- Email: [email protected]
Preparation and characterization of gastrointestinal wafer formulations
Accepted Manuscript Title: Preparation and characterization of gastrointestinal wafer formulations Author: Kirsten Kirsch Ulrike Hanke Werner Weitschies PII: DOI: Reference:
S0378-5173(17)30135-7 http://dx.doi.org/doi:10.1016/j.ijpharm.2017.02.045 IJP 16449
To appear in:
International Journal of Pharmaceutics
Received date: Revised date: Accepted date:
28-11-2016 14-2-2017 17-2-2017
Please cite this article as: Kirsch, K., Hanke, U., Weitschies, W.,Preparation and characterization of gastrointestinal wafer formulations, International Journal of Pharmaceutics (2017), http://dx.doi.org/10.1016/j.ijpharm.2017.02.045 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Preparation and characterization of gastrointestinal wafer formulations
2 3 4
Kirsten Kirsch, Ulrike Hanke, Werner Weitschies1
5 Universitiy of Greifswald, Center of Drug Absorption and Transport, Institute of Pharmacy, Felix-HausdorffStrasse 3, 17487 Greifswald, Germany
ip t
6 7 8
cr
9 10
1
11
[email protected]
12
phone: +49 3834 4204813
13
Abstract
14
Many active pharmaceutical ingredients (API) have a very poor or highly variable bioavailability after oral administration. One
15
possibility to overcome this problem might be found in the application of mucoadhesive dosage forms like gastrointestinal
16
wafers. However, a currently unsolved challenge is the control of the adhesion of the wafer to the intestinal mucus. One
17
suggested solution might be the combination of gastrointestinal wafers and expanding systems. Such a combination requires
18
thin and elastic wafers which are further characterized by an unidirectional drug release. In this study gastrointestinal,
19
twolayered wafers containing a water-insoluble backing layer and a drug-loaded, mucoadhesive layer were fabricated by casting
20
solvent technique. The backing layer consists of Ethocel Standard 10 Premium and the mucoadhesive layer was prepared using
21
a mixture of Methocel E15 Premium LV, polyvinyl alcohol and Macrogol 400. The wafers were characterized regarding their
22
appearance, mechanical properties and dissolution profiles as well as the influence of backing layer thickness on drug transfer
23
and their ability of unidirectional drug release. The wafers with backing layer thickness of 500░µg Ethocel/cm presented
24
adequate mechanical properties, a drug transfer about 73% and unidirectional drug release.
Ac ce pt e
d
M
an
us
Corresponding author
25 26
Keywords
27
Gastrointestinal wafer; Mucoadhesion; Unidirectional drug release; Methocel E15LV
2
28 29 30 1 Page 1 of 24
31
1 Introduction
32 Many active pharmaceutical ingredients (API) have a very poor and/or highly variable bioavailability after
34
oral administration. Reasons are for example low mucosal permeability, a narrow absorption window at
35
particular regions of the gastrointestinal tract (GIT), variable transit times, various fluid volumes, lack of
36
stability in the gastrointestinal environment resulting in a decomposition prior to its absorption and low
37
concentration of API in gastrointestinal contents (Bhasakaran et al., 2012; Dressman and Reppas, 2000;
38
Hens et al., 2016, Koziolek et al., 2015, Tao and Desai, 2005). Additionally, physiological properties are of
39
relevance. For example mucus thickness ranges from 50–450 µm (median 200 µm) and is influenced by
40
hormonal, paracrine and neural stimulation as well as by inflammatory reactions and acids (Allen et al.,
41
1993; Khutoryanskiy, 2011). One strategy to overcome these problems is the usage of mucoadhesive
42
dosage forms like intestinal wafers. Wafers are defined by the U.S. Food and Drug Administration (FDA)
43
(2009) as “a thin slice of material containing a medicinal agent”. Due to their drug release rates and
44
disintegration times, wafers can be classified into rapid disintegrating, meltaway and sustained release
45
wafers. Rapid disintegrating wafers disintegrate within 30-60 s and result in immediately drug release,
46
whereas meltaway wafers stick to the mucosa, disintegrate within 5-30 min and form a gelatinous,
47
mucoadhesive depot at application site. Sustained release wafers are characterized by disintegration times
48
of several hours and a continuous drug release, ideally zero order kinetics (LTS Lohmann Therapie-
49
Systeme, 2010). After swallowing intestinal wafers have the potential to adhere to gastrointestinal (GI)
50
mucosa because of their mucoadhesive properties. Due to the close contact between wafer and mucosa, a
51
high drug concentration gradient is created, resulting in a high drug flux at the absorbing tissue, which is
52
well supplied with blood. These conditions support presumably drug absorption into systemic circulation
53
and enhance oral bioavailability (Andrews et al., 2009; Bernkop-Schnürch, 2005; Boddupalli et al., 2010).
54
However, a challenge is the loss of control over the dosage form after swallowing. It cannot be guaranteed
55
that the wafers adhere in the intended region of the GIT and in the desired way. Especially using
56
multilayered wafer, it cannot be influenced which side of the wafer adhere to the mucus layer and the
57
underlaying epithelial layer. One suggested solution might be the combination of intestinal wafers with
58
expanding systems which can control adhesion process. Such expanding systems are described in the
Ac ce pt e
d
M
an
us
cr
ip t
33
2 Page 2 of 24
patent of Bogdahn et al. (2015) and consist of a shell, an expansion mechanism and a wafer. The shell
60
could be a custom-designed, gastroresistent capsule, which were swallowed and release the wafer after
61
triggering by pH value, pressure or a composition of a fluid surrounding the shell. The expansion
62
mechanism is selected from the group comprising mechanical expansion system, gas driven expansion
63
system, compressed foam or compressed tissue. The wafer is packed in the shell for example lumped
64
together, collapsed, folded or rolled (Bogdahn et al., 2015). The wafers need specific properties for
65
combination with expanding systems. They have to be thin, elastic and folding resistant. Furthermore, an
66
unidirectional drug release profile is required. The aim of this study was to prepare and characterize rapid
67
disintegrating intestinal wafers which can be combined with an expanding system and are characterized by
68
an unidirectional drug release profile.
us
cr
ip t
59
an
69 70 2 Material und methods
M
71 72
2.1 Formulation of wafers
74
Wafers were produced by a casting solvent technique and consisted of a water-insoluble backing layer of
75
EthocelTM Standard 10 Premium (ethyl cellulose, EC) (Colorcon Limited, United Kingdom) and a
76
drug-loaded, mucoadhesive layer.
77
Firstly, the backing layer was prepared by spraying a solution of 4% (w/w) EC in acetone on the release
78
liner according to a defined spraying scheme. Acetone was evaporated by room temperature for 15 min.
79
Polyethylene paper (Polyslik® 111/105, Loparex, Netherlands) was used as release liner. The thickness of
80
the backing layer was expressed as amount of EC per area. It was adjusted to 0-750 µg EC/cm² and was
81
controlled by weighing.
82
Secondly, the drug-loaded, mucoadhesive layer was fabricated. The most suitable formulation was
83
determined
84
MethocelTM E15 Premium LV (hydroxypropyl methylcellulose, HPMC) (DOW Chemical Company, USA),
85
polyvinyl alcohol, partially hydrolyzed (MW approx. 200000) (PVA) (Merck Schuchardt OHG, Germany) and
86
Macrogol 400 (polyethylene glycol, PEG400) (Fagron GmbH & Co.KG, Germany) were produced, whereby
Ac ce pt e
d
73
in
preliminary
tests
(data
not
shown).
For
this
purpose,
different
mixtures
of
3 Page 3 of 24
87
the ratio of one ingredient at a time varied. Produced formulations were tested regarding their disintegration
88
time,
89
MethocelTM E15 Premium LV, PVA and PEG400 with a ratio of 1:2:4 was chosen as most suitable
90
formulation for drug-loaded, mucoadhesive layer. In this study fluorescein sodium (FL) (Fluka Analytical,
91
Germany), quinine anhydrous (QN) (Sigma-Aldrich Chemie GmbH, Germany) and diclofenac sodium
92
(Diclo) (Fagron GmbH & Co.KG, Germany) were used as model drug substances. These substances were
93
chosen because of their different hydrophilicity/lipophilicity and various solubility in aqueous media. The
94
final drug concentration in each wafer was 5 µg/cm² for fluorescein (FL), 100 µg/cm² for quinine (QN) and
95
500 µg/cm² for diclofenac (Diclo). Additionally, placebo wafers were produced. The compositions of all
96
formulations are summarized in Table 1. The polymer mixture was kept overnight and centrifuged by
97
4400 rpm for 50 min to remove all entrapped air bubbles. Then the mixture was cast onto the dried backing
98
layer using a mechanical film casting apparatus equipped with a vacuum suction plate and 300 µm film
99
applicator frame (film applicator CX4, mtv messtechnik OHG, Germany). Casting speed was adjusted to
100
30 mm/s. The casted mixture was dried at 40 °C for 6 h and stored on release liner packed in aluminum foil
101
at room temperature. The resulting polymer film was cut into smaller pieces and peeled off the release liner
102
before usage.
elongation
at
break
and
folding
endurance.
A
mixture
of
d
M
an
us
cr
ip t
strength,
Ac ce pt e
103
tensile
104
2.2 Wafer characterization
105
2.2.1 Appearance
106
The surface uniformity of the produced wafers was visually inspected. It was rated whether the surface was
107
homogenous, smooth, and free of holes and air pockets. Additionally, scanning electron microscopy (SEM)
108
(PhenomTM, FEI CompanyTM, L.O.T.-Oriel GmbH & Co.KG, Germany) was used to observe surface
109
morphology of a placebo wafer with a backing layer thickness of 500 µg EC/cm².
110
The wafer thickness was measured by a mechanical thickness dial gauge (0.01 mm capacity, Kaefer
111
Messuhrenfabrik GmbH & Co.KG, Germany). The wafer (size 2.5 x 4 cm) was placed between to flat
112
contact points and the thickness was read on the analog display. For each formulation the thickness of
113
three wafers was measured on three defined spots and the average was calculated.
4 Page 4 of 24
114
Finally, the mass of the wafers (size 2.5 x 4 cm) was determined using a digital balance (Sartorius GmbH,
115
Germany).
116 2.2.2 Drug content uniformity
118
The model drug substance distribution in the produced wafers was measured to ensure uniformity. Ten
119
samples (size 1 x 1 cm) were collected randomly from each formulation and dissolved by stirring in 10 mL
120
distilled water using a magnetic stirrer. After complete dissolution of the drug-loaded, mucoadhesive layer of
121
the wafer, samples were measured by fluorescence spectroscopy (Varioskan Flash, Thermo Fisher
122
Scientific Germany BV & Co.KG, Germany) (FL λex 490 nm, λem 513 nm and QN λex 347 nm, λem 373 nm) or
123
UV/VIS-spectroscopy (Cary 50 Scan, Varian, Inc., Germany) (Diclo 276 nm) against calibration in the same
124
medium. Wafers passed content uniformity test if they met requirements of the European
125
Pharmacopoeia 8.8 (Ph.Eur. 8.8) chapter 2.09.06 content uniformity.
an
us
cr
ip t
117
M
126 2.2.3 Mechanical properties
128
Mechanical properties include folding endurance, tensile strength and elongation at break. Each parameter
129
was determined using three random samples.
