2024 Volume 72 Issue 1 Pages 102-108
For taste masking of fexofenadine hydrochloride (FXD), ethylcellulose (EC) microparticles with FXD were developed. The amounts of EC, Tween 80, and polyvinyl alcohol (PVA) in the composition had little effect on initial drug release properties. Based on the results of the drug recovery and the drug release properties, FXD(EC200) was the optimal FXD microparticle formulation. From the results of Fourier transform infrared spectroscopy spectra and X-ray diffraction patterns of FXD(EC200), FXD amorphization in the microparticles and interaction between FXD and other components were suggested, and the formation of a solid dispersion of FXD was suggested. Because the possibility of the complex of PVA and FXD on the particle surface was suggested, sodium lauryl sulfate (SLS) was added to the composition. The initial drug release from FXD microparticles with SLS was further suppressed compared with FXD(EC200). From these results, FXD microparticles with SLS can be prepared as a controlled-release formulation and are expected to be useful for masking the bitter tasting particulates.
Population aging is progressing on a global scale due to the extension of life expectancy and decline in birth rates.1) Pharmaceutical formulations that are easy to swallow for older people are desired, and the development of formulations suitable for children is also an important issue. Considering the characteristics of children and older people, it is desirable that oral preparations are easy to swallow and that the dose can be adjusted. Powders and granules are applicable dosage forms for children and older people.2,3) However, unpleasant taste such as bitterness due to the elution of drugs into the oral cavity may cause refusal to take medication. The taste of medicines is very important because human taste is highly subjective and children are more sensitive to bitter substances than adults.4) Various pharmaceutical techniques are used to mask bitterness, which can be classified into sensory, physical, and chemical masking techniques.5) Sensory bitterness masking is a method of making it difficult to perceive bitterness by sweetness, saltiness, sourness, umami, or aroma.6) Physical bitterness masking is a masking technique by controlling the release of bitter-tasting active ingredients.7–9) Chemical masking is a masking technology that suppresses bitterness using chemical factors such as chemical modification of bitter-tasting active ingredients.10)
Solid dispersion, introduced by Sekiguchi and Obi,11) defined as “the dispersion of one or more active ingredients in an inert matrix at solid-state prepared by melting (fusion), solvent or melting–solvent methods” by Chiou and Riegelman.12) Matrix-type microparticles containing uniformly dispersed poorly soluble drugs have been studied for controlled-release dosage forms to improve drug bioavailability and better control drug delivery.13,14) Additionally, the presence of insoluble polymers in the matrix alters the permeability of the matrix, thus altering the rate of drug release into the dissolution medium. Thus, matrix microparticles containing an insoluble polymer may offer benefits in reducing drug release into the oral cavity and controlled drug delivery,15) and the matrix microparticles will be useful as physical bitterness-masking technology.16)
Fexofenadine hydrochloride (FXD) is a non-sedating antihistamine with selective peripheral H1-receptor antagonist activity. FXD is known to have a bitter taste. FXD dry syrup was marketed but recently discontinued. We consider that one of the reasons for this is the difficulty in quality control of the dry syrup formulation. Therefore, we consider that a new formulation with a bitter taste masking effect that is easy to control quality is needed. In this study, matrix microparticles of FXD, which is slightly soluble in water,17) were prepared us7ing insoluble polymer ethylcellulose (EC) as an emulsification-solvent evaporation method. Tween 80 and polyvinyl alcohol (PVA) were used as emulsifiers and emulsion stabilizers, and the effect of the amount of excipient added on formulation characteristics focusing on controlled-release in the oral cavity was evaluated. Sodium lauryl sulfate (SLS), an anionic surfactant, was added to the formulation ingredients to further improve controlled release. SLS is included in the list of Generally Recognized As Safe (GRAS) agents issued by the U.S. Food and Drug Administration, and widely used in cosmetics and oral and topical pharmaceutical formulations.18) SLS has been reported to form complexes between lauryl sulfate anions and protonated drug forms.19) We attempted to use SLS because of the potential for FXD, the hydrochloride salt, to form a complex with SLS. Structural analysis of the formulation was studied using Fourier transform infrared spectroscopy (FTIR) measurement, synchrotron X-ray diffraction measurement and differential scanning calorimetry (DSC) measurement, and the release mechanism based on the interaction between excipients and FXD was discussed.
