Article pubs.acs.org/molecularpharmaceutics
Mechanism of Amorphous Itraconazole Stabilization in Polymer Solid Dispersions: Role of Molecular Mobility Sunny P. Bhardwaj,†,‡ Kapildev K. Arora,†,§ Elizabeth Kwong,∥,⊥ Allen Templeton,∥ Sophie-Dorothee Clas,#,¶ and Raj Suryanarayanan*,† †
Department of Pharmaceutics, University of Minnesota, Minneapolis, Minnesota 55455, United States Merck Research Laboratories, Merck & Co., Kenilworth, New Jersey 07033, United States # Merck Research Laboratories, Merck & Co., West Point, Pennsylvania 19486, United States ∥
ABSTRACT: Physical instability of amorphous solid dispersions can be a major impediment to their widespread use. We characterized the molecular mobility in amorphous solid dispersions of itraconazole (ITZ) with each polyvinylpyrrolidone (PVP) and hydroxypropylmethylcellulose acetate succinate (HPMCAS) with the goal of investigating the correlation between molecular mobility and physical stability. Dielectric spectra showed two mobility modes: α-relaxation at temperatures above the glass transition temperature (Tg) and β-relaxation in the sub-Tg range. HPMCAS substantially increased the α-relaxation time, with an attendant increase in crystallization onset time and a decrease in crystallization rate constant, demonstrating the correlation between α-relaxation and stability. The inhibitory effect on α-relaxation as well as stability was temperature dependent and diminished as the temperature was increased above Tg. PVP, on the other hand, affected neither the α-relaxation time nor the crystallization onset time, further establishing the link between α-relaxation and crystallization onset in solid dispersions. However, it inhibited the crystallization rate, an effect attributed to factors other than mobility. Interestingly, both of the polymers acted as plasticizers of β-relaxation, ruling out the latter’s involvement in physical stability. KEYWORDS: itraconazole, solid dispersion, PVP, HPMCAS, dielectric spectroscopy, molecular mobility, crystallization kinetics, crystallization onset, synchrotron radiation
■
comprehensive picture of the molecular mobility.14,22 Moreover, temperature dependence of molecular mobility in the glassy and supercooled liquid regions could be very different.23−25 These characteristics are not reflected by Tg, making it an unreliable indicator of molecular mobility. Thus, it is important to comprehensively study the molecular dynamics in the amorphous state. Molecular motions in an amorphous matrix can generally be divided into two categories: global mobility or cooperative αrelaxation and noncooperative faster motions (β- or secondary relaxations). Even though α-relaxation is considered to be the major mode responsible for instability,13,26−29 some reports have also hinted at the possible role of β-relaxations.30−33 Dynamic dielectric spectroscopy (DDS) is one of the most widely used techniques to study different molecular motions over a wide temperature range.34−37 Historically, most of the DDS work has been done on single component systems.12,13,26−29 Very few studies have looked at the dielectric relaxations in solid dispersions, and even fewer have attempted
INTRODUCTION Amorphous drug−polymer molecular dispersions have received considerable attention in light of their ability to enhance the aqueous solubility of poorly water-soluble pharmaceuticals. Physical instability leading to crystallization, however, is a major concern as it will lead to the loss of the solubility advantage. Hydrophilic polymers such as poly(vinylpyrrolidone) (PVP), polyethylene glycol, and hydroxypropylmethylcellulose (HPMC) are popular because of their potential ability to form amorphous solid drug dispersions and also stabilize the drug in the amorphous state.1 Different stabilization mechanisms have been proposed for these polymers including specific drug−polymer interactions such as hydrogen bonding, antiplasticization effect, and the polymer’s ability to act as a physical barrier.2−9 However, in light of the incomplete understanding of the stabilization mechanism, polymer selection is still largely empirical. In light of its potential correlation with physical instability, molecular mobility in the amorphous state has been extensively investigated.10−19 Many studies on the stability of solid dispersions have focused on the glass transition temperature (Tg), with the goal of raising Tg by the addition of a polymer.9,20,21 Since amorphous systems are often characterized by multiple relaxations, Tg alone does not provide a © 2014 American Chemical Society
Received: Revised: Accepted: Published: 4228
June 29, 2014 September 5, 2014 October 2, 2014 October 17, 2014 dx.doi.org/10.1021/mp5004515 | Mol. Pharmaceutics 2014, 11, 4228−4237
Molecular Pharmaceutics
Article
a correlation with physical stability.11,38,39 A fundamental unanswered question is: Does the addition of polymer modulate the physical stability of the system through its effect on molecular mobility? In order to comprehensively understand the stabilization mechanism in molecular dispersions, answering this question is imperative. In a previous report, we demonstrated a strong coupling between physical stability and α-relaxation in amorphous itraconazole (ITZ), an antifungal agent.28 In this work, we systematically studied the different relaxations in model ITZ− polymer dispersions. PVP and hydroxypropylmethycellulose acetate succinate (HPMCAS) were the model polymers. There was a pronounced difference in the stabilizing effect of the two polymers, and this could be correlated with the molecular mobility of the two solid dispersions. In addition, the molecular mobility and physical stability exhibited a strikingly similar temperature dependence: additional strong evidence of the potential role of mobility on stability. Another important observation was the difference in the dependence of crystallization onset and kinetics on mobility. While αrelaxation was the major determinant of crystallization onset, factors other than mobility were also influencing the kinetics. These findings not only improve our understanding of the basic mechanism of polymer stabilization but would also enable rationalizing polymer selection in the formulation of solid dispersions.
