Lori E Dodd1, Dean Follmann1, Jing Wang2, Franz Koenig3, Lisa L Korn4, Christian Schoergenhofer5, Michael Proschan1, Sally Hunsberger1, Tyler Bonnett2, Mat Makowski6, Drifa Belhadi7,8, Yeming Wang9,10, Bin Cao9,10, France Mentre7,8, Thomas Jaki11,12. 1. Biostatistics Research Branch, National Institute Allergy and Infectious Diseases, Bethesda, MD, USA. 2. Clinical Monitoring Research Program Directorate, Frederick National Laboratory for Cancer Research, Frederick, MD, USA. 3. Center for Medical Statistics, Informatics and Intelligent Systems; Medical University of Vienna, Vienna, Austria. 4. Department of Medicine (Rheumatology, Allergy, and Immunology Section) and Department of Immunobiology, Yale University, New Haven, CT, USA. 5. Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria. 6. The Emmes Company, LLC, Rockville, MD, USA. 7. Université de Paris, IAME, Inserm, Paris, France. 8. AP-HP, Hôpital Bichat, DEBRC, Paris, France. 9. Center of Respiratory Medicine, Department of Pulmonary and Critical Care Medicine, National Clinical Research Center for Respiratory Diseases, Beijing, China. 10. China-Japan Friendship Hospital, Department of Respiratory Medicine, Capital Medical University, Beijing, China. 11. Department of Mathematics and Statistics, Lancaster University, Lancaster, UK. 12. MRC Biostatistics Unit, University of Cambridge, Cambridge, UK.
Abstract
BACKGROUND: Endpoint choice for randomized controlled trials of treatments for novel coronavirus-induced disease (COVID-19) is complex. Trials must start rapidly to identify treatments that can be used as part of the outbreak response, in the midst of considerable uncertainty and limited information. COVID-19 presentation is heterogeneous, ranging from mild disease that improves within days to critical disease that can last weeks to over a month and can end in death. While improvement in mortality would provide unquestionable evidence about the clinical significance of a treatment, sample sizes for a study evaluating mortality are large and may be impractical, particularly given a multitude of putative therapies to evaluate. Furthermore, patient states in between "cure" and "death" represent meaningful distinctions. Clinical severity scores have been proposed as an alternative. However, the appropriate summary measure for severity scores has been the subject of debate, particularly given the variable time course of COVID-19. Outcomes measured at fixed time points, such as a comparison of severity scores between treatment and control at day 14, may risk missing the time of clinical benefit. An endpoint such as time to improvement (or recovery) avoids the timing problem. However, some have argued that power losses will result from reducing the ordinal scale to a binary state of "recovered" versus "not recovered." METHODS: We evaluate statistical power for possible trial endpoints for COVID-19 treatment trials using simulation models and data from two recent COVID-19 treatment trials. RESULTS: Power for fixed time-point methods depends heavily on the time selected for evaluation. Time-to-event approaches have reasonable statistical power, even when compared with a fixed time-point method evaluated at the optimal time. DISCUSSION: Time-to-event analysis methods have advantages in the COVID-19 setting, unless the optimal time for evaluating treatment effect is known in advance. Even when the optimal time is known, a time-to-event approach may increase power for interim analyses.
