- Email: [email protected]
The Buccal Mucosa as an Alternative Route for the Systemic Delivery of Risperidone
The Buccal Mucosa as an Alternative Route for the Systemic Delivery of Risperidone LARS B. HEEMSTRA, BARRIE C. FINNIN, JOSEPH A. NICOLAZZO Drug Delivery, Disposition and Dynamics, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, Victoria 3052, Australia Received 30 November 2009; revised 11 February 2010; accepted 15 March 2010 Published online 21 April 2010 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.22175 ABSTRACT: The purpose of this study was to investigate the potential of the buccal mucosa for the systemic delivery of risperidone (RISP), and to determine the impact of Azone1 (AZ) on the transport of RISP via this route. The permeability of RISP through porcine buccal mucosa was assessed in modified Ussing chambers at various concentrations to determine the mechanisms involved in transport across the tissue. The effect of AZ was assessed by administering AZ 5% (w/ w) to the tissue as a pretreatment or together with RISP in solution or in a mucoadhesive gel formulation. RISP permeated the buccal mucosa via a passive diffusion mechanism and pretreatment or coadministration of AZ 5% did not significantly modify the permeation of RISP. Application of a RISP mucoadhesive gel resulted in a steady state flux of 64.65 8.0 mg/ cm2/h, which when extrapolated to the in vivo setting, is predicted to result in RISP plasma concentrations of 11.2–56.1 mg/L for mucosal application areas between 2 and 10 cm2. Given that these predicted concentrations are within the therapeutic range of RISP required in humans, delivery of RISP via the buccal mucosa has the potential to result in therapeutically relevant plasma concentrations for the treatment of schizophrenia. ß 2010 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci99:4584–4592, 2010
Keywords: absorption enhancer; Azone1; buccal mucosa; epithelial permeability; mucosal delivery; passive diffusion; risperidone
INTRODUCTION Risperidone (RISP) is an atypical antipsychotic which has been used for the treatment of schizophrenia for many years. RISP is available on the market as both a tablet and oral solution, however, noncompliance is a major issue associated with antipsychotic treatment when administered via the oral route.1,2 In fact, a comprehensive review of five randomized studies demonstrated that there was an enhanced risk of relapse in schizophrenic patients who received intermittent treatment compared to patients who received continuous treatment.3 With this in mind, a long-acting injectable formulation of RISP has been developed (Risperdal Consta1), which has several pharmacokinetic advantages over the use of oral RISP, such as reduced variability in plasma concentrations due to intestinal absorption and first-pass metabolism, and a subsequent reduction in the Correspondence to: Joseph A. Nicolazzo (Telephone: þ61-39903-9605; Fax: þ61-3-9903-9583; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 99, 4584–4592 (2010) ß 2010 Wiley-Liss, Inc. and the American Pharmacists Association
4584
frequency of adverse effects. However, this longacting injectable possesses some disadvantages, including the inability to rapidly discontinue treatment in the case of unexpected side effects and its invasive nature. The need for appropriate delivery mechanisms for these beneficial antipsychotics is therefore becoming an area of research with much focus. For example, very recent studies have highlighted the nasal and transdermal routes as alternative sites for systemically delivering RISP and other antipsychotics,4,5 and it is likely that other nonconventional routes may also prove clinically relevant for this class of compounds. One such route that is often under-represented as an alternative for systemic drug delivery is the buccal mucosa. Delivery of drugs through the buccal mucosa has several advantages over the traditional oral route. Since blood flow from the oral mucosa drains directly into the jugular vein, drugs administered through the buccal mucosa gain direct access to the systemic circulation thereby avoiding hepatic first pass metabolism.6,7 Furthermore, the rate of absorption is not influenced by food or gastric emptying rate.6,7 In addition, the buccal mucosa seems well suited for the use of retentive systems, such as
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
BUCCAL DELIVERY OF RISPERIDONE
mucoadhesive tablets or patch systems. This would provide sustained delivery of drug to the bloodstream without the invasive nature of long-acting injectables,7 which is of extreme benefit to this class of drugs given the improved compliance associated with long-acting antipsychotics.8 This route of delivery is currently used in clinical practice with formulations available for insulin (Oral-LynTM),9 testosterone (Striant1),10 and prochlorperazine (Buccastem1).11 Therefore, there is significant potential to use the buccal route for systemically delivering RISP in a controlled manner, with the ultimate aim of improving patient compliance. While the mucosae of the oral cavity are generally considered more permeable than the skin,12 the buccal mucosa still acts as a barrier and may limit the permeation of a drug administered transmucosally. In order to improve drug delivery via this route, the tissue may be treated with penetration enhancers such as surfactants, fatty acids and alcohols.12 One agent which has been used in our laboratory for modifying buccal penetration of drugs is Azone1 (AZ), and the general mechanism by which drug delivery is affected with this chemical is through altered partitioning of the coadministered drug.13–15 However, the impact of AZ on buccal mucosal delivery depends on the permeant of interest. AZ appears to have no effect on the permeability of the hydrophilic caffeine (CAF) whereas it enhances the permeability and tissue retention of the more lipophilic triamcinolone acetonide (TAC).13,15 Of greater significance to controlled drug delivery is the effect of AZ on estradiol (E2) permeability across the buccal mucosa—pretreatment with AZ significantly reduced the flux of E2 by 67.8% with a 2.2-fold enhancement in the tissue concentrations of E2.14 This suggested that AZ had the potential to enhance the reservoir capacity of the buccal mucosa, leading to a slower release of E2 into the receptor compartment over a prolonged period of time. Such a technology would be of extreme benefit for the delivery of RISP through the buccal mucosa, as this may result in a slower release of RISP into the systemic circulation, resulting in a controlled delivery of this antipsychotic. The purpose of this study, therefore, was to investigate the potential of the buccal mucosa for systemic delivery of RISP using an in vitro model. Due to the limited availability of human buccal tissue, porcine buccal mucosa was used as the model tissue. Porcine buccal mucosa has been shown to exhibit similar morphology and permeability characteristics to human buccal mucosa,16,17 and has been used by many laboratories as a suitable model for assessment of buccal permeability characteristics.18–22 In addition to elucidating the mechanisms involved in the transport of RISP across the buccal mucosa, the potential of AZ to alter the permeability of this DOI 10.1002/jps
4585
compound was investigated. Furthermore, formulation approaches to increase the delivery of RISP through the buccal mucosa were evaluated to determine whether it would be feasible to deliver clinically relevant amounts of RISP to humans via this alternative route.
