Maintaining Supersaturation in Aqueous Drug Solutions - American

Dec 5, 2012 - Department of Industrial and Physical Pharmacy, College of Pharmacy, Purdue University, West Lafayette, Indiana, United States. ‡. Dep...
0 downloads 10 Views 2MB Size
Article pubs.acs.org/crystal

Maintaining Supersaturation in Aqueous Drug Solutions: Impact of Different Polymers on Induction Times Published as part of the Crystal Growth and Design virtual special issue of selected papers presented at the 10th International Workshop on the Crystal Growth of Organic Materials (CGOM10) Grace A. Ilevbare,† Haoyu Liu,‡ Kevin J. Edgar,‡ and Lynne S. Taylor*,† †

Department of Industrial and Physical Pharmacy, College of Pharmacy, Purdue University, West Lafayette, Indiana, United States Department of Sustainable Biomaterials, College of Natural Resources and Environment, Virginia Tech, Blacksburg, Virginia, United States



ABSTRACT: The use of polymeric additives is an increasingly common approach for inhibiting crystallization from the supersaturated solutions used to enhance the delivery of poorly water-soluble drugs. Maintaining supersaturation by employing polymeric additives depends on their ability to inhibit nucleation and crystal growth. In solution crystallization, nucleation initiates the process of crystallization, and therefore adequate control over crystallization from supersaturated solutions cannot be achieved without understanding the mechanism of nucleation inhibition by polymers. In this study, the effectiveness of a group of chemically diverse polymers, including several recently synthesized cellulose derivatives, on induction times in aqueous solutions was quantified. Nucleation was quantified by measuring the induction time for the appearance of particulates from unseeded desupersaturation experiments for three model pharmaceutical compounds: celecoxib, efavirenz, and ritonavir. Induction times in the absence of the polymers varied from approximately 2 min for celecoxib to 2 h for ritonavir. Some polymers were found to extend induction times by up to a factor of 5−6 at the highest supersaturations tested. The effectiveness of the various polymers appeared to depend on the hydrophobicity of the polymer relative to that of the drug. The hydrophobicity of the polymer most likely influences the ability of the polymer to form polymer−solute interactions relative to polymer−solvent and polymer−polymer interactions. Polymer−solute interactions would be expected to hinder the reorganization of a cluster of solute molecules into an ordered crystal structure.



INTRODUCTION The use of high energy forms, such as amorphous solids, to generate supersaturated aqueous solutions of poorly watersoluble drugs has become a common approach for improving drug delivery.1,2 In the supersaturated state, the thermodynamic activity of a drug is increased beyond its solubility limit; thus there is an increased driving force for transit into and across the biological membrane. 3 However, due to the inherent thermodynamic instability of the supersaturated state, which will eventually lead to crystallization of the poorly water-soluble drug, prolonged maintenance of supersaturation in the gastrointestinal (GI) tract may be difficult. It has been shown that the addition of polymeric materials can be used to delay crystallization from supersaturated solutions;4−12 however, to date, very few predictive tools exist for the rational selection of the optimum polymer for a given solute. Stabilization of supersaturated systems by polymeric additives depends on their ability to inhibit nucleation and subsequent crystal growth. Although the inhibition of crystallization from aqueous supersaturated solutions of drugs by polymeric additives has been extensively documented,5,9,12 little quantitative work has been published about the impact of these additives on © 2012 American Chemical Society

nucleation and crystal growth kinetics, and little is known about which polymer properties influence their inhibitory potential. In solution crystallization, nucleation plays a decisive role in determining the final crystal properties, including crystal form and size distribution, and hence is of practical importance in pharmaceutical systems. Drug delivery challenges concerning the concentration threshold above which crystallization of an active pharmaceutical ingredient (API) occurs upon administration are related to the kinetic stability of supersaturated solutions, which is in turn regulated by the nucleation mechanisms and kinetics.13 Thus, optimum control over crystallization from supersaturated solutions is facilitated by understanding the fundamentals of nucleation kinetics and the mechanism of inhibition by additives. While adsorption onto solids is generally accepted as the mode of action by which additives inhibit crystal growth,14−16 the means by which additives influence nucleation is more speculative, partly Received: October 3, 2012 Revised: December 4, 2012 Published: December 5, 2012 740

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751

Crystal Growth & Design

Article

experimental nucleation time, tind, can be defined as the sum of the time for critical nucleus formation (true nucleation time, tn) and growth to detectable size, tg.27

because the molecular-level processes involved are inaccessible to direct experimental observation. However, molecular simulations on a model system suggest that the relative strength of the pairwise intermolecular interactions between solute, solvent, and additive, as well as size factors, influences the impact of the additive on nucleation kinetics.17 Nucleation is energetically more demanding than crystal growth, and there are supersaturations at which crystal growth proceeds while nucleation is unfavorable.14,18,19 In previous studies,20,21 it was noted that polymeric additives were not very effective at inhibiting the crystal growth of the hydrophobic drugs, ritonavir and felodipine, at the high supersaturations expected to be generated by dissolution of an amorphous formulation. For these systems, desupersaturation was found to be rapid in the presence of crystal seeds. Therefore, it seems essential to prevent the initial formation of crystals through the use of effective nucleation inhibitors in order to maximize supersaturation levels for biologically relevant time periods. This in turn necessitates a more fundamental understanding of polymer properties important for the modification of nucleation kinetics. The objectives of this study were to quantify the impact of polymers on the nucleation behavior of three chemically diverse model compounds and to understand the interplay between polymer and drug properties that influences nucleation kinetics. It is hypothesized that the ability of the polymer to inhibit crystal nucleation is dependent upon the strength of interaction of the polymer with the crystallizing solute molecules. The nucleation behavior of the three model compounds, ritonavir, efavirenz, and celecoxib, was investigated in the absence and presence of a group of chemically diverse polymers. Many of the polymers used in this study have been investigated previously in the context of their ability to inhibit crystal growth of ritonavir22 and include a number of recently synthesized cellulose polymers with greater chemical diversity than commercially available cellulose derivatives. Nucleation was experimentally quantified by measuring the induction time for the appearance of particulates. The following section gives a brief overview of the relationship between nucleation− induction time and nucleation theory.

