permeation system reflecting the gastric dissolution process

European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111

Contents lists available at ScienceDirect

European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb

Research Paper

Effects of gastric pH on oral drug absorption: In vitro assessment using a dissolution/permeation system reflecting the gastric dissolution process Makoto Kataoka ⇑, Miho Fukahori, Atsumi Ikemura, Ayaka Kubota, Haruki Higashino, Shinji Sakuma, Shinji Yamashita Faculty of Pharmaceutical Sciences, Setsunan University, 45-1 Nagaotoge-cho, Hirakata, Osaka 573-0101, Japan

a r t i c l e

i n f o

Article history: Received 27 July 2015 Revised 12 January 2016 Accepted in revised form 2 February 2016 Available online 9 February 2016 Keywords: Basic drug Dissolution Poorly water-soluble drug Solubility Stomach

a b s t r a c t The aim of the present study was to evaluate the effects of gastric pH on the oral absorption of poorly water-soluble drugs using an in vitro system. A dissolution/permeation system (D/P system) equipped with a Caco-2 cell monolayer was used as the in vitro system to evaluate oral drug absorption, while a small vessel filled with simulated gastric fluid (SGF) was used to reflect the gastric dissolution phase. After applying drugs in their solid forms to SGF, SGF solution containing a 1/100 clinical dose of each drug was mixed with the apical solution of the D/P system, which was changed to fasted state-simulated intestinal fluid. Dissolved and permeated amounts on applied amount of drugs were then monitored for 2 h. Similar experiments were performed using the same drugs, but without the gastric phase. Oral absorption with or without the gastric phase was predicted in humans based on the amount of the drug that permeated in the D/P system, assuming that the system without the gastric phase reflected human absorption with an elevated gastric pH. The dissolved amounts of basic drugs with poor water solubility, namely albendazole, dipyridamole, and ketoconazole, in the apical solution and their permeation across a Caco-2 cell monolayer were significantly enhanced when the gastric dissolution process was reflected due to the physicochemical properties of basic drugs. These amounts resulted in the prediction of higher oral absorption with normal gastric pH than with high gastric pH. On the other hand, when diclofenac sodium, the salt form of an acidic drug, was applied to the D/P system with the gastric phase, its dissolved and permeated amounts were significantly lower than those without the gastric phase. However, the oral absorption of diclofenac was predicted to be complete (96–98%) irrespective of gastric pH because the permeated amounts of diclofenac under both conditions were sufficiently high to achieve complete absorption. These estimations of the effects of gastric pH on the oral absorption of poorly watersoluble drugs were consistent with observations in humans. In conclusion, the D/P system with the gastric phase may be a useful tool for better predicting the oral absorption of poorly water-soluble basic drugs. In addition, the effects of gastric pH on the oral absorption of poorly water-soluble drugs may be evaluated by the D/P system with and without the gastric phase. Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction The gastrointestinal state in humans is known to affect drug absorption. Many researchers have reported that food consumption and gastric emptying often markedly affect the absorption of orally administered drugs [32,10]. The pH of gastrointestinal fluid has been shown to significantly affect the fraction of the dose absorbed (Fa) of a drug [27,1,28]. The pH of gastric fluid in healthy volunteers is typically acidic (from approximately 1 to 2) [33]; however, it is significantly altered in human subjects with gastric ⇑ Corresponding author. Tel./fax: +81 72 866 3126. E-mail address: [email protected] (M. Kataoka). http://dx.doi.org/10.1016/j.ejpb.2016.02.002 0939-6411/Ó 2016 Elsevier B.V. All rights reserved.

anacidity or hypoacidity or in those being treated with antacids [23,26]. The systemic exposure of poorly water-soluble drugs such as albendazole, dipyridamole, and ketoconazole was previously reported to be significantly influenced by the pH of gastric fluid [27,1,28]. Since these drugs have basic physicochemical properties, drug dissolution in the stomach accretes under acidic conditions, and subsequently induces supersaturated state of drugs in the intestinal fluid. Therefore, the dissolution of poorly water-soluble basic drugs under acidic conditions in the stomach influences their absorption. Recent studies have advocated the interplay between drug dissolution in the intestinal fluid and permeation across the intestinal epithelium as the key factor determining the absorption of a drug

104

M. Kataoka et al. / European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111

after its oral administration [12,5]. Although the dissolution of lipophilic drugs is enhanced by the mixed micelles that form with bile acid and lecithin, owing to the encapsulation of drugs in micelles, the apparent permeabilities of these drugs are known to be significantly reduced by this encapsulation [11]. Since drug dissolution and permeation are closely linked, both processes need to be simultaneously evaluated in order to predict drug absorption. Previous studies using in vitro systems assessed drug dissolution and permeation in the intestinal tract [7,20,25,2]. We established an in vitro system for the simultaneous evaluation of drug dissolution and permeation under physiological conditions, called the dissolution/permeation system (D/P system) [15]. The D/P system enables not only an assessment of the interplay between drug dissolution and permeation, but also the prediction of Fa in humans as a final outcome. We previously demonstrated that the D/P system had the ability to evaluate the effects of various factors, such as food, dose, and formulation, on the oral absorption of poorly water-soluble drugs [16,17,18,19]. However, our previous system focuses on drug dissolution and permeation in the small intestine only. The D/P system equipped with the gastric dissolution process may predict not only the oral absorption of poorly water-soluble basic drugs, but also the effects of pH changes in gastric fluid on the oral absorption of poorly water-soluble drugs. In the present study, the gastric dissolution phase was combined with the D/P system in order to confirm the importance of the gastric dissolution process in predicting the oral absorption of poorly water-soluble drugs using this system. The effects of pH changes in gastric fluid on the oral absorption of poorly water-soluble drugs were also evaluated by the D/P system with the gastric phase. 2. Experimental 2.1. Materials The human colorectal adenocarcinoma cell line, Caco-2, was purchased from the American Type Culture Collection (Rockville, MD) at passage 17. Dulbecco’s Modified Eagle’s Medium (DMEM) was obtained from Sigma–Aldrich (St. Louis, MO). Non-essential amino acids (10 mM), fetal bovine serum (FBS), trypsin-EDTA (trypsin: 0.25%, EDTA: 1 mM), and an antibiotic–antimycotic mixture (penicillin: 10,000 U/mL, streptomycin: 10 mg/mL, amphotericin B: 25 lg/mL; dissolved in 0.85% (w/v) sodium chloride aqueous solution) were purchased from Gibco Laboratories (Lenexa, KS). Cell culture inserts with polyethylene terephthalate filters (pore size: 3.0 lm, growth area: 4.20 cm2) were obtained from Becton Dickinson Bioscience (Bedford, MA). Bovine serum albumin (BSA), dantrolene sodium, dipyridamole, diclofenac sodium, egg-phosphatidylcholine (lecithin), and sodium taurocholate were obtained from Wako Pure Chemical Industries Co., Ltd. (Osaka, Japan). Albendazole was purchased from Sigma– Aldrich (St. Louis, MO). Fluconazole and ketoconazole were obtained from LKT Laboratories, Inc. (MN, USA). DIFLUCANÒ capsules at 50 mg (fluconazole), ESKAZOLEÒ tablets at 200 mg (albendazole), NINAZOLÒ tablets at 200 mg (ketoconazole), and VoltarenÒ tablets at 25 mg (diclofenac sodium) were obtained from Pfizer Japan Inc. (Tokyo, Japan), GlaxoSmithKline K.K. (Tokyo, Japan), T.O. PHARMA Co., Ltd. (Bangkok, Thailand), and Novartis Pharma K.K. (Tokyo, Japan), respectively.

