Drug Res (Stuttg) 2021; 71(05): 265-274
DOI: 10.1055/a-1336-2371
Original Article

Efficacy and Safety of Tocilizumab for Coronavirus Disease 2019 (Covid-19) Patients: A Systematic Review and Meta-analysis

Timotius Ivan Hariyanto
1  Faculty of Medicine, Pelita Harapan University, Karawaci, Tangerang, Indonesia
,
Willie Hardyson
1  Faculty of Medicine, Pelita Harapan University, Karawaci, Tangerang, Indonesia
,
Andree Kurniawan
2  Department of Internal Medicine, Faculty of Medicine, Pelita Harapan University, Karawaci, Tangerang, Indonesia
› Author Affiliations
 

Abstract

Background Currently, the data regarding the effectiveness and safety of tocilizumab as treatment for COVID-19 infection is still conflicting. This study aims to give clear evidence regarding the potential benefit and safety of tocilizumab in improving the outcome of COVID-19 patients.

Methods We systematically searched the PubMed and Europe PMC database using specific keywords related to our aims until November 1st, 2020. All articles published on COVID-19 and tocilizumab were retrieved. Statistical analysis was done using Review Manager 5.4 software.

Results A total of 38 studies with a total of 13 412 COVID-19 patients were included in our analysis. Our meta-analysis showed that tocilizumab treatment is associated with reduction of mortality rate from COVID-19 [OR 0.54 (95% CI 0.42–0.71), p<0.00001, I 2=79%, random-effect modelling], but did not alter the severity of COVID-19 [OR 1.05 (95% CI 0.92–1.20), p=0.47, I 2=84%, random-effect modelling] and length of hospital stay [Mean Difference 1.77 days (95% CI −0.61–4.14 days), p=0.15, I 2=97%, random-effect modelling]. Tocilizumab also does not associated with serious adverse events compared with standard of care treatment [OR 0.91 (95% CI 0.71–1.15), p=0.42, I 2=46%, random-effect modelling].

Conclusion Our study does not support the routine use of tocilizumab for COVID-19 patients. Future studies should focus more on other potential therapies for COVID-19 patients.


#

Introduction

Until now, the number of positive and death cases from coronavirus disease 2019 (COVID-19) is still increasing. This disease has caused significant health and economic burden across the world. The manifestations of the disease may vary from mild respiratory symptoms such as fever, nasal obstruction, and cough to severe life-threatening symptoms such as respiratory distress, shock, arrhythmia, and heart failure [1]. Several comorbid diseases has been demonstrated to be associated with severe COVID-19 infections, such as hypertension, diabetes, dyslipidemia, thyroid disease, cardiovascular disease, anemia, and pulmonary disease [2] [3] [4]. Currently, there are no widely accepted drugs for the management of COVID-19 patients. Several potential agents have been proposed to help in achieving faster recovery time and reducing the mortality rate in COVID-19 patients, and one of the agents is tocilizumab, an IL-6 inhibitor. Tocilizumab has been approved for the treatment of rheumatoid arthritis, juvenile idiopathic arthritis, and giant cell arteritis [5] Recently, tocilizumab has been offered to help in reducing the pro-inflammatory cytokines in COVID-19 and preventing the cytokine storm syndrome that could contribute to the development of the severe outcome. Unfortunately, the evidence regarding the potential benefit and safety of tocilizumab in COVID-19 patients is still conflicting. Therefore, a meta-analysis is required to aid in solving this problem. This article aims to explore the efficacy and safety of tocilizumab administration in patients with COVID-19.


#

Materials and Methods

Eligibility criteria

Studies were included in this review if met the following inclusion criteria: representation for clinical questions (P: positive/confirmed cases of COVID-19; I: receiving tocilizumab as their treatment; C: did not receive tocilizumab or receive only standard of care treatment; O: efficacy of tocilizumab (rate of severe COVID-19, mortality, and length of hospital stay) and serious adverse events of tocilizumab (thromboembolism incident and secondary infection); S: type of study was a randomized control trial, cohort, clinical trial, case-cohort, and cross-over design) and if the full-text article was available. The following types of articles were excluded: articles other than original research (e. g., review articles or commentaries); case reports; articles not in the English language; articles on research in pediatric populations (17 years of age or younger); and articles on research in pregnant women.


