Toward an improved prediction of human in vivo brain penetration

Abstract

The penetration of drugs into the central nervous system is a composite of both the rate of drug uptake across the blood-brain barrier and the extent of distribution into brain tissue compartments. Clinically, positron emission tomography (PET) is the primary technique for deriving information on drug biodistribution as well as target receptor occupancy. In contrast, rodent models have formed the basis for much of the current understanding of brain penetration within pharmaceutical Drug Discovery. Linking these two areas more effectively would greatly improve the translation of candidate compounds into therapeutic agents. This paper examines two of the major influences on the extent of brain penetration across species, namely plasma protein binding and brain tissue binding. An excellent correlation was noted between unbound brain fractions across species (R(2) > 0.9 rat, pig, and human, n = 21), which is indicative of the high degree of conservation of the central nervous system environment. In vitro estimates of human brain-blood or brain-plasma ratios of marketed central nervous system drugs and PET tracers agree well with in vivo values derived from clinical PET and post-mortem studies. These results suggest that passive diffusion across the blood-brain barrier is an important process for many drugs in humans and highlights the possibility for improved prediction of brain penetration across species.