42–62) were seeded on cell culture inserts at a density of 3
105 cells/insert with DMEM supplemented with 10% (v/v) FBS, 1% (v/v) non-essential amino acids, and 0.5% (v/v) antibiotic–antimycotic mixture (culture medium, CM). Fresh CM (1.5 mL in the insert and 2.6 mL in the well) was replenished every 48 h during the initial 6 days and thereafter every 24 h. After 18–21 days in culture, Caco-2 monolayers were utilized in subsequent experiments. 2.2.2. Chambers for the D/P system and the compartment for gastric dissolution A small vessel with a volume set to 6.0 mL, which was consistently stirred at 200 rpm with magnetic stirrers, was used as the gastric compartment (Fig. 1). In the D/P system, the Caco-2 cell monolayer was mounted in side-by-side chambers (Fig. 1). Both sides of the Caco-2 cell monolayer were consistently stirred at 200 rpm with magnetic stirrers. The volumes of the apical and basal sides were set to 8.0 mL and 5.5 mL, respectively. 2.2.3. Preparation of various solutions used in the present study Hank’s balanced salt solution was used as a basic buffer solution in this study (transport medium, TM). Fasted state-simulated intestinal fluid contained 3 mM sodium taurocholate and 0.75 mM lecithin in TM (FaSSIFmod, pH 6.5). Simulated gastric fluid (SGF) and concentrated FaSSIFmod (FaSSIFmod8/6.5) were prepared by TM with an adjusted pH at 1.2 and TM with the addition of sodium taurocholate (4 mM) and lecithin (1 mM), respectively. The pH of FaSSIFmod8/6.5 was adjusted to 7.6 with 4-(2-hydroxyethyl)-1-piper azineethanesulfonic acid (HEPES). When SGF (1.5 mL) was added to FaSSIFmod8/6.5 (6.5 mL) in the apical compartment of the D/P system, the apical solution became FaSSIFmod with pH 6.5. As a basal medium in the D/P system, TM containing BSA (4.5% w/v) with pH adjusted to 7.4 with HEPES was used. 2.2.4. Dissolution study of various drugs in the gastric compartment Drugs were either used as component of a broke tablet (albendazole, diclofenac sodium, and ketoconazole), contents of capsules (dantrolene sodium and fluconazole), or active pharmaceutical ingredients (dipridamole). The small vessel for the gastric compartment was filled with 6 mL SGF. Four percent of the clinical dose of each drug (16 mg

Gastric Dissolution Phase

Small Intestinal Dissolution Phase

Drug Absorption Phase Sampling

Drug (Solid state)

SGF

FaSSIFmod (pH 6.5)

Stirring

Stirring

pH 7.4 (4.5w/v% BSA)

Stirring

Caco-2 Cell Monolayer

2.2. In vitro experiments with the D/P system 2.2.1. Preparation of Caco-2 cell monolayers Caco-2 cell monolayers were obtained in accordance with a previously reported procedure [19]. Briefly, Caco-2 cells (passages

Dissolution/Permeation (D/P) System Fig. 1. Scheme showing the dissolution/permeation system (D/P system) with the gastric dissolution phase.

