Effects of Polymer and Surfactant on the Dissolution and

Article pubs.acs.org/crystal

Effects of Polymer and Surfactant on the Dissolution and Transformation Profiles of Cocrystals in Aqueous Media Amjad Alhalaweh,†,§ Hassan Refat H. Ali,†,‡ and Sitaram P. Velaga*,† †

Pharmaceutical Research Group, Department of Health Sciences, Luleå University of Technology, Luleå, SE-971 87, Sweden Department of Pharmaceutical Analytical Chemistry, Faculty of Pharmacy, Assiut University, Assiut, 71526, Egypt

‡

S Supporting Information *

ABSTRACT: Capturing solubility advantages of cocrystals is of great interest, and thus to understand the mechanism by which different excipients could maintain the supersaturation generated by cocrystals at the course of absorption in aqueous media is essential. To achieve this aim, the impact of different excipients on dissolution behavior of indomethacin−saccharin (IND−SAC) were monitored by measuring the concentrations of cocrystal components in the absence and presence of various concentration of excipients by HPLC, and solid phases were analyzed by differential scanning calorimetry after each experiment and the potential of Raman spectroscopy for monitoring phase transformations in situ was tested. No dissolution advantage was offered by cocrystals in the absence of any solution additive. The polymer and surfactant used in the study increased the solubility of IND but not SAC. This differential solubilization effect is believed to have stabilized the cocrystals for a relevant period for the absorption to take place. This could be attributed to either decreased gap between supersaturation and saturation of the drug or drug interaction with the additives. Understanding the effects of excipients type and concentration on the transformation profile is vital for designing enabling formulations for cocrystals. The eutectic constant may be useful in selecting excipients for stabilizing cocrystals.

■

INTRODUCTION The bioavailability of an orally administered drug is related to its solubility and permeation characteristics.1 The biopharmaceutical classification system (BCS) divides drug substances into four categories, I−IV, on the basis of these characteristics. Because BCS class II drugs are poorly soluble but permeate the gastrointestinal tract relatively easily, the rate and extent of absorption are controlled by the rate of dissolution. As a result, a great deal of current research is focused on improving the dissolution of these drugs or their release from the formulations under development.2,3 The dissolution rate of a drug, as dictated by the Noyes−Whitney equation, is controlled by the surface area of the particles, the bulk and diffusion layer concentrations of the drug, and the thickness of the diffusion layer.4 Thus, in principle, dissolution problems for a drug can be overcome by altering at least one of these parameters. The most common strategy for improving the dissolution of a drug is probably modification of the solid form or the use of excipients in the formulation.5,6 The salt form of the drug is widely used to improve its solubility; however, the salt form cannot be obtained for nonionizable drugs and is often associated with solid-state stability problems.7 Alternatively, the amorphous form of the solid drug can be used. The apparent solubility of the amorphous form is certainly higher than that of its crystalline counterpart, but the poor solid-state stability and difficulties associated with scale-up methods still provide obstacles.8 With this in mind, recent research in solid screening programs has focused on the cocrystal form. © 2013 American Chemical Society

Cocrystals are similar to other crystalline solid forms in that the equilibrium solubility, rather than the apparent solubility, of the drug is improved, but with the added advantage that cocrystals can be made from nonionizable drugs.9−11 Generally, the improved solubility of the amorphous or cocrystal forms cannot be sustained at physiological pHs. Once the drug is released and supersaturation is approached, the drug crystallizes into its less soluble, most stable solid form. Since these processes can occur concurrently and in fractions of seconds, the improved solubility will not be potentially useful.12 Moreover, it has been suggested that amorphous material, in addition to solution-mediated transformation, undergoes solidstate transformation once it has surface contact with the solution.13 Manipulation of the formulation by the addition of polymers or surfactants may overcome the precipitation of the drug and thus sustain the state of drug supersaturation until absorption occurs.6 The use of polymers to inhibit crystallization of the drug in solution has been widely studied.14 Polymers appear to inhibit crystallization of the drug and stabilize its supersaturation by interacting with the solid surface.13 They are also useful for preventing transformation of the drug from the anhydrous form to the hydrate by delaying nucleation or inhibiting crystal growth.15 In contrast, only a very limited Received: October 15, 2013 Revised: November 27, 2013 Published: December 9, 2013 643

