WHO 2019 NCoV Therapeutics 2022.2 Eng

Therapeutics and COVID-19
LIVING GUIDELINE
3 MARCH 2022
Contact
WHO Emerging Diseases Clinical Assessment and Response Network
[email protected]
WHO continues to monitor the situation closely for any changes that may affect this interim guidance. Should any factors change,
WHO will issue a further update. Otherwise, this interim guidance document will expire 2 years after the date of publication.
© World Health Organization 2022. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO licence.
WHO reference number: WHO/2019-nCoV/therapeutics/2022.2
2 of 109
Therapeutics and COVID-19: living guideline - World Health Organization (WHO)
Sections
1. Summary of the guideline .................................................................................................................................................................................................... 4
2. Abbreviations .......................................................................................................................................................................................................................... 6
3. Introduction ............................................................................................................................................................................................................................ 7
4. What triggered this update and what is coming next? .................................................................................................................................................. 8
5. Understanding and applying the WHO severity definitions ........................................................................................................................................ 9
6. Recommendations for therapeutics .................................................................................................................................................................................10
6.1 Molnupiravir (published 3 March 2022) ..............................................................................................................................................................10
6.1.1 Mechanism of action ...................................................................................................................................................................................15
6.2 Janus kinase inhibitors (published 14 January 2022) .......................................................................................................................................16
6.3 Sotrovimab (published 14 January 2022) ............................................................................................................................................................27
6.4 Convalescent plasma (published 7 December 2021)........................................................................................................................................34
6.5 Casirivimab-imdevimab (neutralizing monoclonal antibodies) (published 24 September 2021) ...........................................................42
6.6 IL-6 receptor blockers (published 6 July 2021) ..................................................................................................................................................53
6.7 Ivermectin (published 31 March 2021) ................................................................................................................................................................60
6.8 Hydroxychloroquine (published 17 December 2020) ......................................................................................................................................65
6.9 Lopinavir/ritonavir (published 17 December 2020)..........................................................................................................................................70
6.10 Remdesivir (published 20 November 2020) .....................................................................................................................................................74
6.10.1 Mechanism of action .................................................................................................................................................................................78
6.11 Systemic corticosteroids (published 2 September 2020) ..............................................................................................................................79
7. Methods: how this guideline was created ......................................................................................................................................................................88
8. How to access and use this guideline ..............................................................................................................................................................................92
9. Uncertainties, emerging evidence and future research ...............................................................................................................................................94
10. Authorship, contributions, acknowledgements ..........................................................................................................................................................97
References ............................................................................................................................................................................................................................... 101
3 of 109
1. Summary of the guideline
Clinical question: What is the role of drugs in the treatment of patients with COVID-19?
Context: The evidence base for therapeutics for COVID-19 is evolving with numerous randomized controlled trials (RCTs) refcently
completed and underway. This update adds a new recommendation on molnupiravir in patients with non-severe COVID-19,
informed by data from 6 RCTs with 4796 patients.
New recommendations: The Guideline Development Group (GDG) suggests administering molnupiravir in patients with non-severe
illness, who are at highest risk of hospitalization with implementation of mitigation strategies to reduce potential harms. No
recommendation was made in patients with severe or critical illness as there are no data on molnupiravir for this population.
The Omicron variant has resulted in an update to the recommendations for casirivimab-imdevimab. Additional preclinical evidence
has emerged demonstrating lack of efficacy of casirivimab-imdevimab against the Omicron BA1 variant (see Mechanism of Action)
and reduced neutralizing activity of sotrovimab against the Omicron BA2 variant (see Mechanism of Action). See Section 4 for what
drugs are currently under review.
Understanding the new recommendations: When moving from evidence to recommendations, the GDG considered a combination
of evidence assessing relative benefits and harms, values and preferences, and feasibility issues. For patients with non-severe
illness, the GDG recognized that molnupiravir reduces hospitalization risk and time to symptom resolution, and may have little to no
effect on mortality. The effect on need for invasive mechanical ventilation was uncertain. The GDG also acknowledged that only
patients at the highest risk of being hospitalized are likely to derive important benefits and would want to receive molnupiravir. This
is especially true given the potential for harms associated with molnupiravir. Based on in vitro studies in human cells there is a
theoretical risk of malignancy associated with the drug. No evidence for genetic toxicity was uncovered in animals but this has not
been evaluated in long-term follow-up or human studies. There are also separate theoretical risks that molnupiravir may induce
drug resistance in the virus and/or increase genetic diversity within viral genome sequences that promote the emergence of new
variants.
The conditional recommendation for molnupiravir in non-severe illness acknowledges that it is challenging to accurately identify
those at highest risk of hospitalization, the limited availability of the drug, and that efficacy against emerging variants remains
uncertain. The absence of data on severe and critical COVID-19 prevented the GDG from making any recommendations for these
particular risk categories.
Updates to prior recommendations:
• The conditional recommendations for casirivimab-imdevimab in patients with both non-severe (for those at highest risk of
hospitalization) and severe or critical COVID-19 (for those with seronegative status) are now restricted to cases where rapid
viral genotyping is available and confirms infection with a susceptible SARS-CoV-2 variant (such as Delta). This change follows
pre-clinical evidence that casirivimab-imdevimab lacks efficacy against the Omicron BA1 variant.
• Regarding the previous conditional recommendation against remdesivir in patients with COVID-19, new trial data have
resulted in an ongoing evidence review by the GDG with an anticipated update of the recommendation in the next iteration of
this guideline.
Prior recommendations:
Recommended for patients with severe or critical COVID-19:
• a strong recommendation for systemic corticosteroids;

• a strong recommendation for interleukin-6 (IL-6) receptor blockers (tocilizumab or sarilumab), in combination with
corticosteroids;
• a strong recommendation for baricitinib as an alternative to IL-6 receptor blockers, in combination with corticosteroids.
Recommended for patients with non-severe COVID-19:
• a conditional recommendation for sotrovimab, conditional for those at highest risk of hospitalization.
Not recommended for patients with non-severe COVID-19:
• a conditional recommendation against systemic corticosteroids;

• a strong recommendation against convalescent plasma.
4 of 109
Not recommended for patients with severe and critical COVID-19:
• a recommendation against convalescent plasma, except in the context of a clinical trial;

• a conditional recommendation against ruxolitinib and tofacitinib.
Not recommended, regardless of COVID-19 disease severity:
• a strong recommendation against hydroxychloroquine;

• a strong recommendation against lopinavir/ritonavir;
• a recommendation against ivermectin, except in the context of a clinical trial.
About this guideline: This living guideline from the World Health Organization (WHO) incorporates a new recommendation on
molnupiravir for patients with non-severe COVID-19, and updates existing recommendations. The GDG typically evaluates a drug
when WHO judges sufficient evidence is available to make a recommendation. While the GDG takes an individual patient
perspective in making recommendations, it also considers resource implications, acceptability, feasibility, equity and human rights.
This guideline was developed according to standards and methods for trustworthy guidelines. It is supported by living systematic
reviews and network meta-analyses (LNMAs) (1)(2)(3).
Updates and access: This is the ninth update of the living guideline. It replaces earlier versions (2 September 2020, 20 November
2020, 17 December 2020, 31 March 2021, 6 July 2021, 24 September 2021, 7 December 2021 and 14 January 2022). The current
guideline and its earlier versions are available through the WHO website (4), the BMJ (5), and MAGICapp (online and also as PDF
outputs for readers with limited internet access). The living guideline is written, disseminated, and updated in an online platform
(MAGICapp), with a user-friendly format and easy-to-navigate structure that accommodates dynamically updated evidence and
recommendations, focusing on what is new while keeping existing recommendations updated within the guideline.
This living WHO guideline for COVID-19 treatments is related to the larger, more comprehensive guideline for COVID-19 clinical
management (6). Guidelines for the use of drugs to prevent (rather than treat) COVID-19 are published separately on the WHO
website (7) and by the BMJ (8), supported by a LNMA (9).
5 of 109
2. Abbreviations
ALT alanine aminotransferase
ARDS acute respiratory distress syndrome
CAP community-acquired pneumonia
CI confidence interval
COVID-19 coronavirus disease 2019
DOI declaration of interests
eGFR estimated glomerular filtration rate
FDA United States Food and Drug Administration
GDG guideline development group
GI gastrointestinal
GRADE Grading of Recommendations Assessment, Development and Evaluation
GRC guideline review committee
IL-6 interleukin-6
IMV invasive mechanical ventilation
JAK Janus kinase
LNMA living network meta-analysis
MAGIC Magic Evidence Ecosystem Foundation
MD mean difference
OIS optimal information size
OR odds ratio
PICO population, intervention, comparator, outcome
PMA prospective meta-analysis
RCT randomized controlled trial
RR relative risk/risk ratio
SAE serious adverse event
TACO transfusion-associated circulatory overload
TRALI transfusion-related acute lung injury
WHO World Health Organization
6 of 109
3. Introduction
Info Box
As of 20 February 2022, there have been over 418 million confirmed cases of COVID-19 (10). The pandemic has thus far claimed
approximately 5.8 million lives (10). Vaccination is having a substantial impact on case numbers and hospitalizations in a number of
high-income countries, but limitations in global access to vaccines mean that many populations remain vulnerable (10)(11). Even in
vaccinated individuals, uncertainties remain about the duration of protection and efficacy of current vaccines – and the efficacy of
existing treatments for COVID-19 – against emerging SARS-CoV-2 variants.
Taken together, there remains a need for more effective treatments for COVID-19. The COVID-19 pandemic – and the explosion of
both research and misinformation – has highlighted the need for trustworthy, accessible, and regularly updated living guidance to
place emerging findings into context and provide clear recommendations for clinical practice (12).
This living guideline responds to emerging evidence from RCTs on existing and new drug treatments for COVID-19. More than
5000 trials investigating interventions for COVID-19 have been registered or are ongoing (see Section 9 for emerging
evidence) (13). Among these are large national and international platform trials (such as RECOVERY, WHO SOLIDARITY, REMAP-
CAP, and ACTIV), which recruit large numbers of patients in many countries, with a pragmatic and adaptive design (14)(15)(16)(17).
An overview of ongoing trials is available from the Infectious Diseases Data Observatory, through their living systematic review of
COVID-19 clinical trial registrations (13) and the WHO website.
Several living network meta-analyses associated with this guideline incorporate emerging trial data and allow for analysis of
comparative effectiveness of multiple COVID-19 treatments. To inform the living guidance, we also use additional relevant evidence
on safety, prognosis, and patient values and preferences related to COVID-19 treatments. A recently updated living systematic
review of 232 risk prediction models for COVID-19 did not identify credible and applicable risk prediction tools that could inform
recommendations in this ninth version of the guideline (18).
However, existing and evolving evidence demonstrates remaining uncertainties concerning treatment effects for all outcomes of
importance to patients. There is also a need for better evidence on prognosis and values and preferences of patients with
COVID-19. Moreover, the rapidly evolving evidence landscape requires trustworthy interpretation and expeditious clinical practice
guidelines to inform clinicians and health care decision-makers.
7 of 109
4. What triggered this update and what is coming next?

This ninth version of the WHO living guideline addresses the use of molnupiravir in non-severe COVID-19. It follows the availability of
six RCTs, which were incorporated in an update to the LNMA on drug treatments for COVID-19 (2). It also includes updated
recommendations for casirivimab-imdevimab, driven by the emergence of the Omicron BA1 variant, and an ongoing evidence review by
the GDG for remdesivir with an anticipated update of the recommendation in the next iteration of the guideline.
Fig. 1 shows other therapeutics in progress for this WHO living guideline, also communicated through the WHO portal (4). Each dot
represents a week of time. In deciding which therapeutics to cover, the WHO considers multiple factors, including the extent of
available evidence to inform recommendations, and makes a judgment on whether and when additional evidence might be anticipated.
The WHO has a standing steering committee (see Section 10) to evaluate possibilities for new drug recommendations and updates to
existing drug recommendations.
Fig 1. COVID-19 therapeutics under assessment
8 of 109
5. Understanding and applying the WHO severity definitions
Info Box
This guideline applies to all patients with COVID-19. Recommendations may differ based on the severity of COVID-19, according to
WHO severity definitions (see below) (6). These definitions avoid reliance on access to health care to define patient subgroups.
WHO definitions of disease severity for COVID-19
• Critical COVID-19 – Defined by the criteria for acute respiratory distress syndrome (ARDS), sepsis, septic shock, or other
conditions that would normally require the provision of life-sustaining therapies such as mechanical ventilation (invasive or
non-invasive) or vasopressor therapy.
• Severe COVID-19 – Defined by any of:
◦ oxygen saturation < 90% on room air;

◦ signs of pneumonia;
◦ signs of severe respiratory distress (in adults, accessory muscle use, inability to complete full sentences, respiratory rate > 30
breaths per minute; and, in children, very severe chest wall in-drawing, grunting, central cyanosis, or presence of any other
general danger signs including inability to breastfeed or drink, lethargy, convulsions or reduced level of consciousness).
• Non-severe COVID-19 – Defined as the absence of any criteria for severe or critical COVID-19.
Caution: The GDG noted that the oxygen saturation threshold of 90% to define severe COVID-19 was arbitrary, and should be
interpreted cautiously when defining disease severity. For example, clinicians must use their judgment to determine whether a low
oxygen saturation is a sign of severity or is normal for a given patient with chronic lung disease. Similarly, clinicians may interpret a
saturation of 90–94% on room air as abnormal in the patient with normal lungs, and as an early sign of severe disease in patients
with a downward clinical trajectory. Generally, in cases wherethere is doubt, the GDG suggested erring on the side of considering
disease as severe.
The infographic illustrates these three disease severity groups and key characteristics to apply in practice.
Infographic co-produced by the BMJ and MAGIC; designer Will Stahl-Timmins (see BMJ Rapid Recommendations).
9 of 109
6. Recommendations for therapeutics
6.1 Molnupiravir (published 3 March 2022)

For patients with non-severe COVID-19 (excluding pregnant and breastfeeding women, and children)
Conditional recommendation New
We suggest treatment with molnupiravir, conditional to those at highest risk of hospitalization (conditional recommendation for).
• In the absence of credible tools to predict risk for hospitalization in people infected with SARS-CoV-2, typical characteristics of
people at highest risk include those that lack COVID-19 vaccination, with older age, immunosuppresion and/or chronic diseases
(e.g. diabetes).
• The benefit will be trivial in absolute terms except in those at highest risk for hospitalization, for which the intervention should be
reserved and given early on in disease.
• The panel identified a risk beyond 10% of being hospitalized for COVID-19 to represent a threshold at which most people would
want to be treated with molnupiravir.
• The longer-term harms of molnupiravir remain unknown in the absence of clinical evidence, both for individual patients and at the
population level. These include genotoxicity, emergence of resistance, and emergence of new variants (see Mechanism of Action).
• The conditional recommendation reflects the concern for widespread treatment with molnupiravir before more safety data
become available.
• Use of molnupiravir should be accompanied by mitigation strategies such as avoiding the drug in younger adults, active
pharmacovigilance programmes, and monitoring viral polymerase and spike sequences (see Justification).
• Alternative effective treatments with different safety profiles recommended by WHO, such as neutralizing monoclonal antibodies,
like sotrovimab, may be preferable or antivirals (currently under WHO assessment) if available.
Practical Info
Route, dosage and duration: Additional considerations are available in a linked summary of practical issues (accessible here).
Here follows a brief summary of the key points:
• The recommended dose for molnupiravir is 800 mg tablet every 12 hours daily for 5 days, as per the regimen evaluated in
large trials informing the recommendation.
• Administration should be as early as possible in the time course of the disease. In the included studies, molnupiravir was
administered within 5 days of disease onset.
Evidence To Decision
Benefits and harms

In patients with non-severe COVID-19, molnupiravir probably reduces admission to hospital and time to symptom
resolution, and may reduce mortality. The effect of molnupiravir on mechanical ventilation is very uncertain. Treatment does
not increase the likelihood of adverse effects leading to drug discontinuation.
However, potential long-term harms of molnupiravir remain uncertain and a matter of concern, in the absence of clinical
data. Potential harms include emergence of resistance, and the potential harm coming from the risk of molnupiravir-induced
mutagenesis. These deliberations (see Justification section) were based on molnupiravir's mechanism of action and available
pre-clinical data (see Mechanism of Action section).
The balance between benefits and potential harms was close, but favoured treatment in the highest risk group, if
implemented with other mitigation strategies to avoid harm at individual and population level (see Mitigation strategies
section). There is a risk that monotherapy with molnupiravir (as for other antiviral monotherapies) may be associated with
emergence of drug resistance, as has been seen with other antivirals (see Mechanism of Action section).
The absolute benefits of molnupiravir on hospital admission depend on the prognosis. The GDG defined a threshold of a 6%
absolute reduction in hospital admission to represent what most patients would value as an important benefit. Molnupiravir
would exert such a benefit in patients at highest risk of hospitalization (above 10% baseline risk), such as those that lack
COVID-19 vaccination, older people, or those with immunodeficiencies and/or chronic diseases. The conditional
recommendation for the use of molnupiravir in those at highest risk reflects this threshold: 60 fewer hospitalizations per
1000 patients, and a greater anticipated absolute survival benefit, although this was not possible to quantify in the absence
of data.
10 of 109
The planned subgroup analyses could not be performed in the absence of subgroup data reported publicly or provided by
investigators.
Certainty of the Evidence

The evidence summary was informed by six trials with 4796 participants included in the LNMA, including the MOVe-OUT
study (19).
Certainty of evidence was rated as: moderate for decreased hospitalization (rated down due to serious imprecision); low for
mortality (rated down due to serious imprecision and indirectness); moderate for time to symptom resolution (rated down
due to serious risk of bias); very low for mechanical ventilation (rated down due to extremely serious imprecision and
serious risk of bias); and high for adverse effects leading to drug discontinuation.
Limitations in available empirically developed risk prediction tools for establishing patients’ risk of hospitalization represent
the major source of indirectness for which the GDG rated down the certainty of the evidence (22). In addition, the GDG felt
that there was some indirectness because of the possible emergence of variants (including Omicron) for which the
effectiveness of currently available monoclonal antibodies may be reduced.
The GDG decided against rating certainty down for imprecision for outcomes where low event rates reflected very low
baseline risks (e.g. mortality).
Preference and values

Applying the agreed values and preferences (see Section 7), the GDG inferred that almost all well-informed patients with a
low risk of hospitalization would decline molnupiravir, and only those at highest risk (e.g. unvaccinated, older, or
immunosuppressed) would choose to receive treatment.
In the absence of research evidence, in a previous survey (see recommendation for casirivimab-imdevimab), the GDG
expressed the view that most patients with a risk of hospitalization above 10%, and thus an absolute risk reduction of
approximately 6%, would choose to receive treatment, whereas most of those below that risk level would decline treatment.
A similar survey was completed by the GDG for this recommendation; the GDG expressed the view that most patients
would consider a reduction in the absolute risk of death of 3 per 1000 (increase in survivors from 995 to 998 per 1000
patients) to be important.
Resources and other considerations

Acceptability and feasibility
Molnupiravir is unlikely to be available for all individuals who, given the option, would choose to receive the treatment. This
reinforces that molnupiravir should be reserved for those at highest risk.
Obstacles to access in low- and middle-income countries (LMICs) due to cost and availability are of concern (20). Challenges
in shared decision-making and in communicating the harms versus benefits of molnupiravir may also be increased in LMICs.
For example, those with socioeconomic disadvantages tend to have less access to services, including diagnostic testing and
treatments, in the first 5 days of symptoms, and thus less access to the interventions. Therefore, if patients at highest risk
receive the intervention this may exacerbate health inequity. It is important that countries integrate the COVID-19 clinical
care pathway in the parts of the health system that may provide care for patients with non-severe COVID-19 (i.e. primary
care, community care settings).
The recommendations should provide a stimulus to engage all possible mechanisms to improve global access to the
intervention. As an example of this, on 17 December 2021, WHO published the 7th Invitation to Manufacturers of
therapeutics against COVID-19 to submit an Expression of Interest (EOI) for Product Evaluation to the WHO
Prequalification Unit, which included molnupiravir. If this evaluation demonstrates that a product and its corresponding
manufacturing (and clinical) site(s) meet WHO recommended standards, it will be included in the list of medicinal products
that are considered to be acceptable for procurement by UN organizations and others. Individual countries may formulate
their guidelines considering available resources and prioritize treatment options accordingly.
11 of 109
Access to SARS-CoV-2 diagnostics: Since this recommendation emphasizes the need to administer treatment with
molnupiravir within 5 days of symptom onset; increasing access and ensuring appropriate use of diagnostic tests is essential.
Thus, availability and use of reliable and timely COVID-19 diagnostic tests (including the use of NAAT and Ag-RDTs) is
needed to improve access to drugs, especially those targeting the early phase of disease. The appropriate use of Ag-RDTs by
individuals and trained professionals can improve early diagnosis and earlier access to clinical care, particularly in the
community and in primary health care settings. National programs should optimize their testing systems to reflect local
epidemiology, response objectives, available resources and needs of their populations.
Justification
A combination of the evidence, safety concerns based on preclinical data, values and preferences, and feasibility contributed to
the conditional recommendation for the use of molnupiravir only in patients with non-severe COVID-19 at highest risk of
hospitalization. Typical characteristics of people at highest risk include those who are unvaccinated, older people, or those with
immunodeficiencies and/or chronic diseases (e.g. diabetes).
Only a minority of patients who are at highest risk are likely to achieve sufficient benefit to compensate for the risks, and other
limitations and disadvantages of therapy. These include a lack of reliable tools to identify high-risk patients, limited availability of
the drug, and the safety concerns summarized below.
• The GDG had concerns about the risk of emergent resistance with a new antiviral deployed as monotherapy (see
Mechanism of action section). Significant uncertainty exists regarding how quickly resistance will emerge; in the absence of
sufficient clinical data, the GDG concluded large uncertainties remain.
• Concerning the risk of the drug promoting the emergence of new variants, the GDG noted that there was a low likelihood
that the drug would result in a selective pressure for a new variant; large uncertainty remains in the absence of sufficient
clinical data.
• Molnupiravir is mutagenic in mammalian cells in vitro, but there is no evidence of mutagenicity in animal models or humans.
The GDG therefore acknowledged uncertainty regarding longer term genetic toxicity and potential for malignancy
associated with molnupiravir.
• Given evidence from rat pups of an impact on growth plate thickness, molnupiravir should not be used in children. Similarly,
since molnupiravir elicited embryo-fetal lethality and teratogenicity in offspring when given to pregnant animals, it should
not be used in pregnant or breastfeeding women.
• The GDG acknowledged that spermatogenesis may also be especially prone to the mutagenic effects of molnupiravir, but
that there was uncertainty regarding the consequences to children conceived by fathers receiving or having recently
received molnupiravir.
Applicability
The applicability of this recommendation to children, breastfeeding and pregnant women, is currently uncertain, as the included
RCTs enrolled only non-pregnant adults. However, the GDG concluded that molnupiravir should not be offered to children,
breastfeeding or pregnant women with COVID-19. In addition, men planning to conceive should be oriented on the potential for
temporary genotoxic effect on sperm cell production (see Mitigation strategies section). The unknown long-term risk of
genotoxicity is likely to be higher in younger patients as compared with older patients, thus its use in younger adults not a high
risk should be avoided.
The GDG also had concerns about whether the drug would retain efficacy against emerging variants of concern such as
Omicron. While there is no molecular basis for a loss of efficacy, the GDG noted that the higher viral loads and associated
disease severity may impact the effectiveness of molnupiravir. This represents another area of uncertainty, given currently
available data did not include patients with newer variants, including Omicron (see Section 9).
Clinical Question/ PICO

Population: Patients with non-severe COVID-19
Intervention: Molnupiravir
Comparator: Standard care
12 of 109
Summary
Evidence summary
The LNMA for molnupiravir was informed by six RCTs which enrolled 4827 patients with non-severe illness in
outpatient settings; the LNMA team had access to data for 4796 patients. All RCTs were registered; none were
published in peer-reviewed journals. None of the included studies enrolled children or pregnant women. The appendix
summarizes study characteristics and risk of bias ratings, effect estimates by outcome and associated forest plots for
molnupiravir versus standard care.
For patients with non-severe COVID-19, the GRADE Summary of Findings table shows the relative and absolute effects
of molnupiravir compared with standard care for the outcomes of interest, with certainty ratings, informed by the
LNMA (3).
Subgroup analysis
Five pre-specified subgroup analyses were requested by the GDG:
1. Age: children (≤ 19 years) versus adults (20–60 years) versus older adults (≥ 60 years).
2. Severity of illness at time of treatment initiation: non-severe versus severe versus critical.
3. Time from symptom onset.
4. Serological status (seropositive versus seronegative).
5. Vaccination status (unvaccinated versus vaccinated).
Studies did not enrol children, nor patients with severe or critical illness. All studies enrolled unvaccinated individuals
with time from symptom onset < 5 days. Data regarding serological status were not reported.
Certainty of
Outcome Study results and Comparator Intervention the Evidence Plain language
Timeframe measurements Standard care Molnupiravir (Quality of summary
evidence)
Odds Ratio 0.06 6 0

Mortality (CI 95% 0 — 0.4) per 1000 per 1000 Low
Based on data from Due to serious Molnupiravir may have a
4796 participants in 6 Difference: 6 fewer per 1000 imprecision and small effect on mortality
studies. (Randomized
( CI 95% 6 fewer indirectness 1
controlled) — 4 fewer )
Mechanical
Odds Ratio 1 8 8 Very low
(CI 95% 0.02 — 59.74) per 1000 per 1000 The effect of
ventilation Due to serious
Based on data from molnupiravir on
risk of bias and
1220 participants in 1 Difference: 0 fewer per 1000 mechanical ventilation is
extremely serious
study. (Randomized very uncertain
( CI 95% 8 fewer imprecision 2
controlled)
— 317 more )
Admission to Odds Ratio 0.54 35 19

hospital (CI 95% 0.3 — 0.89) per 1000 per 1000
Moderate Molnupiravir probably
Risk in trials Based on data from
Difference: 16 fewer per Due to serious reduces hospital
4688 participants in 5
studies. (Randomized 1000 imprecision 3 admission
controlled) ( CI 95% 24 fewer
— 4 fewer )
Odds Ratio 0.54

Admission to (CI 95% 0.3 — 0.89)
60 33 Moderate Molnupiravir probably
Based on data from per 1000 per 1000 Due to serious reduces hospital
hospital
Higher risk 4688 participants in 5 imprecision 4 admission
controlled)
13 of 109
Certainty of
Timeframe measurements Standard care Molnupiravir (Quality of summary
evidence)
Difference: 27 fewer per

1000
( CI 95% 41 fewer
— 6 fewer )

Highest risk Based on data from
Difference: 43 fewer per Due to serious reduces hospital
studies. (Randomized 1000 imprecision 5 admission
— 10 fewer )
Adverse effects 0 0
leading to drug per 1000 per 1000 There is little or no
discontinuation Based on data from difference in adverse
4796 participants in 6 High
Difference: 0 fewer per 1000 effects leading to drug
studies. (Randomized discontinuation
( CI 95% 2 fewer
controlled)
— 2 more )
Time to 9 5.6
symptom Lower better (Median) (Mean)
resolution Based on data from:
Difference: MD 3.4 fewer Due to serious reduces duration of
studies. (Randomized ( CI 95% 4.8 fewer risk of bias 6 symptoms
controlled) — 1.7 fewer )
Malignancy In vitro and animal studies suggest the Very low

possibility of carcinogenesis The effect of
No human data
molnupiravir on cancer is
with long-term
uncertain
follow-up
1. Indirectness: serious. The baseline risk across the entire population is very low, meaning that any impact on mortality
will be very small. There are some people with much higher baseline risk, who are not easily identifiable. For these patients,
molnupiravir may have an important impact on mortality. Imprecision: serious. There were only 11 events total (10 in the
control arms and 1 in the molnupiravir arms).
2. Risk of bias: serious. The single trial reporting mechanical ventilation was not blinded. Imprecision: extremely serious.
Very few events, resulted in very large credible intervals that include important and unimportant effects.
3. Imprecision: serious. The upper credible interval includes a small and unimportant effect on hospitalization (4 fewer per
1000).
1000).
1000).
6. Risk of Bias: serious. All three trials were at high risk of bias for deviations from intended intervention (lack of blinding).
One trial was at high risk of bias for possible inadequate randomization concealment.
14 of 109
Mitigation strategies to address safety concerns
New
Info Box
With the safety concerns related to molnupiravir (see Mechanism of action section), the WHO recognizes the need to mitigate
risks, both for individual patients and at the population level.
The conditional recommendation takes into account one such strategy: limiting the intervention to patients that are at higher
risk of hospitalization or death. Typical characteristics of people at highest risk include those with older age, immunodeficiencies
and/or chronic diseases (e.g. diabetes) and lack of COVID-19 vaccination. See WHO recommendations for further information
on COVID-19 vaccination Strategic Advisory Group of Experts on Immunization for more details.
Other mitigation strategies include:
• Decisions around treatment with molnupiravir must be done using a shared decision-making model, ensuring the clinician is
well educated on the potential benefits and harms of therapy and able to explain these to the patient in order to make
well-informed decisions. See Practical information section.
◦ Molnupiravir should not be given to pregnant or breastfeeding women or to children. In case of doubt about pregnancy,
a pregnancy test should be performed prior to treatment initiation. If a woman of child bearing age is considered for
treatment, counselling regarding birth control during treatment and for 4 days after the last dose of molnupiravir should
be facilitated.
◦ Men planning to conceive should be oriented on the potential for temporary genotoxic effect on sperm cell production,
and those who are sexually active with females should be counselled to use birth control during treatment and for at
least 3 months after the last dose of molnupiravir (23).
◦ The unknown long-term risk of genotoxicity is likely to be higher in younger patients as compared with older patients;
thus use in younger adults who are not at high risk should be limited.
• Active sequence monitoring of SARS-CoV-2 detected in clinical respiratory samples (i.e. may include polymerase and spike)
should be arranged for patients receiving therapy, including higher risk individuals (immunocompromised).
• Pharmacovigilance: use of molnupiravir should be accompanied by a robust, active pharmacovigilance programme.
6.1.1 Mechanism of action

Molnupiravir is an orally available antiviral, which was originally designed as an influenza treatment, although not approved. The
drug inhibits replication of SARS-CoV-2 with an in vitro potency broadly, similar to remdesivir, and was re-purposed early in
development as an antiviral for SARS-CoV-2 (24)(25).
Molnupiravir is an orally available prodrug of ß-D-N4-hydroxycytidine (NHC). It is a nucleoside drug, but the mechanism of
action involves lethal mutagensis of the virus. This contrasts with chain-termination seen with other antiviral nucleoside
analogues (e.g. remdesivir and those used in HIV or HCV) (26). NHC is incorporated by the SARS-CoV-2 RdRp, instead of either
C or U nucleosides, into the genomic or subgenomic RNA during copying of the RNA template genome. The resultant NHC-
containing RNAs are then themselves used as a template for production of subsequent RNAs which are predicted to be mutated
and therefore not believed to form functional viruses (26)(27).
Molnupiravir is given orally twice daily unlike remdesivir, which is given by intravenous infusion once daily. In healthy volunteers,
molnupiravir (800mg) achieves maximum plasma concentrations of its active metabolite at 3600 ng/mL (28). This is higher than
that of remdesivir (2200 ng/mL) (29). However, the intracellular half-life of molnupiravir active metabolite is shorter in human
cell lines (3h) compared with that of remdesivir's active metabolite (35h) (28).
High doses of molnupiravir (250 mg/kg twice daily) have been shown to be effective in SARS-CoV-2-infected Syrian golden
hamsters; however, the animal plasma pharmacokinetics were not reported to benchmark against those seen in humans (30).
Evidence of antiviral activity is also available from a study in SARS-CoV-2-infected ferrets at lower doses (31). When
molnupiravir was combined with favipiravir in infected Syrian golden hamsters, the efficacy was greater than when either drug
was given alone (32).
Molnupiravir retains activity against Alpha and Beta variants in vivo (33), and the Delta and Omicron variants in vitro (34)(35).
No data are currently available demonstrating activity against the Delta or Omicron variants in vivo, and while there appears to
15 of 109
be no molecular basis for a loss of activity, there is residual uncertainty around whether a higher replication or transmission rate
may impact efficacy of the drug.
Emergence of resistance: The emergence of resistance to some nucleosides used for other viruses is varied; with some
emerging readily, and others emerging more slowly. The barrier to resistance for a given drug with a given virus is generally
considered to increase with the number of mutations that are required to emerge. Insufficient data are currently available to
ascertain how high the barrier of resistance is with SARS-CoV-2 for molnupiravir. Based on experiences with other nucleoside
antiviral drugs (some have a high barrier to resistance and some have a low barrier to resistance), the drug will place a selective
pressure for resistant mutations within an individual, with the potential to spread at a population level.
Resistance occurs through inherent variability in viral sequences that happen spontaneously as the virus replicates. Chance
variations become selected, known as selective pressure, when they confer a survival advantage in the presence of the drug.
Sometimes, there is a fitness cost to the virus and secondary mutations can subsequently be selected to restore fitness. The
major uncertainty relates to how quickly resistance will emerge rather than whether it will emerge. There may be a higher risk
of resistance in immunocompromised patients because of a longer tail of replication in this group. There may also be a higher
risk of resistance in patients with poor adherence where the virus is exposed to suboptimal drug concentrations. The rate at
which resistance emerges will be slower if drugs are given in combination because more mutations will be required to confer
resistance to multiple drugs than will be required for one drug. Of note, animal studies have also demonstrated drug
combinations to be more effective. The risk of resistance to individual patients is drug failure due to compromised efficacy. If
resistance is transmitted, there is a risk of efficacy failure at a population level and subsequent attempts to combine the drug
may be futile because of “functional monotherapy” with the partner agent. The genetic barrier to resistance cannot be
estimated without data. Non-clinical and/or clinical data are therefore needed, and are not currently available for molnupiravir.
Emergence of new variants: It has been proposed that random mutagenesis arising from the molnupiravir mechanism of action
might increase diversity in the viral sequences that may result in more rapid emergence of new variants (36). Unlike in the
considerations for resistance, there is no conceptual basis for molnupiravir placing a selective pressure on emergence of new
variants. Sequence variation is lower given molnupiravir is only incorporated in place of two of the four nucleotide bases in the
genome than it would be if incorporated in place of any nucleotide. There is no direct evidence to support or refute the variants
hypothesis and as such the risk is currently unquantifiable.
The rate of resistance emergence and the risk of additional diversity in the viral genome leading to new variants, were
acknowledged to be higher with a higher number of patients receiving the intervention.
Non-clinical safety: The GDG reviewed the publically available data on non-clinical safety of molnupiravir from the FDA
meeting documents for molnupiravir Emergence Use Authorization (30 November 2021) (37). The following safety concerns
were highlighted:
• Genetic toxicology data demonstrated that molnupiravir is mutagenic in vitro, but there was no evidence of mutagenicity in
animal models. The GDG acknowledged uncertainties in the available data and concluded that based upon the available
information molnupiravir may or may not be carcinogenic in humans.
• An increase in thickness of growth plate associated with decreased bone formation was observed in rapidly growing rats
but not in mice, rats or dogs. The GDG determined that molnupiravir should not therefore be administered to paediatric
patients.
• Importantly, low concentrations of NHC (0.09% maternal exposures) were detectable in 10-day old rat pups suggesting that
NHC is present in breast milk. The GDG determined molnupiravir should not be administered to breastfeeding women.
• In developmental and reproductive toxicology assessments, reduced foetal body weights were observed in rats and rabbits,
with higher exposures also being associated with embryo-foetal lethality and teratogenicity in rats. Accordingly,
molnupiravir should not be administered during pregnancy.
• There was an absence of available data relating to spermatogenesis, which may be particularly prone to the effect of a
mutagen in adult males. No data are available to quantify the consequences of this for embryo/foetus conceived by fathers
who were receiving or had recently received molnupiravir.
16 of 109
6.2 Janus kinase inhibitors (published 14 January 2022)
Info Box
Recommendations concerning janus kinase (JAK) inhibitors, specifically baricitinib, ruxolitinib and tofacitinib, for patients with
severe or critical COVID-19 were published on 14 January 2022 as the eighth version of the WHO living guideline and in the
BMJ as Rapid Recommendations. It follows the availability of three RCTs for baricitinib, two RCTs for ruxolitinib, and one RCT
for tofacitinib, as per the LNMA on drug therapies (3). No changes were made for the JAK inhibitors recommendations in this
ninth version of the guideline.
Baricitinib, for patients with severe or critical COVID-19
Strong recommendation for
We recommend treatment with baricitinib (strong recommendation for).
• Along with baricitinib, corticosteroids should also be administered in patients with severe or critical COVID-19 (see Section 6.11).
• IL-6 receptor blockers (tocilizumab or sarilumab) have previously been recommended for the treatment of patients with severe or
critical COVID-19 (see Section 6.6). An IL-6 receptor blocker and baricitinib should not be given together, and should be viewed as
alternatives. The choice of whether to use baricitinib or an IL-6 receptor blocker depends on availability, as well as clinical and
contextual factors (see Justification).
Practical Info
Additional considerations are available in a summary of practical issues. Useful information can also be found in the United
States Food and Drug Administration (FDA) fact sheet for health care providers, based on the emergency use authorization
(EUA) of baricitinib (38). Here follows a brief summary of key points:
Route, dosage and duration:
• The recommended dose is 4 mg daily orally in adults with eGFR ≥ 60 mL/min/1.73 m2.
• A duration of 14 days of total treatment or until hospital discharge, whichever is first. The optimal duration of treatment is
unknown, and the proposed duration reflects what was used in the trials providing evidence on treatment effects of
baricitinib.
Dose regimen adjustment:
• Patients with leukopenia, renal impairment or hepatic impairment (note: these parameters should be monitored during
treatment);
• Patients taking strong organic anion transporter 3 (OAT3) inhibitors (e.g. probenecid), there are drug interactions which
warrant dose reductions.
Timing: Baricitinib (like IL-6 receptor blockers) should be initiated at the same time as systemic corticosteroids; specific timing
during hospitalization or the course of illness is not specified.
Benefits and harms

In patients with severe or critical illness, baricitinib probably reduces mortality and duration of mechanical ventilation, and
reduces hospital length of stay. It probably results in little or no increase in serious adverse events.
Subgroup analyses were undertaken for JAK inhibitors as a class (rather than on individual drugs) and revealed no evidence
of a subgroup effect on relative risk in younger (< 70 years) versus older patients; those with critical versus severe
COVID-19; those receiving and not receiving corticosteroids at baseline; and those receiving and not receiving remdesivir at
baseline.
17 of 109

Certainty of evidence was rated as: moderate for decreased mortality (rated down from high for a large ongoing trial that
may change estimates of effect and indirectness due to the relatively short follow-up period close to 28 days possibly being
insufficient to capture all relevant events); high for reduction in hospital length of stay; moderate for reduction in duration of
mechanical ventilation and for little or no increase in serious adverse events, both rated down for serious imprecision; and
low for need for mechanical ventilation, rated down for very serious imprecision.
The GDG noted in particular that the risk of serious infections (bacterial and fungal) may vary considerably in different parts
of the world according to the background prevalence of infections (such as tuberculosis). This may not be so important given
the short course of baricitinib used for treatment of COVID-19, but evidence is limited given the limited geographic spread
of the included trials and short follow-up periods.