t ind = tn + tg

The experimental induction time can also be described as the time that elapses between the creation of supersaturation and the first observable change in some physical property of the crystallizing system, e.g., turbidity. Since the experimentally measured induction time is not a fundamental property and is largely dependent on the method of detection used, reliable methods for the determination of induction time periods are quite important. Assuming that steady-state nucleation is achieved very quickly and that tind is mainly composed of tn (i.e., tn ≫ tg), then the induction time for the formation of a critical nucleus may be expressed as follows:27 t ind ∝ J −1

⎛ −ΔGcrit ⎞ ⎟ J = A exp⎜ ⎝ kT ⎠

(4)

where A is the pre-exponential kinetic factor, ΔGcrit is the free energy change associated with the creation of a critical nucleus, k is the Boltzmann constant, and T is the absolute temperature. This kinetic factor is related to the rate of attachment of molecules to the critical nucleus and thus depends on the molecular mobility.28 Classical nucleation theory (CNT) assumes that clusters evolve in size by a sequence of molecular additions until they reach a critical size but does not provide any information about the structure of aggregates or pathways leading from the solution to the solid crystal.29−31 Alternatively, the two-step nucleation model30,31 (which was initially proposed for protein crystallization) proposes that a sufficient-sized disordered cluster of solute molecules forms first, followed by reorganization of that cluster into an ordered structure.29−31 The applicability of the two-step mechanism to both macromolecules and small organic molecules has been demonstrated, suggesting that this mechanism may be important for crystallization processes from solutions.31 Molecular dynamics simulations also generate a two-step nucleation process.17 The organization time of the cluster has been proposed to be the rate-limiting step of the nucleation process, due to the observed increase in organization time of appropriate lattice structures with greater molecular complexity of the solute.30,31

THEORETICAL CONSIDERATIONS Supersaturation. For dilute solutions, the supersaturation ratio, which provides the thermodynamic driving force for nucleation, is given by the ratio of concentrations: C Ceq

(3)

J is the nucleation rate, which is given by classical nucleation theory as



S=

(2)



(1)

where C is the concentration of the supersaturated solution and Ceq is the equilibrium solubility of the crystalline substance at a given temperature. Nucleation-Induction Time. Determination of the induction time, also known as the nucleation time, is a wellestablished method of characterizing crystal formation kinetics.12,23−26 The true nucleation time, tn, is defined as the time that elapses from the creation of supersaturation until nuclei are formed.26,27 This lag phase represents the time necessary to achieve a steady-state distribution of sizes of newly formed nuclei following the creation of supersaturation.27 However, it is not possible to directly measure nucleation time, because the nuclei formed can only be detected after they have grown to an experimentally detectable size. Therefore, the

EXPERIMENTAL SECTION

Materials. Efavirenz (Figure 1) and ritonavir (Figure 1) were purchased from Attix Corporation, Toronto, Ontario, Canada. Celecoxib (Figure 1) was provided by Pfizer Inc. (Groton, MA). Acetonitrile and methanol were purchased from Macron Chemicals (Phillipsburg, NJ). The commercially available polymers were obtained from various sources: poly(vinyl pyrrolidone) K29/32, poly(acrylic acid) and cellulose acetate phthalate (Sigma-Aldrich Co.,St. Louis MO), poly(vinyl pyrrolidone vinyl acetate) K28 (BASF, Germany), hydroxypropyl methyl cellulose 606 grade and hydroxypropyl methyl cellulose acetate succinate AS-MF grade (Shin-Etsu Chemical Co., Ltd. (Tokyo, Japan)), poly(allylamine) and poly(Nisopropylacrylamide) (Polysciences, Inc. Warrington, PA), hydroxypropyl cellulose (Hercules Polymer and Chemicals, Inc. Florida) and carboxymethyl cellulose acetate butyrate (Eastman Chemical Com741

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751

Crystal Growth & Design

Article

Figure 2. Molecular structure of the novel synthesized cellulose derivatives and the substituent groups. These cellulose derivatives are not regioselectively substituted. The abbreviations are presented in Table 1.

Figure 1. Molecular structures of (a) ritonavir, (b) efavirenz, and (c) celecoxib.

Table 2. Synthesized Polymers and Abbreviations Used in This Study

pany, Kingsport, TN). The synthesis of the cellulose ω-carboxyalkanoates has been described previously.32,33 The abbreviations used in this report for the commercially available polymers are presented in Table 1, while the abbreviations for the synthesized cellulose ester polymers22,32 (Figure 2) are presented in Table 2.

polymer cellulose cellulose cellulose cellulose cellulose cellulose cellulose cellulose cellulose cellulose

Table 1. Synthetic Polymers and Abbreviations Used in This Study polymer

abbreviation

poly(vinylpyrrolidone) (K 29/32) poly(vinylpyrrolidone vinyl acetate) (K 28) poly(allylamine) poly(N-iso-propylacrylamide) poly(acrylic acid) poly(4-vinylpyridine N-oxide) hydroxypropyl cellulose cellulose acetate phthalate hydroxypropyl methyl cellulose (606 grade) hydroxypropyl methyl cellulose acetate succinate (AS-MF) carboxymethyl cellulose acetate butyrate

PVP PVPVA PAlAmn Pn-IPAAmd PAA PVPdn-O HPC CAPh HPMC HPMCAS CMCAB

METHODS Solubility Parameter. The solubility parameter (SP) was used to characterize the relative hydrophobicity of the polymers and the model compounds. SP values were estimated using the method proposed by Fedors.34 The method is based on group additive constants and the contribution of a large number of functional groups was evaluated; therefore it requires only knowledge of the structural formula of the compound. The solubility parameter can be evaluated using34 ∑i Δei ∑i Δvi

