A direct, general approach based on isobolograms for assessing the joint action of drugs in pre-clinical experiments

Pharmacologists and other biologists frequently use methods based on the interpretation of isobolograms to quantify the extent of synergy or antagonism between drugs used in combination in pre-clinical studies. Most methods have been unsatisfactory from a statistical viewpoint, many because they have relied solely on visual evaluation, others because the methods have not taken into account the variability of the measurements. We describe a direct approach for quantifying the joint potency of two drugs, a central feature being the use of simple isobole models that lead directly to response surface models for the expected experimental outcomes. The approach is general in the sense that one can use it for discrete or continuous responses, different underlying probability distributions, linear or non-linear dose-response functions of the drugs used singly, and a variety of experimental designs. Our approach extends the suggestions made by Hewlett for measuring the joint potency of drugs, and is similar in spirit to the approaches proposed by Greco et al. and Weinstein et al. We describe the analysis of data from an in vitro experiment conducted to evaluate the efficacy of the antiviral drugs AZT and ddI used in combination.