- Email: [email protected]
Development and evaluation of a tampering resistant transdermal fentanyl patch
International Journal of Pharmaceutics 488 (2015) 102–107
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Development and evaluation of a tampering resistant transdermal fentanyl patch Bing Cai, Håkan Engqvist * , Susanne Bredenberg Division for Applied Materials Science, Department of Engineering Sciences, The Ångström Laboratory, Uppsala University, Box 534, SE-751 21 Uppsala, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 January 2015 Received in revised form 17 April 2015 Accepted 20 April 2015 Available online 22 April 2015
With the increasing number of misuse and abuse of opioids, the resistance to tampering becomes an important attribute for transdermal opioid patches. In this study, drug-containing geopolymer granules were integrated into an adhesive matrix to improve the resistance of fast drug release against some common abuse techniques. Bench testing showed that fentanyl loaded geopolymer granules had better resistance to tampering compared to a commercial fentanyl patch. Moreover, in a pilot in vivo study on a few rats, the granules showed potential to give similar drug plasma concentrations as the commercial fentanyl patch. After integrating geopolymer granules into an adhesive matrix, the new patch showed a better resistance against the investigated tampering tests compared with the commercially available patch. In this study, we showed that incorporating drug loaded geopolymer granules into a patch adhesive has potential to improve the resistance of the fentanyl patch against tampering without compromising the drug release. ã 2015 Elsevier B.V. All rights reserved.
Keywords: Transdermal patch Tamper-resistance Geopolymer Fentanyl Abuse Drug delivery
1. Introduction The misuse and abuse of prescription opioid products have been medical concerns for many years. Fentanyl is one of the most abused opioids: a survey published in 2010 revealed that the number of emergency visits related to the non-medical use of the fentanyl has increased 105% during 2004–2008 in United States (Anonymous, 2010a). Although the transdermal fentanyl gained a lot of patient compliance for its non-invasive and around-the-clock treatment, the number of abuse cases on these patches increase significantly, causing dose dumping and thus serious side effects (Carson et al., 2010; Moon and Chun, 2011; Prosser et al., 2010; Woodall et al., 2008). Commonly, abusers override the controlled release mechanisms of the patches in order to obtain fast on-set euphoria by oral ingestion, injection and inhalation (Butler et al., 2011; Mastropietro and Omidian, 2013). To address the growing problem of the nonmedical use of fentanyl, there is a need for transdermal formulations that could reduce abuse potential and the risk of dose dumping (Howard and Reidenberg, 2004; Kugelmann and Bartholomaeus, 2006; Tavares et al., 2011). Several tamper-resistant patches have been designed to solve this problem. Patent documents WO2004098568 A2,
* Corresponding author. Tel.: +46 18 471 7130; fax: +46 18 471 3572. E-mail address: [email protected] (H. Engqvist). http://dx.doi.org/10.1016/j.ijpharm.2015.04.061 0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
US7182955 B2 and US8790689 B2 describe transdermal dosage forms with a separated compartment containing antagonist/aversive agent (Hart et al., 2007; Howard and Reidenberg, 2004, 2005). The antagonists or aversive agents are not liberated if the patch is correctly used but will release along with the opioids if the patch formulation is tampered. US7511054 B2 illustrates a dosage form that contains opioid pro-drugs and a form of antagonist poorly absorbed through the skin (Stinchcomb et al., 2009). The antagonist would be minimally delivered transdermally but would take effect when the dosage form is tampered with. As the tamper-resistance of the product will never be absolute, researchers are endeavoring to find better formulations that will further reduce the abuse potential without compromising the efficacy of drug administration. Geopolymers, a type of ceramic materials, are composed of three-dimensional networks of SiO4 and AlO4. Previous studies have suggested that geopolymers could be used as a drug carrier for controlled-release for oral formulation with better tamper-resistance than the compared commercial tablet (Cai et al., 2014; Jämstorp et al., 2010). For the geopolymer-based drug carrier, diffusion is the main rate-limiting step of drug release (Jämstorp et al., 2010, 2011). The physical properties of geopolymer, such as porosity and mechanical strength, could be adjusted by changing its composition and synthesis condition. Our previous study showed that these geopolymer-based formulations could maintain controlled drug release even after milled into fine
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
granules (Cai et al., 2014). Moreover, in comparison to the commercial tablet based on a polymer matrix, these formulations had better resistance against the extraction in heated water. In Cai et al. (2014), geopolymer showed its ability to increase the resistance of oral dosage forms against some common tampering methods and reduce the risk of dose dumping. This study aims to evaluate geopolymer-integrated transdermal patch in its resistance to tampering. To our knowledge, this is the first attempt to integrate ceramics into the matrix of transdermal patches to reduce their abuse potential. The transdermal patch formulation that contains geopolymer granules in the matrix layer was expected to have better tamper-resistance against some common abuse methods without compromising the efficiency of drug delivery, as schematically illustrated in Fig. 1.
