This book deals with advanced methods for adaptive phase I dose-finding clinical trials for
combination of two agents and molecularly targeted agents (MTAs) in oncology. It provides not
only methodological aspects of the dose-finding methods but also software implementations and
practical considerations in applying these complex methods to real cancer clinical trials. Thus
the book aims to furnish researchers in biostatistics and statistical science with a good
summary of recent developments of adaptive dose-finding methods as well as providing
practitioners in biostatistics and clinical investigators with advanced materials for designing
conducting monitoring and analyzing adaptive dose-finding trials. The topics in the book are
mainly related to cancer clinical trials but many of those topics are potentially applicable
or can be extended to trials for other diseases. The focus is mainly on model-based
dose-finding methods for two kinds of phase I trials. One is clinical trials with combinations
of two agents. Development of dose-finding methods for two-agent combination trials requires
reasonable models that can adequately capture joint toxicity probabilities for two agents
taking into consideration possible interactions of the two agents on toxicity probability such
as synergistic or antagonistic effects. Another is clinical trials for evaluating both efficacy
and toxicity outcomes in single- and two-agent combination trials. These methods are often
applied to the phase I trials including MTAs because the toxicity and efficacy for a MTA does
not monotonically increase with dose but the efficacy often increases initially with the dose
and then plateaus. Successful software implementations for several dose-finding methods are
introduced in the book and their operating characteristics in practice are discussed. Recent
advance of the adaptive dose-finding methods in drug developments are also provided.