(189p) A Chemical Engineering Approach to Modelling Drug Dissolution and Transport Phenomena in the Lower Gastrointestinal Tract

The human proximal colon offers opportunities for targeted drug
delivery to treat local conditions such as ulcerative colitis and irritable
bowel disease, as well as more systemic diseases.  Whilst the colon has a lower
surface area than the small intestine, in some cases the long residence time
has certain advantages for treatments. 

Understanding of how drug
dissolution and absorption in the proximal colon might be achieved is hampered
by a lack of in vitro tools which mimic the complex hydrodynamics of the
colonic environment.  The dynamic colon model (DCM), shown in Figure 1, has
been developed to replicate the anatomy, motility patterns and physical
pressures of the human proximal colon [1]. This biorelevant dissolution
apparatus provides additional mechanistic insights on release alongside or in
place of standard USP methods, which rely on simple stirred vessels and low
viscosity buffer solutions which are not representative of conditions within
the colon. More complex in vitro models have been developed for the
colon, such as the three-stage tubular model and the TNO TIM-2 [2, 3]. These models
operate with enhanced biorelevance in many areas such as the incorporation of
fermentation reactors, but are compromised in terms of physical representation
of biological features and biomechanical motility.

The schematic in Figure 1 shows how the DCM replicates the
anatomy of the human proximal colon. The colon receives undigested food
material (chyme) from the ileum through the ileocaecal valve. This is
represented by the ileum terminal on the DCM, the injection point for dosage
forms, dyes etc. The ileocaecal valve joins the colon at the top of the caecum
– the pouch-like base of the proximal colon. The ascending colon, extending
from the caecum to a sharp bend known as the hepatic flexure, is the favoured
section for targeted drug delivery as the colonic contents have lowest
viscosity in this region. The ascending colon has a segmented appearance
comprising saccules called haustra. Contraction of smooth circular muscle
causes a haustrum to distend whilst the subsequent haustrum is relaxed, driving
the luminal contents forward at a rate controlled by a complexity of factors
that maximise water absorption whilst avoiding over-solidification [4]. The DCM
uses a computer controlled hydraulic system to mimic this circular muscle
activity by inflating and deflating the silicone membrane of each haustrum,
thus replicating motility patterns of the human proximal colon observed in
vivo.

In this paper, an extensive analysis of the fluid dynamics
with the DCM is discussed by comparison of results obtained from magnetic
resonance imaging (MRI) and positron emission tomography (PET). These
clinically relevant non-invasive imaging modalities visualise the mixing
processes within the DCM from very distinct perspectives. A comparison will
deliver further insights into fluid flow and how this can impact upon the
disintegration, dissolution and dispersion of solid dosage forms within the
colon. It is anticipated that the morphological MRI data will complement the
functional information gathered by PET, and vice versa, in a way that
progresses the current understanding and interpretation of each individual
dataset. The coupling of MRI and PET techniques will enable discussion in the
context of the synchronisation of colonic wall motion with distribution of
drug/dye molecules throughout the DCM. This forms part of a wider project to
establish an in vitro in vivo correlation (IVIVC) for dissolution
testing and validate the DCM for use as an enhanced biorelevant method for
measurement of dissolution of controlled release formulations, ultimately to
enable improved dosage forms to be developed and tested.

MRI has the capacity to produce a high resolution map that
tracks the dynamic morphology of the interior DCM environment during motility
patterns. Figure 2 depicts MRI snapshots of the partially filled DCM for a
single haustrum unit (Fig. 2 part 1) and the entire DCM tube comprising ten
haustrum units (Fig. 2 part 2). The signature soft tissue contrast exhibited by
MRI easily distinguishes between the water used to hydraulically control the
motion of the haustra walls (d) and the water in the lumen (e) without the
addition of contrasting agents. The critical advantage of MRI is the ability to
synchronise wall motion with fluid flow and map the flow field induced within
the DCM. Recent work has demonstrated that MRI of the human colon can be used
to measure the motion of colonic contents as well as the colonic wall [5,
6]. MRI therefore provides the opportunity to demonstrate the similarities
between the DCM and human colon in mixing and fluid flow.