Applying the agreed upon values and preferences (see Section 7), the GDG inferred that almost all well-informed patients
with severe or critical COVID-19 would want to receive baricitinib due to the likely reduction in mortality, and moderate
certainty evidence of little or no increase in serious adverse events. The benefit of baricitinib on mortality was deemed of
critical importance to patients and the GDG was reassured by the moderate certainty evidence of little or no increase in
serious adverse events. The GDG anticipated little variation in values and preferences between patients for this
intervention.

Resource implications, equity and human rights
Compared with some other candidate treatments for COVID-19, baricitinib is expensive. The recommendation does not
take account of cost-effectiveness. Access to these drugs is challenging in many parts of the world, and, without concerted
effort, is likely to remain so, especially in resource-poor areas. It is therefore possible that this strong recommendation could
exacerbate health inequity. On the other hand, given the demonstrated benefits for patients, it should also provide a
stimulus to engage all possible mechanisms to improve global access to these treatments. Individual countries may
formulate their guidelines considering available resources and prioritize treatment options accordingly. On 17 December
2021, WHO published the 7th Invitation to Manufacturers of therapeutics against COVID-19 to submit an Expression of
Interest (EOI) for Product Evaluation to the WHO Prequalification Unit, which includes baricitinib.
At a time of drug shortage, it may be necessary to prioritize use of baricitinib through clinical triage (6) such as prioritizing
patients with the highest baseline risk for mortality (e.g. those with critical disease over those with severe disease), in whom
the absolute benefit of treatment is therefore greatest. Other suggestions for prioritization, which lack direct evidence,
include focusing on patients with an actively deteriorating clinical course, and avoiding baricitinib in those with established
multi-organ failure (in whom the benefit is likely to be smaller).
As baricitinib is administered orally once daily, hospitalized patients should find it easy to accept this treatment. In patients
who cannot swallow tablets, baricitinib can be crushed, dispersed in water, and given via a nasogastric tube (see Practical
info).
Justification
When moving from evidence to the strong recommendation to use baricitinib in patients with severe or critical COVID-19, the
GDG emphasized the benefits on survival and decreased length of hospital stay with ease of administration and the likelihood of
little or no serious adverse events attributable to the drug. The GDG acknowledged that some serious adverse events, such as
fungal infections, may not have been accurately captured during the relatively short follow-up period in the included trials.
Because of different mechanisms of action, the GDG considered baricitinib separately from other JAK inhibitors (as outlined
below).
Costs and access were important considerations and the GDG recognizes that this recommendation could exacerbate health
inequities. This strong recommendation will provide impetus to address these concerns and maximize access across regions and
18 of 109
countries. The GDG did not anticipate important variability in patient values and preferences, and judged that other contextual
factors would not alter the recommendation (see Evidence to Decision).
Emerging evidence
The GDG carefully considered the implications of a large trial (RECOVERY) having randomized patients to baricitinib or no
baricitinib and the possibility that investigators will report their results in the relatively near future. Uncertainties regarding this
trial include the proportion of patients who also received an IL-6 receptor blocker (see below), and when the information will be
available. The GDG considered the benefits of baricitinib, supported by moderate to high certainty evidence, sufficient for an
immediate strong recommendation for use of the drug, with readiness to update the living guideline as necessary once
RECOVERY trial data are published.
The role of IL-6 receptor blockers and baricitinib
The GDG has previously made a strong recommendation for use of IL-6 receptor blockers (tocilizumab and sarilumab) in
patients with severe or critical COVID-19. The GDG carefully considered whether, on the basis of the current evidence,
baricitinib should be administered as an alternative to IL-6 receptor blockers or in addition to IL-6 receptor blockers. Combining
them may unacceptably increase harms, including secondary bacterial and fungal infections. In the absence of evidence of
incremental benefit, the GDG advises that clinicians do not administer the drugs together.
The issue then arises regarding the basis for choosing between IL-6 receptor blockers and baricitinib (see Research evidence).
The drugs have not undergone direct comparisons, and so the best evidence of their relative effects comes from indirect
comparisons generated by the LNMA that informs these guidelines. Baricitinib may reduce mortality relative to IL-6 receptor
blockers (low certainty) and may reduce the duration of mechanical ventilation (low certainty). There may be little or no
difference in the agents’ impact on mechanical ventilation (low certainty), and there is probably little or no difference in adverse
events leading to discontinuation (moderate certainty) (see Summary of Findings table in Research evidence).
The GDG felt that the low certainty evidence did not warrant a recommendation favouring the use of baricitinib versus IL-6
receptor blockers on the basis of their impact on patient-important benefits and harms. Therefore, when both agents are
available, clinicians should choose between them based on other considerations. These might include experience and comfort
using the drugs; local institutional policies; route of administration (baricitinib is oral; IL-6 receptor blockers are intravenous);
and cost.
Applicability
None of the included RCTs enrolled children, and therefore the applicability of this recommendation to children remains
uncertain. Uncertainty also remains with regard to administration of baricitinib to pregnant or lactating women. The decision
regarding use of this therapeutic should be made between the pregnant individual and their health care provider while
discussing whether the potential benefit justifies the potential risk to the mother and fetus (see Research evidence and Practical
info tabs).

Population: Patients with severe or critical COVID-19
Intervention: Baricitinib
Summary
Evidence summary
The LNMA for baricitinib was informed by three RCTs which enrolled 2659 patients across disease
severities (39)(40)(41). All RCTs were registered, and two were published in peer-reviewed journals (40)(41); one study
was a pre-print (39). All three RCTs enrolled patients in in-patient settings. None of the included studies enrolled
children or pregnant women. The Table shows characteristics of the RCTs.
For patients with severe or critical COVID-19, the GRADE Summary of Findings table shows the relative and absolute
effects of baricitinib compared with standard care for the outcomes of interest, with certainty ratings, informed by the
LNMA (1).
Baseline risk estimates
For severe and critical illness, for the critical outcome of mortality, the applied baseline risk estimate was 13% (130 in
1000). As for other related recommendations in this guideline, the estimate is derived from the SOLIDARITY trial for
severe and critical patients adjusted for treatment effects of corticosteroids. For other outcomes, we used the median of
19 of 109
the control arm of the RCTs that contributed to the evidence (see Section 7).
Subgroup analysis
Four pre-specified subgroup analyses were undertaken for JAK inhibitors as a class rather than for individual drugs:
1. Age: younger adults (< 70 years) versus older adults (≥ 70 years).
3. Concomitant use of corticosteroids at baseline.
4. Concomitant use of remdesivir at baseline.
No evidence of subgroup effects was identified on the relative risk of critical outcomes across all pre-specified effect
modifiers.
Certainty of
Timeframe measurements Standard care Baricitinib (Quality of summary
evidence)
Odds Ratio 0.62 130 85

Mortality (CI 95% 0.44 — 0.85) per 1000 per 1000 Moderate
Based on data from Due to ongoing Baricitinib probably
2659 participants in 3 Difference: 45 fewer per recruitment in a reduces mortality.
studies. (Randomized 1000 large RCT
— 17 fewer )

Mechanical per 1000 per 1000
(CI 95% 0.52 — 1.19) Low
ventilation Based on data from Due to very Baricitinib may reduce
2434 participants in 2 Difference: 21 fewer per serious mechanical ventilation.
studies. (Randomized 1000 imprecision 1
— 19 more )
Adverse effects 0 5
leading to drug per 1000 per 1000 Moderate
Baricitinib probably
discontinuation Based on data from results in little or no
Due to serious
1611 participants in 2 increase in serious
Difference: 5 more per 1000 imprecision 2
studies. (Randomized adverse effects.
( CI 95% 18 fewer
controlled)
— 28 more )
12.8 11.4
Hospital length Lower better days (Median) days (Mean)
of stay Baricitinib reduces
Based on data from:
2652 participants in 3 Difference: MD 1.4 fewer High duration of
( CI 95% 2.4 fewer hospitalization.
Duration of 14.7 11.5

mechanical days (Median) days (Mean)
Lower better Moderate Baricitinib probably
ventilation Based on data from: 328 Due to serious reduces duration of
Difference: MD 3.2 fewer
participants in 2 studies. imprecision 3 mechanical ventilation.
( CI 95% 5.9 fewer
(Randomized controlled) — 0.5 fewer )
20 of 109
Certainty of
Timeframe measurements Standard care Baricitinib (Quality of summary
evidence)
9.9 8.9
Time to clinical days (Median) days (Mean)
Lower better Low
stability Based on data from: Due to very Baricitinib may reduce
2558 participants in 2 Difference: MD 1 fewer serious time to clinical stability.
studies. (Randomized ( CI 95% 2.9 imprecision 4
controlled) fewer — 1.1
more )
1. Imprecision: very serious. Credible interval includes an important decrease and increase in mechanical ventilation.
2. Imprecision: serious. The credible interval includes an important increase in adverse effects.
3. Imprecision: serious. The credible interval includes no important difference.
4. Imprecision: very serious. Credible interval includes important harm and important benefit (using a minimal important
difference threshold of 1 day).

Population: Patients with severe and critical COVID-19
Comparator: Interleukin-6 receptor blockers
Certainty of
Comparator
Outcome Study results and Intervention the Evidence Plain language
IL-6 receptor
Timeframe measurements Baricitinib (Quality of summary
blockers
evidence)
Odds Ratio 0.77 118 96 Low

Mortality (CI 95% 0.53 — 1.1) per 1000 per 1000 Due to serious
Based on data from imprecision and Baricitinib may reduce
2659 participants in 3 Difference: 22 fewer per ongoing mortality.
studies. (Randomized 1000 recruitment in a
controlled) ( CI 95% 52 fewer large RCT 1
— 9 more )
Mechanical
(CI 95% 0.61 — 1.6) per 1000 per 1000 Low
ventilation There may be little or no
Based on data from Due to very
difference on mechanical
2434 participants in 2 Difference: 2 more per 1000 serious
ventilation.
controlled) — 44 more )
Adverse effects 0 1
There is probably little to
discontinuation Based on data from no difference in adverse
Due to serious
2309 participants in 4 effects leading to
studies. (Randomized discontinuation.
( CI 95% 11 fewer
controlled)
— 15 more )
21 of 109
Certainty of
Comparator
IL-6 receptor
blockers
evidence)
8.1 11.2 Very low

Hospital length Lower better days (Median) days (Mean) Due to serious
of stay risk of bias, The impact on hospital
Based on data from:
Difference: MD 3.1 more serious length of stay is very
( CI 95% 3.8 inconsistency, and uncertain.
fewer — 9.9 very serious
controlled)
more ) imprecision 4

mechanical days (Median) days (Mean) Low
Lower better Baricitinib may reduce
Based on data from: 328 MD 2.2 fewer duration of mechanical
Difference: risk of bias and
participants in 2 studies. ( CI 95% 5.3 fewer ventilation.
imprecision 5
8.4 8.9
Lower better Low
stability There may not be an
Based on data from: Due to serious
Difference: MD 0.5 more important impact on
2558 participants in 2 risk of bias and
( CI 95% 2.3 time to clinical stability.
studies. (Randomized imprecision 6
more )

2. Risk of bias: no serious. Most of the data on interleukin-6 receptor blockers comes from trials that were unblinded.
Imprecision: very serious. The credible interval includes important benefit and important harm.
3. Imprecision: serious. The credible interval includes small but important harm.
4. Risk of bias: serious. Most of the data on interleukin-6 receptor blockers comes from trials that were unblinded.
Inconsistency: serious. The trials that studied interleukin-6 receptor blockers had discrepant results: some increased length
of stay, others reduced length of stay. Imprecision: very serious. The credible interval includes important benefit and
important harm.
Imprecision: serious. The credible interval includes no important difference.
Imprecision: serious. Credible interval includes important harm and important benefit (using a minimal important difference
threshold of 1 day).
Ruxolitinib and tofacitinib, for patients with severe or critical COVID-19
Conditional recommendation against
We suggest not to use ruxolitinib or tofacitinib (conditional recommendation against).
• Clinicians should consider using these drugs only if neither baricitinib nor IL-6 receptor blockers (tocilizumab or sarilumab) are
available.
• The GDG emphasized the need for more trial evidence to better inform the recommendations.
22 of 109
Practical Info
Route, dosage and duration: We refer to the table of trial characteristics (ruxolitinib and tofacitinib) to guide the administration
of these agents, in the absence of other available information.
Timing: Ruxolitinib or tofacitinib (like IL-6 receptor blockers) should be initiated with systemic corticosteroids; specific timing
during hospitalization or the course of illness is not specified.
Benefits and harms

The effects of ruxolitinib or tofacitinib on mortality, need for mechanical ventilation and hospital length of stay remain
uncertain. Tofacinib may increase adverse events leading to drug discontinuation.
Subgroup analyses were undertaken for JAK inhibitors as a class (rather than on individual drugs) and revealed no evidence
of a subgroup effect on relative risk in younger (< 70 years) versus older patients; those receiving and not receiving
corticosteroids; those with severe versus critical COVID-19; and those receiving and not receiving remdesivir.

Due to serious imprecision due to small cohorts (ruxolitinib: two RCTs, 475 patients; tofacitinib: one RCT, 289 patients) with
few events and serious indirectness (pertaining to RCTs for ruxolitinib, most patients did not receive corticosteroids),
certainty of evidence was rated as low or very low for all prioritized outcomes for both drugs.

Applying the agreed values and preferences (see Section 7), the GDG inferred that, given the low or very low certainty
evidence on mortality and the other prioritized benefit outcomes and the remaining possibility of serious adverse effects,
the majority of well-informed patients would not want to receive ruxolitinib or tofacitinib. The GDG anticipated, however,
that because benefit has not been excluded, and because a class effect of JAK inhibitors might exist (such that baricitinib
provides indirect evidence of benefit for the other JAK inhibitors), a minority of well-informed patients would choose to
receive one or other drug in circumstances in which neither baricitinib nor IL-6 receptor blockers (tocilizumab or sarilumab)
were available.

The GDG noted that, given the recommendation against use of ruxolitinib or tofacitinib, efforts to ensure access to drugs
should focus on those that are currently recommended.

As ruxolitinib and tofacitinib are administered orally twice daily, this treatment should be easy to accept for hospitalized
patients with severe and critical COVID-19. In patients unable to swallow whole tablets, they can be dispersed in water to
take orally or via nasogastric tube (see Practical info).
Justification
When moving from evidence to the conditional recommendation not to use ruxolitinib or tofacitinib in patients with severe or
critical COVID-19, the GDG emphasized the low to very low certainty evidence for mortality, duration of mechanical ventilation
and possible increase in serious adverse events (particularly for tofacitinib).
The GDG emphasized the need for more trial evidence to better inform the recommendations; this is anticipated through
ongoing trials for these JAK inhibitors.
Applicability
None of the included RCTs enrolled children; therefore, the applicability of this recommendation to children remains uncertain.
Uncertainty also remains with regards to the administration of ruxolitinib or toficitinib to pregnant or lactating women.
23 of 109

Intervention: Ruxolitinib
Summary
Evidence summary
The LNMA on ruxolitinib was informed by two RCTs that enrolled 475 patients across non-severe, severe and critical
illness subgroups (42)(43). Both RCTs were registered, one was published in a peer-reviewed journal, and one was a trial
registration only. Both RCTs enrolled patients in in-patient settings. None of the included studies enrolled children or
pregnant women. The Table shows characteristics of the RCTs.
For patients with severe and critical COVID-19, the GRADE Summary of Findings tables for ruxolitinib shows the
relative and absolute effects compared with usual care for the outcomes of interest, with certainty ratings. See Section 7
for sources of baseline risk estimates informing absolute estimates of effect.
Subgroup analysis
The GDG pre-specified several subgroup analyses of interest across all JAK inhibitors of interest; of these, no significant
relative subgroup effects were found. Please see the Summary accompanying the recommendation for baricitinib for
more details.
Certainty of
Timeframe measurements Standard care Ruxolitinib (Quality of summary
evidence)
130 115
Odds Ratio 0.87 per 1000 per 1000 Very low
Mortality (CI 95% 0.27 — 2.85) Due to serious
The effect of ruxolitinib
Based on data from 472 indirectness and
Difference: 15 fewer per is very uncertain.
participants in 2 studies. very serious
1000
(Randomized controlled) ( CI 95% 91 fewer imprecision 1
— 169 more )
Mechanical Odds Ratio 0.87 116 108 Very low

ventilation (CI 95% 0.36 — 2.04) per 1000 per 1000 Due to serious
The effect of ruxolitinib
Based on data from 472 indirectness and
Difference: 8 fewer per 1000 is very uncertain.
(Randomized controlled) ( CI 95% 71 fewer imprecision 2
— 94 more )
Adverse effects 0 5 Ruxolitinib may not

leading to drug per 1000 per 1000 Low
cause an important
discontinuation Based on data from 484 Due to very
increase in adverse
Difference: 2 more per 1000 serious
participants in 1 study. effects leading to drug
(Randomized controlled) ( CI 95% 11 fewer imprecision 3 discontinuation.
— 15 more )
Hospital length 12.8 11.4 Very low

of stay Lower better days (Median) days (Mean) Due to serious The impact of ruxolitinib
Based on data from: 472 indirectness and on hospital length of
participants in 2 studies. Difference: MD 0.1 more very serious stay is very uncertain.
(Randomized controlled) ( CI 95% 2.1 imprecision 4
fewer — 2.4
more )
24 of 109
Certainty of
Timeframe measurements Standard care Ruxolitinib (Quality of summary
evidence)
Duration of
mechanical Lower better 14.7
days (Median) The effect of ruxolitinib
ventilation Based on data from: 3 Very low
5 on mechanical
participants in 1 study. Insufficient data ventilation is unknown.
(Randomized controlled)
9.9 9.8
Time to clinical days (Median) days (Mean) Very low
The impact of ruxolitinib
stability Lower better Due to serious
on time to clinical
Based on data from: 472 MD 0.1 fewer indirectness and
Difference: stability is very
( CI 95% 2.5 uncertain.
(Randomized controlled) fewer — 2.8 imprecision 6
more )
1. Indirectness: serious. Most patients probably did not receive corticosteroids at baseline. Concomitant use of
corticosteroids potentiates the beneficial effect interleukin-6 receptor blockers. Interleukin-6 is downstream in the Janus
kinase pathway. Therefore, the effect of ruxolitinib may have been larger had most patients received steroids. Further, the
ruxolitinib trial probably included many patients with non-severe disease. A beneficial effect of Janus kinase inhibitors may
be limited to patients with severe or critical disease. Imprecision: very serious. The credible interval includes important harm
and important benefit.
be limited to patients with severe or critical disease. Imprecision: very serious. The credible interval includes important harm
and important benefit.
3. Imprecision: very serious. There was only one event in the single trial that reported this outcome, of 424 patients
enrolled in the study.
be limited to patients with severe or critical disease. Imprecision: very serious. The credible interval includes important
benefit and important harm.
5. Risk of bias: serious. Indirectness: serious. Imprecision: very serious.
be limited to patients with severe or critical disease. Imprecision: very serious. Credible interval includes important harm
and important benefit (using a minimal important difference threshold of 1 day).

Intervention: Tofacitinib
25 of 109
Summary
Evidence summary
The LNMA for tofacitinib was informed by one RCT that enrolled 289 patients across non-severe, severe and critical
illness subgroups (44). The trial was registered and published in a peer-reviewed journal; it excluded children and
pregnant women. Table shows characteristics of the RCT.
For patients with severe or critical COVID-19, the GRADE Summary of Findings table for tofacitinib shows the relative
and absolute effects compared with standard care for the outcomes of interest, with certainty ratings. See Section 7 for
sources of baseline risk estimates informing absolute estimates of effect.
Subgroup analysis
The GDG pre-specified several subgroup analyses of interest across all JAK inhibitors of interest; of these, no significant
relative subgroup effects were found. Please see the Summary accompanying the recommendation for baricitinib for
more details.
Certainty of
Timeframe measurements Standard care Tofacitinib (Quality of summary
evidence)
130 78
Odds Ratio 0.47 per 1000 per 1000
Mortality Very low
(CI 95% 0.11 — 1.63)
Due to extremely The effect of tofacitinib
Based on data from 289 Difference: 52 fewer per serious is uncertain.
participants in 1 study. 1000 imprecision 1
(Randomized controlled) ( CI 95% 113
fewer — 69 more )
116 68
Mechanical Odds Ratio 0.5 per 1000 per 1000 Very low
ventilation (CI 95% 0.17 — 1.37)
Due to extremely The effect of tofacitinib
Based on data from 289 Difference: 48 fewer per serious is uncertain.
participants in 1 study. 1000 imprecision 2
(Randomized controlled) ( CI 95% 94 fewer
— 35 more )
Adverse effects 0 77
leading to drug per 1000 per 1000 Low
Tofacitinib may increase
discontinuation Based on data from 284 Due to very
adverse effects leading
Difference: 77 more per 1000 serious
participants in 1 study. to drug discontinuation.
(Randomized controlled) ( CI 95% 17 more imprecision 3
— 138 more )
12.8 11.7
Hospital length days (Median) days (Mean)
Lower better
Low
of stay Tofacitinib may reduce
Due to very
Based on data from: MD 1.1 fewer duration of
Difference: serious
289 participants in 1 ( CI 95% 2.8 hospitalization.
study. (Randomized imprecision 4
fewer — 0.6
controlled) more )
Duration of
14.7 Very low
The impact of tofacitinib
mechanical days (Median) on duration of
No data mechanical ventilation is
ventilation
unknown.
26 of 109
Certainty of
Timeframe measurements Standard care Tofacitinib (Quality of summary
evidence)
Time to clinical
stability (Randomized controlled) 9.9 Very low The effect of tofacitinib
days (Median) No data on time to clinical
stability is unknown.
1. Imprecision: extremely serious. The credible interval includes important benefit and important harm. There were only 12
events total.
2. Imprecision: extremely serious. Credible interval includes important benefit and important harm. There were only 18
events in total.
3. Imprecision: very serious. Very few events: only 21 in total (16/142 in tofacitinib arm and 5/142 in placebo arm).
4. Imprecision: very serious. Credible interval includes no important difference.

Type I and type II cytokine receptors are a family of receptors employed by over 50 interleukins, interferons, colony stimulating
factors, and hormones (45). The intracellular signalling triggered by these receptors is mediated by Janus kinases (JAKs), a small
family of kinases including JAK1, JAK2, JAK3, and tyrosine kinase 2 (TYK2). Type I cytokines include IL-2, IFN-γ, IL-12, and
TNFb, and type II cytokines include IL-4, IL-5, IL-6, IL-10, and IL-13.
JAK inhibitors are a class of drugs which inhibit intracellular signalling through multifactorial effects on cytokine signalling. As a
consequence, they interfere with many cellular responses, including antiviral responses, angiotensin-converting enzyme 2
(ACE2) expression, T cell function and differentiation, and macrophage activation (45).
Baricitinib, ruxolitinib, and tofacitinib are three of at least nine JAK inhibitors. These three drugs are all generally considered to
be non-specific JAK inhibitors, but differences in the specificity and potency for different JAKs are evident. Baricitinib has been
described as a JAK1/JAK2 inhibitor, ruxolitinib as JAK1/JAK2 > TYK2, and tofacitinib as JAK3/JAK1 > JAK2/TYK2; other
differences have also been previously described (45)(46)(47).
Studies evaluating JAK inhibitors for the treatment of COVID-19 have been conducted at doses that are as high or higher than
those approved for other indications, such as rheumatoid arthritis, myelofibrosis, and ulcerative colitis. Therefore, plausibility is
contingent upon the role of cytokine signalling in COVID-19, and not on whether the pharmacokinetics at the studied dose is
sufficient to inhibit the target proteins. There are notable differences in the approved doses, schedules, pharmacokinetics,
contraindications, and indications of these drugs for other indications. Collectively, these differences limit the confidence to
consider a class-wide recommendation with currently available data.
27 of 109
6.3 Sotrovimab (published 14 January 2022)
Info Box
Recommendations concerning sotrovimab for patients with non-severe COVID-19 were published on 14 January 2022 as
the eighth version of the WHO living guideline and in the BMJ as Rapid Recommendations. It follows the availability of one RCT
for non-severe illness, as per the LNMA on antibody and cellular therapies (2). No changes were made for the sotrovimab
recommendation in this ninth version of the guideline.
For patients with non-severe COVID-19
Conditional recommendation
We suggest treatment with sotrovimab, conditional to those at highest risk of hospitalization (conditional recommendation for).
• Whereas sotrovimab achieves a substantial reduction in the relative risk of hospitalization, the absolute benefit will be trivial in
absolute terms except in those at highest risk for hospitalization, for which the intervention should be reserved.
want to be treated with sotrovimab.
• In the absence of credible tools to predict risk for hospitalization in people infected with COVID-19, typical characteristics of
people at highest risk include those who are unvaccinated, older people, or those with immunodeficiencies and/or chronic diseases
(e.g. diabetes).
• Casirivimab-imdevimab were also conditionally recommended (see Section 6.5) and represent an alternative to sotrovimab; the
two drugs should not be given together. The choice of which monoclonal antibodies to use depends on availability, as well as
clinical and contextual factors, including emerging information about effectiveness with different variants (see Justification).
• Patients with severe or critical COVID-19: based on current evidence, the benefit of sotrovimab in seronegative patients with
severe or critical COVID-19 (see casirivimab-imdevimab recommendation in Section 6.5) remains unclear. This means that careful
clinical judgment needs to be applied if casirivimab-imdevimab is unavailable and sotrovimab is considered. New trial evidence for
sotrovimab in this setting was published after the GDG developed recommendations for this iteration, and will be considered,
alongside other publicly available emerging evidence, when developing future recommendations.
Practical Info
Route, dosage and duration: Additional considerations are available in a summary of practical issues. Here follows a brief
summary of key points.
• The authorized dose for sotrovimab is one single intravenous infusion of 500 mg over 30 minutes, administered as soon as
possible after a positive viral test for SARS-CoV-2 and within 10 days of symptom onset.
• Sotrovimab is available as a concentrated solution, and must be diluted prior to administration.
• Patients should be clinically monitored patients during the infusion, and observed for at least 1 hour after the infusion is
completed.
Benefits and harms

In patients with non-severe illness, sotrovimab probably reduces hospitalization, with little or no impact on infusion
reactions, with no data on time to clinical improvement. Sotrovimab probably has little or no impact on mortality and on
mechanical ventilation.
The planned subgroup analyses could not be performed in the absence of subgroup data reported publicly or provided by
investigators.

The evidence summary was informed by one RCT with 1057 patients (COMET-ICE), included in the LNMA (51). This trial
provides evidence of moderate certainty for reduced hospitalization (due to serious imprecision), high certainty for absence
of infusion reactions, moderate certainty (due to serious indirectness) for no or small difference in mortality and low
28 of 109
certainty (due to serious indirectness and imprecision) for no or small difference in mechanical ventilation.
Limitations in available empirically developed risk prediction tools for establishing patients’ risk of hospitalization represent
the major source of indirectness for which the GDG rated down the certainty of the evidence. In addition, the GDG felt that
there was some indirectness because of the possible emergence of variants for which the effectiveness of currently
available monoclonal antibodies may be reduced.
When rating precision, the GDG considered the estimates of the absolute risks. Accordingly, in the context of very low
event rates (e.g. very low risk of death among patients with non-severe illness), the GDG did not rate down for imprecision
despite relative risks with wide confidence intervals.

Applying the agreed upon values and preferences (see Section 7), the GDG inferred that almost all well-informed patients
with a low risk of hospitalization would decline sotrovimab, and only those at highest risk (e.g. unvaccinated, older, or
immunosuppressed) would choose to receive treatment.
In the absence of research evidence, in a previous survey (see recommendation for casirivimab-imdevimab), the GDG
expressed the view that most patients with a risk of hospitalization above 10%, and thus an absolute risk reduction of
approximately 6%, would choose to receive treatment, whereas most of those below that risk level would decline treatment.
These thresholds were also used here.

The GDG noted that sotrovimab is unlikely to be available for all individuals who, given the option, would choose to receive
the treatment. This further supports the guidance that sotrovimab be reserved for those at highest risk of hospitalization.
Additional challenges include the requirement for intravenous administration to patients who would normally be treated at
home. Specialized clinics may be required to ensure safe and effective administration of sotrovimab. For the intervention to
achieve substantial use, healthcare systems will have to address these challenges.
Given the cost and availability of sotrovimab, and the obstacles to ensuring access in LMICs may prove formidable. For
example, those with socioeconomic disadvantages tend to have less frequent access to services in the first five days of
symptoms and then less access to the interventions. Thus, the panel’s suggestion that patients at highest risk receive the
intervention may exacerbate health inequity. On the other hand, given the demonstrated benefits for patients and the still
low vaccine coverage in some LMICs and low-income countries (LICs), sotrovimab may be a useful tool to reduce
hospitalization from COVID-19 in unvaccinated populations. The recommendations should also provide a stimulus to
engage all possible mechanisms to improve global access to the intervention. Individual countries may formulate their
guidelines considering available resources and prioritize treatment options accordingly. On 17 December 2021, WHO
published the 7th Invitation to Manufacturers of therapeutics against COVID-19 to submit an Expression of Interest (EOI)
for Product Evaluation to the WHO Prequalification Unit, which includes sotrovimab.
Justification
A combination of the evidence, values and preferences, and feasibility contributed to the conditional recommendation for the
use of sotrovimab only in individuals with non-severe COVID-19 at highest risk of hospitalization. Typical characteristics of
people at highest risk include those who are unvaccinated, older people, or those with immunodeficiencies and/or chronic
diseases (e.g. diabetes).
Although there is moderate certainty evidence of a substantial relative risk reduction in hospitalization, only a minority of
patients who are at highest risk are likely to achieve sufficient benefit to compensate for the risks, and other limitations and
disadvantages of this therapy. These include a lack of reliable tools to identify high-risk patients, delivering a parenteral therapy
to patients who are typically cared for in the community, and limited availability of the drug.
The role of sotrovimab and casirivimab-imdevimab
Another combination of monoclonal antibodies, casirivimab-imdevimab, is also conditionally recommended in patients with non-
29 of 109
severe COVID-19 at highest risk of hospitalization. The GDG advised that clinicians do not administer the drugs together, given
an absence of evidence of incremental benefit of the drugs when given together, and the low likelihood of incremental benefit
mechanistically.
The GDG then considered how to choose between the two drugs. No trials provide head-to-head comparisons. With Alpha and
Delta variants, there may be little or no difference in the agents’ impact on critical outcomes, according to an indirect
comparison from the LNMA (see GRADE Summary of Findings table in Research evidence).
Now and in the future, the choice of monoclonal antibodies will depend on emerging information regarding effectiveness with
different variants and their availability, as well as clinical and contextual factors. Of note, the RCTs included in the LNMA were
conducted before the emergence of the Omicron variant. The GDG is now fully considering how the Omicron variant may
impact efficacy.
Following the publication of a previous conditional recommendation for casirivimab-imdevimab, pre-clinical evidence has
emerged suggesting that this monoclonal antibody combination lacks neutralization activity against the Omicron variant in
vitro (48). Sotrovimab has been reported to retain activity against Omicron in pseudo-virus assays, but with higher
concentrations being required for neutralization (49). More data are required to ascertain whether efficacy against the Omicron
variant will be maintained at the studied doses of monoclonal antibodies, and recommendations will be updated when additional
data becomes available.
Applicability
Children and pregnant women: The included RCT enrolled only non-pregnant adults; therefore, the applicability to children and
pregnant women remains uncertain. The GDG had no reason to believe that children or pregnant women with COVID-19 would
respond differently to treatment with sotrovimab. However, for children, as the risk of hospitalization is generally extremely low,
the GDG inferred that, in the absence of immunosuppression or another significant risk factor, children should not receive the
intervention.
Severe and critical COVID-19: A recently published RCT randomized 546 adults hospitalized with COVID-19 to two neutralizing
monoclonal antibody therapies (sotrovimab and BRII-196 plus BRII-198) or placebo (50). The results did not demonstrate
benefits from these antibodies therapies, including a subgroup analysis on patients with seropositive versus seronegative status.
Although the role of sotrovimab in severe and critical COVID-19 is not supported by the new trial, it was not assessed by the
GDG, as they focused on patients with non-severe COVID-19 where evidence was available at the time of recommendation
development; this trial, and any other new evidence that is publicly available, will be fully considered by the GDG for future
recommendations for sotrovimab.

Intervention: Sotrovimab
Summary
The LNMA for sotrovimab was informed by one RCT (COMET-ICE) that randomized 1057 non-hospitalized patients
with symptomatic COVID-19 (≤ 5 days after the onset of symptoms) and at least one risk factor for disease progression.
Of the 1057 randomized patients, the LNMA team had access to data for 1044 patients. Included patients were
randomized to receive a single infusion of sotrovimab at a dose of 500 mg, or placebo. Median age of patients was 53
years; 46% were male. Median duration of follow-up in the intention-to-treat population was 72 days. Vaccinated
patients were excluded from the trial (51).
The GRADE Summary of Findings table shows the relative and absolute effects of sotrovimab compared with standard
care (placebo) for the outcomes of interest, with certainty ratings.
Subgroup analysis
Four pre-specified subgroup analyses were requested by the GDG:

1. Age: children versus adults (< 70 years) versus older adults (≥ 70 years).
3. Time of symptom onset.
4. Serological status.
No data were available in the trial to conduct these analyses.
30 of 109
Certainty of
Timeframe measurements Standard care Sotrovimab (Quality of summary
evidence)
Odds Ratio 0.11 4 0 Moderate

Mortality (CI 95% 0 — 2.6) per 1000 per 1000 Due to serious Sotrovimab probably has
Based on data from
indirectness and little or no impact on
1044 participants in Difference: 4 fewer per 1000 some concerns mortality.
1 study. (Randomized
( CI 95% 4 fewer with imprecision 1
controlled)
— 6 more )

Mechanical (CI 95% 0 — 2.6) per 1000 per 1000 Low
ventilation Due to serious Sotrovimab may have
Based on data from
Difference: 16 fewer per indirectness and little or no impact on
1044 participants in
1000 serious mechanical ventilation.
controlled) ( CI 95% 18 fewer imprecision 2
— 27 more )

Moderate Sotrovimab probably
Risk in trials Based on data from
Difference: 28 fewer per Due to serious reduces admission to
1 study. (Randomized 1000 imprecision 3 hospital.
— 17 fewer )

Higher risk Based on data from
— 28 fewer )

— 46 fewer )
Infusion 0 0
reactions per 1000 per 1000 Sotrovimab results in
Based on data from
1044 participants in High little or no increase in
Difference: 0 fewer per 1000 infusion reactions.
( CI 95% 13 fewer
controlled)
— 13 more )
Time to clinical
improvement The effect of sotrovimab
No data
is unknown.
31 of 109
will be very small. There are some people with much higher baseline risk, which are not easily identifiable. For these patients,
sotrovimab may have an important impact on mortality. Imprecision: no serious. There were only four deaths (all in the
placebo group) in the single trial that studied sotrovimab.
sotrovimab may have an important impact on mechanical ventilation. Imprecision: serious. Very few events – only four
patients received invasive mechanical ventilation in the single trial (all in the placebo group).
3. Imprecision: serious. Few events (25 total: 4 in the sotrovimab group and 21 in the placebo group); does not meet
optimal information size.

Intervention: Sotrovimab
Comparator: Casirivimab-imdevimab
Certainty of
Comparator
Casirivimab-
Timeframe measurements Sotrovimab (Quality of summary
imdevimab
evidence)
2 1 Moderate
Mortality Odds Ratio 0.53 per 1000 per 1000 Due to serious There is probably little or
(CI 95% 0.09 — 2.12) indirectness and no difference on
(Randomized controlled) Difference: 1 fewer per 1000 some concerns mortality.
— 0 more )
Mechanical 6 2 Moderate
ventilation Odds Ratio 0 per 1000 per 1000 Due to serious There is probably little or
(CI 95% 0 — 0.01) indirectness and no difference on
(Randomized controlled) Difference: 4 fewer per 1000 some concerns mechanical ventilation.
— 0 fewer )
35 24
Admission to per 1000 per 1000 Moderate
hospital Odds Ratio 0.68 Due to serious There is probably little or
(CI 95% 0.21 — 1.97) indirectness and no difference in
(Randomized controlled) 1000 some concerns hospitalization.
— 32 more )
Infusion 0 0
reactions per 1000 per 1000 There is little or no
High difference in infusion
Difference: 0 fewer per 1000 reactions.
( CI 95% 13 fewer
— 13 more )
32 of 109
Certainty of
Comparator
Casirivimab-
Timeframe measurements Sotrovimab (Quality of summary
imdevimab
evidence)
No data The effect of sotrovimab

Time to clinical
is unknown.
improvement
the effect on mortality is uncertain. Imprecision: no serious. There was only one death (in the placebo group) in the single
trial that studied sotrovimab.
2. Indirectness: serious. The baseline risk across the entire population is very low, meaning that any impact on mechanical
ventilation will be very small. There are some people with much higher baseline risk, which are not easily identifiable. For
these patients, whether there is an important difference in mechanical ventilation is uncertain. Imprecision: no serious. Very
few events – only two patients received invasive mechanical ventilation in the single trial (both in the placebo group).
3. Indirectness: serious. Patients at high risk are not easily identified. Imprecision: no serious.