105

M. Kataoka et al. / European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111

for albendazole, 4 mg for dipyridamole, 8 mg for fluconazole and ketoconazole, 2 mg for dantrolene sodium, 1 mg for diclofenac sodium) was added to the vessel as a powder. Aliquots of samples (0.1 mL) were routinely collected from the small vessel over a period of 1 h. All samples were immediately filtered through a polytetrafluoroethylene filter (MillexÒ-LH, pore size: 0.45 lm, Millipore, Billerica, MA), and each filtrate (0.05 mL) was mixed with 0.45 mL of the solution consisting of water and acetonitrile (50/50) in order to prevent drug precipitation in SGF. All experiments were performed at 37 °C under stirring at 200 rpm with a magnetic stirrer. 2.2.5. Dissolution and permeation study without the gastric dissolution process A dissolution and permeation study without the gastric dissolution process was performed in accordance with a previously reported procedure [19]. Briefly, the Caco-2 cell monolayer with a support filter was mounted between the chambers of the D/P system after a preincubation with appropriate media for 20 min. The apical and basal sides of the Caco-2 cell monolayer were then filled with FaSSIFmod and the basal medium. An appropriate amount of each drug in its powder form was applied to the apical solution. Aliquots of samples (0.1 mL) were routinely collected from the apical and basal solutions for 2 h. All apical samples were immediately filtered through a polytetrafluoroethylene filter (MillexÒ-LH, pore size: 0.45 lm, Millipore, Billerica, MA) and each filtrate (0.05 mL) was mixed with 0.45 mL of the solution consisting of water and acetonitrile (50/50) in order to prevent drug precipitation in FaSSIFmod. The transepithelial electric resistance (TEER) of the Caco-2 cell monolayer was measured before and after the experiment by MillicellÒ-ERS (Millipore, Billerica, MA). All experiments were performed at 37 °C. 2.2.6. Dissolution and permeation study with the gastric dissolution process The apical and basal sides of the D/P system with an attached Caco-2 cell monolayer were filled with 6.5 mL of FaSSIFmod8/6.5 and 4.0 mL of the basal medium, respectively. A total of 1.5 mL of the drug solution derived from the dissolution study under gastric conditions (see Section 2.2.4.) and 1.5 mL of the basal solution were introduced into the apical and basal sides of the system, respectively. Sample collection and other procedures were performed as described in Section 2.2.5. 2.3. Assay Collected basal samples (0.1 mL) were mixed with 0.9 mL of acetonitrile. The mixture was shaken and the supernatant collected by centrifugation at 15,000 rpm (Himac CF15R, HITACHI, Tokyo, Japan) at 4 °C for 20 min. The amount of each drug in the treated solution from the apical and basal solutions was determined using a UPLC system (ACQUITYÒ UPLC, Waters, MA) equipped with a tandem mass spectrometer (ACQUITYÒ TQD, Waters, MA). A reversephase Waters AcquityÒ UPLC BEH C18 analytical column of 50 mm in length
2.1 mm in I.D. and a 1.7-lm particle size (Waters, MA) was used with a mobile phase consisting of 0.1% (v/v) formic acid in water (solvent A) and acetonitrile containing 0.1% (v/v) formic acid (solvent B) with a gradient time period and maintained at 40 °C. The initial mobile phase was 98% solvent A and 2% solvent B pumped at a flow rate of 0.3 mL/min. Between 0 and 1.0 min, the percentage of solvent B increased linearly to 95%, at which it was maintained for 1.0 min. Between 2.01 and 2.5 min, the percentage of solvent B decreased linearly to 2%. This condition was maintained until 3 min, at which time the next sample was injected into the UPLC system. All treated samples were injected at a volume of 5 lL into the UPLC system. The ion

detection conditions used to determine the concentration of each drug are listed in Table 1. All standard curves were treated in the same manner as the obtained samples (apical and basal). Linearity of apical standard curves (r > 0.99) was observed at a concentration range of 0.1–500 lg/mL for all drugs. Linearity of basal standard curves (r > 0.99) was observed at a concentration range of 0.01–10 lg/mL for albendazole, dantrolene, dipyridamole and ketoconazole, 0.01–20 lg/mL for diclofenac and 0.01–50 lg/mL for fluconazole. The limit of quantification of all drugs was 0.1 mg/mL for apical samples and 0.01 mg/mL for basal samples. All data were expressed as the percentage of the dissolved or permeated drug to the amount applied to the D/P system (% of dose). 2.4. Prediction of oral absorption from in vitro experiments with the D/P system The following equation (Eq. (1)) was used to estimate the fraction of the dose absorbed (Fa%) of each drug in a fasted human (in vivo) from in vitro data:

Fa ð%Þ ¼

Absmax  PAc ; PA50 c þ PAc

ð1Þ

where Absmax is 100 (the maximum Fa), PA is the permeated amount during a 2-h experiment (% of dose/2 h), PA50 is the predicted in vitro permeated amount giving half-maximum Fa, and c is Hill’s coefficient. Based on our previous findings [16], 0.334 and 0.883 were substituted for PA50 and c, respectively. 2.5. Statistical analysis All data are presented as means with the standard deviation (s.d.) for individual groups. Significance was assessed with the unpaired Student’s t-test and p values of 0.05 or less were considered significant. 3. Results 3.1. Dissolution of various drugs under the gastric condition The dissolution of various drugs in the small vessel, reflecting the gastric condition, was monitored for 1 h (Fig. 2). Regarding the basic drugs, ketoconazole and fluconazole, both immediately dissolved in SGF after being added to the vessel in their powder forms due to the acidic conditions present (pH = 1.2) (Fig. 2a). On the other hand, when dantrolene sodium and diclofenac sodium, which are acidic drugs, were added in their powder forms to SGF, the dissolved amounts at 5 min were low at 3.52 ± 0.32% of the dose of diclofenac applied and 0.881 ± 0.080% of the dose of dantrolene applied, and then decreased further (Fig. 2b). These results indicate that both drugs immediately precipitated in gastric fluid after their application.

Table 1 Detection conditions of various drugs by MS/MS. Drugs

Ion mode

Cone voltage (V)

Collision energy (eV)

Precursor ion (m/z)

Product ion (m/z)

Albendazole Dantrolene Diclofenac Dipyridamole Fluconazole Ketoconazole

+ + + + + +

50 30 30 30 30 30

20 48 20 48 24 42

266.09 315.11 296.92 505.33 307.28 532.64

234.17 113.89 216.01 385.48 169.01 81.93

106

M. Kataoka et al. / European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111

(b) Acidic drugs (saltform)

100

100

6

80

80

4

60

2

% of the dose

% of the dose

(a) Basic drugs

60 40 20

40

0 0.0

20

0 0.0

0.5

0

1.0

0.0

0.5

1.0

0.5

Time (h)

1.0

Time (h)

Fig. 2. Time profiles of the dissolution of basic drugs (a) and acidic drugs (salt forms) (b) in the gastric dissolution phase after their application. The basic drugs such as ketoconazole (d) and dipyridamole (N) were applied to the gastric dissolution phase at one-twenty-fifth of the clinical dose. The acidic drugs such as diclofenac sodium (s) and dantrolene sodium (M) were applied to the gastric dissolution phase at one-twenty-fifth of the clinical dose. Data are expressed as the mean ± s.d. of three independent experiments.

