Modeling of dose-response-time data: Four examples of estimating the turnover parameters and generating kinetic functions from response profiles

The most common approach to in vivo pharmacokinetic and pharmacodynamic mod eling involves sequential analysis of the plasma concentration versus time and then response versus time data, such that the plasma kinetic model prov ides an independent function, driving the dynamics. However, response versu s time data, even in the absence of measured drug concentrations; inherentl y contain useful information about the turnover characteristics of response (turnover rate, half-life of response), the drug's biophase kinetics (F, h alf-life) as well as the pharmacodynamic characteristics (potency, intrinsi c activity). Previous analyses have assumed linear kinetics, linear dynamic s, no time lag between kinetics and dynamics (single-valued response), and time constant parameters. However, this report demonstrates that the drug e ffect can be indirect (antinociception, cortisol/adrenocorticotropin(ACTH), body temperature), display nonlinear kinetics, display feedback mechanisms (nonstationarity, cortisol/ACTH) and exhibit hysteresis with the drug leve ls in the biophase (antinociception, body temperature). It is also demonstr ated that crucial;determinants of the success of modeling dose-response-tim e data are the dose selection, multiple dosing, and to some extent differen t input rates and routes. This report exemplifies the possibility of assign ing kinetic forcing functions in pharmacodynamic modeling in both preclinic al and clinical studies for the purpose of characterizing (discrimination b etween turnover and drug-specific parameters) response data and optimizing subsequent clinical protocols, and for identification of inter-individual d ifferences. Copyright (C) 2000 John Wiley & Sons, Ltd.