Free Drug Concepts: A Lingering Problem in Drug Discovery

Abstract

The concepts, while not complex, can be counterintuitive. The processes governing free drug concentrations in vivo are different from those in in vitro assays but are often assumed to be the same. The most common calculation performed (conversion of total drug concentration to free drug concentration) requires an equation in a form that, while mathematically correct, has led to a mechanistic misunderstanding. This misunderstanding results in many published studies being incorrect in their use of pharmacokinetic data as drivers of compound optimization. Pharmacokinetic training focuses on total drug concentrations and parameters, and these also dominate clinical pharmacokinetics where the focus is on a single compound, and for which plasma binding is usually constant. Compound optimization in drug discovery is anchored on the understanding of unbound drug exposure and its relationship to target engagement; thus, comparisons between molecules requires the use of unbound exposure and unbound PK parameters. Even today, authors use the term “Free drug hypothesis”, as though there is some doubt as to its validity. It is not a hypothesis; it is a fundamental principle. Unbound drug concentrations at the target site determine the degree of target engagement Unbound concentrations in vivo are not determined by the extent of plasma or tissue binding; instead, plasma and tissue binding determine the concentrations of bound drug Unbound drug concentrations in vivo are determined by the rate of drug input (absorption) and the rate of elimination (intrinsic clearance) At equilibrium, free drug concentrations in cytosol approximate those in plasma (with a few specific exceptions, described below Figure 1. Clearance terms derived from in vitro and in vivo experiments and their relationships Note: CLint is, by definition, an unbound term and thus CLint measured in vitro is an apparent value until corrected for binding in the assay. However, for the avoidance of doubt, we have used CLint,u to represent the binding-corrected in vitro value. Figure 2. Effect of increasing dose on the total and free AUC and clearance for ceftriaxone. Figure 3. Simulations demonstrating influence of plasma protein binding on pharmacokinetic parameters. Figure 4. Mean Kp_uu (solid bars) and Kp for neutral, acidic (mainly ionised at pH 7.4) and basic drugs (mainly ionised at pH 7.4). Data replotted from Mateus et al. (11) In vivo, protein binding and free fraction have no effect on free drug concentration for oral drugs at steady-state. For oral drugs, intrinsic clearance is the key determinant of free drug concentrations in plasma and tissues. In vitro potency assays represent a closed, static system in which nonspecific binding determines the free drug concentration, and should be corrected for. The only reason to measure fraction unbound in plasma is to convert total plasma concentrations/parameters to free concentrations/parameters. In most situations, free drug concentrations in plasma can be used as a surrogate for free drug concentrations at the site of action. Optimizing for low dose and acceptable dose frequency requires the appropriate balance between unbound exposure and unbound potency, and plasma protein binding has no impact on either of these. clearance intrinsic clearance intrinsic clearance corrected for binding to in vitro system unbound clearance fraction unbound in plasma fraction unbound in tissues ratio of concentrations in tissue and plasma ratio of unbound concentrations in tissue and plasma volume of distribution volume of distribution at steady-state unbound volume of distribution at steady-state This article references 11 other publications. This article has not yet been cited by other publications.