#

Search strategy and study selection

A systematic search of the literature was conducted on PubMed and Europe PMC using the keywords “tocilizumab” OR “anti-IL-6” OR “IL-6 inhibitor” AND “coronavirus disease 2019” OR “COVID-19”, between 2019 and present time (November 1st, 2020) with language restricted to English only. Duplicate results were removed. The remaining articles were independently screened for relevance by its abstracts with two authors. The full text of residual articles was assessed according to the inclusion and exclusion criteria. The references of all identified studies were also analyzed (forward and backward citation tracking) to identify other potentially eligible articles. The study was carried out per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [6]


#

Data extraction and quality assessment

Data extraction was performed independently by two authors, we used standardized forms that include author, year, study design, number of participants, age, gender, number of patients who receive tocilizumab and who did not, tocilizumab dose, and proportion of patients with each outcome of COVID-19.

The outcome of interest was severe COVID-19, mortality, length of hospital stay, and serious adverse events which comprised of thromboembolism incident and secondary infection. Severe COVID-19 was defined as patients who had any of the following features at the time of, or after, admission: (1) respiratory distress (≥30 breaths per min); (2) oxygen saturation at rest≤93%; (3) ratio of the partial pressure of arterial oxygen (PaO2) to a fractional concentration of oxygen inspired air (fiO2)≤300 mmHg; or (4) critical complication (respiratory failure, septic shock, and or multiple organ dysfunction/failure) or admission into ICU. Mortality outcome from COVID-19 was defined as the number of patients who were dead because of COVID-19 infection.

Two investigators independently evaluated the quality of the included cohort and case-control studies using the Newcastle–Ottawa Scale (NOS) [7]. The selection, comparability, and exposure of each study were broadly assessed and studies were assigned a score from zero to nine. Studies with scores≥7 were considered of good quality. They also independently evaluated the quality of the included clinical trial studies using the Revised Cochrane risk-of-bias tool for randomized trials (RoB 2) [8].


#

Statistical analysis

A meta-analysis was performed using Review Manager 5.4 (Cochrane Collaboration) software. Dichotomous variables were calculated using the Mantel-Haenszel formula with a random-effects model regardless of heterogeneity. The effect estimate was reported as risk ratio (RR) along with its 95% confidence intervals (CIs) for dichotomous variables, respectively. For continuous variables, the inverse variance method was used to obtain mean differences (MDs) and its standard deviations (SDs). P-value was two-tailed, and the statistical significance was set at≤0.05. A funnel plot, Begg’s rank correlation method [9], and Egger’s weighted regression method [10] were adopted to statistically assess publication bias (P<0.05 was considered statistically significant). When data were reported as medians and interquartile ranges, we would convert them to means and standard deviations for meta-analytical pooling using the formula by Wan X, et al [11].


#
#

Results

Study selection and characteristics

A total of 4274 records were obtained through systematic electronic searches and other ways. After the removal of duplicates, 4050 records remained. A total of 3956 records were excluded after screening the titles/abstracts because they did not match our inclusion and exclusion criteria. After evaluating 94 full-texts for eligibility, 54 full-text articles were excluded because they do not have the control/comparison group, 2 full-text articles were excluded because the articles were not in English, and finally, 38 studies [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] with a total of 13 412 COVID-19 patients were included in the meta-analysis ([Fig. 1]). Of a total of 38 included studies, 3 were double-blind randomized-controlled trial (RCT), 4 were open-label RCT, 23 were retrospective cohort, 3 studies were prospective cohort, while the remaining 5 studies was a case-control study. The dose and preparation of tocilizumab used were varied among the included studies. Most of the included studies (24 studies) use intravenous tocilizumab at dosage 8 mg/kg, 1–2 doses, while the remaining studies use tocilizumab at 400 mg, 1–2 doses, and subcutaneous tocilizumab at a dosage of 324 mg given as two consecutive injections. The essential characteristics of the included studies are summarized in [Table 1].