Sotrovimab (VIR-7831; GSK4182136) is a single human monoclonal antibody that binds to a conserved epitope of the SARS-
CoV-2 spike protein, preventing the virus from entering cells.
Sotrovimab is an Fc-engineered IgG monoclonal antibody that is expected to have a longer half-life than unmodified IgG
monoclonal antibodies, but there are no publicly available data regarding its half-life in humans. No published pre-clinical or
human pharmacokinetic data were available for review; assessment was therefore based upon public regulatory information and
the GSK Product Monograph (52).
Antiviral activity in a Syrian Golden hamster model of SARS-CoV-2 infection was demonstrated at 5 mg/kg IP but with a version
of the antibody that was not Fc-engineered (52). Neutralization of SARS-CoV-2 (USA WA1/2020) was achieved in Vero E6 cells
with an EC90 value of 0.19 µg/mL (53). Sotrovimab serum concentrations in COMET-ICE (single 500 mg IV infusion) provided
geometric mean Cmax (at the end of a 1 hr IV infusion) of 117.6 µg/mL (N=129, CV% 40) and a geometric mean Day 29 serum
concentration of 24.5 µg/mL (53). Population mean serum concentrations are therefore expected to be 129-fold higher after 29
days than the concentrations needed in vitro to neutralize the original strain of SARS-CoV-2.
A review manuscript with Vir Biotechnology authors indicates less than 10-fold loss of neutralization for sotrovimab for all
tested variants of concern, which did not at the time include Omicron (54).
Information in the FDA Emergency Use Authorization also state “no change” in activity of sotrovimab against Alpha, Beta,
Gamma, Epsilon, Iota, Kappa, Delta (including with K417N), Lambda and Mu in pseudo-typed virus-like particle neutralization
assays (53). Sotrovimab has since been reported to retain activity against Omicron in pseudovirus assays but with higher
concentrations being required for neutralization compared the wild-type virus (49). A better understanding of the
pharmacokinetic-pharmacodynamic relationship will be need to ascertain the consequences of this reduced activity for efficacy.
Resistance to other monoclonal antibodies deployed as single agents has been described to emerge rapidly in patients but this
has not been described clinically for sotrovimab to date. However, an E340A amino acid substitution in the conserved epitope
of the spike protein did emerge under a selective pressure in cell culture, and subsequent characterization using a pseudovirus
assay resulted in a >100-fold reduction in susceptibility to sotrovimab (53). Sixteen other substitutions introduced into the
epitope were also described as reducing neutralization by sotrovimab by between 5.4 and > 297-fold (53). Therefore, resistance
to sotrovimab can emerge under experimental conditions but there is uncertainty regarding the speed at which this will occur in
patients and the ultimate clinical consequences.
33 of 109
6.4 Convalescent plasma (published 7 December 2021)
Info Box
Recommendations concerning convalescent plasma for patients with non-severe, severe and critical COVID-19 were published
on 7 December 2021 as the seventh version of the WHO living guideline and in the BMJ as Rapid Recommendations. It follows
the availability of 16 RCTs across disease severities, as per the LNMA on antibody and cellular therapies (2). No changes were
made for the convalescent plasma recommendations in this ninth version of the guideline.
Recommendation against
We recommend against treatment with convalescent plasma (strong recommendation against).
Practical Info
The GDG made a strong recommendation against using convalescent plasma for the treatment of patients with non-severe
COVID-19 and a recommendation against using convalescent plasma in those with severe or critical COVID-19 outside the
context of a clinical trial. Given this, we will not go into detail regarding the many practical issues related to convalescent plasma
including but not limited to: identification and recruitment of potential donors, collection of plasma, storage and distribution of
plasma, and infusion of convalescent plasma into recipients.
Benefits and harms

In non-severe patients, convalescent plasma does not result in an important impact on mortality. Convalescent plasma
probably does not impact mechanical ventilation. There were no data evaluating the risk of hospitalization with
convalescent plasma and therefore the impact is very uncertain.
Convalescent plasma probably does not result in important increases in risks of transfusion-related acute lung injury (TRALI),
transfusion-associated circulatory overload (TACO), or allergic reactions.

The certainty in mortality was high, whereas mechanical ventilation was moderate due to serious risk of bias. Certainty was
rated as moderate for TRALI and TACO due to serious risk of bias, and for allergic reactions due to concerns regarding risk
of bias and imprecision.

The GDG inferred that, in addition to the agreed upon values and preferences (see Section 7), almost all well-informed
patients would choose against receiving convalescent plasma based on available evidence regarding relative benefits and
harms. From a population perspective, feasibility, acceptability, equity and cost are other important elements to take into
account (see Section 7).
For patients with non-severe illness, the GDG considered that resource and feasibility issues may be amplified in the
outpatient setting, and mobilizing the use of convalescent plasma on a large scale would likely be of questionable feasibility.
34 of 109
The GDG noted that convalescent plasma use is associated with significant resource requirements including identification of
potential donors, testing of donors to ensure adequate titres of anti-SARS-CoV-2 antibodies, collection of donor plasma,
storage of plasma, transportation of plasma to recipient location, and administration of plasma. These resources and
feasibility issues are compounded for those with non-severe disease who are most often outpatients. Also, this process is
costly and time-consuming. Given the number of patients with non-severe disease and the low event rate in this subgroup
of patients, mobilizing the use of convalescent plasma on a large scale would be of questionable feasibility.
Although blood transfusion is acceptable to most, there is a subset of the population that will not accept allogenic blood
transfusion. There are also regulatory challenges in most jurisdictions related to blood product transfusion.
Justification
A combination of the evidence, values and preferences, and feasibility contributed to the strong recommendation against
convalescent plasma in patients with non-severe COVID-19. Most importantly, given there was no benefit demonstrated in any
of the critical or important outcomes for either non-severe or severe or critical COVID-19, the GDG did not see any justification
for the resources (including time and cost) that would be associated with administration of convalescent plasma. The
recommendation also took into account possible associated harms (although not demonstrated in the evidence summary, there
is always a potential for harms with blood product transfusion), the low baseline risk of mortality, mechanical ventilation, and
hospitalization in non-severe illness, and feasibility challenges with the administration of convalescent plasma.
Titres
Titres of neutralizing antibodies varied substantially between included trials, with over half of the trials not reporting or
considering recipient titres at all. In fact, the largest trial (RECOVERY) did not report on donor antibody titres at all. Even when
titres were reported, the method for testing and the volume of plasma infused varied. This made it impossible to provide any
analysis based on donor titre levels or assess for credible subgroup effects.
Applicability
The applicability of this recommendation to children or pregnant women is currently uncertain, as the included RCTs enrolled
non-pregnant adults. The GDG had no reason to think that children with COVID-19 would respond any differently to treatment
with convalescent plasma. However, the risk of hospitalization in children is generally extremely low and the GDG inferred that
in the absence of immunosuppression or another significant risk factor children should not receive the intervention.

Intervention: Convalescent plasma
Summary
Evidence summary
The LNMA on convalescent plasma included 16 RCTs that enrolled 16 236 patients across non-severe, severe, and
critical illness subgroups. All RCTs were registered, and 80% were published in peer-reviewed journals; 20% were pre-
prints. 99% of participants were enrolled from in-patient settings; of them, 15% were admitted to the intensive care unit
(ICU). One percent of patients were enrolled from outpatient settings. None of the included studies enrolled children or
pregnant women. The Table shows characteristics of the RCTs, of which two trials used comparisons to plasma as
placebo and were not included in the evidence summaries. We are aware of two additional published RCTs comparing
convalescent plasma to standard care or placebo (55)(56). These trials were not incorporated in the latest analysis
presented to the GDG, based on which recommendations were made.
For patients with non-severe COVID-19, the GRADE Summary of Findings table shows the relative and absolute effects
of convalescent plasma compared to usual care for the outcomes of interest,with certainty ratings. This evidence
summary was informed by the LNMA (2) pooling data from 1602 patients in 4 RCTs for the outcome of mortality and
less data available for other outcomes, except for allergic reactions (8 RCTs, 243 patients). See Section 7 for sources of
baseline risk estimates informing absolute estimates of effect.
35 of 109
Subgroup analysis
We pre-specified the following subgroup analyses of interest:
1. Age: younger adults (< 70 years) versus older adults (> 70 years).
2. Severity of illness (at time of treatment initiation): non-severe versus severe and critical.
3. Treatment dose: higher titre versus lower titre plasma.
The subgroup analyses were performed on patients across all disease severities. The majority of subgroups did not have
sufficient data across outcomes of interest to pursue subgroup analyses.
Of those that did, we found no significant subgroup effects for severity of illness (p=0.80) and age (p=0.84) on mortality,
and of severity of illness (p=0.17) on mechanical ventilation.
36 of 109
Certainty of
Intervention
Outcome Study results and Comparator the Evidence Plain language
Convalescent
Timeframe measurements Standard care (Quality of summary
plasma
evidence)
Mortality
Odds Ratio 0.83 3 2
(CI 95% 0.43 — 1.46) per 1000 per 1000 Convalescent plasma
closest to 90 days Based on data from High does not result in an
1602 participants in 4 2 important impact on
Difference: 1 fewer per 1000
studies. 1 (Randomized ( CI 95% 2 fewer
mortality.
Mechanical Odds Ratio 0.71 6 4

ventilation (CI 95% 0.18 — 1.77) per 1000 per 1000 Convalescent plasma
Moderate
closest to 90 days Based on data from 705 probably does not
Due to serious
participants in 3 studies. impact mechanical
3
Difference: 2 fewer per 1000 risk of bias 4
(Randomized ventilation.
( CI 95% 5 fewer
Transfusion-
related acute
0 0
per 1000 per 1000
lung injury Convalescent plasma
Based on data from
Moderate
(TRALI) probably does not result
Difference: 0 fewer per 1000 Due to serious
within 28 days 1365 participants in 4 in an important increase
studies. 5 (Randomized ( CI 95% 5 fewer risk of bias 6 in TRALI.
Transfusion-
associated
0 5
per 1000 per 1000
circulatory
Convalescent plasma
overload (TACO) Based on data from
Moderate
Difference: 5 more per 1000 probably does not result
Due to serious
1442 participants in 4 ( CI 95% 3 fewer in an important increase
within 28 days studies. 7 (Randomized risk of bias 8 in TACO.
— 12 more )
controlled)
Allergic Odds Ratio 3.25 3 10

reactions (CI 95% 1.27 — 9.3) per 1000 per 1000 Low Convalescent plasma
within 28 days Based on data from Due to concerns probably does not result
15 243 participants in Difference: 7 more per 1000 with risk of bias in an important increase
8 studies. 9 ( CI 95% 1 more and imprecision 10 in allergic reactions.
(Randomized — 24 more )
controlled)
1. Systematic review. Baseline/comparator: Control arm of reference used for intervention. Supporting references: [57],
[58], [60], [61].
2. Risk of bias: no serious. The GDG did not rate down for risk of bias due to lack of blinding. .
[58], [61].
4. Risk of bias: serious. Imprecision: no serious. The GDG did not rate down for imprecision, because the credible interval
excludes an important benefit and important harm.
[59], [62], [63].
6. Risk of bias: serious. Most patients were enrolled in unblinded studies. Imprecision: no serious. GDG decided not to rate
down for imprecision, because credible interval excludes an important effect and baseline risk is very low.
37 of 109
[59], [62], [63],

down for imprecision, because credible interval excludes an important effect, and baseline risk is very low.
[59], [60], [62], [63], [64], [65], [66].
10. Risk of bias: serious. 2 trials (491 patients; 3% of total) were at low risk of bias vs. 6 trials (14 910 patients) at high risk
of bias. Imprecision: serious. GDG agreed the credible interval includes some concern regarding allergic reactions, though
acknowledges that the baseline risk is low.
For patients with severe or critical COVID-19
Only in research settings
We recommend not to use convalescent plasma for treatment of COVID-19, except in the context of a clinical trial
(recommended only in research settings).
Practical Info
The GDG made a recommendation against using convalescent plasma in those with severe or critical COVID-19 outside the
context of a clinical trial and a strong recommendation against using convalescent plasma for treatment of patients with non-
severe COVID-19. Given this, we will not go into detail regarding the many practical issues related to convalescent plasma
including but not limited to: identification and recruitment of potential donors, collection of plasma, storage and distribution of
plasma, and infusion of convalescent plasma into recipients.
Benefits and harms

In severe or critical patients, convalescent plasma may not result in an important impact on mortality, mechanical
ventilation, time to symptom improvement, length of hospital stay or ventilator-free days.
Convalescent plasma probably does not result in important increases in risks of TRALI, TACO or allergic reactions. However,
there is always potential for harms with blood product transfusion although not demonstrated in the evidence summary.

The certainty in mortality was low due to concerns with indirectness, risk of bias and imprecision. The GDG rated down
certainty to low for mechanical ventilation, length of hospital stay and ventilator-free days for serious risk of bias and
serious imprecision, and to low for time to symptom improvement due to very serious imprecision.
Certainty was rated as moderate for TRALI and TACO due to serious risk of bias, and for allergic reactions due to concerns
regarding risk of bias and imprecision.

The GDG inferred that, in addition to the agreed upon values and preferences (see Section 7), almost all well-informed
patients would choose against receiving convalescent plasma based on available evidence regarding relative benefits and
harms. From a population perspective, feasibility, acceptability, equity and cost are other important elements to take into
account (see Section 7).
38 of 109

The GDG noted that convalescent plasma use is associated with significant resource requirements including identification of
potential donors, testing of donors to ensure adequate titres of anti-SARS-CoV-2 antibodies, collection of donor plasma,
storage of plasma, transportation of plasma to recipient location, and administration of plasma. Also, this process is costly
and time-consuming.
Although blood transfusion is acceptable to most, there is a subset of the population that will not accept allogenic blood
transfusion. There are also regulatory challenges in most jurisdictions related to blood product transfusion.
Justification
After substantial discussion, the GDG decided to make a recommendation against convalescent plasma in patients with severe
or critical COVID-19, except in the context of clinical trials. Given the low certainty evidence suggesting a small or no effect on
mortality, mechanical ventilation, and time to symptom improvement, with possible associate harms (although not demonstrated
in the evidence summary, there is always a potential for harms with blood product transfusion) the panel agreed further research
addressing these patient-important outcomes would be valuable. This research focus on severe or critical COVID-19 was also
informed by the feasibility (patients are already hospitalized) and baseline risk of mortality and requiring life support
interventions (higher in severe or critical COVID-19). The panel identified high titre products as the highest priority for future
research as well as the need of reporting on donor titre and volume infused which can give an idea of dilution of titres in the
recipient. Similarly, the panel identified seronegative COVID-19 patients as the highest priority for future convalescent plasma
research.
A recommendation to only use a drug in the setting of clinical trials is appropriate when there is low certainty evidence, and
future research has a potential for reducing uncertainty about the effects of the intervention and for doing so at a reasonable
cost.

Intervention: Convalescent plasma
Summary
Evidence summary for convalescent plasma
Please see summary for patients with non-severe COVID-19 above. It provides details about the LNMA and 16 included
trials across disease severities, as well as subgroup analyses that did not detect credible effects based on age, severity of
illness, or dosage of convalescent plasma.
The GRADE Summary of Findings table shows the relative and absolute effects of convalescent plasma compared to
usual care for the outcomes of interest for patients with severe and critical COVID-19, with certainty ratings. This
evidence summary was informed by the LNMA (2), pooling data from from 14 366 patients in 10 studies for the
outcome of mortality, with less data available for other outcomes.
Baseline risk estimates
For severe and critical illness, for the critical outcome of mortality, the applied baseline risk estimate was 13% (130 in
1000). As for other related recommendations in this guideline, the estimate is derived from the SOLIDARITY trial for
severe and critical patients adjusted for treatment effects of corticosteroids. For other outcomes, we used the median of
the control arm of the RCTs that contributed to the evidence (see Section 7).
Subgroup analysis
We pre-specified the following subgroup analyses of interest:
1. Age: younger adults (< 70 years) versus older adults (> 70 years).
2. Severity of illness (at time of treatment initiation): non-severe versus severe and critical.
3. Treatment dose: higher titre versus lower titre plasma.
39 of 109
The majority of subgroups did not have sufficient data across outcomes of interest to pursue subgroup analyses.
Of those that did, we found no significant subgroup effects for severity of illness (p=0.80) and age (p=0.84) on mortality,
and of severity of illness (p=0.17) on mechanical ventilation.
Certainty of
Intervention
Convalescent
plasma
evidence)
Mortality
Odds Ratio 0.92 130 121 Very low
(CI 95% 0.7 — 1.12) per 1000 per 1000
closest to 90 days Due to concerns Convalescent plasma
Based on data from
with indirectness, may have a small or no
14 366 participants in Difference: 9 fewer per 1000 risk of bias, and effect on mortality.
10 studies. 1 ( CI 95% 35 fewer imprecision 2
(Randomized controlled) — 13 more )
Mechanical Odds Ratio 0.92 86 80 Low

ventilation (CI 95% 0.46 — 1.68) per 1000 per 1000 Due to serious Convalescent plasma
closest to 90 days Based on data from 623
risk of bias and may not impact
participants in 5 studies.
3
Difference: 6 fewer per 1000 serious mechanical ventilation.
(Randomized ( CI 95% 45 fewer imprecision 4
Transfusion-
related acute
0 0
per 1000 per 1000
lung injury Convalescent plasma
Based on data from
Moderate
(TRALI) probably does not result
Difference: 0 fewer per 1000 Due to serious
within 28 days 1365 participants in 4 in an important increase
studies. 5 (Randomized ( CI 95% 5 fewer risk of bias 6 in TRALI.
Transfusion-
associated
0 5
per 1000 per 1000
circulatory
Convalescent plasma
overload (TACO) Based on data from
Moderate
Difference: 5 more per 1000 probably does not result
Due to serious
1442 participants in 4 ( CI 95% 3 fewer in an important increase
within 28 days studies. 7 (Randomized risk of bias 8 in TACO.
— 12 more )
controlled)
Allergic Odds Ratio 3.25 3 10

reactions (CI 95% 1.27 — 9.3) per 1000 per 1000 Low Convalescent plasma
within 28 days Based on data from Due to concerns probably does not result
15 243 participants in Difference: 7 more per 1000 with risk of bias in an important increase
8 studies. 9 ( CI 95% 1 more and imprecision 10 in allergic reactions.
(Randomized — 24 more )
controlled)
Time to
symptom Lower better
15 15 Low
(Mean) (Mean) Convalescent plasma
improvement Based on data from: 472 Due to very
may not impact time to
participants in 3 studies. serious
11 Difference: MD 0 fewer symptom improvement.
(Randomized imprecision 12
controlled) ( CI 95% 10.4
fewer — 33.6
more )
40 of 109
Certainty of
Intervention
Convalescent
plasma
evidence)
Measured by:days
Length of Lower better 11.7 11 Low
hospital stay Based on data from: Due to serious
days (Mean) days (Mean) Convalescent plasma
may not impact length of
studies. 13 (Randomized serious
Difference: MD 0.7 fewer hospital stay.
controlled) imprecision 14
( CI 95% 2.3
fewer — 1 more )
Ventilator-free Measured by:days

High better
13.7 13 Low
days days (Mean) days (Mean) Due to serious Convalescent plasma
within 28 days Based on data from:
2859 participants in 3 risk of bias and may not impact the
Difference: MD 0.7 fewer serious number of ventilator-
studies. 15 (Randomized
( CI 95% 1.8 imprecision 16 free days.
controlled)
fewer — 0.4
more )
[62], [63], [64], [65], [66], [67], [68], [69], [70].
2. Risk of bias: serious. Indirectness: serious. Imprecision: serious. Credible intervals include both important benefit and
important harm.
[63], [66], [67], [69].
4. Risk of bias: serious. Imprecision: serious. The GDG decided the credible intervals warranted downgrading only once for
imprecision.
[59], [62], [63].
down for imprecision, because credible interval excludes an important effect, and baseline risk is low.
[59], [62], [63].
down for imprecision, because credible interval excludes an important effect, and baseline risk is low.
[59], [60], [62], [63], [64], [65], [66].
10. Risk of bias: serious. 2 trials (491 patients; 3% of total) were at low risk of bias vs. 6 trials (14 910 patients) at high risk
of bias. Imprecision: serious. GDG agreed the credible interval includes some concern regarding allergic reactions, though
acknowledges the baseline risk is low.
11. Systematic review. Baseline/comparator: Control arm of reference used for intervention. Supporting references: [63].
12. Imprecision: very serious.
[63], [66], [67], [68], [69], [70].
14. Risk of bias: serious. All studies except one were not adequately blinded. Imprecision: serious. Credible interval does
not exclude small but important benefit.
[64], [66].
16. Risk of bias: serious. Almost all patients were randomized to trials that were not blinded. Imprecision: serious. Credible
interval does not exclude important benefit.
41 of 109

The proposed primary mechanism of action for convalescent plasma involves the transfer of endogenously produced
neutralizing antibodies present within the plasma from previously infected and recovered patients into patients with active
infection (71). Therefore, the underlying plausibility for this mechanism of action depends upon whether sufficient antibody
concentrations remain following the dilution from donor to recipient. As such, the neutralizing antibody titre within the donor
plasma as well as the volume administered are likely to be important. Data generated in Syrian golden hamsters have
demonstrated efficacy of convalescent plasma against SARS-CoV-2 at a titre of 1:2560, but not at a titre of 1:320, when given at
a volume of 1 mL, which extrapolates based on average blood volume to a human dosing volume of 300 mL (72).
At the extremes of the studies which have investigated convalescent plasma clinically and reported the dose in terms of
neutralizing antibody titre and volume administered, administration of 200 mL would be expected to result in an average
dilution of 25-fold whereas administration of 1000 mL would be expected to result in an average dilution of 5-fold from those
titres present in the circulation of the donor themselves (assuming an average human blood volume of 5 mL (73)). It should be
further recognized that the concentrations (titre) of neutralizing antibodies present within convalescent plasma are highly
variable between donors and that there are different methodologies available to measure it (74).
Antibody titre, methodology employed, and the volume of convalescent plasma administered all vary widely across the studies
that have investigated this approach in COVID-19. It should be further noted that in some trials, the antibody titre reported for
eligibility was higher than the reported antibody titre in the donor plasma that was used because of the differences in
methodology used for the two assessments (e.g. total IgG for donor eligibility with subsequent assessment of the specific
neutralizing antibody titre (75)). There is clear uncertainty surrounding the dose of neutralizing antibodies given in different trials
and this uncertainty is summarised as follows:
For trials in severe/critical patients:
• No cut-off in neutralizing antibody titre of the donor was applied in 9/16 studies.
• Antibody titre of the donor plasma was not recorded in 12/16 trials, meaning the titre may have been high or may have
been low. However, in 3 of the trials in which donor titre was not recorded, a lower cut-off was applied at a titre of either
1:160 (for 2 trials) or 1:400.
• The largest trial (RECOVERY) did not report donor antibody titres although only donors with a titre above 1:100 were
eligible
• One (1/16) trial did not provide information on what volume of plasma was administered meaning volume could have been
high or could have been low.
• Both volume and donor titre were only known for 6/16 trials. Donor titres were 1:80, 1:87, 1:300, 1:320, 1:526, and 1:640
with volumes of 300, 500, 400–600, approx. 480, 750–975, and 300 mL, respectively (estimated dose range of 6-fold).
For trials in non-severe patients:
• Only three trials were conducted in non-severe patients using antibody titres of 1:40, 1:292, and 1:3200 with volumes
administered of 250–300 mL, 400 mL and 250 mL, respectively (estimated dose range of 100-fold).
• Two trials studied both non-severe and severe/critical patients, one of which didn’t record antibody titre, and the other
which used 200–250 +/- 75 mL of plasma with a titre of 1:160.
42 of 109
6.5 Casirivimab-imdevimab (neutralizing monoclonal antibodies) (published 24 September

2021)
Info Box
Recommendations concerning neutralizing monoclonal antibodies (casirivimab-imdevimab) for patients with non-severe, severe
or critical COVID-19 were published on 24 September 2021 as the sixth version of the WHO living guideline and in the BMJ
as Rapid Recommendations. It follows the availability of pre-prints of four trials, that are part of the larger adaptive randomized
master protocol addressing patients with non-severe illness, and of the RECOVERY trial addressing severe and critically ill
patients (9)(10)(11). No changes were made for the casirivimab-imdevimab recommendations in this ninth version of the
guideline.
Following the publication of a previous conditional recommendation for casirivimab-imdevimab, additional preclinical evidence
has emerged (see Mechanism of action) (76). There is a substantial body of pre-clinical in vitro data, and a confirmatory in vivo
evaluation, demonstrating lack of efficacy of casirivimab-imdevimab against the Omicron BA1 variant (see Mechanism of
action). As a result, casirivimab-imdevimab is no longer recommended for COVID-19 treatment except in cases where rapid viral
genotyping is available and confirms infection with a SARS-CoV-2 variant (such as Delta) that is susceptible to the neutralizing
activity of this combination of monoclonal antibodies.
Conditional recommendation Updated
We suggest treatment with casirivimab-imdevimab, conditional to those at highest risk of hospitalization, and where viral
genotyping can confirm a susceptible SARS-CoV-2 variant (i.e. excluding Omicron BA1) (conditional recommendation for).
• Whereas casirivimab-imdevimab achieves a substantial reduction in the relative risk of hospitalization, the absolute benefit will be
trivial in absolute terms for all but those at highest risk for which the intervention should be reserved.
want to be treated with casirivimab-imdevimab.
• In the absence of credible tools to predict risk for hospitalization in people infected with COVID-19, typical characteristics of
people at highest risk include lack of vaccination, older people, or those with immunodeficiencies and/or chronic diseases (e.g.
diabetes).
Practical Info
Dosing and administration route: Intravenous total dose of the monoclonal antibody combination differed in the non-severe
trials, ranging from total dose 1200 mg–8000 mg (600 mg–4000 mg each antibody), demonstrating efficacy at all doses,
including the lowest tested, 1200 mg total dose (600 mg of each antibody). In the face of limited access and resource
considerations, health systems will face choices concerning dose of casirivimab-imdevimab as well as intravenous or
subcutaneous injections. Please see the acceptability and feasibility section (under Evidence to Decision) for some deliberations
to help in making these choices within the possible range of 1200 mg–2400 mg total dose.
Monitoring: Although the available trials have not convincingly shown that casirivimab-imdevimab results in allergic reactions,
the possibility remains. To be administered through an intravenous line containing a sterile in-line or add-on 0.2 micron filter.
Following administration, patients should undergo monitoring for severe anaphylaxis.
Benefits and harms

In non-severe patients, casirivimab-imdevimab probably reduces the risk of hospitalization and duration of symptoms.
Casirivimab-imdevimab is unlikely to have serious adverse effects, including allergic reactions.

Limitations in available empirically developed risk prediction tools for establishing patients’ risk of hospitalization represents
the major source of indirectness for which the GDG rated down the certainty of the evidence (22). In addition, the GDG felt
43 of 109
that there was some indirectness because of the possible emergence of variants in which effectiveness may be reduced. The
GDG thus rated down the certainty of evidence to moderate for hospitalization and duration of symptoms. The GDG rated
down evidence certainty to moderate for allergic reactions because of imprecision but considered the finding of no serious
adverse effects to represent high certainty evidence.

Applying the agreed values and preferences (see Section 7), the GDG inferred that almost all well-informed patients at
typical low risk of hospitalization would decline casirivimab-imdevimab and only those at higher risk (e.g. unvaccinated,
older, or immunosuppressed) would choose the treatment.
The limited availability of casirivimab-imdevimab in relation to the number of infected individuals proved a major concern.
For non-severe illness, GDG members completed a survey in which they provided their views regarding the magnitude of
reduction in hospitalization that would prompt patients to use casirivimab-imdevimab. The panel responses suggested that
the majority of patients with a risk of hospitalization above 10%, and thus an absolute risk reduction of approximately 6%,
would choose to receive treatment while a majority of those below that risk level would decline treatment. Large majorities
of patients with risks substantially higher than 10% would choose to receive treatment and large majorities of those with
substantially lower risks would decline.

The GDG noted that casirivimab-imdevimab is unlikely to be available for all individuals who, given the option, would
choose to receive the treatment. This further supports the guidance that casirivimab-imdevimab be reserved for those at
highest risk of hospitalization.
Major feasibility challenges include limited production of casirivimab-imdevimab and, for outpatients, the requirement for
intravenous administration. Regarding intravenous administration, it is likely that specialized clinics with adequate amounts
of the antibodies and personnel who will ensure safe and effective administration of the intervention will be required. For
the intervention to achieve substantial use, health systems will have to address these challenges.
Choosing a dose: Different doses of the monoclonal antibody combination were used in different trials, and health systems
will face the choice of which dose to use and this can be informed by values and preferences. If one’s priority is to ensure
giving as many people as possible the opportunity to benefit from treatment, one might use the lowest effective dose
offered in the studies of non-severe patients, 1200 mg total dose (600 mg of each antibody) (77). If one’s priority is on
ensuring effectiveness in every individual who receives treatment, and minimizing the risk of emergence of resistance, one
might use a higher total intravenous dose of 2400 mg (1200 mg of each antibody).
Administration route: A similar value and preference issue arises in choosing between intravenous administration – used in
the four trials included in the LNMA (from a larger adaptive randomized master protocol) (78) – and subcutaneous
administration, which has been used in the prophylactic trial (79). Intravenous administration will achieve maximum drug
concentrations faster than subcutaneous administration; however, both will achieve exposure above the proposed
therapeutic threshold. If one’s priority is to ensure maximum effectiveness in every individual who receives treatment, one
might choose intravenous administration. If one’s priority is, in the face of practical difficulties of widespread intravenous
administration in the community, to ensure giving as many people as possible the opportunity to benefit from treatment,
one might ensure the availability of subcutaneous administration as an alternative. Volumes that can be administered
subcutaneously are limited to the lowest dose, which is a total dose 1200 mg (600 mg of each antibody).
Justification
A combination of the evidence, values and preferences, and feasibility contributed to the conditional recommendation for the
use of casirivimab-imdevimab only in patients with non-severe COVID-19 at highest risk of hospitalization. Although there is
moderate certainty evidence of a substantial relative risk reduction in hospitalization, only a minority of patients who are at
highest risk are likely to achieve important benefit. In routine care of those with non-severe COVID-19, there is a lack of tools to
reliably identify those at highest risk of hospitalization. This clinical complexity, combined with the limited availability of the drug
44 of 109
and need for parenteral administration route for a group of patients who are typically cared for in the community, present a
range of challenges for care that need to be addressed by health care systems.
Applicability
The applicability of this recommendation to children is currently uncertain, as the included RCTs enrolled adults. The GDG had
no reason to think that children with COVID-19 would respond any differently to treatment with casirivimab-imdevimab.
However, the risk of hospitalization in children is generally extremely low and the GDG inferred that in the absence of
immunosuppression or another significant risk factor children should not receive the intervention.

Intervention: Casirivimab-imdevimab
Summary
Evidence summary
For patients with non-severe COVID-19, the LNMA (2) pooled data from four trials that enrolled 4722 patients
randomized to casirivimab-imdevimab or usual care (78). All trials were registered and presented in pre-prints when the
data were reviewed by the GDG. The Table shows trial characteristics.
The GRADE Summary of Findings table shows the relative and absolute effects of casirivimab-imdevimab compared to
usual care for the outcomes of interest in patients with non-severe COVID-19, with certainty ratings.
Specific considerations regarding baseline risk estimates informing absolute estimates of effect
For hospital admission, the key outcome driving the recommendation in favour of casirivimab-imdevimab, we used a
baseline risk of 4.2% (42 in 1000) based on the median of the control arm of the four RCTs contributing to the evidence.
These trials recruited patients at elevated risk of being hospitalized to increase statistical power in detecting potential
treatment effects. The baseline risk is therefore appreciably higher than the risk for many patients with non-severe
COVID-19.
Subgroup analysis
We found no evidence of subgroup effects on age or time from onset of illness in patients with non-severe COVID-19.
Certainty of
Intervention
Casirivimab-
imdevimab
evidence)
Odds Ratio 0.57 2 1

Mortality (CI 95% 0.26 — 1.2) per 1000 per 1000 Casirivimab-imdevimab
Moderate
Based on data from do not have an
Due to serious
4722 participants in 4 Difference: 1 fewer per 1000 important effect on
studies. (Randomized indirectness 1 mortality.
( CI 95% 1 fewer
Mechanical
Odds Ratio 0.22 4 1
(CI 95% 0.03 — 1.21) per 1000 per 1000 Casirivimab-imdevimab
ventilation Moderate
Based on data from probably do not have an
Due to serious
3432 participants in 2 Difference: 3 fewer per 1000 important effect on
studies. (Randomized indirectness 2 mechanical ventilation.
( CI 95% 4 fewer
Odds Ratio 0.29 Casirivimab-imdevimab

Admission to
(CI 95% 0.17 — 0.48)
35 10 Moderate
probably reduce
per 1000 per 1000
45 of 109
Certainty of
Intervention
Casirivimab-
imdevimab
evidence)
hospital Based on data from

Risk in trials 4722 participants in 4 Difference: 25 fewer per Due to serious
admission to hospital.
studies. (Randomized 1000 indirectness 3
— 18 fewer )

Moderate Casirivimab-imdevimab
Higher risk Based on data from
Difference: 42 fewer per Due to serious probably reduce
studies. (Randomized 1000 indirectness 4 admission to hospital.
— 30 fewer )

Difference: 69 fewer per Due to serious probably reduce
studies. (Randomized 1000 indirectness 5 admission to hospital.
— 49 fewer )
Adverse effects 2 1 Casirivimab-imdevimab

leading to drug per 1000 per 1000
Based on data from do not result in an
discontinuation High important increase in
Difference: 1 fewer per 1000 adverse effects leading
( CI 95% 2 fewer to drug discontinuation.
controlled)
— 1 more )
Allergic 3 9
per 1000 per 1000 Casirivimab-imdevimab
reactions Based on data from
Moderate
probably do not result in
Due to serious
15 406 participants in an important increase in
4 studies. (Randomized allergic reactions.
( CI 95% 1 fewer
controlled)
— 29 more )
Time to 14 9.9
symptom Lower better (Mean) (Mean)
improvement Based on data from:
Difference: MD 4.1 fewer Due to serious probably reduce time to
studies. (Randomized ( CI 95% 5.7 fewer indirectness 7 symptom improvement.
Duration of 9.6 8.2

hospitalization (Mean) (Mean) Casirivimab-imdevimab
Low
(not in hospital at Lower better may not have an
Due to very
baseline) Based on data from: 111 MD 1.4 fewer important impact on
Difference: serious
participants in 2 studies. ( CI 95% 4.6 fewer duration of
imprecision 8 hospitalization.
(Randomized controlled) — 1.8 more )
46 of 109
1. Indirectness: serious. There is substantial variability in baseline risk of death between patients. Casirivimab-imdevimab
may confer an important benefit in patients at higher risk of death.
2. Indirectness: serious. There is substantial variability in baseline risk of mechanical ventilation between patients.
Casirivimab-imdevimab may confer an important benefit in patients at higher risk of mechanical ventilation.
3. Indirectness: serious. Differences between the population of interest and those studied: the predominant strains
currently circulating are not the same as the ones that were circulating during the studies.
6. Imprecision: serious.
Conditional recommendation Updated
We suggest treatment with casirivimab-imdevimab, conditional to those with seronegative status, and where viral genotyping
can confirm a susceptible SARS-CoV-2 variant (i.e. excluding omicron BA1) (conditional recommendation for).
• With benefits of casirivimab-imdevimab observed only in patients with seronegative status, clinicians will need to identify these
patients by credible tests available at the point of care to appropriately apply this recommendation (see Evidence to Decision
section).
• Treatment with casirivimab-imdevimab is in addition to the current standard of care, which includes corticosteroids and IL-6
receptor blockers.
Practical Info
Dosing and administration route: Intravenous dosing of the monoclonal antibody combination in the RECOVERY trial that
enrolled severe and critical COVID-19 was a total dose of 8000 mg (4000 mg for each antibody), whereas the dose differed in
the four trials in non-severe patients (from a larger adaptive randomized master protocol), ranging from intravenous total dose of
1200 mg–8000 mg. In the face of limited access and resource considerations, health systems will face a choice concerning the
dose of casirivimab-imdevimab. Please see the acceptability and feasibility section (under Evidence to Decision) for deliberations
to help in making these choices within the possible range of 2400 mg–8000 mg total dose.
Diagnostic testing: Tests to identify patients with seronegative status at the time patients present with severe or critical
COVID-19 warrant rapid serological tests with adequate performance characteristics. Health care systems would need to
implement such tests, as outlined in the acceptability and feasibility section.
Monitoring: Although the available trials have not convincingly shown that casirivimab-imdevimab results in allergic reactions,
the possibility remains. Administer through an intravenous line containing a sterile in-line or add-on 0.2 micron filter. Following
infusion, patients should undergo monitoring for allergic reactions.
Benefits and harms

In the overall population of patients with severe and critical COVID-19, casirivimab-imdevimab may not have an impact on
mortality and the impact on mechanical ventilation and duration of hospitalization is very uncertain.
A credible subgroup effect demonstrated that casirivimab-imdevimab probably reduces mortality in patients who are
seronegative, with the absolute effects ranging from 39 fewer per 1000 (95% CI 62 fewer-13 fewer) in the severely ill to 69
47 of 109
fewer (95% CI 110 fewer-23 fewer) in the critically ill. In seronegative patients, the intervention possibly reduces the need
for mechanical ventilation (absolute effect estimate 42 fewer per 1000; 95% CI 74 fewer-6 fewer). Aside from the credible
subgroup effect for serological status, we found no evidence of subgroup effects on age or time from onset of illness in the
non-severe, or on age, time from onset of illness, and severity in the severe and critically ill.

In patients with severe and critical COVID-19, evidence for mortality was of low certainty because of imprecision and high
likelihood that casirivimab-imdevimab has, in the seronegative and seropositive patients included in the overall group, very
different effects. In this population, the evidence regarding the impact of the intervention on need for mechanical
ventilation and duration of hospitalization was very low certainty given additional concerns with risk of bias.
For patients with severe and critical COVID-19 who are seronegative, evidence for mortality was rated as moderate as a
result of concerns regarding imprecision (the confidence interval includes effects as small as 14 in 1000 that some patients
may perceive as trivial) and indirectness (variants may emerge in which casirivimab-imdevimab antibodies may have reduced
effect). For mechanical ventilation, the GDG noted risk of bias from lack of blinding as an additional concern, resulting in low
certainty evidence. For duration of hospitalization, the GDG also found very serious imprecision, resulting in very low
certainty evidence.

Applying the agreed values and preferences (see Section 7), the GDG inferred that most if not all well-informed patients
with severe or critical COVID-19 and seronegative status would choose to receive casirivimab-imdevimab. Other patients –
those whose are seropositive or whose status is uncertain – are likely to decline the intervention.
Although the GDG focused on an individual patient perspective, they also considered a population perspective in which
feasibility, acceptability, equity and cost are important considerations. In this case, feasibility concerns played an important
role in the conditional recommendation. For the severe and critical patients, both limited availability of therapeutics and the
requirement for serological testing as part of clinical decision-making to identify the seronegative patients proved important.

