Background: Since December 2019, COVID-19 has spread to the world which leads to a global health threat. We aimed to investigate the effectiveness of tocilizumab on COVID-19 patients.
Methods: We systematically searched PubMed, EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL) and WHO international Clinical Trials Registry Platform (ICTRP) from their inception to March 10, 2021 for randomized controlled trials (RCTs) on tocilizumab supplementation in adults with COVID-19 disease. The primary outcomes were mortality at 28-30 day and 60-day, incidence of mechanical ventilation (MV), composite outcome of death or MV, time to hospital discharge, and intensive care unit (ICU) admissions. A random-effects meta-analysis model was used to pool studies.
Results: Eleven studies with a total of 6,579 patients were included in our meta-analysis, of which 3,406 and 3,173 were respectively assigned to the tocilizumab and control groups. Tocilizumab could significantly reduce 28-30 day mortality (RR = 0.89, 95% CI 0.80-0.99, P = 0.04), incidence of MV (RR= 0.79, 95% CI 0.71-0.89, P = 0.0001), composition outcome of MV or death (RR = 0.81, 95% CI 0.72-0.90, P = 0.0002), time to hospital discharge (HR = 1.30, 95% CI 1.16-1.45, P ＜ 0.00001 ), ICU admissions (RR = 0.64, 95% CI 0.47-0.88, P = 0.006), serious infection (RR = 0.61, 95% CI 0.40-0.94, P = 0.02) and events of serious adverse advents (RR = 0.64, 95% CI 0.47-0.86, P = 0.004). There was no significant difference between tocilizumab and control groups in 60-day mortality and adverse events (AEs).
Conclusions: Tocilizumab could reduce the short-term mortality, incidence of MV, composite outcome of death or MV, ICU admissions, serious infection and events of serious adverse advents, and shorten the time to hospital discharge in hospitalized patients with COVID-19. The optimal effective dose needs to be confirmed by further studies.
In December 2019, there had been more and more confirmed cases of novel coronavirus pneumonia in Wuhan, China, which then quickly spread to the world leading to a global health threat . The World Health Organization (WHO) officially named this pneumonia coronavirus disease 2019 (COVID-19) on February 11, 2020.  COVID-19 can be mild or progressive to dyspnea and/or hypoxemia, and severe cases often progress to respiratory failure, acute respiratory distress syndrome (ARDS) and septic shock, which can further lead to multiple organ dysfunction syndrome (MODS) or death [3, 4]. Although most COVID-19 patients are self-limited, it still causes serious loss of life worldwide . As of April 26, 2021, more than 14.6 million people have been infected and more than 3 million people have died around the world . All parts of the world are striving to find effective treatments to control the ongoing COVID-19 pandemic .
Recently, many studies have shown that cytokine release syndrome (CRS) is an important cause of death in COVID-19 patients, and IL-6 plays an important role [7–9]. Tocilizumab is a recombinant humanized monoclonal antibody against human interleukin 6 (IL-6) receptor, which has been confirmed that it could reduce the biomarkers of COVID-19 infection and increase the level of lymphocyte count .
There are several meta-analyses of observational studies which have showed that tocilizumab could reduce the mortality of COVID-19 [11–15]. Considering the low levels of evidence in observational studies may confound the findings, the benefits of tocilizumab on mortality of COVID-19 must be cautiously interpreted. Several newly published randomized controlled trials (RCTs) [6, 16–25] and meta-analyses [26–28] of RCTs have investigated the effects of tocilizumab as adjunctive therapy in COVID-19 patients, but reported inconsistent results. Moreover, due to daily update of studies regarding tocilizumab anti-COVID-19, it is mandatory to conduct an updated study on this field. Hence, we conducted an updated meta-analysis to synthesize the evidence of well-conducted RCTs to evaluate the effects of tocilizumab in COIVD-19 treatment.
We systematically searched PubMed, EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL) and WHO international Clinical Trials Registry Platform (ICTRP) from their inception to March 10, 2021 for RCTs, by using a combination of Medical Subject Headings (MeSH) or Emtree and related key words in all fields. The keywords used were “tocilizumab” or “atlizumab” or “Actemra” or “Roactemra” or “lusinex” or “anti-interleukin 6 antibody”, and “COVID-19” or “coronavirus 2019” or “2019-nCoV Infection” or “SARS CoV 2 Infection” or “2019 Novel Coronavirus Disease”. We also scanned the reference lists from relevant studies and key review articles to locate relevant studies.
