Development of novel sustained-release system, disintegration-controlled matrix tablet (DCMT) with solid dispersion granules of nilvadipine

Journal of Controlled Release 108 (2005) 386 – 395 www.elsevier.com/locate/jconrel

Development of novel sustained-release system, disintegration-controlled matrix tablet (DCMT) with solid dispersion granules of nilvadipine Nobuyuki Tanaka a,d, Keiji Imai a, Kazuto Okimoto b, Satoshi Ueda a, Yuji Tokunaga b, Atsuo Ohike a, Rinta Ibuki c, Kazutaka Higaki d, Toshikiro Kimura d,* a

b

d

Fujisawa Pharmaceutical Co., Ltd., Product Development Laboratories, 1-6 Kashima 2-chome, Yodogawa-ku, Osaka 532-8514, Japan Fujisawa Pharmaceutical Co., Ltd., Pharmaceutical Science Laboratories, 1-6 Kashima 2-chome, Yodogawa-ku, Osaka 532-8514, Japan c Fujisawa Pharmaceutical Co., Ltd., Manufacturing Division, 1-6 Kashima 2-chome, Yodogawa-ku, Osaka 532-8514, Japan Department of Pharmaceutics, Faculty of Pharmaceutical Sciences, Okayama University, 1-1-1 Tsushima-naka, Okayama 700-8530, Japan Received 24 March 2005; accepted 29 August 2005 Available online 25 October 2005

Abstract The goal of this study is to develop a novel sustained-release (SR) system for poorly water-soluble drugs by applying solid dispersion (SD) technique for improving the solubility. The developed SR system, disintegration-controlled matrix tablet (DCMT), consists of hydrogenated soybean oil (HSO) as wax and SD granules containing low-substituted hydroxypropylcellulose (L-HPC) as a disintegrant. In this study, nilvadipine (NiD) was chosen as a model compound. Sustained-release profiles of NiD from DCMT were identically controlled in several dissolution mediums in spite of varying pH and agitation speed. The release of NiD from DCMT was sustained more effectively by increasing the amount of wax or by decreasing the amount of disintegrant, and supersaturation of NiD was achieved without any re-crystallization in dissolution medium. The release rate of NiD from DCMT was controlled by the disintegration rate of tablet. The release profile of NiD was described by the Hixson– Crowell’s model better than zero-order kinetics, first-order kinetics and Higuchi’s model, which supports that the release of NiD from DCMT is regulated by the disintegration of the tablet. From this study, it was clarified that DCMT was one of the promising SR systems applying SD for the poorly water-soluble drugs. D 2005 Elsevier B.V. All rights reserved. Keywords: Disintegration; Sustained release; Nilvadipine; Solid dispersion; Wax

1. Introduction * Corresponding author. Tel.: +81 86 251 7948; fax: +81 86 251 7926. E-mail address: [email protected] (T. Kimura). 0168-3659/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2005.08.024

Nilvadipine (NiD) is a dihydropyridine derivative– calcium antagonist that has been developed by Fujisawa Pharmaceutical Co., Ltd. (Osaka and Tokyo,

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

Japan). The immediate release (IR) tablet has been commercialized as an antihypertensive agent, and it is required to be administrated twice a day. Since NiD is a poorly water-soluble compound, of which solubilities at pH 1.2 and pH 6.8 are about 1 Ag/mL, solid dispersion (SD) technique is applied to the IR tablet. The SD technique has been widely applied to various poorly water-soluble compounds [1,2] including NiD [3,4] in order to improve the bioavailability. For the improvement of compliance to patients, however, it is desired to develop a sustained-release (SR) oral dosage form instead of IR tablet. The SR systems are the methods that can achieve therapeutically effective concentrations of drug in the systemic circulation over an extended period of time. Numerous oral SR systems have been developed such as (a) insoluble, slowly eroding, or swelling matrices, (b) polymer-coated tablets, pellets, or granules and (c) osmotically driven systems [5]. For the development of SR dosage form, the solubility of drug is one of the limitations for availability of drug in gastrointestinal (GI) tract. In order to overcome the dissolution limitation of poorly water-soluble drugs, many challenges have been taken such as co-existing with hydrophilic polymer [6], an osmotic pump system by incorporating cyclodextrin derivatives [7] and nanoparticle technologies [8]. A combination of SD and SR techniques is one of the attractive approaches since supersaturation of drugs can be achieved by applying SD. However, it has been known that the supersaturation level is decreased by contacting SD to water for longer period because of re-crystallization of drugs [3,4,9,10]. That is why only a few reports on the application of SD to SR system have been presented. One approach is a direct modification of the character of SD by using water-insoluble or slower dissolving carriers instead of conventional hydrophilic polymers [11,12]. In this technique, a selection of suitable carrier for each drug would be a critical factor. Another approach is a membrane-controlled SR tablet containing SD [13]. Since the release of drug from such a diffusion-controlled system is driven by the gradient of the drug concentration resulting from penetration of water, it may have a risk for the re-crystallization of drug because of contacting SD to water penetrated into the system for longer period. Therefore, a specific formula of SD and/or a