MATERIALS AND METHODS Materials RISP was kindly donated by Acrux Limited (Victoria, Australia) and was manufactured by Alchymars (Alathur, Tamil Nadu, India). AZ was obtained from Yick-Vic Chemicals and Pharmaceuticals (HK) Ltd. (Kwun Tong, Hong Kong). Ethanol (EtOH) 95% (v/v) was obtained from CSR Distilleries (Victoria, Australia) and acetonitrile (ACN) was obtained from Merck (Darmstadt, Germany) and was of HPLC grade. Poloxamer 407 was obtained from BASF (Ludwigshafen, Germany) and Carbopol 974 was obtained from Bronson & Jacobs (New South Wales, Australia). Krebs bicarbonate Ringer (KBR) buffer was prepared as detailed previously,23 and adjusted to physiological pH 7.4 with carbogen (95% O2 þ 5% CO2) bubbling as performed in previous experiments.23 Water was obtained from a Milli-Q water purification system (Millipore, Bedford, MA). All other chemicals were of analytical grade and used as received. Tissue Preparation Cheek tissue from domestic pigs (Sus scrofa domestica) was obtained from a local abattoir immediately after slaughter and transported in ice cold KBR. The buccal epithelium was then carefully separated from underlying submucosa with forceps and surgical scissors. The separated epithelial tissue (approximately 500 mm in thickness) was kept in ice cold KBR and supplied with carbogen bubbling until placed in modified Ussing chambers. Both donor and receptor chambers were filled with 1.5 mL of KBR and the tissue was maintained at 37 8C for 30 min. After this incubation period, permeability experiments were conducted. Assessment of Mechanisms of Transport of RISP To determine if RISP transport across the buccal mucosa was influenced by mechanisms other than passive diffusion, two studies were performed. The first study was a concentration-dependent study to ensure that RISP permeability across the buccal mucosa followed Fickian diffusion, and the second study assessed bidirectional transport to identify whether drug efflux transporters (mainly P-glycoprotein, P-gp) may potentially contribute to the absorption of RISP across the buccal mucosa, if such JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
4586
HEEMSTRA, FINNIN, AND NICOLAZZO
transporters are indeed present in this mucosa. RISP is a known substrate of the drug transporter P-gp,24 and given that P-gp has recently been suggested to be involved in drug absorption in the skin,25 we investigated whether a similar functional activity existed in the buccal mucosa using this probe molecule. Concentration Dependency of RISP Permeation After the 30 min equilibration period, KBR was removed from both chambers. The donor chamber was filled with 1.5 mL of RISP solution (at a concentration of either 2.5, 10, or 20 mg/mL in KBR) and the receptor chamber was filled with KBR. Samples of 200 mL were taken from the receptor chamber at periodic time intervals over 4 h and were replaced with fresh KBR. The results were adjusted to allow for this dilution effect. In addition, 20 mL of donor solution was removed at the start and completion of the experiment. All samples were collected in glass vials, as use of polypropylene vials led to nonspecific adsorption issues. The concentrations of RISP were determined using HPLC (detailed below). A plot comparing the cumulative amount of RISP permeating the buccal tissue over time was constructed and the steady state flux (Jss) was calculated using Eq. (1), where DM is the amount of drug transported during time Dt, and A is the diffusional area (0.64 cm2). Jss ¼
DM ADt
(1)
The apparent permeability coefficient ( PappA) of RISP was then determined by dividing the Jss with the initial donor concentration, and the PappA values obtained from different donor chamber concentrations were compared using a one-way analysis of variance with SPSS for Windows (version 15.0, SPSS, Chicago, IL). Bidirectional Study To delineate between passive diffusion and active efflux at the buccal mucosa, this study involved measuring the permeability of RISP in the directions representing drug transport from the oral cavity to the blood compartment (mucosal to serosal) and from the blood compartment to the oral cavity (serosal to mucosal). In a passive diffusion process, the Jss of RISP is expected to be similar in both directions, whereas if an efflux mechanism was involved, the serosal-to-mucosal PappA should exceed the mucosalto-serosal PappA. A donor solution of 10 mg/mL RISP in KBR was used in these experiments and samples from the receptor chamber were taken and analyzed for RISP concentration by HPLC. PappA was calculated in the serosal-to-mucosal and mucosal-to-serosal directions as detailed above, and an independent samples JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
t-test was used to compare the value between the two groups using SPSS. Effect of AZ Pretreatment Buccal mucosa was mounted in a modified Ussing chamber and excess KBR was removed from the mucosal surface with a Kimwipes1 (KimberleyClark, New South Wales, Australia). A plastic ring was placed over the 0.64 cm2 exposed area, as detailed previously.13 Five microliters of the pretreatment solution (5%, w/w, AZ in 95%, v/v, EtOH or 95%, v/v, EtOH as control) was applied to the exposed area. The solution was allowed 5 min to dry before the chambers were clamped together, filled with 1.5 mL of KBR and allowed to equilibrate with carbogen bubbling for 2 h. After the 2-h pretreatment, both donor and receptor solution were removed and replaced with a solution of RISP (20 mg/mL in KBR) or KBR, respectively. In order to assess the impact of AZ pretreatment on both disappearance of RISP from the donor chamber and appearance of RISP in the receptor chamber, both donor samples (10 mL) and receptor samples (200 mL) were taken over a 4-h period. The same volume of KBR (200 mL) was replaced in the receptor chamber to maintain a constant receptor chamber volume at each postdose time point. Donor and receptor samples were analyzed for RISP concentration by HPLC, and in addition to the PappA, the apparent coefficient of disappearance ( PappD) was also calculated using Eq. (2), PappD ¼
V ku A
(2)
where V represents the volume of the donor chamber, A is the diffusional area and ku is the uptake rate constant (s1) calculated from the negative slope of a ln (RISP mass in donor chamber) versus time plot.26 Given the potential variation in tissue thickness within different areas of the buccal cavity, AZ and EtOH-treated tissues were always taken from the same region of the buccal cavity for comparison. An enhancement ratio (ER) was then calculated with Eq. (3) for the PappA and with Eq. (4) for the PappD. ER ¼
PappA AZ treated tissue PappA EtOH treated tissue
(3)
ER ¼
PappD AZ treated tissue PappD EtOH treated tissue
(4)
Additional experiments were also conducted to determine if a larger amount of AZ (10 mL of a 5%, w/w, solution) resulted in a further enhancement in RISP permeability across the buccal mucosa. In addition to this pretreatment protocol, the impact of coadministration of AZ and RISP on buccal permeation of RISP was examined. Two formulations DOI 10.1002/jps
BUCCAL DELIVERY OF RISPERIDONE
containing 24 mg/mL of RISP in 95% (v/v) EtOH were prepared with one of the formulations containing 5% (w/w) AZ. Ten microliters of each formulation was then applied to the mucosal surface, and after a 5 min period (to allow for EtOH evaporation), the chambers were clamped together and filled with 1.5 mL KBR. The amount of RISP permeating into the receptor chamber over a 4-h period was determined as described above. Permeability of RISP after Application of a Mucoadhesive Gel To evaluate whether delivery of RISP via the buccal mucosa could result in receptor chamber concentrations required for therapeutic efficacy in vivo, a mucoadhesive gel containing a high concentration of RISP was developed and applied to the buccal mucosa. The gel was prepared using the ‘‘cold method.’’27 In a preweighed glass beaker, 4 g of poloxamer 407 was slowly added to 10 g of ice-cold water. The beaker was gently swirled and placed in a refrigerator overnight for complete polymer dissolution. Poloxamers are liquid at room temperature and exhibit low mucoadhesion,28 however, the addition of Carbopol and neutralization to pH 6 causes increased viscosity and bioadhesiveness.29 Therefore, the next day, 200 mg of Carbopol 974 was added to the viscous solution, the pH was adjusted to 6 with 1 M NaOH and the gel was made up to 20 g with water. Once formed, 3 g of the gel was spread out on a glass slab with a metal spatula. To this was slowly added 60 mg of RISP and the gel was triturated to ensure uniform distribution of RISP (which was determined by extraction of RISP from the gel and HPLC analysis). Analysis of the gel formulation by HPLC indicated that the concentration of RISP in the mucoadhesive gel was 24.77 1.08 mg/g (mean SD, n ¼ 5). Freshly separated buccal mucosa was mounted in the modified Ussing chamber. Excess KBR was removed with a Kimwipes1 and 100 mg of the gel formulation was transferred from the glass slide onto the mucosal surface with a metal spatula. The chambers were clamped together and slowly filled with 1.5 mL of KBR in both chambers. Carbogen bubbling was only supplied to the receptor chamber to prevent the gel formulation from detaching from the mucosa in the donor chamber. Samples of 10 mL were taken from the donor chamber at the start and completion of the experiment and receptor samples of 200 mL were taken at periodic time intervals over 4 h and were replaced with fresh KBR. Samples were analyzed by HPLC for RISP concentration and the Jss determined, as described above. Using the in vitrodetermined Jss, the steady state plasma concentration that would be achieved in vivo was calculated according to Eq. (5) (with the assumption that DOI 10.1002/jps
4587
salivary clearance is minimal as the gel would adhere directly to the buccal mucosa): Css ¼
Jss CL
(5)
where Css is the predicted in vivo steady state plasma concentration and CL is the in vivo plasma clearance of RISP in humans (192 mL/min).30 HPLC Analysis Donor and receptor chamber samples were diluted with an equal volume of ACN to precipitate buffer salts. Samples were vortexed and centrifuged for 30 min at 1,989g. After centrifugation, the supernatant was removed and transferred into a clean vial and analyzed with a HPLC system consisting of a Shimadzu LC-20AT pump, SIL-20 AC HT autosampler, CTO-20A column oven and a SPD 20A UV/VIS detector. A symmetry C18 3.9 mm 150 mm (Waters, Milford, MA) column was used for the analysis. The mobile phase consisted of a mixture of a 22 mM potassium phosphate buffer at pH 7.6 and 38% (v/v) ACN and the flow rate was set at 1 mL/min. The detector and the injection volume were 280 nm and 100 mL, respectively. Peak areas were calculated with Shimadzu LCSolution v. 1.24 SP1 software, and were related to RISP concentration after weighting 1/x2, where x is the concentration of RISP. Intra-day precision and accuracy were determined by analysis of six independently prepared RISP solutions at low, medium, and high concentrations, resulting in values of 0.5–7.2% and 101.8–102.8%, respectively. Interday precision was determined by analysis of these solutions on three different days (values ranged from 0.8% to 1.5%). In addition, KBR which had been exposed to buccal tissue for 4 h did not interfere with the HPLC chromatograms. Extraction efficiency was determined by comparing the peak areas of RISP in fresh KBR and KBR that had been exposed to buccal tissue, and this resulted in values of 99.6–102.7%.
RESULTS Concentration-Dependent Permeability of RISP The cumulative amount of RISP that permeated the buccal mucosa over a 4-h experimental period is shown in Figure 1. After a lag time of 1–1.5 h, the cumulative amount of RISP permeating the tissue was linear, in line with what would be expected for a Fickian diffusion process. As the concentration of RISP in the donor chamber increased from 2.5 to 20 mg/mL, the Jss increased in a corresponding manner, as is summarized in Figure 2. The PappA of RISP was calculated to be 0.14 0.02 cm/h for a donor chamber concentration of 2.5 mg/mL, 0.14 0.02 cm/h for a donor chamber concentration of 10 mg/mL, and JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
4588
HEEMSTRA, FINNIN, AND NICOLAZZO
Figure 1. Cumulative amount of RISP permeating the buccal mucosa over 4 h for donor chamber concentrations of 2.5 mg/mL (*), 10 mg/mL (*), and 20 mg/mL (!) RISP. Data are presented as mean SEM (n ¼ 5–7).
