Therapeutics for COVID-19: World Health Organization guidelines as a model for partnered living evidence and guidance

Article type
Authors
Agarwal A1, Agoritsas T2, Diaz J3, Guyatt G4, Rylance J3, Vandvik P5
1MAGIC Evidence Ecosystem Foundation, Oslo, Norway; Department of Health Research Methods, Evidence and Impact, McMaster University, Hamilton, Ontario, Canada
2MAGIC Evidence Ecosystem Foundation, Oslo, Norway; Division of General Internal Medicine and Division of Clinical Epidemiology, University Hospitals of Geneva, Geneva, Switzerland
3World Health Organization, Geneva, Switzerland
4Department of Health Research Methods, Evidence and Impact, McMaster University, Hamilton, Ontario, Canada; Department of Medicine, McMaster University, Hamilton, Ontario, Canada
5MAGIC Evidence Ecosystem Foundation, Oslo, Norway; Department of Medicine, Lovisenberg Diaconal Hospital Trust, Oslo, Norway
Abstract
Background: Evolution in SARS-CoV-2 variants, paired with new clinical trials, has necessitated updated evidence and guidance regarding COVID-19 pharmacotherapies. The World Health Organization (WHO) has responded, leveraging living network meta-analyses (NMAs) to provide iteratively updated practice guidance.

Objectives: To summarize successes and challenges associated with WHO’s living COVID-19 practice guidance

Methods: The living guideline receives methodological support from the MAGIC Evidence Ecosystem Foundation and is informed by living NMAs of randomized trials addressing the effectiveness of COVID-19 pharmacotherapies for patients with nonsevere and severe or critical COVID-19. The guideline adheres to trustworthiness standards.

WHO convened an unconflicted Guideline Development Group (GDG) panel including clinicians, methodologists, content experts, and patient partners. WHO severity definitions facilitated risk-stratified recommendations. Recommendations for nonsevere illness were stratified based on risk of hospitalization as determined by individual patient factors. When moving from evidence to recommendations, the GDG considered a combination of evidence regarding benefits and harms, values and preferences, practical issues, resources, feasibility, and equity. Given lack of empirical data regarding patient values, a key innovation was formal polling of panel members regarding their views of effects that patients feel are important.

Results: Fourteen versions of the living guideline incorporating recommendations for 15 therapeutics have been cross-published across MAGICapp, the WHO website, and The BMJ. For severe or critical illness, corticosteroids, interleukin-6 receptor blockers, and baricitinib are strongly recommended; for severe illness only, remdesivir is conditionally recommended. For nonsevere illness, the GDG strongly recommended nirmatrelvir-ritonavir in patients at high risk of hospitalization and conditionally recommended use in those at moderate risk; the GDG conditionally recommended remdesivir and molnupiravir for patients at high risk. Across iterations, updated evidence led to updated recommendations for nirmatrelvir-ritonavir, remdesivir, molnupiravir, casirivimab-imdevimab, sotrovimab, VV116, ivermectin, and bariticinib and to updated baseline risks for all-cause death and hospital admission.

Ongoing challenges include the absence of accurate prognostic models to estimate risk of hospitalization for patients with nonsevere illness and a paucity of evidence regarding patient preferences.

Conclusions: The WHO living COVID-19 guideline has successfully leveraged living evidence to produce up-to-date trustworthy guidance. Ongoing challenges include limited evidence to inform prognosis and patient preferences.