Subscriber access provided by READING UNIV
Article
Mutual impact of phase separation/crystallization and water sorption in amorphous solid dispersions Christian Luebbert, Maximilian Wessner, and Gabriele Sadowski Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b01076 • Publication Date (Web): 08 Jan 2018 Downloaded from http://pubs.acs.org on January 10, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Molecular Pharmaceutics is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Molecular Pharmaceutics
1
Mutual impact of phase separation/crystallization
2
and water sorption in amorphous solid dispersions
3
Christian Luebbert, Maximilian Wessner, Gabriele Sadowski*
4
*Department of Biochemical and Chemical Engineering, Laboratory of Thermodynamics, TU
5
Dortmund University, Emil-Figge-Str. 70, D-44227 Dortmund, Germany.
6
GRAPHICAL ABSTRACT:
7
8
ABSTRACT: The molecular integration of poorly-water-soluble active pharmaceutical
9
ingredients (APIs) in a suitable polymeric matrix is a possible approach to enhance the
10
dissolution behavior and solubility of these APIs. Like all newly-developed pharmaceutical
11
formulations, these formulations (often denoted as amorphous solid dispersions (ASDs)) need to ACS Paragon Plus Environment
1
Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 30
12
undergo storage stability tests at defined relative humidity (RH) and temperature conditions. In a
13
previous work (Int. J. Pharm. 532 (2017) 635-646), it was shown that thermodynamic modeling
14
can be successfully used to predict the long-term stability of ASDs against API crystallization
15
and moisture-induced amorphous-amorphous phase separation (MIAPS). This works in turn
16
demonstrates the prediction of water sorption in ASDs accounting for potentially occurring API-
17
crystallization and MIAPS. The water sorption and phase behavior of ASDs containing the APIs
18
felodipine and ibuprofen incorporated in three different hydrophilic polymers poly(vinyl
19
pyrrolidone), poly(vinyl acetate), and poly(vinyl pyrrolidone-co-vinyl acetate) at the conditions
20
25 °C/60% RH and 40 °C/75% RH was predicted using the Perturbed-Chain Statistical
21
Associating Fluid Theory (PC-SAFT). The predictions were successfully validated via two-years-
22
lasting water-sorption experiments. It was shown that crystallization of the API and MIAPS on
23
the one hand and water sorption in the ASDs on the other hand dramatically influence each other
24
and that this behavior can even be quantitatively predicted by PC-SAFT which provides valuable
25
insights already at early stages of formulation development.
26
KEYWORDS: amorphous solid dispersion, water sorption, moisture-induced amorphous-
27
amorphous phase separation, long-term stability, PC-SAFT.
28 29
1 INTRODUCTION
30
Poorly water-soluble active pharmaceutical ingredients (APIs) are challenging for
31
pharmaceutical development as sufficient bioavailability and dissolution characteristics can only
32
be achieved by a suitable formulation strategy, e.g. via dissolving the API amorphously in a
33
suitable polymer (so-called amorphous solid dispersion (ASD)). An appropriate polymer
34
stabilizes the amorphous state of the API throughout the shelf-life of the formulation (if the ACS Paragon Plus Environment
2
Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Molecular Pharmaceutics
35
solubility of the crystalline API in the polymeric matrix is sufficiently high) and enhances the
36
dissolution performance in the gastrointestinal tract (e.g. by choosing a hydrophilic polymer).
37
Using those pharmaceutical formulations requires long-term stability tests according to the
38
International Conference on Harmonization (ICH) Guideline Q1A(R2)1, which defines storage
39
temperatures and storage relative humidities (RH). Common storage conditions are 25 °C/60%
40
RH (storage condition for standard long-term stability tests) or 40 °C/75% RH (storage condition
41
for accelerated long-term stability tests). As shown in previous works2–4, RH has a huge impact
42
on the stability of ASD formulations against API crystallization5, moisture-induced amorphous-
43
amorphous phase separation (MIAPS) crystallization and also affects the glass-transition
44
temperature6 of the API/polymer mixture3. The water content in ASDs can be determined
45
gravimetrically (e.g. by dynamic vapor sorption measurements7,8, thermogravimetric analysis, or
46
storage in the presence of a saturated salt solution9), via NMR-, Raman- or IR-spectroscopy10, or
47
via Karl-Fischer-Titration.
48
Andronis et al. showed that already a few mass percent of water uptake in amorphous
49
Indomethacin (IND) accelerated the crystallization velocity by several orders of magnitude11.
50
Konno and Taylor found that the nucleation rate of pure amorphous Felodipine (FEL) increased
51
by factor 20 after the uptake of only 1 wt% water7. Moreover, MIAPS was qualitatively2,12–16
52
observed and recently also quantitatively17 investigated for several API/polymer formulations.
