Maciek 28

Maciek 28, biotechnologia

[ Pobierz całość w formacie PDF ]
Cellulose (2007) 14:65 –71
Springer 2006
DOI 10.1007/s10570-006-9091-y
Application of cellulose acetate butyrate-based membrane for osmotic drug
delivery
Anant Shanbhag, Brian Barclay, Joanna Koziara and Padmaja Shivanand*
Oral Products R&D, ALZA Corporation, 1010 Joaquin Road, Mountain View, CA 94043, USA; *Author for
correspondence (e-mail: pshivana@alzus.jnj.com; phone: +1-650-564-2434; fax: +1-650-564-2686)
Received 9 May 2006; accepted in revised form 31 August 2006
Key words: Acetone levels, Controlled drug delivery, OROS
, Semipermeable membrane
Abstract
The release rate of drugs from an OROS
is controlled by semipermeable membranes composed typically
of cellulose acetate (CA) with various flux enhancers. Cellulose acetate butyrate (CAB) was identified as a
viable alternative. The CAB membrane matched the CA membrane in robustness but had superior drying
properties, offering particular advantages for thermolabile formulations. Studies were conducted to
characterize CAB membrane properties with respect to performance of OROS
systems. Four different
membrane formulations with varying plasticizer type and concentration were investigated. The CAB based
membranes exhibited superior drying characteristics and similar functionality to the CA:polyethylene
glycol (PEG) membranes used as a control. A linear relationship was observed between the level of flux
enhancer and release rate. The stability of the membrane was evaluated based on release profiles after
system storage at various conditions. The CAB membranes appeared to have stability comparable to the
standard CA membrane. A linear relationship between membrane weight and release rate as well as the
time required to release 90% of a drug from the system [T
90
] for a model formulation was observed.
In conclusion, the newly identified alternative membrane composition allows for the use of thinner
membranes, thereby reducing cost of goods, coating time and, most importantly, membrane drying time.
Abbreviations: CA – cellulose acetate; CAB – cellulose acetate butyrate; CAP – cellulose acetate phtha-
late; HDPE – high density polyethylene; HPMCAS – hydroxypropyl methylcellulose acetate succinate;
HPMCP – hydroxypropylmethyl cellulose phthalate; MB – mass balance; PEG – polyethylene glycol;
OROS
– controlled release osmotic delivery technology of ALZA Corporation; PL188 – Poloxamer 188;
RR – release rate; ZO – zero-order
Introduction
their application, see Edgar et al. (2001). Cellulose
esters received a great deal of attention in the
pharmaceutical industry as they provide excellent
properties conducive to controlled drug delivery.
Cellulose esters such as hydroxypropyl methylcel-
lulose acetate succinate (HPMCAS), cellulose
acetate (CA), cellulose acetate butyrate (CAB)
and many others were used as matrix materials
Cellulose is a natural polymeric polysaccharide
composed of
b
-
D
-glucose subunits. Hydroxyl
groups of cellulose can be modified chemically to
form esters or ethers that differ in physicochemical
properties, allowing for a wide range of applica-
tions. For a thorough review of cellulose esters and
66
(Maki et al. 2006; Reddy and Bodor 1994; Streubel
et al. 2000). Cellulose esters provide a number of
advantages as film formers. Depending on the
film application, polymer properties can be fur-
ther modified by varying degree of substitution,
hydroxyl content, or chain length (Edgar et al.
2001). Cellulose acetate phthalate (CAP), hy-
droxypropyl methylcellulose phthalate (HPMCP)
and ethyl cellulose in blends with other polymers are
commonly used for enteric coating (Siepmann et al.
2005; Tezuka et al. 1991; Wu et al. 1997). CA was
one of the first materials used for manufacturing
semipermeable membranes in elementary osmotic
pumps developed by ALZA Corporation. These
membranes continue to be used in commercial
OROS
products, where the semipermeable mem-
brane controls drug release rate. Rate-controlling
membranes are mixtures of CA with a degree of
substitution of approximately 2.5 and various
plasticizers, which in certain cases can also act as
flux enhancers. CA membranes are typically ap-
plied from acetone solutions. Residual solvent can
potentially affect the permeability of themembrane,
which may result in variations in drug release pro-
files. Therefore, OROS
systems are usually dried
under controlled conditions to reduce acetone level
to an appropriate level (commonly below 1000 –
2000 ppm).
