Methods of estimation of IC50 and SC50 parameters for indirect response models from single dose data.

The basic indirect response models are described by several pharmacodynamic parameters of which IC(50) (drug concentration eliciting 50% of the maximum inhibition) and SC(50) (drug concentration eliciting 50% of the maximum stimulation) are not readily derived from the response versus time or response versus concentration plots. We discuss limitations of existing methods of IC(50) and SC(50) estimation from single dose data. A novel approach to such estimations based on the area between the baseline and effect curve is introduced. The effects of pharmacokinetic profile, dose, data variability, and number of data on the parameter estimation are analyzed for the new and one of the previously described methods. Our analysis is based on reconstruction of true IC(50) and SC(50) values from rich and sparse computer-generated data sets modified by two noise levels. Extensions of these methods to two dose data sets are presented. Both methods yielded IC(50) and SC(50) close to the known true values. Depending on the factors mentioned above, these methods exhibited different levels of accuracy.