key: cord-0694632-itd6u0b5
authors: Shang, Lianhan; Lye, David Chien; Cao, Bin
title: Contemporary narrative review of treatment options for COVID‐19
date: 2021-07-08
journal: Respirology
DOI: 10.1111/resp.14106
sha: eaf5714e91ee3d303fc5533709af4a219743be4b
doc_id: 694632
cord_uid: itd6u0b5

The coronavirus disease 2019 (COVID‐19) pandemic is ongoing and many drugs have been studied in clinical trials. From a pathophysiological perspective, anti‐viral drugs may be more effective in the early stage while immunomodulators may be more effective in severe patients in later stages of infection. While drugs such as lopinavir‐ritonavir, hydroxychloroquine and azithromycin have proved to be ineffective in randomized controlled trials, corticosteroids, neutralizing monoclonal antibodies, remdesivir, tocilizumab and baricitinib have been reported to benefit certain groups of patients with COVID‐19. In this review, we will present the key clinical evidence and progress in promising COVID‐19 therapeutics, as well as summarize the experience and lessons learned from the development of the current therapeutics.

At the time of writing, it has been 1 year since the World Health Organization (WHO) announced coronavirus disease 2019 to be a global pandemic in March 2020. From a pathophysiological perspective, there are two mechanisms in COVID-19, direct viral effect and dysregulated immune response. 1, 2 Besides the respiratory tract, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was found in multiple other tissues and organs, such as heart and kidneys. 3, 4 In the early or mild stage of the disease course, the direct viral effect dominates with viral load peaking about 1 week post symptom onset. 2, 5 In the later or severe stage of the disease course, a dysfunctional immune response and hyperinflammation cause further organ damage. 6, 7 Based on this rationale, anti-viral drugs may be more effective in the early stage while immunomodulators may be more effective in severe patients in later stages of infection. Researchers have been trying to find effective treatment for the disease since the report of the first case. While drugs such as lopinavir-ritonavir, hydroxychloroquine and azithromycin have proved to be ineffective in randomized controlled trials (RCTs), corticosteroids, remdesivir, neutralizing monoclonal antibodies, Janus kinase (JAK) inhibitor baricitinib and IL-6 receptor antagonist tocilizumab have been reported to benefit certain groups of COVID-19 patients. [8] [9] [10] [11] There are other drugs with potential benefits that are still under investigations. [8] [9] [10] [11] As of 28 March 2021, there have been more than 120 million confirmed COVID-19 cases. 12 Given the huge number of cases and constant mutations of SARS-CoV-2, SARS-CoV-2 will probably continue to circulate in the future. Having an armamentarium of effective treatment including anti-viral and immunomodulatory drugs will allow us to treat critically ill COVID-19 patients even as many countries race to vaccinate their population and eventually SARS-CoV-2 may become less disruptive to societies when herd immunity is achieved in the medium term.

In this review, we will present the key clinical evidence and progress in promising COVID-19 therapeutics, as well as summarize the experience and lessons learned from the development of the therapeutics. A brief summary of the evidence on COVID-19 therapeutics is presented in Table 1 .

Remdesivir is a broad-spectrum anti-RNA-virus nucleoside analogue, which inhibits virus replication by binding to viral RNA-dependent RNA polymerase. [13] [14] [15] [16] Before COVID-19, remdesivir was previously tested in clinical trials of Ebola treatment and in laboratory studies for Middle East respiratory syndrome coronavirus (MERS-CoV) as well as other coronaviruses, but the drug was not submitted for approval by the US Food and Drug Administration (FDA) for these indications. [17] [18] [19] [20] [21] [22] Remdesivir showed potent antiviral activity against SARS-CoV-2 in vitro and in vivo, and was thus considered as a promising repurposed drug at the beginning of the COVID-19 outbreak. [23] [24] [25] However, the results of the clinical trials for remdesivir varied.

The results of four RCTs comparing remdesivir with placebo/standard care have been reported. [26] [27] [28] [29] All these trials recruited hospitalized patients and used an intravenous regimen of 200 mg remdesivir on the first day and 100 mg remdesivir on the following days (5-day and 10-day regimen vs. control in SIMPLE-2 trial; 10-day regimen in other trials). The first remdesivir trial (n = 237) enrolled severe patients and was underpowered because of rapid control of COVID-19 in China. 26 This study found no statistically significant clinical benefits in time to clinical improvement (defined as 2-point reduction on a 6-category ordinal scale; Table 2 ), 28-day mortality and other clinical outcomes. Nonetheless, in the subgroup analysis for patients admitted within 10 days post symptom onset, individuals treated by remdesivir had numerically faster clinical improvement. The SOLIDARITY trial, with 5475 patients randomized into remdesivir arm and its control, recruited hospitalized COVID-19 patients with a full spectrum of disease severity (about 24%, 67% and 9% of patients receiving no oxygen, supplementary oxygen and mechanical ventilation, respectively). 27 The SOLIDARITY trial found no significant clinical benefits in mortality, initiation of mechanical ventilation and length of hospital stay, regardless of age and respiratory support at trial entry; information on time from disease onset was unavailable in the SOLIDARITY trial. 27 The other two trials, ACTT-1 trial (n = 1062) and SIM-PLE-2 trial (n = 596), reached the prespecified primary endpoints and had positive results. 28, 29 Most of the enrolled patients (n = 957, 90.1%) in ACTT-1 trial had severe disease. 28 In ACTT-1 trial, clinical recovery was defined by meeting the criteria of score 1, 2 or 3 on the National Institute of Allergy and Infectious Diseases (NIAID) 8-category ordinal scale (Table 2 ). 28 In ACTT-1 trial, patients receiving remdesivir had a 5-day shorter median time to clinical recovery (rate ratio 1.29 [95% CI 1.12-1.49]), and lower 14day mortality (hazard ratio 0.55 [95% CI 0.36-0.83]). However, there was no difference in 28-day mortality (hazard ratio 0.73 [95% CI 0.52-1.03]). In a subgroup analysis for the primary outcome-time to clinical recovery-the rate ratio was the largest among patients receiving low-flow oxygen (score 5 on the 8-category ordinal scale) at randomization (rate ratio 1.45 [95% CI 1.18-1.79]), but there was no statistically significant improvement for patients not requiring oxygen as well as patients receiving non-invasive ventilation or high-flow oxygen (scores 4 and 6 on the 8-category ordinal scale, respectively). In subgroup analysis, the 28-day mortality was significantly reduced for the low-flow oxygen group (hazard ratio 0.30 [95% CI 0.14-0.64]). Besides, patients randomized within 10 days of symptom onset showed more significant improvement in time to clinical recovery (rate ratio 1.37 [95% CI 1.14-1.64]) than those randomized beyond 10 days after symptom onset (rate ratio 1.20 [95% CI 0. 94-1.52] ). In other secondary outcomes, remdesivir use was significantly associated with faster time to 1-point and 2-points improvement on the ordinal scale, and shorter length of hospitalization. In addition, remdesivir use was significantly associated with shorter time on oxygen, reduced progression to non-invasive mechanical ventilation or high-flow oxygen and reduced progression to mechanical ventilation.