387

manufacturing method may be required for each drug depending on the character of the drug in order to maintain the supersaturation. In this study, a novel SR system of NiD applying SD technique was developed, and it prevents the recrystallization of NiD during the release by protecting SD from contacting with water for longer period of time. The developed system, disintegration-controlled matrix tablet (DCMT), consists of wax and SD granules containing a disintegrant. The concept of this system is that the wax limits the penetration of water to the surface layer of tablet, and the disintegrant swells with the penetrated water, and then SD granules located on the surface separate from the tablet. A constant rate of tablet disintegration can be achieved by repeating the processes of water penetration and swelling/separating of SD granules. By evaluating the release profile of NiD from DCMT, the mechanisms and factors that could regulate the drug release from this system were clarified in this report.

2. Materials and methods 2.1. Materials NiD was provided by Fujisawa Pharmaceutical Co., Ltd. (Osaka and Tokyo, Japan). Hydroxypropylmethylcellulose (HPMC) and low-substituted hydroxypropylcellulose (L-HPC) were purchased from Shin-Etsu Chemical Co., Ltd. (Tokyo). Lactose, hydrogenated soybean oil (HSO) and magnesium stearate (Mg–St) were purchased from DMV Japan (Tokyo), Miyoshi Oil & Fat Co. Ltd. (Tokyo) and Taihei Chemical Industrial Co., Ltd. (Osaka), respectively. All other materials were of analytical regent grade. 2.2. Preparation of DCMT The preparation of DCMT consists of 3 processes, i.e., preparations of SD granules, HSO-treated granules and tablet. A representative preparation method for each process is as follows. 2.2.1. Preparation of SD granules Forty grams of NiD and 120 g of HPMC were accurately weighed and dissolved in the mixture of 400 mL of ethanol and 400 mL of dichloromethane.

388

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

Then, L-HPC and/or lactose were suspended in the solution. After the evaporation of mixed solvent from the suspended mixture by a vacuum dryer at 40 8C, the dried SD granules were sized using a 500 Am sieve.

where W t is the weight of tested DCMT sampled at time t and Wi is the initial weight of DCMT. The concentrations of NiD were determined using UV spectrometer (HP-8451A, Hewlett Packard) at 246 nm.

2.2.2. Preparation of HSO-treated granules HSO was melted at 85 8C and mixed with prepared SD granules. The mixture was cooled down to room temperature, and it was sized using an 850 Am sieve.

2.5. Evaluation of similarity factor of release profiles

2.2.3. Preparation of DCMT Before compression, the prepared HSO-treated granules were blended with 0.234 g of Mg–St (0.2% (W/W)). The tablet, DCMT, was compressed using convex punches 7 mm in diameter by a tabletting machine (N-30EX, Okada Seiko Co., Ltd., Tokyo). The target compression load was 1000 kg. 2.3. In vitro release study The release test was carried out according to the dissolution test (paddle method) in Japanese Pharmacopoeia (JP). The release media was 900 mL of JP first medium (pH 1.2) and/or JP second medium (pH 6.8), which were maintained at 37 F 0.5 8C. The paddle rotation speeds were 50 and 100 rpm. Samples were taken at different time intervals, and the concentrations of NiD were determined using UV spectrometer (HP-8451A, Hewlett Packard, California, USA) at 246 nm.

In order to compare the release profiles of NiD at different pHs as well as at different paddle rotation speeds, the similarity factor ( f 2) was calculated per the following equation [14]. 82 9 #0:5 < = n X f2 ¼ 50log 41 þ ð1=nÞ jRt  Tt j2  100 : ; t¼1 ð2Þ where R t is the percent dissolved under the standard testing condition, i.e., pH 1.2 at 100 rpm, at time t, and T t is the percent dissolved under the reference testing condition, i.e., pH 6.8 at 100 rpm and/or in pH 1.2 at 50 rpm at time t. 2.6. Evaluation of release kinetics The release kinetics of NiD from DCMT was evaluated according to zero-order kinetics, first-order kinetics, Higuchi’s model and Hixson–Crowell’s model. Zero-order kinetics Q t ¼ Q 0  k0 t