dx.doi.org/10.1021/cg4015256 | Cryst. Growth Des. 2014, 14, 643−648

Crystal Growth & Design

Article

spectroscopy. Three batches were prepared simultaneously and used throughout the study. The cocrystal particles were estimated by scanning electron microscopy (SEM, JSM 6460lv, JEOL, Japan) to be less than 100 μm in size (Supporting Information). Differential Scanning Calorimetry (DSC). Samples (1−3 mg) were scanned on a DSC Q1000 (TA Instruments) under a continuously purged dry nitrogen atmosphere (flow rate 50 mL/ min) from 30 to 300 °C (heating rate 10 °C/min). Nonhermetically sealed aluminum pans were used for all samples. The machine was equipped with a refrigerated cooling system and was precalibrated for temperature and enthalpy using indium. Powder X-ray Diffractometry (PXRD). The solid phases were analyzed in a Siemens DIFFRAC plus 5000 powder diffractometer with Cu Kα radiation (1.54056 Å), which was precalibrated using a silicon standard. Samples were scanned between 5 and 40°2θ, with a step size of 0.02° at 2 step/s. The tube voltage and amperage were set at 40 kV and 40 mA, respectively. The divergence and antiscattering slit settings were variable for illumination of the 20 mm area of the sample. The sample stage was spun at 30 rpm. High Performance Liquid Chromatography (HPLC). The solutions were analyzed by HPLC (series 200 binary LC pump and 200 UV−vis detector, TotalChrom software, Perkin-Elmer, Wellesly, MA). The IND−SAC components were separated over a C18 column (Dalco Chrometch, 5 μm, 150 mm × 4.6 mm). The HPLC analysis was conducted at room temperature with a flow rate of 1 mL/min. UV detection at 319 nm was employed for IND, with a mobile phase of 0.2% w/v phosphoric acid and methanol (25:75).27 SAC was detected at 254 nm with a mobile phase of 20% v/v acetic acid adjusted to pH 3 using a saturated solution of sodium acetate.28 Raman Spectroscopy. Raman spectra of IND−SAC cocrystals, γIND, α-IND, and slurried IND−SAC cocrystals in PB were obtained. The Raman spectra were recorded using 785 nm laser excitation (Kaiser Optical Systems, Inc., Ann Arbor, MI, USA) with a 1/4” diameter immersion optic (Kaiser Optical Systems, Inc., Ann Arbor, MI, USA) attached to an MR probe (Kaiser Optical Systems, Inc., Ann Arbor, MI, USA). The laser power at the samples was about 400 mW. The spectra were collected in the wavenumber range 100−3200 cm−1. The equipment was calibrated daily by recording the Raman spectrum of cyclohexane (1 accumulation, 1 s). The spectrometer was controlled by commercial instrument software (HoloGRAMS, version 4.1), supplied with the machine. All Raman spectra were exported to the Galactic* SPC format using GRAMS AI (Version 8.0, Thermo Electron Corp. Waltham, MA, USA). The raw Raman spectra had a high fluorescence background. A baseline correction was therefore applied to all the Raman spectra to remove this fluorescence background and allow visual interpretation using the GRAMS AI package with no further processing. Powder Dissolution Studies in the Absence and Presence of PVP and SLS. The nonsink powder dissolution study for IND−SAC cocrystals was conducted in a temperature-controlled oven at 25 °C (±0.5 °C) using the phase solubility method. The IND−SAC cocrystals (100 mg) were added to 10 mL of PB buffer (pH 3) in 25 mL glass vials and stirred using small magnetic pellets. At each predetermined time point, an aliquot was withdrawn from the vial, passed through a 0.22 μm membrane filter and appropriately diluted for HPLC analysis. The pH was recorded at each time point and the rest of the suspension was filtered under a vacuum and analyzed by DSC. The concentrations of IND and SAC were determined following the method described above. Three vials were used for each time point, and the standard deviation was calculated. The same procedures and experimental conditions were followed in the absence of excipients and the presence of 250 μg/mL or 2% w/v PVP or 25 and 100 mM SLS. Determination of the Eutectic Point. The eutectic points, that is, the concentrations at which IND−SAC cocrystals and IND were in equilibrium with the solution, were determined by suspending 100 mg of IND in 5 mL of PB (pH 3), adding 100 mg of IND−SAC cocrystals, and stirring for 96 h at 25 °C. The equilibrium concentrations of IND and SAC at the eutectic points were measured by HPLC. The existence of drug and cocrystal solid forms at