The SIMPLE-2 trial recruited moderately ill hospitalized COVID-19 patients, who had pneumonia and peripheral oxygen saturation (SpO 2 ) >94% on room air. 29 About 84% of the recruited patients did not require oxygen therapy at baseline. For the primary outcome-the distribution of clinical status on day 11 assessed by a 7-category ordinal scale ( . Remdesivir also did not confer significant clinical benefits in other outcomes such as time to recovery and time to discontinuation of oxygen support. An additional trial (SIMPLE-1 trial) comparing 5-and 10-day remdesivir without placebo/standard control group found no significant difference in clinical effects in these two regimens for severe COVID-19 patients not requiring mechanical ventilation. 30 The difference in the results of these RCTs may stem from factors such as sample size, definitions of disease severity, enrolment criteria, locations where the trials were conducted, different protocols of standard care and definitions of the outcomes. Mortality is an outcome less likely to be influenced by designs of RCTs. Though a positive signal was indicated in the subgroup analysis of ACTT-1 trial, compared with placebo/standard care, none of the previously mentioned trials found significant clinical benefits of remdesivir on overall 28-day mortality for hospitalized patients as a whole. The WHO living network meta-analysis pooling the results of these RCTs showed remdesivir had no significant clinical benefits on mortality, need for mechanical ventilation, duration of hospital stay and time to symptom resolution/clinical improvement. 8 Given the heterogeneity of the RCTs, the certainty of the evidence in most outcomes was rated low to very low. Informed by the living network meta-analysis and considering costs and resources, the WHO living drug treatment guideline conditionally suggested against using remdesivir for hospitalized COVID-19 patients, regardless of disease severity. 8, 31 The WHO guideline panel highlighted the uncertainty of current evidence and emphasized the importance of further studies. 8, 31 Some other guidelines, such as the Infectious Diseases Society of America (IDSA), the National Institutes of Health (NIH) and the American College of Physicians (ACP), recommended using remdesivir in certain groups of COVID-19 patients. [32] [33] [34] The difference may be due to varying perspectives of considering the evidence as well as methods for pooling the data. For example, the NIH guideline recommended using remdesivir in hospitalized COVID-19 patients requiring supplementary oxygen but not on mechanical ventilation. 34 The NIH guideline panel mainly considered the subgroup analysis of ACTT-1, which provided signals that remdesivir may be effective in accelerating clinical recovery and reducing mortality for patients in their earlier stage of disease requiring low-flow oxygen. 34 Another factor that may influence the efficacy of remdesivir is the timing of using the drug. Among the four trials comparing remdesivir with placebo/standard care, three trials reported median time from symptom onset to randomization, ranging from 8 to 11 days. 26, 28, 29 In the SIMPLE-2 trial targeting moderately ill patients, the median time from symptom onset to the first dose of remdesivir was 8 days in two remdesivir groups and 9 days in the control group. 29 For anti-viral drugs, earlier use may contribute to better effects. The trial to evaluate the safety and efficacy of remdesivir in an outpatient setting for patients with COVID-19 within 7 days of symptom onset and who have at least one risk factor for disease progression is ongoing. 35 The trial may provide further evidence for the timing of drug use and population likely to benefit from remdesivir.

In conclusion, current RCTs of remdesivir focused on hospitalized patients with varying disease severity, and a pooled analysis of low/very low certainty suggested remdesivir had no significant clinical benefits for hospitalized COVID-19 patients in clinical outcomes. However, some individual studies provided positive signals among patients in the early disease stage requiring low-flow oxygen. The interpretation of the evidence may vary from different perspectives.

Neutralizing monoclonal antibodies are a class of immune molecules that can specifically recognize and bind to certain antigens. 36 Neutralizing monoclonal antibodies can be rapidly derived and massively produced, so they are considered as a promising candidate therapy against emerging infectious diseases and have been successfully used in Ebola. 18, 37 Angiotensin-converting enzyme 2 (ACE2) is the receptor for SARS-CoV-2 and the binding affinity of SARS-CoV-2 to ACE2 is much higher than SARS-CoV. 38 ,39 SARS-CoV-2 neutralizing monoclonal antibodies mainly target the spike protein (S protein) of SARS-CoV-2, which prevents the recognition and binding between SARS-CoV-2 and human ACE2 receptor, and prevents viral cell entry. 37, 40 Several neutralizing monoclonal antibodies were issued Emergency Use Authorization (EUA) by the US FDA, including bamlanivimab alone (revoked on 16 April 2021), bamlanivimab with etesevimab, and casirivimab with imdevimab. [41] [42] [43] The information of relevant RCTs is presented in Table 3 .

Bamlanivimab and etesevimab bind to different but overlapping epitopes in SARS-CoV-2 receptor-binding domain (RBD). 43 The BLAZE-1 trial is a double-blind phase II/III RCT on the efficacy and safety of bamlanivimab monotherapy or combined with etesevimab for patients with mild to moderate COVID-19 in the outpatient setting. 41, 44, 45 In the phase II part (n = 577), eligible patients were recruited within 3 days of the first positive SARS-CoV-2 antigen or PCR test. Key exclusion criteria included (1) SpO 2 ≤ 93% on room air or partial pressure of arterial oxygen/fraction of inspired oxygen (PaO 2 /FiO 2 ) < 300 mm Hg or respiratory rate ≥ 30/min or heart rate ≥ 125 bpm or receiving mechanical ventilation and (2) having positive SARS-CoV-2 serology test results. The patients were randomized into three bamlanivimab monotherapy groups of different doses (700, 2800 and 7000 mg), a bamlanivimab (2800 mg) and etesevimab (2800 mg) combination therapy group and a placebo group. Patients were first randomized into the bamlanivimab monotherapy groups or the placebo group ( Note: Data are in n/N (%), median (IQR) or effect size (95% CI). The phase III data of BLAZE-1, as well as the final analysis phase II data and phase III data of casirivimab plus imdevimab were obtained from the FDA fact sheet and press release, so the data were incomplete and might be subjected to changes in later updates of these documents or in formal publications of the studies.

Abbreviations: COVID-19, coronavirus disease 2019; FDA, Food and Drug Administration; mFAS, modified Full Analysis Set; PaO 2 /FiO 2 , partial pressure of arterial oxygen/fraction of inspired oxygen; RCT, randomized controlled trial;

SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; SpO 2 , peripheral oxygen saturation. 2020), and were then randomized into the bamlanivimab and etesevimab combination group or the placebo group (22 August 2020-3 September 2020). The interim analysis of the phase II part, which contained only the three bamlanivimab monotherapy groups and placebo group, showed that 2800 mg bamlanivimab monotherapy was associated with significantly accelerated decline of viral load in nasopharyngeal swab at day 11 (difference in mean log change À0.53 [95% CI À0.98 to À0.08]). 45 The COVID-19-related hospitalization and emergency department visit was 1.6% (5/309) in the pooled bamlanivimab group (1.0% in 700 mg group, 1.9% in 2800 mg group and 2.0% in 700 mg group) versus 6.3% (9/143) in the placebo group. For patients at high risk of disease progression, the hospitalization rate was 4.2% (pooled bamlanivimab group) versus 14.6% (placebo group). However, in the final analysis of the phase II part, which contained all five groups, only the bamlanivimab and etesevimab combination therapy group was associated with significant accelerated viral load decline compared with the placebo group (difference in mean log change À0.57 [95% CI À1.00 to À0.14]). The COVID-19-related hospitalization and emergency department visit were significantly lower in the combination therapy group (À4.9% [95% CI À8.9% to À0.8%]) and 700 mg bamlanivimab monotherapy group (À4.8% [95% CI À8.9% to À0.6%]), but not in the 2800 mg bamlanivimab monotherapy group. 44 The reason why the significant clinical benefits of 2800 mg bamlanivimab monotherapy in accelerating viral load decline in the interim analysis were not confirmed in the final analysis may include the incomplete follow-up of the placebo group in the interim analysis, as well as a higher number of placebo group patients in the final analysis. 46 The phase III data of the BLAZE-1 trial was presented in a press release and FDA fact sheet, which has not been peerreviewed. 43, 47, 48 The phase III part of BLAZE-1 trial enrolled mild to moderate COVID-19 patients at high risk of disease progression, defined as having high BMI, older age or underlying diseases, in the outpatient setting. 43, 47, 48 In the first cohort, patients were randomized to receive 2800 mg bamlanivimab and 2800 mg etesevimab combination therapy (n = 518) or placebo (n = 517). 43 , 47 The trial met its primary endpoint as the combination therapy resulted in about 70% reduction in COVID-19-related hospitalization (≥24 h of acute care) or all-cause death by day 29 (2% vs. 7%, p < 0.001). Besides, the combination therapy also resulted in significantly faster viral load decline. The second cohort of phase III BLAZE-1 trial enrolled 769 patients with the same eligibility criteria to receive 700 mg bamlanivimab and 1400 mg etesevimab combination therapy or placebo in a 2:1 ratio. 48 There was also a significant decrease in the risk of hospitalization or death (0.78% vs. 5.8%, p < 0.0001). 48 The virological and pharmacodynamic/pharmacokinetic data in the BLAZE-4 trial, an ongoing phase II trial exploring the effects of different doses of monoclonal antibody combination therapy, also showed similar effects of 700 mg bamlanivimab/1400 mg etesevimab and 2800 mg bamlanivimab/2800 mg etesevimab. 43, 47 In contrast, the ACTIV-3 trial is a double-blind RCT enrolling hospitalized COVID-19 patients within 12 days of symptom onset. 49 The patients were randomized into the 7000 mg bamlanivimab group or placebo group. The trial did not find favourable effects of bamlanivimab in terms of pulmonary ordinal outcomes at day 5, time to sustained recovery and hospital discharge. Casirivimab plus imdevimab is an antibody cocktail therapy that binds to different sites in SARS-CoV-2 S protein. 50, 51 A phase I/II double-blind RCT enrolled mild to moderate high-risk COVID-19 patients in 1:1:1 ratio to receive 2400 mg casirivimab plus imdevimab (1200 mg casirivimab and 1200 mg imdevimab, n = 266), 8000 mg casirivimab plus imdevimab (4000 mg casirivimab and 4000 mg imdevimab, n = 267) or placebo (n = 266) in the outpatient setting. 42, 52 Compared with placebo, the pooled antibody therapy resulted in a significantly faster viral load decline in nasopharyngeal swab. The largest benefits were observed in seronegative patients or patients with a high viral load at baseline. 42, 52 Lower proportion of medically attended visits was also observed in the pooled antibody therapy group (2.8% vs. 6.5%), and greater benefits were observed among patients with high risk for disease progression (3% vs. 9%). Besides, the efficacy and safety of 2400 and 8000 mg dose were similar in the phase II trial. A phase III trial enrolling more than 4000 participants found that, for high-risk COVID-19 outpatients, both 1200 mg casirivimab plus imdevimab (600 mg casirivimab and 600 mg imdevimab, 1.0% vs. 3.2%, p = 0.0024) and 2400 mg dose (1200 mg casirivimab and 1200 mg imdevimab, 1.3% vs. 4.6%, p < 0.0001) significantly reduced COVID-19-related hospitalization or death (composite outcome). 53, 54 There were several characteristics of the neutralizing monoclonal antibody trials. Current trials target outpatients within short duration of symptom onset, especially those with high risk of disease progression. The phase II trials were dose-ranging and explored the best dosing for phase III trials. The phase II trials mainly explored anti-viral effects (e.g., change in upper respiratory tract viral load) and might set secondary outcomes related to hospitalization/ medical visits. As anti-viral effects do not necessarily translate into clinically important benefits (hospitalization, death, etc.), large-scale phase III trials are needed to validate the clinical benefits in reducing hospitalization and/or death. Phase III trials and the subgroup analyses of phase II trials found larger clinical benefits in high-risk mild/moderate COVID-19 patients. Current data did not find clinical benefits of monoclonal antibodies in hospitalized patients. Therefore, the neutralizing monoclonal antibodies were only approved for EUA in mild to moderate COVID-19 patients at high risk of disease progression. [41] [42] [43] The EUA in these neutralizing monoclonal antibodies did not include hospitalized patients or patients requiring oxygen therapy for COVID-19. [41] [42] [43] Dosing is an important factor in neutralizing monoclonal antibodies. For safety and economic considerations, the best dose should be the lowest effective dose. For example, the EUA dose for bamlanivimab and etesevimab combination therapy was 700 mg bamlanivimab and 1400 mg etesevimab. At the time when EUA was issued, there were only phase III data for 2800/2800 mg combination, and FDA considered the data from in vitro and pharmacokinetic/pharmacodynamic studies, as well as the phase II data in BLAZE-4 trial. 43 The results of the second phase III cohort in BLAZE-1 trial validated the similarity of the two doses. 48 Another important issue is SARS-CoV-2 variants. The neutralizing antibody escape of some SARS-CoV-2 variants may occur under selective pressure of neutralizing monoclonal antibody therapies. 55, 56 Moreover, neutralizing monoclonal antibody therapies may have reduced effects on some naturally existing SARS-CoV-2 variants. 57 For example, while B.1.1.7 variant normally did not influence antibody neutralizing ability, B.1.351 and P.1 variants were partially resistant to casirivimab and completely resistant to bamlanivimab in vitro. 41, [57] [58] [59] In light of the markedly increasing spread of SARS-CoV-2 variants potentially resistant to bamlanivimab, the US government has halted distribution of bamlanivimab monotherapy on 24 March 2021, and the US FDA retracted the EUA of bamlanivimab monotherapy on 16 April 2021. 60, 61 The combination use of two or more potent monoclonal antibodies targeting noncompeting sites on SARS-CoV-2 may reduce the possibility of virus resistance and mutational escape. 50 43 Casirivimab plus imdevimab combination therapy retained neutralization ability on the above-mentioned variants. 42 It is worth noting that the neutralizing effects of monoclonal antibodies on mutant strains are from in vitro studies, and it is still unclear how the changes of in vitro neutralizing activity will influence therapeutic effects in clinical settings.

Current data suggested the efficacy of several neutralizing monoclonal antibody therapies for reducing viral load and preventing hospitalization and/or death among outpatients at high risk of disease progression. However, it should be noted that much of the data have not been peer reviewed. Besides, given the rapid rise of SARS-CoV-2 mutants, neutralizing antibody monotherapy may induce further mutational escape. As SARS-CoV-2 variants are ever changing, it is important to monitor the variant types and their resistance to antibody therapies. Future research should focus on the development of combination antibody therapies.

Convalescent plasma from recovered COVID-19 patients is postulated to have neutralizing antibodies and provide passive immunity. 63 Convalescent plasma was used in previous outbreak of other emerging infectious diseases, such as influenza, Ebola, SARS and MERS. 64, 65 Given the theoretical effects, rapid availability and previous experience in treatment of other infectious diseases, convalescent plasma has been used to treat COVID-19 since the beginning of the outbreak. There have been numerous observational studies as well as 10 RCTs 66-75 comparing the effects and safety of convalescent plasma and placebo/standard care.

The observational studies on convalescent plasma provided positive signals. In a prospective cohort study including 136 severe/critically ill COVID-19 patients treated by convalescent plasma and 251 matched controls, convalescent plasma treatment was associated with significantly reduced 28-day mortality for patients receiving the treatment within 72 h of hospital admission and with high-titer anti-RBD antibody in convalescent plasma. 76 However, the analysis restricted to patients receiving convalescent plasma after 72 h did not show significant mortality benefits. 76 Table 2 ) and time from randomization to discharge. 72 A meta-analysis of 10 RCTs on convalescent plasma did not find significant clinical benefits in mortality (risk ratio 1.02 [95% CI 0.92-1.12]), length of hospital stay, need for mechanical ventilation and other clinically important outcomes. 78 Besides the inherent defects of observational studies, such as poorer control of confounders, there are some other factors which may explain the negative results of RCTs, including the titers of antibodies in the convalescent plasma, the baseline antibody level in recruited patients and the patients' disease severity and stage. 71 Patients were less likely to benefit from convalescent plasma if they received convalescent plasma of low-titer antibodies, already developed antibody, were severe and in later stages of the disease when dysregulated immune response dominated. With sufficient sample size (n = 11,558) and strict quality control of convalescent plasma, the RECOVERY trial partially clarified the issue. 71 In the trial, convalescent plasma was not associated with lower 28-day mortality (rate ratio 1.00 [95% CI 0.93-1.07]), higher proportion of patients discharge within 28 days (rate ratio 0.99 [95% CI 0.94-1.03]) and lower risk of disease progression defined by a composite outcome of receiving invasive mechanical ventilation or death (risk ratio 0.99 [95% CI 0.93-1.05]). In the trial, all donors had high anti-spike IgG levels and the convalescent plasma was prepared in a uniform process, which ensured the quality and antibody titer of the plasma. Besides, the prespecified subgroup analysis on baseline SARS-CoV-2 IgG level (seropositive/seronegative) showed that, among seronegative patients, there was no significant mortality benefit (rate ratio 0.96 [95% CI 0.85-1.07]), or benefit in proportion of discharge within 28 days (rate ratio 1.08 [95% CI 0.99-1.18]), but there were significant benefits in risk of disease progression (risk ratio 0.90 [95% CI 0.83-0.98]). However, there was a 10% difference in the proportion of patients with missing antibody test result between the convalescent plasma (12%) and control groups (22%). Also, seropositive patients tended to be younger, as well as have fewer women, longer time from symptom onset to randomization and higher level of respiratory support at baseline. Therefore, the subgroup analysis can only be considered as a positive signal but not definitive.