103
the geopolymer-integrated patch (Patch A) was 0.314 mg/cm2. A control patch (Patch B) that had the same adhesive matrix and backing layer as Patch A but contained fentanyl base without geopolymer carrier was used for comparison. The drug concentration in the control patch was 0.304 mg/cm2. The drug concentration in Patches A and B were calculated based on patch area, thickness, fentanyl loading and an assumed adhesive layer density of 1. Patch C, a commercial fentanyl patch (Durogesic1), was used for comparison with Patches A and B in their tamperresistance. The adhesive used in Patches A and B is DuroTak1 87-4098, which is an acrylates copolymer containing vinyl acetate similar to the adhesive used in Patch C, DuroTak1 87-4287. 2.3. Evaluation of the resistance to tampering and in vivo drug availability of the geopolymer granules and commercial patches
2. Materials and methods 2.1. Materials Kaolin (Al2Si2O5(OH)4), fumed silica (SiO2, 7 nm particle size), reagent grade sodium hydroxide (NaOH), monopotassium phosphate (KH2PO4), 37% fuming hydrochloric acid (HCl) and 99.5% ethanol were purchased from Sigma–Aldrich. Fentanyl base (MacFarlan and Smith, Edinburgh, UK) was donated by Orexo AB, Sweden. DuroTak1 87-4098 were obtained from National Starch Chemicals (Bridgewater, U.S.). A commercially available fentanyl patch, Durogesic1 (Janssen-Cilag, Belgium) was used for comparison. 2.2. Synthesis The synthesis procedure of geopolymer drug carrier is described in detail in Jämstorp et al. (2010). The geopolymer precursor used in this study had the following composition: Si/Al 2.129, H2O /Al2O3 14.95 and Na2O/Al2O3 1.472 (in molar ratio). The cured geopolymer with 14.9 wt% of fentanyl was ground manually using a mortar and pestle and the residual particles were separated into two particle-size ranges: 315–710 mm and 50–100 mm, respectively, using Retsch1 sieves (F. Kurt Retsch GmbH, Germany). The geopolymer granules with particle size 50–100 mm were integrated into matrix layer to fabricate a transdermal patch intended to resistant against tampering. The drug containing geopolymer carriers with diameter between 50 and 100 mm or fentanyl base were dispersed in IPA (propan-2-ol) within 3 min before mixed with adhesive (DuroTak1 87-4098). The mixture was with the ratio of 1 g geopolymer or 149 mg of fentanyl base to 2.8 g of IPA and 9.98 g of adhesive. The vial contents were mixed via rotation for 3 min until the powder dispersion appeared homogenous. The mixture was immediately laminated on release liner at a 500 mm setting. The casted film was dried in steps: 15 min at room temperature, 5 min at 50 C and then 5 min at 90 C. The dried film was cooled and covered with a backing layer. The finished patches were stored in the ambient temperature with relative humidity <20% before testing. The drug concentration in
As a first step, a preliminary drug release study was performed in order to investigate the tamper resistance of geopolymer granules with the size range 315–710 mm. A single dose of Patch C (Durogesic1 12 mcg/h) was used as comparison to the geopolymer granules with the corresponding dose in terms of extractability in the selected media. The test was carried out in a USP dissolution apparatus II (Sotax AT7 Smart, Sotax AG, Switzerland) equipped with mini vessels and paddles. The measurements were performed using a paddle speed of 50 rpm at 37 C in 200 mL of phosphate buffer at pH 6.8 0.5 or in 200 mL of 50% v/v ethanol solution. The amount of fentanyl was analyzed using isocratic reversed-phase HPLC (Dionex Summit1, Dionex UK Ltd.) with a XTerra MS C18 column (2.1 mm ID 50 mm, 3.5 mm, Waters Corp., Milford, MA, USA). A PDA detector was utilized at a wavelength of 210 nm. The mobile phase was a mixture of acetonitrile and the solvent with 0.1 v/v% TFA, 72 v/v% water and 28 v/v% acetonitrile. The mobile phase was degassed inline at a flow rate 0.25 mL/min. As a second step, a pilot in vivo study on the plasma concentration response