Scholarly article on topic 'Use of polymer combinations in the preparation of solid dispersions of a thermally unstable drug by hot-melt extrusion'

Use of polymer combinations in the preparation of solid dispersions of a thermally unstable drug by hot-melt extrusion Academic research paper on "Nano-technology"

CC BY-NC-ND
0
0
Share paper
Academic journal
Acta Pharmaceutica Sinica B
OECD Field of science
Keywords
{Carbamazepine / "Hot-melt extrusion" / "Thermal unstable drug" / "Solid dispersion" / "Polymer combination" / Stability / Dissolution}

Abstract of research paper on Nano-technology, author of scientific article — Jia Liu, Feng Cao, Can Zhang, Qineng Ping

Abstract The objective of the study was to prepare solid dispersions containing a thermally unstable drug by hot-melt extrusion (HME). Carbamazepine (CBZ) was selected as model drug and combinations of Kollidon VA64 (VA64), Soluplus (SOL) and Eudragit EPO (EPO) were utilized as carriers. Preformulation was conducted to identify the suitability of polymer combinations based on solubility parameters, differential scanning calorimetry (DSC), hot stage microscopy and thermogravimetric analysis. Physicochemical properties of solid dispersions were determined by DSC, X-ray diffraction, fourier transform infrared spectroscopy, dissolution and accelerated stability testing. The results show that drug-polymer miscibility at temperatures below the melting point (T m) of CBZ was improved by combining EPO with VA64 or SOL. With 30% drug loading in a solid dispersion in SOL:EPO (1:1, w/w), CBZ was mainly present in an amorphous form accompanied by a small amount of a microcrystalline form. The dissolution rate of the solid dispersion was significantly increased (approximately 90% within 5min) compared to either the pure drug (approximately 85% within 60min) or the corresponding physical mixture (approximately 80% within 60min) before and after storage. The solid dispersion in SOL:EPO (1:1, w/w) was relatively stable at 40°C/75% RH under CBZ tablet packaging conditions for at least 3 months. In conclusion, polymer combinations that improve drug-polymer miscibility at an HME processing temperature below the T m of a drug appear to be beneficial in the preparation of solid dispersions containing thermally unstable drugs.

Academic research paper on topic "Use of polymer combinations in the preparation of solid dispersions of a thermally unstable drug by hot-melt extrusion"

Acta Pharmaceutica Sinica B ■■■■;■(■):■■■ III

Institute of Materia Medica, Chinese Academy of Medical Sciences Chinese Pharmaceutical Association

Acta Pharmaceutica Sinica B

www.elsevier.com/locate/apsb www.sciencedirect.com

ORIGINAL ARTICLE

Use of polymer combinations in the preparation of solid dispersions of a thermally unstable drug by hot-melt extrusion

Jia Liua, Feng Caoa, Can Zhangb,n, Qineng Ping

aDepartment of Pharmaceutics, School of Pharmacy, China Pharmaceutical University, Nanjing 210009, China Center of Drug Discovery, State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing 210009, China

KEY WORDS

Carbamazepine; Hot-melt extrusion; Thermal unstable drug; Solid dispersion; Polymer combination; Stability; Dissolution

Abstract The objective of the study was to prepare solid dispersions containing a thermally unstable drug by hot-melt extrusion (HME). Carbamazepine (CBZ) was selected as model drug and combinations of Kollidon VA64 (VA64), Soluplus (SOL) and Eudragit EPO (EPO) were utilized as carriers. Reformulation was conducted to identify the suitability of polymer combinations based on solubility parameters, differential scanning calorimetry (DSC), hot stage microscopy and thermogravimetric analysis. Physicochemical properties of solid dispersions were determined by DSC, X-ray diffraction, fourier transform infrared spectroscopy, dissolution and accelerated stability testing. The results show that drug-polymer miscibility at temperatures below the melting point (Tm) of CBZ was improved by combining EPO with VA64 or SOL. With 30% drug loading in a solid dispersion in SOL:EPO (1:1, w/w), CBZ was mainly present in an amorphous form accompanied by a small amount of a microcrystalline form. The dissolution rate of the solid dispersion was significantly increased (approximately 90% within 5 min) compared to either the pure drug (approximately 85% within 60 min) or the corresponding physical mixture (approximately 80% within 60 min) before and after storage. The solid dispersion in SOL:EPO (1:1, w/w) was relatively stable at 40 °C/75% RH under CBZ tablet packaging conditions for at least 3 months. In conclusion, polymer combinations that improve drug-polymer miscibility at an HME processing temperature below the Tm of a drug appear to be beneficial in the preparation of solid dispersions containing thermally unstable drugs.

© 2013 Institute of Materia Medica, Chinese Academy of Medical Sciences and Chinese Pharmaceutical Association. Production and hosting by Elsevier B.V. All rights reserved.

"Corresponding author. Tel./fax: +86 25 83271171. "Corresponding author. Tel./fax: +86 25 83271092.

E-mail addresses: zhangcan@cpu.edu.cn (Can Zhang), pingqn2004@yahoo.cn (Qineng Ping). Peer review under responsibility of Institute of Materia Medica, Chinese Academy of Medical Sciences and Chinese Pharmaceutical Association.

2211-3835 © 2013 Institute of Materia Medica, Chinese Academy of Medical Sciences and Chinese Pharmaceutical Association. Production and hosting by

Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.apsb.2013.06.007

1. Introduction

In drug discovery, the number of active pharmaceutical ingredients (APIs) with poor water-solubility continues to increase with over 50% of APIs now belonging to Class II of the Biopharmaceutics Classification System (BCS). Because the solubility and/or dissolution rate is the rate limiting step to oral absorption for BCS Class II drugs, improvement in either or both properties is considered a key factor for enhancing their bioavailability1. Structural modifications such as formation of a salt, prodrug2 or different polymorph3 and pharmaceutical technologies such as micronization4, formation of solid dispersions5 or inclusion compounds6 have been widely used to enhance the solubility and/or dissolution rate. Among these methods, application of solid dispersion technology is considered to be one of the most attractive options7'8 involving either hot-melt extrusion (HME)9, solvent co-precipitation10, spray drying11 or grinding12.