Cost and availability
Given the cost and availability of casirivimab-imdevimab, and the challenges associated with serological testing, the
obstacles to ensuring access in low- and middle-income countries may prove formidable. Thus, the panel’s suggestion that
patients who are seronegative receive the intervention may exacerbate health inequity. On the other hand, given the
demonstrated benefits for patients, the recommendations should provide a stimulus to engage all possible mechanisms to
improve global access to the intervention. Individual countries may formulate their guidelines considering available
resources and prioritize treatment options accordingly.
Supply of casirivimab-imdevimab is likely to be limited, raising accessibility and possibly rationing challenges. In addition,
benefit requires identification of serological status at the time patients present with severe or critical COVID-19. The
availability of rapid and accurate serological tests as well as dosing and administration route for the drug are therefore key
factors to consider for health care systems.
Rapid serological tests: Tests with performance characteristics similar to the reference standard test used to characterize
seronegative patients in the RECOVERY trial, i.e. the Oxford fluorescent-based ELISA assay for serum IgG against the SARS-
CoV-2 spike protein, with an arbitrary cut-off determined by a panel of positive controls, are available and potentially
affordable. Some lateral flow assays may be suitable and can usually be performed in several minutes (80)(81)(82). Health
care systems must, however, gain expertise in choosing and implementing a rapid test or test, choosing those most
applicable to their setting..
48 of 109
Choosing a dose: The clinical trial in severe and critical patients (RECOVERY) tested a total dose of 8000 mg (4000 mg of
each antibody) casirivimab-imdevimab; clinical trials in non-severe patients have used total doses of 1200 mg–8000 mg
(600 mg–4000 mg of each) with similar effects on decreasing the need for hospitalization. Pharmacokinetic profiles of
casirivimab-imdevimab in non-severe with COVID-19 are available at total doses of 1200 mg–8000 mg (600 mg–4000 mg
of each monoclonal antibody) (78). This study demonstrated that the target therapeutic concentrations were achieved
rapidly in serum and maintained for 28 days even at the lowest total dose of 1200 mg (600 mg of each antibody), although
serum concentrations of the drug were noted to vary considerably between individuals. Therefore, using doses lower than
used in the RECOVERY trial (8000 mg total dose) for treatment of severely and critically ill patients may achieve the same
benefit. On the other hand, it is theoretically plausible but untested that pharmacokinetic differences in severe and critical
patients, when compared with non-severe, may reduce drug exposure (see Mechanism of action). This would increase the
risk of sub-optimal drug exposure in some individuals, which in turn could increase the risk of therapeutic failure and the
emergence of viral resistance.
In the absence of clinical data on treatment of severe and critical patients with doses lower than 8000 mg, making a choice
on which dose to use can be informed by values and preferences. If one’s priority is ensuring effectiveness in every
individual who receives treatment, and minimizing the risk of emergence of resistance, one might use the total intravenous
dose of 8000 mg (4000 mg of each antibody). If one’s priority is, in the face of limited drug availability and high cost, giving
as many people as possible an opportunity to benefit from treatment, one might use an intravenous dose as low as a total of
2400 mg (1200 mg of each antibody).
At a time of drug shortage, it may be necessary to prioritize use of casirivimab-imdevimab through clinical triage. One
possibility is to prioritize patients with the highest baseline risk for mortality (e.g. those with critical disease over those with
severe disease), in whom the absolute benefit of treatment is therefore greatest. For example, despite consistent relative
effects (OR 0.85 for mortality) with casirivimab-imdevimab in seronegative patients, the absolute risk reduction for mortality
in the critically ill would be 69 fewer deaths per 1000 (95% CI 110 to 23 fewer deaths) and in the severely ill would be 39
fewer deaths per 1000 (95% CI 62 to 13 fewer deaths).
Other suggestions for prioritization, which lack direct evidence, include focusing on patients with an actively deteriorating
clinical course and avoiding casirivimab-imdevimab therapy in those with established multi-organ failure (in whom the
benefit is likely to be smaller).
Justification
In patients with severe or critical illness, the conditional recommendation in favour of casirivimab-imdevimab use reflects the
likelihood that any benefits are restricted to patients who are seronegative. In the RECOVERY trial, which provided all the
evidence in severe and critical patients, serological status at baseline was assessed in a pre-planned but retrospective analysis
using a laboratory-based anti-spike protein assay. In order to translate the trial findings into clinical practice, assessment of
serological status will need to become integrated into a clinical decision pathway before treatment is administered. This implies
rapid identification of serological status at the time of presentation of severe or critical illness to guide use in this population.
Several rapid and relatively inexpensive tests with adequate performance characteristics are available and should see increasing
use in settings in which casirivimab-imdevimab is available for administration to these patients.
Applicability
None of the included RCTs enrolled children, and therefore the applicability of this recommendation to children is currently
uncertain. Fortunately, very few children become critically ill with COVID-19. For those who do and are seronegative, it is
possible they may benefit from casirivimab-imdevimab. Lack of data precluded the GDG from making specific recommendations
for other special populations, such as pregnant women.

Population: Patients with severe or critical COVID-19, seronegative
49 of 109
Summary
Evidence summary
The LNMA was informed by one large trial (RECOVERY) in patients with severe and critical illness that enrolled 9785
patients, most of whom received corticosteroids (83). The trial was registered and presented in pre-prints when the data
was reviewed by the GDG. The Table shows trial characteristics.
usual care for the outcomes of interest in patients with severe and critical COVID-19 and seronegative status, with
certainty ratings.
Specific considerations regarding baseline risk estimates informing absolute estimates of effect
In severe and critical COVID-19 patients, for the critical outcome of mortality, the applied baseline risk estimate was
13% (130 in 1000). As for other related recommendations in this guideline, the estimate is derived from the
SOLIDARITY trial for severe and critical patients adjusted for treatment effects of corticosteroids.
To inform baseline risk estimates for mortality in seronegative patients, we identified the control arm of the RECOVERY
trial as the best source. For patients with seronegative status, risk of death in both severe (26%; 260 per 1000) and
critical (46%; 460 per 1000) illness is substantially higher than for the overall population. Thus, seronegative patients
represent a very high risk population, leading to substantial absolute risk reductions in mortality (3.9% in the severe and
6.9% in the critical) despite the modest 15% relative risk reduction.
Subgroup analysis
A highly credible subgroup effect demonstrated that casirivimab-imdevimab likely reduces mortality in patients who are
seronegative but not in those who are seropositive.
The credibility of the subgroup effect was evaluated using the ICEMAN tool (84). The credibility of the subgroup effect
was strongly supported by: an a priori hypothesis with a specified direction; a small number of such hypotheses;
evidence based on a within-study comparison; a suggestion of a similar subgroup effect in mechanical ventilation; and
an interaction p-value of 0.001.
Fig. 2 presents the forest plot depicting the point estimate and confidence interval around the effects on mortality in
patients with seropositive and seronegative status, demonstrating benefit in those with seronegative status, suggesting
harm in those with seropositive status, and no overlap in the confidence intervals, a result corresponding to the p=0.001
in the test of interaction (83).
Fig. 2. Mortality, in seropositive and seronegative patients with severe and critical COVID-19
CI: confidence interval, RR: relative risk.
Very low certainty evidence raises the possibility of shorter hospitalization in seronegative patients. Aside from the
reported subgroup effects on serological status, we found no evidence of subgroup effects on age, time from onset of
illness, and severity (comparing severe and critically ill patients).
Certainty of
Intervention
Casirivimab-
imdevimab
evidence)
Mortality Relative Risk 0.85 260 221 Moderate Casirivimab-imdevimab

Severe disease (CI 95% 0.76 — 0.95) per 1000 per 1000 Due to concerns probably reduce
50 of 109
Certainty of
Intervention
Casirivimab-
imdevimab
evidence)
Based on data from

2823 participants in with imprecision
1000 mortality.
1 study. (Randomized and indirectness 1
( CI 95% 62 fewer
controlled)
— 13 fewer )
460 391
Relative Risk 0.85
Mortality per 1000 per 1000
(CI 95% 0.76 — 0.95) Moderate
Critical disease Based on data from Casirivimab-imdevimab
Difference: 69 fewer per Due to concerns
probably reduce
2823 participants in with imprecision
1000 mortality.
( CI 95% 110 and indirectness 2
controlled) fewer — 23
fewer )
Relative Risk 0.87 320 278

Mechanical (CI 95% 0.77 — 0.98) per 1000 per 1000 Low
ventilation Due to concerns Casirivimab-imdevimab
Based on data from
Difference: 42 fewer per with risk of bias, may reduce mechanical
1000 imprecision, and ventilation.
controlled) ( CI 95% 74 fewer indirectness 3
— 6 fewer )
The median duration of hospital stay Very low

Duration of was 4 days shorter with casirivimab- Due to serious
Based on data from:
hospitalization imdevimab (13 days vs. 17 days). risk of bias, The impact on duration
serious of hospitalization is very
indirectness, and uncertain.
controlled)
very serious
imprecision 4
1. Imprecision: serious. Single study.

2. Imprecision: serious. Single study.
3. Risk of bias: serious. Imprecision: serious.
4. Risk of bias: serious. Indirectness: serious. Imprecision: very serious.

Summary
Evidence summary
The NMA evidence summary was informed by one large trial (RECOVERY) in patients with severe and critical illness that
enrolled 9785 patients, most of whom received corticosteroids (83). The trial was registered and presented in preprints
when the data was reviewed by the GDG. The Table shows trial characteristics.
usual care for the outcomes of interest in patients with severe and critical COVID-19, with certainty ratings.
51 of 109
Certainty of
Intervention
Casirivimab-
imdevimab
evidence)
Mortality Odds Ratio 0.94 130 122

Critical or severe (CI 95% 0.86 — 1.03) per 1000 per 1000 Low Casirivima-imdevimab
disease Based on data from Due to serious may not have an
9785 participants in Difference: 8 fewer per 1000 indirectness and important effect on
( CI 95% 18 fewer imprecision 1 mortality.
controlled)
— 4 more )
Mechanical
(CI 95% 0.87 — 1.04) per 1000 per 1000
ventilation Due to serious The impact on
Based on data from
risk of bias, mechanical ventilation is
6637 participants in Difference: 4 fewer per 1000 indirectness, and very uncertain.
controlled)
— 3 more )
Allergic 3 9
per 1000 per 1000 Casirivimab-imdevimab
reactions Based on data from
Moderate
probably do not result in
Due to serious
15 406 participants in an important increase in
4 studies. (Randomized allergic reactions.
( CI 95% 1 fewer
controlled)
— 29 more )
Adverse effects 2 1 Casirivimab-imdevimab

leading to drug per 1000 per 1000
Based on data from do not result in an
discontinuation High important increase in
Difference: 1 fewer per 1000 adverse effects leading
( CI 95% 2 fewer to drug discontinuation.
controlled)
— 1 more )
Duration of Patients in both groups had the same Very low

Based on data from:
hospitalization median duration of hospitalization (10 Due to serious The impact on duration
days). risk of bias and of hospitalization is very
very serious uncertain.
controlled)
imprecision 4
1. Indirectness: serious. Imprecision: serious.

2. Risk of bias: serious. Indirectness: serious. Imprecision: serious.
3. Imprecision: serious.
4. Risk of bias: serious. Imprecision: very serious.

Casirivimab and imdevimab are two fully human antibodies (REGN10933 and REGN10987). Their mechanism of action is very
plausible: they bind to the SARS-CoV-2 spike protein (85) and have demonstrated antiviral activity in rhesus macaques and
Syrian golden hamsters (86). Pharmacokinetic data in patients with non-severe COVID-19 show that antiviral concentrations of
both antibodies are achieved and maintained for at least 28 days after intravenous administration of the combination at a
total dose of 1200 mg (600 mg each antibody) or above (42). Antiviral concentrations are also achieved and maintained using a
subcutaneous total dose of 1200 mg (600 mg of each antibody) in uninfected individuals for prophylaxis (79). Half-lives range
from 25 to 37 days for both antibodies. Data are currently unavailable for the pharmacokinetics of casirivimab and imdevimab in
52 of 109
severe and critical COVID-19, which are important because serum concentrations of other monoclonal antibodies have been
reported to be lower during systemic inflammation and correlated with albumin and CRP levels (87). Data available also suggest
that when delivered in combination, activity remains for currently circulating variants of concern (88).
While the mechanism is plausible, it was postulated that administration might have differential effects in patients who have
produced their own anti-SARS-CoV-2 spike protein antibodies (hereafter seropositive) compared with those who have not
(hereafter seronegative). It was hypothesized that effects might be larger, or restricted to, seronegative individuals who have not
yet mounted an effective antibody response.
Data describing the in vitro neutralisation of different variants by monocolonal antibodies are collated on the NIH NCATS
OpenData Portal (https://opendata.ncats.nih.gov/variant/activity). Several reports have demonstrated that in vitro neutralisation
of pseudovirus containing the BA1 omicron spike protein and in vitro neutralisation of authentic BA1 omicron virus is
dramatically reduced or lost for casirivumab and imdevimab when studied individually, and completely lost when both
monoclonal antibodies are combined. Furthermore, the combination of casirivumab and imdevimab has no impact upon
subgenomic viral RNA in the lungs or nasal turbinate of K18 human ACE2 transgenic mice infected with BA1 omicron
(https://www.biorxiv.org/content/10.1101/2022.01.23.477397v1.full.pdf+html). Therefore, currently available preclinical data
do not support activity of the casirivumab and imdevimab combination against the BA1 omicron variant.
6.6 IL-6 receptor blockers (published 6 July 2021)
Info Box
The recommendation concerning IL-6 receptor blockers (tocilizumab or sarilumab) was published on 6 July 2021 as the fifth
version of the WHO living guideline and in the BMJ as Rapid Recommendations. It followed the publication of RECOVERY and
REMAP-CAP trial publications in February 2021, and new trial data from 1020 patients randomized head-to-head to either
tocilizumab or sarilumab in REMAP-CAP being made available to the WHO on 1 June 2021. No changes were made for the IL-6
receptor blocker recommendation in this ninth version of the guideline.
WHO has made a strong recommendation for JAK inhibitors, specifically baricitinib, in patients with severe and critical
COVID-19. An IL-6 receptor blocker and baricitinib should not be given together and should be viewed as alternatives. These
new considerations are provided under ‘Justification’ for the recommendation for IL-6 receptor blockers, and are unchanged in
this ninth version of the guideline.
We recommend treatment with IL-6 receptor blockers (tocilizumab or sarilumab) (strong recommendation for).
• Corticosteroids have previously been strongly recommended in patients with severe and critical COVID-19 (see Section 6.11), and
we recommend patients meeting these severity criteria should now receive both corticosteroids and IL-6 receptor blockers.
• Baricitinib, a JAK inhibitor, is now recommended for the treatment of patients with severe and critical COVID-19 (see Section 6.2).
An IL-6 receptor blocker and baricitinib should not be given together and should be viewed as alternatives. The choice of whether
to use baricitinib or an IL-6 receptor blocker depends on availability as well as clinical and contextual factors (see Justification).
Practical Info
Route: IL-6 receptor blockers are administered intravenously for the treatment of patients with severe or critical COVID-19;
subcutaneous administration is not used in this case. IL-6 receptor blocker therapy should be administered in combination with
systemic corticosteroids, which may be administered both orally and intravenously, with due consideration to their high
bioavailability but possible malabsorption in the case of intestinal dysfunction with critical illness.
Duration: Tocilizumab and sarilumab are administered as single intravenous doses, typically over 1 hour. A second dose may be
administered 12 to 48 hours after the first dose; this was offered variably in major clinical trials at the discretion of treating
53 of 109
clinicians if a clinical response was felt to be inadequate. Duration of concurrent systemic corticosteroids is typically up to 10
days, though may vary between 5 and 14 days.
Dose: Tocilizumab is dosed at 8 mg per kilogram of actual body weight, up to a maximum of 800 mg. Sarilumab is most
commonly dosed at 400 mg, consistent with what was used in REMAP-CAP. Renal dose adjustment is not currently warranted
for either drug.
Monitoring: Routine bloodwork including neutrophil count, platelets, transaminases, and total bilirubin should be checked
prior to initiation of therapy. All patients should be monitored for signs and symptoms of infection, given the increased risk
with immunosuppression in addition to systemic corticosteroids. Patients on longer term IL-6 receptor blocker therapy are at
risk of active tuberculosis, invasive fungal infections and opportunistic pathogens. Risks and benefits of therapy should be
considered carefully in patients with any active, severe infection other than COVID-19; caution is advised when considering
the use of tocilizumab in patients with a history of recurring or chronic infections or with underlying conditions which may
predispose them to infections.
Timing: IL-6 receptor blockers should be initiated with systemic corticosteroids; specific timing during hospitalization or the
course of illness is not specified. That being said, IL-6 receptor blockers have been administered early in the course of
hospitalization in the included trials and clinicians may consider this approach if possible. See section on resource implications,
equity and human rights.
Benefits and harms

IL-6 receptor blockers reduce mortality and need for mechanical ventilation based on high certainty evidence. Low certainty
evidence suggests they may also reduce duration of mechanical ventilation and hospitalization (3)(89)(90).
The evidence regarding the risk of serious adverse events (SAEs) is uncertain. Low certainty evidence suggested that the risk
of bacterial infections in the context of immunosuppression treatment with IL-6 receptor blockers may be similar to usual
care (1). However the GDG had some concerns that, given the short-term follow-up of most trials and the challenges
associated with accurately capturing adverse events such as bacterial or fungal infection, the evidence summary may under-
represent the risks of treatment with IL-6 receptor blockers. Furthermore, the trials of IL-6 receptor blockers that inform this
recommendation were mostly performed in high-income countries where the risk of certain infectious complications may be
less than in some other parts of the world, and so the generalizability of the data on adverse events is unclear. We did not
have any data examining differential risk of harm based on whether patients received one or two doses of IL-6 receptor
blocker.
Subgroup analyses indicated no effect modification based on IL-6 receptor blocker drug (sarilumab or tocilizumab) or disease
severity (critical vs severe) and therefore this recommendation applies to all adult patients with either severe or critical
COVID-19 (84). We were unable to examine subgroups based on elevation of inflammatory markers or age due to
insufficient trial data (see Research evidence). Subgroup analyses evaluating baseline steroid use found greater benefit of
IL-6 receptor blockers in patients receiving steroids compared with those who were not (p=0.026), demonstrating that
steroid use does not abolish and might enhance the beneficial effect of IL-6 receptor blockers. Since steroids are already
strongly recommended in patients with severe and critical COVID-19, we did not formally evaluate the credibility of this
subgroup analysis as there would be no rationale for a subgroup recommendation for patients not receiving corticosteroids.

Certainty of evidence was rated as high for mortality and need for mechanical ventilation. Certainty in duration of
mechanical ventilation was rated as low due to serious risk of bias due to concerns regarding lack of blinding in included
trials, and for imprecision as the lower limit of the confidence interval suggested no effect. Certainty in duration of
hospitalization was rated as low due to serious risk of bias from lack of blinding in included trials, and for inconsistency
related to differences in point estimates and lack of overlap in confidence intervals.
Certainty in serious adverse events was rated as very low due to risk of bias related to lack of blinding and ascertainment
bias, and very serious imprecision due to very wide confidence intervals which did not rule out important benefit or harm;
certainty in risk of bacterial or fungal infections was rated as low due to similar concerns regarding serious risk of bias and
serious imprecision.
54 of 109
Certainty in evidence was rated as moderate when comparing the effect on mortality between tocilizumab and sarilumab
due to issues with imprecision.

Applying the agreed values and preferences (see Section 7), the majority of the GDG inferred that almost all well-informed
patients would want to receive IL-6 receptor blockers. The benefit of IL-6 receptor blockers on mortality was deemed of
critical importance to patients, despite the very low certainty around serious adverse events. The GDG anticipated little
variation in values and preferences between patients for this intervention.

The GDG noted that, compared with some other candidate treatments for COVID-19, IL-6 receptor blockers are more
expensive and the recommendation does not take account of cost-effectiveness. Currently, access to these drugs is
challenging in many parts of the world, and without concerted effort is likely to remain so, especially in resource-poor areas.
It is therefore possible that this strong recommendation for IL-6 receptor blockers could exacerbate health inequity. On the
other hand, given the demonstrated benefits for patients, it should also provide a stimulus to engage all possible
mechanisms to improve global access to these treatments. Individual countries may formulate their guidelines considering
available resources and prioritize treatment options accordingly.
At a time of drug shortage, it may be necessary to prioritize use of IL-6 receptor blockade through clinical triage (6). Many
jurisdictions have suggested mechanisms for triaging use of these treatments. These include prioritizing patients with the
highest baseline risk for mortality (e.g. those with critical disease over those with severe disease), in whom the absolute
benefit of treatment is therefore greatest. For example, despite consistent relative effects (OR 0.86 for mortality) with IL-6
receptor blockers, the absolute risk reduction for mortality in the critically ill would be 31 fewer deaths per 1000 (95% CI 11
to 47 fewer deaths) and in the severely ill would be 13 fewer deaths per 1000 (95% CI 5 to 19 fewer deaths).
Other suggestions for prioritization, which lack direct evidence, include focusing on patients with an actively deteriorating
clinical course and avoiding IL-6 receptor blocker therapy in those with established multi-organ failure (in whom the benefit
is likely to be smaller).

As IL-6 receptor blockers require intravenous administration, this treatment would be primarily indicated for patients with
severe and critical COVID-19 who require hospitalization. IL-6 receptor blockers are relatively easy to administer, and only
require one, or at most, two doses.
Justification
When moving from evidence to the strong recommendation to use IL-6 receptor blockers (tocilizumab or sarilumab) in patients
with severe or critical COVID-19, the GDG emphasized the high certainty evidence of improved survival and reduction in need
for mechanical ventilation. Additional trial data from REMAP-CAP (see Research Evidence section) provided more conclusive
evidence regarding the equivalence of tocilizumab and sarilumab.
The GDG acknowledged the uncertain data regarding SAEs and bacterial infections, but felt that the evidence of benefit for the
two most important patient outcomes warranted a strong recommendation. Costs and access were important considerations
and it was recognized that this recommendation could exacerbate health inequities. Hopefully this strong recommendation will
provide impetus to address these concerns and ensure access across regions and countries. The GDG did not anticipate
important variability in patient values and preferences, and judged that other contextual factors would not alter the
recommendation (see Evidence to Decision).
Subgroup analyses
The GDG did not find any evidence of a subgroup effect across patients with different levels of disease severity (severe vs
critical), or by IL-6 receptor blocker drug (tocilizumab vs sarilumab).
There were insufficient data to assess subgroup effect by elevation of inflammatory markers or age. Although the GDG
55 of 109
considered a subgroup analysis of patients receiving corticosteroids at baseline as compared with those that were not, the panel
did not see a need to consider subgroup recommendations for IL-6 receptor blockers in those not receiving corticosteroids as all
severe and critical COVID-19 patients should be receiving corticosteroids (see previous strong recommendation below). Taken
together, the GDG felt that the recommendation applies to both tocilizumab and sarilumab and all adult patients with severe or
critical COVID-19.
The role of IL-6 receptor blockers and baricitinib
WHO has recently made a strong recommendation for baricitinib in patients with severe and critical COVID-19. Both classes of
drugs are immune modulators with overlapping effects on immune responses. There is therefore a reasonable possibility that
their effects on COVID-19 will not be additive. Moreover, it is possible that when given together, adverse effects, including
secondary bacterial or fungal infection, will be greater. In the absence of evidence of incremental benefit of the drugs when
given together, the GDG advises that clinicians do not administer the drugs together.
See justification section for JAK inhibitors and related Summary of Findings table (see Research evidence) for more detailed
discussion regarding selecting between therapies.
Applicability
uncertain. However, the GDG had no reason to think that children with COVID-19 would respond any differently to treatment
with IL-6 receptor blockers. This is especially true given tocilizumab is used in children safely for other indications including
polyarticular juvenile rheumatoid arthritis, systemic onset of juvenile chronic arthritis, and chimeric antigen receptor T-cell
induced cytokine release syndrome. Sarilumab is not approved in children, so if an IL-6 receptor blocker is used in this
population, tocilizumab is preferred. The GDG also recognized that in many settings children are commonly admitted to hospital
with acute respiratory illnesses caused by other pathogens; as a result, it may be challenging to determine who is ill with severe
COVID-19, even with a positive test, and therefore likely to benefit from IL-6 receptor blockade. There were similar
considerations in regard to pregnant women, with no data directly examining this population, but no rationale to suggest they
would respond differently than other adults. The drug may, however, cross the placental membrane, although it is uncertain
what effect transient immunosuppression in the fetus may have and this should be weighed against the potential benefit for the
mother.

Population: Patients with severe and critical COVID-19
Comparator: Interleukin-6 receptor blockers
Certainty of
Comparator
IL-6 receptor
blockers
evidence)

Mortality (CI 95% 0.53 — 1.1) per 1000 per 1000 Due to serious
Based on data from imprecision and Baricitinib may reduce
2659 participants in 3 Difference: 22 fewer per ongoing mortality.
studies. (Randomized 1000 recruitment in a
controlled) ( CI 95% 52 fewer large RCT 1
— 9 more )
Mechanical
(CI 95% 0.61 — 1.6) per 1000 per 1000 Low
ventilation There may be little or no
difference on mechanical
2434 participants in 2 Difference: 2 more per 1000 serious
ventilation.
56 of 109
Certainty of
Comparator
IL-6 receptor
blockers
evidence)
Adverse effects 0 1
There is probably little to
discontinuation Based on data from no difference in adverse
Due to serious
2309 participants in 4 effects leading to
studies. (Randomized discontinuation.
( CI 95% 11 fewer
controlled)
— 15 more )
8.1 11.2 Very low

Hospital length Lower better days (Median) days (Mean) Due to serious
of stay risk of bias, The impact on hospital
Based on data from:
Difference: MD 3.1 more serious length of stay is very
( CI 95% 3.8 inconsistency, and uncertain.
fewer — 9.9 very serious
controlled)
more ) imprecision 4

mechanical days (Median) days (Mean) Low
Lower better Baricitinib may reduce
Based on data from: 328 duration of mechanical
Difference: MD 2.2 fewer risk of bias and
participants in 2 studies. ventilation.
( CI 95% 5.3 fewer imprecision 5
8.4 8.9
Lower better Low
stability There may not be an
Based on data from: Due to serious
Difference: MD 0.5 more important impact on
( CI 95% 2.3 time to clinical stability.
studies. (Randomized imprecision 6
more )

2. Risk of bias: no serious. Most of the data on interleukin-6 receptor blockers comes from trials that were unblinded.
Imprecision: very serious. The credible interval includes important benefit and important harm.
3. Imprecision: serious. The credible interval includes small but important harm.
Inconsistency: serious. The trials that studied interleukin-6 receptor blockers had discrepant results: some increased length
of stay, others reduced length of stay. Imprecision: very serious. The credible interval includes important benefit and
important harm.
Imprecision: serious. The credible interval includes no important difference.
Imprecision: serious. Credible interval includes important harm and important benefit (using a minimal important difference
threshold of 1 day).

Intervention: Interleukin-6 receptor blockers
57 of 109
Summary
Evidence summary
The LNMA (8) on IL-6 receptor blockers was informed by 30 RCTs with 10 618 participants and provided relative
estimates of effect for all patient-important outcomes except mortality, which came from the prospective meta-analysis
(PMA) (90). Of the trials included in the LNMA, all were registered and examined patients with severe or critical illness
related to COVID-19 (trial characteristics table available upon request). Of the trials, 37% were published in peer-
reviewed journals, 3% were available as preprints and 60% were completed but unpublished.
The evidence summary for mortality was based on 27 RCTs and 10 930 participants from the PMA (90). We used the
PMA for mortality as it included some additional unpublished data that reported on this outcome. The GDG recognized
that usual care is likely variable between centres and regions, and has evolved over time. However, given all of the data
come from RCTs, use of these co-interventions that comprise usual care would be expected to be balanced between
study patients randomized to either the intervention or usual care arms.
The GRADE Summary of Findings table shows the relative and absolute effects of IL-6 receptor blockers compared to
usual care for the outcomes of interest in patients with severe and critical COVID-19, with certainty ratings. See Section
7 for sources of baseline risk estimates informing absolute estimates of effect.
Subgroup analysis
All included RCTs evaluated IL-6 receptor blockers exclusively in severely or critically ill adults with COVID-19 requiring
hospitalization. The GDG requested subgroup analyses based on age (< 70 years versus older), disease severity (severe
versus critical), levels of inflammatory markers and baseline corticosteroid use for the following outcomes: mortality,
need for and duration of mechanical ventilation, duration of hospitalization, and risks of SAEs and bacterial infections.
Based on subgroup analyses, the GDG determined that there was no subgroup effect across any pre-specified outcomes
of interest based on disease severity. The GDG considered the results of a subgroup analysis of all included RCTs based
on systemic corticosteroid use for the outcome of mortality. The analysis suggested that the relative effects of IL-6
receptor blockers varied as a function of the use of systemic corticosteroids at baseline. Crucially, steroids did not
abolish and may even enhance the beneficial effect of IL-6 receptor blockers on mortality. For reasons described below,
the GDG did not formally evaluate the credibility of this subgroup analysis.
When comparing tocilizumab and sarilumab, based on the PMA, there was no evidence of a subgroup effect (90).
However, there were more data, and therefore greater precision, for tocilizumab+steroids versus steroids alone (OR
0.77, 95% CI 0.68–0.87) as compared to sarilumab+steroids versus steroids alone (OR 0.92, 95% CI 0.61–1.38). In
addition to these subgroup data, the GDG reviewed head-to-head data from REMAP-CAP investigators which
demonstrated no difference between tocilizumab as compared with sarilumab in a population of patients all receiving
corticosteroids (36.5% mortality with tocilizumab, 33.9% mortality with sarilumab). The NMA estimate of
tocilizumab+steroids versus sarilumab+steroids, incorporating both direct and indirect data, provided moderate certainty
data of no difference between the drugs (OR 1.07, 95% CI 0.86–1.34) (1)(3).
Certainty of
Intervention
IL-6 receptor
blockers
evidence)
Mortality
Odds Ratio 0.86 130 114
(severe and (CI 95% 0.79 — 0.95) per 1000 per 1000
critically ill Based on data from IL-6 receptor blockers
patients) Difference: 16 fewer per High
10 930 participants in reduce mortality.
27 studies. 1 1000
— 6 fewer )
Mechanical
(CI 95% 0.57 — 0.9) per 1000 per 1000
ventilation IL-6 receptor blockers
Based on data from
High reduce need for
5686 participants in 9 Difference: 23 fewer per mechanical ventilation.
studies. 2 (Randomized 1000
— 8 fewer )
58 of 109
Certainty of
Intervention
IL-6 receptor
blockers
evidence)
Adverse events Odds Ratio 0.5 9 5 Very low The effect of IL-6
leading to drug (CI 95% 0.03 — 9.08) per 1000 per 1000 Due to serious receptor blockers on
discontinuation Based on data from 815
risk of bias and adverse events leading
3
Difference: 4 fewer per 1000 very serious to discontinuation is
(Randomized ( CI 95% 9 fewer imprecision 4 uncertain.
Bacterial
(CI 95% 0.72 — 1.29) per 1000 per 1000 IL-6 receptor blockers
infections Due to serious
Based on data from may not increase
risk of bias and
3548 participants in 18 Difference: 5 fewer per 1000 secondary bacterial
serious
studies. (Randomized infections.
Duration of 14.7 13.5 Low

mechanical Lower better (Mean) (Mean) Due to serious IL-6 receptor blockers
ventilation Based on data from:
risk of bias and may reduce duration of
1189 participants in 10 Difference: MD 1.2 lower serious mechanical ventilation.
( CI 95% 2.3 lower imprecision 6
controlled) — 0.1 lower )

Lower better (Mean) (Mean)
hospitalization Due to serious IL-6 receptor blockers
Based on data from:
risk of bias and may reduce duration of
6665 participants in 9 Difference: MD 4.5 lower serious hospitalization.
( CI 95% 6.7 lower inconsistency 7
controlled) — 2.3 lower )
1. . Baseline/comparator: Primary study[15]. Baseline risk for mortality and mechanical ventilation were derived from the
WHO SOLIDARITY trial for patients with severe and critical COVID-19, adjusted for corticosteroids as part of standard of
care (16% baseline risk x RR 0.79 for corticosteroids = 13%). The control arm of the WHO SOLIDARITY trial, performed
across a wide variety of countries and geographical regions, was identified by the GDG panel as generally representing the
most relevant source of evidence for baseline risk estimates for mortality and mechanical ventilation for severely and
critically ill patients with COVID-19.
2. Systematic review [3] . Baseline/comparator: Primary study. Baseline risk for mortality and mechanical ventilation were
derived from the WHO SOLIDARITY trial for patients with severe and critical COVID-19, adjusted for corticosteroids as
part of standard of care (16% baseline risk x RR 0.79 for corticosteroids = 13%). The control arm of the WHO SOLIDARITY
trial, performed across a wide variety of countries and geographical regions, was identified by the GDG panel as generally
representing the most relevant source of evidence for baseline risk estimates for mortality and mechanical ventilation for
severely and critically ill patients with COVID-19.
3. Systematic review. Baseline/comparator: Control arm of reference used for intervention. We used the median event
rate for all patients randomized to usual care across included studies. Supporting references: [3].
4. Risk of bias: serious. We downgraded for some concerns regarding risk of bias due to lack of blinding and ascertainment
bias. Imprecision: very serious. We downgraded due to very wide confidence intervals crossing the null.
5. Risk of bias: serious. We downgraded for some concerns regarding risk of bias due to lack of blinding and ascertainment
bias. Imprecision: serious. Downgraded due to wide confidence intervals crossing the null.
6. Risk of bias: serious. We downgraded for some concerns regarding risk of bias due to lack of blinding. Imprecision:
serious. We downgraded as the lower limit of the confidence interval was close to the null.
7. Risk of bias: serious. We downgraded for some concerns regarding risk of bias due to lack of blinding. Inconsistency:
serious. Downgraded due to differences in point estimates and lack of overlap in confidence intervals.
59 of 109

IL-6 is a pleiotropic cytokine which activates and regulates the immune response to infections. Elevated IL-6 concentrations are
associated with severe outcomes in COVID-19, including respiratory failure and death, although the role of IL-6 in disease
pathogenesis is unclear.
Tocilizumab and sarilumab are monoclonal antibodies approved for use in rheumatoid arthritis. They antagonize the membrane
bound and soluble forms of the IL-6 receptor (IL-6R/sIL-6R). Tocilizumab is approved for intravenous use in rheumatoid arthritis
and sarilumab for subcutaneous use, although in COVID-19 both have been studied intravenously. At the studied doses in
COVID-19, both medicines are expected to achieve very high levels of receptor occupancy based upon studies in rheumatoid
arthritis (29). IL-6 receptor blockers are being repurposed in terms of indication but not in terms of the primary pharmacological
mechanism of action. Efficacy in COVID-19 depends upon the importance of IL-6 signalling in the pathophysiology of the
disease, rather than upon whether the doses used achieve target concentrations.
6.7 Ivermectin (published 31 March 2021)
Info Box
The recommendation concerning ivermectin was published on 31 March 2021 as the fourth version of the WHO living
guideline and in the BMJ as Rapid Recommendations. It followed the increased international attention on ivermectin as a
potential therapeutic option.
No changes were made for the ivermectin recommendation in this ninth version of the guideline. We are aware of a few new,
relatively small trials published since our recommendation was made and that one key trial has since been retracted given
concerns about research fraud (91)(92). However, the updated evidence summary from the LNMA is consistent with our
previously made recommendation. This updated evidence summary will be fully considered by the GDG in subsequent
iterations of the guideline.
For patients with COVID-19, regardless of disease severity
Only in research settings
We recommend not to use ivermectin, except in the context of a clinical trial (recommended only in research settings).
Remark: This recommendation applies to patients with any disease severity and any duration of symptoms.
A recommendation to only use a drug in the setting of clinical trials is appropriate when there is very low certainty evidence and future
research has a large potential for reducing uncertainty about the effects of the intervention and for doing so at reasonable cost.
Practical Info
The GDG made a recommendation against using ivermectin for treatment of patients with COVID-19 outside the setting of a
clinical trial and therefore practical considerations are less relevant for this drug.
60 of 109
Benefits and harms

The effects of ivermectin on mortality, mechanical ventilation, hospital admission, duration of hospitalization and viral
clearance remain uncertain because of very low certainty of evidence addressing each of these outcomes. Ivermectin may
have little or no effect on time to clinical improvement (low certainty evidence). Ivermectin may increase the risk of SAEs
leading to drug discontinuation (low certainty evidence).
Subgroup analyses indicated no effect modification based on dose. We were unable to examine subgroups based on patient
age or severity of illness due to insufficient trial data (see Research evidence). Therefore, we assumed similar effects in all
subgroups. This recommendation applies to patients with any disease severity and any duration of symptoms.

For most key outcomes, including mortality, mechanical ventilation, hospital admission, duration of hospitalization and viral
clearance, the GDG considered the evidence of very low certainty. Evidence was rated as very low certainty primarily
because of very serious imprecision for most outcomes: the aggregate data had wide confidence intervals and/or very few
events. There were also serious concerns related to risk of bias for some outcomes, specifically lack of blinding, lack of trial
pre-registration, and lack of outcome reporting for one trial that did not report mechanical ventilation despite pre-specifying
it in their protocol (publication bias).
For more details, see the Justification section for this recommendation. For other outcomes, including SAEs and time to
clinical improvement, the certainty of the evidence was low.

Applying the agreed values and preferences (see Section 7), the GDG inferred that almost all well-informed patients would
want to receive ivermectin only in the context of a randomized trial, given that the evidence left a very high degree of
uncertainty in effect on mortality, need for mechanical ventilation, need for hospitalization and other critical outcomes of
interest and there was a possibility of harms, such as treatment-associated SAEs. The panel anticipated little variation in
values and preferences between patients when it came to this intervention.