Studies meeting the following criteria were included: (1) participants: patients aged ≥ 18 years with confirmed SARS-coV-2 infection by a positive polymerase-chain -reaction test for SARS-CoV-2 in any body fluid and/or bilateral chest infiltrates on chest x-ray or computed tomography; (2) intervention: tocilizumab administered intravenously (IV), with dosages ranging from 400 to 800 mg; (3) comparison: standard care; (4) outcomes: the primary outcomes were mortality at 28–30 day and 60-day, incidence of mechanical ventilation (MV), composite outcome of death or MV, intensive care unit (ICU) admissions, and time to hospital discharge. The secondary outcomes were time to oxygen supply independency, non-serious adverse events, serious adverse events, serious infection, events of serious adverse advents; (5) study design: RCTs. The language was restricted to English. Two of the authors (JZ and CC) independently evaluated the eligibility of all studies obtained from the databases according to the above selection criteria. Discrepancies regarding study inclusion between reviewers were resolved through discussion.
Two of the authors (JZ and CC) independently extracted data. The following data were extracted from the studies: study name (name of the first author with publication year), country and design, participants (sample size, sex and age), intervention arms and controls (intervention drug, dose, and duration of follow-up) and outcomes (primary and secondary outcomes). The Cochrane Collaboration’s tool for assessing risk of bias was used to appraise the quality of each RCT, which includes the following criteria: adequacy of sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessors, incomplete outcome data, selective reporting and other biases . JZ and CC reviewed all the included studies and rated them “low risk”, “unclear risk” or “high risk” based on the Cochrane risk-of -bias tool.
To evaluate the effect of tocilizumab on COVID-19, we calculated relative risks (RRs) with 95% confidence intervals (CIs) for dichotomous outcomes. As for continuous outcomes, mean differences or standard mean differences between the tocilizumab and placebo groups were used for meta-analysis. Time-to-event outcomes were analyzed by using hazard ratios (HRs). Heterogeneity in results across studies was examined by using Cochran’s Q and I2 statistics . The null hypothesis that the studies are homogeneous was rejected if the P value for heterogeneity was < 0.10 or I2 was > 50%. Studies with an I2 statistic of > 50% were considered to have significant heterogeneity. A random-effects model was used to pool the study estimates for each outcome.
A sensitivity analysis was conducted to assess the influence of individual studies on the pooled result when P was < 0.10 or I2 was > 50%, by excluding each study one by one and recalculating the combined results on the remaining studies . All analyses of data were performed with Review Manager 5.4 (Cochrane Informatics and Knowledge Management Department), available from http://tech. cochrane.org/.
Fig 1 shows a flow diagram of the study selection process. A total of 1,074 records were initially identified from the database search. Of these, 382 records were excluded for duplicates, and 653 records were excluded after screening the titles and abstracts. After full-text screening, a total of 11 studies [6, 16-25] were included in the meta-analysis.
Characteristics of included studies
The characteristics of studies included in our meta-analysis were summarized in Table 1. All 11 RCTs were multicenter studies. Three [16, 18, 19] studies were conducted in multi-countries, while the remaining eight trials were each from France , Itay , USA , Brazil , China [6, 23], UK , India . Overall, a total of 6,579 patients were enrolled in our meta-analysis, 4,906 of them were male (74.6%), and the average age ranged from 54 to 75 years old. A total of 3,406 were administrated with tocilizumab in addition to standard care or placebo, 14 of which were administrated with tocilizumab and favipiravir, and 3,173 were administrated with standard care or placebo, 7 of which were administrated with favipiravir. Except two [21, 22] RCTs used only a single dose of tocilizumab, other nine [6, 16-20, 23-25] RCTs allowed additional dose if needed. The tocilizumab doses vary from 400 mg to 800 mg and were all administrated intravenously infusion for more than one hour. The maximum dose was 480 mg/d in one study , 800 mg/d in eight studies [16-22, 25], 400 mg/d in two [6, 23] studies.