=

ΔEv V

abbreviation CP Adp CA 320S Adp CA 320S Sub CA 320S Seb CAP Adp 0.85 CAB Adp 0.81 CAP Adp 0.33 CAB Adp 0.25 CAP Seb CAB Seb

volume (V), respectively. The group contributions at 25 °C are presented in ref 34. Solubility Studies. The equilibrium solubilities of ritonavir, efavirenz, and celecoxib were determined in the absence and presence of selected polymers. Prior to adding an excess amount of crystalline compound to 100 mM sodium phosphate buffer, pH 6.8, the polymer was predissolved in the buffer at a concentration of 5 μg/mL. The drug−polymer solution was equilibrated at 37 °C for 48 h. Using ultracentrifugation, the supernatant was separated from the excess solid in solution. Ultracentrifugation was performed at 40 000 rpm (equivalent of 274356g) in an Optima L-100 XP ultracentrifuge equipped with Swinging-Bucket Rotor SW 41 Ti (Beckman Coulter, Inc., Brea, CA). The supernatant was diluted using a combination of mobile phase solvents. Solution concentration was determined using an Agilent 1100 high performance liquid chromatography (HPLC) system (Agilent Technologies, Santa Clara, CA). Ritonavir was detected by ultraviolet (UV) absorbance detection at a wavelength of 240 nm, while efavirenz and celecoxib were detected at 247 and 249 nm, respectively. The chromatographic separation was performed with a Zobrax SBC18 analytical column (150 × 2.1 mm I.D., 5 μm, 100 Å) (Agilent Technologies, Santa Clara, CA). A mixture of sodium phosphate buffer (10 mM, pH 6.8) (40%) and acetonitrile (60%) was used as mobile phase and mobile phase flow was



δ=

propionate adipate acetate 320S adipate acetate 320S suberate acetate 320s sebacate acetate propionate 504−0.2 adipate 0.85 acetate butyrate 553−0.4 adipate 0.81 acetate propionate 504−0.2 adipate 0.33 acetate butyrate 553−0.4 adipate 0.25 acetate propionate sebacate 0.24 acetate butyrate sebacate

(5)

where Δei and Δvi are the additive atomic and group contributions to the energy of vaporization (ΔEv) and molar 742

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751

Crystal Growth & Design

Article

maintained at 0.2 mL/min. An injection volume of 20 μL was used. Differential Scanning Calorimetry. Differential scanning calorimetry (DSC) analysis was performed using a TA Instruments Q2000 instrument (TA Instruments, New Castle, DE) attached to a refrigerated cooling accessory (RCS) (TA Instruments, New Castle, DE). Both the DSC and RCS were purged with nitrogen gas. Tin was used for temperature calibration, while cell constant and enthalpy calibrations were performed using indium. Baseline calibration was performed by heating the empty cell from −50 to 300 °C at 20 °C/min. The reference and sample pans were identical and the sample was sealed in an aluminum pan with a pinhole in the lid. The thermogram of the model compound was obtained by heating the sample at rates of 5 °C/min and 20 °C/min for melting point and glass transition temperature determinations, respectively. Melting point and enthalpy of fusion were determined from a first heat scan, while the glass transition temperature was determined from a second heat scan. The temperature range used for efavirenz and ritonavir was 25−150 °C, while the temperature range used for celecoxib was 25−200 °C. Thermal transitions were viewed and analyzed using the analysis software Universal Analysis 2000 for Windows 2000/ XP provided with the instrument. Moisture Sorption Analysis. The moisture sorption profiles of amorphous celecoxib, efavirenz, and ritonavir were determined at 37 °C using a VTI-100 continuous vapor flow sorption instrument (TA Instruments, New Castle, DE). The amorphous samples were prepared using the melt-quench method. A mortar and pestle was used to reduce the size of the amorphous samples prior to moisture sorption analysis. Approximately 15 mg of the amorphous sample was placed into the sample pan and dried at 37 °C and low relative humidity until the weight change was less than 0.01 wt % over 5 min. Moisture sorption analysis at the equilibrium temperature (37 °C) was performed by equilibrating the sample under controlled relative humidity (RH) ranging from 5% RH to 95% RH, in 10% RH intervals. Equilibrium was assumed to be attained when the weight change was less than 0.01 wt % over 5 min at each RH step. Moisture sorption isotherms were obtained on three individually prepared samples of each compound analyzed. Determination of Amorphous “Solubility” − Theoretical and Experimental. The theoretical amorphous solubility of the model compounds was estimated using the experimentally determined crystalline solubility, the estimated free energy difference between crystalline and amorphous forms (ΔGC), and the activity of the amorphous solute saturated with water (I(a2)): ⎡ ΔG ⎤ Camorphous = Ceq exp[−I(a 2)] exp⎢ C ⎥ ⎣ RT ⎦

