Abstract
Enrichment designs with a continuous biomarker require the estimation of a threshold to determine the subpopulation benefitting from the treatment. This paper provides the optimal allocation for inference in a two-stage enrichment design for treatment comparisons, when a continuous biomarker is suspected to affect patient response. Several design criteria, associated with different trial objectives, are optimized under balanced or Neyman allocation and under equality of the first two empirical biomarker’s moments, or under a weaker condition.
Since future patients’ biomarker values are unknown, such designs can be implemented sequentially to asymptotically achieve the optimum. We propose a covariate-adaptive randomization procedure in which, at each step, a distance metric between the current design and the optimum is minimized, and we prove that the underline stochastic process is a Markov chain bounded in probability. Theoretical and simulation results show that this strategy improves the efficiency of a two-stage enrichment clinical trial, especially with smaller sample sizes and under heterogeneous responses.
Organizzatore
Laura Anderlucci