2.4. Disintegration test The disintegration test was carried out by a modified disintegration test of Japanese Pharmacopoeia (JP) using the disintegration apparatus listed in JP at a frequency of 30 cycles per minute. One tablet was tested in 900 mL of JP first medium (pH 1.2), which were maintained at 37 F 0.5 8C. At each sampling time, an aliquot of the fluid was taken out for evaluation of the concentration of NiD, and the tested DCMT was removed from the beaker. The tablet was dried at 30 8C for 12 h, and the disintegration ratio (% disintegrated) was calculated by the following equation. Wi  Wt % Disintegrated ¼  100 Wi

ð1Þ

ð3Þ

where Q t is the amount of drug remaining as a solid state at time t, Q 0 is the initial amount of drug in the pharmaceutical dosage form and k 0 is the zero-order release rate constant. First-order kinetics lnQt ¼ lnQ0  k1 t

ð4Þ

where Q t is the amount of drug remaining as a solid state at time t, Q 0 is the initial amount of drug in the pharmaceutical dosage form and k 1 is the first-order release rate constant. Higuchi’s model pffi Q t ¼ kH t

ð5Þ

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

where Q t is the amount of drug released in time t, and k H is the Higuchi’s (release) rate constant.

389

Table 1 Composition of DCMTs and their formulated amounts Formulation Formulated amounts (mg)

Hixson–Crowell’s model ffiffiffiffiffi ffi pffiffiffiffiffi p 3 Q0  3 Qt ¼ k s t

NiD HPMC Lactose L-HPC

ð6Þ

where Q t is the amount of drug remaining as a solid sate at time t, Q 0 is the initial amount of drug in the pharmaceutical dosage form and k s is the release rate constant.

3. Results and discussion

SD granules 8 HSO-treated 8 granules DCMT 8

HSO

Mg–St

24 24

0–24 0–24

24–128 0 24–128 40–160

0 0

24

0–24

24–128 40–160 0.3–0.6

Total weights range between 136.3 and 320.6 mg.

shown in Fig. 1. Supersaturation in comparison with the intrinsic solubility of NiD was achieved by forming SD with HPMC, and the re-crystallization of NiD was not observed for 4 h.

3.1. Dissolution of NiD from SD granules 3.2. Release of NiD from DCMT The solubilities in water, JP14 first fluid (pH 1.2) and JP14 second fluid (pH 6.8) at 37 8C were 1.1, 1.3 and 1.0 Ag/mL, respectively. In order to improve the solubility of NiD, SD granules were prepared by employing HPMC, one of the most suitable carriers for enhancement of the water-solubility of drugs as well as for prevention of drugs from re-crystallization in the dissolution medium [1,9,15–17]. The formation of SD by using NiD and HPMC has already been confirmed by Okimoto et al. [3]. The dissolution of NiD from SD granules was evaluated in JP14 first fluid (pH 1.2). SD granules containing 8 mg of NiD, 24 mg of HPMC, 36 mg of L-HPC and 12 mg of lactose were tested, and the dissolution profile was

DCMT was prepared by tabletting the HSO-treated SD granules with Mg–St. The composition of DCMT is shown in Table 1 with the amount range of each ingredient tested in this study, and the representative formulations of DCMT are shown in Table 2. In order to evaluate the influences of disintegrant and wax levels on the release rate of NiD from DCMT, in vitro release studies were carried out for DCMTs prepared with varying amounts of L-HPC and HSO. 3.2.1. Influence of disintegrant levels in SD granules on the release of NiD from DCMT To clarify the influence of disintegrant level in SD granules on NiD release profiles, DCMTs were prepared with SD granules containing various amounts (24, 36, 48 and 128 mg) of L-HPC, 80 mg of HSO and 0.2% (W/W) of Mg–St to the total weight of DCMT. The amounts of lactose for SD granules containing 24, 36 and 48 mg of L-HPC were 24, 12 and 0 mg, respectively, to adjust the total weight of SD Table 2 Representative formulations of 3 DCMTs

Fig. 1. Dissolution profiles of NiD from SD granules. Dotted line shows the intrinsic solubility of NiD. The test was performed in accordance with JP14 paddle method; the rotation speed of paddle was 100 rpm, and the dissolution medium (JP14 first medium) was maintained at 37 F 0.5 8C. Each value represents the mean F S.D. (n = 3).

Ingredient

DCMT-1

DCMT-2

DCMT-3

NiD HPMC Lactose L-HPC HSO Mg–St Total (mg/Tab)

8 24 24 24 80 0.3 160.3

8 24 12 36 80 0.3 160.3

8 24 0 88 40 0.3 160.3

The amounts of NiD and excipients are expressed as the unit of mg.

