Impact of Interindividual Differences in Plasma Fraction Unbound on the Pharmacokinetics of a Novel Syk Kinase Inhibitor in Beagle Dogs

Inconsistencies in pharmacokinetic parameters between individual animals in preclinical studies are a common occurrence. Often such differences between animals are simply accepted as experimental variability rather than as indications of specific differences in animal phenotype that could lead to a different interpretation of the data. The fraction unbound in plasma is one factor influencing pharmacokinetic parameters and is typically determined using pooled plasma from multiple animals, making the assumption that there is limited population variance. However, this assumption is not often tested and may not hold true if there are polymorphisms affecting binding or variation in the concentrations of individual plasma proteins that could give rise to different fraction unbound phenotypes in individual animals. During profiling of a novel Syk inhibitor, AZ8399, striking interindividual differences in total plasma clearance and volume of distribution were observed between dogs consistent with differences in fraction unbound between animals. Determination of the fraction unbound showed a ∼5-fold difference in fraction unbound between the animals in the study. Broader analysis of individual dogs across a colony demonstrated a correlation between individual animal fraction unbound with total plasma clearance and volume of distribution. The concentrations of the common drug-binding proteins albumin and α1-acid glycoprotein in plasma were determined, and α1-acid glycoprotein levels were found to correlate with fraction unbound. Finally, single-nucleotide polymorphisms were identified at c.502 and c.522 of exon 5 of the dog α1-acid glycoprotein gene that may be correlated to the α1-acid glycoprotein concentration phenotype observed.

SIGNIFICANCE STATEMENT

The current work demonstrates the potential for significant interindividual differences in plasma fraction unbound in beagle dogs and goes on to examine the underlying cause for the compound described. The findings suggest that the application of a population mean value of fraction unbound generated from a pooled sample may not always be appropriate and could introduce significant errors in scaling of in vitro clearance values, PBPK understanding, and interpretation of PKPD or toxicokinetic data in the context of unbound concentrations.