0.12 0.01 cm/h for a donor chamber concentration of 20 mg/mL (mean SD, n ¼ 5–7). There was no significant difference between the PappA obtained from the three different donor chamber concentrations ( p > 0.05, using a one-way analysis of variance). This study demonstrates that permeation of RISP across the buccal mucosa is a passive process for donor concentrations ranging from 2.5 to 20 mg/mL. Bidirectional Permeability of RISP Figure 3 shows the cumulative amount of RISP permeating the buccal mucosa in the mucosal-to-
Figure 2. Steady state RISP flux (Jss) for the three different donor chamber concentrations of RISP. Data are presented as mean SEM (n ¼ 5–7), and the measured donor chamber concentration is used as the x-value for the data points (r2 ¼ 0.997). JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
Figure 3. Cumulative amount of RISP permeating the buccal mucosa over time in the mucosal-to-serosal (*) and serosal-to-mucosal directions (*). Data are presented as mean SEM (n ¼ 5).
serosal (m-s) and serosal-to-mucosal (s-m) directions. The PappA for RISP was 0.15 0.03 and 0.18 0.03 cm/h for m-s and s-m, respectively (mean SD, n ¼ 5). There was no significant difference in the PappA values between the two groups ( p > 0.05, using an independent samples t-test). This confirms the above findings that RISP permeates the buccal mucosa via a passive diffusion process, and that there does not appear to be an efflux mechanism involved in the transport of RISP at this donor chamber concentration. Effect of AZ Pretreatment The impact of pretreatment of the buccal mucosa with 5% (w/w) AZ on the permeability of RISP is shown in Figure 4a. The cumulative amount of RISP permeating in the presence and absence of AZ is not different, suggesting that AZ neither increases nor decreases RISP permeability across the buccal mucosa. The enhancement ratio (ER) induced by the application of 5 mL of AZ solution was 1.27 0.33 (mean SD, n ¼ 6), which was not significantly different from 1 ( p > 0.05, using a one sampled t-test). Similarly, Figure 4b demonstrates that the disappearance of RISP from the donor chamber is not significantly altered by the presence of AZ, with an ER for the PappD being 1.04 0.12. The impact of AZ on RISP permeability was then assessed with a higher volume of the enhancer solution (10 mL). Figure 5a and b shows the appearance and disappearance, respectively, of RISP in the presence and absence of AZ. As observed with a 5 mL pretreatment, there was no significant effect of a 10 mL pretreatment, with an ER of 1.15 0.21 for the PappA, and 1.25 0.29 for the PappD. DOI 10.1002/jps
BUCCAL DELIVERY OF RISPERIDONE
4589
Figure 4. Amount of RISP permeating the buccal mucosa into the receptor chamber (a) and disappearing from the donor chamber (b) for tissue treated with 5 mL of AZ 5% (w/w) in EtOH (*) or 5 mL of EtOH (*). Data are presented as mean SEM (n ¼ 6).
Figure 5. Amount of RISP permeating the buccal mucosa into the receptor chamber (a) and disappearing from the donor chamber (b) for tissue treated with 10 mL of AZ 5% (w/w) in EtOH (*) or 10 mL of EtOH (*). Data are presented as mean SEM (n ¼ 6).
Given that a 2-h pretreatment with either 5 or 10 mL of AZ did not impact on the permeability of RISP, we attempted to determine whether coapplication of AZ and RISP resulted in a significant enhancement in permeability. The cumulative amounts of RISP in the receptor chamber following coapplication of RISP and AZ in an ethanolic vehicle are shown in Figure 6. There was no significant difference in the receptor chamber concentrations between the formulation with and without AZ, and since RISP did not show a typical permeability profile consistent with Fickian diffusion, a Jss could not be calculated using this application technique.
ability profile of RISP after application of this gel is shown in Figure 7. Since RISP was suspended in the gel, the initial donor concentration could not be determined, and therefore it was not possible to calculate the PappA. However, using Eq. (5), it was possible to predict the steady state plasma concentration of RISP that would be obtained in vivo, assuming the in vitro permeability is reflective of in vivo permeability and salivary clearance is minimal (which can be assumed if the mucoadhesive gel is constantly in contact with the buccal mucosa). Using a CL of 192 mL/min (for intermediate metabolizers of RISP),30 this in vitro flux is predicted to result in plasma concentrations of 11.2, 28.1, and 56.1 mg/L for an applied buccal mucosal surface area of 2, 5, and 10 cm2. Given that the therapeutic plasma range of RISP is between 10 and 90 mg/L,30 it is feasible that application of a mucoadhesive gel (equivalent to 2.5 mg) would result in clinically relevant plasma concentrations for the treatment of schizophrenia.
Permeability of RISP after Application of a Mucoadhesive Gel After application of the mucoadhesive gel to the buccal mucosa (equivalent to a dose of 2.5 mg), RISP permeated the tissue resulting in a Jss of 64.65 8.0 mg/cm2/h (mean SD, n ¼ 7). The permeDOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
4590
HEEMSTRA, FINNIN, AND NICOLAZZO
Figure 6. Cumulative amount of RISP permeating the buccal mucosa over time for tissue treated with 10 mL of a 24 mg/mL RISP solution in EtOH, with (*) or without (*) 5% (w/w) AZ. Data are presented as mean SEM (n ¼ 5).
DISCUSSION The aims of this study were to assess the permeability of RISP through the buccal mucosa and to determine the feasibility of a buccal RISP formulation as a means to deliver therapeutically relevant concentrations of RISP into the systemic circulation using an in vitro model. Furthermore, the potential of AZ to modify the controlled release characteristics of the buccal mucosa for the purposes of RISP delivery was examined. Our initial studies demonstrated that the steady state flux of RISP increased linearly as a function of concentration, suggesting that RISP permeates the buccal mucosa via passive diffusion, in line with what has been observed with other drugs.31–33 While passive diffusion is the main mechanism mediating the absorption of drugs across the buccal mucosa, we
Figure 7. Cumulative amount of RISP permeating the buccal mucosa after application of the buccal mucoadhesive gel. Data are presented as mean SEM (n ¼ 7). JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
were interested to determine whether active efflux pumps were also functional in this tissue, given that efflux pumps have been recently found to affect drug distribution and uptake in the skin.25,34 RISP is a known substrate of P-gp35 and this transporter has been shown to mediate its permeability across Caco-2 cells.36 However, our bidirectional studies did not show a difference in RISP permeability in the m-s and s-m direction, concordant with the concentration dependency studies and contrasting with what would be expected if active transport processes were involved. It is possible that active transport mechanisms were not detected in this study because the transporter was already saturated at the lowest RISP concentration. Studies with lower concentrations of RISP or with the use of specific P-gp inhibitors could be undertaken to determine if efflux transporters do indeed play a functional role in RISP (and other substrate) permeability across the buccal mucosa. Given the potential for RISP to permeate the buccal mucosa, we then measured the impact of AZ pretreatment on the penetration of this antipsychotic, expecting that this enhancer would either increase or decrease the rate as has been observed with TAC and E2. Pretreatment with AZ (at both 5 and 10 mL doses) did not affect the rate of RISP appearance into the receptor chamber nor the rate of RISP disappearance from the donor chamber. Furthermore, whether AZ was presented to the tissue as a pretreatment or coadministered with RISP, there was no effect of this enhancer. This is in contrast with the effect observed with other lipophilic compounds, where there was a significant increase in the rate of TAC and E2 disappearance from the donor chamber with AZ pretreatment, resulting in 4.4- and 3.4-fold increases, respectively, in tissue retention.14,15 Given that this pretreatment did not affect the permeability or tissue retention of CAF,13 it was assumed that the lipophilicity of a permeant was the determining factor for whether AZ affects the tissue retention, and therefore, permeability of the permeant. Given that RISP is similarly lipophilic to E2 and TAC, we expected that AZ would significantly alter the permeability of this antipsychotic. However, unlike TAC and E2, RISP is a weak base with an ionization constant of 8.2,37 and therefore, a proportion of the RISP will be ionized at pH 7.4, which is different to the scenario encountered with TAC and E2, which would exist in their unionized forms. It is more likely, therefore, that the extent of ionization, in addition to the lipophilicity, are important determinants of the ability of AZ to increase tissue retention of a permeant. Further studies would be required, however, to elucidate if this is the case, using a large set of drugs with different degrees of lipophilicity and extents of ionization. DOI 10.1002/jps
BUCCAL DELIVERY OF RISPERIDONE
Application of a mucoadhesive gel containing RISP (equivalent to a dose of 2.5 mg) resulted in a high steady state in vitro flux (64.65 mg/cm2/h). Using Eq. (5), we determined that the application area required to achieve therapeutic plasma concentrations of RISP with this formulation would be 2–10 cm2 (assuming the flux obtained in vitro is representative of that obtained in vivo and that salivary clearance is minimal). Since the total surface area of the buccal mucosa is about 50 cm2,38 it appears possible to use this gel formulation on this area and achieve steady state drug concentrations which are therapeutically relevant. While this is achievable, there are still some practicality issues which need to be dealt with. In particular, it will be necessary to design a formulation with appropriate mucoadhesive properties which will ensure that the concentration of RISP at the buccal mucosal surface remains constant and is not affected by salivary clearance, in order for steady state concentrations to be maintained for prolonged periods of time (up to 24 h). Such studies would be essential prior to considering the utility of this approach for the treatment of schizophrenia in humans. However, these proof-of-principle studies clearly demonstrate that it is possible to deliver RISP systemically through the buccal mucosa to achieve therapeutically relevant plasma concentrations. Additional research may also be undertaken to determine whether paliperidone, the active 9-hydroxy metabolite of RISP, may also be a suitable candidate for buccal delivery, given the interest of this active moiety in recent times.39,40
CONCLUSIONS This study is the first to demonstrate that RISP has sufficient permeability across the buccal mucosa in vitro, which is predicted to be sufficient to result in therapeutically relevant concentrations in vivo. RISP appears to permeate the buccal mucosa via a passive diffusion process with little evidence of efflux mechanisms being involved in its transport. Although AZ has been shown to alter the permeability of other nonionized lipophilic compounds, this permeabilitymodifier had no effect on RISP transport, possibly as a result of RISP existing as an ionized species at the surface of the buccal mucosa at physiological pH.
