Background. Vildagliptin acts by inhibiting dipeptidyl peptidase IV (DPP-4), thereby increases active GLP-1 (glucagon-like peptide 1) concentrations and decreases plasma glucose in diabetic patients.
Objectives. To develop a mechanism-based population PK/PD model that simultaneously describes and predicts vildagliptin pharmacokinetics and its effects on DPP-4 activity and the underlying glucose-insulin-GLP-1 system.
Methods. Data used for model building were from thirteen type 2 diabetic patients that received 10, 25, and 100 mg vildagliptin and placebo for 28 days in a 4-way crossover study (He et al, 2007). Vildagliptin concentrations, DPP-4 activity, and concentrations of active GLP-1, glucose, and insulin were co-modelled. The full population PK/PD model was developed utilizing the MC-PEM algorithm in parallelized S-ADAPT.
Results. A target-mediated drug disposition (TMDD) model that accounts for the high-affinity capacity-limited binding of vildagliptin to DPP-4 in plasma and tissues adequately described the PK and DPP-4 activity. Parallel vildagliptin dissociation from DPP-4 by a slow first-order process and hydrolysis by DPP-4 to an inactive metabolite explained the slightly more than dose proportional increase in vildagliptin concentrations. Active GLP-1 secretion was stimulated by GI intake of nutrients and GLP-1 was metabolized by DPP-4 and an additional non-saturable pathway. Both glucose-dependent insulin secretion and insulin-dependent glucose utilization were stimulated by active GLP-1 in the model. The non-saturable vildagliptin clearance [population mean (inter-individual coefficient of variation)] was 36.4 L/h (25%). Estimates for vildagliptin binding to DPP-4 were 71.9 nM (54%) for the association constant, 1.1 h (94%) for dissociation half-life, and 6.3 h (81%) for the half-life of vildagliptin hydrolysis by DPP-4. Modelling suggests that complete inhibition of DPP-4 resulted in an approximately 2.5-fold increase in active GLP-1 half-life.
Conclusions. Vildagliptin was found to be an inhibitor and a substrate for DPP-4. Effects on glucose and insulin were modelled by stimulation of both insulin secretion and peripheral insulin sensitivity by active GLP-1. The model for the underlying glucose-insulin-GLP-1 system including reciprocal feedbacks can be used as a basis to evaluate other drug effects or even combination therapy. Parallelized S-ADAPT proved to be an excellent choice for estimating a complex population model such as the current PK/PD model.
Reference: He Y-L et al (2007) Clin Pharmacokinet; 46: 577-588.