As for the influence of disease stage and severity on convalescent plasma, the prespecified subgroup analysis in the RECOVERY trial on different levels of respiratory support at baseline did not find statistically significant benefits in any subgroups, but a trend was observed that patients receiving lower level of respiratory support at baseline tended to have numerically better point estimates of the outcomes. 71 The post hoc subgroup analysis on the time from symptom onset to randomization did not find any statistically significant benefits in the subgroups, but patients randomized within 4 days of symptom onset had a numerically lower point estimate of rate ratio in mortality (rate ratio 0.91 [95% CI 0.77-1.07]). One of the potential reasons why no statistically significant clinical benefits were found in the milder subgroup may be that the patients were still comparatively severe in the RECOVERY trial. The subgroup of patients who did not require oxygen at baseline still had a 28-day mortality of 13% (convalescent plasma group) versus 15% (control group). Besides, the subgroup of patients who were within 4 days since symptom onset at baseline had a 28-day mortality of 29% versus 31%. By contrast, an RCT enrolling 160 mild older (mean age: 76.4 years in convalescent plasma group and 77.9 years in placebo group) patients within 72 h of symptom onset found that high-titer convalescent plasma could prevent the development of severe disease. 73 In this RCT, the overall mortality of participants was as low as 3.75%. Therefore, patients with shorter time of disease onset were more likely to benefit from convalescent plasma.

Similar to neutralizing monoclonal antibody therapy, the effects of convalescent plasma may be reduced in SARS-CoV-2 variants. The convalescent plasma from patients infected with original SARS-CoV-2 strains showed modest or no reduction of neutralizing ability for B.1.1.7 variant in two in vitro experiments. 79, 80 However, B.1.351 showed marked resistance to convalescent plasma of early strains. 80, 81 The clinical data of convalescent plasma for SARS-CoV-2 variants were scarce. As B.1.1.7 variant began to rapidly spread in the UK during December, the RECOV-ERY trial did an exploratory subgroup analysis, which divided patients by enrolment before or after 1 December 2020, and did not find difference in mortality in any of the subgroups. 71 However, not all patients received sequencing to identify SARS-CoV-2 variant type in the study, so further clinical data were needed. 71 As the production of convalescent plasma was dynamic, the plasma from patients infected with SARS-CoV-2 variants will be obtained as the variants spread. It may be rational to monitor the types of SARS-CoV-2 variants of plasma donor to optimize the use of convalescent plasma. However, given the limited resources of sequencing, the surveillance may not be feasible for all donors. A specific strategy, such as random sampling, should be identified by the convalescent plasma production teams.

Current data suggest that the patients who are likely to benefit from convalescent plasma therapy are those who are in a very early stage of disease and have mild/moderate disease. As most previous RCTs enrolled severe/critical patients, or did not have enough sample size, further exploration in the above-mentioned population should be conducted. The rescue therapy of convalescent plasma for severe/critically ill is unlikely to bring clinical benefits. As the wide use of convalescent plasma is related to both the welfare of donors and recipients, the optimal indications and impact of SARS-CoV-2 variants should be carefully considered in RCTs and clinical use. 63 

Corticosteroids are a class of potent anti-inflammatory drugs widely used in clinical practice for various diseases. 82 Several meta-analyses of RCTs supported corticosteroids for patients with severe community-acquired pneumonia. [83] [84] [85] In SARS and MERS outbreak, the use of corticosteroids was frequent. There was no RCT evidence for SARS and MERS, and observational studies provided conflicting results. [86] [87] [88] Corticosteroids may curb the overwhelming inflammation in severe COVID-19 patients but may also induce immunosuppression and further aggravate the infection. Besides, the long-term and high-dose use of corticosteroids may bring adverse events such as avascular necrosis and hyperglycaemia. 89 Physicians tended to use corticosteroids for the most severe patients, often complicated with acute respiratory distress syndrome (ARDS) and/or sepsis. 90, 91 Whether corticosteroids should be used for COVID-19 have been heatedly discussed since the beginning of the outbreak. 89 , 92 The RECOVERY trial shed light on the question. 93 The RECOVERY trial randomized hospitalized patients to receive 6 mg dexamethasone for 10 days or until hospital discharge plus standard care (n = 2104) or standard care alone (n = 4321). 93 patients not requiring oxygen did not benefit from corticosteroid therapy. The clinical benefits of corticosteroids shown in the RECOVERY trial were further validated by a meta-analysis including seven RCTs, which suggested corticosteroid treatment resulted in a marked reduction of 28-day mortality for critically ill patients (OR 0.66 [95% CI 0.53-0.82]). 94 Besides, no significant safety issues were identified in the meta-analysis. Low-dose, short-term corticosteroids are recommended for severe/critically ill patients with COVID-19, except those with contraindications. 31, 33, 34, 95 There are several unanswered questions for corticosteroid therapy in COVID-19. The optimal timing for initiating corticosteroid therapy is unclear. The direct virus injury dominates at the beginning of the infection while immune injury occurs later. There is no definite cut-off between the two stages, and the best timing to initiate corticosteroids needs further exploration. In RECOVERY subgroup analysis, the mortality benefit was evident in patients receiving corticosteroids >7 days from symptom onset, which was not shown for those ≤7 days of symptom onset. 93 While the dose and duration in the RECOVERY trial are used as a reference for current treatment regimen, lower doses and shorter duration can be studied. Detailed and long-term side effects were not studied in the RECOVERY trial. Current RCTs did not report enough data on secondary infections. For severe COVID-19 patients in the intensive care unit (ICU), bacterial co-infections are not uncommon. 96, 97 Corticosteroids may increase the risk of secondary infections.

The therapeutic role of corticosteroid in severe COVID-19 patients, especially those who are mechanically ventilated, is conclusive. Further data on optimal timing, dosage and duration, type of corticosteroids and long-term side effects are needed. As it may not be ethical to conduct further RCTs about corticosteroids at this stage, these questions may be answered by large-scale observational studies.

Tocilizumab IL-6 is a proinflammatory cytokine elevated in many severe diseases such as ARDS and sepsis. 98 Elevated IL-6 level was associated with severe COVID-19 and mortality, and some prognostic models of COVID-19 included IL-6 as a biomarker. [99] [100] [101] Based on the rationale, IL-6 blockade has been a popular target for repurposed COVID-19 drugs. Tocilizumab is a recombinant humanized monoclonal IL-6 receptor antagonist previously used in rheumatoid arthritis and cytokine release syndrome induced by chimeric antigen receptor (CAR) T-cell therapy. 102, 103 The efficacy and safety of tocilizumab in COVID-19 were investigated in many observational studies and RCTs, with varying results. The main characteristics and key findings of the 10 RCTs on tocilizumab are presented in Table 4 . 11, [104] [105] [106] [107] [108] [109] [110] [111] [112] While observational studies on tocilizumab provided positive signals in decreasing mortality and preventing disease progression, only one of the five RCTs with results reported in 2020 met the primary endpoints and the meta-analysis of these five trials found no clinical benefits in mortality (Table 4 ). 113 Two RCTs reported in 2021, the REMAP-CAP and RECOVERY trials, had larger sample size and found decreased mortality and lower risk of invasive mechanical ventilation among patients receiving tocilizumab. 11, 109 There are several potential reasons contributing to the discrepancies in the results of these RCTs: (1) Study population: While most enrolled patients in these trials were hospitalized adults with severe COVID-19 needing oxygen therapy, the details of eligibility criteria varied. In some trials, the participants were required not to be on invasive mechanical ventilation before randomization and some trials set threshold of inflammatory biomarkers at baseline. Although most participants in these trials had severe COVID-19, there was huge disparity in actual disease severity, as reflected by the mortality of control groups, ranging from $5% to $30%. The participants in REMAP-CAP and RECOVERY trials were the most severe. Although most patients in the two trials received corticosteroids as standard care, the mortality was still high at about 30%. 11, 109 (2) Standard care, especially the role of corticosteroids: As the understanding of COVID-19 is evolving rapidly, the standard care in the trials has been changing with time. Corticosteroids are confirmed to be an effective treatment for severe COVID-19 patients needing oxygen therapy by RECOVERY trial and meta-analyses. 8,93, 94 The proportion of patients receiving corticosteroids as standard care ranged from <10% to >80%. The proportion of corticosteroids use is especially high in trials enrolling patients after June 2020. Besides, more than 80% participants received corticosteroids treatment in all the 3 RCTs which met the prespecified primary outcomes (EMPACTA trial, REMAP-CAP trial and RECOVERY trial). 11,108,109 A post hoc subgroup analysis of CORIMUNO trial found numerically higher clinical benefits for the combined endpoint of death or ventilatory support in patients receiving corticosteroids. 105 In the RECOVERY trial, tocilizumab did not show statistically significant benefits in mortality (rate ratio 1. 16 11 The subgroup analysis suggested that the clinical benefits of tocilizumab were additional to corticosteroids.