2.
3. 4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
ACKNOWLEDGMENTS Zachary Sum is acknowledged for his assistance with the permeability experiments.
REFERENCES 1. Lieberman JA, Stroup TS, McEvoy JP, Swartz MS, Rosenheck RA, Perkins DO, Keefe RSE, Davis SM, Davis CE, Lebowitz BD, Severe J, Hsiao JK. 2005. Effectiveness of antipsychotic DOI 10.1002/jps
20.
21.
22.
4591
drugs in patients with chronic schizophrenia. N Engl J Med 353:1209–1223. Mo¨ller H. 2007. Long-acting injectable risperidone for the treatment of schizophrenia: Clinical perspectives. Drugs 67: 1541–1566. Kane JM. 1996. Schizophrenia. N Engl J Med 334:34–42. Kumar M, Misra A, Babbar A, Mishra A, Mishra P, Pathak K. 2008. Intranasal nanoemulsion based brain targeting drug delivery system of risperidone. Int J Pharm 358:285–291. Elgorashi AS, Heard CM, Niazy EM, Noureldin OH, Pugh WJ. 2008. Transdermal delivery enhancement of haloperidol from gel formulations by 1,8-cineole. J Pharm Pharmacol 60:689– 692. Rathbone MJ, Drummond BK, Tucker IG. 1994. The oral cavity as a site for systemic drug delivery. Adv Drug Deliv Rev 13: 1–22. Harris D, Robinson JR. 1992. Drug delivery via the mucous membranes of the oral cavity. J Pharm Sci 81:1–10. Nasrallah HA. 2007. The case for long-acting antipsychotic agents in the post-CATIE era. Acta Psychiatr Scand 115: 260–267. Guevara-Aguirre J, Guevara-Aguirre M, Saavedra J, Bernstein G, Rosenbloom AL. 2007. Comparison of oral insulin spray and subcutaneous regular insulin at mealtime in type 1 diabetes. Diabetes Technol Ther 9:372–376. Wang C, Swerdloff R, Kipnes M, Matsumoto A, Dobs A, Cunningham G, Katznelson L, Weber T, Friedman T, Snyder P. 2004. New testosterone buccal system (Striant) delivers physiological testosterone levels: Pharmacokinetics study in hypogonadal men. J Clin Endocrinol Metab 89:3821–3829. Zhang H, Zhang J, Streisand J. 2002. Oral mucosal drug delivery: Clinical pharmacokinetics and therapeutic applications. Clin Pharmacokinet 41:661–680. Nicolazzo JA, Reed BL, Finnin BC. 2005. Buccal penetration enhancers: How do they really work? J Control Release 105:1– 115. Nicolazzo JA, Reed BL, Finnin BC. 2004. Modification of buccal drug delivery following pretreatment with skin penetration enhancers. J Pharm Sci 93:2054–2063. Nicolazzo JA, Reed BL, Finnin BC. 2005. Enhanced buccal mucosal retention and reduced buccal permeability of estradiol in the presence of padimate O and Azone1: A mechanistic study. J Pharm Sci 94:873–882. Nicolazzo JA, Reed BL, Finnin BC. 2005. Enhancing the buccal mucosal uptake and retention of triamcinolone acetonide. J Control Release 105:240–248. Lesch C, Squier C, Cruchley A, Williams D, Speight P. 1989. The permeability of human oral mucosa and skin to water. J Dent Res 68:1345–1349. Squier C, Cox P, Wertz P. 1991. Lipid content and water permeability of skin and oral mucosa. J Invest Dermatol 96: 123–126. Senel S, Kremer MJ, Kas S, Wertz PW, Hincal AA, Squier CA. 2000. Enhancing effect of chitosan on peptide drug delivery across buccal mucosa. Biomaterials 21:2067–2071. Deneer V, Drese G, Roemele´ P, Verhoef J, Lie-A-Huen L, Kingma J, Brouwers J, Junginger H. 2002. Buccal transport of flecainide and sotalol: Effect of a bile salt and ionization state. Int J Pharm 241:127–134. Shojaei A, Berner B, Li X. 1998. Transbuccal delivery of acyclovir: I. In vitro determination of routes of buccal transport. Pharm Res 15:1182–1188. Xiang J, Fang X, Li X. 2002. Transbuccal delivery of 2,3dideoxycytidine: In vitro permeation study and histological investigation. Int J Pharm 231:57–66. De Vries M, Bodde´ H, Verhoef JC, Ponec M, Craane W, Junginger H. 1991. Localization of the permeability barrier inside porcine buccal mucosa: A combined in vitro study of drug
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
4592
23.
24.
25.
26.
27.
28.
29.
30. 31.
HEEMSTRA, FINNIN, AND NICOLAZZO
permeability, electrical resistance and tissue morphology. Int J Pharm 76:25–35. Nicolazzo JA, Reed BL, Finnin BC. 2003. The effect of various in vitro conditions on the permeability characteristics of the buccal mucosa. J Pharm Sci 92:2399–2410. Linnet K, Ejsing TB. 2008. A review on the impact of P-glycoprotein on the penetration of drugs into the brain. Focus on psychotropic drugs. Eur Neuropsychopharmacol 18:157– 169. Ito K, Nguyen HT, Kato Y, Wakayama T, Kubo Y, Iseki S, Tsuji A. 2008. P-Glycoprotein (Abcb1) is involved in absorptive drug transport in skin. J Control Release 131:198–204. Johnson B, Charman W, Porter C. 2003. Application of compartmental modeling to an examination of in vitro intestinal permeability data: Assessing the impact of tissue uptake, Pglycoprotein, and CYP3A. Drug Metab Dispos 31:1151–1160. Schmolka, IR. 1972. Artificial skin I. Preparation and properties of pluronic F-127 gels for treatment of burns. J Biomed Mater Res 6:571–582. Koffi A, Agnely F, Ponchel G, Grossiord J. 2006. Modulation of the rheological and mucoadhesive properties of thermosensitive poloxamer-based hydrogels intended for the rectal administration of quinine. Eur J Pharm Sci 27:328–335. Shin S, Kim J, Oh I. 2000. Mucoadhesive and physicochemical characterization of carbopol-poloxamer gels containing triamcinolone acetonide. Drug Dev Ind Pharm 26:307–312. Moffatt AC, Osselton MD, Widdopp B. 2004. Clarke’s Analysis of Drugs and Poisons. Pharmaceutical Press, London Mashru R, Sutariya V, Sankalia M, Sankalia J. 2005. Transbuccal delivery of lamotrigine across porcine buccal mucosa: In
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
32. 33.
34.
35.
36.
37.
38.
39. 40.
vitro determination of routes of buccal transport. J Pharm Pharm Sci 8:54–62. Adrian C, Olin H, Dalhoff K, Jacobsen J. 2006. In vivo human buccal permeability of nicotine. Int J Pharm 311:196–202. Mashru R, Sutariya V, Sankalia M, Sankalia J. 2005. Effect of pH on in vitro permeation of ondansetron hydrochloride across porcine buccal mucosa. Pharm Dev Technol 10:241–247. Li Q, Kato Y, Sai Y, Imai T, Tsuji A. 2005. Multidrug resistanceassociated protein 1 functions as an efflux pump of xenobiotics in the skin. Pharm Res 22:842–846. Boulton DW, DeVane CL, Liston HL, Markowitz JS. 2002. In vitro P-glycoprotein affinity for atypical and conventional antipsychotics. Life Sci 71:163–169. Cousein E, Barthe´le´my C, Poullain S, Simon N, Lestavel S, Williame V, Joiris E, Danel C, Clavey V, Brossard D, Robert H, Crauste-Manciet S, Vaccher C, Odou P. 2007. P-glycoprotein and cytochrome P450 3A4 involvement in risperidone transport using an in vitro Caco-2/TC7 model and an in vivo model. Prog Neuropsychopharmacol Biol Psychiatry 31:878–886. Mannens G, Meuldermans W, Snoeck E, Heykants J. 1994. Plasma protein binding of risperidone and its distribution in blood. Psychopharmacology 114:566–572. Sudhakar Y, Kuotsu K, Bandyopadhyay AK. 2006. Buccal bioadhesive drug delivery—A promising option for orally less efficient drugs. J Control Release 114:15–40. Citrome L. 2007. Paliperidone: Quo vadis? Int J Clin Pract 61:653–662. Marino J, Caballero J. 2008. Paliperidone extended-release for the treatment of schizophrenia. Pharmacotherapy 28:1283– 1298.
