An R package for implementing simulations for seamless phase II/III clinical trials using early outcomes for treatment selection

  • Authors:
  • Nick Parsons;Tim Friede;Susan Todd;Elsa Valdes Marquez;Jeremy Chataway;Richard Nicholas;Nigel Stallard

  • Affiliations:
  • Warwick Medical School, University of Warwick, Coventry, CV4 7AL, UK;Department of Medical Statistics, University Medical Center Göttingen, Göttingen, Germany;Applied Statistics, University of Reading, UK;Clinical Trial Service Unit and Epidemiological Studies Unit, Oxford, UK;Imperial College Healthcare NHS Trust, London, UK and National Hospital for Neurology and Neurosurgery, London, UK;Imperial College Healthcare NHS Trust, London, UK;Warwick Medical School, University of Warwick, Coventry, CV4 7AL, UK

  • Venue:
  • Computational Statistics & Data Analysis
  • Year:
  • 2012

Quantified Score

Hi-index 0.03

Visualization

Abstract

Adaptive seamless phase II/III clinical trial designs allowing treatment selection at an interim analysis have gained much attention because of their potential benefits compared to more conventional drug development programmes with separate trials for individual phases. A scenario of particular interest is that in which the final outcome in the trial is based on long-term follow-up, but the interim analysis can only realistically be based on early (short-term) outcomes. A new software package (asd) for the statistical software R implements simulations for designs of this type, in addition to the simpler scenario where treatment selection is based on the definitive (final) outcome. The methodology is briefly described and two examples of proposed trial designs in progressive multiple sclerosis are provided, with R code to illustrate application of the methodology.