FXD used as the active pharmaceutical ingredient was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). EC (abt. 49% ethoxy, 45cP), and Tween 80 was purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). SLS was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). PVA (GOHSENOL™ KL-05) was supplied by MITSUBISHI CHEMICAL GROUP (Tokyo, Japan). The saponification degree (mol%) of PVA (KL-05) was 79.8 and a viscosity of 4.6 for 4% PVA (KL-05) solution at 20 °C was as reported by the supplier. The molecular structures of FXD and SLS are shown in Fig. 1.
EC microparticles with FXD (FXD microparticles) were prepared as the composition shown in Table 1. The preparation of FXD microparticles was as follows. FXD was dissolved in 20 mL methanol, and Tween 80, EC, and SLS were added and stirred until dissolved. 0.8% PVA solution (10 mL) was gradually added to that methanol solution over 5 min. Then, methanol and water were removed while the flask was immersed in a 50 °C water bath using an evaporator, and the residue after solvent removal was poured into a balance dish. The balance dish was placed under reduced pressure for 24 h, and the dried mixture was sieved through a sieve (425 µm aperture) to obtain FXD microparticles. Another preparation method, with a composition of FXD(EC100), FXD microparticles was prepared using methanol 16 mL/acetonitrile 4 mL mixture solution. FXD was dissolved methanol/acetonitrile solution, and subsequent preparation methods were the same as above. FXD microparticles prepared using the methanol/acetonitrile solution was named FXD(EC100)-M8A2.
| Formulation | FXD | EC | Tween 80 | SLS | PVA |
|---|---|---|---|---|---|
| mg | |||||
| FXD(EC200) | 40 | 200 | 40 | — | 80 |
| FXD(EC200)-2 | 40 | 200 | 80 | — | 160 |
| FXD(EC100) | 40 | 100 | 40 | — | 80 |
| FXD(EC200)-SLS10 | 40 | 200 | 30 | 10 | 80 |
| FXD(EC200)-SLS20 | 40 | 200 | 20 | 20 | 80 |
The drug content in the formulations was determined by the following method. Purified water (2 mL) was added to 10 mg of prepared FXD microparticles and stirred, then methanol is added to a total volume of 25 mL, and the solution was diluted 10-fold with methanol. The solution was filtered through a membrane filter (0.45 µm) and injected into a HPLC. The amount of FXD (mg) in the formulation was calculated from the drug concentration obtained. The drug recovery was the drug amount recovered as the formulation relative to the amount of FXD used as a raw material. The drug recovery was determined by dividing the amount of FXD in the formulation by the initial amount of FXD.
HPLC was carried out using an LC-6AD pump (Shimadzu, Kyoto, Japan) and a system controller (SCL-10A VP, Shimadzu) equipped with a Capcell Pak C18 MG II column (4.6 × 250 mm, OSAKA SODA CO., LTD., Osaka, Japan) and an SPD-20A UV detector (Shimadzu). Chromatography was carried out at 40 °C. The injection volume was 20 µL. The mobile phase was prepared according to the Japanese Pharmacopoeia 18th edition (JP18). Sodium dihydrogen phosphate dihydrate 7.51 g and sodium perchlorate 0.84 g were dissolved in 1000 mL of water and adjusted to pH 2.0 by adding phosphoric acid. To 650 mL of the solution were added 350 mL of acetonitrile and 3 mL of triethylamine. The flow rate was 1 mL/min and the detection wavelength was 220 nm. Calibration curves were obtained using a linear regression analysis of concentrations plotted against peak area.