(3) Sample size: Several RCTs had sample size not exceeding 200 patients and may not have enough power to detect differences in outcomes.

The meta-analyses in the Australian guidelines for the clinical care of people with COVID-19 (version 39.0) included all the above-mentioned 10 RCTs. 114 outcomes, as some of the largest trials (e.g., RECOVERY trial) did not provide sufficient safety data. As the efficacy of anti-viral drugs (e.g., remdesivir and neutralizing monoclonal antibodies) in severe COVID-19 patients was not yet determined, the use of double immunosuppressants should be carefully considered to reduce the risk of serious infections and prolonged viral replication. The eligibility criteria of the RECOVERY and REMAP-CAP trials provided clues that tocilizumab may work best in hospitalized patients requiring oxygen, with evidence of systemic inflammation, or within 24 h of receiving mechanical ventilation. 11, 109 Current guidelines recommend use of tocilizumab in such populations. 33, 34, 95, 114 Dosing is another question. Although most RCTs adopted 8 mg/kg with maximum of 800 mg dose, the use of second dose varied across these studies. Some studies used only a single dose while some studies relied on physicians' judgement on the need for a second dose. The optimal number of doses and the criteria of giving more than one dose are still not determined. Another question is whether the benefit of tocilizumab can be extended to other IL-6 receptor antagonists. Current RCT data are mainly on tocilizumab while data on sarilumab and other IL-6 receptor antagonists are limited and not supportive. 115 Individual patient-level data meta-analysis of these trials may provide further insights on suitable patients and optimal treatment protocol. A WHO collaborating group which includes principal investigators of many tocilizumab RCTs has initiated such a meta-analysis project. 116 The prospective meta-analysis project set a variety of pre-specified subgroup analyses, such as baseline factors (e.g., age, sex, race and disease severity), dose of therapy and use of corticosteroids.

The JAKs are a class of non-receptor, intracellular tyrosine kinases, which serves as downstream linkers in the signal cascades of various cytokines. [117] [118] [119] [120] [121] JAKs bind to the cytoplasmic domains of type I and type II cytokine receptors and activate signal transducers and activators of transcription (STATs) to transfer extracellular signals. [117] [118] [119] [120] [121] [122] In humans, JAK family is found to consist of JAK1, JAK2, JAK3 and tyrosine kinase 2 (TYK2). [117] [118] [119] [120] [121] [122] Many cytokines related to COVID-19 immune pathogenesis, such as IL-2, IL-6 and interferon-γ, activated JAK-STAT pathway and participated in immune responses. [121] [122] [123] Theoretically, for severe hospitalized COVID-19 patients' immune dysregulation, inhibiting JAK-STAT pathway with JAK inhibitors may curb the cytokine storm and produce clinical benefits.

Baricitinib is an oral JAK inhibitor targeting both JAK1 and JAK2, and used to treat rheumatoid arthritis. 118, 124 For COVID-19 patients, baricitinib may prevent the overwhelming immune damage mediated by proinflammatory cytokines. 121 Besides, baricitinib may prevent the cell entry of SARS-CoV-2 by inhibiting AP2-associated protein kinase 1 (AAK1). 125 Data are presented in the form of n/N (%), mean AE SD or median (IQR), and the comparisons were tocilizumab group versus control group. The unit of data was converted when necessary (e.g., CRP in mg/dl was converted to mg/L).

c In REMAP-CAP trial, only the outcomes in the tocilizumab group and the control group were presented in the table. Data of tocilizumab and sarilumab were not combined.

moderately ill hospitalized COVID-19 patients to receive baricitinib (4 mg per day) in addition to lopinavir-ritonavir for 2 weeks. 128 A historical control group treated by lopinavir-ritonavir and hydroxychloroquine included the last 12 consecutive patients hospitalized before the date of patients treated by baricitinib. 128 In the study, baricitinib group showed significant improvement in outcomes such as fever, SpO 2 , PaO 2 /FiO 2 , C-reactive protein (CRP), ICU admission and discharge rate. This study was among the first studies on the efficacy of baricitinib and showed positive results. However, due to the non-randomized design, small sample size and poor selection of controls, the results are not conclusive and can only be hypothesis-generating. A multicenter retrospective study comparing the effects of baricitinib plus lopinavir-ritonavir (n = 113) versus hydroxychloroquine plus lopinavir-ritonavir (n = 78) showed that the baricitinib group had lower ICU admission and mortality, as well as significant decline in CRP and IL-6. 129 A prospective cohort study enrolled moderate to severe COVID-19 patients and compared the effects of baricitinib plus corticosteroids (n = 62) versus corticosteroids alone (n = 50). The study showed that the baricitinib group had better improvement in respiratory function (measured by SpO 2 /FiO 2 ) from hospital admission to discharge, as well as lower proportion of patients requiring supplementary oxygen at discharge and 1 month after discharge. 130 Several other studies also indicated potential clinical benefits of baricitinib in mitigating the hyperinflammatory status of COVID-19 patients. 123, 131, 132 The results of two RCTs are available. 133, 134 The ACTT-2 trial is a multicenter double-blind, placebocontrolled randomized trial conducted in 67 centers across eight countries. 133 In ACTT-2 trial, 1033 hospitalized patients were randomized to receive baricitinib plus remdesivir (n = 515) or placebo plus remdesivir (n = 518). The clinical status in ACTT-2 trial was measured by the NIAID 8-category ordinal scale used in ACTT-1 trial ( Table 2) . Patients with scores 4, 5, 6 and 7 at baseline accounted for 13.7%, 54.6%, 20.9% and 10.7% of total patients, respectively. The dosage of baricitinib was 4 mg per day for (2 mg per day for patients with estimated glomerular filtration rate < 60 ml/min/1.73 m 2 ) 14 days or until discharge. The trial met its primary endpoint and observed significantly shorter time to recovery (defined by meeting the criteria of scores 1, 2 or 3 on the NIAID 8-category ordinal scale) in baricitinib group (median 7 vs. 8 days, rate ratio 1.16 [95% CI 1.01-1.32]). Prespecified subgroup analysis on baseline score on the 8-category ordinal scale showed that the greatest benefits were observed in patients requiring non-invasive ventilation or high-flow oxygen (i.e., score 6 on the ordinal scale) at baseline (median 10 vs. 18 days, rate ratio 1.51 [95% CI 1.10-2.08]). While no statistically significant benefits in time to recovery were observed in other subgroups (scores 4, 5 and 7), patients requiring lowflow oxygen (score 5) showed the best point estimate for primary outcome (rate ratio 1.17 [95% CI 0.98-1.39]). Besides, significantly higher odds of improvement in clinical status (measured by NIAID 8-category ordinal scale) at day 15 were observed in baricitinib group (OR 1.3 [95% CI 1.0-1.6]). The largest effect was also observed among patients receiving non-invasive ventilation or high-flow oxygen at baseline. Numerically, lower 28-day mortality was observed in baricitinib group overall (5.1% vs. 7.8%, hazard ratio 0.65 [95% CI 0.39-1.09]) and more marked in low-flow and high-flow oxygen subgroups.

The COV-BARRIER was a multicenter double-blind, placebo-controlled randomized trial conducted in 101 centers across 12 countries. 134 The COV-BARRIER trial enrolled laboratory-confirmed, hospitalized COVID-19 patients with pneumonia and ≥1 elevated inflammatory marker (CRP, D-dimer, lactate dehydrogenase or ferritin). 134 The COV-BARRIER trial also used NIAID 8-category ordinal scale to assess the clinical status of patients (Table 2 ). 134 Initially, the trial enrolled patients score 4-6 in the ordinal scale at baseline, and excluded patients requiring invasive machinal ventilation (score 7 in the ordinal scale). In a protocol amendment during the trial, hospitalized patients not requiring oxygen therapy (score 4 in the ordinal scale) at baseline were excluded and the trial focused on patients requiring oxygen therapy (scores 5 and 6 in the ordinal scale) subsequently. Overall, enrolled patients (n = 1525) with scores 4, 5 and 6 in the ordinal scale at baseline were 12.3%, 63.4% and 24.4%, respectively. The primary outcome in the trial was the proportion of patients with disease progression defined by receiving non-invasive ventilation or high-flow oxygen, invasive mechanical ventilation or extracorporeal membrane oxygenation (ECMO) or death (i.e., score 6-8 on the ordinal scale). No statistically significant difference in the primary outcome was found between baricitinib and control groups ( . While no statistically significant mortality benefits were found in patients with score 4 or 5 separately, the largest mortality benefits were shown in patients requiring non-invasive ventilation or high-flow oxygen (score 6) at baseline (17.5% vs. 29.4%, hazard ratio 0.52 [95% CI 0.33-0.80]). A secondary analysis focused on patients requiring oxygen (scoring 5 and 6) also found significant benefits in mortality (5.2% vs. 14.7%, hazard ratio 0.31 [95% CI 0.11-0.88]). Prespecified subgroup analysis on baseline concomitant drug use found significant mortality benefits in patients receiving corticosteroids, not receiving corticosteroids and not receiving remdesivir. For the subgroup of patients receiving remdesivir, a numerical mortality reduction was observed. The definition of time to recovery outcome in the COV-BARRIER trial was the same as the ACTT-2 trial, but no significant benefits were observed (5.2% vs. 14.7%, rate ratio 1.11 [95% CI 0.99-1.24]).

