Pharmacokinetic and pharmacodynamic properties of insulin aspart and human insulin

The preferred approach to determine the pharmacokinetic (PK) and pharmacodynamic (PD) properties of insulin analogues is the euglycemic glucose clamp. Currently, non-compartmental data analytical approaches are used to analyze data. The purpose of the present study is to propose a novel compartmental-model for analysis of data from glucose clamp studies. Data used in this trial only involved 18 of the 20 originally treated subjects. Data was obtained from a crossover trial where 18 healthy subjects each received a single subcutaneous (sc) dose of 1.2 nmol/kg (body weight) insulin aspart (IAsp) or 1.2 nmol/kg human insulin (HI) during a euglycemic glucose clamp after overnight fast. Serum insulin and glucose concentrations were measured and the glucose infusion rate (GIR) was adjusted after dosing, to maintain blood glucose near basal levels. Individual model parameters were estimated for IAsp, HI, and the corresponding glucose and GIR data. We found statistically significant differences between most of the HI and IAsp pharmacokinetic parameters, including the sigmoidicity of the time course of absorption (1.5 for HI vs. 2.1 for IAsp (unit less), P=0.0005, Wilcoxon Signed-rank test), elimination rate constant (0.010 min(-1) for HI vs. 0.016 min(-1) for IAsp (P=0.002)). The PD model parameters were mostly not different, except for the rate of insulin action (0.012 min(-1) for HI vs. 0.017 min(-1) for IAsp (P=0.03)). The model may provide a framework to account for different PK properties when estimating the PD properties of insulin and insulin analogues in glucose clamp experiments.