ABSTRACT
In the previous chapter we discussed the phases-II-III seamless designs, for
which the type-I errors are strongly controlled. In contrast, there is another
school of research on multiple-arm trials mainly for phase-II dose-finding stud-
ies, in which selecting target dose (dose schedule) such as minimum effective
dose (MED) is the main purpose with a reasonable control of the type-I error.
The MED can be defined as the dose where the mean efficacy outcome is equal
to a certain target, with the placebo (or an active control) used as a reference.
Mean efficacy is usually assumed to be nondecreasing with dose. Both efficacy
and safety endpoints are often taken into consideration when selecting a dose
for further studies in phase-III trials, because increasing the dose can result in
both higher efficacy and increased adverse event rates. A common approach is
to quantify efficacy and adverse event rate trade-off through a utility function
that incorporates both efficacy and safety into a measure of overall clinical util-
ity (Berry et al., 2001; Dragalin and Fedorov, 2006; Ivanova et al. 2009, 2012).
Such a utility function is typically umbrella-shaped and the goal is to find a
dose that maximizes the utility of the drug candidate. The objective in phase
II can also be to test efficacy and adverse event rates at the estimated MED or
the optimal dose against a control and recommend for further study in phase-
III trials (Ivanova et al. 2012). Miller et al. (2007) investigated a two-stage
strategy for a dose-ranging study that is optimal across several parametric
models. Dragalin et al. (2008) investigated optimal two-stage designs for two
correlated binary endpoints that follow a bivariate probit model. Bretz et al.
(2005) studied the dose-finding methods under various dose-response scenar-
ios including umbrella-shaped response curves. Ivanova et al. (2012) studied
a Bayesian two-stage adaptive design for finding MED under the same set of
dose-response curves and compared the section probability and power against
uniform allocation method.