instrument was calibrated with sapphire. The amplitude of the temperature modulation was ±0.5 °C with a period of 40 s and an underlying heating rate of 1 °C/min. Powder X-ray Diffractometry (XRD). A powder X-ray diffractometer (D8 ADVANCE; Bruker AXS, Madison, WI) equipped with a variable temperature stage (TTK 450; Anton Paar, Graz-Straßgang, Austria) and Si strip one-dimensional detector (LynxEye; Bruker AXS) was used. Samples were exposed to Cu Kα radiation (40 kV × 40 mA) over the angular range of 5−35° 2θ with a step size of 0.05° and a dwell time of 1 s. For the isothermal crystallization studies at temperatures above Tg, the sample stage was maintained at the desired experimental temperature and XRD patterns were obtained periodically. For the studies below Tg, the powder samples, in tightly closed containers, were stored in ovens maintained at the desired temperature and periodically subjected to XRD. Dielectric Spectroscopy. Using a broadband dielectric spectrometer (Novocontrol Alpha-A high performance frequency analyzer, Novocontrol Technologies, Germany), isothermal dielectric measurements were conducted over the frequency range of 10−1 to 106 Hz at several temperatures from 0 to 105 °C. About 100 mg of sample was placed between two round copper electrodes (20 mm diameter) and a PTFE spacer. The PTFE spacer (thickness, 1 mm; area, 59.69 mm2; and capacity, 1.036 pF) was used to keep the sample confined between electrodes at high temperatures and also to minimize errors due to stray capacitance or edge effects. The sample temperature was maintained with a Novocool Cryosystem temperature controller. Samples of different thickness were analyzed to ensure that, in the frequency region of interest, there was no interference from interfacial polarization. The Havriliak−Negami function (eq 1) was used to fit the dielectric data so as to obtain the average relaxation time (τ) and shape parameters (βHN and γHN).
■
EXPERIMENTAL SECTION Materials. Crystalline ITZ (Bepharm Limited, Shanghai, China; purity ∼98%) was used as obtained without further purification. PVP (grade PF17, pyrogen free with a K value of 17) was obtained from BASF (Ludwigshafen, Germany) and HPMCAS (grade LF) from Shin Etsu Chemical Co. (Tokyo, Japan). All the solvents were of HPLC grade. Spray-Drying. The solute solution in tetrahydrofuran (1% w/v) was introduced at a feed rate of 5 mL/min (Micro-Spray Co-flow spray-dryer/chiller, ProCepT, Zelzate, Belgium). The inlet temperature was maintained at 110 °C during the preparation of solid dispersions and at 105 °C for preparing amorphous ITZ. The resulting outlet temperature range was 55 to 60 °C. The flow rate of the drying gas was maintained at 0.32−0.35 m3/min, while cooling air and side air flow was set at 0.15 m3/min. Atomization air flow was maintained in the 0.003−0.005 m3/min range. Three spray-dried systems were obtained: pure ITZ, ITZ−PVP (90% w/w drug load) solid dispersion, and ITZ−HPMCAS (90% w/w drug load) solid dispersion. All the spray-dried materials were stored in capped glass containers, which were in turn placed in a larger capped glass jar containing anhydrous calcium sulfate and stored at −20 °C. Further handling, including sample preparation for thermal, X-ray, and dielectric analyses, was done in a glovebox maintained at relative humidity < 5% (room temperature). All the samples were found to be X-ray amorphous with a water content