In light of the benefits observed in the ACTT-2 trial, FDA issued EUA on baricitinib and remdesivir combination treatment for hospitalized patients requiring oxygen therapy. 135, 136 However, knowledge gaps and clinical practice issues still exist. First, the exact benefits of baricitinib have not yet been well defined. The ACTT-2 trial found baricitinib was beneficial in time to recovery but not in mortality, while the COV-BARRIER trial reported the converse. Differences in study population, concomitant drugs and definitions of outcomes may contribute to the discrepancy in the results of ACTT-2 and COV-BARRIER trials. The primary outcome is the most important outcome selected for an RCT, and the sample size and power calculation are based on the primary outcome. 137 The analysis on secondary outcomes may have increased risk of false-negative as well as false-positive errors because of factors such as insufficient sample size and multiple comparisons. [137] [138] [139] [140] As mortality was not the primary outcome in ACTT-2 trial, it was not powered to detect the difference in mortality. Besides, the clinical significance of time to recovery with absence of mortality benefits makes baricitinib less useful. In the COV-BARRIER trial, mortality was also not the primary endpoint. However, although the sample size of over 1500 patients was large and the authors used graphical testing procedure to control for type I error of multiple comparisons, as the trial failed to meet its primary endpoint, the confidence in the mortality benefits has been questioned. [137] [138] [139] [140] Nevertheless, a meta-analysis of ACTT-2 and COV-BARRIER trials indicated potential benefits in 28-day mortality (risk ratio 0.63 [95% CI 0.48-0.81]) and risk of invasive mechanical ventilation (risk ratio 0.66 [95% CI 0.46-0.93]). 114 Later trials with larger sample size, such as the ongoing baricitinib arm of the RECOVERY trial, are expected to provide further information on the role of baricitinib.

The second question is the role of baricitinib in clinical practice. As corticosteroids are cheap and widely available, with mortality benefits for hospitalized patients requiring oxygen therapy, physicians may not choose other immune modulators unless corticosteroids are contraindicated. Besides, the ACTT-4 trial, a double-blind RCT comparing the effects of baricitinib and dexamethasone in a head-to-head comparison, was closed prematurely as interim analysis showed that there was unlikely to be statistically significant differences between the two groups in the primary outcome of progression to mechanical ventilation or death. 141 Therefore, similar to tocilizumab, baricitinib may more likely be considered as an addition to corticosteroids but not a replacement. Similar to REMAP-CAP and RECOV-ERY trials for tocilizumab, about 80% of patients in the COV-BARRIER trial received corticosteroids, which suggested its benefits may be additional to corticosteroids. 134 Notably, in both trials, baricitinib showed the greatest benefit in patients requiring non-invasive ventilation or high-flow oxygen therapy, followed by patients requiring low-flow oxygen. 133, 134 Similar to corticosteroids, baricitinib may not bring benefits for patients not requiring oxygen therapy, as these patients may not have severe inflammatory immune responses and use of immunomodulators may hamper virus clearance. The clinical benefits in patients requiring invasive mechanical ventilation were unclear. In the subgroup analysis of ACTT-2 primary outcome, baricitinib did not show significant benefits in patients requiring invasive mechanical ventilation (rate ratio 1.08 [95% CI 0.59-1.97]). In light of this result, the published cohort of COV-BARRIER trial did not enrol these patients. However, as the sample size of this subgroup in the ACTT-2 trial was only 111, further study may be needed.

Other JAK inhibitors such as ruxolitinib and tofacitinib may have a role in treating severe COVID-19 patients. For ruxolitinib, there was only one available pilot RCT with very small sample size (n = 43) on its efficacy for severe COVID-19 patients. 142 The trial did not find statistically significant difference between ruxolitinib and placebo groups in the primary outcome of time to clinical improvement as well as most secondary outcomes such as 28-day mortality and virus clearance time. 142 The most important finding in the trial was that patients in the ruxolitinib group had faster computed tomography improvement and recovery from lymphopenia. 142 No RCT data about tofacitinib are available. As current evidence about other JAK inhibitors is very limited and not adequately powered, it is questionable whether they can be interchangeable for baricitinib. Further RCTs are ongoing and expected to assess the efficacy of ruxolitinib and tofacitinib in the future. 143 In summary, baricitinib may be effective in hospitalized COVID-19 patients requiring oxygen therapy, especially patients receiving non-invasive ventilation or high-flow oxygen therapy. However, given the mixed results of the two available RCTs, the exact benefits of baricitinib need to be confirmed by further studies.

Within 1 year of the pandemic, many drugs have been studied. Therapeutic drug trials during the COVID-19 pandemic provided us with very valuable experience to better confront challenges of other possible emerging infectious diseases in the future.

A large number of clinical trials were registered and conducted, both for novel and repurposed drugs for COVID-19. However, the results of many seemingly promising drugs were unsatisfying. For example, hydroxychloroquine is an antimalarial and anti-inflammatory drug. Although hydroxychloroquine was reported to have in vitro anti-viral effects on SARS-CoV-2, [144] [145] [146] RCTs and meta-analyses showed that hydroxychloroquine had no clinical benefits in reducing mortality, accelerating clinical improvement, lowering length of hospital stay, preventing SARS-CoV-2 infection and other important clinical outcomes, and may increase the incidence of adverse events. 8, 9, 147, 148 Before the conclusive results of RCTs were published, the off-label use of hydroxychloroquine was common for treatment or prophylaxis during the early stage of the pandemic. [149] [150] [151] Similar phenomenon also existed for other drugs, which may be widely used before efficacy was confirmed. As in vitro anti-viral effects do not necessarily translate into clinical effects and off-label use of drugs may cause adverse events, it is necessary to interpret the results of cell culture and animal experiments as well as observational studies cautiously, and evaluate the efficacy and safety of candidate drugs with rigorously designed RCTs. 152 Largescale off-label use of seemingly promising drugs may obscure the actual efficacy of the drugs and hamper the enrolment of RCTs. 152 Large RCTs such as the RECOVERY and SOLIDARITY trials have provided valuable models for pandemic research in the future. 11, 27, 93 It is important to coordinate the research resources of affected regions to maximize sample sizes of RCTs. RCTs with small sample sizes can only provide signals but not conclusive information. They should be discouraged in favour of large collaborative multicenter RCTs. The eligibility criteria of the RCTs are also important. For example, the best timing of anti-viral drugs should be in the early stage of disease. However, the early RCTs of convalescent plasma focused on severe/critically ill patients, which may not reflect the actual therapeutic effects. Observational studies can provide valuable clues for the selection of eligible study populations of RCTs. In addition, the selection of outcome measures should reflect the likely clinical efficacy of the drug, as well as account for pathophysiology and disease stage/severity of the study population. For severe patients, the most important outcome is mortality. However, for patients with mild COVID-19, the risk of hospitalization and progression to severe disease may be equally important. 153 Especially for areas with limited medical resources, lowering hospitalization rate is of great public health significance. 153 For better comparison and synthesis of the results of different RCTs, researchers should consider including the outcomes in the core outcome sets for COVID-19. [154] [155] [156] The working group of the WHO recommended a minimal common outcome measure set for COVID-19 clinical research, including mortality, viral load and clinical course (progression and recovery). 154 As the evidence of COVID-19 therapeutics is rapidly evolving, the importance of living meta-analysis has emerged. There are several big living network meta-analysis programmes with focus on different clinical outcomes. 8, 157 However, given the heterogeneity of the RCTs (e.g., standard care, medical resources and eligibility criteria), the results of meta-analyses should be interpreted cautiously. Global collaboration among study groups and individual patient-level meta-analysis may help better interpret the study results, as it will be feasible to conduct subgroup analyses on disease severity and the effect of concomitant drugs.

