Generalized optimal design for two-arm, randomized phase II clinical trials with endpoints from the exponential dispersion family

Pharm Stat. 2016 Nov;15(6):459-470. doi: 10.1002/pst.1769. Epub 2016 Aug 11.

Abstract

For two-arm randomized phase II clinical trials, previous literature proposed an optimal design that minimizes the total sample sizes subject to multiple constraints on the standard errors of the estimated event rates and their difference. The original design is limited to trials with dichotomous endpoints. This paper extends the original approach to be applicable to phase II clinical trials with endpoints from the exponential dispersion family distributions. The proposed optimal design minimizes the total sample sizes needed to provide estimates of population means of both arms and their difference with pre-specified precision. Its applications on data from specific distribution families are discussed under multiple design considerations. Copyright © 2016 John Wiley & Sons, Ltd.

Keywords: multiple constraints; optimized design; sample size; standard error.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Clinical Trials, Phase II as Topic / methods*
  • Endpoint Determination
  • Humans
  • Randomized Controlled Trials as Topic / methods*
  • Research Design*
  • Sample Size