Preclinical Development – Chemical
Leslie Benet, PhD
Professor
UCSF School of Pharmacy
San Francisco, California
Jae Chang, PhD
Head of Drug Discovery Sciences
Exelixis
Alameda, California
For the past 50 years our field has believed that different models of organ elimination define the clearance relationship of drug elimination in utilizing preclinical measurements to define in vivo predictability. We have proposed that all published human and animal clearance values are arterial clearances and that no model of organ elimination is necessary to describe organ elimination. We show today that the equation the field has believed to be the well-stirred model (WSM) and most often used to predict clinical clearance values is, in fact, an organ of elimination model independent relationship that can be derived by Kirchhoff’s Laws. In this presentation we enumerate the counter-scientific positions one must continue to support when maintaining that organ models of elimination are relevant to define organ clearance. We also demonstrate that the generally accepted WSM equation when basolateral hepatic influx and efflux transporters are clinically relevant, the Extended Clearance Concept (ECC) Equation, requires illogical suppositions and that this equation is not valid. Based on Kirchhoff’s Laws we derive the correct ECC Equation. We cannot identify any useful advance in understanding measured clearance values and defining drug disposition that results from utilization of the WSM, the parallel tube or dispersion models. But if the correct clearance-intrinsic relationships are given either by Kirchhoff’s Laws or by the WSM hypotheses of the 1970s, why does in vitro-in vivo extrapolation (IVIVE) not give valid answers for predicting in vivo clearance from in vitro measures? One possibility is that since arterial clearance is a whole-body parameter, the relevant blood flow limiting this value is cardiac output, not blood flow to elimination organs. But this does not explain the general underprediction for the majority of low clearance compounds. We asked what pharmacokinetic parameter may be the same for the in vitro incubation hepatic elimination and the in vivo hepatic clearance for some drugs (giving good IVIVE predictions) but different for other compounds (giving poor IVIVE predictions) independent of blood flow? In the in vitro incubation, the volume of distribution of the substrate and the enzymes is the same, the volume of fluid in which the incubation is carried out. However, this may not be true in vivo in the liver where drug may distribute into lipophilic portions of the liver into which the enzymes cannot distribute. In this case we are trying to use a one-compartment in vitro measure of intrinsic clearance to predict a multicompartment in vivo intrinsic clearance in the liver. In a second approach, we recognized when applying Kirchhoff’s Laws to the Michaelis-Menten relationship that the rate constants applicable to the Vmax/Km ratio may give a value that is generally greater than the rate constant of elimination determined from the in vitro incubation mixture utilized to predict the intrinsic clearance. In this presentation, I will review attempts to validate the three approaches listed above as well as the relevance of protein binding measurements in correctly improving clinical predictability utilizing preclinical approaches.