In recent years, HME has received widespread attention from the pharmaceutical industry for the production of oral solid dispersions. HME has various advantages over traditional industrial methods including avoidance of solvents and applicability to drugs and adjutants for which a suitable solvent is lacking. It enables sufficient mixing in a short residence time and can be used to produce formulations with controlled, sustained or targeted release13,14. Selected polymers can also serve to mask the bitter taste of certain APIs15,16. In addition, HME is a continuous and controllable process that can be scaled up to a commercially meaningful level17. However, HME is limited in its application to thermally unstable drugs that degrade at the high temperatures and shear forces employed in the extrusion process18. This paper reports a technique to avoid this limitation.

For thermally stable APIs, HME can be conducted at a temperature above their melting point (Tm) that ensures complete conversion of a crystalline to an amorphous form. Such higher temperatures serve to reduce the viscosity of the drug-polymer system to facilitate extrusion. It is also possible to obtain solid dispersions using an extrusion temperature below the Tm of the API by the input of adequate shear forces and selecting a polymer capable of solubilizing the API and providing low viscosity at processing temperatures19-21. However, this is not possible for all drug-polymer combinations since it depends on an interaction between the two components that is very sensitive to their physicochemical properties. The screening of polymers for good miscibility with drug has been attempted by many researchers but few publications have reported improvements in poor miscibility systems.

In the preparation of solid dispersions by HME, single polymers have been generally used as carriers despite the fact that some drugs are only poorly soluble in single polymers. Previous studies have reported that using combinations of polymers can provide a synergistic advantage that includes avoiding unfavorable characteristics of a melt extruded film22, modifying the dissolution profile23 and enhancing the physical stability and oral absorption of solid dispersions24. To further pursue this research, we have investigated the potential miscibility improvement of polymer combinations used to prepare a solid dispersion of the thermally unstable, BCS Class II drug, carbamazepine (CBZ) by HME and combinations of Eudragit EPO (EPO) with either Kollidon VA64 (VA64) or Soluplus (SOL). CBZ exists as at least four anhydrous polymorphs and several solvates among which the anhydrous form III is the only form used in marketed products used to treat epilepsy and trigeminal neuralgia25. The big challenge encountered

o"- NH, Carbamazepine

Figure 1 Molecular structures of carbamazepine and polymers.

in preparing a solid dispersion of CBZ by HME is that the drug substance degrades at its Tm26.

In this study, a solubility parameter was utilized to give a preliminarily indication of the interaction between the drug and polymers. This was followed by thermogravimetric analysis (TGA), differential scanning calorimetry (DSC) and hot stage microscopy (HSM) to assess the thermal stability of materials and drug-polymer miscibility. Solid dispersions of CBZ embedded in single polymers and polymer combinations were then prepared by HME and examined by DSC, X-ray diffraction (XRD), fourier transform infrared spectroscopy (FTIR) and high-performance liquid chromatography (HPLC). In this way, the miscibility of polymer combinations and physicochemical properties of corresponding solid dispersions were fully evaluated.

2. Materials and methods

2.1. Materials

CBZ was purchased from Zhejiang Jiuzhou Pharmaceutical Co., Ltd. (Zhejiang, China). Kollidon VA64 and Soluplus were donated by BASF (Shanghai, China). Eudragit EPO was kindly supplied by Evonik (Shanghai, China). All other chemicals were of analytical or chromatographic grade and used as received. The structures are shown in Fig. 1.

2.2. CBZ-polymer miscibility analysis

2.2.1. Calculation of solubility parameters (S) and glass transition temperature (Tg)

As an indicator of the drug-polymer miscibility, values of S were calculated using the Hoftyzer and van Krevelen group contribution method27 described by Eq. (1).

S2 — 5d2 + Sp2 + Sh2

c ^Fdi c Sd — —TT" ; Sp —

ŒF/)1/2

Sh —

here i is the groups within the molecule, S is the total solubility parameter, Sd is the contribution from dispersion forces, Sp is the contribution from polar interactions, Sh is the contribution of hydrogen bonding, Fdi is the molar attraction constant due to molar dispersion forces, Fpi is the molar attraction constant due to molar

polarization forces, Ehi is the hydrogen bonding energy and V is the molar volume.

The solubility parameters of polymer combinations were calculated using Eq. (2)28

<51,2 = V '1* + V '2^2 (2)

where V is the volume fraction of each compound.

The glass transition temperature (Tg) of polymer combinations was calculated using the Gordon-Taylor equation, Eq. (3)29

(W1T g1 + KW2 Tg2)

T g12 =

(W1 + KW2 )

T g2P2

here p is the density and W is the weight fraction of each compound. The Tg and p values of the three polymers reported

24 26 29 33

in the literature24 26,29 33 were used in the calculation.

2.2.2. Miscibility analysis by modulated temperature DSC (MT-DSC)

This was carried out using a NETZSCHS DSC 204 (NETZSCH group, Germany). Accurately weighed samples (2-3 mg) were placed in sealed aluminum pans and a heat-cool-heat cycle applied involving heating from 30 to 220 °C at 10 °C/min then rapidly cooling to 30 °C and then reheating to 220 °C at 10 °C/min.

2.2.3. Miscibility observation by HSM

HSM was conducted using a Linkham DSC600 as hot stage (Linkham Scientific Instruments Co., Ltd., England) under a Leica DMPL polarizing optical microscope (Leica Microsystems Wetzar Gmbh, Germany). The phase transition during the heat-cool-heat process was observed and images captured under polarizing light for further analyses.