3.2. Dissolved and permeated amounts of various basic drugs in simulated intestinal fluid with or without the gastric dissolution process Drugs were applied to the D/P system as a solid form (without the gastric dissolution process) or solution of SGF obtained from dissolution study with the gastric compartment (with the gastric dissolution process). The time profiles of dissolved and permeated drugs were consecutively monitored for 2 h. In all experiments, no significant decreases were observed in the TEER value during the experiments (data not shown). The time profiles of dissolved basic drugs in the simulated fluid (FaSSIFmod) with or without the gastric dissolution process were determined (Fig. 3). In the case of reflecting gastric conditions, dipyridamole and ketoconazole achieved complete dissolution after the application of each solution to gastric conditions for 30 min, and the dissolved amounts of both drugs gradually

decreased with time. These phenomena indicated that the apical solution in the early period formed a supersaturable state, and dipyridamole and ketoconazole then precipitated with time. In the present study, these observations were defined as the supersaturation and precipitation of drugs, although the characterization of each precipitant has not been performed. When both drugs were directly applied in their solid forms to the apical side of the D/P system (without the gastric dissolution process), each dissolution profile was significantly lower than that obtained with the gastric dissolution process. In the case of albendazole, the time profile of the dissolved amount with the gastric phase was maintained at the same amount during the experiment. However, the dissolved amount with the gastric phase was significantly higher than that without the gastric phase. No significant differences were observed in the time profile of dissolved fluconazole with and without the gastric dissolution process because fluconazole mostly dissolved in FaSSIFmod, irrespective of the gastric dissolution process.

Albendazole

5.0

100

% of the dose

4.0

% of the dose

Dipyridamole

120

3.0 2.0 1.0

80 60 40 20

0.0

0 0.0

0.5

1.0

1.5

2.0

0.0

0.5

Time (h)

Fluconazole

100

1.5

2.0

Ketoconazole

120 100

% of the dose

80

% of the dose

1.0

Time (h)

60 40 20

80 60 40 20

0

0 0.0

0.5

1.0

Time (h)

1.5

2.0

0.0

0.5

1.0

1.5

2.0

Time (h)

Fig. 3. Time profiles of the dissolved amount of basic drugs (albendazole, dipyridamole, fluconazole, and ketoconazole) in the apical side of the D/P system with (d) or without (s) gastric dissolution. Data are expressed as the mean ± s.d. of three independent experiments.

107

M. Kataoka et al. / European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111

No significant differences were noted in the permeated amount of fluconazole with and without the gastric dissolution process, which clearly reflected the dissolution profile in FaSSIFmod (Fig. 4). The permeation of albendazole, dipyridamole, and ketoconazole was significantly enhanced under gastric conditions, reflecting the dissolution profiles in the apical compartment of the D/P system. The permeated amounts of albendazole, dipyridamole, and ketoconazole at 2 h with the gastric dissolution process were approximately 3-, 8-, and 8-fold higher, respectively, than those without the gastric dissolution process (Fig. 4). 3.3. Dissolved and permeated amounts of acidic drugs (salt form) in simulated intestinal fluid with or without the gastric dissolution process Similar experiments to those described in Section 3.2. were also performed with dantrolene sodium and diclofenac sodium, the salt forms of acidic drugs (Figs. 5 and 6). When dantrolene sodium and diclofenac sodium were applied to the D/P system after dissolving in SGF for 30 min, dantrolene and diclofenac slowly dissolved without a supersaturated state (Fig. 5). Regarding dissolution profiles without the gastric dissolution process, the dissolution of dantrolene after its direct application to the apical side of the D/P system as a sodium salt exhibited a supersaturated dissolution profile; it dissolved rapidly in the early period, and its dissolution then gradually decreased with time. On the other hand, diclofenac immediately dissolved after being added to FaSSIFmod and its complete dissolution was maintained during the experiment. The permeation of both acidic drugs was significantly reduced by reflecting gastric conditions (Fig. 6). At the end of the experiment, the permeated amounts of dantrolene with and without the gastric dissolution process were 0.275 ± 0.046% and 0.645 ± 0.166% of the doses applied, respectively. Although the permeated amount of diclofenac (12.9 ± 1.2% of dose/2 h) without gastric dissolution was approximately one half than that without gastric dissolution (22.8 ± 2.7% of dose/2 h), similar to dantrolene,

Permeation data (% of dose/2 h) were substituted for PA in Eq. (1) in order to predict the in vivo oral absorption (Fa%) of each drug under fasting conditions (refer to Section 2.4.). Prediction results are summarized in Table 2. The oral absorption of albendazole was predicted to be 28% and 13% in fasting healthy subjects and fasting subjects with high gastric pH (treated with a H2 blocker), respectively, from their permeated amounts (0.115 ± 0.009% of the dose/2 h and 0.0398 ± 0.0042% of the dose/2 h). These results suggested that an increase in gastric pH resulted in a decrease in the oral absorption of albendazole (from 28% to 13%). Similar decreases in oral absorption were predicted for the basic drugs, dipyridamole and ketoconazole, but not for fluconazole. The oral absorption of fluconazole was predicted to be more than 90%, irrespective of pH changes in the gastric fluid. In contrast, the oral absorption of dantrolene from sodium salt in fasted subjects with high gastric pH (64%) was predicted to be higher than that in healthy subjects (46%). The oral absorption of diclofenac was predicted to be complete, irrespective of the gastric pH, although the permeated amount was significantly affected. 4. Discussion In the present study, we focused on confirming the advantages of predicting the effects of gastric pH on the oral absorption of drugs using the D/P system combined with the gastric phase (Fig. 1). The conditions of the gastric phase need to be set in consideration of physiological conditions and previously reported conditions for the D/P system. Since the relationship of volume

Dipyridamole

1.0 0.8

0.15

% of the dose

% of the dose

3.4. Prediction of oral absorption from in vitro data

Albendazole

0.20

0.10 0.05 0.00 0.0

the absolute amount of permeation was significantly high irrespective of the experimental conditions. In all experiments, no significant decreases were observed in the TEER value during the experiments (data not shown).