Zoom Image
Fig. 1 PRISMA diagram of the detailed process of selection of studies for inclusion in the systematic review and meta-analysis.

Table 1 Characteristics of included studies.

Study

Sample size

Design

Tocilizumab dose

Tocilizumab patients

Non-tocilizumab patients

n (%)

Age (years)

n (%)

Age (years)

Campochiaro C et al. [12] 2020

65

Retrospective cohort

IV: 400 mg, 1–2 doses

32 (49.2%)

64±16.2

33 (50.8%)

63.5±15.1

Canziani LM et al. [13] 2020

128

Case-control

IV: 8 mg/kg, 1–2 doses

64 (50%)

63±12

64 (50%)

64±8

Capra R et al. [14] 2020

85

Retrospective cohort

IV: 400 mg, 1 dose

62 (72.9%)

63.3±14.1

23 (27.1%)

68.3±18.5

Chilimuri S et al. [15] 2020

1225

Retrospective cohort

IV: 400 mg, 1–2 doses

87 (7.1%)

61.6±15.5

1138 (92.9%)

63±14.8

Colaneri M et al. [16] 2020

112

Retrospective cohort

IV: 400 mg, 1 dose

21(18.7%)

62.3±18.6

91 (81.3%)

63.7±16.3

De Rossi et al. [17] 2020

158

Retrospective cohort

IV: 400 mg, 1 dose
SC: 324 mg, 1 dose

90 (56.9%)

62.9±12.5

68 (43.1%)

71±14.6

Eimer J et al. [18] 2020

87

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

29 (33.3%)

56.6±10.3

58 (66.7%)

57.2±9.4

Enzmann MO et al. [19] 2020

150

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

12 (15.3%)

N/A

66 (84.7%)

N/A

Gokhale Y et al. [20] 2020

269

Retrospective cohort

IV: 400 mg, 1 dose

151 (56.1%)

52.3±11.8

118 (43.9%)

55.3±12.5

Guaraldi G et al. [21] 2020

544

Retrospective cohort

IV: 8 mg/kg, 2 doses
SC: 162 mg, 2 doses

179 (32.9%)

63.3±13.3

365 (67.1%)

68±15.5

Gupta S et al. [22] 2020

3924

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

433 (11%)

57±12.5

3491 (89%)

62.3±14.8

Hermine O et al. [23] 2020

130

Open-label RCT

IV: 8 mg/kg, 1–2 doses

63 (48.6%)

65.1±12.7

67 (51.4%)

64.2±11.2

Holt GE et al. [24] 2020

62

Retrospective cohort

IV: 400 mg, 1 dose

32 (51.6%)

N/A

30 (48.4%)

N/A

Ip A et al. [25] 2020

547

Retrospective cohort

IV: 400 mg, 1 dose

134 (24.4%)

61.6±12.5

413 (75.6%)

68±14.1

Kewan T et al. [26] 2020

51

Retrospective cohort

IV: 8 mg/kg, 1 dose

28 (54.9%)

62±13.3

23 (45.1%)

66.6±14.8

Klopfenstein T et al. [27] 2020

206

Case-control

IV: 8 mg/kg, 1–2 doses

30 (14.5%)

75.6±11.3

176 (85.5%)

74.3±11

Lengnan X et al. [28] 2020

19

Retrospective cohort

IV: 400 mg, 1 dose

5 (26.3%)

73.2±4.4

14 (73.7%)

66.2±5

Masia M et al. [29] 2020

138

Prospective cohort

IV: 400 mg if<75 kg and 600 mg if≥75 kg

76 (55%)

65.2±14.9

62 (45%)

65.9±16.8

Martinez-Sanz J et al. [30] 2020

1229

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

260 (21.1%)