number of studies have investigated the potential of excipients in stabilizing or improving the solubility of cocrystals.16 The addition of polymers and surfactants to cocrystal formulations has been shown to increase the speed of drug release from the formulation.16,17 Early cocrystal formulations were mainly empirical. However, the solubility behavior of cocrystals with different ionization properties has recently been thoroughly investigated.18−20 Further, an understanding of the pHdependent solubility and stability of cocrystals has allowed the introduction of strategies that can transform cocrystals from a metastable form in solution to a stable form.9,21 A strategy based on the differential solubilization of cocrystal components has been shown to effectively stabilize the cocrystals.22,23 Thus, increasing our in-depth understanding of cocrystals will help rationalize their formulation. An in vivo study also showed that when indomethacin (IND) was formulated as simple IND−saccharin (SAC) cocrystals, its bioavailability was increased over that of IND alone but was similar to that of the commercial IND formulation Indomee.24 In our recent study of the pH solubility behavior of IND−SAC cocrystals,21 we estimated using relevant mathematical models that IND−SAC cocrystals were more soluble than IND at all pHs, which explains the improved in vivo performance of IND−SAC. It appears possible to obtain better bioavailability than that of Indomee by controlling the transformation of the cocrystals. It was therefore of interest to investigate formulating IND−SAC to capture the full solubility advantages offered by this solid cocrystal form. Our investigations were further motivated by the results of several studies on the stability and dissolution of the amorphous form of IND and its formulations.13,25 The overall objective of the study was to provide insight into rational formulation strategies for capturing the full solubility advantage offered by cocrystals. The specific objectives were (1) to investigate the dissolution and transformation behavior of IND−SAC cocrystals in an aqueous medium through monitoring the dissolution profiles of both the individual cocrystal components and the solid-state in equilibrium, (2) to study the effects of a polymer and a surfactant and their concentrations on the solubility and solution stability of the cocrystals, and (3) to understand the mechanisms by which a polymer and a surfactant influence the supersaturation generated by the cocrystals. The study presents the processes taking place during cocrystal dissolution and challenges involved in cocrystal formulations. It also introduces a rational approach to be followed during cocrystal formulation.

■

MATERIALS AND METHODS

Materials. IND (γ form unless otherwise mentioned), SAC, sodium lauryl sulfate (SLS), polyvinylpyrrolidone (PVP), K29/32 phosphoric acid, sodium acetate, methanol, and ethyl acetate were purchased from Sigma-Aldrich (Stockholm, Sweden). Milli-Q water was used in the study. All chemicals and solvents were used as received. A 200 mM phosphate buffer (PB), pH 3, was used as a medium throughout this study. The α form of IND was prepared as a reference as described earlier, and its phase purity was confirmed by powder X-ray diffractometry (PXRD), differential scanning calorimetry (DSC), and Raman spectroscopy.26 Preparation of IND−SAC Cocrystals by Slurry Cocrystallization. A mixture of IND (3.578 g) and SAC (1.83 g) was slurried in 10 mL of ethyl acetate for 5 days at room temperature. The suspensions were filtered together under a vacuum, dried at room temperature, and collected. The phase purity was verified by PXRD, DSC, and Raman 644

dx.doi.org/10.1021/cg4015256 | Cryst. Growth Des. 2014, 14, 643−648

Crystal Growth & Design

Article

equilibrium was estimated and confirmed by PXRD. These experiments were performed in the presence and absence of different concentrations of SLS or PVP.

■

RESULTS AND DISCUSSION Dissolution Behavior of IND−SAC Cocrystals in Buffer Solutions. The phase purity of IND−SAC cocrystal was initially confirmed by DSC and PXRD (data not shown). HPLC analysis confirmed that the IND−SAC cocrystals had 1:1 stoichiometry (results not shown). It is important to control the phase purity of the cocrystals to avoid any erroneous conclusions based on reduced cocrystal solubility in the presence of excess amounts of the cocrystal components.21 The IND−SAC cocrystals were more soluble than IND at all pH values (about 67 times more soluble at pH 3).21 Thus, IND−SAC cocrystals can potentially generate supersaturation of IND over the whole pH range relevant for oral delivery. The pH of the dissolution medium fell drastically as IND−SAC dissolved in the pH 7.4 PB, which complicated interpretation of the dissolution behavior.29 Hence, pH 3 was specifically chosen in this study so as to retain control of the pH and obtain a deeper understanding of the dissolution behavior of the cocrystals. When Raman spectroscopy was used to investigate the dissolution and transformation kinetics of IND−SAC cocrystals in PB (pH 3) in situ, the characteristic band of the α form of IND at 1650 cm−1 was not recorded until about 50 min of the experiment (Figure 1).30 However, the DSC trace of the solids

Figure 2. DSC thermograms of the solid phases collected at various times during dissolution of IND−SAC cocrystals in PB at pH 3.

