Abstract:
|
Pediatric trials are often required in the later stages of clinical development, more often once a medical product has been recently approved in the corresponding adult patient population. Hence recruitment of pediatric subjects in a controlled clinical trial, especially when there is a placebo or a perceived inferior treatment as the controlled reference comparator, can be extremely difficult given the traditional equal randomization allocation to both study treatments. One practical approach to improve recruitment is to consider an unequal randomization that allows for more pediatric subjects to be allocated to the recently approved medical product than the reference treatment. However, such a randomized allocation is well known not to be optimal in the traditional frequentist design. To address this optimality issue, a Bayesian design and analysis approach is proposed in which an informative prior is placed on the randomized reference treatment to improve efficiency in light of unequal randomization. Operational design characteristics will be investigated as well as regulatory discussion strategies.
|