On determining the levels of treatment to optimize the probability of a favorable response.

Abstract

This paper presents a statistical method for analyzing the dose response surface associated with treatments resulting from combinations of any number of cytotoxic agents. The results of such an analysis are the dosages of each agent in the combination which will maximize the probability of a favorable response. An example utilizing a combination of hexamethylmelamine and ICRF-159 is presented. The predictive ability of the model is verified by a kappa 2 goodness of fit test. The advantages of such a method relate to its potential applicability to more complex experimental situations where multiple drugs are used in combination and/or where variations in timing of drug administration exist.