The dissolution profile of IND−SAC cocrystals in PB (pH 3) was determined to investigate if the supersaturation of the drug can be maintained to a relevant period at the studied conditions. The concentrations of both components of the IND−SAC cocrystals, IND and SAC, were measured at various time points. Figure 3 presents the dissolution profiles of the components. The pH of the solution at all time points was

Figure 1. Raman spectra of IND−SAC cocrystals, γ-IND, α-IND, and slurried IND−SAC cocrystals in PB after 30 and 50 min.

showed melting peaks at 147 °C after 15 min, corresponding to the onset of the α form melting, which suggests rapid cocrystal transformation and crystallization of IND (Figure 2). Clearly, Raman spectroscopy was not able to track the early transformation of IND−SAC under the conditions of the study, possibly as a result of sensitivity issues at very low content of α form; DSC was therefore used as an off-line method for shedding light on the transformation kinetics. Estimation of the cocrystal eutectic constant (Keu) has recently been used to comment on the stability of certain cocrystals.19,31Keu can be determined from the ratio of the concentrations of the coformer and drug at the eutectic point.31 It has been reported that the 1:1 cocrystal is stable if Keu < 1.31 The Keu for IND−SAC cocrystals in PB at pH 3 was 1721 ± 81 (n = 3) in our study, indicating that the cocrystal is not stable at this condition. This high Keu value was the result of the drastic differences in solubility between IND and SAC at pH 3.

Figure 3. The concentrations of (A) IND and (B) SAC in μg/mL, resulting from the dissociation of IND−SAC at various times (in min) in PB at pH 3. 645

dx.doi.org/10.1021/cg4015256 | Cryst. Growth Des. 2014, 14, 643−648

Crystal Growth & Design

Article

to find out whether this amount of PVP was enough to maintain IND supersaturation and to slow the drug crystallization process. The concentrations of both components were determined in the presence of predissolved PVP and were compared with the concentrations in the absence of PVP (Figure 4). No differences in the IND concentrations were

about 2.9 ± 0.1. It is apparent that the IND concentration reached a plateau quickly, while SAC dissolved steadily over a period of four hours (Figure 3). The plateau IND concentration of 1−2 μg/mL agrees very well with the reported apparent solubility of the α form of IND.13 Indeed, traces of the α form of IND were found in the solids separated and filtered after the dissolution experiments at 15 min. As a result, it can be inferred that supersaturation of IND was not reached, even though the solubility of IND−SAC was more than 60-fold higher than that of IND at this pH.21 This could be explained by the huge driving force generated by the highly soluble cocrystals which forced crystallization of IND. Although many articles have predicted faster dissolution and supersaturation of cocrystals in buffered solutions,10 a number of studies have shown no improvement in the dissolution despite higher solubility, similar to our findings.32,33 Indeed, IND is known to have a high crystallization tendency (i.e., it crystallizes readily at lower supersaturation levels) in solution. Amorphous IND undergoes solution-mediated transformation, which profoundly affects the drug supersaturation behavior.13 In our study, the dissolution and crystallization of IND continued, and the system eventually approached a state where IND−SAC cocrystals and solid IND (γ form) were in equilibrium with the concentrations in solution (the eutectic point). Interestingly, it only took a small amount of excess IND to attain the eutectic point, thus losing the solubility advantage of the cocrystals. Such an end-point in the dissolution experiment appears to be unique to cocrystals (which are highly soluble and transform easily), since amorphous solids convert completely to thermodynamically stable crystalline forms at equilibrium in solution. This emphasizes the importance of purity of the cocrystals and request extremely high control on crystallization processes of cocrystals. Dissolution Behavior of IND−SAC Cocrystals in the PVP Solutions. It has been well established that polymers can slow down the solution-mediated phase transformations of various solid forms by various mechanisms, resulting in the generation or sustaining of drug supersaturation for longer periods. PVP is not only used as a stabilizer in the solid state but also is widely used for maintaining supersaturation in solution.33−37 PVP has been shown to inhibit the crystallization of amorphous IND and to sustain the supersaturation of the drug in aqueous solution at 25 °C at a concentration where solubility of IND is not affected.13 In an early study on the use of formulation strategies to access the true advantage of celecoxib−nicotinamide cocrystals, it was shown that PVP with 2% SLS directed the conversion to intermediate compounds that are associated with good absorption characteristics. However, the effect of excipients on the dissolution and conversion behavior of cocrystals has not hitherto been systematically studied. In this paper, IIND−SAC cocrystal dissolution behavior was studied in predissolved PVP to investigate whether the polymer alone could inhibit the crystallization of IND and maintain supersaturation. PVP was chosen because it forms hydrogen bonds with solid forms of IND.25 Although the inhibitory effect of PVP is known to be dependent on its concentration, 250 μg/ mL PVP was used in this study because it has been reported that this concentration is high enough to inhibit crystallization of IND from amorphous solids without affecting the degree of supersaturation.13 The Keu value was determined to be 1721 ± 81 (n = 3) under these conditions, similar to the value without PVP in the solution. The dissolution profile was then analyzed