Morphological Features of FXD MicroparticlesThe morphologies of FXD and FXD microparticles were characterized using an optical microscope (OLYMPUS BX51, Olympus Corporation, Tokyo, Japan) equipped with objective lenses (×20 and ×40 magnification) and a microscope digital camera (Visualix V500FL, Visualix K.K., Hyogo, Japan). From the microscopic images, parallel lines were manually drawn and Feret diameters of 50 particles of FXD and FXD microparticles were measured in random.20)
Drug Release Properties from FXD MicroparticlesDrug release properties were studied according to the paddle method of the Dissolution Test Method (JP18). Since pH of saliva is close to neutral with a pH of 6.8–7.0,21) 900 mL of pH 6.8 phosphate buffer (2nd fluid, described in the Dissolution Test Method in JP18) was used as the dissolution media. For FXD and FXD microparticles, equivalent to 6 mg FXD was weighed and placed in the test solution warmed to 37.0 ± 0.5 °C, and the test was started by rotating the paddle at 50 rpm. Over time, 5 mL of the test solution was collected, filtered through a membrane filter (0.45 µm). Five milliliters of fresh medium was added immediately after each sample was taken. The absorbance of the filtrate was measured at a wavelength of 220 nm using a UV spectrophotometer (UV-1800, Shimadzu).
FTIR MeasurementsSynchrotron FTIR measurement was performed to obtain IR spectra of the ingredients and FXD microparticles at the BL43IR in SPring-8 (Hyogo, Japan). Measurements were made by the transmission method using an FTIR microscope (HYPERION, Bruker Japan, Kanagawa, Japan). BaF2 window was applied. The measurement conditions were as follows; measurement range, 500–4000 cm−1; resolution, 4 cm−1; number of scan times, 256.
Powder X-Ray DiffractionPowder X-ray diffraction patterns of ingredients and FXD microparticles were measured using a SmartLab 9 kW (Rigaku Corporation, Tokyo, Japan). A circular glass sample holder was used as the substrate. The thickness of the filled sample was 0.2 mm. The measurements were performed under the following conditions. Voltage and amperage were 45 kV and 200 mA, diffraction angle (2θ) was 5–30°, step size was 0.02, scanning speed was 10°/min.22)
Synchrotron X-Ray DiffractionFXD and FXD microparticles were irradiated with synchrotron radiation X-rays using BL10C (X-ray wavelength: 0.083 nm) at the Photon Factory (PF) of the High Energy Accelerator Research Organization (KEK). The distance from the sample to the detector was approximately 500 mm and the detector was a pixel array detector (PILATUS3 2M, DECTRIS). A DSC (HCS302-LN190®, Instec, Inc., U.S.A.) was used for temperature scanning from 25 to 260 °C. The rate of temperature increase was 10 °C/min. X-rays were irradiated for 29 s, and spectra were obtained every 30 s.
DSCThe thermal properties of ingredients and FXD microparticles were determined using DSC (Thermo plus EVO II DSC8230, Rigaku Corporation). Samples were weighed at 2 mg, placed in an aluminum pan and sealed, and measured under nitrogen (100 mL/min) at a constant heating rate (10 °C/min) from 25 to 260 °C. Alumina was used as the reference material.
Particle shape is related to various properties of particles. Spherical shape is closely related to flowability, formability, and fillability. The morphological features of FXD and FXD microparticles are shown in Fig. 2. FXD has an elongated, irregular, crystalline shape. FXD microparticles obtained are agglomerated particles and the morphological characteristics are a blocky shape with a rough surface. The Feret diameters of FXD and FXD(EC200) were 20.6 ± 4.0 and 185.0 ± 105.7 µm (mean ± standard deviation (S.D.), n = 50), respectively, indicating an increase in particle size of FXD microparticles. The drug content and the drug recovery of FXD microparticles are shown in Table 2. Drug release profiles from FXD microparticles are shown in Fig. 3. In comparing FXD(EC200) and FXD(EC200)-2, increasing the amount of Tween 80 and PVA as emulsifiers and emulsion stabilizers did not increase the drug recovery (Table 2) or suppress the initial drug release (Fig. 3a).
