included 141 adults. Umifenovir and IFN-α2b versus 10- to 14-day inhalation of IFN-α2b monotherapy showed that the absorption of pneumonia was only seen in the combined therapy group. 14 A comparative study of combined oral umifenovir plus LPV/r (n = 16) demonstrated a greater therapeutic benefit than oral LPV/r monotherapy (n = 17) over 5–21 days of treatment. 13 However, the number of patients included in their study was limited and the treatment duration was quite short, which impacted the strengths of the study. Our findings also demonstrated that the respiratory rate, and the classification of severe and critical conditions are the three main factors that correlated with the endpoint of death, indicating the systemic influence of COVID-19. However, the long-term adverse effects or toxicity of umifenovir on the immune system need to be further evaluated.
Conclusion
Our study of a large patient cohort demonstrated that umifenovir monotherapy did not significantly improve the clinical outcome of COVID-19. Combination of umifenovir with other drugs or vaccines should be assessed in future studies.
Supplemental Material
sj-doc-1-tar-10.1177_17534666231183811 – Supplemental material for Effect of umifenovir (arbidol) versus standard care on clinical outcome in patients with COVID-19: a retrospective cohort study
Supplemental material, sj-doc-1-tar-10.1177_17534666231183811 for Effect of umifenovir (arbidol) versus standard care on clinical outcome in patients with COVID-19: a retrospective cohort study by Wanwan Yi, Chengyou Jia, Ziyu Yang, Hengwei Fan, Ming Li and Zhongwei Lv in Therapeutic Advances in Respiratory Disease
Footnotes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