Figure 4. The concentrations of (A) IND and (B) SAC in μg/mL, resulting from the dissociation of IND−SAC at various times (in min) in PB at pH 3. (Δ) in buffer only, (○) in predissolved 250 μg/mL PVP, and (□) in predissolved 2% w/v PVP.

observed in the presence of the polymer, indicating that it had no inhibitory effect at this concentration. However, the concentration of SAC was not equimolar with that of IND and a small endothermic transition at 147 °C in the DSC thermogram corresponding to α-IND was only detected at 240 min (Supporting Information). The effects of higher concentrations of PVP on cocrystal solubility and solution stability were then investigated. Figure 5 shows Keu as a function of the concentration of the dissolved polymer. The Keu value decreased with increasing PVP concentrations as a result of increasing concentrations of IND; there was little or no effect on SAC concentrations. This indicates that PVP can differentially solubilize the cocrystal components and thus could decrease the supersaturation level with respect to the drug. The dissolution behavior of IND− SAC was investigated at 2% w/v PVP (Figure 4). The concentration of IND was about 12 μg/mL and no crystalline IND was observed at 240 min, as confirmed by DSC (Supporting Information). The SAC concentration was not equimolar with that of IND, and the profile was similar to those 646

dx.doi.org/10.1021/cg4015256 | Cryst. Growth Des. 2014, 14, 643−648

Crystal Growth & Design

Article

had similar concentrations, indicating differential solubilization of the cocrystal components by SLS. Figure 6 shows the calculated Keu at varying SLS concentrations. It is clear that as little as 50 mM predissolved SLS reduced Keu from 1721 to about 9.5. The Keu values gradually decreased with increasing SLS concentrations and remained at about 1 at SLS concentrations of 500 mM or higher (Figure 6). Although it is desirable to stabilize cocrystals thermodynamically at physiological pHs, the formulation requirements may be very demanding. Large amounts of SLS (500−700 mM) are needed to stabilize IND−SAC cocrystals, and this is not feasible from toxicity and formulation points of view. It can be envisaged that IND−SAC cocrystals show considerable supersaturation at SLS concentrations of 25 mM and 100 mM. The dissolution of IND−SAC cocrystals at these SLS concentrations was investigated under conditions similar to those described above. The dissolution profiles of IND and SAC from the cocrystals as a function of time are depicted in Figure 7. The concentration of IND showed a parachuting effect with 25 mM SLS, dropping after the first measurement (2 min) and continuing to decrease. It was evident that IND was crystallized, as confirmed by DSC (Supporting Information). With 100 mM SLS, IND dissolved rapidly, reaching a supersaturated state in 10 min. Interestingly, no traces of αIND were detected by DSC after 240 min, confirming the

Figure 5. Keu values (○) as a function of PVP concentration.