390

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

granules. For the SD granules with 128 mg of L-HPC, no lactose was added. Fig. 2 shows the release profiles of NiD from DCMTs, indicating that the release rates of NiD increased with the increase in the amount of LHPC. HSO was included in the formulation for aiming to limit the penetration of water to the surface layer of tablet. L-HPC was employed to separate the SD granules swollen by penetrated water from the rest of the tablet. Results shown in Fig. 2 clarify that the amount of disintegrant in SD granules influences the release rate of NiD from DCMT, and that the disintegration force is one of the critical factors for regulating the drug release in this system. 3.2.2. Influence of wax levels on the release rate of NiD from DCMT To clarify the influence of wax level on NiD release profiles, the release of NiD from DCMTs prepared with various amounts of HSO was examined. To DCMTs containing 36 mg of L-HPC, 12 mg of lactose was added, and DCMTs with 48 and 128 mg of L-HPC contained no lactose. Fig. 3 (A) shows the effect of HSO on NiD release from DCMTs containing 36 mg of L-HPC. The dissolution of NiD from HSO-treated granules containing 80 mg of HSO and 36 mg of L-HPC, corresponding to a single DCMT, was also investigated (Fig. 3 (A)). Compared with the dissolution from HSO-treated SD granules, the release of NiD from DCMT containing 80 mg of HSO

Fig. 3. Release profiles of NiD from DCMTs containing 36 mg (A), 48 mg (B) and 128 mg (C) of L-HPC with different amount of HSO. The test was performed in accordance with JP14 paddle method; the rotation speed of paddle was 100 rpm, and the dissolution medium (JP14 first medium) was maintained at 37 F 0.5 8C. Each value represents the mean F S.D. (n = 3). Key; (o) SD granules treated with 80 mg of HSO, (5) DCMT with 40 mg of HSO, (D) 60 mg of HSO, ( ) 80 mg of HSO, (E) 100 mg of HSO, (n) 120 mg of HSO, (x) 140 mg of HSO, () 160 mg of HSO.

.

Fig. 2. Release profiles of NiD from DCMTs with different amount of L-HPC. The test was performed in accordance with JP14 paddle method; the rotation speed of paddle was 100 rpm, and the dissolution medium (JP14 first medium) was maintained at 37 F 0.5 8C. Each value represents the mean F S.D. (n = 3). Key; (E) DCMT with 24 mg of L-HPC, ( ) 36 mg of L-HPC, (n) 48 mg of L-HPC, (x) 128 mg of L-HPC.

.

remarkably decreased. This result suggests that compressing HSO-treated granules would be necessary to control the release of NiD. Furthermore, the increase in the amount of HSO from 80 up to 140 mg decreased the release rate of NiD from DCMTs. On the other hand, the decrease in the amount of HSO down to 40 mg also decreased the release rate of NiD.

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

These tablets containing less than 80 mg of HSO did not disintegrate completely, and they remained as the swollen form in dissolution vessels. This might be ascribed to the faster penetration rate of water into DCMT than the disintegration rate of DCMT. If water penetrates deeply into the tablet before disintegration, SD granules located inside the tablet become swollen and form gel. Once gel forms, the tablet is not able to disintegrate any more. That is why SD, which contacts with water for longer time of period, could lead to the re-crystallization of NiD. Similar phenomena were observed for DCMTs containing 48 mg of LHPC, although the release rates tended to be higher than those from DCMTs containing 36 mg of L-HPC (Fig. 3 (B)). Fig. 3 (C) shows that the increase of HSO from 40 to 160 mg decreased the release of NiD from DCMTs containing 128 mg of L-HPC, although the increase in L-HPC amount increased the release comparing with that from DCMTs containing 48 mg of L-HPC. To summarize the results shown in Fig. 3 (A), (B) and (C), the values of 1 / T 50% were plotted against the ratio of HSO amount to the total weight of DCMT (HSO / Total (W/W (%)) in Fig. 4, where T 50% means the time to release 50% NiD. This figure shows that DCMTs containing 128 mg of L-HPC could control the release rate of NiD from DCMTs by changing the amount of HSO. However, this formulation might not be adequate for achieving a well sustained-release profile because the release of NiD completed in 4 h even though the amount of HSO was increased to 160 mg (Fig. 3 (C)). On the other hand, DCMTs contain-

Fig. 4. Relationship between 1 / T 50% and the ratio of HSO amount to the total weight of DCMT. Each value represents the mean F S.D. (n = 3). Dotted line represents the least-squares regression line for 3 or 4 points in each preparation. Key; ( ) DCMT with 36 mg of LHPC, (E) 48 mg of L-HPC, (n) 128 mg of L-HPC.