amorphous solid. The detailed thermodynamic analysis is presented in ref 36. Amorphous samples were prepared by a solvent evaporation method. Prior to use, the samples were analyzed by crosspolarized light microscopy and powder X-ray diffraction analysis to verify their amorphous nature. The experimental amorphous “solubility” of the model compounds was determined by dissolving the solid in 100 mM sodium phosphate buffer at pH 6.8, 37 °C. Solution concentration was measured using a SI Photonics (Tuscon, Arizona) UV spectrometer coupled to a fiber optic probe (path-length 5 mm). Wavelength scans (200−450 nm) were performed at 45 s time intervals. Second derivatives (SIMCA P+ V.12 software, Umetrics Inc., Umea Sweden) of the spectra were taken for the calibration and sample data in order to mitigate particle scattering effects. Calibration solutions were prepared in methanol. Polarized Light Microscopy. Polarized light microscopy (PLM) was used to observe the crystallization behavior of amorphous celecoxib and efavirenz. Amorphous samples were prepared using the melt-quench method. Amorphous particles were placed on a microscopy slide and a small aliquot of sodium phosphate buffer (pH 6.8) was added to the amorphous particles. Images were taken at various time intervals. A Nikon Eclipse E600 Pol microscope with a 20× objective was used with NIS-Elements software package (Version 2.3; Nikon Company, Tokyo, Japan). Induction Time Measurements. Crystal formation was characterized by measuring the induction time from unseededdesupersaturation experiments. The onset of crystal formation at an equilibrium temperature (37 °C) was determined from the increase in intensity of light scattered (extinction) from the drug solutions. Light scattering was detected by monitoring the extinction, at 280 nm for ritonavir and 350 nm for celecoxib and efavirenz, using a SI Photonics UV/vis spectrometer (Tuscon, Arizona), fiber optically coupled with a dip probe (path-length 5 mm); the compounds have no absorbance at these wavelengths. Wavelength scans (200−450 nm) were performed at 30 s time intervals. Solution concentration was also monitored during the desupersaturation experiments by monitoring absorption peaks. Supersaturated solutions were generated by adding a small volume of predissolved drug in methanol to buffer. Solubilized drug (4 mg/mL) in methanol was titrated into buffer solution (50−80 mL) equilibrated at 37 °C, at a pH condition where the compound of interest was unionized, using a syringe pump (Harvard Apparatus, Holliston, MA). Agitation of the solution at 300 rpm was achieved by use of a magnetic stirrer (Corning PC-420D, Fisher Scientific, Pittsburgh, PA). Induction time experiments were performed in the absence and presence of predissolved polymers. A polymer concentration of 5 μg/mL was used and all experiments were performed in triplicate. The induction time for celecoxib, efavirenz, and ritonavir was determined at an initial solution concentration of 22, 19, and 20 μg/mL, respectively. These concentration values correspond to the theoretical amorphous “solubility” of each drug. Data collection commenced immediately after the addition of the drug solution to the test medium. The effect of the ionization state of the polymer on the nucleation behavior of the solute was evaluated by performing experiments at pH 3.8 and 6.8, using 100 mM sodium acetate buffer and 100 mM sodium phosphate, respectively.

(6)

Using the heat of fusion (ΔHf) and melting point (Tm) determined from DSC analysis as described above, the free energy change for transformation of an amorphous solid to a crystalline solid was estimated at 37 °C (T):35 ΔGC =

ΔHf (Tm − T )T Tm2

(7)

The activity of the hydrated amorphous solute in contact with water (I(a2)) was estimated by applying the Gibbs−Duhem equation to the water sorption isotherm data for the 743

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751

Crystal Growth & Design

Article

An important factor contributing to the nucleation mechanism and kinetics is the volume of solution in which nucleation occurs.13 Previous studies have shown that in relatively small volumes (∼1 mL) and at low supersaturations, the probability of nuclei formation in solution is low, leading to a large variation in induction time.37 However, for the relatively large volume (50−80 mL) and high supersaturation ratios (2.4, 14.7, and 15.4 for efavirenz, celecoxib, and ritonavir, respectively) utilized herein, it is reasonable to expect that the probability of nucleation formation will be high (due to random impurities in solution which may induce heterogeneous nucleation which has a lower energetic barrier compared to homogeneous nucleation), resulting in a relatively small variation in induction time measurements. This expectation appears to have been met as reflected by the relatively small variations ( 120 min), while 14 and 15 polymers had an impact on induction times for efavirenz (Figure 8) and celecoxib (Figure 9), respectively, although to varying extents. CAB Adp 0.81, CAP Adp 0.85, CA 320S Seb and CAP Adp 0.33, four of the recently synthesized cellulose derivatives, were the most effective nucleation inhibitors for both efavirenz and ritonavir, increasing the induction time significantly relative to the commercially available and more frequently used polymers, HPMCAS, HPMC and CMCAB. For ritonavir, the induction time was increased by a factor of 2 from 120 min in the absence of a polymer, to approximately 250 min in the presence of CAB Adp 0.81, the most effective inhibitor. For efavirenz, the induction time increased by a maximum factor of approximately 5, from 5 to 25 min with CAB Adp 0.81 again being the most effective inhibitor. The very hydrophilic polymers had no impact on induction times for either compound. Interestingly, CA 320S Sub, CA 320S Seb, CAB Adp 0.81 and CAP Adp 0.85 have been previously identified as effective crystal growth inhibitors for ritonavir, although only at lower supersaturations than those employed in this study.21,22,33 However, the polymer that was the most effective growth rate inhibitor for ritonavir (CA 320S Sub)33 did not correspond to the polymer(s) that extended the induction time by the greatest extent (CAB Adp 0.81 and CAP

a

The ineffective polymers, moderately and very effective polymers are color coded pink, yellow and green, respectively. The arrows indicate the approximate SP value of each model compound. RTV: (Green: > 145 min, yellow: 120 - 144 min); EFZ: (Green: > 17 min, yellow: 5 16 min); CLB: (Green > 7 min, yellow: 2 - 6 min).

Figure 8. Induction times for efavirenz from unseeded-desupersaturation experiments, in the absence and presence of polymers (n = 3). An initial concentration of 19 μg/mL efavirenz and a polymer concentration of was 5 μg/mL were used.

Adp 0.33). For celecoxib a more chemically diverse and more hydrophilic group of polymers  HPMC, HPMCAS, CA 320S Adp and Pn-IPAAmd  was the most effective at extending the induction time. For celecoxib, it is very clear from Figure 9, 747

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751

Crystal Growth & Design

Article

Figure 10. Induction time of efavirenz in the absence and presence of CAP Adp 0.85, CAB Adp 0.81 and HPMC at two pH conditions (3.8 and 6.8), initial drug concentration was 19 μg/mL and polymer concentration was 5 μg/mL.