130
Folding endurance was tested manually. The wafer was folded repeatedly at the same place until breaking
131
and the number of folds was counted. Wafers which could be folded for more than 100times without
132
breaking passed the folding endurance test.
133
Tensile strength σ (MPa) and elongation at break ε (%) were evaluated using a texture analyzer (TAplus,
134
Lloyd Instruments an AMETEK Company, Germany) connected to a data acquisition and analysis software
135
(Nexygen Plus 3.0 Software, AMETEK Company, Germany). The wafer (size 2.5 x 6 cm) was placed
136
between two clamps positioned 5 cm apart. The lower clamp was stationary and the upper clamp was
137
moved at a rate of 1 mm/s to a distance of 200 mm. The force required to break (F, N) the wafer and the
138
elongation at breaking point (l, mm) were measured. Tensile strength was calculated using Formula 1,
139
where A (mm²) is the cross-section of the wafer and elongation at break was calculated using Formula 2,
140
where l0 (mm) is the original length of the wafer.
Ac ce pt e
d
127
141 5 Page 5 of 24
142
σ (MPa) = F (N)/A (mm²)
Formula 1
ε (%) = l (mm)/l0 (mm) · 100 (%)
Formula 2
143 144 145 2.3 Disintegration of mucoadhesive layer
147
The disintegration time of the mucoadhesive layer only of the placebo wafers (size 2.5 x 4 cm) was
148
determined either in 10 mL distilled water or on mucosa simulating alginate gel film (size 5 x 7 cm). The
149
alginate gel film was prepared from an aqueous sodium alginate solution (3% (w/w)) (Fagron GmbH & Co.
150
KG, Germany) and gelled for 10 min with aqueous calcium chloride solution (6% (w/w)) (AppliChem GmbH,
151
Germany). Alginate gel film was chosen because human mucosa is covered by mucus, which is a
152
viscoelastic gel and contains 90-98% water. The used alginate film contained 97% water and had
153
viscoelastic properties as well. Further, alginate gel films and blocks are often used models to simulate
154
tissue surfaces (Ahearne et al., 2005; Neubert, 2009). The endpoint of the disintegration was inspected
155
visually and was defined referring to Ph.Eur. 8.8 chapter 2.09.02 disintegration of suppositories. The
156
endpoint was defined as the time point where the softening of the wafer occurred accompanied by
157
appreciable change of shape and the softening was such that the wafer no longer had a solid core. All tests
158
were performed in triplicate.
cr
us
an
M
d
Ac ce pt e
159
ip t
146
160
2.4 Investigation of the influence of the backing layer thickness on drug transfer
161
The influence of the backing layer thickness on drug transfer was investigated using an in vitro model
162
(Figure 1). The model consists of a tube with a diameter of 3.5 cm which was coated with a mucosa
163
simulating alginate gel film as acceptor compartment. The alginate gel film was made of an aqueous
164
sodium alginate solution (3% (w/w)) and gelled for 10 min with aqueous calcium chloride solution
165
(6% (w/w)). A small balloon was used as expanding system in order to initiate the contact between wafer
166
and simulated mucosa during the contact time with defined pressure. The evaluated wafer had a size of
167
1 x 11.5 cm (width x length) in order to fit to tube perimeter and balloon size. At the beginning of the
168
experiment, the wafer was fixed on the balloon and both were slid into the tube. Then the balloon was
169
expanded by compressed air and by this way the wafer was pressed on the alginate gel film. After defined 6 Page 6 of 24
period of time the balloon was collapsed and pulled out of the tube. At the end of the experiment, the
171
amount of model drug substance which was transmitted to the alginate gel film, was remaining on the
172
balloon and was adhering to the tube was determined. For this purpose, the mucosa simulating alginate gel
173
film was dissolved in 50 mL of five times concentrated phosphate buffer pH 6.8 (USP), the balloon was
174
incubated in 25 mL of five times concentrated phosphate buffer pH 6.8 (USP) and the tube was rinsed three
175
times with five times concentrated phosphate buffer pH 6.8 (USP). Aliquots were measured by fluorescence
176
spectroscopy (FL λex 490 nm, λem 513 nm and QN λex 347 nm, λem 373 nm) or UV/VIS-spectroscopy (Diclo
177
276 nm) against calibration in the same medium. The test parameters were set to a contact time of 5 min
178
and a pressure of 0.1 bar. The alginate gel film was casted using a 1000 µm film applicator frame. All tests
179
were carried out sixfold and significance of influence of backing layer thickness was verified by ANOVA (
180
= 0.05).
an
us
cr
ip t
170
181
M
182 2.5 Dissolution test
184
Dissolution tests of wafers with a backing layer thickness of 500 µg EC/cm² (size 24 cm²) were performed
185
using paddle apparatus (DT80, Erweka, Germany) (75 rpm) with 750 mL 0.07 M phosphate buffer pH 6.8
186
(USP) at 37 °C. Aliquots were taken over a period of 60 min, the withdrawn volume was replaced by fresh
187
buffer and the model drug substance concentrations were determined either by using fluorescence
188
spectroscopy (FL λex 490 nm, λem 513 nm and QN λex 347 nm, λem 373 nm) or UV/VIS-spectroscopy (Diclo
189
276 nm) against a calibration in the same medium. All tests were carried out in triplicate.
Ac ce pt e
190
d
183
191
2.6 Unidirectional drug release
192
The ability of the wafers to improve unidirectional drug release was studied using an Ussing chamber
193
(EasyMount 6-channal Ussing Chamber System, Physiologic Instruments Inc., USA). The tested wafers
194
(size 1 cm²) had a backing layer thickness of 500 µg EC/cm². The wafers were mounted between both
195
compartments of the Ussing chamber, thereby the backing layer was facing compartment 1 (donor
196
compartment) and the drug-loaded, mucoadhesive layer was facing compartment 2 (acceptor
197
compartment). Subsequently, both compartments were filled with modified Ringer buffer pH 7.4. The buffer 7 Page 7 of 24
was prepared with the following composition using distilled water as solvent: 1.60 g/L sodium
199
hydrogencarbonate (NaHCO3) (AppliChem GmbH, Germany), 0.06 g/L sodium dihydrogenphosphate
200
(NaH2PO4) (Sigma Aldrich, Germany), 6.50 g/L sodium chloride (NaCl) (AppliChem GmbH, Germany),
201
0.40 g/L potassium chloride (KCl) (AppliChem GmbH, Germany), 0.23 g/L calcium chloride (CaCl2)
202
(AppliChem GmbH, Germany), 0.23 g/L magnesium chloride (MgCl2) (AppliChem GmbH, Germany),
203
0.14 g/L di-sodium hydrogenphosphate (Na2HPO4) (AppliChem GmbH, Germany) and 1.80 g/L glucose
204
anhydrous (AppliChem GmbH, Germany). The Ussing chamber was incubated at 37 °C and the medium
205
was constantly moved by gas bubbles (Carbogen, Air Liquide Deutschland GmbH, Germany). Over a
206
period of 2 h 200 µL aliquots were taken from both compartments, withdrawn volumes were replaced by
207
fresh buffer and model drug substance concentrations were measured either using fluorescence
208
spectroscopy (FL λex 490 nm, λem 513 nm and QN λex 347 nm, λem 373 nm) or UV/VIS-spectroscopy (Diclo
209
276 nm) against calibration in the same medium. The percentage of the model drug substance distribution
210
in the donor and acceptor compartment was calculated. Thereby, the leakage from the backing layer and
211
the drug release from the drug-loaded, mucoadhesive layer were determined. All tests were performed at
212
least in triplicate.
d
M
an
us
cr
ip t
198
214 215 216
3. Results
Ac ce pt e
213
217
3.1 Wafer characterization
218
3.1.1 Appearance
219
All wafers had a smooth, homogenous and air bubble free surface as it is demonstrated in Figure 2. The
220
results of the thickness and weight measurements are summarized in Table 2. The weight and the
221
thickness of the placebo and drug-loaded wafers increased with increasing thickness of the backing layer.
222
The total weight of the placebo wafers increased from 7.4 mg/cm² to 8.2 mg/cm² and the total thickness
223
from 88 µm to 95 µm.
224 225
3.1.2 Drug content uniformity 8 Page 8 of 24
226
The average content of ten samples from the polymer film of each wafer formulation was between 90% and
227
110% of declared content and no sample was between 75% and 125% of the declared content.
228
Accordingly, all investigated wafers passed the test of content uniformity.
229 3.1.3 Mechanical properties
231
Placebo and drug-loaded wafers were tested regarding their tensile strength, elongation at break and
232
folding endurance. The results are presented in Table 2. The placebo wafers had a tensile strength
233
between 2.4 MPa and 3.1 MPa. The elongation at break decreased from 63.8% to 8.3% with increasing
234
backing layer thickness. Wafers containing FL presented an increased tensile strength, whereas wafers
235
containing QN or Diclo showed a decreased tensile strength compared to placebo wafers. FL wafers also
236
exhibited a decreasing elongation at break from 91.3% to 6.9% with increasing backing layer thickness. For
237
QN wafers and Diclo wafers no tendency depending on backing layer thickness could be observed.
238
Furthermore, all tested formulations presented a folding endurance over 100.
M
an
us
cr
ip t
230
239 3.2 Disintegration of mucoadhesive layer
241
The mucoadhesive layer of the placebo wafers disintegrated in 10 mL of distilled water in 63 ± 6 sec
242
(MW ± SD, n = 3) and on mucosa simulating alginate gel film within 65 ± 5 sec (MW ± SD, n = 3).
Ac ce pt e
243
d
240
244
3.3 Investigation of the influence of the backing layer thickness on drug transfer
245
The influence of the backing layer thickness on drug transfer from wafer to mucosa simulating alginate gel
246
film was determined, whereby thickness varied between 0 and 750 µg EC/cm². With increasing backing
247
layer thickness, the model drug substance amount which was transferred to the simulated mucosa and to
248
the tube increased about 35.9% (FL), 38.7% (QN) or 27.8% (Diclo) comparing wafers with a backing layer
249
thickness of 0 µg EC/cm² and 750 µg EC/cm² (Figure 3). The increase of transmitted drug substance
250
amount with increasing backing layer thickness was significant for all tested substances.
