The dynamics of Aβ distribution after γ-secretase inhibitor treatment, as determined by experimental and modelling approaches in a wild type rat

• Published in: Journal of pharmacokinetics and pharmacodynamics Vol. 39; no. 3; pp. 227 - 237
• Main Authors: Tai, Leon M., Jacobsen, Helmut, Ozmen, Laurence, Flohr, Alexander, Jakob-Roetne, Roland, Caruso, Antonello, Grimm, Hans-Peter
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01-06-2012
• Subjects: Amyloid beta-Peptides - metabolism; Amyloid Precursor Protein Secretases - antagonists & inhibitors; Amyloid Precursor Protein Secretases - metabolism; Animals; Biochemistry; Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Brain - drug effects; Brain - enzymology; Brain - metabolism; Male; Models, Biological; Original Paper; Pharmacology/Toxicology; Pharmacy; Protease Inhibitors - pharmacology; Rats; Rats, Wistar; Treatment Outcome; Veterinary Medicine/Veterinary Science; Gamma-secretase inhibitor; Amyloid beta
• ISSN: 1567-567X; 1573-8744
• DOI: 10.1007/s10928-012-9246-4

Inhibition of the enzyme(s) that produce the Amyloid beta (Aβ) peptide, namely BACE and γ-secretase, is considered an attractive target for Alzheimer’s disease therapy. However, the optimal pharmacokinetic–pharmacodynamic modelling method to describe the changes in Aβ levels after drug treatment is unclear. In this study, turnover models were employed to describe Aβ levels following treatment with the γ-secretase inhibitor RO5036450, in the wild type rat. Initially, Aβ level changes in the brain, cerebral spinal fluid (CSF) and plasma were modeled as separate biological compartments, which allowed the estimation of a compound
IC
50
and Aβ turnover. While the data were well described, the model did not take into consideration that the CSF pool of Aβ most likely originates from the brain via the CSF drainage pathway. Therefore, a separate model was carried out, with the assumption that CSF Aβ levels originated from the brain. The optimal model that described the data involved two brain Aβ 40 sub-compartments, one with a rapid turnover, from which CSF Aβ 40 is derived, and a second quasi-static pool of ~20%. Importantly, the estimated in vivo brain
IC
50
was in a good range of the in vitro
IC
50
(ratio, 1.4). In conclusion, the PK/PD models presented here are well suited for describing the temporal changes in Aβ levels that occur after treatment with an Aβ lowering drug, and identifying physiological parameters.