Karin E. Fattinger, Davide Verotta, Herve C. Porchet, Alain Munafo, Jean-Yves le Cotonnec and Lewis B. Sheiner
A pharmacodynamic analysis of the input-response relationship between the GnRH antagonist antide and luteinizing hormone and testosterone concentrations is presented. A control compartmental model is developed using pharmacokinetic and pharmacodynamic data from two phase I studies where different short intravenous infusions of the drug were given to healthy male volunteers. Because of the control interdependence between serum luteinizing hormone and testosterone concentrations, a separation principle similar to that suggested previously to analyze physiological pharmacokinetic data is used for model exploration. That is, testosterone and luteinizing hormone are first modeled separately. We condition on observed luteinizing hormone when modeling testosterone (testosterone conditional model) and then condition on observed testosterone and predicted (from a pharmacokinetic model) antide concentrations when modeling luteinizing hormone (luteinizing hormone conditional model). The submodels reveal that the effect of luteinizing hormone on testosterone production depends on previous exposure to luteinizing hormone and that LH production depends not on current but on previous testosterone exposure resulting in an overshoot of luteinizing hormone after termination of antide suppression. Both submodels are then combined into one global compartmental model, which in addition contains a model for diurnal variation in testosterone production. The final combined model of luteinizing hormone and testosterone describes the observed data well and can be used to predict luteinizing hormone and testosterone responses to nonstudied antide dosages. However, the sensitivity of predictions to model assumptions limits the range of valid extrapolation, and this, too, is illustrated.