Samples were placed on open glass slides, fixed on the hot stage and heated from 30 to 190 °C at 10 °C/min. The temperature was maintained at 190 °C for 15 min after which samples were rapidly cooled to 30 °C and then reheated to 190 °C at 10 °C/min.

2.3. Preparation and characterization of solid dispersions

2.3.2. Characterization of solid dispersions DSC was performed as in Section 2.2.2 except that only a heat cycle from 30 to 220 °C at 10 °C/min was performed under a nitrogen atmosphere.

XRD was carried out at room temperature using a D/max 2500VL/PC powder X-ray diffractometer (Rigaku, Japan) operating at 40 kV and 40 mA to determine the presence of crystals in extrudates. Samples were scanned over a 20 range of 3^-0° with a step size of 0.02° and a step time of 0.3 s.

FTIR was performed in the 4000^-00 cm-1 region using a 983G FTIR spectrometer (Perkin Elmer, USA) operating at 4 cm-1 resolution and 64 scans per spectrum. Powder samples were mixed with KBr (1%, w/w), compressed into pellets and scanned immediately.

2.4. Solubilization and stability studies 2.4.1. Dissolution testing

This was performed according to Dissolution Test Method 2 as described in the paragraph on CBZ tablets in the Chinese Pharmacopoeia (2010)34 using a ZRS-8G dissolution tester (Tianda Tianfa Technology Co., Ltd., China) with a rotation speed of 150 rpm. Samples containing the equivalent of 100 mg CBZ were accurately weighed and added to the dissolution vessel containing 1L 0.065 M HCl. Aliquots (10 mL) were taken at various times and immediately filtered through a 0.45 pm filter. Filtrates were diluted 1:10 with fresh dissolution medium and then analyzed for CBZ by measuring absorbance at 285 nm using a Rayleigh UV 9600 spectrophotometer (Beijing Ruili Analysis Equipment Co., Ltd., China). Each study was performed in triplicate.

2.4.2. HPLC analysis

CBZ and related substances in extrudates were determined using a Shimadzu LC-2010C HT HPLC system (Shimadzu Corporation, Japan) with an Inertsil® ODS-SP C18 column (150 mm x 4.6 mm, 5 pm; GL Science, USA) and a mobile phase of methanol:water (55:45, v/v) delivered at 1.0mL/min. The injection volume was 20 pL and detection was at 285 nm for CBZ and at 230 nm for related substances.

2.3.1. Preparation by HME

Prior to HME, TGA was used to evaluate the thermal stability of the drug and polymers. Samples were placed in open aluminium pans of a NETZSCHS DG 209 (NETZSCH group, Germany) and heated from room temperature to 250 °C at 10 °C/min under a nitrogen atmosphere.

HME was performed using a co-rotating (diameter 5 14 mm) twin screw HAAKE MiniCTW extruder (Thermo Scientific, Germany) equipped with a 2 mm round opening die. Considering the high content of CBZ tablets (100 or 200 mg per tablet), all extrudates were prepared with a high drug loading of 30%.

Drug and polymers were accurately weighed, mixed and then continuously fed into the extruder operating at a screw speed of 30 rpm and an extrusion temperature of 185 °C or 165 °C which is below the Tm of CBZ. Extrudates were collected as strands, air-cooled and then milled in a mortar until the resulting powder passed through an 80-mesh sieve.

2.4.3. Accelerated stability testing

Extrudates were sealed in bottles protected from light and stored at 40 °C/75% RH for 3 months (storage conditions for CBZ tablets). Any recrystallization of CBZ in the products was then determined by XRD. The content of CBZ and related substances and dissolution of the products were also determined.

2.4.4. Hygroscopicity

Hygroscopicity of CBZ, polymers and extrudates was determined gravimetrically. Accurately weighed samples (approximately 1 g) were placed in chambers under controlled relative humidity (RH) and weighed after storage for 7 days at room temperature. RH in chambers was produced by equilibrating with saturated salt solutions as follows: K2CO3 44% RH; NaBr 60% RH; NaCl 75% RH; and KNO3 92.5% RH at room temperature.

Table 1 Values of solubility parameters (5), glass transition temperatures (Tg) and their differences from CBZ values (A5 and ATg) for CBZ and polymers.

Sample Sd (MPa1/2) Sp (MPa1/2) <5h (MPa1/2) S (MPa1/2) AS (MPa1/2) Pliterature (g/cm3) Tg literature (1C) ATg (°C)

CBZ 22.0 7.0 9.6 25.0 - 1.338 59 -

VA64 19.4 9.7 9.6 23.7 1.3 1.167 101 42

SOL 18.5 10.9 10.2 22.1 2.9 0.99 71 12

EPO 17.3 5.4 8.8 20.2 4.8 1.03 48 11

VA64:EPO (1:2, w/w) - - - 22.1 2.9 - 57.0a 2.0

SOL:EPO (1:1, w/w) - - - 29.9 4.9 - 57.5a 1.5

Calculated by Eq. (3).

3. Results and discussion

3.1. Miscibility analysis based on 5 and Tg values

It is generally believed that favorable interactions and a uniform phase will result when the difference in 5 values (A5) between two components is less than 7 MPa1/2, while unfavorable interactions and phase separation will result when A5>10MPa1/2 35. The calculated solubility parameters are presented in Table 1. The 5s of CBZ, VA64, SOL and EPO are similar and are close to the reported data24,33,36,37. The A5s values between CBZ and the three polymers were in the range 1.68-4.84 MPa1/2, being less than 7 MPa1/2 indicates likely miscibility. However, the combinations of EPO with VA64 or SOL appear to increase the A5s between CBZ and polymer mixtures which does not imply miscibility improvement.