251 252
3.4 Dissolution test
9 Page 9 of 24
253
In Figure 4 the release profiles of drug-loaded wafers with a backing layer thickness of 500 µg EC/cm² are
254
shown. Wafers loaded with FL showed complete drug release within 30 min, whereby only 87% of declared
255
drug amount was recovered. QN wafers presented complete drug release of 98% after 7.5 min and Diclo
256
wafers presented 99% drug release within 15 min.
257 3.5 Unidirectional drug release
259
The ability of wafers to improve unidirectional drug transport was determined using an Ussing chamber
260
mounting the wafer between both compartments. Percentage of the model drug substance distribution
261
between donor and acceptor compartment was calculated and is presented in Figure 5. The investigated
262
wafers showed a rapid model drug substance accumulation in the acceptor compartment, followed by a
263
slow increase of the model drug substance concentration in the donor compartment and a slow decrease in
264
the acceptor compartment after 30 to 60 min.
an
us
cr
ip t
258
M
265 266 4 Discussion
d
267
Ac ce pt e
268 269
The produced wafers were intended to be flexible, rapidly disintegrating and to provide an unidirectional
270
drug release profile (immediate release wafers). For this purpose, twolayered wafers consisting of a
271
water-insoluble backing layer and a drug-loaded, mucoadhesive layer were developed.
272
The backing layer is intended to act as a barrier between the drug-loaded layer and the intestinal luminal
273
fluid in vivo. For the production EthocelTM Standard 10 Premium was used which is an EC containing an
274
ethoxyl content of 48.0 to 49.5% (The Dow Chemical Company, 2005). EC was selected because it is
275
insoluble in water and literature data confirm that it prevents drug release from the backing layer side of
276
wafers and promotes unidirectional drug release (Gupta et al., 2013; Kalavadia et al., 2014; Toorisaka et
277
al., 2012; Whitehead et al., 2004).
278
The drug-loaded, mucoadhesive layer should promote a high drug concentration gradient at intestinal
279
mucosa. The polymers used for production should be mucoadhesive, film forming, biocompatible and
280
non-toxic. The polymer selection based on theories of mucoadhesion which claim that mucoadhesive 10 Page 10 of 24
polymers are hydrophilic networks containing several polar, non-charged and/or non-ionic, hydrogen bond
282
forming, functional groups (Andrews et al., 2009; Boddupalli et al., 2010; Dodou et al., 2005; Khutoryanskiy,
283
2011; Shaikh et al., 2011). Because of these MethocelTM E15 Premium LV and PVA were used for the
284
fabrication of the drug-loaded, mucoadhesive layer. Additionally, PEG400 was applied in this layer because
285
it is as good plasticizer (Honary and Orafai, 2002; Saringat et al., 2005).
286
The produced twolayered wafers were characterized regarding their appearance and mechanical
287
properties. All wafers had a homogenous and smooth surface. The weight and the total thickness of the
288
wafers increased with increasing backing layer thickness. The tensile strength increased with increasing
289
backing layer thickness as well. This is on good accordance to the observations reported by Bhasakaran et
290
al. (2012). One reason could be that the surface connection between the backing layer and the
291
mucoadhesive layer build up a composite material which has different properties than the single layers. The
292
mucoadhesive layer remains unchanged, but the backing layer becomes more resistant to mechanical
293
stress with increasing thickness. Consequently, the entire twolayered wafer becomes more resistant to
294
mechanical stress. Furthermore, the integration of model drug substances resulted in an increasing weight
295
and total thickness and influenced mechanical properties. The tensile strength decreased and the
296
elongation at break increased with incorporation of FL. Wafers containing QN and Diclo as model drug
297
substances presented no change of the mechanical properties. A reason could be the solubility depending
298
distribution of the model drug substances in the wafer matrix. All polymers used for the wafer matrix were
299
hydrophilic. The hydrophilic FL (LogP 0.67) (PubChem; Sigma Aldrich Chemie GmbH, 2011) is probably
300
dissolved in the polymeric wafer matrix and intercalates between the polymer chains. So, it acts as a
301
plasticizer. In contrast, QN (logP 3.44) (DrugBank Version 4.3) and Diclo (logP 1.27) (Smith, 2014) are
302
more likely suspended in the matrix and have no effect on polymer chain formation.
cr
us
an
M
d
Ac ce pt e
303
ip t
281
304
To determine the optimal backing layer thickness drug transfer experiments with an in vitro model were
305
conducted (Figure 1). For this purpose, the drug transfer from the wafer to a mucosa simulating alginate gel
306
film was measured. The amount of model drug substance transferred to the alginate gel increased with
307
increasing backing layer thickness. One reason for this observation could be that the drug-loaded,
308
mucoadhesive layer disintegrates and dissolves during the contact time, whereas the water-insoluble 11 Page 11 of 24
backing layer remains unchanged and intact. As a result, the backing layer minimizes the contact between
310
the model drug substance and the expanding system and supports the drug transfer to the mucosa
311
simulating alginate gel film. Further barrier function increased with increasing thickness of the backing layer.
312
The residual amounts on the expander essentially resulted from the fact that the disintegrating wafer
313
spreads over the boundary of the backing layer. With respect to the obtained results a backing layer
314
thickness of 500 µg EC/cm² was selected for all following experiments.
315
ip t
309
lnvestigated wafers showed a complete release of the incorporated model drug substances within 30 min.
317
However, FL loaded wafers provided a prolonged release in comparison to wafers containing QN and Diclo.
318
One explanation might be found again in the solubility depending distribution of the model drug substance
319
within the matrix. Because of these, the diffusion rate of FL from the dosage form into the medium was the
320
rate limitating step. By QN and Diclo, the erosion rate of the polymer matrix was the rate limitating step (The
321
Dow Chemical Company, 2000). Another explanation could be that FL penetrate into backing layer because
322
of its solubility, interacts with the EC and retained there. Additionally, the wafers partially curled up after
323
medium contact and the backing layer was facing the medium. This resulted in a reduced contact area
324
between the drug-loaded layer and the medium and as a consequence in a prolonged release.
us
an
M
d
Ac ce pt e
325
cr
316
326
Finally, the unidirectional drug release behavior of wafers with a backing layer thickness of 500 µg EC/cm²
327
was determined using the Ussing chamber. Over a period of 2 h aliquots were taken from both
328
compartments and leakage from the backing layer and the drug transfer from the mucoadhesive layer were
329
determined. The investigated wafers provided a rapid model drug substance transfer into the acceptor
330
compartment. After 30 to 60 min, a slow increase of the model drug substance concentration in the donor
331
compartment followed due to redistribution by diffusion through the backing layer. In conclusion, the
332
backing layer prevents drug release in the donor compartment and promotes unidirectional drug transport.
333 334 335
5 Conclusion
336 12 Page 12 of 24
The aim of this study was to develop a rapidly disintegrating intestinal wafer which can be combined with an
338
expanding system and is characterized by an unidirectional drug release. The developed wafers were
339
characterized regarding their appearance, mechanical properties, dissolution profiles as well as their drug
340
transfer and drug release properties. Wafers with a backing layer thickness of 500 µg EC/cm² exhibited
341
adequate mechanical properties, a drug transfer of about 73% to a mucosa simulating alginate gel and
342
unidirectional drug release. In combination with expanding systems such wafer can enhance drug
343
absorption and enable oral intake of drugs with poor oral bioavailability. However, in subsequent
344
experiments more biorelevant in vitro as well as in vivo studies should be developed to proof this concept.
cr
ip t
337
us
345
347
Conflict of interest
348
The authors report no conflict of interest.
an
346
M
349 350
d
352
References
Ac ce pt e
351
353 354 355
Ahearne, M., Yang, Y., El Haj, A. J., Then, K. Y., & Liu, K.-K. (2005). Characterizing the viscoelastic properties of thin hydrogel-based constructs for tissue engineering application. Journal of the Royal Society Interface, 2, S. 455463. doi:10.1098/rsif.2005.0065
356 357
Allen, A., Flenström, G., Garner, A., & Kivilaakso, E. (1993). Gastroduodenal Mucosal Protection. Physiological Reviews, 73(4), S. 823-857.
358 359
Andrews, G. P., Laverty, T. P., & Jones, D. S. (2009). Mucoadhesive polymeric platforms for controlled drug delivery. European Journal of Pharmaceutics and Biopharmaceutics, 71, pp. 505-518.
360 361
Bernkop-Schnürch, A. (2005). Mucoadhesive systems in oral drug delivery. Drug Discovery Today: Technologies, 2(1), pp. 83-87.
362 363 364
Bhasakaran, S., Moris, S., & Rout, A. (2012). Gastrointestinal Mucoadhesive Patch System for Oral Administration of Metronidazole. International Journal of Research in Pharmaceutical and Biomedical Sciences, 3(4), pp. 14971505.
365 366
Boddupalli, B. M., Mohammed, Z. N., Nath, R. A., & Banji, D. (2010). Mucoadhesive drug delivery system: An overview. Journal of Advanced Pharmaceutical Technology & Research, 4, pp. 381-387.
367
Bogdahn, M., Kirsch, K., Grimm, M., Koziolek, M., & Weitschies, W. (22. 12 2015). Patentnr. PCT/EP2015/002601. 13 Page 13 of 24
Dodou, D., Breedveld, P., & Wieringa, P. A. (2005). Mucoadhesives in the gastrointestinal tract: revisiting the literature for novel applications. European Journal of Pharmaceutics and Biopharmaceutics, pp. 1-16.
370 371
Dressman, J. B., & Reppas, C. (2000). In vitro-in vivo correlations for lipophilic, poorly water-soluble drugs. European Journal of Pharmaceutical Sciences, 11(Suppl. 2), pp. S73-S80.
372 373
DrugBank Version 4.3. (n.d.). LogP quinine anhydrous. Retrieved 12 01, 2015, from http://www.drugbank.ca/drugs/DB00468
374 375
Grabovac, V., Föger, F., & Bernkop-Schnürch, A. (2008). Design and in vivo evaluation of a patch delivery system for insulin based on thiolated polymers. International Journal of Pharmaceutics, 348, pp. 169-174.
376 377
Gupta, V., Hwang, B. H., Lee, J., Anselmo, A. C., Doshi, N., & Mitragotri, S. (2013). Mucoadhesive intestinal devices for oral delivery of salmon calcitonin. Journal of Controlled Release, 172, pp. 753-762.
378 379 380
Hens, B., Corsetti, M., Spiller R, Marciani, L., Vanuytsel, T., Tack, J., . . . Augustijns, P. (2016). Exploring Gastrointestinal Variables Affecting Drug and Formulation Behavior: Methodologies, Challenges and Opportunities. International Journal of Pharmaceutics. doi:10.1016/j.ijpharm.2016.11.063
381 382 383
Honary, S., & Orafai, H. (2002). The Effect of Different Plasticizer Molecular Weights and Concentrations on Mechanical and Thermomechanical Properties of Free Films. Drug Development and Industrial Pharmacy, 28(6), pp. 711-715.