.

391

ing 36 or 48 mg of L-HPC showed the good linearity when more than 50% of HSO was contained, and they could sustain the release of NiD for longer period of time (Fig. 3 (A) and (B)) at the same time. The linearity between 1 / T 50% and HSO / Total (W/W)% found at the different ranges depending on the amounts of L-HPC shows that there is a suitable ratio between the amounts of HSO and L-HPC for controlling the release of NiD from DCMTs. From these results, it was clarified that the amount of wax, HSO, and its suitable ratio to disintegrant, LHPC, in DCMT were critical factors to regulate the release rate of NiD from DCMTs without any recrystallization of NiD. 3.3. Influence of pH and agitation speed on the release rate After oral administration, pharmaceutical dosage forms are exposed to various physiological factors such as gastrointestinal fluids with different pHs and GI motility, and they may vary depending on sites of gastrointestinal tract, days and/or patients. Therefore, the release kinetics of drugs from SR dosage form is desired to be constant under any physiological conditions. In the present study, the effect of pH (Fig. 5 (A)) or agitation intensity (Fig. 5 (B)) on the release of NiD was investigated for DCMT containing 80 mg of HSO and 24 mg (DCMT-1) or 36 mg (DCMT-2) of L-HPC. The composition of these DCMTs is listed in Table 2 with that for DCMT-3. Release tests were performed at pH 1.2 or pH 6.8 (Fig. 5 (A)), and at the paddle rotation speed of 100 or 50 rpm (Fig. 5 (B)). To compare the release profiles of NiD under each condition, a similarity factor ( f 2) was calculated. The similarity factor has been adopted by Center for Drug Evaluation and Research (FDA), Human Medicines Evaluation Unit of the European Agency for the Evaluation of Medical Products (EMEA) and National Institute of Health Sciences of Japan (NIHS) as a criterion for the assessment of similarity between two in vitro release profiles. The f 2 values between the 2 different pHs were 78.0 and 77.9 for DCMT-1 and DCMT-2, respectively. On the other hand, 63.3 for DCMT-1 and 54.4 for DCMT-2 were calculated between the 2 different agitating speeds. Since the release profiles are judged to be similar to each other when the f 2 value ranges between 50 and 100

392

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

release of NiD from DCMTs, but it was also suggested that compressing the granules would be necessary to control the release of NiD. Therefore, to confirm the importance of compressing the granules, the release of NiD from DCMT-2 was compared with its intermediate products, i.e., SD granules and HSOtreated granules, as well as the compressed SD granules (Fig. 6). All samples contained the same SD granules, and the HSO-treated granules corresponded to a single DCMT-2 with the exception of Mg–St. The dissolution profile of NiD from HSO-treated SD granules was very similar to that from SD granules that were not treated with HSO, suggesting that HSO does not cover the surface of granules completely enough to suppress the penetration of water. Furthermore, the dissolution profile of NiD from the compressed SD granules was similar to that from SD granules as well, indicating that the reduction of surface area by compression did not affect the release of NiD. Taken all together, it is suggested that compressing the HSOtreated granules could make the wax covering of the granules contact each other and form a hydrophobic wax matrix layer, which can prevent water from penetrating into the tablet. Fig. 5. Effects of pHs (A) or paddle rotation speeds (B) on release profiles of NiD from DCMT. (A) JP14 first or second medium was used at 37 F 0.5 8C and the rotation speed of paddle was 100 rpm. Keys; (n) DCMT-1 in JP14 first medium (pH 1.2), (5) DCMT-1 in JP14 second medium (pH 6.8), ( ) DCMT-2 in JP14 first medium, (o) DCMT-2 in JP14 second medium. (B) JP14 first medium maintained at 37 F 0.5 8C was used as the dissolution medium. Keys; (n) DCMT-1 at 100 rpm, (5) DCMT-1 at 50 rpm, ( ) DCMT-2 at 100 rpm, (o) DCMT-2 at 50 rpm. Each value represents the mean F S.D. (n = 3).

.

3.4.2. Comparisons of disintegration profile of DCMT with release profile of NiD from DCMT In order to investigate the relationship between the disintegration of DCMT and the release of NiD from DCMT, the release profile of NiD from DCMT-2 was

.

