Peter A G Sandercock1, Janet Darbyshire2, David DeMets3, Robert Fowler4, David G Lalloo5, Mohammed Munavvar6, Natalie Staplin7, Adilia Warris8, Janet Wittes9, Jonathan R Emberson10. 1. Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, UK. 2. University College London, London, UK. 3. School of Medicine and Public Health, University of Wisconsin-Madison, Madison, USA. 4. Department of Medicine and Interdepartmental Division of Critical Care Medicine, University of Toronto, Toronto, Canada. 5. Liverpool School of Tropical Medicine, Liverpool, UK. 6. Lancashire Teaching Hospitals and University of Central Lancashire, Preston, UK. 7. Medical Research Council Population Health Research Unit, Clinical Trial Service Unit and Epidemiological Studies Unit, Nuffield Department of Population Health, University of Oxford, Oxford, UK. 8. MRC Centre for Medical Mycology, University of Exeter, Exeter, UK. 9. Wittes LLC, Washington D.C., USA. 10. Medical Research Council Population Health Research Unit, Clinical Trial Service Unit and Epidemiological Studies Unit, Nuffield Department of Population Health, University of Oxford, Oxford, UK. jonathan.emberson@ndph.ox.ac.uk.
Abstract
AIM: To inform the oversight of future clinical trials during a pandemic, we summarise the experiences of the Data Monitoring Committee (DMC) for the Randomised Evaluation of COVID therapy trial (RECOVERY), a large-scale randomised adaptive platform clinical trial of treatments for hospitalised patients with COVID-19. METHODS AND FINDINGS: During the first 24 months of the trial (March 2020 to February 2022), the DMC oversaw accumulating data for 14 treatments in adults (plus 10 in children) involving > 45,000 randomised patients. Five trial aspects key for the DMC in performing its role were: a large committee of members, including some with extensive DMC experience and others who had broad clinical expertise; clear strategic planning, communication, and responsiveness by the trial principal investigators; data collection and analysis systems able to cope with phases of very rapid recruitment and link to electronic health records; an ability to work constructively with regulators (and other DMCs) to address emerging concerns without the need to release unblinded mortality results; and the use of videoconferencing systems that enabled national and international members to meet at short notice and from home during the pandemic when physical meetings were impossible. Challenges included that the first four treatments introduced were effectively 'competing' for patients (increasing pressure to make rapid decisions on each one); balancing the global health imperative to report on findings with the need to maintain confidentiality until the results were sufficiently certain to appropriately inform treatment decisions; and reliably assessing safety, especially for newer agents introduced after the initial wave and in the small numbers of pregnant women and children included. We present a series of case vignettes to illustrate some of the issues and the DMC decision-making related to hydroxychloroquine, dexamethasone, casirivimab + imdevimab, and tocilizumab. CONCLUSIONS: RECOVERY's streamlined adaptive platform design, linked to hospital-level and population-level health data, enabled the rapid and reliable assessment of multiple treatments for hospitalised patients with COVID-19. The later introduction of factorial assessments increased the trial's efficiency, without compromising the DMC's ability to assess safety and efficacy. Requests for the release of unblinded primary outcome data to regulators at points when data were not mature required significant efforts in communication with the regulators by the DMC to avoid inappropriate early trial termination.
AIM: To inform the oversight of future clinical trials during a pandemic, we summarise the experiences of the Data Monitoring Committee (DMC) for the Randomised Evaluation of COVID therapy trial (RECOVERY), a large-scale randomised adaptive platform clinical trial of treatments for hospitalised patients with COVID-19. METHODS AND FINDINGS: During the first 24 months of the trial (March 2020 to February 2022), the DMC oversaw accumulating data for 14 treatments in adults (plus 10 in children) involving > 45,000 randomised patients. Five trial aspects key for the DMC in performing its role were: a large committee of members, including some with extensive DMC experience and others who had broad clinical expertise; clear strategic planning, communication, and responsiveness by the trial principal investigators; data collection and analysis systems able to cope with phases of very rapid recruitment and link to electronic health records; an ability to work constructively with regulators (and other DMCs) to address emerging concerns without the need to release unblinded mortality results; and the use of videoconferencing systems that enabled national and international members to meet at short notice and from home during the pandemic when physical meetings were impossible. Challenges included that the first four treatments introduced were effectively 'competing' for patients (increasing pressure to make rapid decisions on each one); balancing the global health imperative to report on findings with the need to maintain confidentiality until the results were sufficiently certain to appropriately inform treatment decisions; and reliably assessing safety, especially for newer agents introduced after the initial wave and in the small numbers of pregnant women and children included. We present a series of case vignettes to illustrate some of the issues and the DMC decision-making related to hydroxychloroquine, dexamethasone, casirivimab + imdevimab, and tocilizumab. CONCLUSIONS: RECOVERY's streamlined adaptive platform design, linked to hospital-level and population-level health data, enabled the rapid and reliable assessment of multiple treatments for hospitalised patients with COVID-19. The later introduction of factorial assessments increased the trial's efficiency, without compromising the DMC's ability to assess safety and efficacy. Requests for the release of unblinded primary outcome data to regulators at points when data were not mature required significant efforts in communication with the regulators by the DMC to avoid inappropriate early trial termination.
Authors: Claire Snowdon; Peter Brocklehurst; Robert C Tasker; Martin Ward Platt; Diana Elbourne Journal: PLoS One Date: 2018-07-26 Impact factor: 3.240
Authors: Peter Horby; Wei Shen Lim; Jonathan R Emberson; Marion Mafham; Jennifer L Bell; Louise Linsell; Natalie Staplin; Christopher Brightling; Andrew Ustianowski; Einas Elmahi; Benjamin Prudon; Christopher Green; Timothy Felton; David Chadwick; Kanchan Rege; Christopher Fegan; Lucy C Chappell; Saul N Faust; Thomas Jaki; Katie Jeffery; Alan Montgomery; Kathryn Rowan; Edmund Juszczak; J Kenneth Baillie; Richard Haynes; Martin J Landray Journal: N Engl J Med Date: 2020-07-17 Impact factor: 91.245