Inside the DCM, PET offers the complementary ability to
image the key underlying mass transport processes that are driven by the
anatomical motility mapped by MRI; such as convection of a dynamically
disintegrating extended release dosage form within the lumen. PET enables the
local spatial concentrations of a tracer (analogous to a drug) to be mapped
during the different stages of a motility pattern. For example, Stamatopoulos,
et al. observed the accumulation of tracer molecules around the caecum and
ileum terminal likely due to lack of propulsion forces at such early stages in
development of an antegrade propagating wave [1].  Unlike the dynamically
changing colonic environment, conditions were fixed in these studies to permit
scrutiny of the interplay between a range of parameters such as fill level,
media viscosity and motility pattern (occlusion degree, velocity of propagating
wave, contraction/relaxation sequence, etc.). Media viscosity has been shown to
influence the convective mass transport of tracer as can be seen in Figure 3
[1]. Increasing viscosity from 0.25 to 0.50 % (w/w) NaCMC (Fig. 3 Parts 1 and 2
respectively) decreases the uniformity of tracer distribution along the lumen.
This is shown by lower intensity of the radioactive signal as x-axis position
increases from 0-20 cm. Furthermore, the amount of tracer remaining around the
caecum increases with viscosity, whilst a transition to plug flow is also
observed in the distribution of the remaining tracer along the lumen following
5-10 antegrade propagating sequences (Fig 3. Parts 2c & 2d). As the wave
reaches the end of the lumen, media flows up the wall of the rigid siphon
without passing over the flexure, before gravity forces induce backflow. This
backflow is stronger, closer to the siphon. For this reason, the higher
viscosity media used to generate Figure 3 part 2 imparts more resistance to
backflow, thus a more significant accumulation of tracer is observed around  x
= 13±1.5 cm.  These phenomena have also been observed during in vivo
scintigraphy studies [7, 8, 9].

Acknowledgements

Conner O’Farrell is sponsored by the EPSRC Centre for Doctoral
Training in Formulation Engineering (EP/L015153/1), AstraZeneca AB R&D,
Gothenburg.  We acknowledge the Sir Peter Mansfield Imaging Centre at
the University of Nottingham for their support in carrying out the MRI
experiments.

References

text-align:left;text-indent:-18.0pt">1.    
Stamatopoulos, K., Development of a Biorelevant Dynamic Model of Human
Proximal Colon: a Tool for Designing Colon-Specific Drug Delivery Systems, in
School of Chemical Engineering. 2016, University of Birmingham: Birmingham.

text-align:left;text-indent:-18.0pt">2.    
Spratt, P., Nicolella, C. & Pyle, D. L., An Engineering Model of the
Human Colon. Food and Bioproducts Processing, 2005. 83: 147-157.

text-align:left;text-indent:-18.0pt">3.    
Venema, K., The TNO In Vitro Model of the Colon (TIM-2). Verhoeckx, K.,
Cotter, P., Lópezexpósito, I., Kleiveland, C., Lea, T., Mackie, A., Requena,
T., Swiatecka, D. & Wichers, H. (eds.) The Impact of Food Bioactives on
Health: in vitro and ex vivo models. 2015, Cham: Springer International
Publishing.

text-align:left;text-indent:-18.0pt">4.    
Sarna, S. K. & Shi, X-Z. Chapter 39 - Function and Regulation of
Colonic Contractions in Health and Disease A2 - Johnson, Leonard R. Physiology
of the Gastrointestinal Tract (Fourth Edition). 2006, Burlington: Academic
Press.

text-align:left;text-indent:-18.0pt">5.    
Hoad, C.L., et al., Colon wall motility: comparison of novel
quantitative semi-automatic measurements using cine MRI. Neurogastroenterology
and Motility, 2016. 28(3): p. 327-335.

text-align:left;text-indent:-18.0pt">6.    
Pritchard, S.E., et al., Assessment of motion of colonic contents in the
human colon using MRI tagging. Neurogastroenterology and Motility, 2017. 29(9).

text-align:left;text-indent:-18.0pt">7.    
Munjeri, O., Collett, J. H., Fell, J. T., Sharma, H. L., Smith, A-M., In
vivo behaviour of hydrogel beads based on amidated pectins. Drug Deliv. J.
Deliv.Targeting Therap. Agents, 1998. 5: p. 239-241

text-align:left;text-indent:-18.0pt">8.    
Cook, I. J., Furukawa, Y., Panagopoulos, V., Collins, P. J. & Dent,
J., Relationships between spatial patterns of colonic pressure and individual
movements of content. Am J Physiol Gastrointest Liver Physiol, 2000. 278: p.
G329-41.

text-align:left;text-indent:-18.0pt">9.    
Dinning, P. G., Szczesniak, M. M. & Cook, I. J. 2008. Proximal
colonic propagating pressure waves sequences and their relationship with
movements of content in the proximal human colon. Neurogastroenterol Motil,
2000. 20: 512-20.