DOI 10.1002/jps
Keywords: absorption enhancer; Azone1; buccal mucosa; epithelial permeability; mucosal delivery; passive diffusion; risperidone
INTRODUCTION Risperidone (RISP) is an atypical antipsychotic which has been used for the treatment of schizophrenia for many years. RISP is available on the market as both a tablet and oral solution, however, noncompliance is a major issue associated with antipsychotic treatment when administered via the oral route.1,2 In fact, a comprehensive review of five randomized studies demonstrated that there was an enhanced risk of relapse in schizophrenic patients who received intermittent treatment compared to patients who received continuous treatment.3 With this in mind, a long-acting injectable formulation of RISP has been developed (Risperdal Consta1), which has several pharmacokinetic advantages over the use of oral RISP, such as reduced variability in plasma concentrations due to intestinal absorption and first-pass metabolism, and a subsequent reduction in the Correspondence to: Joseph A. Nicolazzo (Telephone: þ61-39903-9605; Fax: þ61-3-9903-9583; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 99, 4584–4592 (2010) ß 2010 Wiley-Liss, Inc. and the American Pharmacists Association
4584
frequency of adverse effects. However, this longacting injectable possesses some disadvantages, including the inability to rapidly discontinue treatment in the case of unexpected side effects and its invasive nature. The need for appropriate delivery mechanisms for these beneficial antipsychotics is therefore becoming an area of research with much focus. For example, very recent studies have highlighted the nasal and transdermal routes as alternative sites for systemically delivering RISP and other antipsychotics,4,5 and it is likely that other nonconventional routes may also prove clinically relevant for this class of compounds. One such route that is often under-represented as an alternative for systemic drug delivery is the buccal mucosa. Delivery of drugs through the buccal mucosa has several advantages over the traditional oral route. Since blood flow from the oral mucosa drains directly into the jugular vein, drugs administered through the buccal mucosa gain direct access to the systemic circulation thereby avoiding hepatic first pass metabolism.6,7 Furthermore, the rate of absorption is not influenced by food or gastric emptying rate.6,7 In addition, the buccal mucosa seems well suited for the use of retentive systems, such as
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
BUCCAL DELIVERY OF RISPERIDONE
mucoadhesive tablets or patch systems. This would provide sustained delivery of drug to the bloodstream without the invasive nature of long-acting injectables,7 which is of extreme benefit to this class of drugs given the improved compliance associated with long-acting antipsychotics.8 This route of delivery is currently used in clinical practice with formulations available for insulin (Oral-LynTM),9 testosterone (Striant1),10 and prochlorperazine (Buccastem1).11 Therefore, there is significant potential to use the buccal route for systemically delivering RISP in a controlled manner, with the ultimate aim of improving patient compliance. While the mucosae of the oral cavity are generally considered more permeable than the skin,12 the buccal mucosa still acts as a barrier and may limit the permeation of a drug administered transmucosally. In order to improve drug delivery via this route, the tissue may be treated with penetration enhancers such as surfactants, fatty acids and alcohols.12 One agent which has been used in our laboratory for modifying buccal penetration of drugs is Azone1 (AZ), and the general mechanism by which drug delivery is affected with this chemical is through altered partitioning of the coadministered drug.13–15 However, the impact of AZ on buccal mucosal delivery depends on the permeant of interest. AZ appears to have no effect on the permeability of the hydrophilic caffeine (CAF) whereas it enhances the permeability and tissue retention of the more lipophilic triamcinolone acetonide (TAC).13,15 Of greater significance to controlled drug delivery is the effect of AZ on estradiol (E2) permeability across the buccal mucosa—pretreatment with AZ significantly reduced the flux of E2 by 67.8% with a 2.2-fold enhancement in the tissue concentrations of E2.14 This suggested that AZ had the potential to enhance the reservoir capacity of the buccal mucosa, leading to a slower release of E2 into the receptor compartment over a prolonged period of time. Such a technology would be of extreme benefit for the delivery of RISP through the buccal mucosa, as this may result in a slower release of RISP into the systemic circulation, resulting in a controlled delivery of this antipsychotic. The purpose of this study, therefore, was to investigate the potential of the buccal mucosa for systemic delivery of RISP using an in vitro model. Due to the limited availability of human buccal tissue, porcine buccal mucosa was used as the model tissue. Porcine buccal mucosa has been shown to exhibit similar morphology and permeability characteristics to human buccal mucosa,16,17 and has been used by many laboratories as a suitable model for assessment of buccal permeability characteristics.18–22 In addition to elucidating the mechanisms involved in the transport of RISP across the buccal mucosa, the potential of AZ to alter the permeability of this DOI 10.1002/jps
4585
compound was investigated. Furthermore, formulation approaches to increase the delivery of RISP through the buccal mucosa were evaluated to determine whether it would be feasible to deliver clinically relevant amounts of RISP to humans via this alternative route.
MATERIALS AND METHODS Materials RISP was kindly donated by Acrux Limited (Victoria, Australia) and was manufactured by Alchymars (Alathur, Tamil Nadu, India). AZ was obtained from Yick-Vic Chemicals and Pharmaceuticals (HK) Ltd. (Kwun Tong, Hong Kong). Ethanol (EtOH) 95% (v/v) was obtained from CSR Distilleries (Victoria, Australia) and acetonitrile (ACN) was obtained from Merck (Darmstadt, Germany) and was of HPLC grade. Poloxamer 407 was obtained from BASF (Ludwigshafen, Germany) and Carbopol 974 was obtained from Bronson & Jacobs (New South Wales, Australia). Krebs bicarbonate Ringer (KBR) buffer was prepared as detailed previously,23 and adjusted to physiological pH 7.4 with carbogen (95% O2 þ 5% CO2) bubbling as performed in previous experiments.23 Water was obtained from a Milli-Q water purification system (Millipore, Bedford, MA). All other chemicals were of analytical grade and used as received. Tissue Preparation Cheek tissue from domestic pigs (Sus scrofa domestica) was obtained from a local abattoir immediately after slaughter and transported in ice cold KBR. The buccal epithelium was then carefully separated from underlying submucosa with forceps and surgical scissors. The separated epithelial tissue (approximately 500 mm in thickness) was kept in ice cold KBR and supplied with carbogen bubbling until placed in modified Ussing chambers. Both donor and receptor chambers were filled with 1.5 mL of KBR and the tissue was maintained at 37 8C for 30 min. After this incubation period, permeability experiments were conducted. Assessment of Mechanisms of Transport of RISP To determine if RISP transport across the buccal mucosa was influenced by mechanisms other than passive diffusion, two studies were performed. The first study was a concentration-dependent study to ensure that RISP permeability across the buccal mucosa followed Fickian diffusion, and the second study assessed bidirectional transport to identify whether drug efflux transporters (mainly P-glycoprotein, P-gp) may potentially contribute to the absorption of RISP across the buccal mucosa, if such JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
4586
HEEMSTRA, FINNIN, AND NICOLAZZO
transporters are indeed present in this mucosa. RISP is a known substrate of the drug transporter P-gp,24 and given that P-gp has recently been suggested to be involved in drug absorption in the skin,25 we investigated whether a similar functional activity existed in the buccal mucosa using this probe molecule. Concentration Dependency of RISP Permeation After the 30 min equilibration period, KBR was removed from both chambers. The donor chamber was filled with 1.5 mL of RISP solution (at a concentration of either 2.5, 10, or 20 mg/mL in KBR) and the receptor chamber was filled with KBR. Samples of 200 mL were taken from the receptor chamber at periodic time intervals over 4 h and were replaced with fresh KBR. The results were adjusted to allow for this dilution effect. In addition, 20 mL of donor solution was removed at the start and completion of the experiment. All samples were collected in glass vials, as use of polypropylene vials led to nonspecific adsorption issues. The concentrations of RISP were determined using HPLC (detailed below). A plot comparing the cumulative amount of RISP permeating the buccal tissue over time was constructed and the steady state flux (Jss) was calculated using Eq. (1), where DM is the amount of drug transported during time Dt, and A is the diffusional area (0.64 cm2). Jss ¼