Figure 9. Induction times for celecoxib from unseeded-desupersaturation experiments, in the absence and presence of polymers (n = 3). An initial concentration corresponding to 22 μg/mL celecoxib and a polymer concentration of 5 μg/mL were used.

polymer (Figure 10), is comparable at pH 3.8 and 6.8, as expected. So far we have evaluated the induction times at the supersaturation that we predict would be created upon dissolution of the amorphous form. However, it is clear that for celecoxib, which has a short induction time in the absence of a polymer, the gains in induction time in the presence of the most effective polymer are rather modest (i.e., an increase from 2 to 13 min) with respect to the biological time frame over which absorption occurs (3−4 h). It was thus of interest to evaluate whether greater overall gains could be achieved by generating a lower extent of supersaturation that could perhaps be maintained for a longer period of time in the presence of a polymer. By performing unseeded desupersaturation-induction time experiments at initial supersaturations corresponding to half of the theoretical amorphous solubility (∼11 μg/mL) for celecoxib, the effectiveness of HPMC was found to increase significantly. At these lower concentrations, the induction time of celecoxib was 27.5 ± 0.7 min in the absence of polymer. HPMC extended the induction time to 855.5 ± 6.0 min, increasing the induction time by a factor of 31. To further illustrate the potential benefits of using a polymeric inhibitor at a lower supersaturation, the supersaturation factor (SF) of celecoxib in the presence of HPMC at initial concentrations of 11 μg/mL (S of 7.3) and 22 μg/mL (S of 14.7) was estimated. Figure 11 shows the supersaturation ratio vs time profiles of celecoxib. The AUCsp,240min values for celecoxib, which represents the overall concentration enhancement over a time period of biological relevance, were 1745 and 1426, respectively, in the presence of HPMC at an initial S of 7.3 and 14.7 while the AUCs,240min of the saturated solution was 239.0, yielding SF values of 5.9 and 7.3 at initial S values of 14.7 and 7.3, respectively. In summary, higher levels of inhibition were achieved at lower supersaturations leading to an overall enhancement in the extent of supersaturation as a function of time.

that the more hydrophobic polymers are ineffective, while the most hydrophilic polymers have some effect and moderately hydrophobic polymers are the most effective at increasing the induction times. Celecoxib had the shortest induction time of the three model compounds, and the most effective polymers were only able to increase the induction time from 2 min to approximately 13 min. Summarizing, in general, the hydrophilic polymers were less effective/ineffective in inhibiting crystal formation of the more hydrophobic compounds, efavirenz and ritonavir (SP values of 20.03 MPa1/2and 21.89 MPa1/2, respectively), while the more hydrophobic cellulose-based polymers (both commercially available and in-house synthesized cellulose-based polymers) were effective nucleation inhibitors. In contrast, for the more hydrophilic celecoxib (SP value of 23.38 MPa1/2), the hydrophilic polymers and some of the moderately hydrophobic cellulose-based polymers were effective in increasing induction times, while the more hydrophobic polymers were ineffective. A side-by-side comparison of the impact of the investigated polymers on induction times of the model compounds is presented in Table 5, where the very effective, moderately effective and ineffective polymers are color coded green, yellow and pink, respectively. Effect of Polymer Ionization and Initial Drug Supersaturation on Induction Times. The impact of polymer ionization state on efavirenz induction time was investigated for CAB Adp 0.81 and CAP Adp 0.85, the two most effective polymers, by running experiments at two pH conditions, 3.8 (polymer COOH group is predominantly un-ionized) and 6.8 (polymer is ionized). A column chart comparing the effectiveness of CAB Adp 0.81, CAP Adp 0.85 and HPMC as nucleation inhibitors for efavirenz at an initial concentration of 19 μg/mL is shown in Figure 10; HPMC was added as a control since this polymer has no ionizable functional groups. At pH 3.8, the effectiveness (ratio of the induction time in the presence of polymer to the induction time in the absence of polymer) of CAP Adp 0.81 and CAP Adp 0.85 decreased by ∼50%, while the effectiveness of HPMC, a nonionizable



DISCUSSION Preventing crystallization is an essential component of any drug delivery strategy that utilizes supersaturation to enhance mass transport across a biological membrane. It is clear from the results of this study that polymers should not be selected 748

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751

Crystal Growth & Design

Article

(Table 5). The average SP value of the polymers within the aforementioned range is 20.87 MPa1/2, which is comparable to the SP value of ritonavir (20.03 MPa1/2). In contrast, for the more hydrophilic celecoxib with a SP value of 23.38 MPa1/2, polymers with SP values ranging from 21.27 to 32.24 MPa1/2 were effective, where the average SP value for this group of polymers is 24.28 MPa1/2. In general, for the three model compounds investigated in this study, the average SP value for the group of effective polymers is close to the SP value of the drug molecule (Table 6). Table 6. Comparison of the SP Values of the Model Compounds and Average SP Value of the Effective Polymersa

Figure 11. Degree of supersaturation vs time profile for celecoxib in the presence of HPMC at initial concentrations of 22 μg/mL and 11 μg/mL. The saturation curve is also shown. The shaded portion under each curve was used to estimate the area under the curve. Areas under the graphs at initial concentrations of 11 and 22 μg/mL are 1745 and 1426, respectively, and area under the saturation curve is 239.

compound

SP value of compound

effective polymer sp Range

average SP value of effective polymers

ritonavir efavirenz celecoxib

20.03 21.89 23.38

19.62−22.44 19.62−23.28 21.27−32.24

20.87 21.73 24.28

a

SP unit in MPa1/2.

For efavirenz and ritonavir, the synthetic polymers were ineffective in inhibiting nucleation, even when they had solubility parameters similar to the drug molecule as for PnIPAAmd and PAlAmn (Figures 7 and 8). However, in addition to drug-polymer interactions, the structure of the polymer monomer units may also influence the ability of the polymer to disrupt the packing of solute molecules. The ineffectiveness of the synthetic polymers (Figure 3) perhaps can be attributed to their flexible structures compared to the relatively rigid structure of the cellulose-based polymers. Higher levels of disruption can be caused by steric effects due to the bulky substituent groups of the cellulose esters and lower degrees of freedom of the polymer molecules compared to that of the smaller-flexible synthetic polymers. 17,46,47 Interestingly, although Pn-IPAAmd had no effect on the induction time of ritonavir, it has been found to be a highly effective crystal growth rate inhibitor of this compound.22 Clearly the properties of a polymer that are important for growth rate modification are not necessarily the same as for nucleation inhibition. The amphiphilic nature of the synthesized cellulose-based polymers appears to further enhance their ability to inhibit crystal formation. At pH 6.8, the synthesized cellulose-based polymers (which contain ionizable carboxylic acids) were more effective inhibitors relative to neutral cellulose derivatives. The pKa of the adipate substituent is approximately 4.43,48 so the degree of ionization of the CO2H-containing substituents (adipate, suberate and sebacate) will be nearly complete (∼100%) at pH 6.8.49 The presence of an ionized carboxylate group will lead to unfavorable charge−charge repulsion between monomer units of the polymer in solution, thus preventing self-association of additive molecules, which will maximize polymer interaction with solute molecules; this apparently increases their impact on solute nucleation. Thus at pH 3.8, where the CO2H substituent is less than 25% ionized, the effectiveness of the carboxylic acid bearing polymers is severely reduced, (Figure 10) which may be attributed to increased self-association of additive molecules due to reduced ionization. Using molecular-dynamic simulations, Anwar et al.17 highlighted the importance of the self-affinity of the additive