[14], it was concluded that the release profiles of NiD from the two DCMTs were similar at different pHs as well as at different agitating speeds. This result suggests that the release kinetics of NiD from these DCMTs might not be changed so much by the physiological factors in gastrointestinal tract. 3.4. Mechanism for release properties of DCMT 3.4.1. Role of compression in the control of NiD release from DCMT As described in the section 3.2.2, the amount of HSO is one of the critical factors controlling the

Fig. 6. Effect of compression on the release of NiD. The test was performed in accordance with JP14 paddle method; the rotation speed of paddle was 100 rpm, and the dissolution medium (JP14 first medium) was maintained at 37 F 0.5 8C. Each value represents the mean F S.D. (n = 3). Key; (5) SD granules, (n) compressed SD granules, (o) SD granules treated with HSO, ( ) DCMT-2.

.

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

393

compared with the disintegration profile of the tablet (Fig. 7). The release profile of NiD was almost superimposed on the disintegration profile of DCMT, i.e. the decrease of tablet weight–time profile, strongly indicating that the release rate of NiD from DCMT was controlled by the disintegration rate of the tablet, which is different from an insoluble matrix system and/or an erosion system. 3.4.3. Kinetics of release profile To investigate the mechanism of NiD release from DCMTs, the release study was performed for DCMTs, of which the weight and the diameter were fixed to 160.3 mg and 7 mm, respectively, by considering the appropriate size for oral administration to human (Fig. 8 (A)), and the regression analysis of the release profiles was performed by zero-order kinetic model, first-order kinetic model, Higuchi’s model and Hixson–Crowell’s model (Table 3). Several theories and kinetic models have been established to describe the drug release kinetics, and they are practically useful for understanding the mechanism of drug release from pharmaceutical dosage forms. Zero-order kinetics can be used to describe the drug release from several types of modified release pharmaceutical dosage form such as matrix tablets for drugs with low solubility [18], osmotic systems. First-order kinetics [19,20] can describe the release profile from the pharmaceutical dosage forms such as those containing water-soluble drugs in por-

Fig. 7. Comparison between the disintegration profile of DCMT-2 and the release profile of NiD from DCMT-2. The test was performed using the disintegration apparatus in JP14. The frequency was 30 cycle per minute, and the disintegration medium (JP14 first medium) was maintained at 37 F 0.5 8C. Each value represents the mean F S.D. (n = 3). Key; ( ) Disintegration of DCMT-2, (o) Release of NiD from DCMT-2.

.

Fig. 8. Release profiles of NiD from DCMTs (A) and cube root plot of released drug based on Hixson–Crowell’s equation (B). The test was performed in accordance with JP14 paddle method. The rotation speed of paddle was 100 rpm, and the dissolution medium (JP14 first medium) was maintained at 37 F 0.5 8C. Each value represents the mean F S.D. (n = 3). Solid lines in (B) represent leastsquares regression lines. Keys; (E) DCMT-1, ( ) DCMT-2, (n) DCMT-3.

.

ous matrices [21], where drugs would be released at the rates proportional to the amounts of drug remaining in the interior of dosage form. Higuchi’s model [22], based on the Fick’s low, can be applied for the drug release that is regulated by the diffusion of drugs within the formulation. In this case, the cumulative released amounts versus square root of time give the straight line. Drug release from matrix tablets containing water-soluble drugs can be described by Higuchi’s model [23–26]. Hixson–Crowell’s cube root model [27] can be applied to the pharmaceutical dosage form whose drug release rate is proportional to the surface area of dosage form such as the erosion-dependent release systems [28,29].

394

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

Table 3 Release rate constant and linearity of NiD release profiles from DCMTs based on four release models Release model Zero-order First-order Higuchi Hixson–Crowell

r2 k 0 (mg/h) r2 k 1 (h 1) r2 k H (mg/h1 / 2) r2 k S (mg1 / 3 / h)

DCMT-1

DCMT-2

DCMT-3

0.993 0.710 0.849 0.414 0.963 2.502 0.991 0.081

0.963 1.210 0.765 0.516 0.991 3.893 0.997 0.165

0.921 2.020 0.784 0.514 0.974 5.666 0.995 0.431

Release rate constants (k 0, k 1, k H and k s) and the values of squared correlation coefficient (r 2 ) were obtained by the regression analysis for each model.