DM ADt
(1)
The apparent permeability coefficient ( PappA) of RISP was then determined by dividing the Jss with the initial donor concentration, and the PappA values obtained from different donor chamber concentrations were compared using a one-way analysis of variance with SPSS for Windows (version 15.0, SPSS, Chicago, IL). Bidirectional Study To delineate between passive diffusion and active efflux at the buccal mucosa, this study involved measuring the permeability of RISP in the directions representing drug transport from the oral cavity to the blood compartment (mucosal to serosal) and from the blood compartment to the oral cavity (serosal to mucosal). In a passive diffusion process, the Jss of RISP is expected to be similar in both directions, whereas if an efflux mechanism was involved, the serosal-to-mucosal PappA should exceed the mucosalto-serosal PappA. A donor solution of 10 mg/mL RISP in KBR was used in these experiments and samples from the receptor chamber were taken and analyzed for RISP concentration by HPLC. PappA was calculated in the serosal-to-mucosal and mucosal-to-serosal directions as detailed above, and an independent samples JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
t-test was used to compare the value between the two groups using SPSS. Effect of AZ Pretreatment Buccal mucosa was mounted in a modified Ussing chamber and excess KBR was removed from the mucosal surface with a Kimwipes1 (KimberleyClark, New South Wales, Australia). A plastic ring was placed over the 0.64 cm2 exposed area, as detailed previously.13 Five microliters of the pretreatment solution (5%, w/w, AZ in 95%, v/v, EtOH or 95%, v/v, EtOH as control) was applied to the exposed area. The solution was allowed 5 min to dry before the chambers were clamped together, filled with 1.5 mL of KBR and allowed to equilibrate with carbogen bubbling for 2 h. After the 2-h pretreatment, both donor and receptor solution were removed and replaced with a solution of RISP (20 mg/mL in KBR) or KBR, respectively. In order to assess the impact of AZ pretreatment on both disappearance of RISP from the donor chamber and appearance of RISP in the receptor chamber, both donor samples (10 mL) and receptor samples (200 mL) were taken over a 4-h period. The same volume of KBR (200 mL) was replaced in the receptor chamber to maintain a constant receptor chamber volume at each postdose time point. Donor and receptor samples were analyzed for RISP concentration by HPLC, and in addition to the PappA, the apparent coefficient of disappearance ( PappD) was also calculated using Eq. (2), PappD ¼
V ku A
(2)
where V represents the volume of the donor chamber, A is the diffusional area and ku is the uptake rate constant (s1) calculated from the negative slope of a ln (RISP mass in donor chamber) versus time plot.26 Given the potential variation in tissue thickness within different areas of the buccal cavity, AZ and EtOH-treated tissues were always taken from the same region of the buccal cavity for comparison. An enhancement ratio (ER) was then calculated with Eq. (3) for the PappA and with Eq. (4) for the PappD. ER ¼
PappA AZ treated tissue PappA EtOH treated tissue
(3)
ER ¼
PappD AZ treated tissue PappD EtOH treated tissue
(4)
Additional experiments were also conducted to determine if a larger amount of AZ (10 mL of a 5%, w/w, solution) resulted in a further enhancement in RISP permeability across the buccal mucosa. In addition to this pretreatment protocol, the impact of coadministration of AZ and RISP on buccal permeation of RISP was examined. Two formulations DOI 10.1002/jps
BUCCAL DELIVERY OF RISPERIDONE
containing 24 mg/mL of RISP in 95% (v/v) EtOH were prepared with one of the formulations containing 5% (w/w) AZ. Ten microliters of each formulation was then applied to the mucosal surface, and after a 5 min period (to allow for EtOH evaporation), the chambers were clamped together and filled with 1.5 mL KBR. The amount of RISP permeating into the receptor chamber over a 4-h period was determined as described above. Permeability of RISP after Application of a Mucoadhesive Gel To evaluate whether delivery of RISP via the buccal mucosa could result in receptor chamber concentrations required for therapeutic efficacy in vivo, a mucoadhesive gel containing a high concentration of RISP was developed and applied to the buccal mucosa. The gel was prepared using the ‘‘cold method.’’27 In a preweighed glass beaker, 4 g of poloxamer 407 was slowly added to 10 g of ice-cold water. The beaker was gently swirled and placed in a refrigerator overnight for complete polymer dissolution. Poloxamers are liquid at room temperature and exhibit low mucoadhesion,28 however, the addition of Carbopol and neutralization to pH 6 causes increased viscosity and bioadhesiveness.29 Therefore, the next day, 200 mg of Carbopol 974 was added to the viscous solution, the pH was adjusted to 6 with 1 M NaOH and the gel was made up to 20 g with water. Once formed, 3 g of the gel was spread out on a glass slab with a metal spatula. To this was slowly added 60 mg of RISP and the gel was triturated to ensure uniform distribution of RISP (which was determined by extraction of RISP from the gel and HPLC analysis). Analysis of the gel formulation by HPLC indicated that the concentration of RISP in the mucoadhesive gel was 24.77 1.08 mg/g (mean SD, n ¼ 5). Freshly separated buccal mucosa was mounted in the modified Ussing chamber. Excess KBR was removed with a Kimwipes1 and 100 mg of the gel formulation was transferred from the glass slide onto the mucosal surface with a metal spatula. The chambers were clamped together and slowly filled with 1.5 mL of KBR in both chambers. Carbogen bubbling was only supplied to the receptor chamber to prevent the gel formulation from detaching from the mucosa in the donor chamber. Samples of 10 mL were taken from the donor chamber at the start and completion of the experiment and receptor samples of 200 mL were taken at periodic time intervals over 4 h and were replaced with fresh KBR. Samples were analyzed by HPLC for RISP concentration and the Jss determined, as described above. Using the in vitrodetermined Jss, the steady state plasma concentration that would be achieved in vivo was calculated according to Eq. (5) (with the assumption that DOI 10.1002/jps
4587
salivary clearance is minimal as the gel would adhere directly to the buccal mucosa): Css ¼
Jss CL
(5)
where Css is the predicted in vivo steady state plasma concentration and CL is the in vivo plasma clearance of RISP in humans (192 mL/min).30 HPLC Analysis Donor and receptor chamber samples were diluted with an equal volume of ACN to precipitate buffer salts. Samples were vortexed and centrifuged for 30 min at 1,989g. After centrifugation, the supernatant was removed and transferred into a clean vial and analyzed with a HPLC system consisting of a Shimadzu LC-20AT pump, SIL-20 AC HT autosampler, CTO-20A column oven and a SPD 20A UV/VIS detector. A symmetry C18 3.9 mm 150 mm (Waters, Milford, MA) column was used for the analysis. The mobile phase consisted of a mixture of a 22 mM potassium phosphate buffer at pH 7.6 and 38% (v/v) ACN and the flow rate was set at 1 mL/min. The detector and the injection volume were 280 nm and 100 mL, respectively. Peak areas were calculated with Shimadzu LCSolution v. 1.24 SP1 software, and were related to RISP concentration after weighting 1/x2, where x is the concentration of RISP. Intra-day precision and accuracy were determined by analysis of six independently prepared RISP solutions at low, medium, and high concentrations, resulting in values of 0.5–7.2% and 101.8–102.8%, respectively. Interday precision was determined by analysis of these solutions on three different days (values ranged from 0.8% to 1.5%). In addition, KBR which had been exposed to buccal tissue for 4 h did not interfere with the HPLC chromatograms. Extraction efficiency was determined by comparing the peak areas of RISP in fresh KBR and KBR that had been exposed to buccal tissue, and this resulted in values of 99.6–102.7%.
RESULTS Concentration-Dependent Permeability of RISP The cumulative amount of RISP that permeated the buccal mucosa over a 4-h experimental period is shown in Figure 1. After a lag time of 1–1.5 h, the cumulative amount of RISP permeating the tissue was linear, in line with what would be expected for a Fickian diffusion process. As the concentration of RISP in the donor chamber increased from 2.5 to 20 mg/mL, the Jss increased in a corresponding manner, as is summarized in Figure 2. The PappA of RISP was calculated to be 0.14 0.02 cm/h for a donor chamber concentration of 2.5 mg/mL, 0.14 0.02 cm/h for a donor chamber concentration of 10 mg/mL, and JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
4588
HEEMSTRA, FINNIN, AND NICOLAZZO
Figure 1. Cumulative amount of RISP permeating the buccal mucosa over 4 h for donor chamber concentrations of 2.5 mg/mL (*), 10 mg/mL (*), and 20 mg/mL (!) RISP. Data are presented as mean SEM (n ¼ 5–7).
