Calculation of agonist efficacy, apparent affinity, and receptor population changes after administration of insurmountable antagonists: comparison of different analytical approaches.

The determination of the affinity and efficacy of an agonist in functional (as opposed to radioligand binding) experiments is necessary to explain its potency. Using modeled dose-response data both in their ideal form and with an added average deviation as well as previously published experimental data, a variety of analytical approaches were compared which differed in goodness-of-fit, ease of handling, and range of successful application. A nonlinear curve-fitting algorithm that analyzed several dose-response curves simultaneously and fitted them to an extended version of the operational model of Black and Leff (1983) (Proc R Soc Lond B 220:141-162) was demonstrated to be superior to the other approaches using the above criteria. However, judging from the limitations of the analytical approaches, claims of efficacy or affinity differences between agonists that are based on less than 10-fold numerical differences in the same behavioral paradigm should be viewed with skepticism. It was also found that simple inspection of dose-response curves obtained before and after administration of an insurmountable antagonist give estimates of fair accuracy under most circumstances.