The Tg values of the three amorphous polymers VA64, SOL and EPO were 101, 71 and 48 1C, respectively. The ATg values between CBZ and the three polymers were 42, 12 and 11 1C, respectively. Interestingly, the Tg values of polymer combinations are very close to that of CBZ.

Since the Tg value indicates a temperature above which polymer chains become flexible, more interactions are expected to occur in the heating process if components have similar Tg values. After comparing a series of ratios according to the dispersion state of CBZ and the moldability of extrudates, VA64:EPO (1:2, w/w) and SOL:EPO (1:1, w/w) were finally chosen as the most promising carriers. The results suggest that ATg may be useful in predicting miscibility in the thermal process.

3.2. CBZ-polymer miscibility under thermal processing conditions

3.2.1. MT-DSC studies

MT-DSC studies were performed to identify whether an amorphous solid dispersion of CBZ was formed by HME. Compared to the sharp melting peak of pure CBZ, the endothermic peaks in physical mixtures broadened during the first heating cycle and then disappeared in the second heating cycle. This is caused by gradual dissolution of the crystalline drug in the molten polymers and complete conversion to the amorphous state during the DSC heating process (Fig. 2).

3.2.2. HSM studies

HSM can provide visual evidence to confirm the results of MT-DSC. Images taken under polarizing light are shown in Fig. 3. Pure CBZ in form III (P-monoclinic crystal) began to melt and

Figure 2 MT-DSC thermograms of (a) CBZ; (b) and (c) physical mixtures of CBZ and VA64:EPO (1:2, w/w); (d) and (e) physical mixtures of CBZ and SOL:EPO (1:1, w/w); (b) and (d) the first heating process; (c) and (e) the second heating process.

transform to form I (needle crystal) at approximately 165 1C and subsequently underwent complete melting at 190 °C. During the rapid cooling process, recrystallization to needle crystals which quickly covered the whole field of view occurred. For both physical mixtures, extensive solubilization of CBZ in molten VA64:EPO (1:2, w/w) and SOL:EPO (1:1, w/w) occurred at approximately 165 °C and was complete at 190 °C. Upon rapid cooling and reheating, the crystals of CBZ were no longer observed, indicating miscibility of CBZ and polymer combinations and the formation of amorphous CBZ after the heat-cool cycle. The visual observations correlated well with the results of MT-DSC and provide strong evidence for the formation of a solid dispersion containing amorphous CBZ by HME.

It was noticed that CBZ crystals were almost melted and transformed to the amorphous state in extrudates produced at 165 1C. For physical mixtures of the same formulations, however, CBZ crystals were not dissolved completely when the temperature was maintained at 165 1C for 15 min which is obviously longer than the residence time of the HME process. This clearly demonstrates that the dissolution of CBZ in molten polymers is facilitated by the input of adequate shear forces in the HME process.

3.3. Stability of CBZ and formation of solid dispersions

3.3.1. The effect of processing temperatures

TGA was used to evaluate the thermal stability of materials during

the heating process. As illustrated in Fig. 4A, VA64, SOL and

Polymer combination of solid dispersions of carbamazepine

Figure 3 HSM images of phase transition.

EPO were thermally stable up to 230, 220 and 200 °C, respectively, although 2.81% and 2.01% free water was lost at temperatures below 100 °C for VA64 and SOL, respectively. Mass loss or degradation of CBZ was observed at its Tm (190 °C) indicating the extrusion temperature should be set below 190 °C. The thermal stability of CBZ in different formulations during the HME process and the appearance of corresponding extrudates are presented in Table 2.

Formulations 1-6 consist of CBZ and single polymers (VA64, SOL or EPO). Initially, HME was attempted at 185 °C but obvious browning of extrudates was evident accompanied by respectively 18.5%, 11.6% and 9.4% drug degradation. It has been reported that, in the HME process, the decomposition of thermally unstable substances may occur at temperatures much lower than that predicted by TGA of the pure drug because the crystalline drug dissolves in the molten polymer and is gradually transformed to

Figure 4 TGA of (A) CBZ and polymers and (B) physical mixtures (PM) of formulations 7 and 8.

Table 2 Formulations, CBZ content remaining after HME and appearance and moldability of extrudates.

Formulation Polymer Process temp. (°C) CBZ content (%) Appearance Moldability

1 VA64 185 81.5 Transparent GOOD

2 VA64 165 ND CBZ crystals present GOOD

3 SOL 185 88.4 Transparent GOOD

4 SOL 165 ND CBZ crystals present GOOD

5 EPO 185 90.6 Opaque POOR

6 EPO 165 ND Opaque POOR

7 VA64:EPO (1:2, w/w) 165 98.6 Semi-translucent GOOD

8 SOL:EPO (1:1, w/w) 165 96.7 Semi-translucent GOOD

ND: not determined.

the amorphous state. In addition, the intensive mixing and high screw speed also probably contribute to decreasing the temperature of onset of degradation38. In order to confirm this, TGA was conducted on physical mixtures corresponding to formulations 7 and 8 and degradation observed at approximately 172 1C and 175 1C, respectively, which are much lower than the degradation temperature of pure CBZ (Fig. 4B).

Based on these observations, HME was carried out at 165 1C. For formulations 2 and 4, CBZ was not completely dissolved in the single polymers (VA64 or SOL) as small amounts of CBZ particles were observed in the extrudates. This phenomenon indicates the poor miscibility of CBZ and VA64 (SOL) which is assumed to result from the high melt viscosity of the polymers under such HME conditions. We attempted to facilitate the dissolution rate of CBZ by speeding up the extruder screw speed or extending the residence time but significant degradation occurred (data not shown). For formulations 5 and 6, extrudates could not be collected conveniently possibly due to the low Tg (48 1C) of EPO which caused very low melt viscosity and poor moldability at 185 °C.