0.6 0.4 0.2 0.0

0.5

1.0

1.5

2.0

0.0

0.5

Time (h)

Fluconazole

6.0 4.0 2.0 0.0 0.0

1.5

2.0

Ketoconazole

2.0

% of the dose

% of the dose

8.0

1.0

Time (h)

1.5 1.0 0.5 0.0

0.5

1.0

Time (h)

1.5

2.0

0.0

0.5

1.0

1.5

2.0

Time (h)

Fig. 4. Time profiles of the permeation of basic drugs (albendazole, dipyridamole, fluconazole, and ketoconazole) to the basal side of the D/P system with (d) or without (s) gastric dissolution. Data are expressed as the mean ± s.d. of three independent experiments.

M. Kataoka et al. / European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111

100

30

80

20

60

10

40

0

Dantrolene

Diclofenac

120 100

0.0

0.5

1.0

1.5

% of the dose

% of the dose

108

2.0

20

80 60 40 20

0

0 0.0

0.5

1.0

1.5

2.0

0.0

0.5

Time (h)

1.0

1.5

2.0

Time (h)

Fig. 5. Time profiles of the dissolved amount of acidic drugs (dantrolene sodium and diclofenac sodium) in the apical side of the D/P system with (d) or without (s) gastric dissolution. Data are expressed as the mean ± s.d. of three independent experiments.

Dantrolene

1.0

25

% of the dose

0.8

% of the dose

Diclofenac

30

0.6 0.4 0.2

20 15 10 5 0 0.0

0.0 0.0

0.5

1.0

1.5

2.0

Time (h)

0.5

1.0

1.5

2.0

Time (h)

Fig. 6. Time profiles of the permeation of acidic drugs (dantrolene sodium and diclofenac sodium) to the basal side of the D/P system with (d) or without (s) gastric dissolution. Data are expressed as the mean ± s.d. of three independent experiments.

Table 2 Effects of the gastric dissolution phase on the predicted absorption of various drugs.

a b c d *

Drugs

Polarity

BCS classa

Clinical dose (mg)

Amount applied to the D/P system (mg)

Gastric dissolution phaseb

Permeated amount (% of the dose applied/2 h)

Predicted absorptionc (%)

Absorption ratiod (GD/GD+)

Albendazole

Base

II

400

4

Base

II

100

1

Fluconazole

Base

I

200

2

Ketoconazole

Base

II

200

2

Dantrolene Na

Acid (salt)

II or IV

50

0.5

Diclofenac Na

Acid (salt)

II

25

0.25

0.0398 ± 0.0042 0.115 ± 0.009* 0.0879 ± 0.0116 0.730 ± 0.101* 4.52 ± 0.39 5.55 ± 0.86 0.220 ± 0.025 1.71 ± 0.14* 0.645 ± 0.166 0.275 ± 0.046* 22.8 ± 2.7 12.9 ± 1.2*

13 28 24 67 91 92 41 81 64 46 98 96

0.5

Dipyridamole

 +  +  +  +  +  +

0.4 1.0 0.5 1.4 1.0

See the text. Drug permeation was assessed by the D/P system with (+) or without () the gastric dissolution phase. Predicted by Eq. (1). Ratio between predicted absorption from the permeated amount in the D/P system with (GD+) and without (GD) the gastric phase. Significant difference from the permeated amount without the gastric dissolution phase (p < 0.05).

between the intestine and apical compartment of the D/P system was 1/100, the dose applied to the D/P system was fixed at 1% of the clinical dose [15]. The volume of gastric fluid was previously reported to be approximately 50 mL under fasted conditions [33,24]. Assuming that the intake volume of water to take medicine is 250 mL, the volume of gastric fluid immediately increases to 300 mL after the intake of medicine with water. The half-life of the gastric emptying of an ingested solution was previously reported to be approximately 4–13 min [29,33,8]. The effective volume of gastric fluid was assumed to be 150 mL due to the

volume of and rapid emptying from the stomach. Therefore, the fluid volume of the gastric phase was set to 6 mL (one twenty-fifth of 150 mL) and the amount of the drug applied was set to one twenty-fifth of the clinical dose. Although the agitation speed of SGF may be important, the stirring rate of the gastric phase was set to 200 rpm, similar to the D/P system, due to the first attempt to combine the gastric phase with the D/P system. Various drugs with different physicochemical properties were used in the present study (Table 2). Briefly, fluconazole and four drugs (albendazole, dipyridamole, ketoconazole, and diclofenac