384 385 386
Kalavadia, S., Dash, R. P., Misra, M., & Nivsarkar, M. (2014). Design and in vivo evaluation of gastrointestinal mucoadhesive patch system (GMPS) loaded with chitosan nanoparticles. International Journal of Pharmaceutical Development & Technology, 4(4), pp. 258-266.
387 388
Khutoryanskiy, V. V. (2011). Advances in Mucoadhesion and Mucoadhesive Polymers. Macromolecular Bioscience, 11, pp. 748-764.
389 390 391
Koziolek, M., Grimm, M., Becker, D., Iordanov, V., Zou, H., Shimizu, J., . . . Weitschies, W. (2015). Investigation of pH and Temperature Profiles in the GI Tract of Fasted Human Subjects Using the Intellicap(R) System. Journal of Pharmaceutical Sciences, 104(9), pp. 2855-2863.
392 393
LTS Lohmann Therapie-Systeme. (2010). Your active ingredient in its most appealing form. Retrieved 01 15, 2013, from www.lts-corp.com
394 395
Neubert, A. (2009). Entwicklung eines In vitro-Modells zur Untersuchung der Wirkstofffreisetzung und -verteilung aus Arzneistoff-freisetzenden Stents. Ernst-Moritz-Arndt-Universität Greifswald, Greifswald.
396 397
PubChem. (n.d.). LogP Fluorescein Sodium. Retrieved 11 28, 2014, from http://pubchem.ncbi.nlm.nih.gov/compound/10608'sectionVapor-Pressur
398 399 400
Saringat, H. B., Alfadol, K. I., & Khan, G. M. (2005). The influence of different plasticizers on some physical and mechanical properties of hydroxypropyl methylcellulose free films. Pakistan Journal of Pharmaceutical Sciences, 18(3), S. 25-38.
401 402
Shaikh, R., Singh, T. R., Garland, M. J., Woolfson, A. D., & Donnelly, R. F. (2011). Mucoadhesive drug delivery systems. Journal of Pharmacy and Bioallied Sciences, 3(1), pp. 89-100.
Ac ce pt e
d
M
an
us
cr
ip t
368 369
14 Page 14 of 24
Shen, Z., & Mitragotri, S. (2002). Intestinal Patches for Oral Drug Delivery. Pharmaceutical Research, 19(4), S. 391395.
405
Sigma Aldrich Chemie GmbH. (2011). Sicherheitsdatenblatt Fluorescein-Natrium.
406 407
Smith, H. (2014). Formulation, in vitro release and transdermal diffusion of diclofenac salts by implementation of the delivery gap principle. Retrieved 12 01, 2015, from http://dspace.nwu.ac.za/handle/10384/12003
408 409
Tao, S. L., & Desai, T. A. (2005). Gastrointestinal patch systems for oral drug delivery. Drug Discovery Today, 10(13), pp. 909-915.
410 411
Teutonico, D., Montanari, S., & Ponchel, G. (2011). Concentration and surface of absorption: Concepts and applications to gastrointestinal patches delivery. International Journal of Pharmaceutics, 413, pp. 87-92.
412 413
The Dow Chemical Company. (2000). Using METHOCEL Cellulose Ethers for Controlled Release of Drugs in Hydrophilic Matrix Systems. 1-36.
414
The Dow Chemical Company. (2005). ETHOCEL Ethylcellulose Polymers Technical Handbook. 1-28.
415 416
The European Directorate for the Quality of Medicines & Health Care. (2016). PhEur - European Pharmacopoeia 8.8. Retrieved from http://online6.edqm.eu/ep805/
417 418
Toorisaka, E., Watanabe, K., Ono, H., Hirata, M., Kamiya, N., & Goto, M. (2012). Intestinal patches with an immobilized solid-in-oil formulation for oral protein delivery. Acta Biomaterialia, 8, pp. 653-658.
419 420 421 422
U.S. Food and Drug Administration. (2009). Data Standards Manual (monographs) - Dosage Form. Retrieved 07 21, 2016, from http://www.fda.gov/Drugs/DevelopmentApprovalProcess/FormsSubmissionRequirements/ElectronicSubmis sions/DataStandardsManualmonographs/ucm071666.htm
423 424
Whitehead, K., Shen, Z., & Mitragotri, S. (2004). Oral delivery of macromolecules using intestinal patches: applications for insulin delivery. Journal of controlled release, 98, pp. 37-45.
426 427 428
cr
us
an
M
d
Ac ce pt e
425
ip t
403 404
Figure captions
429
Figure 1: Schematic overview of in vitro model for measuring drug transfer from wafers to mucosa simulating alginate
430
gel film as acceptor compartment consisting of a tube (diameter 3.5 cm), mucosa simulating alginate gel film and a
431
balloon that can be expanded with compressed air.
432 433
Figure 2: Scanning electron microscopic (SEM) image at the edge (A) and from above (B, C) of a placebo wafer
434
consisting of a water-insoluble backing layer (C) of 500 µg ethyl cellulose/cm² and a mucoadhesive layer (B), which
15 Page 15 of 24
TM
435
was fabricated using a mixture of Methocel
436
1:2:4.
E15 Premium LV, polyvinyl alcohol and Macrogol 400 with a ratio of
437 Figure 3: Influence of backing layer thickness on drug transfer from wafer to simulated mucosa evaluated by an in vitro
439
model using the following parameters: contact time 5 min, pressure 0.1 bar; drug content 5 µg/cm² fluorescein sodium,
440
100 µg/cm² quinine (anhydrous), or 500 µg/cm² diclofenac sodium and alginate gel film casted by1000 µm film
441
applicator frame (MW ± SD, n = 6).
ip t
438
cr
442 443
Figure 4: Drug release profiles of wafers with a backing layer thickness of 500 µg ethyl cellulose (EC)/cm² (size
444
24 cm²) loaded either with 5 µg/cm² fluorescein sodium
445
diclofenac sodium
446
SD, n = 3).
us
(A), 100 µg/cm² quinine (anhydrous)
(B) or 500 µg/cm²
an
(C) in 0.07 M phosphate buffer (USP) pH 6.8 at 37 °C determined with paddle apparatus (MW ±
447 448
Figure 5: Distribution of fluorescein sodium
449
wafers with a backing layer thickness of 500 µg ethyl cellulose (EC)/cm² investigated in modified Ringer buffer pH 7.4
450
at 37 °C using Ussing chamber (MW ± SD, n = 3).
453 454 455 456 457
(C) from
M
(B) and diclofenac sodium
d Ac ce pt e
451 452
(A), quinine (anhydrous)
TM
458
Table 1: Summary of produced wafer formulations using a mixture of Methocel
459
and Macrogol 400 with a ratio of 1:2:4, different backing layer thicknesses as well as model drug substance
460
concentrations (EC=ethyl cellulose, FL=Fluorescein sodium, QN=Quinine anhydrous, Diclo=Diclofenac sodium).
E15 Premium LV, polyvinyl alcohol
461
twolayered wafers
backing layer (µg EC/cm²)
model drug substance (µg/cm²) FL
QN
Diclo
0
5
100
500
0
0
0
5
100
500
0
0
0
5
100
500
(placebo and drug-loaded) 300 400
16 Page 16 of 24
500 750
0
0
0
5
100
500
0
0
0
5
100
500
462 463
ip t
464 465 466
cr
467 468
us
469 470
an
471 472 473
M
474 475 476
d
477
Ac ce pt e
478 479 480 481 482 483 484 485
TM
486
Table 2: Characterization of wafers consisting of a water-insoluble backing layer of Ethocel
487
(ethyl cellulose, EC) of different thicknesses expressed as amount of EC per area and a drug-loaded, mucoadhesive
488
layer, which was produced using a mixture of Methocel
489
ratio of 1:2:4 and as well as containing a variety of model drug substances (MW ± SD, n = 3-6).
TM
Standard 10 Premium
E15 Premium LV, polyvinyl alcohol and Macrogol 400 with a
490 total weight (mg/cm²)
total thickness (µm)
tensile strenght (MPa)
elongation at break (%)
folding endurance
0
7.5 ± 0.2
82 ± 6
2.4 ± 0.3
63.8 ± 24.9
> 100
300
7.4 ± 0.1
88 ± 9
2.8 ± 0.0
38.6 ± 4.5
> 100
backing layer (µg EC/cm²) placebo
17 Page 17 of 24
400
7.7 ± 0.6
87 ± 4
3.1 ± 0.3
11.6 ± 2.6
> 100
500
7.8 ± 0.3
91 ± 8
3.1 ± 0.2
12.4 ± 1.0
> 100
750
8.2 ± 1.1
95 ± 17
2.6 ± 0.7
8.3 ± 0.3
> 100
fluorescein sodium (5 µg/cm²) 6.8 ± 0.6
71 ± 5
3.1 ± 0.2
91.3 ± 5.0
> 100
300
7.4 ± 0.4
85 ± 6
3.0 ± 0.3
47.8 ± 10.0
> 100
400
7.9 ± 0.6
87 ± 3
3.6 ± 0.1
11.7 ± 0.3
> 100
500
7.3 ± 0.1
88 ± 6
3.9 ± 0.5
9.5 ± 3.0
> 100
750
8.6 ± 0.2
87 ± 6
3.2 ± 0.1
6.9 ± 0.4
quinine (anhydrous) (100 µg/cm²) 77 ± 3
2.4 ± 0.3
40.3 ± 4.6
> 100
300
7.2*
91 ± 9
1.9 ± 0.1
63.6 ± 37.2
> 100
400
8.6 ± 1.0
94 ± 5
1.4 ± 0.1
35.7 ± 15.6
> 100
500
8.4 ± 1.2
112 ± 8
1.4 ± 0.1
750
8.0*
106 ± 10
1.5 ± 0.1
300
8.2 ± 1.1
97 ± 12
400
8.3 ± 0.6
93 ± 11
500
8.5 ± 0.7
97 ± 6
750
8.7 ± 0.2
98 ± 8
us
> 100
13.8 ± 2.2
> 100
an
83 ± 5
1.9 ± 0.2
80.9 ± 39.0
> 100
1.9 ± 0.3
106.8 ± 14.4
> 100
1.8 ± 0.2
101.2 ± 18.7
> 100
2.3 ± 0.3
67.9 ± 7.8
> 100
2.2 ± 0.2
10.7 ± 1.9
> 100
M
7.3 ± 0.5
75.1 ± 26.7
d
0
cr
6.8 ± 0.4
Ac ce pt e
492
> 100
0
diclofenac sodium (500 µg/cm²)
491
ip t
0
18 Page 18 of 24
Ac
ce
pt
ed
M
an
us
cr
i
*Graphical Abstract (for review)
Page 19 of 24
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 20 of 24
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 21 of 24
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 22 of 24
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 23 of 24
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 24 of 24
S0378-5173(17)30135-7 http://dx.doi.org/doi:10.1016/j.ijpharm.2017.02.045 IJP 16449
To appear in:
International Journal of Pharmaceutics
Received date: Revised date: Accepted date:
28-11-2016 14-2-2017 17-2-2017
Please cite this article as: Kirsch, K., Hanke, U., Weitschies, W.,Preparation and characterization of gastrointestinal wafer formulations, International Journal of Pharmaceutics (2017), http://dx.doi.org/10.1016/j.ijpharm.2017.02.045 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Preparation and characterization of gastrointestinal wafer formulations
2 3 4
Kirsten Kirsch, Ulrike Hanke, Werner Weitschies1
5 Universitiy of Greifswald, Center of Drug Absorption and Transport, Institute of Pharmacy, Felix-HausdorffStrasse 3, 17487 Greifswald, Germany
ip t
6 7 8
cr
9 10
1
11
[email protected]
12
phone: +49 3834 4204813
13
Abstract
14
Many active pharmaceutical ingredients (API) have a very poor or highly variable bioavailability after oral administration. One
15
possibility to overcome this problem might be found in the application of mucoadhesive dosage forms like gastrointestinal
16
wafers. However, a currently unsolved challenge is the control of the adhesion of the wafer to the intestinal mucus. One
17
suggested solution might be the combination of gastrointestinal wafers and expanding systems. Such a combination requires
18
thin and elastic wafers which are further characterized by an unidirectional drug release. In this study gastrointestinal,
19
twolayered wafers containing a water-insoluble backing layer and a drug-loaded, mucoadhesive layer were fabricated by casting
20
solvent technique. The backing layer consists of Ethocel Standard 10 Premium and the mucoadhesive layer was prepared using
21
a mixture of Methocel E15 Premium LV, polyvinyl alcohol and Macrogol 400. The wafers were characterized regarding their
22
appearance, mechanical properties and dissolution profiles as well as the influence of backing layer thickness on drug transfer
23
and their ability of unidirectional drug release. The wafers with backing layer thickness of 500░µg Ethocel/cm presented
24
adequate mechanical properties, a drug transfer about 73% and unidirectional drug release.