0.12 0.01 cm/h for a donor chamber concentration of 20 mg/mL (mean SD, n ¼ 5–7). There was no significant difference between the PappA obtained from the three different donor chamber concentrations ( p > 0.05, using a one-way analysis of variance). This study demonstrates that permeation of RISP across the buccal mucosa is a passive process for donor concentrations ranging from 2.5 to 20 mg/mL. Bidirectional Permeability of RISP Figure 3 shows the cumulative amount of RISP permeating the buccal mucosa in the mucosal-to-
Figure 2. Steady state RISP flux (Jss) for the three different donor chamber concentrations of RISP. Data are presented as mean SEM (n ¼ 5–7), and the measured donor chamber concentration is used as the x-value for the data points (r2 ¼ 0.997). JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
Figure 3. Cumulative amount of RISP permeating the buccal mucosa over time in the mucosal-to-serosal (*) and serosal-to-mucosal directions (*). Data are presented as mean SEM (n ¼ 5).
serosal (m-s) and serosal-to-mucosal (s-m) directions. The PappA for RISP was 0.15 0.03 and 0.18 0.03 cm/h for m-s and s-m, respectively (mean SD, n ¼ 5). There was no significant difference in the PappA values between the two groups ( p > 0.05, using an independent samples t-test). This confirms the above findings that RISP permeates the buccal mucosa via a passive diffusion process, and that there does not appear to be an efflux mechanism involved in the transport of RISP at this donor chamber concentration. Effect of AZ Pretreatment The impact of pretreatment of the buccal mucosa with 5% (w/w) AZ on the permeability of RISP is shown in Figure 4a. The cumulative amount of RISP permeating in the presence and absence of AZ is not different, suggesting that AZ neither increases nor decreases RISP permeability across the buccal mucosa. The enhancement ratio (ER) induced by the application of 5 mL of AZ solution was 1.27 0.33 (mean SD, n ¼ 6), which was not significantly different from 1 ( p > 0.05, using a one sampled t-test). Similarly, Figure 4b demonstrates that the disappearance of RISP from the donor chamber is not significantly altered by the presence of AZ, with an ER for the PappD being 1.04 0.12. The impact of AZ on RISP permeability was then assessed with a higher volume of the enhancer solution (10 mL). Figure 5a and b shows the appearance and disappearance, respectively, of RISP in the presence and absence of AZ. As observed with a 5 mL pretreatment, there was no significant effect of a 10 mL pretreatment, with an ER of 1.15 0.21 for the PappA, and 1.25 0.29 for the PappD. DOI 10.1002/jps
BUCCAL DELIVERY OF RISPERIDONE
4589
Figure 4. Amount of RISP permeating the buccal mucosa into the receptor chamber (a) and disappearing from the donor chamber (b) for tissue treated with 5 mL of AZ 5% (w/w) in EtOH (*) or 5 mL of EtOH (*). Data are presented as mean SEM (n ¼ 6).
Figure 5. Amount of RISP permeating the buccal mucosa into the receptor chamber (a) and disappearing from the donor chamber (b) for tissue treated with 10 mL of AZ 5% (w/w) in EtOH (*) or 10 mL of EtOH (*). Data are presented as mean SEM (n ¼ 6).
Given that a 2-h pretreatment with either 5 or 10 mL of AZ did not impact on the permeability of RISP, we attempted to determine whether coapplication of AZ and RISP resulted in a significant enhancement in permeability. The cumulative amounts of RISP in the receptor chamber following coapplication of RISP and AZ in an ethanolic vehicle are shown in Figure 6. There was no significant difference in the receptor chamber concentrations between the formulation with and without AZ, and since RISP did not show a typical permeability profile consistent with Fickian diffusion, a Jss could not be calculated using this application technique.
ability profile of RISP after application of this gel is shown in Figure 7. Since RISP was suspended in the gel, the initial donor concentration could not be determined, and therefore it was not possible to calculate the PappA. However, using Eq. (5), it was possible to predict the steady state plasma concentration of RISP that would be obtained in vivo, assuming the in vitro permeability is reflective of in vivo permeability and salivary clearance is minimal (which can be assumed if the mucoadhesive gel is constantly in contact with the buccal mucosa). Using a CL of 192 mL/min (for intermediate metabolizers of RISP),30 this in vitro flux is predicted to result in plasma concentrations of 11.2, 28.1, and 56.1 mg/L for an applied buccal mucosal surface area of 2, 5, and 10 cm2. Given that the therapeutic plasma range of RISP is between 10 and 90 mg/L,30 it is feasible that application of a mucoadhesive gel (equivalent to 2.5 mg) would result in clinically relevant plasma concentrations for the treatment of schizophrenia.
Permeability of RISP after Application of a Mucoadhesive Gel After application of the mucoadhesive gel to the buccal mucosa (equivalent to a dose of 2.5 mg), RISP permeated the tissue resulting in a Jss of 64.65 8.0 mg/cm2/h (mean SD, n ¼ 7). The permeDOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
4590
HEEMSTRA, FINNIN, AND NICOLAZZO
Figure 6. Cumulative amount of RISP permeating the buccal mucosa over time for tissue treated with 10 mL of a 24 mg/mL RISP solution in EtOH, with (*) or without (*) 5% (w/w) AZ. Data are presented as mean SEM (n ¼ 5).
DISCUSSION The aims of this study were to assess the permeability of RISP through the buccal mucosa and to determine the feasibility of a buccal RISP formulation as a means to deliver therapeutically relevant concentrations of RISP into the systemic circulation using an in vitro model. Furthermore, the potential of AZ to modify the controlled release characteristics of the buccal mucosa for the purposes of RISP delivery was examined. Our initial studies demonstrated that the steady state flux of RISP increased linearly as a function of concentration, suggesting that RISP permeates the buccal mucosa via passive diffusion, in line with what has been observed with other drugs.31–33 While passive diffusion is the main mechanism mediating the absorption of drugs across the buccal mucosa, we
Figure 7. Cumulative amount of RISP permeating the buccal mucosa after application of the buccal mucoadhesive gel. Data are presented as mean SEM (n ¼ 7). JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
were interested to determine whether active efflux pumps were also functional in this tissue, given that efflux pumps have been recently found to affect drug distribution and uptake in the skin.25,34 RISP is a known substrate of P-gp35 and this transporter has been shown to mediate its permeability across Caco-2 cells.36 However, our bidirectional studies did not show a difference in RISP permeability in the m-s and s-m direction, concordant with the concentration dependency studies and contrasting with what would be expected if active transport processes were involved. It is possible that active transport mechanisms were not detected in this study because the transporter was already saturated at the lowest RISP concentration. Studies with lower concentrations of RISP or with the use of specific P-gp inhibitors could be undertaken to determine if efflux transporters do indeed play a functional role in RISP (and other substrate) permeability across the buccal mucosa. Given the potential for RISP to permeate the buccal mucosa, we then measured the impact of AZ pretreatment on the penetration of this antipsychotic, expecting that this enhancer would either increase or decrease the rate as has been observed with TAC and E2. Pretreatment with AZ (at both 5 and 10 mL doses) did not affect the rate of RISP appearance into the receptor chamber nor the rate of RISP disappearance from the donor chamber. Furthermore, whether AZ was presented to the tissue as a pretreatment or coadministered with RISP, there was no effect of this enhancer. This is in contrast with the effect observed with other lipophilic compounds, where there was a significant increase in the rate of TAC and E2 disappearance from the donor chamber with AZ pretreatment, resulting in 4.4- and 3.4-fold increases, respectively, in tissue retention.14,15 Given that this pretreatment did not affect the permeability or tissue retention of CAF,13 it was assumed that the lipophilicity of a permeant was the determining factor for whether AZ affects the tissue retention, and therefore, permeability of the permeant. Given that RISP is similarly lipophilic to E2 and TAC, we expected that AZ would significantly alter the permeability of this antipsychotic. However, unlike TAC and E2, RISP is a weak base with an ionization constant of 8.2,37 and therefore, a proportion of the RISP will be ionized at pH 7.4, which is different to the scenario encountered with TAC and E2, which would exist in their unionized forms. It is more likely, therefore, that the extent of ionization, in addition to the lipophilicity, are important determinants of the ability of AZ to increase tissue retention of a permeant. Further studies would be required, however, to elucidate if this is the case, using a large set of drugs with different degrees of lipophilicity and extents of ionization. DOI 10.1002/jps
BUCCAL DELIVERY OF RISPERIDONE
Application of a mucoadhesive gel containing RISP (equivalent to a dose of 2.5 mg) resulted in a high steady state in vitro flux (64.65 mg/cm2/h). Using Eq. (5), we determined that the application area required to achieve therapeutic plasma concentrations of RISP with this formulation would be 2–10 cm2 (assuming the flux obtained in vitro is representative of that obtained in vivo and that salivary clearance is minimal). Since the total surface area of the buccal mucosa is about 50 cm2,38 it appears possible to use this gel formulation on this area and achieve steady state drug concentrations which are therapeutically relevant. While this is achievable, there are still some practicality issues which need to be dealt with. In particular, it will be necessary to design a formulation with appropriate mucoadhesive properties which will ensure that the concentration of RISP at the buccal mucosal surface remains constant and is not affected by salivary clearance, in order for steady state concentrations to be maintained for prolonged periods of time (up to 24 h). Such studies would be essential prior to considering the utility of this approach for the treatment of schizophrenia in humans. However, these proof-of-principle studies clearly demonstrate that it is possible to deliver RISP systemically through the buccal mucosa to achieve therapeutically relevant plasma concentrations. Additional research may also be undertaken to determine whether paliperidone, the active 9-hydroxy metabolite of RISP, may also be a suitable candidate for buccal delivery, given the interest of this active moiety in recent times.39,40
CONCLUSIONS This study is the first to demonstrate that RISP has sufficient permeability across the buccal mucosa in vitro, which is predicted to be sufficient to result in therapeutically relevant concentrations in vivo. RISP appears to permeate the buccal mucosa via a passive diffusion process with little evidence of efflux mechanisms being involved in its transport. Although AZ has been shown to alter the permeability of other nonionized lipophilic compounds, this permeabilitymodifier had no effect on RISP transport, possibly as a result of RISP existing as an ionized species at the surface of the buccal mucosa at physiological pH.
