Guoqiao Wang1, Gary R Cutter2, Stacey S Cofield2, Fred Lublin3, Jerry S Wolinsky4, Tarah Gustafson3, Stephen Krieger3, Amber Salter1. 1. Division of Biostatistics, Washington University in St. Louis, St. Louis, MO, USA. 2. Department of Biostatistics, The University of Alabama at Birmingham, Birmingham, AL, USA. 3. Icahn School of Medicine at Mount Sinai, New York City, NY, USA. 4. McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX, USA.
Abstract
BACKGROUND: In randomized clinical trials, when treatments do not work equally effectively across stratifications of participants, observed event rates may differ from those hypothesized leading to deviations in estimated power. OBJECTIVES: To investigate the effect of distributions of baseline Expanded Disability Status Scale (EDSS) proportions in relapsing-remitting multiple sclerosis (RRMS) on the trial outcome, confirmed disability progression rate (CDPR), and power. METHODS: We reported CDPRs in the CombiRx trial by baseline EDSS and by groups (1st (0, 1), 2nd (1.5, 2), 3rd (2.5, 3), and 4th (⩾3.5)) and investigated the effect of different combinations of baseline EDSS proportions on the trial outcome and power. RESULTS: There were 244 (25.4%) participants in the 1st group, 368 (38.4%) in the 2nd group, 223 (23.3%) in the 3rd group, and 124 (12.9%) in the 4th group with CDPRs of 40.1%, 13.9%, 11.2%, and 16.9%, respectively. Both CDPR and power increased when the proportion of the 1st group increased in hypothetical trials with equal sample sizes in each arm, and a 10% increase in the 1st group led to a 5% increase in power. CONCLUSION: Various baseline EDSS proportions yielded different CDPRs and power, suggesting caution in interpretation of treatment effects across trials that enrolled participants with different proportions of baseline EDSS.
RCT Entities:
BACKGROUND: In randomized clinical trials, when treatments do not work equally effectively across stratifications of participants, observed event rates may differ from those hypothesized leading to deviations in estimated power. OBJECTIVES: To investigate the effect of distributions of baseline Expanded Disability Status Scale (EDSS) proportions in relapsing-remitting multiple sclerosis (RRMS) on the trial outcome, confirmed disability progression rate (CDPR), and power. METHODS: We reported CDPRs in the CombiRx trial by baseline EDSS and by groups (1st (0, 1), 2nd (1.5, 2), 3rd (2.5, 3), and 4th (⩾3.5)) and investigated the effect of different combinations of baseline EDSS proportions on the trial outcome and power. RESULTS: There were 244 (25.4%) participants in the 1st group, 368 (38.4%) in the 2nd group, 223 (23.3%) in the 3rd group, and 124 (12.9%) in the 4th group with CDPRs of 40.1%, 13.9%, 11.2%, and 16.9%, respectively. Both CDPR and power increased when the proportion of the 1st group increased in hypothetical trials with equal sample sizes in each arm, and a 10% increase in the 1st group led to a 5% increase in power. CONCLUSION: Various baseline EDSS proportions yielded different CDPRs and power, suggesting caution in interpretation of treatment effects across trials that enrolled participants with different proportions of baseline EDSS.
Authors: Giancarlo Comi; Douglas Jeffery; Ludwig Kappos; Xavier Montalban; Alexey Boyko; Maria A Rocca; Massimo Filippi Journal: N Engl J Med Date: 2012-03-15 Impact factor: 91.245
Authors: J W Lindsey; T F Scott; S G Lynch; S S Cofield; F Nelson; R Conwit; T Gustafson; G R Cutter; J S Wolinsky; F D Lublin Journal: Mult Scler Relat Disord Date: 2012-02-23 Impact factor: 4.339
Authors: Paul O'Connor; Jerry S Wolinsky; Christian Confavreux; Giancarlo Comi; Ludwig Kappos; Tomas P Olsson; Hadj Benzerdjeb; Philippe Truffinet; Lin Wang; Aaron Miller; Mark S Freedman Journal: N Engl J Med Date: 2011-10-06 Impact factor: 91.245
Authors: Fred D Lublin; Stacey S Cofield; Gary R Cutter; Robin Conwit; Ponnada A Narayana; Flavia Nelson; Amber R Salter; Tarah Gustafson; Jerry S Wolinsky Journal: Ann Neurol Date: 2013-03-11 Impact factor: 10.422
Authors: Jeffrey A Cohen; Alasdair J Coles; Douglas L Arnold; Christian Confavreux; Edward J Fox; Hans-Peter Hartung; Eva Havrdova; Krzysztof W Selmaj; Howard L Weiner; Elizabeth Fisher; Vesna V Brinar; Gavin Giovannoni; Miroslav Stojanovic; Bella I Ertik; Stephen L Lake; David H Margolin; Michael A Panzara; D Alastair S Compston Journal: Lancet Date: 2012-11-01 Impact factor: 79.321
Authors: Ludwig Kappos; Mark S Freedman; Chris H Polman; Gilles Edan; Hans-Peter Hartung; David H Miller; Xavier Montalbán; Frederik Barkhof; Ernst-Wilhelm Radü; Carola Metzig; Lars Bauer; Vivian Lanius; Rupert Sandbrink; Christoph Pohl Journal: Lancet Neurol Date: 2009-09-10 Impact factor: 44.182
Authors: Paul O'Connor; Massimo Filippi; Barry Arnason; Giancarlo Comi; Stuart Cook; Douglas Goodin; Hans-Peter Hartung; Douglas Jeffery; Ludwig Kappos; Francis Boateng; Vitali Filippov; Maria Groth; Volker Knappertz; Christian Kraus; Rupert Sandbrink; Christoph Pohl; Timon Bogumil; P O'Connor; M Filippi; B Arnason; S Cook; D Goodin; H-P Hartung; H-P Harung; L Kappos; D Jeffery; G Comi Journal: Lancet Neurol Date: 2009-09-02 Impact factor: 44.182