65.3±15.5

969 (78.9%)

68.3±17

Menzella F et al. [31] 2020

79

Prospective cohort

IV: 8 mg/kg, 2 doses
SC: 162 mg, 2–4 doses

41 (51.8%)

63.3±10.6

38 (48.2%)

70.3±11.3

Mikulska M et al. [32] 2020

196

Prospective cohort

IV: 8 mg/kg, 1–2 doses
SC: 162 mg, 1–2 doses

130 (66.3%)

64.5±12.4

66 (33.7%)

73.5±14.4

Moiseev S et al. [33] 2020

137

Retrospective cohort

IV: 400 mg, 1 dose

83 (60.5%)

55.6±11.1

54 (39.5%)

56.3±14

Moreno-Perez O et al. [34] 2020

236

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

77 (32.6%)

62.3±14

159 (67.4%)

57±19.2

Perrone F et al. [35] 2020

301

Open-label RCT

IV: 8 mg/kg, 1–2 doses

180 (59.8%)

N/A

121 (40.2%)

N/A

Potere N et al. [36] 2020

80

Case-control

SC: 162 mg, 2 doses

40 (50%)

59.8±16.9

40 (50%)

59.1±17

Price CC et al. [37] 2020

239

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

153 (64%)

N/A

86 (46%)

N/A

Rodriguez-Bano J et al. [38] 2020

432

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

88 (20.3%)

64.6±11.8

344 (79.7%)

68±12.5

Rojas-Marte G et al. [39] 2020

193

Case-control

IV: 8 mg/kg, 1–2 doses

96 (49.7%)

58.8±13.6

97 (50.3%)

62±14

Roomi S et al. [40] 2020

176

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

134 (78.8%)

65.4±10.5

36 (21.2%)

58±13.2

Rosas I et al. [41] 2020

438

Double-blind RCT

IV: 8 mg/kg, 1–2 doses

294 (67.1%)

60.9±14.6

144 (32.9%)

60.6±13.7

Rossi B et al. [42] 2020

246

Case-control

IV: 400 mg, 1 dose

106 (43%)

64.3±13

140 (57%)

70.1±16.5

Roumier M et al. [43] 2020

59

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

30 (50.8%)

58.8±12.4

29 (49.2%)

71.2±15.4

Ruiz-Antoran B et al. [44] 2020

506

Retrospective cohort

IV: 8 mg/kg, 1–2 doses

268 (52.9%)

65±11.7

238 (47.1%)

71.3±14.2

Salama C et al. [45] 2020

377

Double-blind RCT

IV: 8 mg/kg, 1–2 doses

249 (66%)

56±14.3

128 (34%)

55.6±14.9

Salvarani C et al. [46] 2020

126

Open-label RCT

IV: 8 mg/kg, 1–2 doses

60 (47.6%)

62.1±16.2

66 (52.4%)

61.6±14

Somers EC et al. [47] 2020

154

Retrospective cohort

IV: 8 mg/kg, 1 dose

78 (50.6%)

55±14.9

76 (49.4%)

60±14.5

Stone JH et al. [48] 2020

243

Double-blind RCT

IV: 8 mg/kg, 1 dose

161 (66.2%)

59.2±17.2

82 (33.8%)

56.3±17.1

Wang D et al. [49] 2020

65

Open-label RCT

IV: 400 mg, 1–2 doses

34 (52.3%)

64.1±9.6

31 (47.7%)

62±11.1


#

Quality of study assessment

Studies with various study designs including a clinical trial, cohort, and case-control were included in this review and assessed accordingly with the appropriate scale or tool. Newcastle Ottawa Scales (NOS) were used to assess the cohort and case-control studies ([Table 2]). All included studies were rated ‘good’. For clinical trial studies, the Revised Cochrane risk-of-bias tool for randomized trials (RoB 2) was used and all of the included trials showed a low risk of bias ([Table 3]). In conclusion, all studies were seemed fit to be included in the meta-analysis.