53
Additive approaches (weighted sum of water sorption of the individual components)8,18,19 do
54
not provide correct measures of the water sorption20 in ASD formulations as they do not at all
55
account for the interactions between all formulation components API, polymer, and water.
56
Instead it is important to note, that the molecular state of an ASD, i.e. whether it is (I)
57
homogeneous and amorphous, or (II) demixed into two amorphous phases, or (III) contains API
58
crystals strongly depends on the amount of absorbed water and vice versa. Thus, phase behavior
ACS Paragon Plus Environment
3
Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 30
59
of ASDs and their water sorption are inextricably linked and can only be understood when
60
considered simultaneously2–4.
61
So far, no systematic work has been published modeling the water sorption in ASDs
62
considering the physical state of the ASD (amorphous, MIAPS and/or API crystallization) along
63
with long-term sorption experiments. Therefore, in this work we predicted the water content in
64
ASD formulations using the Perturbed-Chain Statistical Associating Fluid Theory21 (PC-SAFT)
65
and explicitly considering the physical state of the ASDs. PC-SAFT has been already
66
successfully used to model the phase behavior of ASDs22–25, the aqueous solubility of
67
crystalline26 and amorphous27 APIs, the dissolution of APIs in aqueous media28 and the influence
68
of RH on the API/polymer phase behavior2,3.
69
The water-sorption predictions in this work were verified by two-years enduring long-term
70
sorption measurements. Poly (vinyl pyrrolidone) (PVP) and poly(vinyl acetate) (PVAC) were
71
selected as representatives for very hydrophilic and hydrophobic polymeric excipients,
72
respectively. The copolymer poly(vinyl pyrrolidone-co-vinylacetate) (PVPVA) was selected as
73
example for slightly hydrophilic excipients.
74 75
2 WATER SORPTION IN ASDs
76
In order to correctly describe the water sorption in ASDs, all possible phase transitions
77
(MIAPS and crystallization) must be considered as they certainly influence the water sorption.
78
ASD schematic phase diagrams along with the resulting water-sorption profiles at different
79
temperatures are shown in Fig 1.
ACS Paragon Plus Environment
4
Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Molecular Pharmaceutics
80 81 82 83 84 85 86 87 88 89 90 91
Fig. 1. Schematic phase behavior and water-sorption profiles of an API/polymer formulation at constant RH and varying temperatures (Figures a, b, and c). Upper figures: Temperature-dependent solubility and MIAPS in the API/polymer/water system as function of the API mass fraction in the water-free formulation ( ). The solubility line is black; the MIAPS region is marked in gray. Lower figures: amount of water absorbed in the ASD at indicated temperature as function of . Solid lines: water sorption of homogeneous formulations; dashed-dotted lines: water sorption of formulations undergoing MIAPS; dotted lines: water sorption after crystallization. Thermodynamically-metastable states are thin lines, whereas thermodynamically-stable states are thick lines. Thin vertical lines connect identical states in upper and lower figures.
92
The upper part in Fig. 1 depicts the solubility temperature and the MIAPS region of an
93
API/polymer ASD as function of the API mass fraction in the water-free formulation ( ).
94
Such diagrams show the equilibrium solubility line (formulations stored below this line will
95
eventually crystallize) and the MIAPS region (formulations stored within this region will
96
eventually undergo MIAPS) as described in detail in earlier works2–4. These diagrams do account
97
for the water sorption in the formulation at constant RH and are therefore the phase diagrams of
98
the water-containing formulations (although for practical reasons, is shown on the x-axis).
99
The water sorption depends on the amount of the (hydrophobic) API in the (more or less
100
hydrophilic) polymer as well as on the physical state of the formulations. This is shown in the
101
three lower diagrams of Fig. 1. Herein, the total mass fraction of water absorbed in the ACS Paragon Plus Environment
5
Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 30
102
formulations ( ) is depicted for three different storage temperatures Ta, Tb and Tc (depicted
103
in the upper diagrams) of the same API/polymer system and as function of .
104
Fig. 1a illustrates the phase behavior and the corresponding sorption behavior at constant RH
105
and a storage temperature Ta which lies above both, the MIAPS region and the solubility line.
106
Thus, neither crystallization nor MIAPS is expected to occur. The schematic corresponding water
107
sorption profile shown in the lower diagram of Fig. 1a is characterized by a steady decrease of
108
the amount of absorbed water from a high value in the API-free hydrophilic polymer on the left-
109
hand side of the diagram to a very low value in the pure, hydrophobic API on the very right side
110
of the diagram.
111
At storage temperature Tb (Fig. 1b), crystallization and MIAPS may occur depending on the
112
API content in the formulation . In regions II and IV, the formulations stored at Tb are
113
located above solubility line and MIAPS region (Fig. 1b) and thus will remain amorphous at the
114
given RH. The water sorption in these regions will therefore follow the same line as in Fig. 1a.
