Safety and Efficacy Modelling in Anti-Diabetic Drug Development

Abstract: A central aim in drug development is to ensure that the new drug is efficacious and safe in the intended patient population.Mathematical models describing the pharmacokinetic-pharmacodynamic (PK-PD) properties of a drug are valuable to increase the knowledge about drug effects and disease and can be used to inform decisions. The aim of this thesis was to develop mechanism-based PK-PD-disease models for important safety and efficacy biomarkers used in anti-diabetic drug development.Population PK, PK-PD and disease models were developed, based on data from clinical studies in subjects with varying degrees of renal function, non-diabetic subjects with insulin resistance and patients with type 2 diabetes mellitus (T2DM), receiving a peroxisome proliferator-activated receptor (PPAR) ?/? agonist, tesaglitazar.The PK model showed that a decreased renal elimination of the metabolite in renally impaired subjects leads to increased levels of metabolite undergoing interconversion and subsequent accumulation of tesaglitazar. Tesaglitazar negatively affects the glomerular filtration rate (GFR), and since renal function affects tesaglitazar exposure, a PK-PD model was developed to simultaneously describe this interrelationship. The model and data showed that all patients had decreases in GFR, which were reversible when discontinuing treatment.The PK-PD model described the interplay between fasting plasma glucose (FPG), glycosylated haemoglobin (HbA1c) and haemoglobin in T2DM patients. It provided a mechanistically plausible description of the release and aging of red blood cells (RBC), and the glucose dependent glycosylation of RBC to HbA1c. The PK-PD model for FPG and fasting insulin, incorporating components for ?-cell mass, insulin sensitivity and impact of disease and drug treatment, realistically described the complex glucose homeostasis in the heterogeneous patient population.The mechanism-based PK, PK-PD and disease models increase the understanding about T2DM and important biomarkers, and can be used to improve decision making in the development of future anti-diabetic drugs.