Ac ce pt e
d
M
an
us
Corresponding author
25 26
Keywords
27
Gastrointestinal wafer; Mucoadhesion; Unidirectional drug release; Methocel E15LV
2
28 29 30 1 Page 1 of 24
31
1 Introduction
32 Many active pharmaceutical ingredients (API) have a very poor and/or highly variable bioavailability after
34
oral administration. Reasons are for example low mucosal permeability, a narrow absorption window at
35
particular regions of the gastrointestinal tract (GIT), variable transit times, various fluid volumes, lack of
36
stability in the gastrointestinal environment resulting in a decomposition prior to its absorption and low
37
concentration of API in gastrointestinal contents (Bhasakaran et al., 2012; Dressman and Reppas, 2000;
38
Hens et al., 2016, Koziolek et al., 2015, Tao and Desai, 2005). Additionally, physiological properties are of
39
relevance. For example mucus thickness ranges from 50–450 µm (median 200 µm) and is influenced by
40
hormonal, paracrine and neural stimulation as well as by inflammatory reactions and acids (Allen et al.,
41
1993; Khutoryanskiy, 2011). One strategy to overcome these problems is the usage of mucoadhesive
42
dosage forms like intestinal wafers. Wafers are defined by the U.S. Food and Drug Administration (FDA)
43
(2009) as “a thin slice of material containing a medicinal agent”. Due to their drug release rates and
44
disintegration times, wafers can be classified into rapid disintegrating, meltaway and sustained release
45
wafers. Rapid disintegrating wafers disintegrate within 30-60 s and result in immediately drug release,
46
whereas meltaway wafers stick to the mucosa, disintegrate within 5-30 min and form a gelatinous,
47
mucoadhesive depot at application site. Sustained release wafers are characterized by disintegration times
48
of several hours and a continuous drug release, ideally zero order kinetics (LTS Lohmann Therapie-
49
Systeme, 2010). After swallowing intestinal wafers have the potential to adhere to gastrointestinal (GI)
50
mucosa because of their mucoadhesive properties. Due to the close contact between wafer and mucosa, a
51
high drug concentration gradient is created, resulting in a high drug flux at the absorbing tissue, which is
52
well supplied with blood. These conditions support presumably drug absorption into systemic circulation
53
and enhance oral bioavailability (Andrews et al., 2009; Bernkop-Schnürch, 2005; Boddupalli et al., 2010).
54
However, a challenge is the loss of control over the dosage form after swallowing. It cannot be guaranteed
55
that the wafers adhere in the intended region of the GIT and in the desired way. Especially using
56
multilayered wafer, it cannot be influenced which side of the wafer adhere to the mucus layer and the
57
underlaying epithelial layer. One suggested solution might be the combination of intestinal wafers with
58
expanding systems which can control adhesion process. Such expanding systems are described in the
Ac ce pt e
d
M
an
us
cr
ip t
33
2 Page 2 of 24
patent of Bogdahn et al. (2015) and consist of a shell, an expansion mechanism and a wafer. The shell
60
could be a custom-designed, gastroresistent capsule, which were swallowed and release the wafer after
61
triggering by pH value, pressure or a composition of a fluid surrounding the shell. The expansion
62
mechanism is selected from the group comprising mechanical expansion system, gas driven expansion
63
system, compressed foam or compressed tissue. The wafer is packed in the shell for example lumped
64
together, collapsed, folded or rolled (Bogdahn et al., 2015). The wafers need specific properties for
65
combination with expanding systems. They have to be thin, elastic and folding resistant. Furthermore, an
66
unidirectional drug release profile is required. The aim of this study was to prepare and characterize rapid
67
disintegrating intestinal wafers which can be combined with an expanding system and are characterized by
68
an unidirectional drug release profile.
us
cr
ip t
59
an
69 70 2 Material und methods
M
71 72
2.1 Formulation of wafers
74
Wafers were produced by a casting solvent technique and consisted of a water-insoluble backing layer of
75
EthocelTM Standard 10 Premium (ethyl cellulose, EC) (Colorcon Limited, United Kingdom) and a
76
drug-loaded, mucoadhesive layer.
77
Firstly, the backing layer was prepared by spraying a solution of 4% (w/w) EC in acetone on the release
78
liner according to a defined spraying scheme. Acetone was evaporated by room temperature for 15 min.
79
Polyethylene paper (Polyslik® 111/105, Loparex, Netherlands) was used as release liner. The thickness of
80
the backing layer was expressed as amount of EC per area. It was adjusted to 0-750 µg EC/cm² and was
81
controlled by weighing.
82
Secondly, the drug-loaded, mucoadhesive layer was fabricated. The most suitable formulation was
83
determined
84
MethocelTM E15 Premium LV (hydroxypropyl methylcellulose, HPMC) (DOW Chemical Company, USA),
85
polyvinyl alcohol, partially hydrolyzed (MW approx. 200000) (PVA) (Merck Schuchardt OHG, Germany) and
86
Macrogol 400 (polyethylene glycol, PEG400) (Fagron GmbH & Co.KG, Germany) were produced, whereby
Ac ce pt e
d
73
in
preliminary
tests
(data
not
shown).
For
this
purpose,
different
mixtures
of
3 Page 3 of 24
87
the ratio of one ingredient at a time varied. Produced formulations were tested regarding their disintegration
88
time,
89
MethocelTM E15 Premium LV, PVA and PEG400 with a ratio of 1:2:4 was chosen as most suitable
90
formulation for drug-loaded, mucoadhesive layer. In this study fluorescein sodium (FL) (Fluka Analytical,
91
Germany), quinine anhydrous (QN) (Sigma-Aldrich Chemie GmbH, Germany) and diclofenac sodium
92
(Diclo) (Fagron GmbH & Co.KG, Germany) were used as model drug substances. These substances were
93
chosen because of their different hydrophilicity/lipophilicity and various solubility in aqueous media. The
94
final drug concentration in each wafer was 5 µg/cm² for fluorescein (FL), 100 µg/cm² for quinine (QN) and
95
500 µg/cm² for diclofenac (Diclo). Additionally, placebo wafers were produced. The compositions of all
96
formulations are summarized in Table 1. The polymer mixture was kept overnight and centrifuged by
97
4400 rpm for 50 min to remove all entrapped air bubbles. Then the mixture was cast onto the dried backing
98
layer using a mechanical film casting apparatus equipped with a vacuum suction plate and 300 µm film
99
applicator frame (film applicator CX4, mtv messtechnik OHG, Germany). Casting speed was adjusted to
100
30 mm/s. The casted mixture was dried at 40 °C for 6 h and stored on release liner packed in aluminum foil
101
at room temperature. The resulting polymer film was cut into smaller pieces and peeled off the release liner
102
before usage.
elongation
at
break
and
folding
endurance.
A
mixture
of
d
M
an
us
cr
ip t
strength,
Ac ce pt e
103
tensile
104
2.2 Wafer characterization
105
2.2.1 Appearance
106
The surface uniformity of the produced wafers was visually inspected. It was rated whether the surface was
107
homogenous, smooth, and free of holes and air pockets. Additionally, scanning electron microscopy (SEM)
108
(PhenomTM, FEI CompanyTM, L.O.T.-Oriel GmbH & Co.KG, Germany) was used to observe surface
109
morphology of a placebo wafer with a backing layer thickness of 500 µg EC/cm².
110
The wafer thickness was measured by a mechanical thickness dial gauge (0.01 mm capacity, Kaefer
111
Messuhrenfabrik GmbH & Co.KG, Germany). The wafer (size 2.5 x 4 cm) was placed between to flat
112
contact points and the thickness was read on the analog display. For each formulation the thickness of
113
three wafers was measured on three defined spots and the average was calculated.
4 Page 4 of 24
114
Finally, the mass of the wafers (size 2.5 x 4 cm) was determined using a digital balance (Sartorius GmbH,
115
Germany).
116 2.2.2 Drug content uniformity
118
The model drug substance distribution in the produced wafers was measured to ensure uniformity. Ten
119
samples (size 1 x 1 cm) were collected randomly from each formulation and dissolved by stirring in 10 mL
120
distilled water using a magnetic stirrer. After complete dissolution of the drug-loaded, mucoadhesive layer of
121
the wafer, samples were measured by fluorescence spectroscopy (Varioskan Flash, Thermo Fisher
122
Scientific Germany BV & Co.KG, Germany) (FL λex 490 nm, λem 513 nm and QN λex 347 nm, λem 373 nm) or
123
UV/VIS-spectroscopy (Cary 50 Scan, Varian, Inc., Germany) (Diclo 276 nm) against calibration in the same
124
medium. Wafers passed content uniformity test if they met requirements of the European
125
Pharmacopoeia 8.8 (Ph.Eur. 8.8) chapter 2.09.06 content uniformity.
an
us
cr
ip t
117
M
126 2.2.3 Mechanical properties
128
Mechanical properties include folding endurance, tensile strength and elongation at break. Each parameter
129
was determined using three random samples.