115
However, formulations with API contents in the regions III and V will not remain amorphous, but
116
undergo a phase transition upon storage.
117
As long as no phase transition occurs, the water content in the (thus metastable) formulations
118
will correspond to that of the thin solid line in the lower Fig. 1b. Upon formation of crystalline
119
API (region V), the overall water content decreases compared to the amorphous state as crystals
120
do not incorporate water in the crystal lattice. The formulation will therefore absorb less water
121
(dotted line in lower Fig. 1b) than in the fully amorphous state. Similarly, during MIAPS (region
122
III) two amorphous (liquid) phases (API-rich and API-poor) evolve and these phases absorb
123
different amounts of water as they differ in hydrophilicity. The final water content in those
124
demixed formulations will again be different form the one of the homogeneous ASD and will
125
correspond to that of the dash-dotted line in the lower Fig. 1b. ACS Paragon Plus Environment
6
Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Molecular Pharmaceutics
126
At storage temperature Tc, the solubility line now lies far on the left side of the diagram (upper
127
Fig. 1c). This means, that the formulations remain amorphous only for the very low API
128
concentrations in region VI. At least after infinite time (in equilibrium), API crystallization is
129
expected to occur in regions VII, VIII, IX and MIAPS will occur in region VIII. As long as the
130
formulations remain homogeneous and amorphous (i.e. before reaching the equilibrium state),
131
water will be absorbed according to the same line as in lower Fig. 1a (thin solid line in lower Fig.
132
1c). Formulations undergoing MIAPS will absorb water according to the gray dash-dotted line
133
lower Fig. 1c. Hence, as MIAPS is metastable at this temperature, crystallization is
134
superimposing MIAPS and is thus the thermodynamically-stable state (regions VII-IX). After
135
reaching the equilibrium state, these formulations will show API crystals in equilibrium with an
136
amorphous phase which has a composition according to the intersection between regions VI and
137
VII. This means, the API is almost completely depleted from the amorphous phase which in turn
138
becomes much more hydrophilic than the original formulation. Thus, for this specific case, API
139
crystallization enhances the water content in the formulation according to the dotted line in the
140
lower Fig. 1c.
141
It becomes obvious that the water sorption may extremely vary depending on the physical state
142
of the formulation. API crystallization might decrease (Fig. 1b, region V) or increase (Fig. 1c,
143
regions VII-IX) the water content in formulations, depending on the hydrophilicities of the
144
involved API and polymer and the solubility of the API in the polymer/water mixture at the
145
considered storage condition.
146 147
2.1 THERMODYNAMIC MODELING
148
To predict the sorption behavior described in the previous section, the thermodynamic
149
equilibrium of the water vapor (V), the amorphous formulation (L) and the crystallized API
ACS Paragon Plus Environment
7
Molecular Pharmaceutics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 30
150
(solid, S) needs to be considered. The water sorption in homogenous amorphous formulations
151
(Fig. 1, regions I, II, IV, VI) can be calculated by considering the equilibrium of V and L, only.
152
At thermodynamic equilibrium, for each component i the chemical potential µ needs to be the
153
same in the two phases. Assuming that API and polymer cannot evaporate and water is the only
154
component present in both phases V and L leads to the following equilibrium condition: =
155
(1)
156
Whereas it can be assumed that the chemical potential of water in the vapor phase is not
157
influenced by the presence of air, the chemical potential of water in the amorphous phase does
158
depend on the presence and amount of both, the API and the polymer. Thus, the vapor-liquid
159
equilibrium condition given in Eq. 1 results in Eq. 2 =
160 161
=
(2)
where the relative humidity RH is defined as the ratio of the water partial pressure .
in the
In Eq. 2, and are the
162
vapor phase and the vapor pressure of pure water
163
water mole fraction and the water activity coefficient in the amorphous formulation, respectively.
164
accounts for the intermolecular interactions among API, polymer, and water and is
165
obtained in this work from the Perturbed-Chain Statistical Associating Fluid Theory (PC-
166
SAFT)21,29,30.
167
Crystallizing API (regions V, VII-IX) needs to be considered by additionally taking into
168
account the equilibrium of the API between the solid phase S and the amorphous phase L leading
169
to Eq. 3:
170 171 172
! =
(3)
Assuming that the solid phase S consists of pure crystalline API which is in equilibrium with an amorphous phase L containing all three components polymer, water and API leads to Eq.431:
ACS Paragon Plus Environment
8
Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Molecular Pharmaceutics
#
= $ exp +−
173
%&'
/ ∆.%&'
01
/ ∆56,%&'
1
21 − 1 / 4 −
0
%&'
/ 1%&'
8ln 2
1
4−
/ 1%&'
1
+ 1