2.
3. 4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
ACKNOWLEDGMENTS Zachary Sum is acknowledged for his assistance with the permeability experiments.
REFERENCES 1. Lieberman JA, Stroup TS, McEvoy JP, Swartz MS, Rosenheck RA, Perkins DO, Keefe RSE, Davis SM, Davis CE, Lebowitz BD, Severe J, Hsiao JK. 2005. Effectiveness of antipsychotic DOI 10.1002/jps
20.
21.
22.
4591
drugs in patients with chronic schizophrenia. N Engl J Med 353:1209–1223. Mo¨ller H. 2007. Long-acting injectable risperidone for the treatment of schizophrenia: Clinical perspectives. Drugs 67: 1541–1566. Kane JM. 1996. Schizophrenia. N Engl J Med 334:34–42. Kumar M, Misra A, Babbar A, Mishra A, Mishra P, Pathak K. 2008. Intranasal nanoemulsion based brain targeting drug delivery system of risperidone. Int J Pharm 358:285–291. Elgorashi AS, Heard CM, Niazy EM, Noureldin OH, Pugh WJ. 2008. Transdermal delivery enhancement of haloperidol from gel formulations by 1,8-cineole. J Pharm Pharmacol 60:689– 692. Rathbone MJ, Drummond BK, Tucker IG. 1994. The oral cavity as a site for systemic drug delivery. Adv Drug Deliv Rev 13: 1–22. Harris D, Robinson JR. 1992. Drug delivery via the mucous membranes of the oral cavity. J Pharm Sci 81:1–10. Nasrallah HA. 2007. The case for long-acting antipsychotic agents in the post-CATIE era. Acta Psychiatr Scand 115: 260–267. Guevara-Aguirre J, Guevara-Aguirre M, Saavedra J, Bernstein G, Rosenbloom AL. 2007. Comparison of oral insulin spray and subcutaneous regular insulin at mealtime in type 1 diabetes. Diabetes Technol Ther 9:372–376. Wang C, Swerdloff R, Kipnes M, Matsumoto A, Dobs A, Cunningham G, Katznelson L, Weber T, Friedman T, Snyder P. 2004. New testosterone buccal system (Striant) delivers physiological testosterone levels: Pharmacokinetics study in hypogonadal men. J Clin Endocrinol Metab 89:3821–3829. Zhang H, Zhang J, Streisand J. 2002. Oral mucosal drug delivery: Clinical pharmacokinetics and therapeutic applications. Clin Pharmacokinet 41:661–680. Nicolazzo JA, Reed BL, Finnin BC. 2005. Buccal penetration enhancers: How do they really work? J Control Release 105:1– 115. Nicolazzo JA, Reed BL, Finnin BC. 2004. Modification of buccal drug delivery following pretreatment with skin penetration enhancers. J Pharm Sci 93:2054–2063. Nicolazzo JA, Reed BL, Finnin BC. 2005. Enhanced buccal mucosal retention and reduced buccal permeability of estradiol in the presence of padimate O and Azone1: A mechanistic study. J Pharm Sci 94:873–882. Nicolazzo JA, Reed BL, Finnin BC. 2005. Enhancing the buccal mucosal uptake and retention of triamcinolone acetonide. J Control Release 105:240–248. Lesch C, Squier C, Cruchley A, Williams D, Speight P. 1989. The permeability of human oral mucosa and skin to water. J Dent Res 68:1345–1349. Squier C, Cox P, Wertz P. 1991. Lipid content and water permeability of skin and oral mucosa. J Invest Dermatol 96: 123–126. Senel S, Kremer MJ, Kas S, Wertz PW, Hincal AA, Squier CA. 2000. Enhancing effect of chitosan on peptide drug delivery across buccal mucosa. Biomaterials 21:2067–2071. Deneer V, Drese G, Roemele´ P, Verhoef J, Lie-A-Huen L, Kingma J, Brouwers J, Junginger H. 2002. Buccal transport of flecainide and sotalol: Effect of a bile salt and ionization state. Int J Pharm 241:127–134. Shojaei A, Berner B, Li X. 1998. Transbuccal delivery of acyclovir: I. In vitro determination of routes of buccal transport. Pharm Res 15:1182–1188. Xiang J, Fang X, Li X. 2002. Transbuccal delivery of 2,3dideoxycytidine: In vitro permeation study and histological investigation. Int J Pharm 231:57–66. De Vries M, Bodde´ H, Verhoef JC, Ponec M, Craane W, Junginger H. 1991. Localization of the permeability barrier inside porcine buccal mucosa: A combined in vitro study of drug
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
4592
23.
24.
25.
26.
27.
28.
29.
30. 31.
HEEMSTRA, FINNIN, AND NICOLAZZO
permeability, electrical resistance and tissue morphology. Int J Pharm 76:25–35. Nicolazzo JA, Reed BL, Finnin BC. 2003. The effect of various in vitro conditions on the permeability characteristics of the buccal mucosa. J Pharm Sci 92:2399–2410. Linnet K, Ejsing TB. 2008. A review on the impact of P-glycoprotein on the penetration of drugs into the brain. Focus on psychotropic drugs. Eur Neuropsychopharmacol 18:157– 169. Ito K, Nguyen HT, Kato Y, Wakayama T, Kubo Y, Iseki S, Tsuji A. 2008. P-Glycoprotein (Abcb1) is involved in absorptive drug transport in skin. J Control Release 131:198–204. Johnson B, Charman W, Porter C. 2003. Application of compartmental modeling to an examination of in vitro intestinal permeability data: Assessing the impact of tissue uptake, Pglycoprotein, and CYP3A. Drug Metab Dispos 31:1151–1160. Schmolka, IR. 1972. Artificial skin I. Preparation and properties of pluronic F-127 gels for treatment of burns. J Biomed Mater Res 6:571–582. Koffi A, Agnely F, Ponchel G, Grossiord J. 2006. Modulation of the rheological and mucoadhesive properties of thermosensitive poloxamer-based hydrogels intended for the rectal administration of quinine. Eur J Pharm Sci 27:328–335. Shin S, Kim J, Oh I. 2000. Mucoadhesive and physicochemical characterization of carbopol-poloxamer gels containing triamcinolone acetonide. Drug Dev Ind Pharm 26:307–312. Moffatt AC, Osselton MD, Widdopp B. 2004. Clarke’s Analysis of Drugs and Poisons. Pharmaceutical Press, London Mashru R, Sutariya V, Sankalia M, Sankalia J. 2005. Transbuccal delivery of lamotrigine across porcine buccal mucosa: In
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 99, NO. 11, NOVEMBER 2010
32. 33.
34.
35.
36.
37.
38.
39. 40.
vitro determination of routes of buccal transport. J Pharm Pharm Sci 8:54–62. Adrian C, Olin H, Dalhoff K, Jacobsen J. 2006. In vivo human buccal permeability of nicotine. Int J Pharm 311:196–202. Mashru R, Sutariya V, Sankalia M, Sankalia J. 2005. Effect of pH on in vitro permeation of ondansetron hydrochloride across porcine buccal mucosa. Pharm Dev Technol 10:241–247. Li Q, Kato Y, Sai Y, Imai T, Tsuji A. 2005. Multidrug resistanceassociated protein 1 functions as an efflux pump of xenobiotics in the skin. Pharm Res 22:842–846. Boulton DW, DeVane CL, Liston HL, Markowitz JS. 2002. In vitro P-glycoprotein affinity for atypical and conventional antipsychotics. Life Sci 71:163–169. Cousein E, Barthe´le´my C, Poullain S, Simon N, Lestavel S, Williame V, Joiris E, Danel C, Clavey V, Brossard D, Robert H, Crauste-Manciet S, Vaccher C, Odou P. 2007. P-glycoprotein and cytochrome P450 3A4 involvement in risperidone transport using an in vitro Caco-2/TC7 model and an in vivo model. Prog Neuropsychopharmacol Biol Psychiatry 31:878–886. Mannens G, Meuldermans W, Snoeck E, Heykants J. 1994. Plasma protein binding of risperidone and its distribution in blood. Psychopharmacology 114:566–572. Sudhakar Y, Kuotsu K, Bandyopadhyay AK. 2006. Buccal bioadhesive drug delivery—A promising option for orally less efficient drugs. J Control Release 114:15–40. Citrome L. 2007. Paliperidone: Quo vadis? Int J Clin Pract 61:653–662. Marino J, Caballero J. 2008. Paliperidone extended-release for the treatment of schizophrenia. Pharmacotherapy 28:1283– 1298.
DOI 10.1002/jps