In conclusion, after 1 year, current evidence indicates that neutralizing monoclonal antibodies can reduce the risk of hospitalization and disease progression in high-risk outpatients while corticosteroids provide mortality benefits in severe COVID-19 patients requiring oxygen therapy.

Remdesivir improves clinical recovery and may reduce mortality if given early especially in patients on low-flow oxygen, while baricitinib improves recovery and may reduce mortality in patients on high-flow oxygen. Tocilizumab may provide additional benefits in critically ill mechanically ventilated patients on corticosteroids. For this and future pandemics, it is important to conduct RCTs with large sample sizes to test the efficacy and safety of potential therapeutic drugs. The eligibility criteria, primary and secondary outcomes should be carefully selected based on the pathophysiology of the infection and pharmacology of the potential therapeutic drugs to best assess clinical efficacy.

Dr David Chien Lye declared to be a member of Gilead global advisory panel on remdesivir 2020 (no honorarium accepted).

David Chien Lye https://orcid.org/0000-0003-0324-0205 Bin Cao https://orcid.org/0000-0001-6991-0350

Dr Lianhan Shang is a clinical researcher in respiratory medicine, with his research conducted in China-Japan Friendship Hospital. His research interests mainly include clinical research on respiratory medicine, especially respiratory infectious diseases, as well as evidence synthesis study and clinical research methodology. He has published eight peer-reviewed articles and reviews.

Professor David Chien Lye is an infectious disease physician with research interests in COVID-19, dengue and antimicrobial resistance. In particular, he is active in RCTs in optimizing treatment for infectious diseases. He is director of research and training at National Centre for Infectious Diseases, senior consultant at Tan 

COVID-19 and the heart

Pathophysiology, transmission, diagnosis, and treatment of coronavirus disease 2019 (COVID-19): a review

The Covid-Pathology Team. Autopsy of COVID-19 patients in China

Extrapulmonary manifestations of COVID-19

SARS-CoV-2, SARS-CoV, and MERS-CoV viral load dynamics, duration of viral shedding, and infectiousness: a systematic review and meta-analysis

Immunopathogenesis and treatment of cytokine storm in COVID-19

Viral presence and immunopathology in patients with lethal COVID-19: a prospective autopsy cohort study

Drug treatments for covid-19: living systematic review and network meta-analysis

Effect of hydroxychloroquine in hospitalized patients with Covid-19

Azithromycin in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial

Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial

World Health Organization. WHO Coronavirus (COVID-19) Dashboard

Mechanism of SARS-CoV-2 polymerase stalling by remdesivir

The antiviral compound remdesivir potently inhibits RNA-dependent RNA polymerase from Middle East respiratory syndrome coronavirus

Coronavirus susceptibility to the antiviral remdesivir (GS-5734) is mediated by the viral polymerase and the proofreading exoribonuclease. mBio

Remdesivir is a direct-acting antiviral that inhibits RNA-dependent RNA polymerase from severe acute respiratory syndrome coronavirus 2 with high potency

PREVAIL IV: a randomized, double-blind, twophase, phase 2 trial of remdesivir versus placebo for reduction of Ebola virus RNA in the semen of male survivors

A randomized, controlled trial of Ebola virus disease therapeutics

Prophylactic and therapeutic remdesivir (GS-5734) treatment in the rhesus macaque model of MERS-CoV infection

Middle East respiratory syndrome (MERS), and coronavirus disease 2019 (COVID-19): a systematic review of in vitro, in vivo, and clinical trials

Broad-spectrum antiviral GS-5734 inhibits both epidemic and zoonotic coronaviruses

Therapeutic efficacy of the small molecule GS-5734 against Ebola virus in rhesus monkeys

Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro

Remdesivir is a delayed translocation inhibitor of SARS-CoV-2 replication

Clinical benefit of remdesivir in rhesus macaques infected with SARS-CoV-2

Remdesivir in adults with severe COVID-19: a randomised, double-blind, placebocontrolled, multicentre trial

Repurposed antiviral drugs for Covid-19 -interim WHO Solidarity Trial results

Remdesivir for the treatment of Covid-19 -final report

Effect of remdesivir vs standard care on clinical status at 11 days in patients with moderate COVID-19: a randomized clinical trial

Remdesivir for 5 or 10 days in patients with severe Covid-19

A living WHO guideline on drugs for covid-19

Should remdesivir be used for the treatment of patients with COVID-19? Rapid, living practice points from the American College of Physicians (version 2)

Infectious Diseases Society of America Guidelines on the Treatment and Management of Patients with COVID-19. Infectious Diseases Society of America

Update). National Institutes of Health

Study to Evaluate the Efficacy and Safety of Remdesivir (GS-5734™) Treatment of Coronavirus Disease 2019 (COVID-19) in an Outpatient Setting

Neutralizing monoclonal antibodies for COVID-19 treatment and prevention

Human monoclonal antibodies: on the menu of targeted therapeutics against COVID-19

Structural basis of receptor recognition by SARS-CoV-2

Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation

Monoclonal antibodies for treating COVID-19

Fact Sheet for Health Care Providers Emergency Use Authorization (EUA) of Bamlanivimab

Fact sheet for Health Care Providers Emergency Use Authorization (EUA) of REGEN-COV™ (Casirivimab with Imdevimab)

Fact Sheet for Health Care Providers Emergency Use Authorization (EUA) of Bamlanivimab and Etesevimab

Effect of bamlanivimab as monotherapy or in combination with etesevimab on viral load in patients with mild to moderate COVID-19: a randomized clinical Trial

SARS-CoV-2 neutralizing antibody LY-CoV555 in outpatients with Covid-19

Neutralizing monoclonal antibody for mild to moderate COVID-19

New Data Show Treatment with Lilly's Neutralizing Antibodies Bamlanivimab (LY-CoV555) and Etesevimab (LY-CoV016) Together Reduced Risk of COVID-19 Hospitalizations and Death by 70 Percent

s Bamlanivimab and Etesevimab Together Reduced Hospitalizations and Death in Phase 3 Trial for Early COVID-19

A neutralizing monoclonal antibody for hospitalized patients with Covid-19

Antibody cocktail to SARS-CoV-2 spike protein prevents rapid mutational escape seen with individual antibodies

Studies in humanized mice and convalescent humans yield a SARS-CoV-2 antibody cocktail

REGN-COV2, a neutralizing antibody cocktail, in outpatients with Covid-19

Phase 3 Trial Shows REGEN-COV™ (Casirivimab with Imdevimab) Antibody Cocktail Reduced Hospitalization or Death by 70% in Non-Hospitalized COVID-19 Patients

COV-2067 Phase 3 Trial in High-Risk Outpatients Shows that REGEN-COVTM (2400 mg and 1200 mg IV doses) Significantly Reduces Risk of Hospitalization or Death While Also Shortening Symptom Duration

Escape from neutralizing antibodies by SARS-CoV-2 spike protein variants

Neutralising antibody escape of SARS-CoV-2 spike protein: risk assessment for antibody-based Covid-19 therapeutics and vaccines

Circulating SARS-CoV-2 variants B.1.1.7, 501Y.V2, and P.1 have gained ability to utilize rat and mouse Ace2 and altered in vitro sensitivity to neutralizing antibodies and ACE2-Ig

SARS-CoV-2 variants B.1.351 and B.1.1.248: escape from therapeutic antibodies and antibodies induced by infection and vaccination

V2 and 501Y.V3 variants of SARS-CoV-2 lose binding to Bamlanivimab in vitro

Update on COVID-19 Variants and Impact on Bamlanivimab Distribution

COVID-19) Update: FDA Revokes Emergency Use Authorization for Monoclonal Antibody Bamlanivimab

The dual function monoclonal antibodies VIR-7831 and VIR-7832 demonstrate potent in vitro and in vivo activity against SARS-CoV-2

Different aspects of convalescent plasma therapy for COVID-19 treatment; a critical review

Convalescent plasma against COVID-19: a broad-spectrum therapeutic approach for emerging infectious diseases

Current views on the potentials of convalescent plasma therapy (CPT) as coronavirus disease 2019 (COVID-19) treatment: a systematic review and meta-analysis based on recent studies and previous respiratory pandemics

Convalescent plasma in the management of moderate covid-19 in adults in India: open label phase II multicentre randomised controlled trial (PLACID Trial)