M. Kataoka et al. / European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111

sodium) were classified as classes I and II, respectively, in accordance with the Biopharmaceutics Classification System (BCS) [4,9]. Due to oral absorption (70%) and poor water solubility [22], dantrolene sodium may be classified as class II or IV. The gastric dissolution profiles of ketoconazole and fluconazole were determined; both basic drugs immediately dissolved after being added to SGF. In the case of the acidic drugs, dantrolene and diclofenac, both showed the highest dissolved amounts at the first time point measured (5 min), and their dissolution subsequently decreased with time, indicating that acidic drugs as salt forms rapidly dissolve in SGF, but immediately precipitate due to their physicochemical properties. Therefore, the intestinal dissolution processes of drugs following their oral administration may be strongly affected by gastric dissolution. Since the pH of gastric fluid is known to be significantly lower than that of intestinal fluid, the solubility of basic drugs is markedly higher in the former than in the latter. Therefore, poorly water-soluble basic drugs in intestinal fluid may dissolve in the stomach after their oral administration and then precipitate in the intestinal tract, a process that may induce a supersaturated state of such drugs in the intestinal fluid, resulting in high Fa regardless of poor solubility. Previous studies demonstrated that the systemic exposure of poorly water-soluble basic drugs such as albendazole, dipyridamole, and ketoconazole was significantly higher in fasted healthy volunteers than in fasted human subjects with gastric anacidity or hypoacidity or in those being treated with antacids [27,1,28]. The gastric pH of subjects treated with H2 blockers was previously reported to range between 6 and 7 [23,26]. Sugano [30] calculated the Fa of poorly water-soluble basic drugs in subjects with normal and high gastric pH on the basis of a computational oral absorption simulation theory. In our in vitro system, dissolved concentration of dipyridamole and ketoconazole in FaSSIFmod in the early period was higher than that in the later period (Fig. 3). Providing that dissolved drug in the apical side can only permeate to the basal side through the Caco-2 monolayer, the permeation rate will depend on the concentration of dissolved drug at a certain time (dissolution and precipitation of drugs). The rate of permeation of both drugs from 0.25 to 0.50 h (dipridamole; 0.541 ± 0.118% of dose/h, ketoconazole; 1.62 ± 0.20% of dose/h) was significantly faster than that from 1.5 to 2.0 h (dipridamole; 0.342 ± 0.027% of dose/h, ketoconazole; 0.406 ± 0.072% of dose/h) (Fig. 4). These results implied that supersaturation of such drugs occurred in the apical compartment of the D/P system. The advantages of predicting oral absorption in humans with the D/P system have already been reported [16,17,18,19,3]. In previous studies, oral absorption was successfully evaluated based on the relationship between oral absorption and the permeated amount in the D/P system. This relationship was obtained from 13 drugs with various physicochemical properties [16]. Therefore, the oral absorption of various drugs was predicted with Eq. (1) in accordance with previous studies. The oral absorption of albendazole, dipyridamole, and ketoconazole in fasted humans with normal gastric acidity was predicted to be 28%, 67%, and 81%, respectively (Table 2). In contrast, the oral absorption of these drugs in fasted humans with elevated gastric pH was predicted to be approximately half than that in normal subjects. The absorption of fluconazole was not affected by pH changes in gastric fluid because oral absorption was predicted to be high (>90%) with and without the gastric dissolution process in the D/P system. The oral absorption of albendazole in humans at a dose of 20 mg/kg was previously shown to be reduced to 70% by a pretreatment with cimetidine [28]. In another study using rabbits at a dose of 5 mg/kg, which corresponded to the typically recommended dose of albendazole (400 mg), oral absorption was reported to be threefold lower with elevated pH than that with low pH [21]. The effects of gastric pH on the oral absorption of dipyridamole have

109

been clinically reported at doses of 50 mg and 200 mg. Russell et al. [27] demonstrated that the AUC following the administration of 50 mg of dipyridamole to humans with elevated gastric pH was decreased to 63% by a pretreatment with famotidine. Furthermore, the AUC from 0 to 12 h after the oral administration of buffered dipyridamole with tartaric acid (200 mg) to subjects with elevated gastric pH was 2-fold higher than that when 200 mg of dipyridamole was administered in its typical tablet form to subjects with the same gastric condition [6]. These clinical findings suggested that the absorption of dipyridamole was significantly affected by the dissolution process in the stomach at a dose ranging between 50 mg and 200 mg. The absorption of ketoconazole (200 mg) in human subjects with elevated gastric pH administered a H2 blocker was markedly lower (approximately one-tenth) than that in those with normal gastric pH; however, this change in absorption was not observed for fluconazole (200 mg) [1]. The pharmacokinetics of ketoconazole are known to be strongly affected by first-pass metabolism via CYP3A. Higashino et al. [9] revealed that the intestinal first-pass metabolism of ketoconazole in rats was significantly inhibited by a supersaturated solution of itself, indicating nonlinear pharmacokinetics. These findings suggest that similar pharmacokinetics exist in humans. The results obtained in the present study on the effects of gastric pH on the oral absorption of four basic drugs were consistent with in vivo findings. When acidic drugs were administered in their free forms, it was not possible to anticipate the effects of gastric pH on the oral absorption of these drugs because of their physicochemical properties based on the Henderson–Hasselbalch equation and physiological pH range. Therefore, in order to confirm the effects of gastric dissolution on the oral absorption of poorly water-soluble drugs, dantrolene sodium and diclofenac sodium were used as the salt forms of acidic drugs. In the gastric dissolution process, the dissolved amounts of both drugs with slight supersaturation after their addition to the gastric vessel were 0.596–2.38% of the applied dose at 1 h, respectively, indicating that these drugs immediately precipitated as their free forms once applied due to acidic conditions (Fig. 1b). After the addition of gastric fluid with drugs to the apical side of the D/P system, both drugs gradually dissolved in FaSSIFmod with time (Fig. 5). When dantrolene and diclofenac were directly added to the D/P system in their salt forms, supersaturation and rapid precipitation were observed for dantrolene, but not for diclofenac (Fig. 5). Due to differences in the dissolution processes, the permeation of both drugs applied to the apical side in their salt forms was significantly enhanced (Fig. 6). The oral absorption of dantrolene with or without the gastric dissolution process was predicted to be 46% and 64%, respectively, based on the permeated amount in 2 h. It was possible to predict that the enhancement in the dissolution rates with the salt forms was diminished by exposure to acidic conditions, which led to decreases in the oral absorption of poorly water-soluble acidic drugs after their administration as salt forms. Kambayashi and Dressman [14] reported an in vitro–in silico–in vivo approach that considered the dissolution and precipitation of drugs in order to predict the oral pharmacokinetic profile of dantrolene sodium. In that study, the precipitation of dantrolene was also observed in fasted state SGF (FaSSGF) and FaSSIF following the addition of the commercial product of its sodium salt. Therefore, the gastric dissolution process may be important not only for basic drugs, but also for the salt forms of acidic drugs. Although significant differences were observed in the permeated diclofenac amounts with or without the gastric dissolution process, its oral absorption was predicted to be very high under both conditions (98% without and 96% with the gastric dissolution process). A bioequivalence study in healthy volunteers with normal and decreased gastric acidity has been performed for four enteric-coated diclofenac sodium products [31]. The time for the first appearance in the