BACKGROUND: Endpoint choice for randomized controlled trials of treatments for novel coronavirus-induced disease (COVID-19) is complex. Trials must start rapidly to identify treatments that can be used as part of the outbreak response, in the midst of considerable uncertainty and limited information. COVID-19 presentation is heterogeneous, ranging from mild disease that improves within days to critical disease that can last weeks to over a month and can end in death. While improvement in mortality would provide unquestionable evidence about the clinical significance of a treatment, sample sizes for a study evaluating mortality are large and may be impractical, particularly given a multitude of putative therapies to evaluate. Furthermore, patient states in between "cure" and "death" represent meaningful distinctions. Clinical severity scores have been proposed as an alternative. However, the appropriate summary measure for severity scores has been the subject of debate, particularly given the variable time course of COVID-19. Outcomes measured at fixed time points, such as a comparison of severity scores between treatment and control at day 14, may risk missing the time of clinical benefit. An endpoint such as time to improvement (or recovery) avoids the timing problem. However, some have argued that power losses will result from reducing the ordinal scale to a binary state of "recovered" versus "not recovered." METHODS: We evaluate statistical power for possible trial endpoints for COVID-19 treatment trials using simulation models and data from two recent COVID-19 treatment trials. RESULTS: Power for fixed time-point methods depends heavily on the time selected for evaluation. Time-to-event approaches have reasonable statistical power, even when compared with a fixed time-point method evaluated at the optimal time. DISCUSSION: Time-to-event analysis methods have advantages in the COVID-19 setting, unless the optimal time for evaluating treatment effect is known in advance. Even when the optimal time is known, a time-to-event approach may increase power for interim analyses.
Entities:
Keywords:
COVID-19; WHO ordinal scale; censoring; clinical trials; endpoints; log-rank test; proportional odds model
Authors: Katharine J Bar; Pamela A Shaw; Grace H Choi; Nicole Aqui; Andrew Fesnak; Jasper B Yang; Haideliza Soto-Calderon; Lizette Grajales; Julie Starr; Michelle Andronov; Miranda Mastellone; Chigozie Amonu; Geoff Feret; Maureen DeMarshall; Marie Buchanan; Maria Caturla; James Gordon; Alan Wanicur; M Alexandra Monroy; Felicity Mampe; Emily Lindemuth; Sigrid Gouma; Anne M Mullin; Holly Barilla; Anastasiya Pronina; Leah Irwin; Raeann Thomas; Risa A Eichinger; Faye Demuth; Eline T Luning Prak; Jose L Pascual; William R Short; Michal A Elovitz; Jillian Baron; Nuala J Meyer; Kathleen O Degnan; Ian Frank; Scott E Hensley; Donald L Siegel; Pablo Tebas Journal: J Clin Invest Date: 2021-12-15 Impact factor: 14.808
Authors: Daniel D Murray; Abdel G Babiker; Jason V Baker; Christina E Barkauskas; Samuel M Brown; Christina C Chang; Victoria J Davey; Annetine C Gelijns; Adit A Ginde; Birgit Grund; Elizabeth Higgs; Fleur Hudson; Virginia L Kan; H Clifford Lane; Thomas A Murray; Roger Paredes; Mahesh Kb Parmar; Sarah Pett; Andrew N Phillips; Mark N Polizzotto; Cavan Reilly; Uriel Sandkovsky; Shweta Sharma; Marc Teitelbaum; B Taylor Thompson; Barnaby E Young; James D Neaton; Jens D Lundgren Journal: Clin Trials Date: 2021-10-10 Impact factor: 2.486
Authors: Youngji Jo; Lise Jamieson; Ijeoma Edoka; Lawrence Long; Sheetal Silal; Juliet R C Pulliam; Harry Moultrie; Ian Sanne; Gesine Meyer-Rath; Brooke E Nichols Journal: Open Forum Infect Dis Date: 2021-01-29 Impact factor: 3.835
Authors: Elias Laurin Meyer; Peter Mesenbrink; Tobias Mielke; Tom Parke; Daniel Evans; Franz König Journal: Trials Date: 2021-03-04 Impact factor: 2.279
Authors: Subham Das; Anu Kunnath Ramachandran; Sumit Raosaheb Birangal; Saleem Akbar; Bahar Ahmed; Alex Joseph Journal: Med Drug Discov Date: 2021-03-10
Authors: Nigel Stallard; Lisa Hampson; Norbert Benda; Werner Brannath; Thomas Burnett; Tim Friede; Peter K Kimani; Franz Koenig; Johannes Krisam; Pavel Mozgunov; Martin Posch; James Wason; Gernot Wassmer; John Whitehead; S Faye Williamson; Sarah Zohar; Thomas Jaki Journal: Stat Biopharm Res Date: 2020-07-29 Impact factor: 1.452