130
Folding endurance was tested manually. The wafer was folded repeatedly at the same place until breaking
131
and the number of folds was counted. Wafers which could be folded for more than 100times without
132
breaking passed the folding endurance test.
133
Tensile strength σ (MPa) and elongation at break ε (%) were evaluated using a texture analyzer (TAplus,
134
Lloyd Instruments an AMETEK Company, Germany) connected to a data acquisition and analysis software
135
(Nexygen Plus 3.0 Software, AMETEK Company, Germany). The wafer (size 2.5 x 6 cm) was placed
136
between two clamps positioned 5 cm apart. The lower clamp was stationary and the upper clamp was
137
moved at a rate of 1 mm/s to a distance of 200 mm. The force required to break (F, N) the wafer and the
138
elongation at breaking point (l, mm) were measured. Tensile strength was calculated using Formula 1,
139
where A (mm²) is the cross-section of the wafer and elongation at break was calculated using Formula 2,
140
where l0 (mm) is the original length of the wafer.
Ac ce pt e
d
127
141 5 Page 5 of 24
142
σ (MPa) = F (N)/A (mm²)
Formula 1
ε (%) = l (mm)/l0 (mm) · 100 (%)
Formula 2
143 144 145 2.3 Disintegration of mucoadhesive layer
147
The disintegration time of the mucoadhesive layer only of the placebo wafers (size 2.5 x 4 cm) was
148
determined either in 10 mL distilled water or on mucosa simulating alginate gel film (size 5 x 7 cm). The
149
alginate gel film was prepared from an aqueous sodium alginate solution (3% (w/w)) (Fagron GmbH & Co.
150
KG, Germany) and gelled for 10 min with aqueous calcium chloride solution (6% (w/w)) (AppliChem GmbH,
151
Germany). Alginate gel film was chosen because human mucosa is covered by mucus, which is a
152
viscoelastic gel and contains 90-98% water. The used alginate film contained 97% water and had
153
viscoelastic properties as well. Further, alginate gel films and blocks are often used models to simulate
154
tissue surfaces (Ahearne et al., 2005; Neubert, 2009). The endpoint of the disintegration was inspected
155
visually and was defined referring to Ph.Eur. 8.8 chapter 2.09.02 disintegration of suppositories. The
156
endpoint was defined as the time point where the softening of the wafer occurred accompanied by
157
appreciable change of shape and the softening was such that the wafer no longer had a solid core. All tests
158
were performed in triplicate.
cr
us
an
M
d
Ac ce pt e
159
ip t
146
160
2.4 Investigation of the influence of the backing layer thickness on drug transfer
161
The influence of the backing layer thickness on drug transfer was investigated using an in vitro model
162
(Figure 1). The model consists of a tube with a diameter of 3.5 cm which was coated with a mucosa
163
simulating alginate gel film as acceptor compartment. The alginate gel film was made of an aqueous
164
sodium alginate solution (3% (w/w)) and gelled for 10 min with aqueous calcium chloride solution
165
(6% (w/w)). A small balloon was used as expanding system in order to initiate the contact between wafer
166
and simulated mucosa during the contact time with defined pressure. The evaluated wafer had a size of
167
1 x 11.5 cm (width x length) in order to fit to tube perimeter and balloon size. At the beginning of the
168
experiment, the wafer was fixed on the balloon and both were slid into the tube. Then the balloon was
169
expanded by compressed air and by this way the wafer was pressed on the alginate gel film. After defined 6 Page 6 of 24
period of time the balloon was collapsed and pulled out of the tube. At the end of the experiment, the
171
amount of model drug substance which was transmitted to the alginate gel film, was remaining on the
172
balloon and was adhering to the tube was determined. For this purpose, the mucosa simulating alginate gel
173
film was dissolved in 50 mL of five times concentrated phosphate buffer pH 6.8 (USP), the balloon was
174
incubated in 25 mL of five times concentrated phosphate buffer pH 6.8 (USP) and the tube was rinsed three
175
times with five times concentrated phosphate buffer pH 6.8 (USP). Aliquots were measured by fluorescence
176
spectroscopy (FL λex 490 nm, λem 513 nm and QN λex 347 nm, λem 373 nm) or UV/VIS-spectroscopy (Diclo
177
276 nm) against calibration in the same medium. The test parameters were set to a contact time of 5 min
178
and a pressure of 0.1 bar. The alginate gel film was casted using a 1000 µm film applicator frame. All tests
179
were carried out sixfold and significance of influence of backing layer thickness was verified by ANOVA (
180
= 0.05).
an
us
cr
ip t
170
181
M
182 2.5 Dissolution test
184
Dissolution tests of wafers with a backing layer thickness of 500 µg EC/cm² (size 24 cm²) were performed
185
using paddle apparatus (DT80, Erweka, Germany) (75 rpm) with 750 mL 0.07 M phosphate buffer pH 6.8
186
(USP) at 37 °C. Aliquots were taken over a period of 60 min, the withdrawn volume was replaced by fresh
187
buffer and the model drug substance concentrations were determined either by using fluorescence
188
spectroscopy (FL λex 490 nm, λem 513 nm and QN λex 347 nm, λem 373 nm) or UV/VIS-spectroscopy (Diclo
189
276 nm) against a calibration in the same medium. All tests were carried out in triplicate.
Ac ce pt e
190
d
183
191
2.6 Unidirectional drug release
192
The ability of the wafers to improve unidirectional drug release was studied using an Ussing chamber
193
(EasyMount 6-channal Ussing Chamber System, Physiologic Instruments Inc., USA). The tested wafers
194
(size 1 cm²) had a backing layer thickness of 500 µg EC/cm². The wafers were mounted between both
195
compartments of the Ussing chamber, thereby the backing layer was facing compartment 1 (donor
196
compartment) and the drug-loaded, mucoadhesive layer was facing compartment 2 (acceptor
197
compartment). Subsequently, both compartments were filled with modified Ringer buffer pH 7.4. The buffer 7 Page 7 of 24
was prepared with the following composition using distilled water as solvent: 1.60 g/L sodium
199
hydrogencarbonate (NaHCO3) (AppliChem GmbH, Germany), 0.06 g/L sodium dihydrogenphosphate
200
(NaH2PO4) (Sigma Aldrich, Germany), 6.50 g/L sodium chloride (NaCl) (AppliChem GmbH, Germany),
201
0.40 g/L potassium chloride (KCl) (AppliChem GmbH, Germany), 0.23 g/L calcium chloride (CaCl2)
202
(AppliChem GmbH, Germany), 0.23 g/L magnesium chloride (MgCl2) (AppliChem GmbH, Germany),
203
0.14 g/L di-sodium hydrogenphosphate (Na2HPO4) (AppliChem GmbH, Germany) and 1.80 g/L glucose
204
anhydrous (AppliChem GmbH, Germany). The Ussing chamber was incubated at 37 °C and the medium
205
was constantly moved by gas bubbles (Carbogen, Air Liquide Deutschland GmbH, Germany). Over a
206
period of 2 h 200 µL aliquots were taken from both compartments, withdrawn volumes were replaced by
207
fresh buffer and model drug substance concentrations were measured either using fluorescence
208
spectroscopy (FL λex 490 nm, λem 513 nm and QN λex 347 nm, λem 373 nm) or UV/VIS-spectroscopy (Diclo
209
276 nm) against calibration in the same medium. The percentage of the model drug substance distribution
210
in the donor and acceptor compartment was calculated. Thereby, the leakage from the backing layer and
211
the drug release from the drug-loaded, mucoadhesive layer were determined. All tests were performed at
212
least in triplicate.
d
M
an
us
cr
ip t
198
214 215 216
3. Results
Ac ce pt e
213
217
3.1 Wafer characterization
218
3.1.1 Appearance
219
All wafers had a smooth, homogenous and air bubble free surface as it is demonstrated in Figure 2. The
220
results of the thickness and weight measurements are summarized in Table 2. The weight and the
221
thickness of the placebo and drug-loaded wafers increased with increasing thickness of the backing layer.
222
The total weight of the placebo wafers increased from 7.4 mg/cm² to 8.2 mg/cm² and the total thickness
223
from 88 µm to 95 µm.
224 225
3.1.2 Drug content uniformity 8 Page 8 of 24
226
The average content of ten samples from the polymer film of each wafer formulation was between 90% and
227
110% of declared content and no sample was between 75% and 125% of the declared content.
228
Accordingly, all investigated wafers passed the test of content uniformity.
229 3.1.3 Mechanical properties
231
Placebo and drug-loaded wafers were tested regarding their tensile strength, elongation at break and
232
folding endurance. The results are presented in Table 2. The placebo wafers had a tensile strength
233
between 2.4 MPa and 3.1 MPa. The elongation at break decreased from 63.8% to 8.3% with increasing
234
backing layer thickness. Wafers containing FL presented an increased tensile strength, whereas wafers
235
containing QN or Diclo showed a decreased tensile strength compared to placebo wafers. FL wafers also
236
exhibited a decreasing elongation at break from 91.3% to 6.9% with increasing backing layer thickness. For
237
QN wafers and Diclo wafers no tendency depending on backing layer thickness could be observed.
238
Furthermore, all tested formulations presented a folding endurance over 100.
M
an
us
cr
ip t
230
239 3.2 Disintegration of mucoadhesive layer
241
The mucoadhesive layer of the placebo wafers disintegrated in 10 mL of distilled water in 63 ± 6 sec
242
(MW ± SD, n = 3) and on mucosa simulating alginate gel film within 65 ± 5 sec (MW ± SD, n = 3).
Ac ce pt e
243
d
240
244
3.3 Investigation of the influence of the backing layer thickness on drug transfer
245
The influence of the backing layer thickness on drug transfer from wafer to mucosa simulating alginate gel
246
film was determined, whereby thickness varied between 0 and 750 µg EC/cm². With increasing backing
247
layer thickness, the model drug substance amount which was transferred to the simulated mucosa and to
248
the tube increased about 35.9% (FL), 38.7% (QN) or 27.8% (Diclo) comparing wafers with a backing layer
249
thickness of 0 µg EC/cm² and 750 µg EC/cm² (Figure 3). The increase of transmitted drug substance
250
amount with increasing backing layer thickness was significant for all tested substances.
251 252
3.4 Dissolution test
9 Page 9 of 24
253
In Figure 4 the release profiles of drug-loaded wafers with a backing layer thickness of 500 µg EC/cm² are
254
shown. Wafers loaded with FL showed complete drug release within 30 min, whereby only 87% of declared
255
drug amount was recovered. QN wafers presented complete drug release of 98% after 7.5 min and Diclo
256
wafers presented 99% drug release within 15 min.