| Formulation | FXD content | FXD recovery |
|---|---|---|
| % | ||
| FXD(EC200) | 9.6 ± 0.3 | 73.2 ± 2.8 |
| FXD(EC200)-2 | 8.2 ± 0.1 | 66.9 ± 1.4 |
| FXD(EC100) | 13.0 ± 0.7 | 54.2 ± 4.8 |
| FXD(EC100)-M8A2 | 11.1 ± 3.3 | 35.7 ± 15.7 |
| FXD(EC200)-SLS10 | 11.1 ± 0.2 | 87.1 ± 2.2 |
| FXD(EC200)-SLS20 | 11.3 ± 0.7 | 87.6 ± 10.9 |
Each value represents the mean ± S.D. (n = 3).
Each data point represents mean ± S.D. (n = 3).
Next, FXD(EC100) was prepared with half the amount of EC without changing the amount of Tween 80 and PVA. The drug release in the initial 1 min was slightly reduced in FXD(EC100) (Fig. 3b), but the drug recovery decreased. For further controlled drug release from FXD microparticles, FXD(EC100)-M8A2 were prepared using an methanol/acetonitrile mix solution. Drug release from FXD(EC100)-M8A2 after 2 min was suppressed (Fig. 3b), but drug recovery was reduced. From these results, FXD and acetonitrile may have poor affinity, and FXD(EC100)-M8A2 was unable to retain FXD diffusing into the aqueous phase. Based on these results on drug recovery and drug release properties, FXD(EC200) was the optimal FXD microparticle formulation with the highest drug recovery and ability to inhibit the initial drug release.
Structural Analysis of FXD MicroparticlesFTIR studies were performed to detect possible molecular interactions between FXD and other components in the FXD microparticles. Figure 4 presents FTIR spectra of FXD, Tween 80, PVA, EC, and FXD(EC200). In the FXD spectrum, characteristic peaks were observed at 3294 cm−1 (O–H stretching), 1707 cm−1 (C=O carboxylic acid stretching), and 1279 cm−1 (C–N stretching of tertiary amine).23) The Tween 80 spectrum shows intense, sharp absorption peaks due to the different functional groups present in the molecules. The band at 2924 cm−1 was due to -CH2- asymmetric expansion, whereas the band at 2862 cm−1 was due to -CH2- symmetric expansion. The band at 1736 cm−1 can be attributed to C=O and the band at 1126 cm−1 was due to stretching of C–O–C.24) For PVA, the broad peak at 3375 cm−1 was associated with O–H stretching, the peak at 2939 cm−1 with C–H stretching, the 1734 cm−1 peak with C=O groups, the peak at 1431 cm−1 with CH2 bending, and the 1238 cm−1 peak with C–O–C groups.25–27) For EC, a distinct peak at 3479 cm−1 was observed, which was due to -OH groups present on the closed ring structure of repeating units in the polymer.28) The peaks observed at 2981 cm−1 and 2879 cm−1 may be due to -CH stretching. The peak present at 1375 cm−1 was due to -CH3 bending, and the peak at 1444 cm−1 was due to -CH2 bending. The broad peak at 1169 cm−1 may be due to C–O–C stretch in the cyclic.29,30) In contrast, FXD(EC200) spectra showed disappearance of characteristic peaks of FXD at 3294 cm−1 (O–H stretching) and 1707 cm−1 (C=O carboxylic acid stretching). The peak at 1279 cm−1 (C–N stretching of tertiary amine) was included in a broad peak. This may be a result of FXD complexation with other components via hydrogen bonds, van der Waals forces, and hydrophobic interactions.
