ABSTRACT
Studying and understanding the joint effect of combined treatments is important in pharmacology and in the development of combination therapies for many diseases. The Loewe additivity model has been considered as one of the best general reference models for evaluating drug interactions. The Loewe additivity model can be characterized by the interaction index, which can be obtained based on the available concentration-effect data, the assumed relationship between concentration and effects, and the experimental designs. In this chapter, we present how to estimate the interaction index and construct its 95% condence interval (95% CI) under three different settings. In the case where only a small number of combination doses are studied, one may rst estimate the concentration-effect curves for each single drug, and then calculate the interaction index and its 95% CI for each combination dose (see Section 4.2). When there are more combination doses, this approach tends to be more varying as it depends only on the measurement at a single combination dose level. To gain efciency, one can assume a model and pool the data at various combination doses to form a better estimate of the interaction index. Two commonly used approaches are the
CONTENTS
4.1 Introduction ..................................................................................................56 4.2 Condence Interval for Interaction Index When a Single
Combination Dose Is Studied ....................................................................58 4.3 Condence Interval for Interaction Index When Ray Design Is
Applied .......................................................................................................... 61 4.4 Condence Interval for Interaction Index When Response
Surface Models (RSMs) Are Applied ........................................................64 4.4.1 Model of Greco et al. (1990) ............................................................65 4.4.2 Model of Machado and Robinson (1994) ......................................66 4.4.3 Model of Plummer and Short (1990) .............................................66