Membrane characteristics such as morphology,
permeability, or residual solvent level can differ
significantly depending on the technique used to
form the film (Chainey et al. 1985; Sun et al. 1999).
Therefore, characterization of tested membranes is
conducted on spray-coated membranes applied
onto OROS
systems, rather than on solvent-cas-
ted films. Membrane properties are evaluated
based on OROS
performance (e.g., release rate,
time required to release 90% of a drug from the
system [T
90
] or time required to release 50% of a
drug from the system [T
50
]). In the present study
model cores (Formulations A and B) were used, as
they are well characterized and robust. Conse-
quently, any indications of membrane instability
and changes in performance would result in
detectable differences in drug release parameters.
The desired drug release rate can be achieved by
adjusting type and amount of flux enhancer as well
as membrane thickness (Razaghi and Schwartz
2002; Makhija and Vavia 2003; Okimoto et al.
1998; Prabakaran et al. 2004; Rani et al. 2003).
Systems requiring fast controlled release (short
T
90
) would have relatively thin membranes con-
taining high levels of flux enhancers. However, for
systems requiring long delivery times, a thicker
membrane is required to achieve the target release
rate. Thicker membranes require longer coating
runs and longer drying times that can significantly
prolong manufacturing times. Such increases in
process times are normally undesirable. To over-
come recognized limitations of CA membranes,
studies were conducted to identify an alternative
membrane. CAB exhibited good solubility in
organic solvents, while being more hydrophobic
than CA (Eastman 2004). Therefore CAB based
membranes were investigated to achieve low
permeability. Present studies addressed the early
feasibility and development of CAB based,
low-permeability membranes for application in
osmotic drug delivery.
Materials and equipment
CA (398 –10) and CA butyrate (171 –15PG) were
purchased from Eastman Chemicals (Kingsport,
TN). Polyethylene glycol (PEG 3350), Poloxamer
188 (PL 188) and Poloxamer 338 (PL 338) (block
copolymers of ethylene oxide and propylene oxide)
were obtained from BASF (Shreveport, LA).
Acetone was obtained from Univar, Inc. (San
Jose, CA). Solvents for analytical purposes were
high-performance liquid chromatography (HPLC)
grade and obtained from JT Baker and Company.
Model systems used for the study were OROS
Push-Pull formulation A and B.
Membranes were applied using a 12 –in or 24-in
Vector Hi-coater (Vector Corp, IA).
Experimental
The compositions of tested membranes are listed
in Table 1. Coating solutions were prepared at 5%
(w/w) solids and all coatings were conducted using
the same processing parameters. A 0.64-mm orifice
was manually drilled through the membrane in the
center on the drug layer using a Servo drill. The
drilled systems were then dried for 3 days at 45C/
45% relative humidity (RH) before being charac-
terized for release rate. A stability study was
performed using International Conference on
Harmonization (ICH) guidelines at 40C/75%
67
Table 1. Membrane formulations for flux enhancer screening
study.
range for acetyl content, an excellent solubility in
organic solvents, and an ability to form a semi-
permeable film. In the case of a semipermeable
CA-based membrane, as the acetyl content in-
creases, its permeability decreases and solvent
resistance increases (Eastman 2004). Therefore, for
CA-based membranes, permeability can be ad-
justed by modifying flux enhancer levels and/or
altering membrane thickness; however, this results
in longer coating and drying time. Studies were
conducted to find a viable alternative coating
material that would match the CA membrane in
robustness, while imparting a lower water flux.
Initial experiments focused on more hydrophobic
polymers as complete replacements for CA.
Attention was focused particularly on polymers
with a hydrophobic backbone in combination with
flux enhancers to modulate water permeability.
From the screening study, CAB in combination
with Poloxamer was identified as a lead candidate.
CA butyrate exhibited close similarities to CA in
terms of its excellent solubility in organic solvents
and physical properties, while being more hydro-
phobic and demonstrating superior drying prop-
erties. Flux enhancers are commonly used along
with the polymers to adjust the permeability of the
membrane (Guo 1993, 1994). The flux enhancer
type and level of incorporation have been shown
to affect permeability and mechanical properties of
films (Carmen and Roland 1996; Crawford and
Esmerian 1971; Eastman 2004). Therefore, care
needs to be taken during flux enhancer selection to
obtain a membrane with acceptable characteris-
tics. Depending on the desired permeability,
hydrophobic or hydrophilic flux enhancers can be
used to either enhance or reduce water perme-
ability.