in experiments conducted at 250 μg/mL of PVP. The results indicate that conversion of cocrystals took place, but PVP could have kinetically inhibited the crystallization of IND at high concentrations and could maintain a supersaturation level at these concentrations for a time period. The maintained supersaturation is of a great importance to avoid erratic absorption of the drug. Dissolution Behavior of IND−SAC Cocrystals in the Presence of SLS. The micellar surfactant solubilization of solids is a useful method of thermodynamically stabilizing cocrystals in aqueous media.22,23 This approach is based on the differential solubilization of the cocrystal components, where the surfactant preferentially increases the solubility of the poorly soluble component through micelle formation, resulting in the stabilization or minimization of the thermodynamic driving force behind conversion of the cocrystals.19,31 The surfactant SLS was predissolved at various concentrations in the range of 0−800 mM, and the eutectic points were determined. Figure 6 shows the concentrations of IND and SAC as a function of SLS concentration at the eutectic points. It can be seen that the concentration of IND dramatically increased relative to that of SAC with increasing SLS concentrations. From about 500 mM SLS, IND and SAC

Figure 7. The concentrations of (A) IND and (B) SAC in μg/mL, resulting from the dissociation of IND−SAC at various times (in min) in PB at pH 3. (Δ) in buffer only, (○) in predissolved 25 mM SLS, and (□) in predissolved 100 mM SLS.

Figure 6. Keu values (○) as a function of SLS concentration. The dotted line represents the theoretical presentation of Keu = 1 at various concentrations of SLS. 647

dx.doi.org/10.1021/cg4015256 | Cryst. Growth Des. 2014, 14, 643−648

Crystal Growth & Design

Article

(8) Serajuddin, A. J. Pharm. Sci. 1999, 88 (10), 1058−1066. (9) Bethune, S. J.; Huang, N.; Jayasankar, A.; Rodri guez-Hornedo, N. Cryst. Growth Des. 2009, 9 (9), 3976−3988. (10) Thakuria, R.; Delori, A.; Jones, W.; Lipert, M. P.; Roy, L.; Rodríguez-Hornedo, N. Int. J. Pharm. 2013, In press. (11) Alhalaweh, A.; George, S.; Basavoju, S.; Childs, S. L.; Rizvi, S. A.; Velaga, S. P. CrystEngComm 2012, 14 (15), 5078−5088. (12) Alhalaweh, A.; Andersson, S.; Velaga, S. P. Eur. J. Pharm. Sci. 2009, 38 (3), 206−214. (13) Alonzo, D. E.; Zhang, G. G. Z.; Zhou, D.; Gao, Y.; Taylor, L. S. Pharm. Res. 2010, 27 (4), 608−618. (14) Konno, H.; Taylor, L. S. J. Pharm. Sci. 2006, 95 (12), 2692− 2705. (15) Wikström, H.; Carroll, W. J.; Taylor, L. S. Pharm. Res. 2008, 25 (4), 923−935. (16) Remenar, J. F.; Peterson, M. L.; Stephens, P. W.; Zhang, Z.; Zimenkov, Y.; Hickey, M. B. Mol. Pharm. 2007, 4 (3), 386−400. (17) Childs, S. L.; Kandi, P.; Lingireddy, S. R. Mol. Pharmacol. 2013, 10 (8), 3112−3127. (18) Alhalaweh, A.; George, S.; Bostro m, D.; Velaga, S. P. Cryst. Growth Des. 2010, 10 (11), 4847−4855. (19) Alhalaweh, A.; Sokolowski, A.; Rodriguez-Hornedo, N.; Velaga, S. P. Cryst. Growth Des. 2011, 11 (9), 3923−3929. (20) Nehm, S.; Rodriguez-Spong, B.; Rodriguez-Hornedo, N. Cryst. Growth Des. 2006, 6 (2), 592−600. (21) Alhalaweh, A.; Rodriguez-Hornedo, N.; Roy, L.; Velaga, S. P. Mol Pharm. 2012, 9 (9), 2605−26012. (22) Huang, N.; Rodri guez-Hornedo, N. Cryst. Growth Des. 2010, 10 (5), 2050−2053. (23) Huang, N.; Rodríguez-Hornedo, N. CrystEngComm 2011, 13, 5409−5422. (24) Jung, M. S.; Kim, J. S.; Kim, M. S.; Alhalaweh, A.; Cho, W.; Hwang, S. J.; Velaga, S. P. J. Pharm. Pharmacol. 2010, 62 (11), 1560− 1568. (25) Taylor, L. S.; Zografi, G. Pharm. Res. 1997, 14 (12), 1691−1698. (26) Allesø, M.; Velaga, S.; Alhalaweh, A.; Cornett, C.; Rasmussen, M. A.; Berg, F.; Diego, H. L.; Rantanen, J. Anal Chem. 2008, 80 (20), 7755−7764. (27) Hess, S.; Teubert, U.; Ortwein, J.; Eger, K. Eur. J. Pharm. Sci. 2001, 14 (4), 301−311. (28) Food Additives Analytical ManualA Collection of Analytical Methods for Selected Food Additives, 5th ed.; Warner, C., Modderman, J., Fazio, T., Beroza, M., Schwartzmann, G., Fominaya, K., Eds.; AOAC International: Arlington, 1993; Vol. 1, p 85. (29) Basavoju, S.; Boström, D.; Velaga, S. P. Pharm. Res. 2008, 25 (3), 530−541. (30) Ali, H. R. H.; Alhalaweh, A.; Velaga, S. P. Drug Dev. Ind. Pharm. 2013, 39 (5), 625−634. (31) Good, D.; Rodri guez-Hornedo, N. Cryst. Growth Des. 2010, 10 (3), 1028−1032. (32) Shiraki, K.; Takata, N.; Takano, R.; Hayashi, Y.; Terada, K. Pharm. Res. 2008, 25 (11), 2581−2592. (33) Gupta, P.; Kakumanu, V. K.; Bansal, A. K. Pharm. Res. 2004, 21 (10), 1762−1769. (34) Gupta, P.; Thilagavathi, R.; Chakraborti, A. K.; Bansal, A. K. Mol Pharm. 2005, 2 (5), 384−391. (35) Matsumoto, T.; Zografi, G. Pharm. Res. 1999, 16 (11), 1722− 1728. (36) Liu, X.; Lu, M.; Guo, Z.; Huang, L.; Feng, X.; Wu, C. Pharm. Res. 2012, 29, 806−817. (37) Good, D.; Miranda, C.; Rodríguez-Hornedo, N. CrystEngComm 2011, 13 (4), 1181−1189.