3.3.2. The effect of polymer combinations on formation of solid dispersions and moldability of extrudates

To confirm the miscibility improvement of polymer combinations, formulations 7 and 8, consisting of CBZ and polymer combinations VA64/EPO or SOL/EPO, were extruded at 165 °C. Semi-translucent and homogeneous extrudates were obtained with only slight degradation of CBZ. The greater miscibility of these polymer combinations may be due to decreases in the viscosity of the molten systems. As the torque value is proportional to viscosity, the impact of EPO on the torque values of molten polymer combinations was investigated. Incorporation of EPO was found to produce a significant

plasticizing effect on the viscosity of VA64 and SOL as indicated by large reductions in torque values (Fig. 5).

The plasticizing effect of EPO on the viscosity of a molten polymer combination leads to an improvement in the mixing homogeneity within the chamber and a decrease in the thickness of the mass transfer boundary layer around drug particles, thus increasing the dissolution rate of CBZ in the molten polymer. A similar effect of viscosity on drug-polymer miscibility had been previously reported38. Suzuki and Sunada39 used DSC to show that high viscosity limits the miscibility of nifedipine and HPMC. Fu et al.32 also reported that the high viscosity of VA64 under extrusion at 120-130 °C prevented uniform mixing with other ingredients due to its high softening point (approximately 180 °C) and resulted in a lower drug release.

The results of these studies suggest solubility parameters are limited in their ability to predict miscibility of molten systems as the thermal properties of polymers such as melt viscosity are not taken into account. Furthermore, the dissolution capability of a drug in a molten polymer could also be underestimated even when thermal factors are considered. As indicated in the HSM mis-cibility analysis, mixing homogeneity was limited due to the absence of any shear forces. Hence it is reasonable to suggest that the thermal properties of polymers such as melt viscosity and Tg should be combined with solubility parameters to improve mis-cibility prediction.

Although the dissolution of CBZ should be improved by decreasing the melt viscosity of the carrier, the moldability of resulting extrudates will suffer if the viscosity is decreased too much. Thus, taking into account both the dispersion state of CBZ and the moldability of extrudates, extrudates made from combinations of EPO and VA64 (SOL) with a fixed CBZ loading of 30% were evaluated. VA64:EPO (1:2, w/w) and SOL:EPO (1:1, w/w)

Figure 5 Torque values of (A) VA64/EPO combinations and (B) SOL/EPO combinations at different temperatures.

Figure 6 DSC curves of (A) CBZ, polymers and physical mixtures (PM) and extrudates (Ext) from formulation 7 and (B) physical mixtures (PM) of formulation 8 and extrudates (Ext) from SOL:EPO (1:1, w/w) with different drug loading.

were identified as the most suitable carriers (data not shown) and the physicochemical properties, dissolution and stability of corresponding solid dispersions were fully determined.

3.4. Physical characterization 3.4.1. DSC profiles

DSC profiles of CBZ, polymers, extrudates and corresponding physical mixtures of formulations 7 and 8 are shown in Fig. 6A. In the thermogram of CBZ, the small endothermic peak at 169.8 °C followed by a sharp endothermic peak at 190.4 °C suggest a polymorphic transformation of CBZ. These two endothermic peaks were attributed to the melting of the original form III, and the melt-recrystallized form I of CBZ, respectively40.

Although peak areas were small, endothermic peaks at approximately 150-160 °C corresponding to the crystalline drug were still present in the DSC thermograms of extrudates indicating that most CBZ was dispersed in an amorphous state with a small amount remaining in a crystalline state. By comparing the endothermic enthalpy of extrudates with that of pure CBZ, it was calculated that approximately 8.2% and 7.7% undissolved drug crystals are present in the extrudates from formulations 7 and 8, respectively.

Since incomplete dissolution of drug is usually due to overloading, solid dispersions with different drug loading were prepared in SOL:EPO (1:1, w/w) and the corresponding DSC curves analyzed. As shown in Fig. 6B, the enthalpy and onset temperature of drug endothermic peaks gradually decreased and finally disappeared as drug loading decreased from 30% to 15%. This suggests that the polymer combination was saturated with CBZ at the 30% loading. Generally, complete transformation to the amorphous

0 10 20 30 40

Figure 7 XRD profiles of CBZ, polymers, physical mixtures (PM) and extrudates (Ext) from formulations 7 and 8.

state results in better solubilization but, considering the high CBZ content of tablets (100 mg or 200 mg per tablet), a high drug loading may be required in preparing such tablets by HME if drug stability can be maintained over the desired shelf life.

3.4.2. X-ray diffraction

XRD profiles, shown in Fig. 7, reveal no evidence of CBZ crystals in either extrudate. This inconsistency probably results from the lower sensitivity of XRD compared to DSC to detect small-sized crystals even when their concentration in the sample is above the limit of detection41. Since micron sized crystals were also shown

4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 600 Wavenumber (cm1)

Figure 8 FTIR spectra of CBZ, polymers, physical mixtures (PM) and extrudates (Ext) from formulations 7 and 8.

to be present in extrudates by HSM, It can be concluded that CBZ was mainly present in an amorphous form accompanied by a small amount of a microcrystalline form.

3.4.3. FTIR characterization

Possible interactions between drug and polymers in extrudates were investigated by FTIR. As shown in Fig. 8, CBZ form III has characteristic peaks (cm-1) at 3465 (free anti-NH stretching), 3157 (hydrogen bonded syn-NH stretching), 1677 (amide C=O stretching), 1605 and 1594 (aromatic and C=C stretching)42. In CBZ, the NH2 and C=O groups are capable of forming hydrogen bonds with polymers.