Assessment of risk of bias
There was a high risk of bias in blinding of participants and personnel and blinding of outcome assessment because of open-labelled design in eight studies [6, 16, 17, 20, 22-25], there was an unclear risk of bias in allocation concealment because of no mentioned of allocation in one study , shown in Fig 2.
Nine studies [16-22, 24, 25] with 6,493 patients were included in the meta-analysis. Overall, there was a significant difference between tocilizumab and control groups at 28–30day mortality (RR = 0.89, 95% CI 0.80-0.99, P = 0.04). Two trials with 507 patients contributed to the 60-day mortality, and no statistically difference was found (RR = 0.88, 95% CI 0.54-1.43, P = 0.60), shown in Fig 3.
Incidence of mechanical ventilation
Eight trials [16-19, 21, 22, 24, 25] examined the incidence of MV between tocilizumab and control groups. The pooled analysis including 5,369 participants showed that tocilizumab could significantly decrease the incidence of MV (RR = 0.79, 95% CI 0.71-0.89, P = 0.0001), shown in Fig 4.
Composite outcome of death or MV
Eight RCTs [16-19, 21-23, 25] including 5,241 adults examined the composite outcome of death or MV. The pooled analysis showed that there was a significant difference between tocilizumab and control groups (RR = 0.81, 95% CI 0.72-0.90, P = 0.0002) (supplemental Fig 5).
Time to hospital discharge
Pooled analysis of five trials [16-19, 21] with 1,943 cases showed that there was a statistically significant difference in time to hospital discharge between tocilizumab and control groups (HR = 1.30, 95% CI 1.16-1.45, P＜ 0.00001 ) (supplemental Fig 6).
Four trials [17, 18, 20, 24] with 499 cases were included in the meta-analysis. Overall, there was statistically significant difference between tocilizumab and control groups in ICU admissions (RR= 0.64, 95% CI 0.47-0.88, P = 0.006) (supplemental Fig 7).
Time to oxygen supply independency
Our meta-analysis including three RCTs [17, 21, 22] with 502 cases showed that there was no significant difference in time to oxygen supply independency between tocilizumab and control groups (HR = 1.21, 95% CI 0.94-1.57, P = 0.14) (supplemental Fig 8).
Five RCTs [17-21] including 1,311 cases were included in the meta-analysis. Overall, there was statistically significant difference between tocilizumab and control groups in serious infection (RR = 0.61, 95% CI 0.40-0.94, P = 0.02) (supplemental Fig 9).
Non-serious adverse events and serious adverse events
Eight studies [6, 17-20, 22-24] including 1463 patients showed that there was no difference between tocilizumab and control groups in non-SAE (RR = 1.19, 95% CI 0.94-1.50, P = 0.14) (supplemental Fig 10). Nine trials [6, 16-22, 24] with 2,440 participants showed that there was no significant difference between tocilizumab and control groups in SAE (RR = 0.91, 95% CI 0.76-1.08, P = 0.28) (supplemental Fig 11).
Events of serious adverse advents
Our meta-analysis including four trials [17, 18, 21, 24] with 991 cases showed that there was a significant difference between tocilizumab and control groups in events of serious adverse advents (RR = 0.64, 95% CI 0.47-0.86, P = 0.004) (supplemental Fig 12).
Our meta-analysis investigated the effect of adjunctive tocilizumab on COVID-19 patients in hospital and found that tocilizumab supplementation could reduce 28–30 days mortality, the incidence of mechanical ventilation, composite outcome of death or mechanical ventilation, ICU admissions, serious infection and events of SAE, and shorten the time to discharge. There was no evidence that tocilizumab could increase the adverse advents and reduce 60-day mortality and time to oxygen supply independency.
The results of our meta-analysis are not completely consistent with recently published meta-analyses [26, 28, 31, 32], which showed that tocilizumab had no effect on 28–30 days mortality in COVID-19 patients. There are several strengths in our meta-analysis. Firstly, with the updating of the literature, our meta-analysis was able to include the most recently published RCTs. Secondly, given the clinical heterogeneity across included studies, we used the random-effects model to pool the results, which is generally a plausible match to the underlying population effect distribution .
