Get access

Comparison of analysis approaches for phase III clinical trials in amyotrophic lateral sclerosis

Authors

  • Brian C. Healy PhD,

    Corresponding author
    1. Biostatistics Center, Massachusetts General Hospital, 50 Staniford Street, Suite 560, Boston, Massachusetts 02114, USA
    2. Partners MS Center, Brigham and Women's Hospital, Brookline Massachusetts, USA
    • Biostatistics Center, Massachusetts General Hospital, 55 Staniford Street, Boston, Massachusetts 02114, USA
    Search for more papers by this author
  • David Schoenfeld PhD

    1. Biostatistics Center, Massachusetts General Hospital, 50 Staniford Street, Suite 560, Boston, Massachusetts 02114, USA
    2. Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts, USA
    Search for more papers by this author

Abstract

Introduction: In this study we explore several methods for incorporating survival information in the analysis of Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS) scores. Methods: ALSFRS scores and patient survival times were simulated based on estimates from a recent clinical trial. Six analysis approaches were applied to the data. Each approach was based on ALSFRS scores, the survival time, or a combination of the 2. The power of each approach to detect potential treatment effects was estimated. Results: When the treatment acted solely on the change in ALSFRS, the shared parameter model provided the most power, although all of the models based on random effects were similar. As the effect on survival increased, rank-based analysis showed potential gains in power. Survival analysis was superior under a small effect on ALSFRS and a larger effect on mortality. Conclusions: The shared parameter model and rank-based approach can offer improvements in power over traditional approaches. Muscle Nerve 46: 501–511, 2012

Ancillary