Polymer peaks indicative of an interaction with drug include the hydrogen bonding acceptor C=O stretching and the hydrogen bond donor O-H stretching. The relevant peaks are as follows: VA64—the vinyl pyrrolidone monomer C=O at 1678 cm-1 and the vinyl acetate monomer C=O at 1740 cm-1; EPO—the ester C=O at 1731 cm-1; SOL—the vinyl acetate monomer ester C=O at 1741 cm-1, the vinyl caprolactam monomer C=O at 1643 cm-1 and the O-H group at 3464 cm-1. Any changes in the wave numbers or shapes of these peaks reflect an interaction between CBZ and the respective polymers.

The FTIR spectra of physical mixtures appear as the summations of the individual spectra of components indicating no interactions between CBZ and polymers in physical mixtures. In the spectrum of the extrudate from formulation 7, the peaks at

3465 cm-1 and 3157 cm-1 were replaced by a weak peak at 3485 cm-1 which is characteristic of CBZ form I 43. This indicates that most CBZ (initially in form III) is transformed into an amorphous form after HME and only a small amount is present as crystalline form I. The sharp peak at 1678 cm-1 is broader and shifted to 1682 cm-1 which could indicate the formation of an intermolecular hydrogen bond between CBZ and VA64. There was no evidence of an interaction between CBZ and EPO as the peak at 1731 cm-1 remained unchanged.

The residual CBZ crystals in extrudates from formulation 8 were present in both form I and form III as weak peaks at 3485 and

3466 cm-1 were observed. Four characteristic peaks at 1741, 1731, 1677 and 1643 cm-1 were replaced by three slightly broader peaks at 1735, 1683 and 1633 cm-1 reflecting a possible interaction among CBZ, SOL and EPO. This interaction between drug and polymers can facilitate miscibility and, by inhibiting the movement of molecules to some extent, improve the physical stability of extrudates.

0 10 20 30 40 50 60 70 80 90 100 110 120 Time (min)

Figure 9 Dissolution profiles of CBZ, physical mixtures (PM) and extrudates (Ext) from formulations 7 and 8 (n = 3).

3.5. Dissolution study

The dissolution profiles of pure CBZ, extrudates and corresponding physical mixtures are shown in Fig. 9. In comparison to pure CBZ, the dissolution rate of physical mixtures was slightly increased probably because the hydrophilic polymers can wet the surface of drug particles and act to solubilize them. The dissolution of extrudates was markedly enhanced with total release occurring within 20 min. This clearly shows that a remarkable improvement in dissolution performance was achieved by HME.

3.6. Accelerated stability testing

The results of storage under accelerated stability conditions for 3 months are summarized in Table 3. Some characteristic peaks of crystalline drug were observed to grow in the XRD profiles of extrudates from formulation 7 after 1 month, indicating recrys-tallization of amorphous drug during storage. Despite this, a decrease in the dissolution performance did not occur. Two possible explanations for the lack of change dissolution could be first, that the recrystallized CBZ crystals were small in size and sufficiently dispersed within the hydrophilic polymers, and second, that solubilization of drug due to the hydrophilic polymers occurred as previously demonstrated in the dissolution study.

In the case of extrudates from formulation 8, no substantial recrystallization was observed by XRD over the 3 months storage suggesting CBZ is more stable in this formulation. This may be because SOL can engage in more extensive hydrogen bonding with CBZ and EPO, resulting in less molecular mobility. In addition, there were no significant variations in content of drug and related substances nor in dissolution profiles after storage. Taken together, these results imply that extrudates from formulation 8 are stable over the storage period and that the small quantity of CBZ microcrystals (approximately 7.7%) found after storage has little effect on the stability of the extrudates.

3.7. Hygroscopicity

It is well documented that hygroscopicity plays an important role in determining the stability of solid dispersions since water may act as a plasticizer and enhance the mobility of the polymer chains thus decreasing the Tg of amorphous substances and encouraging

Table 3 Physicochemical stability of extrudates from formulations 7 and 8 after accelerated stability testing.

Sample Condition, time (40 °C/75% RH) Appearance XRD Q5/Q2o (%) CBZ/Related substances (%)

Extrudates from formulation 7 Package, 0 month Loose powder A/C 96.3/98.3 98.6/0.42

Package, 1 month Loose powder C/A 89.9/99.8 98.8/0.51

Package, 2 months Loose powder C/A 90.5/99.8 98.4/0.52

Package, 3 months Loose powder C/A 91.0/99.2 98.2/0.53

Extrudates from formulation 8 Package, 0 month Loose powder A/C 85.2/99.8 96.7/0.40

Package, 1 month Loose powder A/C 90.9/98.8 98.2/0.42

Package, 2 months Loose powder A/C 90.1/99.9 96.6/0.50

Package, 3 months Loose powder A/C 92.4/99.2 96.4/0.49

A/C: amorphous with a small amount of microcrystals, C/A: predominantly crystalline; Q5 and Q2q: percent drug release after 5 and 20 min, respectively.

Figure 10 Moisture uptake (hygroscopicity) of CBZ, polymers and extrudates (Ext) from formulations 1, 3, 5, 7 and 8.

recrystallization. EPO is less hygroscopic than either VA64 or SOL so that hygroscopicity of extrudates should be decreased by introducing EPO into formulations. To investigate this, the hygroscopicity of CBZ, polymers and exrtudates was investigated at different RH values (Fig. 10).

VA64 became transparent and SOL agglomerated when stored at 75% RH and 92.5% RH for 7 days, while EPO and CBZ remained dry powders at all RH consistent with their lower hygroscopicity. It is interesting to note that the order of hygro-scopicity of extrudates is the same as that of the polymers indicating the hygroscopicity of an extrudate is decreased by introducing a polymer with lower hygroscopicity such as EPO.

Although conventional plasticizers such as citrate esters44,45, surfactants46 and low-molecular weight PEGs47 can be used to decrease the viscosity of molten polymers, the toxicity of plasticizers restricts their application. Combing polymers with different Tg values may be a better choice since they not only avoid the disadvantage of conventional plasticizers but also impart synergistic advantages such as decreasing the hygroscopicity of solid dispersions as demonstrated in our study.