As shown in Table 3, Hixson–Crowell’s model gave the highest value of the squared correlation coefficient (r 2) for every DCMT, indicating that Hixson–Crowell’s model would be the most suitable model for describing the release of NiD from DCMTs. Fig. 8 (B) shows the plots of NiD release based on Hixson– Crowell’s model and the fitting patterns obtained by the regression analysis with Eq. (6). This result suggests that the release of NiD from DCMTs is not controlled by the drug diffusion in the tablet, but controlled by the disintegration of the tablet. From these results, the mechanism of NiD release from DCMT can be considered as follows: 1) the penetration of water into DCMT is effectively limited by wax

1. Penetration of water

2. Swelling of surface

layer, 2) the disintegrant in SD granules is swollen by the penetrated water, 3) the swollen SD granules separate from the surface of the tablet, and 4) drug dissolves from the separated SD granules (Fig. 9).

4. Conclusion To achieve the SR system for poorly water-soluble drugs, DCMT that consists of wax and SD granules containing a disintegrant was prepared for NiD. The release of NiD from DCMTs was successfully sustained without any re-crystallization. Wax layer effectively limits the penetration of water into the tablet, and the disintegrant contained in SD granules is gradually swollen by the penetrated water, and then the granules are separated from the DCMT, which is a rate-limiting step for NiD to release from DCMT. This disintegration-controlled release mechanism gives the release kinetics following Hixson–Crowell’s model by repeating the processes of water penetration and swelling/separating of SD granules under various conditions. Although the function of disintegration is often applied to immediate release dosage forms, the concept of DCMT, disintegration-controlled SR system, is quite novel and DCMT is a promising SR system, which can be combined with SD, for poorly watersoluble drugs.

3. Separation of swollen granules

Water

DCMT

4. Dissolution of drug from separated granules

Released drug Fig. 9. Proposed mechanism of drug release from DCMT.

N. Tanaka et al. / Journal of Controlled Release 108 (2005) 386–395

References [1] K. Yamashita, T. Nakate, K. Okimoto, A. Ohike, Y. Tokunaga, R. Ibuki, K. Higaki, T. Kimura, Establishment of new preparation method for solid dispersion formulation of tacrolimus, Int. J. Pharm. 267 (2003) 79 – 91. [2] F. Cui, M. Yang, Y. Jiang, D. Cun, W. Lin, Y. Fan, Y. Kawashima, Design of sustained-release nitrendipine microspheres having solid dispersion structure by quasi-emulsion solvent diffusion method, J. Control. Release 91 (2003) 375 – 384. [3] K. Okimoto, M. Miyake, R. Ibuki, M. Yasumura, N. Ohnishi, T. Nakai, Dissolution mechanism and rate of solid dispersion particles of nilvadipine with hydroxypropylmethylcellulose, Int. J. Pharm. 159 (1997) 85 – 93. [4] N. Hirasawa, S. Ishise, H. Mityata, K. Danjo, Physicochemical characterization and drug release studies of nilvadipine solid dispersions using water-insoluble polymer as a carrier, Drug Dev. Ind. Pharm. 29 (3) (2003) 339 – 344. [5] S.A. Charman, W.N. Charman, Oral modified-release delivery systems, in: M.J. Rathbone, J. Hadgraft, M.S. Roberts (Eds.), Modified-Release Drug Delivery Technology, Drugs and The Pharmaceutical Sciences, vol. 126, Marcel Dekker, Inc., New York, 2003, pp. 1 – 10. [6] T. Ishikawa, Y. Watanabe, K. Takayama, H. Endo, M. Matsumoto, Effect of hydroxypropylmethylcellulose (HPMC) on the release profiles and bioavailability of a poorly water-soluble drug from tablets prepared using macrogol and HPMC, Int. J. Pharm. 202 (2000) 173 – 178. [7] K. Okimoto, M. Miyake, N. Ohnishi, R.A. Rajewski, V.J. Stella, T. Irie, K. Uekama, Design and evaluation of an osmotic pump tablet (OPT) for prednisolone, a poorly water soluble drug, using (SBE)7m-h-CD, Pharm. Res. 15 (10) (1998) 1562 – 1568. [8] G.J. Vergote, C. Vervaet, I. Van Driessche, S. Hoste, S. De Smedt, J. Demeester, R.A. Jain, S. Ruddly, J.P. Remon, An oral controlled release matrix pellet formulation containing nanocrystalline ketoprofen, Int. J. Pharm. 219 (2001) 81 – 87. [9] I. Sugimoto, K. Sasaki, A. Kuchiki, T. Ishihara, H. Nakagawa, Stability and bioavailability of nifedipine in fine granules, Chem. Pharm. Bull. 30 (12) (1982) 4479 – 4488. [10] K. Takayama, N. Nambu, T. Nagai, Dissolution kinetics for coprecipitates of indomethacin with polyvinylpyrrolidone, Chem. Pharm. Bull. 28 (1980) 3304 – 3309. [11] T. Ozeki, H. Yuasa, Y. Kanaya, Application of the solid dispersion method to the controlled release of medicine: IX. Difference in the release of flurbiprofen from solid dispersions with poly(ethylene oxide) and hydroxypropylcellulose and the interaction between medicine and polymers, Int. J. Pharm. 155 (1997) 209 – 217. [12] M. Yang, F. Cui, B. You, L. Wang, P. Yue, Y. Kawashima, A novel pH-dependent gradient-release delivery system for nitrendipine II. Investigations of the factors affecting the release behaviors of the system, Int. J. Pharm. 286 (2004) 99 – 109. [13] J. Kercˇ, Three-phase pharmaceutical form — Threeform — with controlled release of amorphous active ingredient for once-daily administration, in: M.J. Rathbone, J. Hadgraft,