110

M. Kataoka et al. / European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111

plasma of one product was significantly faster than those of the other products and was also faster than the usual gastric residence time, irrespective of gastric acidity, suggesting that this product disintegrated in the stomach even though it was coated with enteric material. However, no significant differences were noted in the AUC or Cmax between normal and decreased gastric acidity. These clinical findings indicated that the oral absorption of diclofenac sodium was not affected by the gastric dissolution process and supported our estimation. However, transit times and fluid properties in the stomach need to be considered for an accurate evaluation, not only for basic drugs, but also for the salt forms of acidic drugs. Therefore, the exposure times of drugs and gastric conditions need to be established in order to perform highly accurate predictions of oral absorption. Previous studies estimated gastric half-life to be between 4 and 13 min [29,33,24]. These estimated fast transfers are shorter than the exposure time (30 min) of drugs under the gastric conditions used in this study; however, exposure times under gastric conditions may not affect the dissolution profile in the apical side of the D/P system because no significant changes were observed in the dissolved amounts of basic drugs in the gastric vessel over a period of 1 h, but not for acidic salts. Since the dissolved amounts of both drugs in the apical side of the D/P system were significantly greater than those in the gastric vessel, the precipitation of drugs in the gastric vessel from 5 to 30 min did not affect subsequent dissolution under small intestinal conditions. FaSSGF (pH 1.6) has been recommended by Jantratid et al. [13] for estimating the intragastric dissolution of drugs. In addition, the stirring rate of the gastric vessel (200 rpm in this study) should be carefully fixed to mimic agitation of gastric fluid in human stomach. Therefore, in order to evaluate the absorption of drugs from formulations and preformulations, parameters related to gastric emptying and dissolution, such as the transit time, pH, and agitation of gastric fluid, in fasted humans need to be reflected in the D/P system with the gastric dissolution process. 5. Conclusion The gastric dissolution process significantly affected dissolved amount of drugs in the apical medium and permeation to the basal side of the D/P system, which led to changes in the prediction of the oral absorption of poorly water-soluble drugs. The effects of gastric pH on oral absorption in humans, particularly basic drugs, may be assessed by the D/P system with the gastric dissolution process. Therefore, the D/P system with the gastric dissolution process has the advantage of better predicting the oral absorption of poorly water-soluble basic drugs. In addition, the effects of gastric pH on the oral absorption of poorly water-soluble drugs may be evaluated from the D/P system with and without the gastric phase. However, experimental conditions such as the pH of gastric fluid and time of exposure to gastric fluid need to be carefully fixed in order to perform accurate predictions of drug absorption and will be a subject for future studies. References [1] R.A. Blum, D.T. D’Andrea, B.M. Florentino, J.H. Wilton, D.M. Hilligoss, M.J. Gardner, E.B. Henry, H. Goldstein, J.J. Schentag, Increased gastric pH and the bioavailability of fluconazole and ketoconazole, Ann. Intern. Med. 114 (1991) 755–757. [2] E. Borbás, A. Balogh, K. Bocz, J. Müller, É. Kiserdei, T. Vigh, B. Sinkó, A. Marosi, A. Halász, Z. Dohányos, L. Szente, G.T. Balogh, Z.K. Nagy, In vitro dissolution– permeation evaluation of an electrospun cyclodextrin-based formulation of aripiprazole using lFluxTM, Int. J. Pharm. 491 (2015) 180–189. [3] P. Buch, P. Langguth, M. Kataoka, S. Yamashita, IVIVC in oral absorption for fenofibrate immediate release tablets using a dissolution/permeation system, J. Pharm. Sci. 98 (2009) 2001–2009.