Randomized controlled trial of convalescent plasma therapy against standard therapy in patients with severe COVID-19 disease

Convalescent Plasma for COVID-19: A multicenter

Efficacy of convalescent plasma therapy compared to fresh frozen plasma in severely ill COVID-19 Patients: a pilot randomized controlled trial

Convalescent plasma for COVID-19. A randomized clinical trial

Convalescent plasma in patients admitted to hospital with COVID-19 (RECOVERY): a randomised controlled, open-label, platform trial

Effect of convalescent plasma therapy on time to clinical improvement in patients with severe and life-threatening COVID-19: a randomized clinical Trial

Early high-titer plasma therapy to prevent severe Covid-19 in older adults

Clinical and immunological benefits of convalescent plasma therapy in severe COVID-19: insights from a single center open label randomised control trial

A randomized trial of convalescent plasma in Covid-19 severe pneumonia

Treatment of coronavirus disease 2019 patients with convalescent plasma reveals a signal of significantly decreased mortality

Clinical efficacy of convalescent plasma therapy on treating COVID-19 patients: evidence from matched study and a meta-analysis

Association of convalescent plasma treatment with clinical outcomes in patients with COVID-19: a systematic review and meta-analysis

Reduced neutralization of SARS-CoV-2 B.1.1.7 variant by convalescent and vaccine sera

Antibody resistance of SARS-CoV-2 variants B.1.351 and B.1.1.7

SARS-CoV-2 501Y.V2 escapes neutralization by South African COVID-19 donor plasma

Use of glucocorticoids in the critical care setting: science and clinical evidence

Corticosteroid therapy for patients hospitalized with community-acquired pneumonia: a systematic review and meta-analysis

Efficacy and safety of corticosteroids for community-acquired pneumonia: a systematic review and meta-analysis

Corticosteroids for pneumonia. Cochrane Database Syst Rev

SARS: systematic review of treatment effects

Treatment of severe acute respiratory syndrome with glucosteroids: the Guangzhou experience

Corticosteroid therapy for critically ill patients with

Middle East respiratory syndrome

Clinical evidence does not support corticosteroid treatment for 2019-nCoV lung injury

Severe community-acquired pneumonia: current management and future therapeutic alternatives

The role of glucocorticoids as adjunctive treatment for sepsis in the modern era

On the use of corticosteroids for 2019-nCoV pneumonia

Dexamethasone in hospitalized patients with Covid-19

Association between administration of systemic corticosteroids and mortality among critically ill patients with COVID-19: a meta-analysis

Management of hospitalised adults with coronavirus disease-19 (COVID-19): a European Respiratory Society living guideline

Bacterial co-infection and secondary infection in patients with COVID-19: a living rapid review and meta-analysis

Incidence of co-infections and superinfections in hospitalized patients with COVID-19: a retrospective cohort study

Cytokine elevation in severe and critical COVID-19: a rapid systematic review, meta-analysis, and comparison with other inflammatory syndromes

Circulating levels of interleukin-6 and interleukin-10, but not tumor necrosis factor-alpha, as potential biomarkers of severity and mortality for COVID-19: systematic review with metaanalysis

Development and validation of an online model to predict critical COVID-19 with immune-inflammatory parameters

Interleukin-6-based mortality prediction model for COVID-19: validation and update in multicentre and second wave cohorts

New insights and long-term safety of tocilizumab in rheumatoid arthritis

Tocilizumab for the treatment of chimeric antigen receptor T cell-induced cytokine release syndrome

Tocilizumab in hospitalized patients with severe Covid-19 pneumonia

Effect of tocilizumab vs usual care in adults hospitalized with COVID-19 and moderate or severe pneumonia: a randomized clinical trial

Effect of tocilizumab vs standard care on clinical worsening in patients hospitalized with COVID-19 pneumonia: a randomized clinical Trial

Efficacy of tocilizumab in patients hospitalized with Covid-19

Tocilizumab in patients hospitalized with Covid-19 pneumonia

Interleukin-6 receptor antagonists in critically ill patients with Covid-19

Effect of tocilizumab on clinical outcomes at 15 days in patients with severe or critical coronavirus disease 2019: randomised controlled trial

Tocilizumab plus standard care versus standard care in patients in India with moderate to severe COVID-19-associated cytokine release syndrome (COVINTOC): an open-label, multicentre, randomised, controlled, phase 3 trial

Tocilizumab in patients with moderate or severe COVID-19: a randomized, controlled, open-label, multicenter trial

Efficacy and safety of tocilizumab in COVID-19 patients: a living systematic review and meta-analysis

National COVID-19 Clinical Evidence Taskforce. Australian Guidelines for the Clinical Care of People with COVID-19

Sarilumab in patients admitted to hospital with severe or critical COVID-19: a randomised, double-blind, placebo-controlled, phase 3 trial

Anti-Interleukin-6 Therapies for Hospitalized Patients with COVID-19: A Protocol for a Prospective Meta-Analysis of Randomized Trials

The JAK-STAT pathway at twenty

JAK inhibition as a therapeutic strategy for immune and inflammatory diseases

Mechanisms and consequences of Jak-STAT signaling in the immune system

JAK/STAT -emerging players in metabolism

Targeting JAK-STAT signaling to control cytokine release syndrome in COVID-19

Type I/II cytokines, JAKs, and new strategies for treating autoimmune diseases

Characterization of the cytokine storm reflects hyperinflammatory endothelial dysfunction in COVID-19

Baricitinib: a review in rheumatoid arthritis

Baricitinib as potential treatment for 2019-nCoV acute respiratory disease

Mechanism of baricitinib supports artificial intelligence-predicted testing in COVID-19 patients

COVID-19: combining antiviral and anti-inflammatory treatments

Baricitinib therapy in COVID-19: a pilot study on safety and clinical impact

Beneficial impact of baricitinib in COVID-19 moderate pneumonia; multicentre study

Baricitinib improves respiratory function in patients treated with corticosteroids for SARS-CoV-2 pneumonia: an observational cohort study

Use of baricitinib in patients with moderate to severe coronavirus disease

Baricitinib restrains the immune dysregulation in patients with severe COVID-19

Baricitinib plus remdesivir for hospitalized adults with Covid-19

Baricitinib plus standard of care for hospitalized adults with COVID-19

An EUA for baricitinib (Olumiant) for COVID-19

Fact Sheet for Healthcare Providers Emergency Use Authorization (EUA) of Baricitinib

The primary outcome measure and its importance in clinical trials

Secondary endpoints cannot be validly analyzed if the primary endpoint does not demonstrate clear statistical significance

Interpreting the results of secondary end points and subgroup analyses in clinical trials: should we lock the crazy aunt in the attic?

The primary outcome fails -what next?

National Institutes of Health. NIH Closes Enrollment in Trial Comparing COVID-19 Treatment Regimens

Ruxolitinib in treatment of severe coronavirus disease 2019 (COVID-19): a multicenter, single-blind, randomized controlled trial

Janus kinase signaling as risk factor and therapeutic target for severe SARS-CoV-2 infection

Hydroxychloroquine, a less toxic derivative of chloroquine, is effective in inhibiting SARS-CoV-2 infection in vitro

Clinical evidence for repurposing chloroquine and hydroxychloroquine as antiviral agents: a systematic review

COVID-19 prevention and treatment: a critical analysis of chloroquine and hydroxychloroquine clinical pharmacology

Prophylaxis for covid-19: living systematic review and network meta-analysis

Adverse effects of remdesivir, hydroxychloroquine, and lopinavir/ritonavir when used for COVID-19: systematic review and meta-analysis of randomized trials

Use of repurposed and adjuvant drugs in hospital patients with covid-19: multinational network cohort study

Off label, compassionate and irrational use of medicines in Covid-19 pandemic, health consequences and ethical issues

Off-label prescribing in the midst of a pandemic: the case of hydroxychloroquine

Treating COVID-19-off-label drug use, compassionate use, and randomized clinical trials during pandemics

Guidelines should not pool evidence from uncomplicated and severe COVID-19

WHO Working Group on the Clinical Characterisation and Management of COVID-19 Infection. A minimal common outcome measure set for COVID-19 clinical research

Core outcome set for clinical trials on coronavirus disease 2019 (COS-COVID). Engineering (Beijing)

Core outcome set for clinical trials of COVID-19 based on traditional Chinese and Western medicine

Interventions for treatment of COVID-19: second edition of a living systematic review with meta-analyses and trial sequential analyses (the LIVING Project)