In the initial flux enhancer screening study, a
higher amount of water was required to dissolve
the PEG, which is insoluble in pure acetone.
However Poloxamer dissolves upon heating the
pure acetone to 28C. The solvent system can affect
the physical appearance, mechanical properties,
morphology and permeability of the membrane
(Yuan et al. 2001). All the coatings were slightly
opaque and did not exhibit any visual imperfec-
tions. The opacity of the membrane may be
attributed to nonoptimal coating conditions and/
or potential material incompatibilities, which are
currently being investigated. In the current study
average release rate was measured to identify the
Coating ID Membrane composition
%
Water
%
Solids
Run I
CAB 171-15:Poloxamer
188 (80:20)
0.5
5
Run II CAB 171-15: Poloxamer
188 (75:25)
0.5
5
Run III CAB 171-15: Poloxamer
338 (80:20)
0.5
5
Run IV CAB 171-15:PEG
3350 (80:20)
5
5
Run V CA: PEG 3350 (99:1)
5
5
RH and 25 C/60% RH. The samples were pack-
aged in thermo-sealed 45-ml HDPE bottles with
0.75 g desiccant. Release rate and physical
appearance were monitored at various time points
during the course of the study to assess stability.
The drying study was performed by drying
OROS
Push-Pull formulations A and B for the
required period of time in a humidity-controlled
oven at various conditions. The initial residual
solvent level was estimated from coated, undrilled
systems stored in airtight containers at 4C. The
residual solvent in the systems was quantified by
solvent extraction followed by gas chromatography
(GC) analysis.
The effect of flux enhancer level on release rate
was studied at three different concentrations of
PL188: 15, 20 and 25% (w/w). The effect of
membrane weight on system functionality was also
assessed using cores coated with a CAB: PL188
(80:20 w/w) membrane. The systems were drilled
and dried at 37C /ambient humidity for 3 days
prior to testing.
The effect of all tested parameters on system
performance was determined based on release rate
parameters. Release rate was monitored using a
United States Pharmacopoeia (USP) Type VII
apparatus. Systems were released in 50 ml of
artificial gastric fluid pH 1.2 (AGF) at 37C, and
the medium was changed at every time interval.
Aliquots were sampled at 2-h intervals for 24 h
and analyzed for drug content using an Agilent
HPLC system.
Results and discussion
CA is a semisynthetic, generally regarded as safe
(GRAS) polymer with a narrow specification
68
promising membrane compositions with lower
permeability. The flux enhancer screening study
was performed using Formulation A systems, and
release rates of the systems (n = 5) were deter-
mined for all membrane compositions after 3-day
and 7-day drying at 45C/45% RH. A system
functionality summary is listed in Table 2. All the
CAB-based membranes exhibited surprisingly ra-
pid drying characteristics with no residual solvents
being detected after 3 days of drying at 45C/45%
RH. In contrast, the CA: PEG membrane required
more than a 3-day drying to reduce the residual
solvent levels to approximately 1000 ppm. The
target weight for the CAB membrane was 17.2 mg,
which is about 30% lower than the target weight of
24.5 mg for the CA: PEG (99:1 w/w) membrane
used as a control. The membrane weights corre-
spond to thicknesses of 0.15 mm and 0.097 mm for
CA-based and CAB-based membranes, respec-
tively. Despite the thinner membrane, CAB: PL188
(80:20 w/w) and CAB: PEG (80:20 w/w) coated
systems exhibited lower average release rates than
the CA: PEG control (Table 2). The product of
average release rate and membrane weight can be
used as a predictor of the permeability of the
membranes. Two of the four membrane formula-
tions tested exhibited lower permeability than the
CA: PEG control. Data further suggest that
membranes were stable and no dramatic differ-
ences in release rate after 3-day and 7-day drying
were noted. Stability of the membrane composi-
tions was assessed for Formulation A per ICH
guidelines at 40C/75% RH and 25C/60% RH.