Ivermectin is a relatively inexpensive drug and is widely available, including in low-income settings. The low cost and wide
availability do not, in the GDG's view, mandate the use of a drug in which any benefit remains very uncertain and ongoing
concerns regarding harms remain. Although the cost may be low per patient, the GDG raised concerns about diverting
attention and resources away from care likely to provide a benefit such as corticosteroids in patients with severe COVID-19
and other supportive care interventions. Also, use of ivermectin for COVID-19 would divert drug supply away from
pathologies for which it is clearly indicated, potentially contributing to drug shortages, especially for helminth control and
elimination programmes. Other endemic infections that may worsen with corticosteroids should be considered. If steroids
are used in the treatment of COVID-19, empiric treatment with ivermectin may still be considered in Strongyloidiasis
endemic areas, at the discretion of clinicians overseeing treatment, albeit not for treatment of COVID-19 itself.
Justification
When moving from evidence to a recommendation on the use of ivermectin in patients with COVID-19 only in the context of a
clinical trial, the GDG emphasized the high degree of uncertainty in the most critical outcomes such as mortality and need for
mechanical ventilation. It also noted the evidence suggesting possible harm associated with treatment, with increased adverse
events. The GDG did not anticipate important variability in patient values and preferences. Other contextual factors, such as
resource considerations, accessibility, feasibility and impact on health equity did not alter the recommendation.
Compared with previous drugs evaluated as part of the WHO Therapeutics and COVID-19: living guideline, currently there are far
fewer RCT data available for ivermectin. The existing data on ivermectin also have a substantially higher degree of uncertainty,
with included trials having enrolled substantially fewer patients with far fewer events.
61 of 109
High degree of uncertainty

The certainty in effect estimates for ivermectin on the main outcomes of interest, including mortality, is very low and therefore
the effect of ivermectin on these outcomes remains uncertain. There are two domains that contribute to this uncertainty:
serious risk of bias; and serious imprecision. Although 16 RCTs contributed to the evidence summary informing this drug, only
five directly compared ivermectin with standard of care and reported mortality (93)(94)(95)(96)(97)(98)(99). Of note, and in
keeping with our methodology, the LNMA team excluded quasi-randomized trials, or any RCT that did not use explicit
randomization techniques. Of these five RCTs, two (93)(94) were at high risk of bias, due to inadequate blinding. One of these
two trials (93) also started enrolling and randomizing patients prior to the protocol being publicly posted, another factor that
contributes to an increased risk of bias. The potential impact of risk of bias is exemplified by subgroup analyses for mortality
based on trial risk of bias. As demonstrated in the forest plot (Fig. 3), the pooled estimate across all five RCTs that directly
compare ivermectin with standard care suggests a reduction in mortality with ivermectin, but this effect is not apparent if we
only consider the trials at low risk of bias (which together contribute nearly two-thirds of the evidence). This finding increases
the degree of uncertainty regarding the true effect of ivermectin on mortality. Consistent with the direct evidence, a similar
phenomenon is observed with the indirect evidence comparing ivermectin to standard of care (via comparisons against
hydroxychloroquine and lopinavir/ritonavir). The indirect evidence suggesting a reduction in mortality with ivermectin is driven
almost entirely by one study which is at high risk of bias (91) due to a lack of detailed description of blinding or randomization
and the lack of a publicly available study protocol (figure not shown).
Fig. 3. Forest plot demonstrating direct comparison of ivermectin versus standard of care for mortality with subgroup analysis
by risk of bias
IV: inverse variance.
In addition to concerns related to risk of bias, for the outcome of mortality, there are very serious concerns related to
imprecision. According to GRADE, imprecision is evaluated based on both a confidence interval approach and an evaluation of
information size (event number), ensuring there is adequate information on which to make informed judgments (100). In this
case, despite confidence intervals that suggest benefit with ivermectin, the information size is very low. For mortality (and
ignoring the concerns related to risk of bias discussed above), there were nine deaths across all 511 patients randomized to
ivermectin (1.76%) and 22 deaths across all 404 patients randomized to standard of care (5.45%). This is an extremely small
number of events on which to base conclusions, and far below the optimal information size. In fact, performing
a theoretical exercise in which a change of three events (deaths) is made from those randomized to standard of care to those
randomized to ivermectin eliminates any statistical significance, a finding that suggests that results could reasonably be due to
chance alone. Furthermore, the evidence informing this comparison is from multiple small trials, adding to the risk of
unrecognized imbalances in study arms. Given the strong likelihood that chance may be playing a role in the observed findings,
the panel believed there was very serious imprecision further lowering the overall certainty in findings.
This combination of serious risk of bias and very serious imprecision contributed to very low certainty of evidence for mortality
despite a point estimate and confidence interval that appear to suggest benefit with ivermectin. As a result, the panel concluded
that the effect of ivermectin on mortality is uncertain. Similar considerations were applied to the other critical outcomes
including mechanical ventilation, hospital admission, and duration of hospitalization and resulted in very low certainty for these
outcomes as well.
62 of 109
Subgroup analyses
We conducted subgroup analysis only for effect by ivermectin dose and the panel did not find any evidence of a subgroup
effect (see Research evidence). A lack of within-trial comparisons prevented subgroup analyses by age or disease severity.
Therefore, the panel did not make any subgroup recommendation for this drug. In other words, the recommendation against
ivermectin except in the context of clinical trials is applicable across disease severity, age groups, and all dose regimens of
ivermectin.
Applicability
None of the included RCTs enrolled children under 15, and therefore the applicability of this recommendation to children is
currently uncertain. However, the panel had no reason to think that children with COVID-19 would respond any differently to
treatment with ivermectin. There were similar considerations for pregnant women, with no data directly examining this
population, but no rationale to suggest they would respond differently to other adults.

Population: Patients with COVID-19 (all disease severities)
Intervention: Ivermectin
Summary
Evidence summary
The LNMA on ivermectin was based on 16 RCTs and 2407 participants. Of the included studies, 75% examined patients
with non-severe disease and 25% included both severe and non-severe patients. A number of the included studies did
not report on our outcomes of interest. Of the studies, 25% were published in peer-reviewed journals, 44% were
available as preprints and 31% were completed but unpublished (see Table on trial characteristics). We excluded a
number of quasi-RCTs (101)(102)(103)(104).
The GRADE Summary of Findings table shows the relative and absolute effects of ivermectin compared to usual care for
the outcomes of interest in patients with COVID-19, with certainty ratings. See Section 7 for sources of baseline risk
estimates informing absolute estimates of effect.
Subgroup analysis
The NMA team performed subgroup analyses which could result in distinct recommendations by subgroups. From the
available data, subgroup analyses were only possible by dose of ivermectin and considering the outcomes of mortality,
mechanical ventilation, admission to hospital, and adverse events leading to drug discontinuation. The ivermectin dose
subgroup analyses were performed from the direct comparison of ivermectin versus usual care. For these analyses,
meta-regression was used to evaluate the effect of cumulative dose as a continuous variable, and further adding a co-
variate for single vs multiple dosing regimens. This approach was based on input from the pharmacology experts (led by
Professor Andrew Owen) who performed pharmacokinetic simulations across trial doses, and found that cumulative
ivermectin dose was expected to correlate with key pharmacokinetic parameters when single- and multiple-dose studies
were segregated. It should be noted that the included trials did not directly assess the pharmacokinetics of ivermectin,
and our approach was based upon simulations validated where possible against published pharmacokinetics in humans.
The panel used a pre-specified framework incorporating the ICEMAN tool to assess the credibility of subgroup
findings (84).
The GDG panel requested subgroup analyses based on: age (considering children vs younger adults vs older adults [70
years or older]); illness severity (non-severe vs severe vs critical COVID-19); time from onset of symptoms; and use of
concomitant medications. However, there was insufficient within-trial data to perform any of these subgroup analyses,
based on our pre-specified protocol. The panel recognized that usual care is likely variable between centres and regions,
and has evolved over time. However, given all of the data come from RCTs, use of these co-interventions that comprise
usual care should be balanced between study patients randomized to either the intervention or usual care arms.
Certainty of
Timeframe measurements Standard care Ivermectin (Quality of summary
evidence)
Odds Ratio 0.19 Very low

Mortality
(CI 95% 0.09 — 0.36)
70 14 Due to serious The effect of ivermectin
Based on data from per 1000 per 1000 risk of bias and on mortality is uncertain.
very serious
imprecision 2
63 of 109
Certainty of
evidence)
7 studies. 1 1000
(Randomized ( CI 95% 63 fewer
20 10
Mechanical Odds Ratio 0.51 per 1000 per 1000 Very low
ventilation (CI 95% 0.12 — 1.77) Due to very The effect of ivermectin
Based on data from 687 serious on mechanical
participants in 5 studies. imprecision and ventilation is uncertain.
1000
(Randomized controlled) ( CI 95% 18 fewer publication bias 3
— 15 more )
500 618
Viral clearance Odds Ratio 1.62 per 1000 per 1000 Low
7 days (CI 95% 0.95 — 2.86) Ivermectin may increase
Due to serious
Based on data from 625 or have no effect on viral
Difference: 118 more per inconsistency and
participants in 6 studies. clearance.
1000 imprecision 4
— 241 more )
Hospital 50 18
admission Very low
(CI 95% 0.08 — 1.48) The effect of ivermectin
(outpatients only) Due to extremely
Based on data from 398 Difference: 32 fewer per on hospital admission is
serious
participants in 1 study. 1000 uncertain.
imprecision 5
— 23 more )
Serious adverse Odds Ratio 3.07 9 27

per 1000 per 1000 Low Ivermectin may increase
events (CI 95% 0.77 — 12.09)
Due to very the risk of serious
Based on data from 584
Difference: 18 more per 1000 serious adverse events leading
( CI 95% 2 fewer imprecision 6 to drug discontinuation.
— 89 more )
11 10.5
Time to clinical Measured by: days days (Mean) days (Mean) Low Ivermectin may have
improvement Lower better
Due to very little or no difference on
Based on data from: 633 MD 0.5 fewer
Difference: serious time to clinical
( CI 95% 1.7 imprecision 7 improvement.
(Randomized controlled) fewer — 1.1
more )
12.8 11.7 Very low

Duration of Measured by: days days (Mean) days (Mean) Due to serious
hospitalization Lower better The effect of ivermectin
imprecision,
Based on data from: 252 MD 1.1 fewer on hospital length of
Difference: inconsistency and
participants in 3 studies. stay is uncertain.
( CI 95% 2.3 serious risk of bias
(Randomized controlled) fewer — 0.1 8
more )
64 of 109
Certainty of
evidence)
Time to viral Measured by: days 7.3 5.7 Very low

days (Mean) days (Mean) Due to very We are uncertain
clearance Lower better
serious whether ivermectin
Based on data from: 559
Difference: MD 1.6 fewer imprecision and improves or worsens
serious risk of bias time to viral clearance.
(Randomized controlled) ( CI 95% 4.1 9
fewer — 3 more )
1. Systematic review [1] . Baseline/comparator: Control arm of reference used for intervention. We elected to use the
control arm of the WHO SOLIDARITY trial, reflecting usual care across countries participating in the trial.
2. Risk of bias: serious. The large trial contributing most of the effect estimate was driven by studies that were not blinded.
Imprecision: very serious. The number of total events was very small.
3. Imprecision: very serious. Very few events and credible intervals that include both important benefit and harm.
Publication bias: serious.
4. Inconsistency: serious. The point estimates varied widely and credible intervals do not substantially overlap.
Imprecision: serious. Credible interval includes no effect.
5. Imprecision: extremely serious. Credible interval includes important benefit and harm.
6. Imprecision: very serious. Credible interval includes little to no difference.
8. Risk of bias: serious. Result driven by one study that was not blinded. Inconsistency: serious. Despite overlapping
confidence intervals, point estimates discrepant. Imprecision: serious. Credible intervals include no difference.
9. Risk of bias: serious. Concerns around risk of bias. Imprecision: very serious. Credible interval includes important benefit
and important harm.

Ivermectin is an antiparasitic agent that interferes with nerve and muscle function of helminths through binding glutamate-gated
chloride channels (105). Based on in vitro experiments, some have postulated that ivermectin may have a direct antiviral effect
against SARS-CoV-2. However, in humans the concentrations needed for in vitro inhibition are unlikely to be achieved by the
doses proposed for COVID-19 (106)(107)(108). Ivermectin had no impact on SARS-CoV-2 viral RNA in the Syrian golden
hamster model of SARS-CoV-2 infection (109). The proposed mechanism remains unclear: multiple targets have been proposed
based upon either analogy to other viruses with very different life cycles, or, like several hundred other candidates, simulations
indicating molecular docking with multiple viral targets including spike, RdRp and 3CLpro (110)(111)(112)(113)(114). No direct
evidence for any mechanism of antiviral action against SARS-CoV-2 currently exists.
Some have proposed, based predominantly upon research in other indications, that ivermectin has an immunomodulatory effect,
but again the mechanism remains unclear. Historical data showed that ivermectin improved survival in mice given a lethal dose
of lipopolysaccharide (115), and has benefits in murine models of atopic dermatitis and allergic asthma (116)(117). For SARS-
CoV-2, one hypothesis suggests immunomodulation mediated by allosteric modulation of the alpha-7 nicotinic acetylcholine
receptor (indirectly by modulating the activity of ligands of the receptor). Although investigators have demonstrated this
action in vitro, concentrations used in these experiments have been even higher than those required for an antiviral effect (118),
and therefore very unlikely to be achieved in humans. In the Syrian golden hamster model of SARS-CoV-2 infection, ivermectin
resulted in some changes in pulmonary immune phenotype consistent with allosteric modulation of the alpha-7 nicotinic
acetylcholine receptor (109). However, ivermectin did not appear to rescue body weight loss which is a hallmark of disease in
this model, and drug concentrations were not measured to extrapolate to those achieved in humans. Taken together, there
remains great uncertainty regarding the relevance of any immunomodulatory or anti-inflammatory action of ivermectin.
65 of 109
6.8 Hydroxychloroquine (published 17 December 2020)

Info Box
The recommendation concerning hydroxychloroquine was published 17 December 2020 as the third version of the WHO living
guideline and in the BMJ as Rapid Recommendations. It followed the pre-print publication of the WHO SOLIDARITY trial on 15
October, 2020, reporting results on treatment with hydroxychloroquine, remdesivir and lopinavir/ritonavir in hospitalized
patients with COVID-19 (15). No changes were made for the hydroxychloroquine recommendation in this ninth version of the
guideline.
We recommend not to use hydroxychloroquine or chloroquine (strong recommendation against).
Practical Info
The GDG made a strong recommendation against using hydroxychloroquine or chloroquine for treatment of patients with
COVID-19. The use of hydroxychloroquine may preclude the use of other important drugs that also prolong the QT interval,
such as azithromycin and fluoroquinolones. Concomitant use of drugs that prolong the QT interval should be done with extreme
caution.
Benefits and harms

Hydroxychloroquine and chloroquine probably do not reduce mortality or mechanical ventilation and may not reduce
duration of hospitalization. The evidence does not exclude the potential for a small increased risk of death and mechanical
ventilation with hydroxychloroquine. The effect on other less important outcomes, including time to symptom resolution,
admission to hospital, and duration of mechanical ventilation, remains uncertain.
Hydroxychloroquine may increase the risk of diarrhoea and nausea/vomiting; a finding consistent with evidence from its use
in other conditions. Diarrhoea and vomiting may increase the risk of hypovolaemia, hypotension and acute kidney injury,
especially in settings where health care resources are limited. Whether or not and to what degree hydroxychloroquine
increases the risk of cardiac toxicity, including life-threatening arrhythmias, is uncertain.
Subgroup analyses indicated no effect modification based on severity of illness (comparing either critical vs severe/non-
severe or non-severe vs critical/severe) or age (comparing those aged < 70 years vs older). Further, the cumulative dose and
predicted Day 3 serum trough concentrations did not modify the effect for any outcome. Therefore, we assumed similar
effects in all subgroups.
We also reviewed evidence comparing the use of hydroxychloroquine plus azithromycin vs hydroxychloroquine alone. There
was no evidence that the addition of azithromycin modified the effect of hydroxychloroquine for any outcome (very low
certainty).

For the key outcomes of mortality and mechanical ventilation, the panel considered the evidence to be of moderate
certainty. There were residual concerns about lack of blinding in the largest trials and the imprecision. For example, the
credible interval around the pooled effect leaves open the possibility of a very small reduction in mortality. The quality of
evidence was low for diarrhoea and nausea/vomiting because of lack of blinding in many of the trials and because the total
number of patients enrolled in trials reporting these outcomes was smaller than the optimal information size (although the
credible interval laid entirely on the side of harm for both outcomes).
66 of 109
For all other outcomes, the certainty of the evidence was low or very low. The primary concerns with the data were
imprecision (credible intervals included both important benefit and important harm) as well as risk of bias (lack of blinding).

not want to receive hydroxychloroquine given the evidence suggesting there was probably no effect on mortality or need
for mechanical ventilation and there was a risk of adverse events including diarrhoea and nausea and vomiting. The panel
did not expect there would be much variation in values and preferences between patients when it came to this intervention.

Hydroxychloroquine and chloroquine are relatively inexpensive compared with other drugs used for COVID-19 and are
already widely available, including in low-income settings. Despite this, the panel felt that almost all patients would choose
not to use hydroxychloroquine or chloroquine because the harms outweigh the benefits. Although the cost may be low per
patient, the GDG panel raised concerns about diverting attention and resources away from care likely to provide a benefit
such as corticosteroids in patients with severe COVID-19 and other supportive care interventions.
Justification
When moving from evidence to the strong recommendation against the use of hydroxychloroquine or chloroquine for patients
with COVID-19, the panel emphasized the moderate certainty evidence of probably no reduction in mortality or need for
mechanical ventilation. It also noted the evidence suggesting possible harm associated with treatment, with increased nausea
and diarrhoea. The GDG did not anticipate important variability in patient values and preferences, and other contextual factors,
such as resource considerations, accessibility, feasibility and impact on health equity (see summary of these factors under
Evidence to decision).
Subgroup analyses
The panel did not find any evidence of a subgroup effect across patients with different levels of disease severity, between adults
and older adults, and by different doses, and therefore did not make any subgroup recommendation for this drug. In other
words, the strong recommendation is applicable across disease severity, age groups, and all doses and dose schedules of
hydroxychloroquine.
The trials included patients from around the world, with all disease severities, and treated in different settings (outpatient and
inpatient). Although the trials did not report subgroup effects by time from symptom onset, many of the trials enrolled patients
early in the disease course. The GDG panel therefore felt that the evidence applies to all patients with COVID-19.
Applicability
Special populations
uncertain. However, the panel had no reason to think that children with COVID-19 would respond any differently to treatment
with hydroxychloroquine. There were similar considerations in regards to pregnant women, with no data directly examining this
population, but no rationale to suggest they would respond differently than other adults. Hydroxychloroquine crosses the
placental barrier and there are concerns that it may lead to retinal damage in neonates. Although hydroxychloroquine has been
used in pregnant women with systemic autoimmune diseases, such as systemic lupus erythematosus, pregnant women may
have even more reasons than other patients to be reluctant to use hydroxychloroquine for COVID-19.
In combination with azithromycin

There was no evidence from the NMA that the addition of azithromycin modified the effect of hydroxychloroquine for any
outcome. As there were no trial data suggesting that azithromycin favourably modifies the effect of hydroxychloroquine, the
recommendation against hydroxychloroquine and chloroquine applies to patients whether or not they are concomitantly
receiving azithromycin.
67 of 109
Uncertainties
Please see end of document for residual uncertainties (Section 9). The GDG panel felt that it was unlikely future studies would
identify a subgroup of patients that are likely to benefit from hydroxychloroquine or chloroquine.

Intervention: Hydroxychloroquine
Summary
Evidence summary
The LNMA on hydroxychloroquine was based on 30 RCTs with 10 921 participants, providing relative estimates of
effect for patient-important outcomes (see Table). Five of the trials (414 total participants) randomized some patients to
chloroquine.
The GRADE Summary of Findings table shows the relative and absolute effects of hydroxychloroquine compared to
usual care for the outcomes of interest in patients with COVID-19, with certainty ratings. See Section 7 for sources of
baseline risk estimates informing absolute estimates of effect.
Subgroup analysis
For hydroxychloroquine, the GDG panel requested subgroup analyses based on age (considering children vs younger
adults [e.g. < 70 years] vs older adults [e.g. 70 years or older]), illness severity (non-severe vs severe vs critical
COVID-19) and based on whether or not it was co-administered with azithromycin.
The panel also requested a subgroup analysis based on high dose vs low dose hydroxychloroquine. A categorical
approach to hydroxychloroquine dosing proved impossible because the trials used varying loading doses, continuation
doses and durations. Therefore, in collaboration with a pharmacology expert (Professor Andrew Owen), we modelled the
expected serum concentrations over time. We hypothesized that higher trough concentrations early in the treatment
course (e.g. trough concentration on Day 3) might be more effective than lower early trough concentrations. We also
hypothesized that higher maximum serum concentrations (e.g. peak concentration on the last day) might result in higher
risk of adverse effects than lower maximum serum concentrations. In our pharmacokinetic model, the cumulative dose
was highly correlated with all measures of serum concentrations on Day 3 and the final day of treatment, and therefore
we decided to use cumulative dose as the primary analysis. Day 3 trough concentration was least strongly correlated
with total cumulative dose (R2 = 0.376) and therefore we performed a sensitivity subgroup analysis with predicted Day
3 trough concentrations for efficacy outcomes.
Certainty of
Intervention
Hydroxychloro
quine
evidence)
Odds Ratio 1.11 106 116

Hydroxychloroquine
Based on data from Due to borderline
probably does not
10 859 participants in Difference: 10 more per 1000 risk of bias and
reduce mortality.
29 studies. 1 ( CI 95% 5 fewer imprecision 2
(Randomized controlled) — 28 more )
Mechanical
Odds Ratio 1.2 105 123 Moderate
(CI 95% 0.83 — 1.81) per 1000 per 1000 Hydroxychloroquine
ventilation Due to borderline
Based on data from probably does not
risk of bias and
6379 participants in 5 Difference: 18 more per 1000 reduce mechanical
serious
studies. (Randomized ventilation.
Viral clearance Odds Ratio 1.08 483 502 Very low The effect of
7 days (CI 95% 0.25 — 4.78) Due to very hydroxychloroquine on
per 1000 per 1000
68 of 109
Certainty of
Intervention
Hydroxychloro
quine
evidence)
Based on data from 280

participants in 4 studies. Difference: 19 more per 1000 serious viral clearance is very
4
(Randomized ( CI 95% 294 imprecision 5 uncertain.
more )
47 19 Very low
Admission to Odds Ratio 0.39 per 1000 per 1000 Due to very The effect of
hospital (CI 95% 0.12 — 1.28)
serious hydroxychloroquine on
Based on data from 465 Difference: 28 fewer per imprecision and admission to hospital is
participants in 1 study. 1000 serious uncertain.
(Randomized controlled) ( CI 95% 41 fewer indirectness 6
— 12 more )
46 56 Very low
Cardiac toxicity per 1000 per 1000 The effect of
Based on data from Due to serious
hydroxychloroquine on
imprecision, risk
3287 participants in 7 cardiac toxicity is
Difference: 10 more per 1000 of bias, and
studies. (Randomized uncertain.
( CI 95% 0 more indirectness 7
controlled)
— 30 more )
149 255
Diarrhoea Low
(CI 95% 1.4 — 2.73) Hydroxychloroquine may
Due to serious
Based on data from 979 increase the risk of
Difference: 106 more per imprecision and
participants in 6 studies. diarrhoea.
1000 risk of bias 8
(Randomized controlled) ( CI 95% 48 more
— 174 more )
Nausea/
(CI 95% 1.26 — 2.41) per 1000 per 1000
vomiting Due to serious Hydroxychloroquine may
Based on data from
imprecision and increase the risk of
1429 participants in 7 Difference: 62 more per 1000 serious risk of bias nausea and vomiting.
( CI 95% 23 more 9

Delirium per 1000 per 1000 Due to very
(CI 95% 0.77 — 3.28) The effect of
serious
Based on data from 423 hydroxychloroquine on
Difference: 33 more per 1000 imprecision and
participants in 1 study. delirium is uncertain.
serious
— 116 more ) indirectness 10
Time to clinical 11 9 Very low The effect of

improvement Lower better days (Mean) days (Mean) Due to serious hydroxychloroquine on
Based on data from: 479 risk of bias, time to clinical
participants in 5 studies. Difference: MD 2 fewer imprecision, and improvement is
(Randomized controlled) ( CI 95% 4 fewer indirectness 11 uncertain.
— 0.1 more )
69 of 109
Certainty of
Intervention
Hydroxychloro
quine
evidence)

Lower better days (Mean) days (Mean) Hydroxychloroquine may
hospitalization Due to serious
Based on data from: have no effect on
imprecision and
5534 participants in 5 Difference: MD 0.1 more duration of
serious risk of bias
studies. (Randomized hospitalization.
( CI 95% 1.9 12
controlled) fewer — 2 more )
9.7 10.6
Time to viral days (Mean) days (Mean) Very low
The effect of
clearance Lower better Due to serious
hydroxychloroquine on
Based on data from: 440 MD 0.7 fewer risk of bias and
Difference: time to viral clearance is
participants in 5 studies. ( CI 95% 4.3 very serious
uncertain.
(Randomized controlled) fewer — 4.8 imprecision 13
more )
Adverse events Two of 108 patients randomized to

hydroxychloroquine discontinued The effect of
leading to drug Based on data from: 210
Very low
treatment because of adverse effects. hydroxychloroquine on
discontinuation participants in 3 studies. Due to extremely
None of 102 patients did so in the adverse events leading
serious
(Randomized controlled) placebo/standard care group. to drug discontinuation
imprecision 14 is uncertain.
1. Systematic review [1] . Baseline/comparator: Primary study. Baseline risk for mortality and mechanical ventilation were
derived from the WHO SOLIDARITY trial for patients with severe and critical COVID-19.
2. Imprecision: serious. The 95% CI crosses the minimally important difference (2% reduction in mortality). .
3. Imprecision: serious. Wide confidence intervals.
4. Systematic review. We used the median event rate for all patients randomized to usual care across included studies.
Baseline/comparator: Control arm of reference used for intervention. Supporting references: [1],
5. Imprecision: very serious. Wide confidence intervals.
6. Indirectness: serious. Imprecision: very serious.
7. Risk of bias: serious. Unblinded studies -> cardiac toxicity differential detection. Indirectness: serious. Studies measured
serious cardiac toxicity differently. Imprecision: serious.
8. Risk of bias: serious. Concerns mitigated because of large effect and indirect evidence showing consistent results.
Imprecision: serious. OIS not met. Upgrade: large magnitude of effect.
Imprecision: serious. OIS not met. Upgrade: large magnitude of effect.
10. Indirectness: serious. This outcome was not collected systematically and the definition of delirium was not specified.
Imprecision: very serious.
11. Risk of bias: serious. Indirectness: serious. Studies measured clinical improvement differently. Imprecision: serious.
12. Risk of bias: serious. Imprecision: serious. Wide confidence intervals.
13. Risk of bias: serious. Imprecision: very serious.
14. Imprecision: extremely serious.
70 of 109
6.9 Lopinavir/ritonavir (published 17 December 2020)

Info Box
The recommendation concerning lopinavir/ritonavir was published 17 December 2020 as the third version of the WHO living
October 2020, reporting results on treatment with lopinavir/ritonavirsivir, remdesivir and hydroxychloroquine in hospitalized
patients with COVID-19 (15). No changes were made for the lopinavir/ritonavir recommendation in this ninth version of the
guideline.
We recommend not to use lopinavir/ritonavir (strong recommendation against).
Benefits and harms

The GDG panel found a lack of evidence that lopinavir/ritonavir improved outcomes that matter to patients such as reduced
mortality, need for mechanical ventilation, time to clinical improvement and others. For mortality and need for mechanical
ventilation this was based on moderate certainty evidence, for the other outcomes low or very low certainty evidence.
There was low certainty evidence that lopinavir/ritonavir may increase the risk of diarrhoea and nausea and vomiting, a
finding consistent with the indirect evidence evaluating its use in patients with HIV. Diarrhoea and vomiting may increase
the risk of hypovolaemia, hypotension and acute kidney injury, especially in settings where health care resources are limited.
There was an uncertain effect on viral clearance and acute kidney injury.
Subgroup analysis indicated no effect modification based on severity of illness (comparing either critical vs severe/non-
severe or non-severe vs critical/severe) or age (comparing those aged < 70 years versus those 70 years and older). As there
was no evidence of a statistical subgroup effect, we did not formally evaluate using the ICEMAN tool.

The evidence is based on a linked systematic review and NMA of seven RCTs; pooling data from 7429 patients hospitalized
with various severities of COVID-19 and variably reporting the outcomes of interest to the guideline panel (1). The panel
agreed that there was moderate certainty for mortality and need for mechanical ventilation, low certainty for diarrhoea,
nausea and duration of hospitalization and very low certainty in the estimates of effect for viral clearance, acute kidney
injury and time to clinical improvement. Most outcomes were lowered for risk of bias and imprecision (wide confidence
intervals which do not exclude important benefit or harm).

not want to receive lopinavir/ritonavir given the evidence suggested there was probably no effect on mortality or need for
mechanical ventilation and there was a risk of adverse events including diarrhoea and nausea and vomiting. The panel did
not expect there would be much variation in values and preferences between patients when it came to this intervention.
71 of 109

Although the cost of lopinavir/ritonavir is not as high as some other investigational drugs for COVID-19, and the drug is
generally available in most health care settings, the GDG raised concerns about opportunity costs and the importance of not
drawing attention and resources away from best supportive care or the use of corticosteroids in severe COVID-19.
Justification
When moving from evidence to the strong recommendation against the use of lopinavir/ritonavir for patients with COVID-19,
the panel emphasized the moderate certainty evidence of probably no reduction in mortality or need for mechanical ventilation.
It also noted the evidence suggesting possible harm associated with treatment, with increased nausea and diarrhoea. The GDG
did not anticipate important variability in patient values and preferences, and other contextual factors, such as resource
considerations, accessibility, feasibility and impact on health equity would not alter the recommendation (see summary of these
factors under Evidence to Decision).
Subgroup analysis
The panel did not find any evidence of a subgroup effect across patients with different levels of disease severity, or between
adults and older adults and therefore did not make any subgroup recommendation for this drug. Although the trials did not
report subgroup effects by time from symptom onset, many of the trials enrolled patients early in the disease course. The strong
recommendation is applicable across disease severity and age groups.
Applicability
uncertain. However, the panel had no reason to think that children with COVID-19 would respond any differently to treatment
with lopinavir/ritonavir. There were similar considerations in regards to pregnant women, with no data directly examining this
population, but no rationale to suggest they would respond differently than other adults. In patients using lopinavir/ritonavir for
HIV infection, it should generally be continued while receiving care for COVID-19.
Uncertainties
Please see end of document for residual uncertainties (Section 9). The GDG panel felt that it was unlikely future studies would
identify a subgroup of patients that are likely to benefit from lopinavir/ritonavir.
Additional considerations
In patients who have undiagnosed or untreated HIV, use of lopinavir/ritonavir alone may promote HIV resistance to important
antiretrovirals. Widespread use of lopinavir/ritonavir for COVID-19 may cause drug shortages for people living with HIV.

Intervention: Lopinavir/ritonavir
Summary
Evidence summary
The LNMA on lopinavir/ritonavir was based on 7 RCTs with 7429 participants. Of note, none of the included studies
enrolled children or adolescents under the age of 19 years old (see Table). The GRADE Summary of Findings table shows
the relative and absolute effects of lopinavir/ritonavir compared to usual care for the outcomes of interest in patients
with COVID-19 across all disease severities, with certainty ratings. See section 7 for sources of baseline risk estimates
informing absolute estimates of effect.
Subgroup analysis
For lopinavir/ritonavir, the GDG panel requested subgroup analyses based on age (considering children vs younger
adults [e.g. under 70 years] vs older adults [e.g. 70 years or older]), and illness severity (non-severe vs severe vs critical
COVID-19). The GDG discussed other potential subgroups of interest including time from onset of symptoms until
initiation of therapy and concomitant medications, but recognized that these analyses would not be possible without
access to individual participant data and/or more detailed reporting from the individual trials.
72 of 109
Certainty of
Intervention
Lopinavir/
ritonavir
evidence)
Odds Ratio 1 106 106

Lopinavir/ritonavir
Based on data from Due to borderline
probably has no effect
8061 participants in 4 Difference: 0 fewer per 1000 risk of bias and
on mortality.
studies. 1 (Randomized ( CI 95% 17 fewer imprecision 2
Mechanical
(CI 95% 0.98 — 1.36) per 1000 per 1000 Moderate Lopinavir/ritonavir
ventilation Based on data from Due to borderline probably does not
7579 participants in 3 Difference: 17 more per 1000 risk of bias and reduce mechanical
( CI 95% 2 fewer imprecision 3 ventilation.
controlled)
— 38 more )
483 246
Viral clearance (CI 95% 0.04 — 1.97) Low The effects of lopinavir/
Based on data from 171 Due to very ritonavir on viral
participants in 2 studies. serious clearance is very
4
1000
(Randomized ( CI 95% 447 imprecision 5 uncertain.
more )
45 25
Acute kidney Relative Risk per 1000 per 1000 Very low
injury Due to serious The effect of lopinavir/
Based on data from 259 risk of bias and ritonavir on acute kidney
participants in 2 studies. very serious injury is uncertain.
1000
— 20 more )
67 235 Moderate
Odds Ratio 4.28 per 1000 per 1000 Due to serious
Diarrhoea (CI 95% 1.99 — 9.18) risk of bias and Lopinavir/ritonavir may
Based on data from 370 imprecision; increase the risk of
Difference: 168 more per
participants in 4 studies. upgraded due to diarrhoea.
1000
(Randomized controlled) large magnitude
( CI 95% 58 more
— 330 more ) of effect 7
17 177
Nausea/ Relative Risk per 1000 per 1000
Moderate
vomiting Lopinavir/ritonavir may
Based on data from 370 Difference: 160 more per Due to serious
increase the risk of
1000 risk of bias and
participants in 4 studies. nausea/vomiting.
(Randomized controlled) ( CI 95% 100 imprecision 8
more — 210
more )
Very low The effect of lopinavir/

Time to clinical Lower better
11 10 Due to serious ritonavir improves on
Based on data from: days (Mean) days (Mean) risk of bias and time to clinical
improvement
199 participants in 1 very serious improvement is very
study. imprecision 9 uncertain.
73 of 109
Certainty of
Intervention
Lopinavir/
ritonavir
evidence)
Difference: MD 1 fewer
( CI 95% 4.1
(Randomized controlled) fewer — 3.2
more )

Lower better days (Mean) days (Mean) Low Lopinavir/ritonavir may
hospitalization Based on data from: Due to serious have no effect on
5239 participants in 2 Difference: MD 0.3 lower risk of bias and duration of
( CI 95% 3 lower imprecision 10 hospitalization.
controlled) — 2.5 higher )
1. Systematic review. Baseline/comparator: Primary study[15]. Baseline risk for mortality and mechanical ventilation were
derived from the WHO SOLIDARITY trial for patients with severe and critical COVID-19. Supporting references: [1].
2. Imprecision: serious. The 95% CI crosses the minimally important difference (2% reduction in mortality).
3. Imprecision: serious. Wide confidence intervals.
4. Systematic review. Baseline/comparator: Control arm of reference used for intervention. We used the median event
rate for all patients randomized to usual care across included studies. Supporting references: [1].
6. Risk of bias: serious. Imprecision: very serious. Wide confidence intervals.
Imprecision: serious. Few patients and events. Upgrade: large magnitude of effect.
Imprecision: serious. Few patients and events. Upgrade: large magnitude of effect.
9. Risk of bias: serious. Imprecision: very serious. Wide confidence intervals, low number of patients.
6.10 Remdesivir (published 20 November 2020)
Info Box
The recommendation concerning remdesivir was published 20 November 2020 as the second version of the WHO living
October 2020, reporting results on treatment with remdesivir, hydroxychloroquine and lopinavir/ritonavir in hospitalized
patients with COVID-19 (15). No changes were made for the remdesivir recommendation in this ninth version of the guideline.
Of note, this recommendation is under review given new trials, and an update is planned in the next iteration of this guideline.
The current recommendation provided is based on the initial assessment made by the GDG, and does not represent best
current evidence.
Conditional recommendation against In review
We suggest not to use remdesivir (conditional recommendation against).
74 of 109
Practical Info
The GDG made a conditional recommendation against using remdesivir for treatment of hospitalized patients with COVID-19. If
administration of remdesivir is considered, it should be noted that its use is contraindicated in those with liver (ALT > 5 times
normal at baseline) or renal (eGFR < 30 mL/minute) dysfunction. To date, it can only be administered intravenously, and it has
relatively limited availability.
Benefits and harms

The GDG panel found a lack of evidence that remdesivir improved outcomes that matter to patients such as reduced
mortality, need for mechanical ventilation, time to clinical improvement and others. However, the low certainty evidence for
these outcomes, especially mortality, does not prove that remdesivir is ineffective; rather, there is insufficient evidence to
confirm that it improves patient-important outcomes.
There was no evidence of increased risk of SAEs from the trials. However, further pharmacovigilance is needed because
SAEs are commonly underreported and rare events could be missed, even in large RCTs.
A subgroup analysis indicated that remdesivir treatment possibly increased mortality in the critically ill and possibly reduced
mortality in the non-severely and severely ill. The panel judged the overall credibility of this subgroup effect (evaluated using
the ICEMAN tool) to be insufficient to make subgroup recommendations. The overall low certainty evidence on the benefits
and harms of remdesivir, driven by risk of bias and imprecision limitations in the included studies, also contributed to the
judgment.

The evidence is based on a linked systematic review and NMA of four RCTs; pooling data from 7333 patients hospitalized
with various severities of COVID-19 and variably reporting the outcomes of interest to the guideline panel (1). The panel
agreed that there was low certainty in the estimates of effect for all patient-important outcomes across benefits and harms,
mostly driven by risk of bias and imprecision (wide confidence intervals which do not exclude important benefit or harm).
There was very low certainty evidence for viral clearance and delirium.

Applying the agreed values and preferences (see Section 7), the GDG inferred that most patients would be reluctant to use
remdesivir given the evidence left high uncertainty regarding effects on mortality and the other prioritized outcomes. This
was particularly so as any beneficial effects of remdesivir, if they do exist, are likely to be small and the possibility of
important harm remains. The panel acknowledged, however, that values and preferences are likely to vary, and there will be
patients and clinicians who choose to use remdesivir given the evidence has not excluded the possibility of benefit.