Table 2 Newcastle-Ottawa quality assessment of observational studies.

First author, year

Study design

Selection

Comparability

Outcome

Total score

Result

Campochiaro C et al. [12] 2020

Cohort

****

**

***

9

Good

Canziani LM et al. [13] 2020

Cohort

****

**

***

9

Good

Capra R et al. [14] 2020

Cohort

***

**

**

7

Good

Chilimuri S et al. [15] 2020

Cohort

***

**

***

8

Good

Colaneri M et al. [16] 2020

Cohort

****

**

***

9

Good

De Rossi N et al. [17] 2020

Cohort

****

**

***

9

Good

Eimer J et al. [18] 2020

Cohort

**

**

***

7

Good

Enzmann MO et al. [19] 2020

Cohort

**

**

***

7

Good

Gokhale Y et al. [20] 2020

Cohort

***

**

***

8

Good

Guaraldi et al. [21] 2020

Cohort

****

**

***

9

Good

Gupta S et al. [22] 2020

Cohort

****

**

***

9

Good

Holt GE et al. [24] 2020

Cohort

**

**

***

7

Good

Ip A et al. [25] 2020

Cohort

***

**

***

8

Good

Kewan T et al. [26] 2020

Cohort

***

**

***

8

Good

Klopfenstein T et al. [27] 2020

Cohort

***

**

**

7

Good

Lengnan X et al. [28] 2020

Cohort

***

**

***

8

Good

Masia M et al. [29] 2020

Cohort

***

**

***

8

Good

Martinez-Sanz J et al. [30] 2020

Cohort

****

**

***

9

Good

Menzella F et al. [31] 2020

Cohort

***

**

***

8

Good

Mikulska M et al. [32] 2020

Cohort

****

**

***

9

Good

Moiseev S et al. [33] 2020

Cohort

**

**

***

7

Good

Moreno-Perez O et al. [34] 2020

Cohort

**

**

***

7

Good

Potere N et al. [36] 2020

Cohort

***

**

**

7

Good

Price CC et al. [37] 2020

Cohort

***

**

***

8

Good

Rodriguez-Bano J et al. [38] 2020

Cohort

****

**

***

9

Good

Rojas-Marte G et al. [39] 2020

Case-control

***

**

***

8

Good

Roomi S et al. [40] 2020

Cohort

***

**

***

8

Good

Rossi B et al. [42] 2020

Case-control

***

**

***

8

Good

Roumier M et al. [43] 2020

Cohort

***

**

**

7

Good

Ruiz-Antoran B et al. [44] 2020

Cohort

***

**

***

8

Good

Salama C et al. [45] 2020

Cohort

***

**

***

8

Good

Somers EC et al. [46] 2020

Cohort

***

**

***

8

Good

Table 3 Risk of bias assessment for clinical trial studies using RoB-2 tool.


#

Tocilizumab and outcomes

Tocilizumab efficacy

Our pooled analysis showed that tocilizumab administration was associated with reduction of mortality rate from COVID-19 [OR 0.54 (95% CI 0.42–0.71), p<0.00001, I 2=79%, random-effect modelling] ([Fig. 2a]). However, tocilizumab administration did not alter the severity of COVID-19 [OR 1.05 (95% CI 0.92–1.20), p=0.47, I 2=84%, random-effect modelling] ([Fig. 2b]) and length of hospital stay [Mean Difference 1.77 days (95% CI −0.61–4.14 days), p=0.15, I 2=97%, random-effect modelling] ([Fig. 2c]).

Zoom Image
Fig. 2 Forest plot that demonstrates the association of tocilizumab with the mortality a, severe COVID-19 b, length of hospital stay c, and serious adverse events d in COVID-19 infection.