257 3.5 Unidirectional drug release
259
The ability of wafers to improve unidirectional drug transport was determined using an Ussing chamber
260
mounting the wafer between both compartments. Percentage of the model drug substance distribution
261
between donor and acceptor compartment was calculated and is presented in Figure 5. The investigated
262
wafers showed a rapid model drug substance accumulation in the acceptor compartment, followed by a
263
slow increase of the model drug substance concentration in the donor compartment and a slow decrease in
264
the acceptor compartment after 30 to 60 min.
an
us
cr
ip t
258
M
265 266 4 Discussion
d
267
Ac ce pt e
268 269
The produced wafers were intended to be flexible, rapidly disintegrating and to provide an unidirectional
270
drug release profile (immediate release wafers). For this purpose, twolayered wafers consisting of a
271
water-insoluble backing layer and a drug-loaded, mucoadhesive layer were developed.
272
The backing layer is intended to act as a barrier between the drug-loaded layer and the intestinal luminal
273
fluid in vivo. For the production EthocelTM Standard 10 Premium was used which is an EC containing an
274
ethoxyl content of 48.0 to 49.5% (The Dow Chemical Company, 2005). EC was selected because it is
275
insoluble in water and literature data confirm that it prevents drug release from the backing layer side of
276
wafers and promotes unidirectional drug release (Gupta et al., 2013; Kalavadia et al., 2014; Toorisaka et
277
al., 2012; Whitehead et al., 2004).
278
The drug-loaded, mucoadhesive layer should promote a high drug concentration gradient at intestinal
279
mucosa. The polymers used for production should be mucoadhesive, film forming, biocompatible and
280
non-toxic. The polymer selection based on theories of mucoadhesion which claim that mucoadhesive 10 Page 10 of 24
polymers are hydrophilic networks containing several polar, non-charged and/or non-ionic, hydrogen bond
282
forming, functional groups (Andrews et al., 2009; Boddupalli et al., 2010; Dodou et al., 2005; Khutoryanskiy,
283
2011; Shaikh et al., 2011). Because of these MethocelTM E15 Premium LV and PVA were used for the
284
fabrication of the drug-loaded, mucoadhesive layer. Additionally, PEG400 was applied in this layer because
285
it is as good plasticizer (Honary and Orafai, 2002; Saringat et al., 2005).
286
The produced twolayered wafers were characterized regarding their appearance and mechanical
287
properties. All wafers had a homogenous and smooth surface. The weight and the total thickness of the
288
wafers increased with increasing backing layer thickness. The tensile strength increased with increasing
289
backing layer thickness as well. This is on good accordance to the observations reported by Bhasakaran et
290
al. (2012). One reason could be that the surface connection between the backing layer and the
291
mucoadhesive layer build up a composite material which has different properties than the single layers. The
292
mucoadhesive layer remains unchanged, but the backing layer becomes more resistant to mechanical
293
stress with increasing thickness. Consequently, the entire twolayered wafer becomes more resistant to
294
mechanical stress. Furthermore, the integration of model drug substances resulted in an increasing weight
295
and total thickness and influenced mechanical properties. The tensile strength decreased and the
296
elongation at break increased with incorporation of FL. Wafers containing QN and Diclo as model drug
297
substances presented no change of the mechanical properties. A reason could be the solubility depending
298
distribution of the model drug substances in the wafer matrix. All polymers used for the wafer matrix were
299
hydrophilic. The hydrophilic FL (LogP 0.67) (PubChem; Sigma Aldrich Chemie GmbH, 2011) is probably
300
dissolved in the polymeric wafer matrix and intercalates between the polymer chains. So, it acts as a
301
plasticizer. In contrast, QN (logP 3.44) (DrugBank Version 4.3) and Diclo (logP 1.27) (Smith, 2014) are
302
more likely suspended in the matrix and have no effect on polymer chain formation.
cr
us
an
M
d
Ac ce pt e
303
ip t
281
304
To determine the optimal backing layer thickness drug transfer experiments with an in vitro model were
305
conducted (Figure 1). For this purpose, the drug transfer from the wafer to a mucosa simulating alginate gel
306
film was measured. The amount of model drug substance transferred to the alginate gel increased with
307
increasing backing layer thickness. One reason for this observation could be that the drug-loaded,
308
mucoadhesive layer disintegrates and dissolves during the contact time, whereas the water-insoluble 11 Page 11 of 24
backing layer remains unchanged and intact. As a result, the backing layer minimizes the contact between
310
the model drug substance and the expanding system and supports the drug transfer to the mucosa
311
simulating alginate gel film. Further barrier function increased with increasing thickness of the backing layer.
312
The residual amounts on the expander essentially resulted from the fact that the disintegrating wafer
313
spreads over the boundary of the backing layer. With respect to the obtained results a backing layer
314
thickness of 500 µg EC/cm² was selected for all following experiments.
315
ip t
309
lnvestigated wafers showed a complete release of the incorporated model drug substances within 30 min.
317
However, FL loaded wafers provided a prolonged release in comparison to wafers containing QN and Diclo.
318
One explanation might be found again in the solubility depending distribution of the model drug substance
319
within the matrix. Because of these, the diffusion rate of FL from the dosage form into the medium was the
320
rate limitating step. By QN and Diclo, the erosion rate of the polymer matrix was the rate limitating step (The
321
Dow Chemical Company, 2000). Another explanation could be that FL penetrate into backing layer because
322
of its solubility, interacts with the EC and retained there. Additionally, the wafers partially curled up after
323
medium contact and the backing layer was facing the medium. This resulted in a reduced contact area
324
between the drug-loaded layer and the medium and as a consequence in a prolonged release.
us
an
M
d
Ac ce pt e
325
cr
316
326
Finally, the unidirectional drug release behavior of wafers with a backing layer thickness of 500 µg EC/cm²
327
was determined using the Ussing chamber. Over a period of 2 h aliquots were taken from both
328
compartments and leakage from the backing layer and the drug transfer from the mucoadhesive layer were
329
determined. The investigated wafers provided a rapid model drug substance transfer into the acceptor
330
compartment. After 30 to 60 min, a slow increase of the model drug substance concentration in the donor
331
compartment followed due to redistribution by diffusion through the backing layer. In conclusion, the
332
backing layer prevents drug release in the donor compartment and promotes unidirectional drug transport.
333 334 335
5 Conclusion
336 12 Page 12 of 24
The aim of this study was to develop a rapidly disintegrating intestinal wafer which can be combined with an
338
expanding system and is characterized by an unidirectional drug release. The developed wafers were
339
characterized regarding their appearance, mechanical properties, dissolution profiles as well as their drug
340
transfer and drug release properties. Wafers with a backing layer thickness of 500 µg EC/cm² exhibited
341
adequate mechanical properties, a drug transfer of about 73% to a mucosa simulating alginate gel and
342
unidirectional drug release. In combination with expanding systems such wafer can enhance drug
343
absorption and enable oral intake of drugs with poor oral bioavailability. However, in subsequent
344
experiments more biorelevant in vitro as well as in vivo studies should be developed to proof this concept.
cr
ip t
337
us
345
347
Conflict of interest
348
The authors report no conflict of interest.
an
346
M
349 350
d
352
References
Ac ce pt e
351
353 354 355
Ahearne, M., Yang, Y., El Haj, A. J., Then, K. Y., & Liu, K.-K. (2005). Characterizing the viscoelastic properties of thin hydrogel-based constructs for tissue engineering application. Journal of the Royal Society Interface, 2, S. 455463. doi:10.1098/rsif.2005.0065
356 357
Allen, A., Flenström, G., Garner, A., & Kivilaakso, E. (1993). Gastroduodenal Mucosal Protection. Physiological Reviews, 73(4), S. 823-857.
358 359
Andrews, G. P., Laverty, T. P., & Jones, D. S. (2009). Mucoadhesive polymeric platforms for controlled drug delivery. European Journal of Pharmaceutics and Biopharmaceutics, 71, pp. 505-518.
360 361
Bernkop-Schnürch, A. (2005). Mucoadhesive systems in oral drug delivery. Drug Discovery Today: Technologies, 2(1), pp. 83-87.
362 363 364
Bhasakaran, S., Moris, S., & Rout, A. (2012). Gastrointestinal Mucoadhesive Patch System for Oral Administration of Metronidazole. International Journal of Research in Pharmaceutical and Biomedical Sciences, 3(4), pp. 14971505.
365 366
Boddupalli, B. M., Mohammed, Z. N., Nath, R. A., & Banji, D. (2010). Mucoadhesive drug delivery system: An overview. Journal of Advanced Pharmaceutical Technology & Research, 4, pp. 381-387.
367
Bogdahn, M., Kirsch, K., Grimm, M., Koziolek, M., & Weitschies, W. (22. 12 2015). Patentnr. PCT/EP2015/002601. 13 Page 13 of 24
Dodou, D., Breedveld, P., & Wieringa, P. A. (2005). Mucoadhesives in the gastrointestinal tract: revisiting the literature for novel applications. European Journal of Pharmaceutics and Biopharmaceutics, pp. 1-16.
370 371
Dressman, J. B., & Reppas, C. (2000). In vitro-in vivo correlations for lipophilic, poorly water-soluble drugs. European Journal of Pharmaceutical Sciences, 11(Suppl. 2), pp. S73-S80.
372 373
DrugBank Version 4.3. (n.d.). LogP quinine anhydrous. Retrieved 12 01, 2015, from http://www.drugbank.ca/drugs/DB00468
374 375
Grabovac, V., Föger, F., & Bernkop-Schnürch, A. (2008). Design and in vivo evaluation of a patch delivery system for insulin based on thiolated polymers. International Journal of Pharmaceutics, 348, pp. 169-174.
376 377
Gupta, V., Hwang, B. H., Lee, J., Anselmo, A. C., Doshi, N., & Mitragotri, S. (2013). Mucoadhesive intestinal devices for oral delivery of salmon calcitonin. Journal of Controlled Release, 172, pp. 753-762.
378 379 380
Hens, B., Corsetti, M., Spiller R, Marciani, L., Vanuytsel, T., Tack, J., . . . Augustijns, P. (2016). Exploring Gastrointestinal Variables Affecting Drug and Formulation Behavior: Methodologies, Challenges and Opportunities. International Journal of Pharmaceutics. doi:10.1016/j.ijpharm.2016.11.063
381 382 383
Honary, S., & Orafai, H. (2002). The Effect of Different Plasticizer Molecular Weights and Concentrations on Mechanical and Thermomechanical Properties of Free Films. Drug Development and Industrial Pharmacy, 28(6), pp. 711-715.