Powder X-ray diffraction measurements were performed to investigate the crystallinity of FXD in the microparticles. Figure 5 presents powder X-ray diffraction patterns of FXD, PVA, EC, and FXD microparticles. FXD showed sharp crystalline peaks,31) and PVA and EC showed broad peaks. The diffraction patterns of FXD(EC200) and FXD(EC100) showed a broad peak with disappearance of the FXD crystal peaks, suggesting FXD amorphization in FXD microparticles.17) From the results of FTIR spectra and X-ray diffraction patterns of FXD microparticles, the interaction between FXD and other components suggested that a solid dispersion of FXD was formed in the FXD particulates. On the other hand, in the diffraction patterns of FXD(EC200) and FXD(EC100), small peaks were observed at 11.3 and 17.0° (2θ). To investigate the origin of these peaks, synchrotron X-ray diffraction and DSC were performed. Figure 6 presents synchrotron X-ray diffraction patterns and a DSC thermogram of FXD during heat treatment. The X-ray diffraction pattern of FXD (Fig. 6a) showed crystalline peaks consistent with the powder X-ray diffraction pattern and the FXD crystalline peaks disappeared above 230 °C. The DSC thermogram of FXD (Fig. 6b) showed a sharp endothermic peak at 200.2 °C, corresponding to its melting point,32) and an obtuse endothermic peak at 235.7 °C. Because the FXD crystalline peaks disappeared above 230 °C, the endothermic peak at 235.7 °C was thought to be due to the decomposition of FXD. Figure 7a presents synchrotron X-ray diffraction patterns of FXD(EC200). The peaks at 6.0 nm−1, 12.0 nm−1, 15.1 nm−1 (q) disappeared above 150 °C. In a previous report, PVA heated at 10 °C/min revealed a glass transition at 45.3 °C and the onset of melting at 169.4 °C.33) In the DSC thermogram of PVA (Fig. 7b), PVA heated revealed a glass transition at 46.7 °C and the onset of melting at 156.1 °C. On the other hand, EC is known to be a highly or fully amorphous material and its glass transition temperature (Tg) has been determined to be within the region of 130–150 °C, although the measured value appears to be dependent on the type of EC and the measurement method used.34) In this study, the Tg of EC appears near 165 °C. Hence, the peaks at 6.0, 12.0, 15.1 nm−1 (q) may have originated from a complex of PVA and other components. In this study, FXD microparticles were prepared using a solvent co-evaporation method. This method has a slower solidification rate than the fusion method.35) During the solidification of FXD microparticles, FXD may have leached into the solvent and some of it adheres to particle surface. Because PVA is thought to form a hydration layer at the particle-solvent interface,36) it is possible that PVA and FXD form a complex on the particle surface (see supplementary data, S1). From these results, the origin of the peaks at 6.0, 12.0, 15.1 nm−1 (q) may be a complex of PVA and FXD. A possibility that PVA and Tween 80 form complex with FXD cannot be denied. This point needs to be clarified in the future. In the synchrotron X-ray diffraction patterns of FXD(EC200) (Fig. 7a), a diffraction peak appeared at 7.7 nm−1 above 200 °C, and its diffraction intensity increased with increasing temperature. A peak was observed at 209.7 °C in the DSC curve of FXD(EC200), and a diffraction peak appearing above 200 °C was due to the decomposition products of the components.