A novel therapy typically requires higher certainty evidence of important benefits than currently available for remdesivir,
preferably supported wherever possible by cost-effectiveness analysis. In the absence of this information, the GDG raised
concerns about opportunity costs and the importance of not drawing attention and resources away from best supportive
care or the use of corticosteroids in severe COVID-19. It was noted that remdesivir is administered only by the intravenous
route currently, and that global availability is currently limited.
Justification
When moving from evidence to the conditional recommendation against the use of remdesivir for patients with COVID-19, the
panel emphasized the evidence of possibly no effect on mortality, need for mechanical ventilation, recovery from symptoms and
other patient-important outcomes, albeit of low certainty; it also noted the anticipated variability in patient values and
preferences, and other contextual factors, such as resource considerations, accessibility, feasibility and impact on health equity
(see summary of these factors under Evidence to Decision).
75 of 109
Importantly, given the low certainty evidence for these outcomes, the panel concluded that the evidence did not prove that
remdesivir has no benefit; rather, there is no evidence based on currently available data that it improves patient-important
outcomes. Especially given the costs and resource implications associated with remdesivir, but consistent with the approach that
should be taken with any new drug, the panel felt the responsibility should be on demonstrating evidence of efficacy, which is
not established by the currently available data. The panel noted that there was no evidence of increased risk of SAEs in patients
receiving remdesivir, at least from the included trials. Further pharmacovigilance is required to confirm this, as SAEs are
commonly underreported and rare events would be missed, even in large RCTs.
Subgroup analysis
The panel carefully considered a potential subgroup effect across patients with different levels of disease severity, suggesting a
possible increase in mortality in the critically ill and a possible reduction in mortality in the non-severely and severely ill. For this
analysis, critical illness was defined as those requiring invasive or non-invasive ventilation; severe illness as those requiring
oxygen therapy (but not meeting critical illness criteria); and non-severe as all others. Patients requiring high-flow nasal cannula
represented a small proportion and were characterized as either severe (SOLIDARITY) (15) or critical (ACTT-1) (119). The
analysis focused on within-study subgroup comparisons across the different severities, and therefore the SIMPLE-MODERATE
trial could not be included in the subgroup analysis as it only enrolled patients with non-severe COVID-19. The panel reviewed
the results of both the random effects frequentist analysis and the post hoc Bayesian analysis which incorporated meta-
regression using study as a random effect.
The GDG panel judged the credibility in the subgroup analysis assessing differences in mortality by severity of illness to be
insufficient to make subgroup recommendations. Important factors influencing this decision included a lack of a
priori hypothesized direction of subgroup effect by trial investigators, little or no previously existing supportive evidence for the
subgroup finding, and relatively arbitrary cut points used to examine the subgroups of interest. The overall low certainty
evidence for the benefits and harms of remdesivir, driven by risk of bias and imprecision limitations, also contributed to the
judgment. The panel highlighted that despite the conditional recommendation against remdesivir, they support further
enrolment into RCTs evaluating remdesivir, especially to provide higher certainty of evidence for specific subgroups of patients.
The panel had a priori requested analyses of other important subgroups of patients including children and older persons, but
there were no data to address these groups specifically. None of the included RCTs enrolled children, and although older people
were included in the trials, their outcomes were not reported separately. Also, there is no pharmacokinetic or safety data on
remdesivir for children. Given this, the applicability of this recommendation to children is currently uncertain.

Intervention: Remdesivir
Summary
Evidence summary
Based on 4 RCTs with 7333 participants (15)(119)(120)(121), the LNMA provided relative estimates of effect for patient-
important outcomes . Of note, none of the included studies enrolled children or adolescents under the age of 19 years
old (see Table).The GRADE Summary of Findings table shows the relative and absolute effects of remdesivir compared
to usual care for the outcomes of interest in patients COVID-19 across all disease severities, with certainty ratings. See
Section 7 for sources of baseline risk estimates informing absolute estimates of effect.
Subgroup analysis
The GDG panel requested subgroup analyses based on age (considering children vs adults vs older people), illness
severity (non-severe vs severe vs critical COVID), and duration of remdesivir therapy (5 days vs longer than 5 days). The
GDG discussed other potential subgroups of interest including time from onset of symptoms until initiation of therapy,
and concomitant medications (especially corticosteroids); however, the GDG recognized these analyses would not be
possible without access to individual participant data. To this last point, the panel recognized that usual care is likely
variable between centres, regions and evolved over time. However, given all of the data come from RCTs, use of these
co-interventions that comprise usual care should be balanced between study patients randomized to either the
intervention or usual care arms.
Following the panel's request, the NMA team performed subgroup analyses in order to assess for effect modification
which, if present, could mandate distinct recommendations by subgroups. From the data available from the included
trials, subgroup analysis was only possible for severity of illness and the outcome of mortality. This subgroup analysis
was performed using a random effects frequentist analysis based on the three WHO severity definitions. A post hoc
76 of 109
Bayesian analysis was also performed, which incorporated meta-regression using study as a random effect. This latter
approach has the advantage of more accurately accounting for within-study differences but can only compare two
subgroups at a time. The panel used a pre-specified framework incorporating the ICEMAN tool to assess the credibility
of subgroup findings (84).
Certainty of
Timeframe measurements Standard care Remdesivir (Quality of summary
evidence)

Mortality per 1000 per 1000 Low
(CI 95% 0.7 — 1.12)
28 days Due to serious Remdesivir possibly has
Based on data from
Difference: 10 fewer per risk of bias and little or no effect on
1000 serious mortality.
studies. 1 (Randomized
controlled)
— 11 more )

Mechanical per 1000 per 1000 Low
(CI 95% 0.76 — 1.03)
ventilation Due to serious Remdesivir possibly has
Based on data from
Difference: 10 fewer per risk of bias and little or no effect on
1000 serious mechanical ventilation.
controlled)
— 3 more )
Serious adverse
events leading
Odds Ratio 1 15 15 Remdesivir possibly has
(CI 95% 0.37 — 3.83) per 1000 per 1000 Low
to little or no effect on
discontinuation serious adverse events
1894 participants in 3 Difference: 0 fewer per 1000 serious
leading to
studies. 4 (Randomized ( CI 95% 9 fewer imprecision 5 discontinuation.
483 498
Viral clearance Odds Ratio 1.06 per 1000 per 1000 Very low
7 days (CI 95% 0.06 — 17.56) The effect of remdesivir
Due to very
Based on data from 196 Difference: 15 more per 1000 on viral clearance is
serious
participants in 1 study. uncertain.
( CI 95% 430 imprecision 6
(Randomized controlled) fewer — 460
more )
Acute kidney
(CI 95% 0.51 — 1.41) per 1000 per 1000
injury Due to serious Remdesivir possibly has
Based on data from
imprecision and little or no effect on
studies. (Randomized Difference: e8 )fewer per 1000 serious acute kidney injury.
( CI 95% 27 fewer indirectness 7
Odds Ratio 1.22 Very low

Delirium (CI 95% 0.48 — 3.11)
16 19 Due to very We are uncertain
Based on data from per 1000 per 1000 serious whether remdesivir
1048 participants in imprecision and increases or decreases
1 study. (Randomized Difference: 3 more per 1000 serious delirium.
controlled) ( CI 95% 8 fewer indirectness 8
— 32 more )
77 of 109
Certainty of
Timeframe measurements Standard care Remdesivir (Quality of summary
evidence)
Time to clinical
days
Measured by: 11 9 Low
Lower better days days
improvement Due to serious Remdesivir possibly has
Based on data from:
imprecision and little or no effect on time
1882 participants in 3 Difference: MD 2 lower serious to clinical improvement.
( CI 95% 4.2 lower indirectness 9
Duration of
days
Measured by: 12.8 12.3 Low
Lower better days days Remdesivir possibly has
hospitalization Due to serious
Based on data from: little or no effect on
imprecision and
1882 participants in 3 Difference: MD 0.5 lower duration of
serious
studies. (Randomized hospitalization.
( CI 95% 3.3 lower indirectness 10
Duration of Measured by: days 14.7 13.4

days days Low
ventilation Lower better Remdesivir possibly has
Due to very
Based on data from: 440 little or no effect on
Difference: MD 1.3 lower serious
participants in 2 studies. duration of ventilation.
( CI 95% 4.1 lower imprecision 11
— 1.5 higher )
1. Systematic review [1] . Baseline/comparator: Primary study[15]. Baseline risk for mortality and mechanical ventilation
were derived from the WHO SOLIDARITY trial for patients with severe and critical COVID-19.
2. Risk of bias: serious. We rated two trials as high risk of bias due to high or probably high risk of bias in deviations from
the intended intervention. Imprecision: serious. The 95% CI crosses the minimally important difference (2% reduction in
mortality).
4. Systematic review [1] . Baseline/comparator: Control arm of reference used for intervention. We used the median
event rate for all patients randomized to usual care across included studies.
7. Indirectness: serious. Studies used change in serum creatinine rather than patient-important measures of acute kidney
injury. Imprecision: serious. Wide 95% credible intervals.
8. Indirectness: serious. Differences between the outcomes of interest and those reported (e.g short-term/surrogate, not
patient-important). Imprecision: very serious.
10. Indirectness: serious. Imprecision: serious. Wide confidence intervals.

Remdesivir is a novel monophosphoramidate adenosine analogue prodrug which is metabolized to an active tri-phosphate form
that inhibits viral RNA synthesis. Remdesivir has in vitro and in vivo antiviral activity against several viruses, including SARS-
CoV-2. Remdesivir is widely used in many countries, with several guidelines recommending its use in patients with severe or
critical COVID-19 (122)(123).
78 of 109
6.11 Systemic corticosteroids (published 2 September 2020)
Info Box
The recommendations for corticosteroids were first published as WHO living guidelines 2 September 2020, and as BMJ Rapid
Recommendations 5 September 2020. It followed the publication of the preliminary report of the RECOVERY trial, later
published as a peer-reviewed paper (14). No changes were made for the corticosteroids recommendations in this ninth version
of the guideline.
Whereas the recommendations remain unchanged, the evidence summary for corticosteroids in patients with COVID-19 was
updated before the sixth iteration of this living guideline. The baseline risk estimates for mortality are now based on the WHO
SOLIDARITY trial (as for other drugs in this guideline) (15) rather than the initial ISARIC cohort study (124) that likely
overestimates current mortality risks at the global level. The update was also needed to inform the baseline risk for mortality in
the evidence summary informing the strong recommendation for IL-6 receptor blockers, in addition to standard of care for
patients with severe or critical COVID-19, where corticosteroids provide a relative reduction in mortality by 21%.
We recommend treatment with systemic corticosteroids (strong recommendation for).
Practical Info
Route: Systemic corticosteroids may be administered both orally and intravenously. Of note, while the bioavailability of
dexamethasone is very high (that is, similar concentrations are achieved in plasma after oral and intravenous intake), critically ill
patients may be unable to absorb any nutrients or medications due to intestinal dysfunction. Clinicians therefore may consider
administering systemic corticosteroids intravenously rather than orally if intestinal dysfunction is suspected.
Duration: While more patients received corticosteroids in the form of dexamethasone 6 mg daily for up to 10 days, the total
duration of regimens evaluated in the seven trials varied between 5 and 14 days, and treatment was generally discontinued at
hospital discharge (that is, the duration of treatment could be less than the duration stipulated in the protocols).
Dose: The once daily dexamethasone formulation may increase adherence. A dose of 6 mg of dexamethasone is equivalent (in
terms of glucocorticoid effect) to 150 mg of hydrocortisone (that is, 50 mg every 8 hours), 40 mg of prednisone, or 32 mg of
methylprednisolone (8 mg every 6 hours or 16 mg every 12 hours).
Monitoring: It would be prudent to monitor glucose levels in patients with severe and critical COVID-19, regardless of whether
the patient is known to have diabetes.
Timing: The timing of therapy from onset of symptoms was discussed by the panel. The RECOVERY investigators reported a
subgroup analysis suggesting that the initiation of therapy 7 days or more after symptom onset may be more beneficial than
treatment initiated within 7 days of symptom onset. A post hoc subgroup analysis within the PMA did not support this
hypothesis. While some panel members believed that postponing systemic corticosteroids until after viral replication is
contained by the immune system may be reasonable, many noted that, in practice, it is often impossible to ascertain symptom
onset and that signs of severity often appear late (that is, denote a co-linearity between severity and timing). The panel
concluded that, given the evidence, it was preferable to err on the side of administering corticosteroids when treating patients
with severe or critical COVID-19 (even if within 7 days of symptoms onset) and to err on the side of not giving corticosteroids
when treating patients with non-severe disease (even if after 7 days of symptoms onset).
Benefits and harms

Panel members who voted for a conditional recommendation argued that the trials evaluating systemic corticosteroids for
79 of 109
COVID-19 reported limited information regarding potential harm. Between the two panel meetings, indirect evidence
regarding the potential harmful effects of systemic corticosteroids from studies in sepsis, ARDS and community-acquired
pneumonia (CAP) was added to the summary of findings table (125)(126). While generally of low certainty, these data were
reassuring and suggested that corticosteroids are not associated with an increased risk of adverse events, beyond likely
increasing the incidence of hyperglycaemia (moderate certainty evidence; absolute effect estimate 46 more per 1000
patients, 95% CI: 23 more to 72 more) and hypernatraemia (moderate certainty evidence; 26 more per 1000 patients, 95%
CI: 13 more to 41 more). Panel members also noted that, given the expected effect of systemic corticosteroids on mortality,
most patients would not refuse this intervention to avoid adverse events believed to be markedly less important to most
patients than death.
In contrast with new agents proposed for COVID-19, clinicians have a vast experience of systemic corticosteroids and the
panel was reassured by their overall safety profile. Moreover, the panel was confident that clinicians using these guidelines
would be aware of additional potential side-effects and contraindications to systemic corticosteroid therapy, which may
vary geographically in function of endemic microbiological flora. Notwithstanding, clinicians should exercise caution in use
of corticosteroids in patients with diabetes or underlying immunocompromise.
Ultimately, the panel made its recommendation on the basis of the moderate certainty evidence of a 28-day mortality
reduction of 8.7% in the critically ill and 6.7% in patients with severe COVID-19 who were not critically ill, respectively. In
the fifth iteration of this living guideline, mortality baseline risk estimates were updated based on the WHO SOLIDARITY
trial, considered to represent the best source of prognosis across countries facing the COVID-19 pandemic. This resulted in
an overall 3.3% reduction in 28-day mortality for patients with severe or critical COVID-19, still with moderate certainty
evidence and considered by the panel to represent a clear benefit to patients, with no impact on the established
recommendations.

The panel took an individual patient perspective to values and preferences but, given the burden of the pandemic for health
care systems globally, also placed a high value on resource allocation and equity. The benefits of corticosteroids on mortality
was deemed of critical importance to patients, with little or no anticipated variability in their preference to be offered
treatment if severely ill from COVID-19.

Resource implications, feasibility, equity and human rights
In this guideline, the panel took an individual patient perspective, but also placed a high value on resource allocation. In such
a perspective, attention is paid to the opportunity cost associated with the widespread provision of therapies for COVID-19.
In contrast to other candidate treatments for COVID-19 that, generally, are expensive, often unlicensed, difficult to obtain
and require advanced medical infrastructure, systemic corticosteroids are low cost, easy to administer, and readily available
globally (127). Dexamethasone and prednisolone are among the most commonly listed medicines in national essential
medicines lists; listed by 95% of countries. Dexamethasone was first listed by WHO as an essential medicine in 1977, while
prednisolone was listed 2 years later (128).
Accordingly, systemic corticosteroids are among a relatively small number of interventions for COVID-19 that have the
potential to reduce inequities and improve equity in health. Those considerations influenced the strength of this
recommendation.
Acceptability
The ease of administration, the relatively short duration of a course of systemic corticosteroid therapy, and the generally
benign safety profile of systemic corticosteroids for up to 7–10 days led the panel to conclude that the acceptability of this
intervention was high.
Justification
This recommendation was achieved after a vote, which concerned the strength of the recommendation in favour of systemic
corticosteroids. Of the 23 voting panel members, 19 (83%) voted in favour of a strong recommendation, and 4 (17%) voted in
80 of 109
favour of a conditional recommendation. The reasons for the four cautionary votes, which were shared by some panel members
who voted in favour of a strong recommendation, are summarized below.
Applicability
Panel members who voted for a conditional recommendation argued that many patients who were potentially eligible for the
RECOVERY trial were excluded from participating in the evaluation of corticosteroids by their treating clinicians and that
without detailed information on the characteristics of excluded patients, this precluded, in their opinion, a strong
recommendation. Other panel members felt that such a proportion of excluded patients was the norm rather than the exception
in pragmatic trials and that, while detailed information on the reasons for excluding patients were not collected, the main
reasons for refusing to offer participation in the trial were likely related to safety concerns of stopping corticosteroids in patients
with a clear indication for corticosteroids (confirmed as per personal communication from the RECOVERY Principal Investigator).
Panel members noted that there are few absolute contraindications to a 7–10 day course of corticosteroid therapy, that
recommendations are intended for the average patient population, and that it is understood that even strong recommendations
should not be applied to patients in whom the intervention is contraindicated as determined by the treating clinician.
Eventually, the panel concluded that this recommendation applies to patients with severe and critical COVID-19 regardless of
hospitalization status. The underlying assumption is that these patients would be treated in hospitals and receive respiratory
support in the form of oxygen; non-invasive or invasive ventilation if these options were available. Following GRADE guidance,
in making a strong recommendation, the panel has inferred that all or almost all fully informed patients with severe COVID-19
would choose to take systemic corticosteroids. It is understood that even in the context of a strong recommendation, the
intervention may be contraindicated for certain patients. Absolute contraindications for 7–10 day courses of systemic
corticosteroid therapy are rare. In considering potential contraindications, clinicians must determine if they warrant depriving a
patient of a potentially life-saving therapy.
The applicability of the recommendation is less clear for populations that were under-represented in the considered trials, such
as children, patients with tuberculosis, and those who are immunocompromised. Notwithstanding, clinicians will also consider
the risk of depriving these patients of potentially life-saving therapy. In contrast, the panel concluded that the recommendation
should definitely be applied to certain patients who were not included in the trials, such as patients with severe and critical
COVID-19 who could not be hospitalized or receive oxygen because of resource limitations.
The recommendation does not apply to the following uses of corticosteroids: transdermal or inhaled administration, high-dose
or long-term regimens, or prophylaxis.

Population: Patients with severe or critical COVID-19 (updated baseline mortality risk)
Intervention: Systemic corticosteroids
Summary
Evidence summary
This guideline was triggered on 22 June 2020 by the publication of the preliminary report of the RECOVERY trial, later
published as a peer-reviewed paper (14). Corticosteroids are listed in the WHO Model List of Essential Medicines,
readily available globally at a low cost, and of considerable interest to all stakeholder groups. The guideline panel was
informed by combining two meta-analyses which pooled data from eight randomized trials (7184 participants) of
systemic corticosteroids for COVID-19 (1)(129). The panel discussions were also informed by two other meta-analyses,
which were already published and pooled data about the safety of systemic corticosteroids in distinct but relevant
patient populations.
The GRADE Summary of Findings table shows the relative and absolute effects of systemic corticosteroids compared to
usual care for the outcomes of interest in patients with severe and critical COVID-19, with certainty ratings. Below we
provide more details about the trials and meta-analysis as well as a subgroup analysis that informed the
recommendation. See Section 7 for sources of baseline risk estimates informing absolute estimates of effect.
On 17 July 2020, the panel reviewed evidence from eight RCTs (7184 patients) evaluating systemic corticosteroids
versus usual care in COVID-19. RECOVERY, the largest of the seven trials, from which mortality data were available by
subgroup (severe and non-severe), evaluated the effects of dexamethasone 6 mg given once daily (oral or intravenous)
for up to 10 days in 6425 hospitalized patients in the United Kingdom (2104 were randomized to dexamethasone and
4321 were randomized to usual care) (14). At the time of randomization, 16% were receiving invasive mechanical
81 of 109
ventilation or extracorporeal membrane oxygenation; 60% were receiving oxygen only (with or without non-invasive
ventilation); and 24% were receiving neither.
The data from seven other smaller trials included 63 non-critically ill patients and approximately 700 critically ill patients
(definitions of critical illness varied across studies). For the latter, patients were enrolled up to 9 June 2020, and
approximately four-fifths were invasively mechanically ventilated; approximately half were randomized to receive
corticosteroid therapy, and half randomized to no corticosteroid therapy. Corticosteroid regimens included:
methylprednisolone 40 mg every 12 hours for 3 days and then 20 mg every 12 hours for 3 days (GLUCOCOVID) (130);
dexamethasone 20 mg daily for 5 days followed by 10 mg daily for 5 days (two trials, DEXA-COVID19,
CoDEX) (131)(132); hydrocortisone 200 mg daily for 4 to 7 days followed by 100 mg daily for 2 to 4 days and then 50
mg daily for 2 to 3 days (one trial, CAPE-COVID) (133); hydrocortisone 200 mg daily for 7 days (one trial, REMAP-
CAP) (16); methylprednisolone 40 mg every 12 hours for 5 days (one trial, Steroids-SARI) (134).
Seven of the trials were conducted in individual countries (Brazil, China, Denmark, France, Spain) whilst REMAP-CAP
was an international study (recruiting in 14 European countries, Australia, Canada, New Zealand, Saudi Arabia and the
United Kingdom). All trials reported mortality 28 days after randomization, except for one trial at 21 days and another at
30 days. Because the mortality data from one trial (GLUCOCOVID, n=63) were not reported by subgroup, the panel
reviewed only the data pertaining to the outcome of mechanical ventilation from this trial (130). An additional trial,
which randomized hospitalized patients with suspected SARS-CoV-2 infection, published on 12 August 2020
(MetCOVID) (135), was included as a supplement in the PMA publication, as it was registered after the searches of trial
registries were performed. The supplement showed that inclusion would not change results other than reduce
inconsistency.
Subgroup analyses
While all other trials evaluated systemic corticosteroids exclusively in critically ill patients, the RECOVERY trial enrolled
hospitalized patients with COVID-19. The panel considered the results of a subgroup analysis of the RECOVERY trial
suggesting that the relative effects of systemic corticosteroids varied as a function of the level of respiratory support
received at randomization. On the basis of the peer-reviewed criteria for credible subgroup effects (84), the panel
determined that the subgroup effect was sufficiently credible to warrant separate recommendations for severe and non-
severe COVID-19.
However, acknowledging that during a pandemic, access to health care may vary considerably over time as well as
between different countries, the panel decided against defining patient populations concerned by the recommendations
on the basis of access to health interventions (i.e. hospitalization and respiratory support). Thus, the panel attributed the
effect modification in the RECOVERY trial to illness severity.
The panel also acknowledged the existence of variable definitions for severity and use of respiratory support
interventions. The WHO clinical guidance for COVID-19 published on 27 May 2020 (version 3) defined severity of
COVID-19 by clinical indicators, but modified the oxygen saturation threshold from 94% to 90%, in order to align with
previous WHO guidance (6). See Section 5 for the WHO severity criteria and Infographic for three disease severity
groups for which the recommendations apply in practice.
Certainty of
Intervention
Systemic
corticosteroids
evidence)

Mortality (CI 95% 0.7 — 0.9) per 1000 per 1000 Systemic corticosteroids
28 days Moderate probably reduce the risk
Based on data from
Difference: 34 fewer per Due to serious of 28-day mortality in
studies. 1 1000 risk of bias 2 patients with critical
illness due to COVID-19.
Follow up: 28 days. ( CI 95% 48 fewer
— 16 fewer )
Need for Relative Risk 0.74 116 86 Systemic corticosteroids

invasive (CI 95% 0.59 — 0.93) per 1000 per 1000 Moderate
probably reduce the
mechanical Based on data from Due to serious
need of mechanical
ventilation 5481 participants in 2 risk of bias 3 ventilation.
28 days studies.
82 of 109
Certainty of
Intervention
Systemic
corticosteroids
evidence)

1000
Follow up: 28 days.
( CI 95% 48 fewer
— 8 fewer )
Gastrointestinal Relative Risk 1.06 48 51 Low

bleeding (CI 95% 0.85 — 1.33) per 1000 per 1000 Due to serious Corticosteroids may not
Based on data from indirectness, Due increase the risk of
5403 participants in 30 Difference: 3 more per 1000 to serious gastrointestinal bleeding.
studies. ( CI 95% 7 fewer imprecision 4
— 16 more )
Super-infections Relative Risk 1.01 186 188 Low

(CI 95% 0.9 — 1.13) per 1000 per 1000 Due to serious Corticosteroids may not
6027 participants in 32 Difference: 2 more per 1000 to serious super-infections.
— 24 more )
Relative Risk 116 286 332

Hyperglycaemia (CI 95% 1.08 — 1.25) per 1000 per 1000 Moderate Corticosteroids probably
Based on data from Due to serious increase the risk of
8938 participants in 24 Difference: 46 more per 1000 indirectness 6 hyperglycaemia.
studies. ( CI 95% 23 more
— 72 more )

Hypernatremia (CI 95% 1.32 — 2.03) per 1000 per 1000 Moderate Corticosteroids probably
5015 participants in 6 Difference: 26 more per 1000 indirectness 7 hypernatremia.
— 41 more )
Neuromuscular Relative Risk 1.09 69 75 Low

(CI 95% 0.86 — 1.39) per 1000 per 1000 Corticosteroids may not
weakness Due to serious
Based on data from indirectness, Due
Difference: 6 more per 1000 neuromuscular
6358 participants in 8 to serious
weakness.
— 27 more )
Neuropsychiatric Relative Risk 0.81 35 28 Low

effects (CI 95% 0.41 — 1.63) per 1000 per 1000 Due to serious Corticosteroids may not
1813 participants in 7 Difference: 7 fewer per 1000 to serious neuropsychiatric effects.
— 22 more )
Duration of days Low

Measured by:
Lower better
13 12 Due to serious
Steroids may result in an
hospitalization days days important reduction in
Based on data from: risk of bias and
the duration of
6425 participants in serious
hospitalizations.
1 study. imprecision 10
83 of 109
1. Systematic review [1] . Baseline/comparator: Primary study[15]. Baseline risk estimate for mortality updated as of May
2021: now from WHO SOLIDARITY (considered the best source) with 14.6% mortality at 28 days in severe and critically ill
patients. This estimate adjusted for 50% receiving corticosteroids as standard of care in SOLIDARITY.
2. Risk of bias: serious. Lack of blinding.
3. Risk of bias: serious. Lack of blinding.
6. Indirectness: serious.
10. Risk of bias: serious. Lack of blinding. Imprecision: serious. Confidence interval includes no benefit.
For patients with non-severe COVID-19 infection
Conditional recommendation against
We suggest not to use systemic corticosteroids (conditional recommendation against).
Practical Info
With the conditional recommendation against the use of corticosteroids in patients with non-severe COVID-19 the following
practical information apply in situations where such treatment is to be considered:
Route: Systemic corticosteroids may be administered both orally and intravenously. Of note, while the bioavailability of
dexamethasone is very high (i.e. similar concentrations are achieved in plasma after oral and intravenous intake), critically ill
patients may be unable to absorb any nutrients or medications due to intestinal dysfunction. Clinicians therefore may consider
administering systemic corticosteroids intravenously rather than orally if intestinal dysfunction is suspected.
Duration: While more patients received corticosteroids in the form of dexamethasone 6 mg daily for up to 10 days, the total
duration of regimens evaluated in the seven trials varied between 5 and 14 days, and treatment was generally discontinued at
hospital discharge (i.e. the duration of treatment could be less than the duration stipulated in the protocols).
Dose: The once daily dexamethasone formulation may increase adherence. A dose of 6 mg of dexamethasone is equivalent (in
terms of glucocorticoid effect) to 150 mg of hydrocortisone (e.g. 50 mg every 8 hours), or 40 mg of prednisone, or 32 mg of
methylprednisolone (e.g. 8 mg every 6 hours or 16 mg every 12 hours). It would be prudent to monitor glucose levels in patients
with severe and critical COVID-19, regardless of whether the patient is known to have diabetes.
Timing: The timing of therapy from onset of symptoms was discussed by the panel. The RECOVERY investigators reported a
subgroup analysis suggesting that the initiation of therapy 7 days or more after symptom onset may be more beneficial than
treatment initiated within 7 days of treatment onset. A post hoc subgroup analysis within the PMA did not support this
hypothesis. While some panel members believed that postponing systemic corticosteroids until after viral replication is
contained by the immune system may be reasonable, many noted that, in practice, it is often impossible to ascertain symptom
onset and that signs of severity frequently appear late (i.e. denote a co-linearity between severity and timing). The panel
concluded that, given the evidence, it was preferable to err on the side of administering corticosteroids when treating patients
with severe or critical COVID-19 (even if within 7 days of symptoms onset) and to err on the side of not giving corticosteroids
when treating patients with non-severe disease (even if after 7 days of symptoms onset).
Other endemic infections that may worsen with corticosteroids should be considered. For example, for Strongyloides
stercoralis hyperinfection associated with corticosteroid therapy, diagnosis or empiric treatment may be considered in endemic
areas if steroids are used.
84 of 109
Benefits and harms

The panel made its recommendation on the basis of low certainty evidence suggesting a potential increase of 3.9% in
28-day mortality among patients with COVID-19 who are not severely ill. The certainty of the evidence for this specific
subgroup was downgraded due to serious imprecision (i.e. the evidence does not allow to rule out a mortality reduction) and
risk of bias due to lack of blinding. In making a conditional recommendation against the indiscriminate use of systemic
corticosteroids, the panel inferred that most fully informed individuals with non-severe illness would not want to receive
systemic corticosteroids, but many could want to consider this intervention through shared decision-making with their
treating physician (136)(6).
Note: WHO recommends antenatal corticosteroid therapy for pregnant women at risk of preterm birth from 24 to 34
weeks’ gestation when there is no clinical evidence of maternal infection, and adequate childbirth and newborn care is
available. However, in cases where the woman presents with mild or moderate COVID-19, the clinical benefits of antenatal
corticosteroid might outweigh the risks of potential harm to the mother. In this situation, the balance of benefits and harms
for the woman and the preterm newborn should be discussed with the woman to ensure an informed decision, as this
assessment may vary depending on the woman’s clinical condition, her wishes and that of her family, and available health
care resources.

See Benefits and Harms section.

The weak or conditional recommendation was driven by likely variation in patient values and preferences. The panel judged
that most individuals with non-severe illness would decline systemic corticosteroids. However, many may want them after
shared decision-making with their treating physician.

Resource implications, feasibility, equity and human rights
The panel also considered that in order to help guarantee access to systemic corticosteroids for patients with severe and
critical COVID-19, it is reasonable to avoid administering this intervention to patients who, given the current evidence,
would not appear to derive any benefit from this intervention.
Justification
This recommendation was achieved by consensus.
Applicability
This recommendation applies to patients with non-severe disease regardless of their hospitalization status. The panel noted that
patients with non-severe COVID-19 would not normally require acute care in hospital or respiratory support, but that in some
jurisdictions, these patients may be hospitalized for isolation purposes only, in which case they should not be treated with
systemic corticosteroids. The panel concluded that systemic corticosteroids should not be stopped for patients with non-severe
COVID-19 who are already treated with systemic corticosteroids for other reasons (e.g. patients with chronic obstructive
pulmonary disease or other chronic autoimmune diseases need not discontinue a course of systemic oral corticosteroid). If the
clinical condition of patients with non-severe COVID-19 worsens (i.e. increase in respiratory rate, signs of respiratory distress or
hypoxaemia) they should receive systemic corticosteroids (see recommendation for severe and critical COVID-19).
85 of 109

Intervention: Systemic corticosteroids
Summary
Evidence summary
Please see evidence summary above (placed under recommendation for patients with severe and critical COVID-19 to
find more information about the eight RCTs pooled into two systematic reviews with meta-analysis. It also provides
information about additional systematic reviews used to inform safety outcomes and results of subgroup analyses
resulting in separate recommendations for patients with non-severe COVID-19 and those with severe and critical
illness.
The GRADE Summary of Findings table shows the relative and absolute effects of systemic corticosteroids compared to
usual care for the outcomes of interest in patients with non- severe COVID-19, with certainty ratings.
Certainty of
Intervention
Systemic
corticosteroids
evidence)
Mortality
Relative Risk 1.22 23 28 Low Systemic corticosteroids
(CI 95% 0.93 — 1.61) per 1000 per 1000
28 days Due to serious may increase the risk of
Based on data from
risk of bias and 28-day mortality in
Difference: 5 more per 1000 serious patients with non-severe
1 study. 1
( CI 95% 2 fewer imprecision 2 COVID-19.
Follow up: 28 days.
— 14 more )
Need for
invasive Relative Risk 0.74
116 86
(CI 95% 0.59 — 0.93) per 1000 per 1000
mechanical Systemic corticosteroids
Moderate
ventilation Based on data from probably reduce the
Difference: 30 fewer per Due to serious
28 days 5481 participants in 2 need for mechanical
studies. 1000 risk of bias 3 ventilation.
Follow up: 28 days. ( CI 95% 48 fewer
— 8 fewer )
Gastrointestinal Relative Risk 1.06 48 51 Low

bleeding (CI 95% 0.85 — 1.33) per 1000 per 1000 Due to serious Corticosteroids may not
Based on data from indirectness and increase the risk of
5403 participants in 30 Difference: 3 more per 1000 serious gastrointestinal bleeding.
studies. 4 ( CI 95% 7 fewer imprecision 5
— 16 more )
Super-infections Relative Risk 1.01 186 188 Low

(CI 95% 0.9 — 1.13) per 1000 per 1000 Due to serious Corticosteroids may not
6027 participants in 32 Difference: 2 more per 1000 to serious super-infections.
— 24 more )
Relative Risk 1.16

Hyperglycaemia
(CI 95% 1.08 — 1.25)
286 332 Moderate Corticosteroids probably
per 1000 per 1000 Due to serious increase the risk of
Based on data from
8938 participants in 24 indirectness 7 hyperglycaemia.
studies.
86 of 109
Certainty of
Intervention
Systemic
corticosteroids
evidence)
Difference: 46 more per 1000

( CI 95% 23 more
— 72 more )

Hypernatremia (CI 95% 1.32 — 2.03) per 1000 per 1000 Moderate Corticosteroids probably
5015 participants in 6 Difference: 26 more per 1000 indirectness 8 hypernatremia.
— 41 more )
Neuromuscular Relative Risk 1.09 69 75 Low

(CI 95% 0.86 — 1.39) per 1000 per 1000 Corticosteroids may not
weakness Due to serious
Based on data from indirectness and
Difference: 6 more per 1000 neuromuscular
6358 participants in 8 serious
weakness.
— 27 more )
Neuropsychiatric Relative Risk 0.81 35 28 Low

effects (CI 95% 0.41 — 1.63) per 1000 per 1000 Due to serious Corticosteroids may not
Based on data from indirectness and increase the risk of
1813 participants in 7 Difference: 7 fewer per 1000 serious neuropsychiatric effects.
— 22 more )
days
Measured by:
Duration of Lower better 13 12 Low
Steroids may result in an
hospitalization days days Due to serious
Based on data from: important reduction in
risk of bias and
6425 participants in the duration of
serious
1 study. hospitalizations.
imprecision 11
1. Systematic review [1]. Baseline/comparator: Primary study [15]. We derived baseline risk for mortality and mechanical
ventilation from the control arm of the WHO SOLIDARITY trial.
2. Risk of bias: serious. lack of blinding. Imprecision: serious.
3. Risk of bias: serious. lack of blinding.
11. Risk of Bias: serious. lack of blinding. Imprecision: serious. confidence interval includes no benefit.
87 of 109
7. Methods: how this guideline was created

This living WHO guideline was developed according to standards and methods for trustworthy guidelines, making use of an innovative
process to achieve efficiency in dynamic updating of recommendations. The methods are aligned with the WHO Handbook for guideline
development and according to a pre-approved protocol (planning proposal) by the Guideline Review Committee (GRC) (136).
Related guidelines
This living WHO guideline for COVID-19 treatments is related to the larger, more comprehensive guidance for COVID-19 Clinical
management: living guideline, which has a wider scope of content and has been regularly updated (6). The first eight versions of this
WHO Therapeutics and COVID-19: living guideline, addressing corticosteroids, remdesivir, hydroxychloroquine, lopinavir/ritonavir,
ivermectin, IL-6 receptor blockers, casirivimab-imdevimab (neutralizing monoclonal antibodies), convalescent plasma, JAK inhibitors and
sotrovimab can be accessed via the WHO website (4).
Guidelines regarding the use of drugs to prevent (rather than treat) COVID-19 are included in a separate document, WHO Living
guideline: Drugs to prevent COVID-19, that can be accessed via the WHO website and the BMJ (8).
Timing
This guideline is living – dynamically updated and globally disseminated once new evidence warrants a change in
recommendations (137). The aim is for a six-week timeframe from the public availability of trial data that trigger the guideline
development process to WHO publication, while maintaining standards for trustworthy guidelines (WHO Handbook for guideline
development) (136)(138).
Stepwise approach
Here we outline the approach, involving simultaneous processes, taken to improve efficiency and timeliness of development and
dissemination of living, trustworthy guidance.
Step 1: Evidence monitoring and mapping and triggering of evidence synthesis

Comprehensive daily monitoring of all emerging RCTs occurs on a continuous basis, within the context of the living systematic review
and network meta-analysis (NMA), using experienced information specialists, who review all relevant information sources for new RCTs
addressing interventions for COVID-19. Incorporating pre-print data, which have not yet undergone peer review, promote rapid data
sharing in a public health emergency and its inclusion can accelerate the assessment and clinical use of COVID-19 therapeutic
interventions. Guidelines are periodically updated to assess data that have undergone peer review in the intervening period and new
data. Once practice-changing evidence, or increasing international interest, are identified, the WHO Therapeutics Steering Committee
triggers the guideline development process. The trigger for producing or updating specific recommendations is based on the following
(any of the three may initiate a recommendation):
• likelihood to change practice;

• sufficient RCT data on therapeutics to inform the high-quality evidence synthesis living systematic review;
• relevance to a global audience.
Step 2: Convening the GDG

WHO selected GDG members to ensure global geographical representation, gender balance, and appropriate technical and clinical
expertise, and patient representatives. For each intervention, the technical unit collected and managed declarations of interests (DOIs)
and found no GDG member to have a conflict of interest. In addition to the distribution of a DOI form, during the meeting, the WHO
Secretariat described the DOI process and an opportunity was given to GDG members to declare any interests not provided in written
form. No verbal conflicts were declared. Web searches did not identify any additional interests that could be perceived to affect an
individual’s objectivity and independence during the development of the recommendations.
The pre-selected expert GDG (see Section 10) convened on 16 and 22 December 2021 to address molnupiravir. The meeting involved a
review of the basics of GRADE methodology including formulating population, intervention, comparator, outcome (PICO) questions and
subgroups of interests, and prioritization of patient-important outcomes (see step 4 below). The GDG subsequently reviewed analyses,
including pre-specified subgroup analyses presented in summary of findings tables, considered an individual patient perspective and
feasibility issues specific to this intervention, and formulated recommendations. The GDG also reviewed the mechanism of actions and
non-clinical evidence around safety.
Step 3: Evidence synthesis

The living systematic review/NMA team, as requested by the WHO Therapeutics Steering Committee, performed an independent
systematic review to examine the benefits and harms of the intervention (1). The systematic review team includes systematic review
experts, clinical experts, clinical epidemiologists and biostatisticians. Team members have expertise in GRADE methodology and rating
certainty of evidence specifically in NMAs. The NMA team considered deliberations from the initial GDG meeting, specifically focusing
88 of 109
on the outcomes and subgroups prioritized by the GDG. The methods team rated credibility of subgroups using the ICEMAN tool (84).
Step 4: Final recommendations

The GRADE approach provided the framework for establishing evidence certainty and generating both the direction and strength of
recommendations (139)(140). A priori voting rules informed procedures if the GDG failed to reach consensus. There was no need for
voting.
The following key factors informed transparent and trustworthy recommendations:
• absolute benefits and harms for all patient-important outcomes through structured evidence summaries (e.g. GRADE summary of
findings tables) (141);
• quality/certainty of the evidence (139)(142);
• values and preferences of patients (143);
• resources and other considerations (including considerations of feasibility, applicability, equity) (143);
• effect estimates and confidence intervals for each outcome, with an associated rating of certainty in the evidence, as presented in
summary of findings tables. If such data are not available, the GDG reviews narrative summaries (141);
• recommendations are rated as either conditional or strong, as defined by GRADE. If the GDG members disagree regarding the
evidence assessment or strength of recommendations, WHO will apply voting according to established rules (140)(143).
When possible, we used research evidence to inform discussion around these key factors. If not available, discussion of these factors
was informed by expert opinion, supported by surveys of the GDG members as outlined below.
Benefits and harms

The GDG members prioritized outcomes (rating from 9 [critical] to 1 [not important] ) in patients with non-severe COVID-19 and in
patients with severe and critical COVID-19, taking a patient perspective (Tables 1 and 2 below). The GDG's questions were structured
using the PICO format (see evidence profile under the recommendations). The prioritization was performed through a survey, most
lately in May 2021, followed by a GDG discussion. These prioritized outcomes were used to update the LNMA (2).
Selecting and rating the importance of outcomes

GDG members prioritized outcomes from the perspective of patients with non-severe illness (Table 1) and severe and critical illness
(Table 2).
Table 1. GDG outcome rating from the perspective of patients with non-severe illness
Outcome Mean SD Range
Admission to hospital 8.5 0.7 7-9

Death 8.1 1.9 3-9
Quality of life 7.5 1.3 5-9
Serious adverse effects (e.g. adverse events leading to drug discontinuation) 7.4 1.8 3-9
Time to symptom resolution 7.3 1.7 4-9
Duration of hospitalization 6.6 0.9 5-8
Duration of oxygen support 6.6 1.2 5-9

Need for invasive mechanical ventilation 5.9 2.3 1-8
New non-SARS-CoV-2 infection 5.6 2.1 3-9
Time to viral clearance 5.5 2.4 1-9

Duration of invasive mechanical ventilation 5.4 2.1 1-8
SD: standard deviation.
Note: 7 to 9 – critical; 4 to 6 – important; 1 to 3 – of limited importance.
89 of 109
Table 2. GDG outcome rating from the perspective of patients with severe and critical illness
Outcome Mean SD Range
Death 9.0 0 9
Need for invasive mechanical ventilation 8.2 0.9 6-9
Duration of invasive mechanical ventilation 7.6 0.9 6-9
Quality of life 6.9 1.3 5-9
Duration of hospitalization 6.7 1.2 4-9
Serious adverse effects (e.g. adverse events leading to drug discontinuation) 6.7 1.8 3-9
Time to symptom resolution 6.5 1.6 4-9
New non-SARS-CoV-2 infection 6.4 1.8 3-9
Duration of oxygen support 6.3 1.3 4-9
Time to viral clearance 4.7 2.3 1-9
SD: standard deviation.