384 385 386
Kalavadia, S., Dash, R. P., Misra, M., & Nivsarkar, M. (2014). Design and in vivo evaluation of gastrointestinal mucoadhesive patch system (GMPS) loaded with chitosan nanoparticles. International Journal of Pharmaceutical Development & Technology, 4(4), pp. 258-266.
387 388
Khutoryanskiy, V. V. (2011). Advances in Mucoadhesion and Mucoadhesive Polymers. Macromolecular Bioscience, 11, pp. 748-764.
389 390 391
Koziolek, M., Grimm, M., Becker, D., Iordanov, V., Zou, H., Shimizu, J., . . . Weitschies, W. (2015). Investigation of pH and Temperature Profiles in the GI Tract of Fasted Human Subjects Using the Intellicap(R) System. Journal of Pharmaceutical Sciences, 104(9), pp. 2855-2863.
392 393
LTS Lohmann Therapie-Systeme. (2010). Your active ingredient in its most appealing form. Retrieved 01 15, 2013, from www.lts-corp.com
394 395
Neubert, A. (2009). Entwicklung eines In vitro-Modells zur Untersuchung der Wirkstofffreisetzung und -verteilung aus Arzneistoff-freisetzenden Stents. Ernst-Moritz-Arndt-Universität Greifswald, Greifswald.
396 397
PubChem. (n.d.). LogP Fluorescein Sodium. Retrieved 11 28, 2014, from http://pubchem.ncbi.nlm.nih.gov/compound/10608'sectionVapor-Pressur
398 399 400
Saringat, H. B., Alfadol, K. I., & Khan, G. M. (2005). The influence of different plasticizers on some physical and mechanical properties of hydroxypropyl methylcellulose free films. Pakistan Journal of Pharmaceutical Sciences, 18(3), S. 25-38.
401 402
Shaikh, R., Singh, T. R., Garland, M. J., Woolfson, A. D., & Donnelly, R. F. (2011). Mucoadhesive drug delivery systems. Journal of Pharmacy and Bioallied Sciences, 3(1), pp. 89-100.
Ac ce pt e
d
M
an
us
cr
ip t
368 369
14 Page 14 of 24
Shen, Z., & Mitragotri, S. (2002). Intestinal Patches for Oral Drug Delivery. Pharmaceutical Research, 19(4), S. 391395.
405
Sigma Aldrich Chemie GmbH. (2011). Sicherheitsdatenblatt Fluorescein-Natrium.
406 407
Smith, H. (2014). Formulation, in vitro release and transdermal diffusion of diclofenac salts by implementation of the delivery gap principle. Retrieved 12 01, 2015, from http://dspace.nwu.ac.za/handle/10384/12003
408 409
Tao, S. L., & Desai, T. A. (2005). Gastrointestinal patch systems for oral drug delivery. Drug Discovery Today, 10(13), pp. 909-915.
410 411
Teutonico, D., Montanari, S., & Ponchel, G. (2011). Concentration and surface of absorption: Concepts and applications to gastrointestinal patches delivery. International Journal of Pharmaceutics, 413, pp. 87-92.
412 413
The Dow Chemical Company. (2000). Using METHOCEL Cellulose Ethers for Controlled Release of Drugs in Hydrophilic Matrix Systems. 1-36.
414
The Dow Chemical Company. (2005). ETHOCEL Ethylcellulose Polymers Technical Handbook. 1-28.
415 416
The European Directorate for the Quality of Medicines & Health Care. (2016). PhEur - European Pharmacopoeia 8.8. Retrieved from http://online6.edqm.eu/ep805/
417 418
Toorisaka, E., Watanabe, K., Ono, H., Hirata, M., Kamiya, N., & Goto, M. (2012). Intestinal patches with an immobilized solid-in-oil formulation for oral protein delivery. Acta Biomaterialia, 8, pp. 653-658.
419 420 421 422
U.S. Food and Drug Administration. (2009). Data Standards Manual (monographs) - Dosage Form. Retrieved 07 21, 2016, from http://www.fda.gov/Drugs/DevelopmentApprovalProcess/FormsSubmissionRequirements/ElectronicSubmis sions/DataStandardsManualmonographs/ucm071666.htm
423 424
Whitehead, K., Shen, Z., & Mitragotri, S. (2004). Oral delivery of macromolecules using intestinal patches: applications for insulin delivery. Journal of controlled release, 98, pp. 37-45.
426 427 428
cr
us
an
M
d
Ac ce pt e
425
ip t
403 404
Figure captions
429
Figure 1: Schematic overview of in vitro model for measuring drug transfer from wafers to mucosa simulating alginate
430
gel film as acceptor compartment consisting of a tube (diameter 3.5 cm), mucosa simulating alginate gel film and a
431
balloon that can be expanded with compressed air.
432 433
Figure 2: Scanning electron microscopic (SEM) image at the edge (A) and from above (B, C) of a placebo wafer
434
consisting of a water-insoluble backing layer (C) of 500 µg ethyl cellulose/cm² and a mucoadhesive layer (B), which
15 Page 15 of 24
TM
435
was fabricated using a mixture of Methocel
436
1:2:4.
E15 Premium LV, polyvinyl alcohol and Macrogol 400 with a ratio of
437 Figure 3: Influence of backing layer thickness on drug transfer from wafer to simulated mucosa evaluated by an in vitro
439
model using the following parameters: contact time 5 min, pressure 0.1 bar; drug content 5 µg/cm² fluorescein sodium,
440
100 µg/cm² quinine (anhydrous), or 500 µg/cm² diclofenac sodium and alginate gel film casted by1000 µm film
441
applicator frame (MW ± SD, n = 6).
ip t
438
cr
442 443
Figure 4: Drug release profiles of wafers with a backing layer thickness of 500 µg ethyl cellulose (EC)/cm² (size
444
24 cm²) loaded either with 5 µg/cm² fluorescein sodium
445
diclofenac sodium
446
SD, n = 3).
us
(A), 100 µg/cm² quinine (anhydrous)
(B) or 500 µg/cm²
an
(C) in 0.07 M phosphate buffer (USP) pH 6.8 at 37 °C determined with paddle apparatus (MW ±
447 448
Figure 5: Distribution of fluorescein sodium
449
wafers with a backing layer thickness of 500 µg ethyl cellulose (EC)/cm² investigated in modified Ringer buffer pH 7.4
450
at 37 °C using Ussing chamber (MW ± SD, n = 3).
453 454 455 456 457
(C) from
M
(B) and diclofenac sodium
d Ac ce pt e
451 452
(A), quinine (anhydrous)
TM
458
Table 1: Summary of produced wafer formulations using a mixture of Methocel
459
and Macrogol 400 with a ratio of 1:2:4, different backing layer thicknesses as well as model drug substance
460
concentrations (EC=ethyl cellulose, FL=Fluorescein sodium, QN=Quinine anhydrous, Diclo=Diclofenac sodium).
E15 Premium LV, polyvinyl alcohol
461
twolayered wafers
backing layer (µg EC/cm²)
model drug substance (µg/cm²) FL
QN
Diclo
0
5
100
500
0
0
0
5
100
500
0
0
0
5
100
500
(placebo and drug-loaded) 300 400
16 Page 16 of 24
500 750
0
0
0
5
100
500
0
0
0
5
100
500
462 463
ip t
464 465 466
cr
467 468
us
469 470
an
471 472 473
M
474 475 476
d
477
Ac ce pt e
478 479 480 481 482 483 484 485
TM
486
Table 2: Characterization of wafers consisting of a water-insoluble backing layer of Ethocel
487
(ethyl cellulose, EC) of different thicknesses expressed as amount of EC per area and a drug-loaded, mucoadhesive
488
layer, which was produced using a mixture of Methocel
489
ratio of 1:2:4 and as well as containing a variety of model drug substances (MW ± SD, n = 3-6).
TM
Standard 10 Premium
E15 Premium LV, polyvinyl alcohol and Macrogol 400 with a
490 total weight (mg/cm²)
total thickness (µm)
tensile strenght (MPa)
elongation at break (%)
folding endurance
0
7.5 ± 0.2
82 ± 6
2.4 ± 0.3
63.8 ± 24.9
> 100
300
7.4 ± 0.1
88 ± 9
2.8 ± 0.0
38.6 ± 4.5
> 100
backing layer (µg EC/cm²) placebo
17 Page 17 of 24
400
7.7 ± 0.6
87 ± 4
3.1 ± 0.3
11.6 ± 2.6
> 100
500
7.8 ± 0.3
91 ± 8
3.1 ± 0.2
12.4 ± 1.0
> 100
750
8.2 ± 1.1
95 ± 17
2.6 ± 0.7
8.3 ± 0.3
> 100
fluorescein sodium (5 µg/cm²) 6.8 ± 0.6
71 ± 5
3.1 ± 0.2
91.3 ± 5.0
> 100
300
7.4 ± 0.4
85 ± 6
3.0 ± 0.3
47.8 ± 10.0
> 100
400
7.9 ± 0.6
87 ± 3
3.6 ± 0.1
11.7 ± 0.3
> 100
500
7.3 ± 0.1
88 ± 6
3.9 ± 0.5
9.5 ± 3.0
> 100
750
8.6 ± 0.2
87 ± 6
3.2 ± 0.1
6.9 ± 0.4
quinine (anhydrous) (100 µg/cm²) 77 ± 3
2.4 ± 0.3
40.3 ± 4.6
> 100
300
7.2*
91 ± 9
1.9 ± 0.1
63.6 ± 37.2
> 100
400
8.6 ± 1.0
94 ± 5
1.4 ± 0.1
35.7 ± 15.6
> 100
500
8.4 ± 1.2
112 ± 8
1.4 ± 0.1
750
8.0*
106 ± 10
1.5 ± 0.1
300
8.2 ± 1.1
97 ± 12
400
8.3 ± 0.6
93 ± 11
500
8.5 ± 0.7
97 ± 6
750
8.7 ± 0.2
98 ± 8
us
> 100
13.8 ± 2.2
> 100
an
83 ± 5
1.9 ± 0.2
80.9 ± 39.0
> 100
1.9 ± 0.3
106.8 ± 14.4
> 100
1.8 ± 0.2
101.2 ± 18.7
> 100
2.3 ± 0.3
67.9 ± 7.8
> 100
2.2 ± 0.2
10.7 ± 1.9
> 100
M
7.3 ± 0.5
75.1 ± 26.7
d
0
cr
6.8 ± 0.4
Ac ce pt e
492
> 100
0
diclofenac sodium (500 µg/cm²)
491
ip t
0
18 Page 18 of 24
Ac
ce
pt
ed
M
an
us
cr
i
*Graphical Abstract (for review)
Page 19 of 24
Ac ce p
te
d
M
an
us
cr
ip t
Figure(s)
Page 20 of 24
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 21 of 24
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 22 of 24
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 23 of 24
Ac
ce
pt
ed
M
an
us
cr
i
Figure(s)
Page 24 of 24