Formulation A cores were chosen to assess stability
of the membrane compositions, as this core for-
mulation is robust, well characterized, and has
proven stability. Neither the CAB-based mem-
brane nor the CA: PEG control exhibited any
discernible changes in appearance upon stability
testing. The release rates were determined for the
systems (n = 10) initially and at each sampling
point for the stability study. The CAB:PL 188
(80:20) membrane did not show any changes on
stability up to 6 months at 40C/75% RH in com-
parison to the CA:PEG membrane used as a con-
trol. Furthermore, the real time data at 25C/60%
RH confirmed stability of the CAB: PL188
(80:20 w/w) membrane through 16 months (data
not shown).
The results from the drying studies conducted
on the Formulation B are shown in Figure 1. The
CAB membrane exhibited remarkably rapid
drying characteristics in comparison to the CA:
PEG membrane. It could be hypothesized that
rapid drying of CAB membrane might be related
to differences in material properties and thickness
of the membrane (thinner membranes allowing for
faster diffusion of the solvent). The results of the
flux enhancer study are graphically depicted in
Figures 2 and 3. The release rate from a CA: PEG
membrane used as a control is depicted in
Figure 3B. A linear relationship was observed
between the level of flux enhancer and the average
release rate. This indicates the ability to modify
membrane permeability and consequently the
10000
8000
CAB:PL 188 (80:20 w/w)
CA:PEG (99:1 w/w)
6000
4000
2000
0
0
1
2
3
4
5
6
drying time (days)
Figure 1. Drying curves for CA and CAB-based membranes.
CA:PEG (99:1) average membrane weight: 40 mg, dried at
40C/ambient humidity and the CAB:PL188 (80:20), average
membrane weight: 33 mg, dried at 37C/ambient humidity.
Table 2. Summary of flux enhancer screening study.
Coating ID
Run I
Run II
Run III
Run IV
Run V
Drying time, (days)
3
7
3
7
3
7
3
7
3
7
Average residual acetone (ppm)
0
0
0
0
0
0
0
0
1165
415
Average membrane weight (MW), (mg) 17.4 17.6 21.0 21.0 16.8 16.2 15.9 16.1
23.8
22.6
Average release rate (RR), (mg/h)
0.44 0.46 0.96 0.98 0.73 0.78 0.38 0.39
0.56
0.56
RR*MW
7.6
8.1 20.1 20.6 12.2 12.6
6.0
6.3
13.3
12.6
Release rate variability, %
7.6
2.7
5.2
9.6
6.9 11.8
9.6
5.9
2.5
2.3
69
4
good correlation was observed, when average re-
lease rate was plotted against the inverse of mem-
brane weight. This suggests that the membrane
permeability may be further fine tuned by adjusting
the membrane thickness (i.e. membrane weight). A
linear relationship with a negative slope was ob-
served when [T
90
] was plotted as a function of in-
verse of membrane weight, indicating the ability to
achieve prolonged durations of release with
increasing membrane weight and, consequently,
membrane thickness.
Recent studies have indicated a low level of
extractables from CAB (Ma et al. 2004), thereby
alluding to the safety of this material for clinical
applications. The key application of CAB-based
membrane will be in enabling controlled osmotic
delivery of thermolabile formulations, which
are sensitive to drying conditions, and products
requiring less permeable membranes. The
3
2
1
R
2
= 0.9965
0
0
5
10
15
20
25
30
Poloxamer (wt%)
Figure 2. Effect of flux enhancer (Poloxamer 188) level on drug
release rate from CAB-based OROS
Formulation A.
delivery rate of drugs by altering the level of flux
enhancer in the membrane.
The effect of membrane weight on average release
rate and [T
90
] for the OROS
Formulation B are
shown in Figures 4 and 5. A linear relationship with
A
120
100
80
60
40
75:25
80:20
85:15
20
0
0 2 4 6 8 02468 024
Time (hrs)
B
90
80
70
60
50
40
30
20
10
0
0 2 4 6 8 02468024
Time (hrs)
Figure 3. Release rate profiles from Formulation B. Release rate values were adjusted for membrane weights and mass balance. (A)
Effect of flux enhancer (Poloxamer 188) on cumulative release profiles of CAB-based OROS
(B) Release rate profile for CA:PEG-
based OROS
.
[ Pobierz całość w formacie PDF ]

zanotowane.pl

doc.pisz.pl

pdf.pisz.pl

mement.xlx.pl