4. Conclusions

The selection of polymer and processing temperature are important concerns in the preparation of solid dispersions containing a thermally unstable drug by HME. Compared to single polymers,

polymer combinations may benefit the manufacture of amorphous solid dispersions by improving drug-polymer miscibility and decreasing the processing temperature. Using a high drug loading, solid dispersions containing amorphous and microcrystalline CBZ could be obtained by HME using SOL:EPO (1:1, w/w). These dispersions had superior in vitro dissolution of the poorly water-soluble drug and remained physicochemically stable for up to 3 months under accelerated stability testing conditions used for CBZ tablets.

Acknowledgments

We would like to thank the Resin Division Technical Service Center of Evonik Industries (Shanghai, China) for providing the extrusion equipment and Eudragit EPO. We also thank the Technical Service Center of BASF, Shanghai, China for supplying Kollidon VA64 and Soluplus.

References

1. Singh A, Worku ZA, van den Mooter G. Oral formulation strategies to improve solubility of poorly water-soluble drugs. Expert Opin Drug Deliv 2011;8:1361-78.

2. Stella VJ, Nti-Addae KW. Prodrug strategies to overcome poor water solubility. Adv Drug Deliv Rev 2007;59:677-94.

3. Blagden N, de Matas M, Gavan PT, York P. Crystal engineering of active pharmaceutical ingredients to improve solubility and dissolution rates. Adv Drug Deliv Rev 2007;59:617-30.

4. Xia D, Cui F, Piao H, Cun D, Jiang Y, Ouyang M, et al. Effect of crystal size on the in vitro dissolution and oral absorption of nitrendipine in rats. Pharm Res 2010;27:1965-76.

5. Hussain MD, Saxena V, Brausch JF, Talukder RM. Ibuprofen-phospholipid solid dispersions: improved dissolution and gastric tolerance. Int J Pharm 2012;422:290-4.

6. Patel JS, Patel RP. Preparation, characterization and in vitro dissolution study of Nitrazepam: cyclodextrin inclusion complex. J Pharm Bioallied Sci 2012;4(Suppl):S106-7.

7. Leuner C, Dressman J. Improving drug solubility for oral delivery using solid dispersions. Eur J Pharm Biopharm 2000;50:47-60.

8. Serajuddin AT. Solid dispersions of poorly water-soluble drugs: early promises, subsequent problems, and recent breakthroughs. J Pharm Sci 1999;88:1058-66.

9. Zajc N, Obreza A, Bele M, Srcic S. Physical properties and dissolution behaviour of nifedipine/mannitol solid dispersions prepared by hot melt method. Int J Pharm 2005;291:51-8.

10. Vemavarapu C, Mollan MJ, Needham TE. Coprecipitation of pharmaceutical actives and their structurally related additives by the RESS process. Powder Tech 2009;189:444-53.

11. Badens E, Majerik V, Horväth G, Szokonya L, Bosc N, Teillaud E, et al. Comparison of solid dispersions produced by supercritical antisolvent and spray-freezing technologies. Int J Pharm 2009;377:25-34.

12. Inam MA, Ouattara S, Frances C. Effects of concentration of dispersions on particle sizing during production of fine particles in wet grinding process. Powder Tech 2011;208:329-36.

13. Almeida A, Possemiers S, Boone MN, de Beer T, Quinten T, van Hoorebeke L, et al. Ethylene vinyl acetate as matrix for oral sustained release dosage forms produced via hot-melt extrusion. Eur J Pharm Biopharm 2011;77:297-305.

14. Verhoeven E, de Beer TR, Schacht E, van den Mooter G, Remon JP, Vervaet C. Influence of polyethylene glycol/polyethylene oxide on the release characteristics of sustained-release ethylcellulose mini-matrices produced by hot-melt extrusion: in vitro and in vivo evaluation. Eur J Pharm Biopharm 2009;72:463-70.

15. Gryczke A, Schminke S, Maniruzzaman M, Beck J, Douroumis D. Development and evaluation of orally disintegrating tablets (ODTs) containing Ibuprofen granules prepared by hot melt extrusion. Colloids Surf B 2011;86:275-84.

16. Witzleb R, Kanikanti VR, Hamann HJ, Kleinebudde P. Solid lipid extrusion with small die diameters-electrostatic charging, taste masking and continuous production. Eur J Pharm Biopharm 2011;77:170-7.

17. Crowley MM, Zhang F, Repka MA, Thumma S, Upadhye SB, Battu SK, et al. Pharmaceutical applications of hot-melt extrusion: part I. Drug Dev Ind Pharm 2007;33:909-26.

18. Dinunzio JC, Brough C, Hughey JR, Miller DA, Williams 3rd RO, McGinity JW. Fusion production of solid dispersions containing a heat-sensitive active ingredient by hot melt extrusion and Kinetisol dispersing. Eur J Pharm Biopharm 2010;74:340-51.

19. Chokshi RJ, Shah NH, Sandhu HK, Malick AW, Zia H. Stabilization of low glass transition temperature indomethacin formulations: impact of polymer-type and its concentration. J Pharm Sci 2008;97:2286-98.

20. Qi S, Gryczke A, Belton P, Craig DQM. Characterization of solid dispersions of paracetamol and EUDRAGIT®E prepared by hot-melt extrusion using thermal, microthermal and spectroscopic analysis. Int J Pharm 2008;354:158-67.

21. Liu HJ, Wang P, Zhang XY, Shen F, Gogos CG. Effects of extrusion process parameters on the dissolution behavior of indomethacin in Eudragit® EPO solid dispersions. Int J Pharm 2010;383:161-9.