Note: 7 to 9 – critical; 4 to 6 – important; 1 to 3 – of limited importance.
Derivation of absolute effects for drug treatments
For patients with non-severe illness, we used the median of the control arm of the RCTs that contributed to the evidence, identified in
the LNMA (1)(2).
For patients with severe and critical illness, the GDG identified the control arm of the WHO SOLIDARITY trial, performed across a wide
variety of countries and geographical regions, as representing the most relevant source of evidence for baseline risk estimates for
mortality and mechanical ventilation. Systemic corticosteroids now represent standard of care in patients with severe and critical
COVID-19 (see strong recommendation issued by WHO September 2020). Therefore, the baseline risk estimates in the evidence
summaries for JAK inhibitors, convalescent plasma and IL-6 receptor blockers were adjusted for treatment effects of corticosteroids for
the outcome of mortality and mechanical ventilation. The applied baseline risk estimate for mortality was 13% (130 in 1000). For other
outcomes, we used the median of the control arm of the RCTs that contributed to the evidence.
Specific deliberations on baseline risk are presented for each recommendation.
The GDG acknowledged that baseline risks, and thus absolute effects, may vary significantly geographically and over time. Thus, users
of this guideline may prefer estimating absolute effects by using local event rates.
Values and preferences

We had insufficient information to provide the GDG with an evidence-based description of patient experiences or values and
preferences regarding treatment decisions for COVID-19 drug treatments. The GDG, therefore, relied on their own judgments of what
well-informed patients would value after carefully balancing the benefits, harms, and burdens of treatment. Judgments on values and
preferences were crucially informed through the experiences of former COVID-19 patients, represented in the GDG.
The GDG agreed that the following values and preferences would be typical of well-informed patients:
• Most patients would be reluctant to use a medication for which the evidence left high uncertainty regarding effects on outcomes
they consider important. This was particularly so when evidence suggested treatment effects, if they do exist, are small, and the
possibility of important harm remains.
• In an alternative situation with larger benefits and less uncertainty regarding both benefits and harms, more patients would be
inclined to choose the intervention.
In addition to taking an individual patient perspective, the GDG also considered a population perspective in which feasibility,
acceptability, equity and cost were important considerations.
Specific deliberations on values and preferences and associated feasibility and resource related considerations are presented for each
recommendation.
90 of 109
Step 5: External and internal review

An external review group reviewed the final guideline document to identify factual errors, and to comment on clarity of language,
contextual issues and implications for implementation. The technical unit collected and managed declarations of interests (DOIs) of the
external reviewers and found no external reviewer to have a conflict of interest. However, for certain therapeutics, pharmaceutical
company technical representative may be asked to comment on a new drug from the industry perspectives, in line with the WHO
Handbook (page 70), as comments from such individuals or organizations on a draft guideline may be helpful in anticipating and
dealing with controversy, identifying factual errors, and promoting engagement with all stakeholders. Comments on contextual issues
were considered taking into account their interests. The conflict of interest of such individuals will be transparent, as their affiliation
will appear in the acknowledgement section.
The guideline was then reviewed and approved by the WHO GRC and the Publication Review Committee.
91 of 109
8. How to access and use this guideline

This is a living guideline from WHO. The recommendations included here will be updated, and new recommendations will be added for
other drugs for COVID-19.
How to access the guideline:
• WHO website in PDF format (4): This is a full read out of the MAGICapp content for those without reliable web access. It can also
be downloaded directly from MAGICapp (see cogwheel on top right).
• MAGICapp in online, multilayered formats: This is the fullest version of the guideline, as detailed below.
• BMJ Rapid Recommendations (5): Designed with clinical readers in mind and including an interactive infographic to summarize all
treatments included.
• WHO Academy app: Mobile application available for health workers and public on Apple Store and Google Play with a full Case
Management section which includes Guidance, Training and Tools, including the latest training modules on Therapeutics for
COVID-19. Includes treatment and other guidelines and training materials from WHO on COVID-19 for use offline.
• WHO COVID-19 Clinical Care Pathway is a new tool that summarizes these recommendations in a concise and easy to understand
manner for health workers. It links this guideline to WHO guidelines on Diagnostic testing for SARS-CoV-2 and Antigen-detection
in the diagnosis of SARS-CoV-2 infection to aid in implementation.
How to navigate this guideline
The guideline is written, disseminated, and updated in MAGICapp, with a format and structure that ensures user-friendliness and ease
of navigation (138). It accommodates dynamic updating of evidence and recommendations that can focus on what is new while keeping
existing recommendations, as appropriate, within the guideline.
The purpose of the online formats and additional tools, such as the infographics, is to make it easier to navigate and make use of the
guideline in busy clinical practice. The online multilayered formats are designed to allow end-users to find recommendations first and
then drill down to find supporting evidence and other information pertinent to applying the recommendations in practice, including
tools for shared decision-making (clinical encounter decision aids) (138).
Fig. 4 shows how the online multilayered formats are designed to allow end-users to find recommendations first and then drill down to
find supporting information pertinent to applying the recommendations in practice. End-users will also need to understand what is
meant by strong and weak/conditional recommendations (displayed immediately below) and certainty of evidence (the extent to which
the estimates of effect from research represent true effects from treatment).
For each recommendation additional information is available through the following tabs:
• Research evidence: Readers can find details about the research evidence underpinning the recommendations as GRADE Summary
of Findings tables and narrative evidence summaries (shown in Fig. 4).
• Evidence to decision: The absolute benefits and harms are summarized, along with other factors such as the values and preferences
of patients, practical issues around delivering the treatment as well as considerations concerning resources, applicability, feasibility,
equity and human rights. These latter factors are particularly important for those in need of adapting the guidelines for the national
or local context.
• Justification: Explanation of how the GDG considered and integrated evidence to decision factors when creating the
recommendations, focussing on controversial and challenging issues.
• Practical information: For example, dosing, duration and administration of drugs, or how to apply tests to identify patients in
practice.
• Decision aids: Tools for shared decision-making in clinical encounters (19).
92 of 109
Fig. 4. Example of how research evidence is available one click away, with narrative evidence summary giving additional details to
GRADE Summary of Findings table
Additional educational modules and implementation tools for health workers:
• WHO COVID-19 essential supplies forecasting tool (COVID-ESFT) assists governments, partners, and other stakeholders to
forecast the necessary volume of personal protective equipment, diagnostic test equipment, consumable medical supplies,
biomedical equipment for case management, and essential drugs for supportive care and treatment of COVID-19.
• WHO Clinical care for severe acute respiratory infection toolkit: COVID-19 adaptation provides algorithms and practical tools for
clinicians working in acute care hospitals managing adult and paediatric patients with acute respiratory infection, including severe
pneumonia, acute respiratory distress syndrome, sepsis and septic shock. This includes information on screening, testing,
monitoring and treatments.
• WHO Openwho.org clinical management course series hosts a full course series on COVID-19 which covers a holistic pathway of
care for a patient, from screening and triage to rehabilitation, testing and treatments and palliative care.
This living guideline from WHO is also used to inform the activities of the WHO Prequalification of Medicinal Products.
93 of 109
9. Uncertainties, emerging evidence and future research

The guideline recommendations for COVID-19 therapeutics demonstrate remaining uncertainties concerning treatment effects for all
outcomes of importance to patients. There is also a need for better evidence on prognosis and on values and preferences of patients
with COVID-19.
Here we outline key uncertainties for molnupiravir identified by the GDG, adding to those for JAK inhibitors, sotrovimab, convalescent
plasma casirivimab-imdevimab, ivermectin, corticosteroids, remdesivir, hydroxychloroquine, lopinavir/ritonavir, and IL-6 receptor
blockers identified when recommendations were initially formulated in previous versions of the living guideline. These uncertainties
may inform future research, i.e. the production of higher certainty and more relevant evidence to inform policy and practice. We also
outline emerging evidence in the rapidly changing landscape of trials for COVID-19.
Ongoing uncertainties and opportunities for future research
Molnupiravir
• need for clinical data to investigate safety and applicability concerns (including in children, lactating or pregnant women, and men;
and long-term impact on mutagenesis and cancer risk);
• accurate clinical prediction guides to establish the individual patient risk of hospitalization in patients presenting with non-severe
COVID-19 in order to best identify patients that would most benefit from this intervention;
• data to inform individual and population-level concerns, such as the emergence of resistance and efficacy against new variants;
• comparative effectiveness of molnupiravir compared with other treatment options (eg. monoclonal antibodies or other antivirals) in
the non-severe population, including combination therapy;
• the relative intracellular nucleotide ratios of endogenous: molnupiravir cell lines and animal models to assess genetic toxicity;
• how readily mutations arise under a selective pressure with NHC in vitro and molnupiravir in animal models and patients with SARS-
CoV-2 infection;
• if mutations arising under selective pressure in vitro, in vivo or in humans:
◦ confer a decreased antiviral activity for NHC;
◦ arise in the spike protein and/or do they confer an increase in replicative potential/transmission.
JAK inhibitors
• incremental benefit for patients receiving baricitinib and IL-6 receptor blockers together, rather than either drug individually;
• relative benefits of tofacitinib and ruxolitinib to baricitinib;
• safety and efficacy in children, and pregnant and lactating women.
Sotrovimab
• accurate clinical prediction guides to establish the individual patient risk of hospitalization in patients presenting with non-severe
COVID-19, in order to best identify patients who would most benefit from this intervention;
• efficacy and safety for severe and critical seronegative COVID-19 patients, for patients infected with emerging variants, as well as
in children and pregnant women.
Convalescent plasma
• effects in severe and critical illness (low to moderate certainty evidence for most patient-important outcomes);
• long-term mortality and functional outcomes in COVID-19 survivors;
• safety and efficacy in children, pregnant, and lactating women;
• effects of high-titre convalescent plasma on mortality and other patient-important outcomes;
• effects in patients with seronegative antibody status.
Casirivimab-imdevimab
• accurate clinical prediction guides to establish individual patient risk of hospitalization in patients presenting with non-severe
COVID-19 in order to best identify patients that would most benefit from this intervention;
• dosing and administration routes in non-severe and severe/critical COVID-19 patients;
• safety and efficacy in children and pregnant women.
94 of 109
IL-6 receptor blockers

• safety data in terms of nosocomial infections;
• data in children, pregnant patients and those that are already immunocompromised;
• patients with non-severe COVID-19;
• immunity and the risk of a subsequent infection, which may impact the risk of death after 28 days;
• outcomes by different IL-6 receptor blocker dosing and optimal timing of drug initiation.
Ivermectin
Given the very low certainty in estimates for most critical outcomes of interest, the GDG felt that further high-quality clinical trials
examining this drug would be essential before any recommendation for use as part of clinical care. This includes further RCTs examining
both inpatients and outpatients and those with varying disease severities and using different ivermectin dosing regimens. The focus of
these studies should be on outcomes important to patients such as mortality, quality of life, need for hospitalization, need for invasive
mechanical ventilation and time to clinical or symptom improvement. Also, a better characterization of potential harms with ivermectin
in patients with COVID-19 would be important.
Hydroxychloroquine
Although some uncertainty remains, the GDG panel felt that further research was unlikely to uncover a subgroup of patients that would
benefit from hydroxychloroquine on the most important outcomes (mortality, mechanical ventilation) given the consistent results in
trials across disease severity and location.
Lopinavir/ritonavir
Although some uncertainty remains, the GDG panel felt that further research was unlikely to uncover a subgroup of patients that would
benefit from lopinavir/ritonavir on the most important outcomes (mortality, mechanical ventilation) given the consistent results in trials
across disease severity and location.
Remdesivir
• critical outcomes of interest, particularly those that impact resource allocation, such as the need for mechanical ventilation,
duration of mechanical ventilation and duration of hospitalization;
• specific subgroups, such as different severities of illness, different time (days) since onset of illness, children and older adults,
pregnant women, and duration of therapy;
• long-term outcomes such as mortality at extended endpoints or long-term quality of life;
• long-term safety and rare but important side-effects;
• patient-reported outcomes such as symptom burden;
• outcomes, when used in combination with other agents, such as, but not limited to, corticosteroids;
• impact on viral shedding, viral clearance, patient infectivity.
Corticosteroids

• patients with non-severe COVID-19 (i.e. pneumonia without hypoxaemia);
• outcomes, when used in combination with additional therapies for COVID-19, such as novel immunomodulators. It will become
increasingly important to ascertain how these interact with systemic corticosteroids. All investigational therapies for severe and
critical COVID-19 (including remdesivir) should be compared with systemic corticosteroids or evaluated in combination with
systemic corticosteroids vs systemic corticosteroids alone;
• immunity and the risk of a subsequent infection, which may impact the risk of death after 28 days;
• outcomes, by different steroid preparation, dosing, and optimal timing of drug initiation.
Emerging evidence
The unprecedented volume of planned and ongoing studies for COVID-19 interventions – over 5000 RCTs as of 4 January 2022 –
implies that more reliable and relevant evidence will emerge to inform policy and practice (13). An overview of registered and ongoing
trials for COVID-19 therapeutics and prophylaxis is available from the Infectious Diseases Data Observatory, through their living
systematic review of COVID-19 clinical trial registrations (13), the WHO website and other repositories, such as the COVID-NMA
initiative.
Whereas most of these studies are small and of variable methodological quality, a number of large, international platform trials (e.g.
RECOVERY, SOLIDARITY, and DISCOVERY) are better equipped to provide robust evidence for a number of potential treatment
95 of 109
options (14)(15)(16)(17). Such trials can also adapt their design, recruitment strategies, and selection of interventions based on new
insights, exemplified by the uncertainties outlined above.
96 of 109
10. Authorship, contributions, acknowledgements

Authorship, contributions, acknowledgements
WHO would like to thank the collaborative efforts of all those involved to make this process rapid, efficient, trustworthy and
transparent.
WHO Therapeutics Steering Committee (updated for molnupiravir)

The committee includes representatives from various WHO departments at headquarters and the regions and has been approved by the
WHO Director of the Country Readiness Department, and the WHO Chief Scientist. The WHO Secretariat meets on a regular basis to
discuss when to trigger guideline updates based on evidence updates from the WHO rapid review team, and other sources of evidence
and selects the members of the Guideline Development Group (GDG) for the living guideline.
Janet V Diaz (Lead, Clinical Team for COVID-19 Response, Health Emergencies Programme, Geneva); John Appiah (Lead, Case
Management, WHO Regional Office for Africa); Lisa Askie (Quality Assurance of Norms and Standards Department); Silvia Bertagnolio
(Communicable and Noncommunicable Diseases Division/Clinical Team for COVID-19 Response); Chiori Kodama (WHO Regional Office
for the Eastern Mediterranean); Krutika Kuppalli (Clinical Team for COVID-19 Response, Health Emergencies Programme, Geneva);
Marta Lado Castro-Rial (Clinical Team for COVID-19 Response, Health Emergencies Programme, Geneva); Lorenzo Moja (Health
Products Policy and Standards Department); Olufemi Oladapo (Sexual and Reproductive Health and Research Department); Dina Pfeifer
(WHO Regional Office for Europe/Health Emergencies Programme); J Pryanka Relan (Clinical Team for COVID-19 Response, Health
Emergencies Programme, Geneva); Ludovic Reveiz (Evidence and Intelligence for Action in Health Department, Incident Management
Systems for COVID-19, Pan American Health Organization); Vaseeharan Sathiyamoorthy (Research for Health, Science Division);
Anthony Solomon (Neglected Tropical Diseases); Pushpa Wijesinghe (Lead, Case Management, Regional Office for South-East Asia).
Supporting project officers: Julie Viry and Anne Colin (Clinical Team for COVID-19 Response, Health Emergencies Programme, Geneva).
The WHO Therapeutics Steering Committee is fully responsible for decisions about guidance production and convening the GDG.
Special thanks to the WHO Pharmacovigilance team for their support and contributions to this update: Noha Iessa and Shanti Pal.
Guideline Development Group (GDG) for molnupiravir recommendation. For list of GDG members of previous recommendations, see
here.
Wagdy Amin (Ministry of Health and Population, Egypt); Fabien Alberto Jaimes Barragan (Antioquia University Medellin, Colombia);
Frederique Jacquerioz Bausch (Department of Primary Care, University Hospital of Geneva, Switzerland); Maurizio Cecconi (Humanitas
Research Hospital Milan, Italy); Duncan Chanda (Institute for Medical Research and Training, Zambia); Vu Quoc Dat (Department of
Infectious Diseases, Hanoi Medical University, Hanoi, Viet Nam); Ann De Sutter (Ghent Faculty of Medicine and Health Science,
Belgium); Heike Geduld (Emergency Medicine, Stellenbosch University, South Africa); Patrick Gee (patient panel member, Virginia,
United States); Matthias Gotte (University of Alberta, Canada); Nerina Harley (Royal Melbourne Hospital and Epworth Healthcare,
Melbourne, Australia); Manai Hela (Emergency Medical Service Tunis, Tunisia); Beverley Hunt (King's College London, United Kingdom);
Fyezah Jehan (Aga Khan University, Pakistan); Sushil Kumar Kabra (All India Institute of Medical Sciences, New Delhi, India); Yae Jean
Kim (Sungkyunkwan University School of Medicine, Samsung Medical Center, Seoul, Republic of Korea); Niranjan Kissoon (Department
of Paediatrics and Emergency Medicine, University of British Columbia, Vancouver, Canada); Sanjeev Krishna (St George’s University of
London, United Kingdom); Arthur Kwizera (Makerere University College of Health Sciences, Mulago National Referral Hospital ICU,
Uganda); Yee-Sin Leo (National Centre for Infectious Diseases, Singapore); Thiago Lisboa (Coraçao Hospital, Sao Paulo, Brazil); Imelda
Mahaka (Pangaea Zimbabwe AIDS Trust, Harare, Zimbabwe); Emmanuel Nsutebu (Sheikh Shakhbout Medical City, Abu Dhabi); Natalia
Pshenichnaya (Central Research Institute of Epidemiology of Rospotrebnadzor, Moscow, Russian Federation); Rohit Sarin (National
Institute of Tuberculosis and Respiratory Diseases, New Delhi, India); Manu Shankar-Hari (King’s College London, United Kingdom);
Yinzhong Shen (Shanghai Public Health Clinical Center, Fudan University, Shanghai, China); Shalini Sri Ranganathan (University of
Colombo, Sri Lanka); Miriam Stegemann (Charité - Universitätsmedizin Berlin, Germany); Ronald Swanstrom (Department of
Biochemistry & Biophysics of the University of North Carolina, United States); Tim Uyeki (Influenza Division, U.S. Centers for Disease
Control and Prevention, United States); Sridhar Venkatapuram (King’s College London, United Kingdom); Ananda Wijewickrama
(Ministry of Health, Sri Lanka).
Guideline Development Group (GDG) for JAK inhibitors recommendation. For list of GDG members of previous recommendations,
see here.
Wagdy Amin (Ministry of Health and Population, Egypt); Carolyn S Calfee (University of California, San Francisco); Duncan Chanda
(Institute for Medical Research and Training, Zambia); Vu Quoc Dat (Department of Infectious Diseases, Hanoi Medical University,
Hanoi, Viet Nam); Ann De Sutter (Ghent Faculty of Medicine and Health Science, Belgium); Beverley Hunt (King's College London,
United Kingdom); Heike Geduld (Emergency Medicine, Stellenbosch University, South Africa); Yae Jean Kim (Sungkyunkwan University
School of Medicine, Samsung Medical Center, Seoul, Republic of Korea); Sanjeev Krishna (St George’s University of London, United
Kingdom); Natalia Pshenichnaya (Central Research Institute of Epidemiology of Rospotrebnadzor, Moscow, Russian Federation); Saniya
Sabzwari (Aga Khan University, Karachi, Pakistan); Rohit Sarin (National Institute of Tuberculosis and Respiratory Diseases, New Delhi,
India); Yinzhong Shen (Shanghai Public Health Clinical Center, Fudan University, Shanghai, China); Shalini Sri Ranganathan (University of
97 of 109
Colombo, Sri Lanka); Miriam Stegemann (Charité - Universitätsmedizin Berlin, Germany); Sridhar Venkatapuram (King’s College London,
United Kingdom); Ananda Wijewickrama (Ministry of Health, Sri Lanka); Yee-Sin Leo (National Centre for Infectious Diseases,
Singapore).
Guideline Development Group (GDG) for sotrovimab recommendation. For list of GDG members of previous recommendations, see
here.
Wagdy Amin (Ministry of Health and Population, Egypt); Fabien Alberto Jaimes Barragan (Antioquia University Medellin, Colombia);
Duncan Chanda (Institute for Medical Research and Training, Zambia); Vu Quoc Dat (Department of Infectious Diseases, Hanoi Medical
University, Hanoi, Vietnam); Ann De Sutter (Ghent Faculty of Medicine and Health Science, Belgium); Heike Geduld (Emergency
Medicine, Stellenbosch University, South Africa); Nerina Harley (Royal Melbourne Hospital and Epworth Healthcare, Melbourne,
Australia); Beverley Hunt (King's College, London); Fyezah Jehan (Aga Khan University, Pakistan); Sushil Kumar Kabra (All India Institute
of Medical Sciences, New Delhi, India); Yae Jean Kim (Sungkyunkwan University School of Medicine, Samsung Medical Center, Seoul,
Republic of Korea); Niranjan Kissoon (Department of Paediatrics and Emergency Medicine, University of British Columbia, Vancouver,
Canada); Sanjeev Krishna (St George’s University of London, United Kingdom); Thiago Lisboa (Coraçao Hospital, Sao Paulo, Brazil);
Natalia Pshenichnaya (Central Research Institute of Epidemiology of Rospotrebnadzor, Moscow, Russian Federation); Rohit Sarin
(National Institute of Tuberculosis and Respiratory Diseases, New Delhi, India); Manu Shankar-Hari (King’s College London, United
Kingdom), Yinzhong Shen (Shanghai Public Health Clinical Center, Fudan University, Shanghai, China); Shalini Sri Ranganathan
(University of Colombo, Sri Lanka); Ronald Swanstrom (University of North Carolina, United States); Miriam Stegemann (Charité -
Universitätsmedizin Berlin, Germany); Sridhar Venkatapuram (King’s College London, United Kingdom); Yee-Sin Leo (National Centre for
Infectious Diseases, Singapore).
Guideline Development Group (GDG) for convalescent plasma guideline. For list of GDG members of previous recommendations, see
here.
Wagdy Amin (Ministry of Health and Population, Egypt); Erlina Burhan (Infection Division Department of Pulmonology and Respiratory
Medicine, Faculty of Medicine Universitas Indonesia); Carolyn S Calfee (University of California, San Francisco); Maurizio Cecconi
(Humanitas Research Hospital Milan, Italy); Vu Quoc Dat (Department of Infectious Diseases, Hanoi Medical University, Hanoi, Viet
Nam); Heike Geduld (Emergency Medicine, Stellenbosch University, South Africa); Patrick Gee (patient panel member, United States of
America); Nerina Harley (Royal Melbourne Hospital and Epworth Healthcare, Melbourne, Australia); Madiha Hashmi (Ziauddin
University, Karachi, Pakistan); Sushil Kumar Kabra (All India Institute of Medical Sciences, New Delhi, India); Seema Kanda (patient panel
member, Ontario, Canada); Leticia Kawano-Dourado (Research Institute, Hospital do Coração, São Paulo, Brazil); Niranjan Kissoon
(Department of Paediatrics and Emergency Medicine, University of British Columbia, Vancouver, Canada); Greta Mino (Alcivar Hospital
in Guayaquil, Ecuador); Natalia Pshenichnaya (Central Research Institute of Epidemiology of Rospotrebnadzor, Moscow, Russian
Federation); Nida Qadir (Pulmonary and Critical Care Medicine, David Geffen School of Medicine, University of California, Los Angeles,
United States); Saniya Sabzwari (Aga Khan University, Karachi, Pakistan); Rohit Sarin (National Institute of Tuberculosis and Respiratory
Diseases, New Delhi, India); Yinzhong Shen (Shanghai Public Health Clinical Center, Fudan University, Shanghai, China); Shalini Sri
Ranganathan (University of Colombo, Sri Lanka); Miriam Stegemann (Charité - Universitätsmedizin Berlin, Germany); Sridhar
Venkatapuram (King’s College, London); Ananda Wijewickrama (Ministry of Health, Sri Lanka).
Methods chairs
Gordon Guyatt (casirivimab-imdevimab, JAK inhibitors), Bram Rochwerg (IL-6 receptor blockers, ivermectin, remdesivir and lopinavir/
ritonavir, convalescent plasma, molnupiravir), Reed Siemieniuk (hydroxychloroquine), Francois Lamontagne (corticosteroids, sotrovimab).
Clinical chairs
Michael Jacobs (casirivimab-imdevimab, IL-6 receptor blockers, ivermectin, remdesivir, hydroxychloroquine, lopinavir/ritonavir, JAK
inhibitors), Yee-Sin Leo (corticosteroids), Leticia Kawano-Dourado (convalescent plasma, sotrovimab, molnupiravir).
Methods resource persons

Arnav Agarwal (University of Toronto, Canada); Thomas Agoritsas (University Hospitals of Geneva, Switzerland); Romina Brignardello-
Petersen (McMaster University, Canada); Gordon H Guyatt (McMaster University, Canada); George Tomlinson (Department of Medicine,
University Health Network, Toronto, Canada); Per Olav Vandvik (MAGIC, University of Oslo Norway), Linan Zeng (West China Second
University Hospital, Sichuan University, Chengdu, China; McMaster University, Canada).
Guideline Support Collaboration Committee which provides the coordination between WHO and MAGIC to allow the rapid
development of the WHO guideline and its dissemination into the various publication platforms: Thomas Agoritsas (MAGIC, University
Hospitals of Geneva); Janet Diaz (World Health Organization), Helen McDonald (British Medical Journal); Gordon Guyatt (McMaster
University, Canada); Per Olav Vandvik (MAGIC, University of Oslo), Julie Viry (World Health Organization).
Temporary advisors:
Special thanks to Professor Andrew Owen (Department of Molecular and Clinical Pharmacology, University of Liverpool) for his
contributions in the pharmacokinetics of ivermectin, IL-6 receptor blockers, casirivimab-imdevimab, convalescent plasma and the
98 of 109
monoclonal antibodies, JAK inhibitors, sotrovimab and molnupiravir.
Special thanks to Professor Craig Thompson (University of Oxford) for his contributions in diagnostic testing as it relates to the
casrivimab-imdevimab guideline.
External reviewers
Special thanks to the external reviewers for their insights on convalescent plasma, casrivimab-imdevimab, IL-6 receptor blockers,
ivermectin, sotrovimab, and molnupiravir.
Aula Abbara (Médecins Sans Frontières); Yaseen Arabi (King Saud Bin Abdulaziz University for Health Sciences, Saudi Arabia); Marcio da
Fonseca (Médecins Sans Frontières); Richard Kojan (Alliance for International Medical Action); Carolina Nanclares (Médecins Sans
Frontières); Saschveen Singh (Médecins Sans Frontières).
Special thanks to Paula Dakin (Regeneron Pharmaceuticals Inc) who was invited to comment on on casirimab-imdevimab (version six) to
identify factual errors, and to comment on clarity of language, contextual issues and implications for implementation. Their comments
were considered taking into account the interests of Regeneron Pharmaceuticals Inc.
Special thanks to Lisa Burry (Department of Pharmacy, Mont Sinai Hospital, Toronto) for her contributions to the Practical information
sheets as clinical pharmacists.
Infographic
Special thanks to the BMJ for providing the infographic for this guideline.
Funding
Many thanks to the Bill & Melinda Gates Foundation, Norwegian Directorate of Public Health and Germany.
Special thanks to the MAGIC Evidence Ecosystem Foundation, which provides pro bono methodologic support.
Meta-analysis teams
Special thanks to the McMaster University LNMA team

Arnav Agarwal (University of Toronto, Canada); Thomas Agoritsas (MAGIC; University Hospitals of Geneva, Switzerland); Jessica J
Bartoszko (McMaster University, Canada); Romina Brignardello-Petersen (McMaster University, Canada); Derek K Chu (McMaster
University, Canada); Rachel Couban (McMaster University, Canada); Andrea Darzi (McMaster University, Canada); Tahira Devji
(McMaster University, Canada); Bo Fang (Chongqing Medical University, China); Carmen Fang (William Osler Health Network, Canada);
Signe Agnes Flottorp (University of Oslo, Norway); Farid Foroutan (McMaster University, Canada); Long Ge (Lanzhou University, China);
Gordon H Guyatt (McMaster University, Canada); Mi Ah Han (Chosun University, Republic of Korea); Diane Heels-Ansdell (McMaster
University, Canada); Kimia Honarmand (Western University, Canada); Liangying Hou (Lanzhou University, China); Xiaorong Hou
(Chongqing Medical University, China); Quazi Ibrahim (McMaster University, Canada); Ariel Izcovich (Servicio de Clinica Médica del
Hospital Alemán, Argentina); Elena Kum (McMaster University, Canada); Francois Lamontagne (Université de Sherbrooke, Canada); Qin
Liu (Chongqing Medical University, China); Mark Loeb (McMaster University, Canada); Maura Marcucci (McMaster University, Canada);
Shelley L McLeod (Sinai Health, Canada); Sharhzad Motaghi, (McMaster University, Canada); Srinivas Murthy (University of British
Columbia, Canada); Reem A Mustafa (McMaster University, Canada); John D Neary (McMaster University, Canada); Hector Pardo-
Hernandez (Sant Pau Biomedical Research Institute, Spain); Anila Qasim (McMaster University, Canada); Gabriel Rada (Epistemonikos
Foundation, Chile); Irbaz Bin Riaz (Mayo Clinic Rochester, United States of America); Bram Rochwerg (McMaster University, Canada);
Behnam Sadeghirad (McMaster University, Canada); Nigar Sekercioglu (McMaster University, Canada); Lulu Sheng (Chongqing Medical
University, China); Reed AC Siemieniuk (McMaster University, Canada); Ashwini Sreekanta (McMaster University, Canada); Charlotte
Switzer (McMaster University, Canada); Britta Tendal (Monash University, Australia); Lehana Thabane (McMaster University, Canada);
George Tomlinson (University of Toronto, Canada); Tari Turner (Monash University, Australia); Per Olav Vandvik (MAGIC; University of
Oslo, Norway); Robin WM Vernooij (University Medical Center Utrecht, Netherlands); Andrés Viteri-García (Epistemonikos Foundation,
Chile); Ying Wang (McMaster University, Canada); Liang Yao (McMaster University, Canada); Zhikang Ye (McMaster University, Canada);
Dena Zeraatkar (McMaster University, Canada) (1)(2)(3).
Special thanks to WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group for their publication: Association
of administration of interleukin-6 antagonists with mortality and other outcomes among hospitalized patients with COVID-19: a
prospective meta-analysis (90).
Manu Shankar-Hari; Claire L Vale; Peter J Godolphin; David Fisher; Julian PT Higgins; Francesca Spiga; Jelena Savović; Jayne Tierney;
Nor Arisah Misnan; Gabriel Baron; Julie S Benbenishty; Lindsay R Berry; Niklas Broman; Alexandre Biasi Cavalcanti; Roos Colman;
Stefanie L De Buyser; Lennie PG Derde; Pere Domingo; Sharifah Faridah Syed Omar; Ana Fernandez-Cruz; Thijs Feuth; Felipe Garcia;
99 of 109
Rosario Garcia-Vicuna; Isidoro Gonzalez-Alvaro; Anthony C Gordon; Richard Haynes; Olivier Hermine; Peter W Horby; Nora K Horick;
Kuldeep Kumar: Bart N Lambrecht; Martin J Landray; Lorna Leal; David J Lederer; Elizabeth Lorenzi; Xavier Mariette; Nicolas Merchante;
Nor Arisah Misnan; Shalini V Mohan; Michael C Nivens; Jarmo Oksi; Jose A Perez-Molina; Reuven Pizov; Raphael Porcher; Simone
Postma; Reena Rajasuriar; Athimalaipet V Ramanan; Pankti D Reid; Abraham Rutgers; Aranzazu Sancho-Lopez; Todd B Seto; Sumathi
Sivapalasingam; Arvinder Singh Soin; Natalie Staplin; John H Stone; Garth W Strohbehn; Jonas Sunden-Cullberg; Julian Torre-Cisneros;
Larry W Tsai; Hubert van Hoogstraten; Tom van Meerten; Viviane Cordeiro Veiga; Peter Westerwheel; Srinivas Murthy; Janet V Diaz;
John C Marshall; Jonathan A C Sterne.
We would like to thank Hetero, Dr Reddy's Laboratories and MSD (Known as Merck in the United States and Canada) and Ridgeback
Biotherapeutics for sharing pre-published data that was used to conduct the meta-analysis which informs this most recent WHO Living
guideline update (V9).
100 of 109
References
1. Siemieniuk RAC, Bartoszko JJ, Ge L, Zeraatkar D, Izcovich A, Pardo-Hernandez H, et al. : Drug treatments for covid-19: living
systematic review and network meta-analysis. BMJ 2020;370: m2980 Pubmed Journal
2. Siemieniuk RAC, Bartoszko JJ, Díaz Martinez JP, Kum E, Qasim A, Zeraatkar D, et al. : Antibody and cellular therapies for treatment of
covid-19: a living systematic review and network meta-analysis. BMJ 2021;374 n2231. Journal
3. Zeraatkar D, Cusano E, Diaz Martinez JP, Qasim A, Mangala S, Kum E, et al. : Tocilizumab and sarilumab alone or in combination with
corticosteroids for COVID-19: a systematic review and network meta-analysis. medRxiv 2021 (https://www.medrxiv.org/content/
10.1101/2021.07.05.21259867v1, accessed 30 June 2021). Journal
4. Therapeutics and COVID-19. Geneva: World Health Organization; 2021 (https://www.who.int/teams/health-care-readiness-clinical-