arbitrarily for this purpose, and that the effectiveness of a given polymer seems to be linked to the properties of the crystallizing solute. On the basis of the data presented in Figures 7−9 and Table 5, it is apparent that the hydrophobicity of the polymer relative to that of the drug molecule appears to be a key parameter in determining the impact of a given polymer on the nucleation rate. From a theoretical perspective and considering only nonspecific interactions, the hydrophobicity of the polymer relative to the drug affects the affinity of polymer segments for the drug molecules. This will impact the ability of the polymer to mix with drug prenucleation aggregates which in turn would be expected to influence the effectiveness of the polymer as a nucleation inhibitor. Here we assume that polymers most likely affect nucleation by hindering the reorganization of a cluster of solute molecules into an ordered structure, which is proposed to be the rate-limiting step of the two-step nucleation model. An effective nucleation inhibitor should thus interact favorably with the solute.17 A strong affinity of the polymer for the solute, relative to that for the solvent and other polymer molecules in solution, not only ensures that the polymer molecules interact with the solute aggregates, but also leads to disruption of the reorganization of the solute clusters.17 If a polymer is very hydrophilic relative to the solute molecules, it would be expected to have a higher affinity for water molecules, while if the polymer is too hydrophobic, it is likely to interact more favorably with other polymer molecules. On the basis of this reasoning, it is apparent that the ability of the polymer to inhibit crystal nucleation will be dependent on both the properties of the drug and polymer; a polymer should have hydrophobicity similar to that of the solute to maximize nonspecific interactions, or alternatively be able to form highly favorable specific interactions with the drug to be an effective nucleation inhibitor. These expectations are in general supported by the results of this study. Cellulose polymers that had a SP values close to that of the drug molecule were effective nucleation inhibitors; that is polymers with relatively similar hydrophobicity to the drug molecule were effective. For ritonavir (SP 20.03 MPa1/2), cellulose polymers with a SP value ranging from 19.62 to 22.44 MPa1/2 were effective crystallization inhibitors, albeit to varying extents 749

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751

Crystal Growth & Design

Article

Notes

molecule relative to its affinity for the solvent molecules and the crystallizing solute molecules. For the rapid crystallizers, efavirenz and celecoxib, at the high supersaturations generated in this study, the impact of the polymers on nucleation inhibition was limited from a practical perspective, with the systems crystallizing within minutes even in the presence of the best inhibitor. On the basis of theoretical models, nucleation rate will decrease with a reduction in supersaturation; therefore, higher levels of inhibition by the polymer are likely at lower supersaturations, where the thermodynamic driving force is lower. This was indeed demonstrated to be the case for celecoxib, where at the lower supersaturation tested, it is clear that considerably extended induction times can be achieved. Thus, a more moderate supersaturation level can be sustained for much longer than a higher supersaturation level, leading to a higher area under the curve for the supersaturation vs time curve and a higher supersaturation factor (Figure 11). This example thus demonstrates that although it is tempting to try and achieve a very high level of supersaturation in order to enhance biological absorption, this strategy is a double-edged sword, since it leads to higher nucleation tendency. The rate of drug crystallization will ultimately determine the success of a given supersaturating dosage form and as demonstrated for celecoxib, it may be appropriate to try and achieve a lower level of supersaturation that can be sustained for a longer time period. Supersaturation levels may be potentially controlled in vivo using controlled release dosage forms, or less soluble high energy forms. This is an important finding for pharmaceutical formulation development, where a supersaturating dosage form may fail due to rapid crystallization as a result of extreme supersaturation.

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors thank Eastman Chemical Company and Pfizer Inc. for their kind donations of CMCAB and celecoxib, respectively. Support of the National Science Foundation through grant DMR-0804609 is gratefully acknowledged.