22. Prodduturi S, Urman K, Otaigbe J, Repka M. Stabilization of hot-melt extrusion formulations containing solid solutions using polymer blends. AAPS Pharm Sci Tech 2007;8. Article 50.

23. Kalivoda A, Fischbach M, Kleinebudde P. Application of mixtures of polymeric carriers for dissolution enhancement of fenofibrate using hot-melt extrusion. Int J Pharm 2012;429:58-68.

24. Sakurai A, Sakai T, Sako K, Maitani Y. Polymer combination increased both physical stability and oral absorption of solid dispersions containing a low glass transit ion temperature drug: physico-chemical characterization and in vivo study. Chem Pharm Bull 2012;60:459-64.

25. Sehic S, Betz G, Hadzidedic S, El-Arini SK, Leuenberger H. Investigation of intrinsic dissolution behavior of different carbamaze-pine samples. Int J Pharm 2010;386:77-90.

26. Liu X, Lu M, Guo Z, Huang L, Feng X, Wu C. Improving the chemical stability of amorphous solid dispersion with cocrystal technique by hot melt extrusion. Pharm Res 2012;29:806-17.

27. Hoftyzer PJ, Krevelen DWV. Properties of polymers. Amsterdam: Elsevier; 1976.

28. Zheng JM. Polymer science in pharmaceutics. Beijing: Chinese Medical Science Press; 2000.

29. Foster A, Hempenstall J, Tucker I, Rades T. Selection of excipients for melt extrusion with two poorly water soluble drugs by solubility parameter calculation and thermal analysis. Int J Pharm 2001;226:147-61.

30. Chokshi RJ, Sandhu HK, Iyer RM, Shah NH, Malick AW, Zia H. Characterization of physico-mechanical properties of indomethacin and polymers to assess their suitability for hot-melt extrusion process as a means to manufacture solid dispersion/solution. J Pharm Sci 2005;94:2463-74.

31. Flicker F, Eberle VA, Betz G. Variability in commercial carbamaze-pine samples-impact on drug release. Int J Pharm 2011;410:99-106.

32. Fu J, Xu L, Wang X, Zhang S, Tao X, Zhao X, et al. Nimodipine (NM) tablets with high dissolution containing NM solid dispersions prepared by hot-melt extrusion. Drug Dev Ind Pharm 2011;37:934-44.

33. Tian Y, Booth J, Meehan E, Jones DS, Li S, Andrews GP. Construction of drug-polymer thermodynamic phase diagrams using Flory-Huggins interaction theory: identifying the relevance of temperature and drug weight fraction to phase separation within solid dispersions. Mol Pharm 2013;10:236-48.

34. The State Pharmacopoeia Committee of China. The pharmacopoeia of the People's Republic of China. Beijing: Chinese Medicine Science and Technology Press; 2010.

35. Greenlagh DJ, Williams AC, Timmins P, York P. Solubility parameters as predictors of miscibility in solid dispersions. J Pharm Sci 1999;88:1182-90.

36. Patterson JE, James MB, Forster AH, Lancaster RW, Butler JM, Rades T. Preparation of glass solutions of three poorly water soluble drugs by spray drying, melt extrusion and ball milling. Int J Pharm 2007;336:22-34.

37. Sathigari SK, Radhakrishnan VK, Davis VA, Parsons DL, Babu RJ. Amorphous-state characterization of efavirenz—polymer hot-melt extrusion systems for dissolution enhancement. J Pharm Sci 2012;101:3456-64.

38. Hughey JR, Keen JM, Brough C, Saeger S, McGinity JW. Thermal processing of a poorly water-soluble drug substance exhibiting a high melting point: the utility of KinetiSol® dispersing. Int J Pharm 2011;419:222-30.

39. Suzuki H, Sunada H. Comparison of nicotinamide, ethylurea and polyethylene glycol as carriers for nifedipine solid dispersion systems. Chem Pharm Bull 1997;45:1688-93.

40. Grzesiak AL, Lang M, Kim K, Matzger AJ. Comparison of the four anhydrous polymorphs of carbamazepine and the crystal structure of form I. J Pharm Sci 2003;92:2260-71.

41. Munson EJ. Analytical techniques in solid-state characterization. In: Qiu Y, Chen Y, Zhang GGZ, Liu L, Porter WR, editors. Developing solid oral dosage forms: pharmaceutical theory and practice. 1st ed. Oxford: Elsevier; 2009. p. 61-74.

42. Thomas S, Mathela CS, Agarwal A, Paul SK. Identification and structural elucidation of an unknown impurity in carbamazepine active pharmaceutical ingredient by liquid chromatography-tandem mass spectrometry and semi-preparative chromatographic isolation. J Pharm BiomedAnal 2011;56:423-8.

43. Rustichelli C, Gamberini G, Ferioli V, Gamberini MC, Ficarra R, Tommasini S. Solid-state study of polymorphic drugs:carbamazepine. J Pharm Biomed 2000;23:41-54.

44. Repka MA, Gerding TG, Repka SL, McGinity JW. Influence of plasticizers and drugs on the physical-mechanical properties of hydroxypropylcellulose films prepared by hot melt extrusion. Drug Dev Ind Pharm 1999;25:625-33.

45. Aitken-Nichol C, Zhang F, McGinity JW. Hot melt extrusion of acrylic films. Pharm Res 1996;13:804-8.

46. Ghebremeskel AN, Vemavarapu C, Lodaya M. Use of surfactants as plasticizers in preparing solid dispersions of poorly soluble API: stability testing of selected solid dispersions. Pharm Res 2006;23:1928-36.

47. Follonier N, Doelker E, Cole ET. Evaluation of hot-melt extrusion as a new technique for the production of polymer-based pellets for sustained release capsules containing high loadings of freely soluble drugs. Drug Dev Ind Pharm 1994;20:1323-39.