Effective Implementation of Bayesian Adaptive Randomization in Early Phase Clinical Development

We illustrate the  use of placebo, low and high dose of an experimental treatment to evaluate 
Bayesian Adaptive Randomization(BAR) characteristics in a proof-of-concept dose selection study in 
oncology. Our proposed design links the treatment assignment probabilities to the performance 
of the respective arms. A smooth adaptation of assignment probabilities to treatment arms is 
guaranteed by a tuning parameter controlling how the data influence the randomization. 
We compare the design's operating characteristics with those of the screening design, used to 
calculate the maximum sample size.
We concluded that BAR is a flexible tool which may fit easily to the needs of early
development and may be ethically appealing as more subjects are assigned to more active treatment arms.


Pantelis Vlachos
Cytel Inc