All published meta-analyses have different degrees of benefit from tocilizumab. Tleyjeh et al  showed that tocilizumab could reduce the risk of MV and the composition outcome of MV or death. Lin et al  found that COVID-19 patients receiving tocilizumab had a lower rate of MV, ICU admission, composite outcome of MV or death compared with control group. Sophie et al  showed evidence of a beneficial effect of tocilizumab compared with control on MV. Chia et al  found that COVID-19 patients with use of tocilizumab could benefit in the composite endpoint of MV and/or death. Moreover, Ghosn et al  including 8 RCTs had the same result that tocilizumab could reduce all-cause mortality at 28-day in hospitalized COVID-19 patients. Rezaei et al  including forty-five studies with 13,189 patients showed that tocilizumab can reduce mortality rates in severe to critical COVID-19 patients. Researches [35, 36] have shown that interleukin-6 is an important cytokine associated with mortality and severity of COVID-19. One genomic analysis  showed that genetic variants in interleukin-6 inflammatory pathway are associated with life-threatening in COVID-19. All of the studies support the therapeutic strategy of inhibiting interleukin-6 in severe COVID-19 patients. Tocilizumab is an anti-IL-6 receptor-a monoclonal antibody, which has been identified to specifically bind soluble and membranal IL-6R and inhibit signal transduction . Combined with our meta-analysis, we conclude that tocilizumab could reduce the short-term mortality, the composition outcome of MV or death, risk of MV, ICU admission in moderate to critical COVID-19 patients.
Our study analyzed time to discharge and found that tocilizumab could significantly shorten time to hospital discharge. We also analyzed time to oxygen supply independency, and found that there was no significant difference between tocilizumab and control group. A retrospective study  found that patients had lowered their oxygen intake after using tocilizumab compared with control group in severe COVID-19 patients. Considering the data that could be analyzed was limited, we only included three RCTs [17, 21, 22] in this meta-analysis. Considering the limited studies and patients, we could not conclude that tocilizumab has no effect on time to oxygen supply independency. There are need more effective and larger RCTs to confirm this issue.
As for safety, all published meta-analyses [26–28, 31, 32] got the similar conclusions that tocilizumab was safe and did not increase adverse events and serious adverse events compared with control group in COVID-19 patients which were consistent with our results. We also found that tocilizumab could reduce the serious infection and events of SAEs. Lin et al  and Ghosn et al  also separately found that tocilizumab could reduce serious infection and SAEs, which were consistent with our results.
Although a major strength of this meta-analysis was that we included the largest number of RCTs in currently, this study also had limitations. First, there are many ongoing RCTs, also indicate that follow-up of ongoing trials is needed and meta-analyses need to be updated when necessary. Second, the dose of tocilizumab varied from 400–800 mg/d in included studies, the optimal effective dose of tocilizumab is still uncertain. Thirdly, given the limited data on oxygen supply independency and 60-day mortality, further studies are warranted.
Tocilizumab could reduce the short-term mortality, the incidence of mechanical ventilation, the composite outcome of death or MV, ICU admissions, serious infection and events of SAE, and shorten the time to discharge, and could not increase the adverse advents of the hospitalized patients with COVID-19. But the optimal effective dose needs to be confirmed by further studies.
Cochrane Central Register of Controlled Trials
international Clinical Trials Registry Platform
randomized controlled trials
intensive care unit
World Health Organization
acute respiratory distress syndrome
multiple organ dysfunction syndrome
cytokine release syndrome
Medical Subject Headings
serious adverse events
Ethics approval and consent to participate
Consent for publication
Availability of data and materials
Competing of interest
The authors declare that they have no competing interests.
The authors declare that this research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Chun Chen and Jing Zhang initiated and coordinated the study. Chun Chen, Jing Zhang and ZeMei Zhou were responsible for the literature research, data extraction, statistical analysis, and Chun Chen wrote the first draft. Studies reviewed by Jing Zhang. All authors read and approved the final manuscript.
Due to technical limitations, table 1 is only available as a download in the Supplemental Files section.