Can parallel multi-compartment In vitro drug binding measurements improve the prediction of In vivo PK measurements?

Wright, Wayne Ludlow

Thesis

Can parallel multi-compartment In vitro drug binding measurements improve the prediction of In vivo PK measurements?

Download PDF

Creator

Wright, Wayne Ludlow

Rights statement

Strathclyde Thesis Copyright

Awarding institution

University of Strathclyde

Date of award

2023

Thesis identifier

T17269

Person Identifier (Local)

201563834

Qualification Level

Doctoral (Postgraduate)

Qualification Name

Doctor of Philosophy (PhD)

Department, School or Faculty

Strathclyde Institute of Pharmacy and Biomedical Sciences

Abstract

Measurement of unbound drug concentration (Cu), plasma protein binding (PPB), volume of distribution (VD) are important pharmacokinetic parameters for understanding drug distribution in-vivo. However, experimentation is generally performed in-vitro under non-competitive conditions across single tissues; this does not adequately reflect the in-vivo situation where drug binds competitively between blood components (e.g., plasma) and tissues (proteins and lipids). Several assays are therefore required across multiple tissues and species which can be time consuming. The aim of this thesis was to evaluate a new device (Competitive Rapid Equilibrium Dialysis – “CRED”) which allows for plasma protein and tissue binding to be conducted in parallel from several tissues (as occurs in-vivo). The unbound drug compartments of the CRED are interconnected thereby mimicking how unbound drug is the fraction available for in vivo drug distribution. This thesis investigates the use of the CRED device toward the goal of improving in-vitro predictions of in-vivo drug distribution by comparison of plasma protein binding and volume of distribution data obtained using column chromatography and observed literature respectively. The 6 compartment CRED system was adopted in order to study a set of tool compounds that span a wide range of physicochemical properties, and therefore binding affinities. Chapter 2 explores a simple model combining human serum albumin (HSA) and phosphatidylcholine (PC) acting as competitive binding surrogates of plasma and tissues protein, respectively. A correlation coefficient R2 of 0.7 was obtained for plasma protein binding (PPB) when compared to HSA binding measured using column chromatography. Applying a more complex mixture of phospholipids (phosphatidylethanolamine, phosphatidylserine and phosphatidylinositol) to better reflect the lipid content of tissues in-vivo, the volume of distribution correlation coefficient R2 was improved from 0.5 (PC only) to 0.80 (comparison to literature data). Some researchers have reported a general increase in PPB with increasing size of preclinical species and finally to humans. Investigation of PPB across species in parallel enabled this to be further investigated and was found to be dependent on both compound and protein load which aids the predictability in-vivo (Chapter3). The major bottleneck using CRED is the time to equilibrium which is a particular drawback when studying multiple compounds. The throughput was improved using a cassette of compounds along with a highly optimised LC-MS approach for rapid analysis from 4.0hrs to under 1hr per 96 well block (RapidSep, Chapter4). Overall, the competitive binding environment provides the platform to improve in vivo drug distribution and facilitate the advancement of better chemical leads.

Advisor / supervisor