In vitro simulation of the human proximal colon to advance biopharmaceutics modelling

O'Farrell, Connor ORCID: 0000-0002-4888-8892 (2024). In vitro simulation of the human proximal colon to advance biopharmaceutics modelling. University of Birmingham. Ph.D.

Preview

OFarrell2024PhD.pdf
Text - Accepted Version
Available under License All rights reserved.
Download (12MB) | Preview

Abstract

The Dynamic Colon Model (DCM) is an advanced in vitro model of the human proximal colon. This thesis takes an existing DCM through design, development and validation work to full application as an advanced biopharmaceutical dissolution apparatus. The model was improved in terms of robustness, longevity and industry-integratability. The number of motility patterns until failure was increased from < 80 (n = 3) to > 14, 400.

Magnetic resonance imaging (MRI) showed that the DCM mimicked wall motion, luminal flow patterns and velocities of the contents of the human proximal colon. The segmented architecture and peristaltic activity of the DCM generated flow profiles that were distinct from compendial dissolution apparatuses. Different motility patterns and media viscosities were shown to be classifiable by the degree of mixing-related motion
using a new MR tagging method. Furthermore, flow studies were used to inform the design of a digital twin that robustly replicated flow patterns in the DCM under different physiological conditions (media viscosity, volume, and peristaltic wave speed).

Five biorelevant motility patterns extracted from in vivo data were applied to the DCM which significantly affected theophylline release from a controlled release tablet. The concentration profiles of theophylline were markedly localized when measured at different segments of the DCM tube, highlighting the importance of a segmented lumen in intestine models and in generating spatial information to support simple temporal dissolution profiles.

It is recommended that viscosity and shear rates are considered when designing future dissolution test methodologies for hydrophilic matrix-based controlled-release formulations. Mean wall shear rates were strongly time and viscosity dependent although peaks were comparable to previously published estimates of the USPII operated at 25 and 50 rpm. The shear stresses invoked by the unstimulated, healthy adult human colon may be lower than those in the USP II at 25 rpm and thus insufficient to achieve total release of a therapeutic compound from a dosage form. When operated under stimulated conditions, measured drug release in the DCM was between that measured at 25 and 50 rpm in the USP II.

Ultimately, an in silico physiologically-based biopharmaceutic (PBB) model was developed for theophylline in Gastroplus®. To predict theophylline pharmacokinetics from a controlled release tablet formulation, Uniphyllin Continus®, dissolution profiles from the DCM were integrated as global mean data and regional release profiles. The latter used spatiotemporal data collected in the DCM to model drug dissolution and distribution throughout the continuous caecum-ascending colon lumen. Predicted absorption rate was inherently sensitive to hydrodynamics (using the USP II apparatus as a control). Motility patterns which caused drug to accumulate in the caecum resulted in a lower predicted absorption rate than those which performed better at homogenising the lumen, which may have consequences for formulation performance in subjects with impaired motility. Recommendations are made to further compartmentalise the colon in PBB modelling platforms to improve prediction performance for controlled release formulations.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):