unit/covid-19/therapeutics, accessed 30 November 2021). Website
5. Lamontagne F, Agoritsas T, Macdonald H, Leo Y-S, Diaz J, Agarwal A, et al. : A living WHO guideline on drugs for covid-19. BMJ
2020;370 m3379 Pubmed Journal
6. Living guidance for clinical management of COVID-19. Geneva: World Health Organization; 2021 (WHO/2019-nCoV/clinical/2021.2;
https://www.who.int/publications/i/item/WHO-2019-nCoV-clinical-2021-2, accessed 7 December 2021). Website
7. WHO Living guideline: Drugs to prevent COVID-19. Geneva: World Health Organization; 2021 (WHO-2019-nCoV-
prophylaxes-2021.1; https://www.who.int/publications/i/item/WHO-2019-nCoV-prophylaxes-2021-1, accessed 6 July 2021). Website
8. Lamontagne F, Agoritsas T, Siemieniuk R, Rochwerg B, Bartoszko J, Askie L, et al. : A living WHO guideline on drugs to prevent
covid-19. BMJ 2021;372: n526. Pubmed Journal
9. Bartoszko JJ, Siemieniuk RAC, Kum E, Qasim A, Zeraatkar D, Ge L, et al. : Prophylaxis against covid-19: living systematic review and
network meta-analysis. BMJ 2021;373 n949. Journal Website
10. WHO Coronavirus (COVID-19) Dashboard [online database]. Geneva: World Health Organization; 2021 (https://covid19.who.int,
accessed 4 December 2021). Website
11. Coronavirus (COVID-19) Vaccinations [online resource]. Our World in Data; 2021 (https://ourworldindata.org/covid-vaccinations,
accessed 4 December 2021). Website
12. Naci H, Kesselheim AS, Røttingen JA, Salanti G, Vandvik PO, Cipriani A : Producing and using timely comparative evidence on drugs:
lessons from clinical trials for COVID-19. BMJ 2020;371: m3869. Pubmed Journal
13. Maguire BJ, Guérin PJ : A living systematic review protocol for COVID-19 clinical trial registrations. Wellcome Open Res 2020;5 60.
Pubmed Journal
14. RECOVERY Collaborative Group : Dexamethasone in hospitalized patients with COVID-19 - preliminary report. New Eng J Med
2020;384(8):693-704. Pubmed Journal
15. WHO SOLIDARITY Trial Consortium : Repurposed antiviral drugs for COVID-19 - interim WHO SOLIDARITY trial results. New Eng
J Med 2021;384: 497-511. Pubmed Journal
16. The Writing Committee for the REMAP-CAP Investigators : Effect of hydrocortisone on mortality and organ support in patients with
severe COVID-19: The REMAP-CAP COVID-19 corticosteroid domain randomized clinical trial. JAMA 2020;324(14):1317-1329.
Pubmed Journal
17. ACTIV-3/TICO LY-CoV555 Study Group : A neutralizing monoclonal antibody for hospitalized patients with COVID-19. New Eng J
101 of 109
Med 2021;384(10):905-914. Pubmed Journal
18. Wynants L, Van Calster B, Collins GS, Riley RD, Heinze G, Schuit E, et al. : Prediction models for diagnosis and prognosis of covid-19:
systematic review and critical appraisal. BMJ 2020;369 m1328 Pubmed Journal
19. Jayk Bernal A, Gomes da Silva MM, Musungaie DB, Kovalchuk E, Gonzalez A, Delos Reyes V, et al. : Molnupiravir for Oral Treatment
of Covid-19 in Nonhospitalized Patients. New Eng J Med 2021;386(6):509-520. Pubmed Journal
20. Hill A, Ellis L, Wang J, Pepperrell T : Prices versus costs of production for molnupiravir as a COVID-19 treatment. Research Square
2022; Journal Website
21. Antigen-detection in the diagnosis of SARS-CoV-2 infection. Geneva: World Health Organization; 2021 (https://www.who.int/
publications/i/item/antigen-detection-in-the-diagnosis-of-sars-cov-2infection-using-rapid-immunoassays, accessed 20 February 2022).
Website
22. Hippisley-Cox J, Coupland CA, Mehta N, Keogh RH, Diaz-Ordaz K, Khunti K, et al. : Risk prediction of covid-19 related death and
hospital admission in adults after covid-19 vaccination: national prospective cohort study. BMJ 2021;374 n2244. Pubmed Journal
23. European Medicines Agency : Use of molnupiravir for the treatment of COVID-19. (https://www.ema.europa.eu/en/documents/
referral/lagevrio-also-known-molnupiravir-mk-4482-covid-19-article-53-procedure-assessment-report_en.pdf, accessed 10 February
2022). Website
24. Sheahan TP, Sims AC, Zhou S, Graham RL, Pruijssers AJ, Agostini ML, et al. : An orally bioavailable broad-spectrum antiviral inhibits
SARS-CoV-2 in human airway epithelial cell cultures and multiple coronaviruses in mice. Science translational medicine
2020;12(541):eabb5883. Pubmed Journal
25. Tao S, Zandi K, Bassit L, Ong YT, Verma K, Liu P, et al. : Comparison of anti-SARS-CoV-2 activity and intracellular metabolism of
remdesivir and its parent nucleoside. Current research in pharmacology and drug discovery 2021;2 100045. Pubmed Journal
26. Kabinger F, Stiller C, Schmitzová J, Dienemann C, Kokic G, Hillen HS, et al. : Mechanism of molnupiravir-induced SARS-CoV-2
mutagenesis. Nature structural & molecular biology 2021;28(9):740-746. Pubmed Journal
27. Gordon CJ, Tchesnokov EP, Schinazi RF, Götte M : Molnupiravir promotes SARS-CoV-2 mutagenesis via the RNA template. The
Journal of biological chemistry 2021;297(1):100770. Pubmed Journal
28. Painter WP, Holman W, Bush JA, Almazedi F, Malik H, Eraut NCJE, et al. : Human Safety, Tolerability, and Pharmacokinetics of
Molnupiravir, a Novel Broad-Spectrum Oral Antiviral Agent with Activity Against SARS-CoV-2. Antimicrobial agents and chemotherapy
2021;65(5):e02428-20. Pubmed Journal
29. Humeniuk R, Mathias A, Cao H, Osinusi A, Shen G, Chng E, et al. : Safety, Tolerability, and Pharmacokinetics of Remdesivir, An
Antiviral for Treatment of COVID-19, in Healthy Subjects. Clinical and translational science 2020;13(5):896-906. Pubmed Journal
30. Rosenke K, Hansen F, Schwarz B, Feldmann F, Haddock E, Rosenke R, et al. : Orally delivered MK-4482 inhibits SARS-CoV-2
replication in the Syrian hamster model. Research square 2020; Pubmed Journal
31. Cox RM, Wolf JD, Plemper RK : Therapeutically administered ribonucleoside analogue MK-4482/EIDD-2801 blocks SARS-CoV-2
transmission in ferrets. Nature microbiology 2021;6(1):11-18. Pubmed Journal
32. Abdelnabi R, Foo CS, Kaptein SJF, Zhang X, Do TND, Langendries L, et al. : The combined treatment of Molnupiravir and Favipiravir
results in a potentiation of antiviral efficacy in a SARS-CoV-2 hamster infection model. EBioMedicine 2021;72 103595. Pubmed Journal
33. Abdelnabi R, Foo CS, De Jonghe S, Maes P, Weynand B, Neyts J : Molnupiravir Inhibits Replication of the Emerging SARS-CoV-2
Variants of Concern in a Hamster Infection Model. The Journal of infectious diseases 2021;224(5):749-753. Pubmed Journal
34. Prince T, Donovan-Banfield I, Goldswain H, Penrice-Randal R, Turtle L, Fletcher T, et al. : Antiviral activity of molnupiravir precursor
102 of 109
NHC against Variants of Concern (VOCs) and its therapeutic window in a human lung cell model. bioRxiv 2021; Journal Website
35. Vangeel L, De Jonghe S, Maes P, Slechten B, Raymenants J, André E, et al. : Remdesivir, Molnupiravir and Nirmatrelvir remain active
against SARS-CoV-2 Omicron and other variants of concern. bioRxiv 2021; Journal Website
36. Haseltine WA : Supercharging New Viral Variants: The Dangers Of Molnupiravir (Part 1). Forbes 2021; Website
37. November 30, 2021: Antimicrobial Drugs Advisory Committee Meeting Announcement. Food and Drug Administration
(https://www.fda.gov/advisory-committees/advisory-committee-calendar/november-30-2021-antimicrobial-drugs-advisory-committee-
meeting-announcement-11302021, accessed 20 February 2022). Website
38. Fact sheet for healthcare providers - Emergency Use Authorization (EUA) of baricitinib. United States Food and Drug Administration.
(https://www.fda.gov/media/143823/download, accessed 4 January 2022). Website
39. Ely EW, Ramanan AV, Kartman CE, de Bono S, Liao R, Piruzeli MLB, et al. : Baricitinib plus standard of care for hospitalised adults
with COVID-19 on invasive mechanical ventilation or extracorporeal membrane oxygenation: results of a randomised, placebo-
controlled trial. medRxiv 2021; Journal Website
40. Kalil AC, Patterson TF, Mehta AK, Tomashek KM, Wolfe CR, Ghazaryan V, et al. : Baricitinib plus Remdesivir for Hospitalized Adults
with Covid-19. New Eng J Med 2021;384(9):795-807. Pubmed Journal
41. Marconi VC, Ramanan AV, de Bono S, Kartman CE, Krishnan V, Liao R, et al. : Efficacy and safety of baricitinib for the treatment of
hospitalised adults with COVID-19 (COV-BARRIER): a randomised, double-blind, parallel-group, placebo-controlled phase 3 trial. Lancet
Respir Med 2021;9(12):1407-1418. Pubmed Journal
42. Cao Y, Wei J, Zou L, Jiang T, Wang G, Chen L, et al. : Ruxolitinib in treatment of severe coronavirus disease 2019 (COVID-19):
A multicenter, single-blind, randomized controlled trial. J Allergy Clin Immunol 2020;146(1):137-146.e3. Pubmed Journal
43. Study to assess the efficacy and safety of ruxolitinib in patients with COVID-19 associated cytokine storm (RUXCOVID).
ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). 2000 Feb 29. Identifier: NCT04362137.
(https://clinicaltrials.gov/ct2/show/results/NCT04362137?view=results, accessed 4 January 2022). Website
44. Guimarães PO, Quirk D, Furtado RH, Maia LN, Saraiva JF, Antunes MO, et al. : Tofacitinib in patients hospitalized with COVID-19
pneumonia. New Eng J Med 2021;385(5):406-415. Pubmed Journal
45. Mayence A, Vanden Eynde JJ : Baricitinib: A 2018 Novel FDA-Approved Small Molecule Inhibiting Janus Kinases. Pharmaceuticals
2019;12(1):37. Pubmed Journal
46. Fragoulis GE, McInnes IB, Siebert S : JAK-inhibitors. New players in the field of immune-mediated diseases, beyond rheumatoid
arthritis. Rheumatology 2019;58(Suppl 1):i43-i54. Pubmed Journal
47. Schwartz DM, Kanno Y, Villarino A, Ward M, Gadina M, O'Shea JJ : JAK inhibition as a therapeutic strategy for immune and
inflammatory diseases. Nat Rev Drug Discov 2017;17(1):78. Pubmed Journal
48. Wilhelm A, Widera M, Grikscheit K, Toptan T, Schenk B, Pallas C, et al. : Reduced neutralization of SARS-CoV-2 omicron variant by
vaccine sera and monoclonal antibodies. medRxiv 2021; Journal Website
49. Cathcart AL, Havenar-Daughton C, Lempp FA, Ma D, Schmid MA, Agostini ML, et al. : The dual function monoclonal antibodies
VIR-7831 and VIR-7832 demonstrate potent in vitro and in vivo activity against SARS-CoV-2. bioRxiv 2021; Journal Website
50. ACTIV-3/Therapeutics for Inpatients with COVID-19 (TICO) Study Group : Efficacy and safety of two neutralising monoclonal
antibody therapies, sotrovimab and BRII-196 plus BRII-198, for adults hospitalised with COVID-19 (TICO): a randomised controlled trial.
Lancet Infect Dis 2021;S1473-3099(21):00751-9. Pubmed Journal
51. Gupta A, Gonzalez-Rojas Y, Juarez E, Crespo Casal M, Moya J, Falci DR, et al. : Early treatment for COVID-19 with SARS-CoV-2
103 of 109
neutralizing antibody sotrovimab. New Eng J Med 2021;385(21):1941-1950. Pubmed Journal
52. Sotrovimab for injection. In: COVID-19 vaccines and treatments portal. Ottawa: Health Canada; 2021 (Product monograph;
https://covid-vaccine.canada.ca/info/pdf/sotrovimab-pm-en.pdf, accessed 10 December 2021). Website
53. United States Food and Drug Administration : Fact sheet for healthcare providers - Emergency Use Authorization (EUA) of
sotrovimab. (https://www.fda.gov/media/149534/download, accessed 4 January 2022). Website
54. Corti D, Purcell LA, Snell G, Veesler D : Tackling COVID-19 with neutralizing monoclonal antibodies. Cell 2021;184(12):3086-3108.
Pubmed Journal
55. Kirenga B, Byakika-Kibwika P, Muttamba W, Kayongo A, Loryndah NO, Mugenyi L, et al. : Efficacy of convalescent plasma for
treatment of COVID-19 in Uganda. BMJ Open Respir Res 2021;8(1):e001017. Pubmed Journal
56. Korley FK, Durkalski-Mauldin V, Yeatts SD, Schulman K, Davenport RD, Dumont LJ, et al. : Early convalescent plasma for high-risk
outpatients with COVID-19. New Eng J Med 2021; 1951-1960. Pubmed Journal
57. Libster R, Pérez Marc G, Wappner D, Coviello S, Bianchi A, Braem V, et al. : Early high-titer plasma therapy to prevent severe
COVID-19 in older adults. New Eng J Med 2021;384(7):610-618. Journal Website
58. Avendano-Sola C, Ramos-Martinez A, Munez-Rubio E, Ruiz-Antoran B, Malo de Molina R, Torres F, et al. : Convalescent plasma for
COVID-19: A multicenter, randomized clinical trial (preprint). medRxiv 2020; Journal Website
59. Salman OH, Mohamed HSA : Efficacy and safety of transfusing plasma from COVID-19 survivors to COVID-19 victims with severe
illness. A double-blinded controlled preliminary study. Egypt J Anaesth 2020;36(1):264-272. Journal
60. RECOVERY Collaborative Group : Convalescent plasma in patients admitted to hospital with COVID-19 (RECOVERY): a randomised
controlled, open-label, platform trial. Lancet 2021;397(10289):2049-2059. Pubmed Journal
61. Agarwal A, Mukherjee A, Kumar G, Chatterjee P, Bhatnagar T, Malhotra P : Convalescent plasma in the management of moderate
COVID-19 in adults in India: open label phase II multicentre randomised controlled trial (PLACID Trial). BMJ 2020;371 m3939. Journal
Website
62. Begin P, Callum J, Jamula E, Cook R, Heddle NM, Tinmouth A, et al. : Convalescent plasma for hospitalized patients with COVID-19:
an open-label, randomized controlled trial. Nat Med 2021; 2012-2024. Pubmed Journal Website
63. Simonovich VA, Burgos Pratx LD, Scibona P, Beruto MV, Vallone MG, Vázquez C, et al. : A randomized trial of convalescent plasma in
COVID-19 severe pneumonia. New Eng J Med 2020;384(7):619-629. Journal Website
64. REMAP-CAP Investigators, Estcourt LJ : Convalescent plasma in critically ill patients with COVID-19. medRxiv 2021; Journal
Website
65. Li L, Zhang W, Hu Y, Tong X, Zheng S, Yang J, et al. : Effect of convalescent plasma therapy on time to clinical improvement in
patients with severe and life-threatening COVID-19: a randomized clinical trial. JAMA 2020; 460-470. Pubmed Journal Website
66. Sekine L, Arns B, Fabro BR, Cipolatt MM, Machado RRG, Durigon EL, et al. : Convalescent plasma for COVID-19 in hospitalised
patients: an open-label, randomised clinical trial. Eur Respir J 2021; 2101471. Pubmed Journal
67. Pouladzadeh M, Safdarian M, Eshghi P, Abolghasemi H, Bavani AG, Sheibani B, et al. : A randomized clinical trial evaluating the
immunomodulatory effect of convalescent plasma on COVID-19-related cytokine storm. Intern Emerg Med 2021;16 1-11. Pubmed
Journal
68. Ray Y, Paul SR, Bandopadhyay P, D’Rozario R, Sarif J, Lahiri A, et al. : Clinical and immunological benefits of convalescent plasma
therapy in severe COVID-19: insights from a single center open label randomised control trial. medRxiv 2020; Journal Website
104 of 109
69. AlQahtani M, Abdulrahman A, Almadani A, Alali SY, Al Zamrooni AM, Hejab AH, et al. : Randomized controlled trial of convalescent
plasma therapy against standard therapy in patients with severe COVID-19 disease. Sci Rep 2021;11(1):9927. Pubmed Journal
70. Gharbharan A, Jordans CCE, GeurtsvanKessel C, den Hollander JG, Karim F, Mollema FPN, et al. : Effects of potent neutralizing
antibodies from convalescent plasma in patients hospitalized for severe SARS-CoV-2 infection. Nat Commun 2021;12(1):3189. Pubmed
Journal
71. Casadevall A, Pirofski L-A : The convalescent sera option for containing COVID-19. J Clin Investig 2020;130(4):1545-1548. Pubmed
Journal
72. Haagmans BL, Noack D, Okba NMA, Li W, Wang C, Bestebroer T, et al. : SARS-CoV-2 neutralizing human antibodies protect against
lower respiratory tract disease in a hamster model. J Infect Dis 2021;223(12):2020-2028. Pubmed Journal
73. Sharma R, Sharma S : Physiology, Blood Volume. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing. 2021; Pubmed
74. Lamikanra A, Nguyen D, Simmonds P, Williams S, Bentley EM, Rowe C, et al. : Comparability of six different immunoassays
measuring SARS-CoV-2 antibodies with neutralizing antibody levels in convalescent plasma: From utility to prediction. Transfusion
2021;61(10):2837-2843. Pubmed Journal
75. O'Donnell MR, Grinsztejn B, Cummings MJ, Justman JE, Lamb MR, Eckhardt CM, et al. : A randomized double-blind controlled trial
of convalescent plasma in adults with severe COVID-19. J Clin Investig 2021;131(13):e150646. Pubmed Journal
76. World Health Organization : Enhancing readiness for omicron (B.1.1.529): technical brief and priority actions for member states.
(https://www.who.int/publications/m/item/enhancing-readiness-for-omicron-%28b.1.1.529%29-technical-brief-and-priority-actions-
for-member-states, accessed 7 January 2022). Website
77. United States Food and Drug Administration : Emergency use authorization (EUA) of REGEN-COV TM (casirivimab and imdevimab).
(https://www.fda.gov/media/145611/download, accessed 1 September 2021). Website
78. Weinreich DM, Sivapalasingam S, Norton T, Ali S, Gao H, Bhore R, et al. : REGEN-COV antibody cocktail clinical outcomes study in
COVID-19 outpatients. medRxiv (http://medrxiv.org/content/early/2021/06/06/2021.05.19.21257469.abstract, accessed 1 August
2021). Journal Website
79. O’Brien MP, Forleo-Neto E, Sarkar N, Isa F, Hou P, Chan K-C, et al. : Subcutaneous REGEN-COV antibody combination in early
SARS-CoV-2 infection. medRxiv (https://www.medrxiv.org/content/10.1101/2021.06.14.21258569v1, accessed 13 September 2021).
Journal Website
80. National SARS-CoV-2 Serology Assay Evaluation Group : Performance characteristics of five immunoassays for SARS-CoV-2: a head-
to-head benchmark comparison. Lancet Infect Dis 2020;20(12):1390-1400. Pubmed Journal
81. Peto T, UK COVID-19 Lateral Flow Oversight Team : COVID-19: Rapid antigen detection for SARS-CoV-2 by lateral flow assay: A
national systematic evaluation of sensitivity and specificity for mass-testing. EClinicalMedicine 2021;36 100924. Pubmed Journal
82. Moshe M, Daunt A, Flower B, Simmons B, Brown JC, Frise R, et al. : SARS-CoV-2 lateral flow assays for possible use in national
covid-19 seroprevalence surveys (React 2): diagnostic accuracy study. BMJ 2021;372 n423. Pubmed Journal
83. RECOVERY Collaborative Group : Casirivimab and imdevimab in patients admitted to hospital with COVID-19 (RECOVERY): a
randomised, controlled, open-label, platform trial. medRxiv (https://www.medrxiv.org/content/10.1101/2021.06.15.21258542v1,
accessed 1 August 2021). Journal Website
84. Schandelmaier S, Briel M, Varadhan R, Schmid CH, Devasenapathy N, Hayward RA, et al. : Development of the Instrument to assess
the Credibility of Effect Modification Analyses (ICEMAN) in randomized controlled trials and meta-analyses. CMAJ
2020;192(32):E901-E906. Pubmed Journal
85. Hansen J, Baum A, Pascal KE, Russo V, Giordano S, Wloga E, et al. : Studies in humanized mice and convalescent humans yield a
SARS-CoV-2 antibody cocktail. Science 2020;369(6506):1010-1014. Pubmed Journal
105 of 109
86. Baum A, Ajithdoss D, Copin R, Zhou A, Lanza K, Negron N, et al. : REGN-COV2 antibodies prevent and treat SARS-CoV-2 infection
in rhesus macaques and hamsters. Science 2020;370(6520):1110-1115. Pubmed Journal
87. Ryman JT, Meibohm B : Pharmacokinetics of Monoclonal Antibodies. CPT Pharmacometrics Syst Pharmacol 2017;6(9):576-588.
Pubmed Journal
88. Casirivimab and imdevimab for injection. In: COVID-19 vaccines and treatments portal. Ottawa: Health Canada; 2021 (Product
monograph; https://covid-vaccine.canada.ca/info/pdf/casirivimab-imdevimab-pm-en.pdf, accessed 10 September 2021). Website
89. WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group : Anti-interleukin-6 therapies for hospitalized
patients with COVID-19: a protocol for a prospective meta-analysis of randomized trials. (https://www.who.int/publications/i/item/
WHO-2019-nCoV-PMA_protocols-anti-IL-6-2021.1, accessed 10 June 2021).
90. The WHO Rapid Evidence Appraisal for COVID-19 Therapies [REACT] Working Group : Association of administration of
interleukin-6 antagonists with mortality and other outcomes among hospitalized patients with COVID-19: a prospective meta-analysis.
JAMA 2021;326(6):499-518. Journal Website
91. [WITHDRAWN] Elgazzar A, Hany B, Youssef SA, Hany B, Hafez M, Moussa H : Efficacy and safety of ivermectin for treatment and
prophylaxis of COVID-19 pandemic. Research Square 2021; Journal Website
92. Reardon S : Flawed ivermectin preprint highlights challenges of COVID drug studies. Nature 2021;596(7871):173-174. Pubmed
Journal
93. Kirti R, Roy R, Pattadar C, Raj R, Agarwal N, Biswas B, et al. : Ivermectin as a potential treatment for mild to moderate COVID-19 – a
double blind randomized placebo-controlled trial. medRxiv 2021; Journal Website
94. Niaee MS, Gheibi N, Namdar P, Allami A, Zolghadr L, Javadi A, et al. : Ivermectin as an adjunct treatment for hospitalized adult
COVID-19 patients: a randomized multi-center clinical trial. Research Square 2021; Journal Website
95. Mohan A, Tiwari P, Suri T, et al. : Ivermectin in mild and moderate COVID-19 (RIVET-COV): a randomized, placebo-controlled trial.
Research Square 2021; Journal Website
96. López-Medina E, López P, Hurtado IC, Dávalos DM, Ramirez O, Martínez E, et al. : Effect of ivermectin on time to resolution of
symptoms among adults with mild COVID-19: a randomized clinical trial. JAMA 2021;325(14):1426-1435. Journal Website
97. Beltran-Gonzalez JL, Gonzalez-Gamez M, Mendoza-Enciso EA, Esparza-Maldonado RJ, Hernanez-Palacios D, Duenas-Campos S, et
al. : Efficacy and safety of ivermectin and hydroxychloroquine in patients with severe COVID-19. A randomized controlled trial. medRxiv
2021; Journal Website
98. Abd-Elsalam S, Noor RA, Badawi R, Khalaf M, Esmail ES, Soliman S, et al. : Clinical study evaluating the efficacy of ivermectin in
COVID-19 treatment: A randomized controlled study. J Med Virol 2021;93(10):5833-5838. Journal Website
99. Vallejos J, Zoni R, Bangher M, Villamandos S, Bobadilla A, Plano F, et al. : Ivermectin to prevent hospitalizations in patients with
COVID-19 (IVERCOR-COVID19) a randomized, double-blind, placebo-controlled trial. BMC Infect Dis 2021;21(1):635. Journal Website
100. Guyatt GH, Oxman AD, Kunz R, Brozek J, Alonso-Coello P, Rind D, et al. : GRADE guidelines 6. Rating the quality of evidence -
imprecision. J Clin Epidemiol 2011;64(12):1283-1293. Pubmed Journal
101. Okumuş N, Demirtürk N, Çetinkaya RA, Güner R, Avci IY, Orhan S, et al. : Evaluation of the effectiveness and safety of adding
ivermectin to treatment in severe COVID-19 patients. BMC Infect Dis 2021;21 411. Journal
102. Podder CS, Chowdhury N, Sina MI, Haque WM : Outcome of ivermectin treated mild to moderate COVID-19 cases: a single-
centre, open-label, randomised controlled study. IMC J Med Sci 2020;14(2):11-18. Journal
103. Hashim HA, Maulood MF, Rasheed AM, Fatak DF, Kabah KK, Abdulamir AS : Controlled randomized clinical trial on using
106 of 109
ivermectin with doxycycline for treating COVID-19 patients in Baghdad, Iraq. medRxiv 2020; Journal Website
104. Chowdhury AT, Shahbaz M, Karim MR, Islam J, Dan G, He S : A comparative study on ivermectin-doxycycline and
hydroxychloroquine-azithromycin therapy on COVID-19 patients. EJMO 2021;5(1):63-70. Journal
105. Stromectol® (ivermectin). Package insert [online]. Netherlands: MSD BV, 2009 (https://www.accessdata.fda.gov/drugsatfda_docs/
label/2009/050742s026lbl.pdf, accessed 20 March 2021). Website
106. Jermain B, Hanafin PO, Cao Y, Lifschitz A, Lanusse C, Rao GG : Development of a minimal physiologically-based pharmacokinetic
model to simulate lung exposure in humans following oral administration of ivermectin for COVID-19 drug repurposing. J Pharm Sci
2020;109(12):3574-3578. Pubmed Journal
107. Arshad U, Pertinez H, Box H, Tatham L, Rajoli RKR, Curley P, et al. : Prioritization of anti-SARS-Cov-2 drug repurposing
opportunities based on plasma and target site concentrations derived from their established human pharmacokinetics. Clin Pharmacol
Ther 2020;108(4):775-790. Pubmed Journal
108. Peña-Silva R, Duffull SB, Steer AC, Jaramillo-Rincon SX, Gwee A, Zhu X : Pharmacokinetic considerations on the repurposing of
ivermectin for treatment of COVID-19. Br J Clin Pharmacol 2021;87(3):1589-1590. Pubmed Journal
109. de Melo GD, Lazarini F, Larrous F, Feige F, Kornobis E, Levallois S, et al. : Attenuation of clinical and immunological outcomes during
SARS-CoV-2 infection by ivermectin. EMBO Mol Med 2021;13(8):e14122. Journal
110. Parvez MSA, Karim MA, Hasan M, Jaman J, Karim Z, Tahsin T, et al. : Prediction of potential inhibitors for RNA-dependent RNA
polymerase of SARS-CoV-2 using comprehensive drug repurposing and molecular docking approach. Int J Biol Macromol 2020;163
1787-1797. Pubmed Journal
111. Mody V, Ho J, Wills S, Mawri A, Lawson L, Ebert MCCJC, et al. : Identification of 3-chymotrypsin like protease (3CLPro) inhibitors
as potential anti-SARS-CoV-2 agents. Commun Biol 2021;4(1):93. Pubmed Journal
112. Arouche TDS, Martins AY, Ramalho TDC, Júnior RNC, Costa FLP, Filho TSDA, et al. : Molecular docking of azithromycin, ritonavir,
lopinavir, oseltamivir, ivermectin and heparin interacting with coronavirus disease 2019 main and severe acute respiratory syndrome
coronavirus-2 3C-like proteases. J Nanosci Nanotechnol 2021;21(4):2075-2089. Pubmed Journal
113. Kalhor H, Sadeghi S, Abolhasani H, Kalhor R, Rahimi H : Repurposing of the approved small molecule drugs in order to inhibit
SARS-CoV-2 S protein and human ACE2 interaction through virtual screening approaches. J Biomol Struct Dyn 2020; 1-16. Pubmed
Journal
114. Lehrer S, Rheinstein PH : Ivermectin docks to the SARS-CoV-2 spike receptor-binding domain attached to ACE2. In Vivo
34(5):3023-3026. Pubmed Journal
115. Zhang X, Song Y, Ci X, An N, Ju Y, Li H, et al. : Ivermectin inhibits LPS-induced production of inflammatory cytokines and improves
LPS-induced survival in mice. Inflamm Res 2008;57(11):524-529. Pubmed Journal
116. Ventre E, Rozières A, Lenief V, Albert F, Rossio P, Laoubi L, et al. : Topical ivermectin improves allergic skin inflammation. Allergy
2017;72(8):1212-1221. Pubmed Journal
117. Yan S, Ci X, Chen NA, Chen C, Li X, Chu X, et al. : Anti-inflammatory effects of ivermectin in mouse model of allergic asthma.
Inflamm Res 2011;60(6):589-596. Pubmed Journal
118. Krause RM, Buisson B, Bertrand S, Corringer PJ, Galzi JL, Changeux JP, et al. : Ivermectin: a positive allosteric effector of the alpha7
neuronal nicotinic acetylcholine receptor. Mol Pharmacol 1998;53(2):283-294. Pubmed
119. Wang Y, Zhang D, Du G, Du R, Zhao J, Jin Y, et al. : Remdesivir in adults with severe COVID-19: a randomised, double-blind,
placebo-controlled, multicentre trial. Lancet 2020;395(10236):1569-1578. Pubmed Journal
107 of 109
120. Beigel JH, Tomashek KM, Dodd LE, Mehta AK, Zingman BS, Kalil AC, et al. : Remdesivir for the Treatment of Covid-19 - Final
Report. New Eng J Med 2020;383: 1813-1826. Pubmed Journal
121. Spinner CD, Gottlieb RL, Criner GJ, Arribas López JR, Cattelan AM, Soriano Viladomiu A, et al. : Effect of remdesivir vs standard
care on clinical status at 11 days in patients with moderate COVID-19: a randomized clinical trial. JAMA 2020;324(11):1048-1057.
Pubmed Journal
122. Rochwerg B, Agarwal A, Zeng L, Leo Y-S, Appiah JA, Agoritsas T, et al. : Remdesivir for severe covid-19: a clinical practice guideline.
BMJ 2020;370 m2924. Pubmed Journal
123. COVID-19 Treatment Guidelines Panel : Coronavirus disease 2019 (COVID-19) treatment guidelines. National Institutes of Health;
2020 (https://www.covid19treatmentguidelines.nih.gov/, accessed 01 July 2021). Website
124. Docherty AB, Harrison EM, Green CA, Hardwick HE, Pius R, Norman L, et al. : Features of 20 133 UK patients in hospital with
COVID-19 using the ISARIC WHO Clinical Characterisation Protocol: prospective observational cohort study. BMJ 2020;369: m1985.
Pubmed Journal
125. Ye Z, Wang Y, Colunga-Lozano LE, Prasad M, Tangamornsuksan W, Rochwerg B, et al. : Efficacy and safety of corticosteroids in
COVID-19 based on evidence for COVID-19, other coronavirus infections, influenza, community-acquired pneumonia and acute
respiratory distress syndrome: a systematic review and meta-analysis. CMAJ 2020;192(27):E756-E767. Pubmed Journal
126. Rochwerg B, Oczkowski SJ, Siemieniuk RAC, Agoritsas T, Belley-Cote E, D'Aragon F, et al. : Corticosteroids in sepsis: an updated
systematic review and meta-analysis. Crit Care Med 2018;46(9):1411-1420. Pubmed Journal
127. World Health Organization : Q&A: Dexamethasone and COVID-19. (https://www.who.int/news-room/questions-and-answers/

item/coronavirus-disease-covid-19-dexamethasone, accessed 20 February 2021). Website
128. Persaud N, Jiang M, Shaikh R, Bali A, Oronsaye E, Woods H, et al. : Comparison of essential medicines lists in 137 countries. Bull
World Health Org 2019;97(6):394-404. Pubmed Journal
129. The WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT)Working Group : Association between administration of
systemic corticosteroids and mortality among critically ill patients with COVID-19: a meta-analysis. JAMA 2020;324(13):1330-1341.
Pubmed Journal
130. Corral-Gudino L, Bahamonde A, Arnaiz-Revillas F, Gómez-Barquero J, Abadía-Otero J, García-Ibarbia C, et al. : Methylprednisolone

in adults hospitalized with COVID-19 pneumonia: An open-label randomized trial (GLUCOCOVID). Wien Klin Wochenschr
2021;133(7-8):303-311. Journal
131. Efficacy of dexamethasone treatment for patients with ARDS caused by COVID-19 (DEXA-COVID19). ClinicalTrials.gov [Internet].
Bethesda (MD): National Library of Medicine (US). 2000 Feb 29. Identifier NCT04325061. (https://clinicaltrials.gov/ct2/show/
NCT04325061, accessed 31 August 2020). Website
132. Tomazini BM, Maia IS, Cavalcanti AB, Berwanger O, Rosa RG, Veiga VC, et al. : Effect of dexamethasone on days alive and
ventilator-free in patients with moderate or severe acute respiratory distress syndrome and COVID-19: the CoDEX randomized clinical
trial. JAMA 2020;324(13):1307-1316. Pubmed Journal
133. Dequin P-F, Heming N, Meziani F, Plantefève G, Voiriot G, Badié J, et al. : Effect of hydrocortisone on 21-day mortality or
respiratory support among critically ill patients with COVID-19: a randomized clinical trial. JAMA 2020;324(13):1298-1306. Pubmed
Journal
134. Glucocorticoid therapy for COVID-19 critically ill patients with severe acute respiratory failure (Steroids-SARI). ClinicalTrials.gov
[Internet]. Bethesda (MD): National Library of Medicine (US). 2000 Feb 29. Identifier NCT04244591. (https://www.clinicaltrials.gov/ct2/
show/NCT04244591, accessed 31 August 2020).
135. Jeronimo CMP, Farias MEL, Val FFA, Sampaio VS, Alexandre MAA, Melo GC, et al. : Methylprednisolone as adjunctive therapy for
patients hospitalized with COVID-19 (Metcovid): a randomised, double-blind, phase IIb, placebo-controlled trial. Clin Infect Dis
2020;72(9):e373-e381. Pubmed Journal
108 of 109
136. Handbook for guideline development. Geneva: World Health Organization; 2008 (https://www.who.int/publications/guidelines/
handbook_2nd_ed.pdf?ua=1, accessed 18 February 2021).
137. Qaseem A, Forland F, Macbeth F, Ollenschläger G, Phillips S, van der Wees P, et al. : Guidelines International Network: toward
international standards for clinical practice guidelines. Ann Int Med 2012;156(7):525-531. Pubmed Journal
138. Vandvik PO, Brandt L, Alonso-Coello P, Treweek S, Akl EA, Kristiansen A, et al. : Creating clinical practice guidelines we can trust,
use, and share: a new era is imminent. Chest 2013;144(2):381-389. Pubmed Journal
139. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. : GRADE: an emerging consensus on rating quality of
evidence and strength of recommendations. BMJ 2008;336(7650):924-926. Pubmed Journal
140. Guyatt GH, Oxman AD, Kunz R, Falck-Ytter Y, Vist GE, Liberati A, et al. : Going from evidence to recommendations. BMJ
2008;336(7652):1049-1051. Pubmed Journal
141. Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. : GRADE guidelines: 1. Introduction-GRADE evidence profiles and
summary of findings tables. J Clin Epidemiol 2011;64(4):383-394. Pubmed Journal
142. Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J, et al. : GRADE guidelines: 3. Rating the quality of evidence.
J Clin Epidemiol 2011;64(4):401-406. Pubmed Journal
143. Andrews JC, Schünemann HJ, Oxman AD, Pottie K, Meerpohl JJ, Coello PA, et al. : GRADE guidelines: 15. Going from evidence to
recommendation-determinants of a recommendation's direction and strength. J Clin Epidemiol 2013;66(7):726-735. Pubmed Journal
109 of 109

WHO 2019 NCoV Therapeutics 2022.2 Eng

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

WHO 2019 NCoV Therapeutics 2022.2 Eng

Uploaded by

Copyright:

Available Formats

Therapeutics and COVID-19

WHO reference number: WHO/2019-nCoV/therapeutics/2022.2

1. Summary of the guideline .................................................................................................................................................................................................... 4

4. What triggered this update and what is coming next? .................................................................................................................................................. 8

5. Understanding and applying the WHO severity definitions ........................................................................................................................................ 9

6. Recommendations for therapeutics .................................................................................................................................................................................10

6.1 Molnupiravir (published 3 March 2022) ..............................................................................................................................................................10

6.1.1 Mechanism of action ...................................................................................................................................................................................15

6.2 Janus kinase inhibitors (published 14 January 2022) .......................................................................................................................................16

6.2.1 Mechanism of action ...................................................................................................................................................................................27

6.3 Sotrovimab (published 14 January 2022) ............................................................................................................................................................27

6.3.1 Mechanism of action ...................................................................................................................................................................................33

6.4 Convalescent plasma (published 7 December 2021)........................................................................................................................................34

6.4.1 Mechanism of action ...................................................................................................................................................................................42

6.5 Casirivimab-imdevimab (neutralizing monoclonal antibodies) (published 24 September 2021) ...........................................................42

6.5.1 Mechanism of action ...................................................................................................................................................................................52

6.6 IL-6 receptor blockers (published 6 July 2021) ..................................................................................................................................................53

6.6.1 Mechanism of action ...................................................................................................................................................................................60

6.7 Ivermectin (published 31 March 2021) ................................................................................................................................................................60

6.7.1 Mechanism of action ...................................................................................................................................................................................65

6.8 Hydroxychloroquine (published 17 December 2020) ......................................................................................................................................65

6.9 Lopinavir/ritonavir (published 17 December 2020)..........................................................................................................................................70

6.10 Remdesivir (published 20 November 2020) .....................................................................................................................................................74

6.10.1 Mechanism of action .................................................................................................................................................................................78

6.11 Systemic corticosteroids (published 2 September 2020) ..............................................................................................................................79

7. Methods: how this guideline was created ......................................................................................................................................................................88

8. How to access and use this guideline ..............................................................................................................................................................................92

9. Uncertainties, emerging evidence and future research ...............................................................................................................................................94

10. Authorship, contributions, acknowledgements ..........................................................................................................................................................97

References ............................................................................................................................................................................................................................... 101

1. Summary of the guideline

Updates to prior recommendations:

Recommended for patients with severe or critical COVID-19:

• a strong recommendation for systemic corticosteroids;

Recommended for patients with non-severe COVID-19:

Not recommended for patients with non-severe COVID-19:

• a conditional recommendation against systemic corticosteroids;

Not recommended for patients with severe and critical COVID-19:

• a recommendation against convalescent plasma, except in the context of a clinical trial;

Not recommended, regardless of COVID-19 disease severity:

• a strong recommendation against hydroxychloroquine;

ALT alanine aminotransferase

ARDS acute respiratory distress syndrome

CAP community-acquired pneumonia

COVID-19 coronavirus disease 2019

DOI declaration of interests

eGFR estimated glomerular filtration rate

FDA United States Food and Drug Administration

GDG guideline development group

GRADE Grading of Recommendations Assessment, Development and Evaluation

GRC guideline review committee

IMV invasive mechanical ventilation

JAK Janus kinase

LNMA living network meta-analysis

MAGIC Magic Evidence Ecosystem Foundation

OIS optimal information size

PICO population, intervention, comparator, outcome

PMA prospective meta-analysis

RCT randomized controlled trial

RR relative risk/risk ratio

SAE serious adverse event

TACO transfusion-associated circulatory overload

TRALI transfusion-related acute lung injury

WHO World Health Organization