#

Tocilizumab safety

Our meta-analysis showed that tocilizumab administration was not associated with serious adverse events [OR 0.91 (95% CI 0.71–1.15), p=0.42, I 2=46%, random-effect modelling] ([Fig. 2d]). Subgroup analysis showed that tocilizumab administration was not associated with thromboembolism incident [OR 1.02 (95% CI 0.69–1.50), p=0.93, I 2=12%, random-effect modelling], nor secondary infection [OR 0.86 (95% CI 0.63–1.18), p=0.36, I 2=57%, random-effect modelling].


#
#

Subgroup analysis

Subgroup analysis for clinical trial studies showed a higher OR for mortality rate outcome [OR 0.90 (95% CI 0.64–1.26), p=0.54, I 2=0%, random-effect modelling] compared to observational studies [OR 0.50 (95% CI 0.38–0.67), p<0.00001, I 2=80%, random-effect modelling]. Subgroup analysis for clinical trial studies showed a lower OR for severe COVID-19 outcome [OR 0.81 (95% CI 0.53–1.23), p=0.32, I 2=23%, random-effect modelling] compared to observational studies [OR 1.11 (95% CI 0.96–1.28), p=0.15, I 2=86%, random-effect modelling]. Subgroup analysis for clinical trial studies showed a lower Mean Difference for length of hospital stay outcome [Mean Difference −1.43 days (95% CI −5.13–2.26 days), p=0.45, I 2=95%, random-effect modelling] compared to observational studies [Mean Difference 2.70 days (95% CI −0.59–5.99 days), p=0.11, I 2=97%, random-effect modelling]. Subgroup analysis for clinical trial studies showed a lower OR for serious adverse events outcome [OR 0.52 (95% CI 0.29–0.92), p=0.02, I 2=38%, random-effect modelling] compared to observational studies [OR 1.04 (95% CI 0.80–1.35), p=0.76, I 2=41%, random-effect modelling].


#

Publication Bias

The funnel-plot analysis showed a qualitatively symmetrical inverted funnel-plot for the association between tocilizumab administration and mortality ([Fig. 3a]), severe COVID-19 ([Fig. 3b]), length of hospital stay ([Fig. 3c]), and serious adverse events ([Fig. 3d]). Meanwhile, rank-correlation Begg’s test and regression-based Egger’s test were not statistically significant for all outcomes, showing no indication of publication bias ([Table 4]).

Zoom Image
Fig. 3 Funnel plot analysis for mortality a, severe COVID-19 b, length of hospital stay c, and serious adverse events d outcome.

Table 4 Summary of meta-analysis.