(1) Kwong, A. D.; Kauffman, R. S.; Hurter, P.; Mueller, P. Discovery and development of telaprevir: an NS3−4A protease inhibitor for treating genotype 1 chronic hepatitis C virus. Nat. Biotechnol. 2011, 29, 933−1003. (2) Bollag, G.; Hirth, P.; Tsai, J.; Zhang, J.; Ibrahim, P. N.; Cho, H.; Spevak, W.; Zhang, C.; Zhang, Y.; Habets, G.; Burton, E. A.; Wong, B.; Tsang, G.; West, B. L.; Powell, B.; Shellooe, R.; Marimuthu, A.; Nguyen, H.; Zhang, K. Y. J.; Artis, D. R.; Schlessinger, J.; Su, F.; Higgins, B.; Iyer, R.; D’Andrea, K.; Koehler, A.; Stumm, M.; Lin, P. S.; Lee, R. J.; Grippo, J.; Puzanov, I.; Kim, K. B.; Ribas, A.; McArthur, G. A.; Sosman, J. A.; Chapman, P. B.; Flaherty, K. T.; Xu, X.; Nathanson, K. L.; Nolop, K. Clinical efficacy of a RAF inhibitor needs broad target blockade in BRAF-mutant melanoma. Nature 2010, 467, 596−599. (3) Gao, P.; Guyton, M. E.; Huang, T.; Bauer, J. M.; Stefanski, K. J.; Lu, Q. Enhanced oral bioavailability of a poorly water soluble drug PNY-91325 by supersaturatable formulations. Drug Dev. Ind. Pharm. 2004, 30, 221−229. (4) Simonelli, A. P.; Mehta, S. C.; Higuchi, W. I. Inhibition of sulfathiazole crystal growth by polyvinylpyrrolidone. J. Pharm. Sci. 1970, 57, 633−638. (5) Vandecruys, R.; Peeters, J.; Verreck, G.; Brewster, M. E. Use of a screening method to determine excipients which optimize the extent and stability of supersaturated drug solutions and application of this system to solid formulation design. Int. J. Pharm. 2007, 342, 168−175. (6) Femi-Oyewo, M. N.; Spring, M. S. Studies on paracetamol crystals produced by growth in aqueous solutions. Int. J. Pharm. 1994, 112, 17−28. (7) Simonelli, A. P.; Mehta, S. C.; Higuchi, W. I. Inhibition of sulfathiazole crystal growth by polyvinylpyrrolidone. J. Pharm. Sci. 1970, 59, 633−638. (8) Lindfors, L.; Forssen, S.; Westergren, J.; Olsson, U. Nucleation and crystal growth in supersaturated solutions of model drug. J. Colloid Interface Sci. 2008, 325, 404−413. (9) Alonzo, D. E.; Zhang, G. G. Z.; Zhou, D.; Gao, Y.; Taylor, L. S. Understanding the behavior of amorphous pharmaceutical systems during dissolution. Pharm. Res. 2009, 27, 608−618. (10) Dai, W. G.; Dong, L. C.; Shi, X. F.; Nguyen, J.; Evans, J.; Xu, Y. D.; Creasey, A. A. Evaluation of drug precipitation of solubility enhancing liquid formulations using milligram quantities of a new molecular entity (NME). J. Pharm. Sci. 2007, 96, 2957−2969. (11) Gao, P.; Akrami, A.; Alvarez, F.; Hu, J.; Li, L.; Ma, C.; Surapaneni, S. Characterization and optimization of AMG 517 supersaturatable self-emulsifying drug delivery system (S-SEDDS) for improved oral absorption. J. Pharm. Sci. 2009, 98, 516−528. (12) Raghavan, S. L.; Trividic, A.; Davis, A. F.; Hadgraft, J. Crystallization of hydrocortisone acetate: influence of polymers. Int. J. Pharm. 2001, 212, 213−221. (13) Rodriguez-Hornedo, N.; Murphy, D. Significance of controlling crystallization mechanisms and kinetics in pharmaceutical systems. J. Pharm. Sci. 1999, 88, 651−660. (14) Mullin, J. W. Crystallization, 4th ed.; Oxford: Elsiever Butterworth-Heinemann, 2001. (15) Hasegawa, A.; Taguchi, M.; Suzuki, R.; Miyata, T.; Nakagawa, H.; Sugimoto, I. Supersaturation mechanism of drugs from solid dispersion with enteric coating agents. Chem. Pharm. Bull. 1988, 36, 4941−4950.



CONCLUSION In this study, the impact of 21 polymers, added at low levels, on the solution induction times of three drug compounds (celecoxib, efavirenz and ritonavir) was quantified, enabling key polymer properties important for nucleation inhibition of the drug molecules to be elucidated. The polymers that were effective inhibitors were found to have (1) an optimal level of hydrophobicity  the effective polymers had a similar hydrophobicity to the drug molecule, and (2) structure  the cellulose derivatives with bulky side groups were more effective inhibitors compared to the synthetic polymers, even when the level of hydrophobicity of the synthetic polymers was similar to that of the cellulose-based polymers. These factors are likely to affect nucleation by promoting polymer-solute interactions, thereby disrupting the reorganization of a cluster of solute molecules into an ordered crystal structure. These insights into the key factors that affect nucleation inhibition by cellulose derivatives are important for the development of new excipients with superior crystallization inhibitory properties. Furthermore, they bolster our understanding of the mechanisms by which polymers inhibit drug crystallization.



REFERENCES

AUTHOR INFORMATION

Corresponding Author

*Address: Department of Industrial and Physical Pharmacy, College of Pharmacy, Purdue University, 575 Stadium Mall Drive, West Lafayette, Indiana 47907, USA. Tel: +1-765-4966614; fax: +1-765-494-6545; e-mail: [email protected]. 750

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751

Crystal Growth & Design

Article

for improved oral bioavailable of celecoxib. Biol. Pharm. Bull. 2004, 27, 1993−1999. (40) Strickley, R. G. Currently marketed oral lipid-based dosage forms: Drug products and excipients. In Oral Lipid-Based Formulations: Enhancing the Bioavailability of Poorly Water Soluble Drugs; Hauss, D., Ed.; CRC Press: Boca Raton, FL, 2007. (41) Casas-Breva, I.; Peris-Vicente, J.; Rambla-Alegre, M.; CardaBroch, S.; Esteve-Romero, J. Monitoring of HAART regime antiretrovirals in serum of acquired immunodeficiency syndrome patients by micellar liquid chromatography. Analyst 2012, 137, 4327− 4334. (42) Hancock, B. C.; Parks, M. What is the true solubility advantage for amorphous pharmaceuticals? Pharm. Res. 2000, 17, 397−404. (43) Alonzo, D. E.; Gao, Y.; Zhou, D.; Mo, H.; Zhang, G. G. Z.; Taylor, L. S. Dissolution and precipitation behavior of amorphous solid dispersions. J. Pharm. Sci. 2011, 100, 3316−3331. (44) Baird, J. A.; Van Eerdenbrugh, B.; Taylor, L. S. A classification system to assess the crystallization tendency of organic molecules from undercooled melts. J. Pharm. Sci. 2010, 99, 3787−3806. (45) Van Eerdenburgh, B.; Baird, J. A.; Taylor, L. S. Crystallization tendency of active pharmaceutical ingredients following rapid solventevaporation − classification and comparison with crystallization tendency from undercooled melts. J. Pharm. Sci. 2010, 99, 3826−3838. (46) Shen, T.; Langan, P.; French, A. D.; Johnson, G. J.; Gnanakaran, S. Conformational flexibility of soluble cellulose oligomers: Chain length and temperature dependence. J. Am. Chem. Soc. 2009, 131, 14786−14794. (47) Kramarenko, E. Y.; Winkler, R. G.; Khalatur, P. G.; Khokhlov, A. R.; Reineker, P. Molecular dynamics simulation study of adsorption of polymer chains with variable degree of rigidity. I. Static properties. J. Chem. Phys. 1995, 104, 4806−4813. (48) Christensen, J. J.; Hansen, L. D.; Izatt, R. M. Handbook of Proton Ionization Heats and Related Thermodynamic Quantities; John Wiley and Sons: New York, 1976. (49) Hoogendam, C. W.; De Keizer, A.; Cohen Stuart, M. A.; Bijsterbosch, B. H.; Smit, J. A. H.; Van Dijk, J. A. P. P.; van der Horst, P. M.; Batelaan, J. G. Persistence length of carboxymethyl cellulose as evaluated from size exclusion chromatography and potentiometric titrations. Macromolecules 1998, 31, 6297.