[4] B. Chuasuwan, V. Binjesoh, J.E. Polli, H. Zhang, G.L. Amidon, H.E. Junginger, K.K. Midha, V.P. Shah, S. Stavchansky, J.B. Dressman, D.M. Barends, Biowaiver monographs for immediate release solid oral dosage forms: diclofenac sodium and diclofenac potassium, J. Pharm. Sci. 98 (2009) 1206–1219. [5] A. Dahan, A. Hoffman, The effect of different lipid based formulations on the oral absorption of lipophilic drugs: the ability of in vitro lipolysis and consecutive ex vivo intestinal permeability data to predict in vivo bioavailability in rats, Eur. J. Pharm. Biopharm. 67 (2007) 96–105. [6] H. Derendorf, C.P. VanderMaelen, R.S. Brickl, T.R. MacGregor, W. Eisert, Dipyridamole bioavailability in subjects with reduced gastric acidity, J. Clin. Pharmacol. 45 (2005) 845–850. [7] M.J. Ginski, J.E. Polli, Prediction of dissolution–absorption relationships from a dissolution/Caco-2 system, Int. J. Pharm. 177 (1999) 117–125. [8] B. Hens, J. Brouwers, B. Anneveld, M. Corsetti, M. Symillides, M. Vertzoni, C. Reppas, D.B. Turner, P. Augustijns, Gastrointestinal transfer: in vivo evaluation and implementation in in vitro and in silico predictive tools, Eur. J. Pharm. Sci. 63 (2014) 233–242. [9] H. Higashino, T. Hasegawa, M. Yamamoto, R. Matsui, Y. Masaoka, M. Kataoka, S. Sakuma, S. Yamashita, In vitro–in vivo correlation of the effect of supersaturation on the intestinal absorption of BCS Class 2 drugs, Mol. Pharm. 11 (2014) 746–754. [10] J. Hirtz, The gastrointestinal absorption of drugs in man: a review of current concepts and methods of investigation, Br. J. Clin. Pharmacol. 19 (Suppl 2) (1985) 77S–83S. [11] F. Ingels, S. Deferme, E. Destexhe, M. Oth, G. Van den Mooter, P. Augustijns, Simulated intestinal fluid as transport medium in the Caco-2 cell culture model, Int. J. Pharm. 232 (2002) 183–192. [12] F. Ingels, B. Beck, M. Oth, P. Augustijns, Effect of simulated intestinal fluid on drug permeability estimation across Caco-2 monolayers, Int. J. Pharm. 274 (2004) 221–232. [13] E. Jantratid, N. Janssen, C. Reppas, J.B. Dressman, Dissolution media simulating conditions in the proximal human gastrointestinal tract: an update, Pharm. Res. 25 (2008) 1663–1676. [14] A. Kambayashi, J.B. Dressman, An in vitro–in silico–in vivo approach to predicting the oral pharmacokinetic profile of salts of weak acids: case example dantrolene, Eur. J. Pharm. Biopharm. 84 (2013) 200–207. [15] M. Kataoka, Y. Masaoka, Y. Yamazaki, T. Sakane, H. Sezaki, S. Yamashita, In vitro system to evaluate oral absorption of poorly water-soluble drugs: simultaneous analysis on dissolution and permeation of drugs, Pharm. Res. 20 (2003) 1674–1680. [16] M. Kataoka, Y. Masaoka, S. Sakuma, S. Yamashita, Effect of food intake on the oral absorption of poorly water-soluble drugs: in vitro assessment of drug dissolution and permeation assay system, J. Pharm. Sci. 95 (2006) 2051–2061. [17] M. Kataoka, S. Itsubata, Y. Masaoka, S. Sakuma, S. Yamashita, In vitro dissolution/permeation system to predict the oral absorption of poorly water-soluble drugs: effect of food and dose strength on it, Biol. Pharm. Bull. 34 (2011) 401–407. [18] M. Kataoka, K. Sugano, C. da Costa Mathews, J.W. Wong, K.L. Jones, Y. Masaoka, S. Sakuma, S. Yamashita, Application of dissolution/permeation system for evaluation of formulation effect on oral absorption of poorly water-soluble drugs in drug development, Pharm. Res. 29 (2012) 485–1494. [19] M. Kataoka, K. Yano, Y. Hamatsu, Y. Masaoka, S. Sakuma, S. Yamashita, Assessment of absorption potential of poorly water-soluble drugs by using the dissolution/permeation system, Eur. J. Pharm. Biopharm. 85 (2013) 1317– 1324. [20] M. Kobayashi, N. Sada, M. Sugawara, K. Iseki, K. Miyazaki, Development of a new system for prediction of drug absorption that takes into account drug dissolution and pH change in the gastro-intestinal tract, Int. J. Pharm. 221 (2001) 87–94. [21] 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. [22] T. Krause, M.U. Gerbershagen, M. Fiege, R. Weisshorn, F. Wappler, Dantrolene – a review of its pharmacology, therapeutic use and new developments, Anaesthesia 59 (2004) 364–373. [23] P. Lelawongs, J.A. Barone, J.L. Colaizzi, A.T. Hsuan, W. Mechlinski, R. Legendre, J. Guarnieri, Effect of food and gastric acidity on absorption of orally administered ketoconazole, Clin. Pharm. 7 (1988) 228–235. [24] D.M. Mudie, K. Murray, C.L. Hoad, S.E. Pritchard, M.C. Garnett, G.L. Amidon, P.A. Gowland, R.C. Spiller, G.E. Amidon, L. Marciani, Quantification of gastrointestinal liquid volumes and distribution following a 240 mL dose of water in the fasted state, Mol. Pharm. 11 (2014) 3039–3047. [25] S.A. Motz, U.F. Schaefer, S. Balbach, T. Eichinger, C.M. Lehr, Permeability assessment for solid oral drug formulations based on Caco-2 monolayer in combination with a flow through dissolution cell, Eur. J. Pharm. Biopharm. 66 (2007) 286–295. [26] C.U. Nwokolo, A. Sawyerr, J.T. Smith, R.E. Pounder, Intravenous pentagastrin can induce the illusion of ’tolerance’ to a single dose of an H2-blocker in man, Aliment. Pharmacol. Ther. Suppl. 1 (1990) 75–83. [27] T.L. Russell, R.R. Berardi, J.L. Barnett, T.L. O’Sullivan, J.G. Wagner, J.B. Dressman, PH-related changes in the absorption of dipyridamole in the elderly, Pharm. Res. 11 (1994) 136–143. [28] H.G. Schipper, R.P. Koopmans, J. Nagy, J.J. Butter, P.A. Kager, C.J. Van Boxtel, Effect of dose increase or cimetidine co-administration on albendazole bioavailability, Am. J. Trop. Med. Hyg. 63 (2000) 270–273.

M. Kataoka et al. / European Journal of Pharmaceutics and Biopharmaceutics 101 (2016) 103–111 [29] T. Shingaki, T. Takashima, Y. Wada, M. Tanaka, M. Kataoka, A. Ishii, Y. Shigihara, Y. Sugiyama, S. Yamashita, Y. Watanabe, Imaging of gastrointestinal absorption and biodistribution of an orally administered probe using positron emission tomography in humans, Clin. Pharmacol. Ther. 91 (2012) 653–659. [30] K. Sugano, Computational oral absorption simulation of free base drugs, Int. J. Pharm. 398 (2010) 73–82. [31] M.E. Van Gelderen, M. Olling, D.M. Barends, J. Meulenbelt, P. Salomons, A.G. Rauws, The bioavailability of diclofenac from enteric coated products in

111

healthy volunteers with normal and artificially decreased gastric acidity, Biopharm. Drug Dispos. 15 (1994) 775–788. [32] P.G. Welling, Effects of food on drug absorption, Pharmacol. Ther. 43 (1989) 425–441. [33] S. Yamashita, M. Kataoka, H. Higashino, S. Sakuma, T. Sakamoto, H. Uchimaru, H. Tsukikawa, M. Shiramoto, H. Uchiyama, H. Tachiki, S. Irie, Measurement of drug concentration in the stomach after intragastric administration of drug solution to healthy volunteers: analysis of intragastric fluid dynamics and drug absorption, Pharm. Res. 30 (2013) 951–958.