Outcomes

Effect size (95% Confidence Interval), p-value

Heterogeneity (I2), p-value

Begg’s test

Egger’s test

Number of Studies

Mortality

OR=0.54 [0.42–0.71],<0.00001

79%,<0.00001

0.968

0.284

37

Severe COVID-19

OR=1.05 [0.92–1.20], 0.47

84%,<0.00001

0.464

0.150

30

Length of hospital stay

Mean Difference=1.77 [−0.61–4.14], 0.15

97%,<0.00001

0.836

0.213

17

Thrombosis incident

OR=1.02 [0.69–1.50], 0.93

12%, 0.33

0.916

0.978

9

Secondary infection

OR=0.86 [0.63–1.18], 0.36

57%, 0.02

0.558

0.451

16


#
#

Discussion

Based on a contrite meta-analysis of available data, tocilizumab seems to be beneficial only in reducing the mortality rate from COVID-19 infection, but it did not alter the severity outcome of COVID-19 and the duration of hospital stay. However, our subgroup analysis that involves only clinical trial studies showed that tocilizumab failed to reduce the mortality rate from COVID-19 and cannot alter the severity outcome and length of hospital stay in COVID-19 patients. Tocilizumab also appears to be relatively safe in COVID-19 patients, compared with standard of care treatment as it is not associated with serious adverse events such as thromboembolism incident and secondary infection. Several reasons may be proposed to explain the lack of efficacy from tocilizumab administration in COVID-19 patients. First, interleukin-6 and other inflammatory proteins that are observed to be present at elevated levels in patients with COVID-19 represent host responses to the infection, similar to the elevations in cytokine levels seen in patients with endocarditis, sepsis, and other infections, rather than components of a self-amplifying inflammatory loop that would benefit from suppression [49]. Second, severe COVID-19 symptoms may not be caused by cytokine storm syndrome like we used to think before. Recently published systematic review and meta-analysis showed that the descriptor cytokine storm does not appropriately describe the milieu in COVID-19-induced organ dysfunction. The mean IL-6 concentration in COVID-19 patients is relatively low (36.7 pg/mL (95% CI 21.6–62.3 pg/mL), when compared with other conditions which received benefit from tocilizumab administration such as chimeric antigen receptor (CAR) T cell-induced cytokine release syndrome (difference 3074 pg/mL, 95% CI 325–26735 pg/mL; p<0·0001), or when compared with other severe conditions such as ARDS unrelated to COVID-19 (mean 460.1 pg/mL, 95% CI 216.3–978.7 pg/mL; difference 423.4 pg/mL, 95% CI 106.9–1438.1 pg/mL; p<0·0001), and sepsis (mean 983.6 pg/mL, 95% CI 550.1–1758.4 pg/mL; difference 947 pg/mL, 95% CI 324–2648 pg/mL; p<0·0001). Even in patients with hypoinflammatory ARDS, the mean IL-6 was still 5 times higher than the concentration in patients with COVID-19 [50]. Alternative mechanisms of COVID-19-induced organ dysfunction may play a part. Therefore, IL-6 may not play such a significant role in the pathogenesis of COVID-19, and inhibiting IL-6 through tocilizumab administration will not significantly alter the outcomes of COVID-19.

Our study was not without limitations. First, there was significant heterogeneity noted in our studies. One plausible rationale for this is the fact that the therapies for COVID-19 are rapidly evolving and hence the SOC differed significantly from one study to another. Moreover, the unaccounted confounders, especially in the included observational studies can also explain the heterogeneity noted in our study. Second, there was a significant variation in the follow-up of patients. Third, the studies did not consistently measure serum IL-6 and hence a correlation between IL-6 level and drug activity could not be established. Last, there was no standardization in the number of medication dosage, route of administration, and timing of administration. This can also account for the difference in outcomes noted across studies.

Despite the limitations, our study has significant strengths. First, we included a total of 38 studies with over 13 000 COVID-19 patients. This is by far the largest analysis comparing the addition of tocilizumab to the standard of care treatment. Moreover, we also included 5 recently published clinical trial studies in our analysis and performed a subgroup analysis that only consist of clinical trial studies to give more complete data regarding the benefit of tocilizumab administration in COVID-19 patients.


#

Conclusion

In conclusion, tocilizumab is not effective and failed to improve the outcome of COVID-19 patients compared with standard of care treatment, although it is relatively safe and did not cause significant serious adverse events. Our study does not support the routine use of tocilizumab for COVID-19 patients. Physicians may hence consider giving other potential agents for the treatment of COVID-19 patients, in addition to standard of care treatment. Future studies should focus more on other potential therapies besides tocilizumab.


#
#

Conflict of Interest

The authors declare that they have no conflict of interest.


Correspondence

Andree Kurniawan
Department of Internal Medicine
Faculty of Medicine
Pelita Harapan University
Boulevard Jendral Sudirman Street
Karawaci
15811 Tangerang
Indonesia   
Phone: +628158891655   

Publication History

Received: 25 November 2020

Accepted: 07 December 2020

Publication Date:
05 January 2021 (online)

© 2021. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


Zoom Image
Fig. 1 PRISMA diagram of the detailed process of selection of studies for inclusion in the systematic review and meta-analysis.
Zoom Image
Fig. 2 Forest plot that demonstrates the association of tocilizumab with the mortality a, severe COVID-19 b, length of hospital stay c, and serious adverse events d in COVID-19 infection.
Zoom Image
Fig. 3 Funnel plot analysis for mortality a, severe COVID-19 b, length of hospital stay c, and serious adverse events d outcome.