[14] [15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

395

M.S. Roberts (Eds.), Modified-Release Drug Delivery Technology, Drugs and The Pharmaceutical Sciences, vol. 126, Marcel Dekker, Inc., New York, 2003, pp. 115 – 123. J.W. Moore, H.H. Flanner, Mathematical comparison of dissolution profiles, Pharm. Technol. 20 (6) (1996) 64 – 74. H. Suzuki, H. Sunada, Influence of water-soluble polymers on the dissolution of nifedipine solid dispersions with combined carriers, Chem. Pharm. Bull. 46 (1998) 482 – 487. N. Kohri, Y. Yamayoshi, H. Xin, K. Iseki, N. Sato, S. Todo, K. Miyazaki, Improving the oral bioavailability of albendazole in rabbits by the solid dispersion technique, J. Pharm. Pharmacol. 51 (1999) 159 – 164. I. Kushida, M. Ichikawa, N. Asakawa, Improvement of dissolution and oral absorption of ER-34122, a poorly water-soluble dual 5-lipoxygenase/cyclooxygenase inhibitor with anti-inflammatory activity by preparing solid dispersion, J. Pharm. Sci. 91 (1) (2002) 258 – 266. C.G. Varelas, D.G. Dixon, C. Steiner, Zero-order release from biphasic polymer hydrogels, J. Control. Release 34 (1995) 185 – 192. M. Gibaldi, S. Feldman, Establishment of sink conditions in dissolution rate determinations — theoretical considerations and application to nondisintegrating dosage forms, J. Pharm. Sci. 56 (10) (1967) 1238 – 1242. J.G. Wagner, Interpretation of percent dissolved-time plots derived from in vitro testing of conventional tablets and capsules, J. Pharm. Sci. 58 (10) (1969) 1253 – 1257. N.V. Mulye, S.J. Turco, A simple model based on first order kinetics to explain release of highly water soluble drugs from porous dicalcium phosphate dehydrate matrices, Drug Dev. Ind. Pharm. 21 (8) (1995) 943 – 953. T. Higuchi, Mechanism of sustained-action medication. Theoretical analysis of rate of release of solid drugs dispersed in solid matrices, J. Pharm. Sci. 52 (12) (1963) 1145 – 1149. S.J. Desai, P. Singh, A.P. Simonelli, W.I. Higuchi, Investigation of factors influencing release of solid drug dispersed in inert matrices: III. Quantitative studies involving the polyethylene plastic matrix, J. Pharm. Sci. 55 (11) (1966) 1230 – 1234. S.J. Desai, P. Singh, A.P. Simonelli, W.I. Higuchi, Investigation of factors influencing release of solid drug dispersed in inert matrices: IV. Some studies involving the polyvinyl chloride matrix, J. Pharm. Sci. 55 (11) (1966) 1235 – 1239. J.B. Schwartz, A.P. Simonelli, W.I. Higuchi, Drug release from wax matrices: I. Analysis of data with first-order kinetics and with the diffusion-controlled model, J. Pharm. Sci. 57 (2) (1968) 274 – 277. J.B. Schwartz, A.P. Simonelli, W.I. Higuchi, Drug release from wax matrices: II. Application of a mixture theory to the sulfanilamide-wax system, J. Pharm. Sci. 57 (2) (1968) 278 – 282. A.W. Hixson, J.H. Crowell, Dependence of reaction velocity upon surface and agitation, Ind. Eng. Chem. 23 (1931) 923 – 931. G. Yan, H. Li, R. Zhang, D. Ding, Preparation and evaluation of a sustained-release formulation of nifedipine HPMC tablets, Drug Dev. Ind. Pharm. 26 (6) (2000) 681 – 686. P. Costa, J.M.S. Lobo, Modeling and comparison of dissolution profiles, Eur. J. Pharm. Sci. 13 (2001) 123 – 133.