(16) Somasundaran, P.; Krishnakumar, S. Adsorption of surfactants and polymers at the solid-liquid interface. Colloids Surf., A 1997, 123 − 124, 491−513. (17) Anwar, J.; Boateng, P. K.; Tamaki, R.; Odedra, S. Mode of action and design rule for additives that modulate crystal nucleation. Angew. Chem., Int. Ed. 2009, 48, 1596−1600. (18) Myerson, A. S. Handbook of Industrial Crystallization; Butterworth-Heinemann Ltd: Oxford, 1993. (19) Nyvlt, J.; Sohnel, O.; Matuchova, M.; Broul, M. The Kinetics of Industrial Crystallization; Elsevier: New York, 1985. (20) Alonzo, D. E.; Raina, S.; Zhou, D.; Gao, Y.; Zhang, G. G. Z.; Taylor, L. S. Characterizing the impact of hydroxypropylmethyl cellulose on the growth and nucleation kinetics of felodipine from supersaturated solutions. Cryst. Growth Des. 2012, 12, 1538−1547. (21) Ilevbare, G. A.; Liu, H.; Edgar, K. J.; Taylor, L. S. Inhibition of solution crystal growth of ritonavir by cellulose polymers − factors influencing polymer effectiveness. CrystEngComm 2012, 14, 6503− 6510. (22) Ilevbare, G. A.; Liu, H.; Edgar, K. J.; Taylor, L. S. Understanding polymer properties important for crystal growth inhibition − Impact of chemically diverse polymers on solution crystal growth of ritonavir. Cryst. Growth Des. 2012, 12, 3133−3143. (23) Cerdeira, M.; Candal, R. J.; Herrera, M. L. Analytical techniques for nucleation studies in lipids: advantages and disadvantages. J. Food Sci. 2004, 69, 185−191. (24) Verdoes, D.; Kashchiev, D.; van Rosmalen, G. M. Determination of nucleation and growth rates from induction times in seeded and unseeded precipitation of calcium carbonate. J. Cryst. Growth 1992, 118, 401−413. (25) Van der Leeden, M. C.; Kashchiev, D.; van Rosmalen, G. M. Effect of additives on nucleation rate, crystal growth rate and induction time in precipitation. J. Cryst. Growth 1993, 130, 221−232. (26) Kuldipkumar, A.; Kwon, G. S.; Zhang, G. G. Z. Determining the growth mechanism of tolazamide by induction time measurement. Cryst. Growth Des. 2007, 7, 234−242. (27) Sohnel, O.; Mullin, J. W. Interpretation of crystallization induction periods. J. Colloid Interface Sci. 1988, 123, 43−50. (28) Oxtoby, D. W.; Kaschiev, D. A General Relation between the Nucleation Work and the Size of the Nucleus in Multicomponent Nucleation. J. Chem. Phys. 1994, 100, 7665−7771. (29) Schuth, F. Nucleation and Crystallization of Solids from Solution. Curr. Opin. Solid State Mater. Sci. 2001, 5, 389−395. (30) Erdemir, D.; Lee, A. Y.; Myerson, A. S. Nucleation of crystals from solution: classical and two-step models. Acc. Chem. Res. 2009, 42, 621−629. (31) Vekilov, P. G. Nucleation. Cryst. Growth Des. 2010, 10, 5007− 5019. (32) Kar, N.; Liu, H.; Edgar, K. J. Synthesis of Cellulose Adipate Derivatives. Biomacromolecules 2011, 12, 1106−1115. (33) Liu, H.; Ilevbare, G. A.; Cherniawski, B. P.; Ritchie, E. T.; Taylor, L. S.; Edgar, K. J. Synthesis and structure-property evaluation of cellulose ω-carboxyesters for amorphous solid dispersions. Carbohydr. Polym. 2012, DOI: 10.1016/j.carbpol.2012.11.049. (34) Fedors, R. F. A method for estimating both the solubility parameter and molar volumes of liquids. Polym. Eng. Sci. 1974, 14, 147−154. (35) Hoffman, J. D. Thermodynamic driving force in nucleation and growth processes. J. Chem. Phys. 1958, 29, 1192. (36) Murdande, S. B.; Pikal, M. J.; Shanker, R. M.; Bogner, R. H. Solubility advantage of amorphous pharmaceuticals I. A thermodynamic analysis. J. Pharm. Sci. 2009, 99, 1254. (37) Jiang, S.; Ter Horst, J. H. Crystal Nucleation Rates from Probability Distributions of Induction Times. Cryst. Growth Des. 2010, 11, 256−261. (38) Bevernage, J.; Forier, T.; Brouwers, J.; Tack, J.; Annaert, P.; Augustijns, P. Excipient-mediated supersaturation stabilization in human intestinal fluids. Mol. Pharmaceutics 2011, 8, 564−570. (39) Subramanian, N.; Ray, S.; Ghosal, S. K.; Bhadra, R.; Moulik, S. P. Formulation design of self-microemulsifying drug delivery systems 751

dx.doi.org/10.1021/cg301447d | Cryst. Growth Des. 2013, 13, 740−751