Basic Concepts in Physiologically Based Pharmacokinetic Modeling in Drug Discovery and Development

Mathematical models are frequently used to help understand the PK of drugs following intravenous (i.v.) and oral dosing in animals and humans. Models describing other routes of administration, e.g., transdermal delivery of drugs across skin, are also available.1 These models are often used to describe the relationship between the plasma or relevant tissue concentration of the drug and time and are built using compartments or “building blocks”. A classical PK model typically has a central compartment representing plasma that is linked to one or two peripheral compartments via rate constants.2 When defined in the terms of rate constants, the model parameters do not generally have any physiological meaning but can be transformed to provide more interpretable PK descriptors, e.g., clearance and volume of distribution. Clearance refers to the volume of plasma cleared of drug per unit time via metabolic or excretion processes. The volume of distribution refers to the volume of plasma required to occupy the total amount of drug in the body at the concentration observed in plasma. Both the clearance and volume of distribution can be used to calculate the effective half-life or “residence” time of the drug. In this context, these models are useful as they offer a concise and standard representation of both the preclinical and clinical experimental results. In a preclinical setting, PK parameters from different in vivo studies can be used to rank compounds for further investigation or can be linked to physicochemical, in vitro or structural properties to guide optimization of PK properties for new compounds. In the clinical setting, PK parameters for different subjects can be compared and potentially related to demographic characteristics, or PK parameters derived from one study design can be used to simulate plasma concentrations for alternative doses or dosing regimens. Typically, population PK models, which aim to describe the covariates of variability in drug concentrations and PK parameters among individuals in the target patient population, are used to inform initial dose selection or personalize dosage in subgroups of patients. Demographic and physiological variables, including body weight and metabolic functions, are often evaluated as covariates as it is known that these can affect dose– concentration relationships. However, these more empirical approaches cannot accommodate all prior information on both the drug and the physiology, thus restricting the ability to predict PK for a similar drug or to extrapolate the PK to different physiological conditions. Although PBPK models are built using a similar mathematical framework, they are parameterized using known physiology and consist of a larger number of compartments which correspond to the different organs or tissues in the body. These compartments are connected by flow rates that parallel the circulating blood system. These models, like the more empirical models, provide estimates of common PK parameters, e.g., clearance, volume of distribution, and effective half-life. However, these more physiologically relevant models provide a quantitative mechanistic framework by which scaled drug-specific parameters (using in vitro-in vivo extrapolation (IVIVE) techniques) can be used to predict the plasma and, importantly, tissue concentration–time profiles of new drugs, following i.v. or oral administration. By their very nature, they can be used to extrapolate a dose in healthy volunteers to one in a disease population if the relevant physiological properties of the target population are available. For example, data relating to reduced cytochrome P450 (CYP) expression in patients with chronic kidney disease can be incorporated into a PBPK model (along with other parameters that are known to be affected) to predict a dose adjustment of a drug relative to that in a healthy volunteer population.3