From the synchrotron X-ray diffraction and DSC measurements, the possibility of complex formation of PVA and FXD was considered. The hydrochloride salt, FXD, is known to be positively charged by ionization of tertiary amines (Fig. 1a). Therefore, we considered if the addition of SLS, an anionic surfactant (Fig. 1b), may inhibit the formation of a complex between FXD and PVA on the surface of the microparticles. Based on the composition of FXD(EC200), the amount of Tween 80 was reduced and SLS was added to the components (Table 1). The morphological characteristics of FXD(EC200)-SLS10 and FXD(EC200)-SLS20 were similar to those of FXD(EC200). The Feret diameters of FXD(EC200)-SLS10 and FXD(EC200)-SLS20 were 192.9 ± 88.5 µm and 187.0 ± 90.3 µm (mean ± S.D., n = 50), respectively; both were equivalent to the particle size of FXD (EC200). On the other hand, The FXD content and the FXD recovery of FXD(EC200)-SLS10 and FXD(EC200)-SLS20 increased compared with those of FXD(EC200) (Table 2). Drug release profiles from FXD(EC200)-SLS10 and FXD(EC200)-SLS20 are shown in Fig. 3c. The initial drug release was further suppressed compared with FXD(EC200). The drug release rate from FXD(EC200)-SLS20 after 1 min was 25.8%, which was clearly inhibited compared with the drug release rate of FXD bulk powder (60.8%) after 1 min, and FXD(EC200)-SLS20 can reduce the bitter taste caused by drugs in the oral cavity. Powder X-ray diffraction patterns of FXD(EC200)-SLS10 and FXD(EC200)-SLS20 are shown in Fig. 5. The diffraction patterns of FXD(EC200)-SLS10 and FXD(EC200)-SLS20 showed a broad peak. In the DSC thermogram of FXD(EC200)-SLS20 (Fig. 7b), no peaks were observed that may be attributed to FXD crystals. Thus, suggesting FXD amorphization in FXD microparticles with SLS. Furthermore, the small peaks at 11.3 and 17.0° (2θ) in Fig. 5 were not indicated. Hence, it was considered that the formation of the complex between PVA and FXD may have been inhibited by adding SLS in FXD microparticles. SLS may be present at the interface between microparticles and the solvent as a surfactant. It is possible that the electrostatic interaction between the negatively charged SLS and the positively charged FXD inhibited FXD from protruding onto the surface of the microparticles, reducing contact between FXD and PVA. (See supplementary data, S1) The suppression of FXD protrusion onto the particle surface may contribute to the inhibition of initial drug release.
This study reports the preparation of EC microparticles for taste masking of FXD. The amounts of EC, Tween 80, and PVA in the compositions had little effect on the initial drug release properties. Based on the results of the drug recovery and drug release properties, FXD(EC200) was the optimal FXD microparticles with the highest drug recovery and ability to inhibit the initial drug release. From the results of FTIR spectra and X-ray diffraction patterns of FXD(EC200), FXD amorphization in the microparticles and the interaction between FXD and other components was suggested, and the formation of a solid dispersion of FXD was suggested. From synchrotron X-ray diffraction and DSC, the possibility of the complex of PVA and FXD on the particle surface was suggested. Therefore, in order to inhibit the formation of a complex between FXD and PVA, SLS was added to the composition of FXD(EC200). The initial drug release from FXD(EC200)-SLS20 was further suppressed compared with FXD(EC200). From the powder X-ray diffraction pattern, it was considered thar the formation of the complex between PVA and FXD was inhibited by adding SLS in FXD microparticles. From these results, FXD microparticles with SLS can be prepared as a controlled-release formulation and expected to be useful in masking bitter tasting particulates.
We would like to thank MITSUBISHI CHEMICAL GROUP (Tokyo, Japan) for providing polyvinyl alcohol. The synchrotron FTIR experiments were performed using a BL43IR at SPring-8 with the approval of the SPring-8 Proposal Review Committee (2022A1293, 2022B1465, 2023A1195). The synchrotron X-ray diffraction experiments were performed using BL10C at Photon Factory with the approval of the Photon Factory Proposal Review Committee (21G094, 23G102). This work was partially supported by JSPS KAKENHI Grant Number JP22K06705 (Yuri Ikeuchi-Takahashi). This work was partially supported by JSPS KAKENHI Grant Number JP23K07775 (Yasuko Obata).
The authors declare no conflict of interest.
This article contains supplementary materials.