maintenance of supersaturation over the entire period of the experiment (Supporting Information). The release of IND and SAC from the cocrystals followed a 1:1 molar ratio. The degree of supersaturation has an impact on the transformation of the cocrystals to their corresponding drugs. This can be affected either by using formulation excipients that kinetically or thermodynamically stabilize the cocrystals during the course of absorption or by using cocrystals that generate less supersaturation than those associated with fast transformation. The Keu approach is valuable for assessing cocrystal solution behavior and in the selection of formulation excipients on a rational basis.

■

CONCLUSIONS IND−SAC cocrystals are known to improve the solubility of IND. This advantage has not been fully realized in previous formulations because of the speedy conversion of the cocrystals to the crystallized drug component as a result of the high supersaturation level created by the cocrystal with respect to the drug. The types and concentrations of the excipients have a great impact on the solubility of the cocrystals and the stability of the solution. This study presents a rational formulation approach for choosing appropriate excipients, based on Keu, that can be used in the future development of cocrystal formulations. The study has shown that the supersaturation can be maintained using the proposed formulation for a relevant time for absorption to take place.

■

ASSOCIATED CONTENT

S Supporting Information *

Figure 1. SEM photographs of IND-SAC cocrystals powder used in the dissolution study. Figures 2−5. DSC curves of cocrystals. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address §

Department of Pharmacy, Uppsala University, Uppsala, Sweden.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors would like to thank the Kempe Foundation (Kempestiftelserna) for an instrumentation grant. H.R.H.A. and S.P.V. are also thankful for the project grant from the Swedish Research Council (Vetenskapsrådet).

■

REFERENCES

(1) Amidon, G. L.; Lennernäs, H.; Shah, V. P.; Crison, J. R. Pharm. Res. 1995, 12 (3), 413−420. (2) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery Rev. 1997, 23 (1−3), 3−25. (3) Panakanti, R.; Narang, A. S. Pharm. Res. 2012, 29, 2639−2659. (4) Noyes, A. A.; Whitney, W. R. J. Am. Chem. Soc. 1897, 19 (12), 930−934. (5) Blagden, N.; De Matas, M.; Gavan, P.; York, P. Adv. Drug Delivery Rev. 2007, 59 (7), 617−630. (6) Dai, W. G.; Pollock-Dove, C.; Dong, L. C.; Li, S. Adv. Drug Delivery Rev. 2008, 60 (6), 657−672. (7) Serajuddin, A. Adv. Drug Delivery Rev. 2007, 59 (7), 603−616. 648

dx.doi.org/10.1021/cg4015256 | Cryst. Growth Des. 2014, 14, 643−648