of one forth of Patch C (Durogesic1 12 mcg/h) and geopolymer granules (size range of 315–710 mm) with corresponding dose was investigated on male Sprague Dawley rats (Taconic, Denmark). The animals were housed in ventilated cabinet with 12 h lighting arrangement. A 5-week acclimatization period was allowed before the test commenced. The fur on the back of the rats was shaved and treated with depilatory agent the day before the experiment. The granules were attached to the skin on the back of the rats using a circular-shaped placebo elastic textile patch (Cederroth AB, Sweden) with 1.2 cm in diameter in order to hold the granules in place. Due to the size variance between rats, the applied dose was adjusted to around 1500 mg/kg for comparison. Blood samples were taken after 0.25, 0.5, 1, 2, 4, 6 and 24 h and stored at +4 C for maximum 30 min until centrifugation at 6000 rpm for 5 min. Plasma was stored initially at –20 C for 1–3 days and then at –80 C for 4 days until analyzed. Before analysis, the plasma samples were thawed and mixed on a shaker. A 50 mL of plasma was transferred into a microwell plate and precipitated with 200 mL of acetonitile. Internal standard solution was added into the plate and the mixture was mixed on a shaker for 30 min. The plasma samples were then centrifuged at
Fig. 1. Illustration of the geopolymer-integrated fentanyl transdermal patch.
104
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
4000 rpm for 15 min. The supernatant was transferred to another microwell plate and the acetonitile was evaporated under nitrogen at 45 C until approximately 50–100 mL was left. The residue was diluted in 50 mL acetonitile:water (50:50) with 0.1% acetic acid. The content was transferred to 96-deepwell plate and mixed on a shaker. The concentration of fentanyl in plasma was determined using reversed-phase UHPLC (Surveyor MS, Thermo Scientific, Bremen, Germany) with an Eclipse Plus C18 column (2.1 mm ID 50 mm, 3.5 mm, Agilent Technologies, Palo Alto, CA, USA) and Heated Electrospray Tandem Mass Spectrometry (HESI-MSMS, Thermo TSQ Quantum Ultra, Thermo Scientific, Bremen, Germany). The mobile phase was a mixture of 80 v/v% acetonitirle, water 20 v/v% and 0.1% acetic acid. The mobile phase was degassed inline at a flow rate 0.4 mL/min. All animal procedures were approved by the local ethical committee in Uppsala (permit number: C C279/10) and followed the guidelines of European Communities Council Directive (86/609/EEC). 2.4. Evaluation of the resistance of transdermal formulations against tampering The cross-section morphology of the house-made patches (Patches A and B) was obtained by the images from a Leo 1550 FEG scanning electron microscope (SEM, Zeiss, Germany). The patches were firstly coated with a thin layer of gold/palladium prior to observation to avoid the charging effect of the samples. One eighth of Patch C (Durogesic1 25 mcg/h) and Patches A and B with corresponding dose were evaluated by in vitro tampering tests aiming to evaluate the difficulty level of by passing or compromising controlled release mechanisms of the transdermal patch to achieve rapid or immediate drug release. These test conditions were selected and designed according to the previous published studies and regulations (Anonymous, 2010b, 2013; Goliber, 2005), which simulate some common abuse methods: extraction, chewing, smoking and swallowing (Butler et al., 2011; Katz et al., 2006). The patches were extracted under the following conditions in two-hour course: 40% ethanol aqueous solution (ambient temperature), distillated water (70 C) and HCl solution at pH 1 0.5 (ambient temperature). Drug release in the solvents with 40% ethanol and 70 C water simulated the extraction in the commercial alcohol drinks and in water at elevated temperature, respectively. The solution at pH 1 represented the low-pH solvent that can extract fentanyl effectively. All extractions were performed with the same drug-to-solvent ratio of 1 mg fentanyl to 19 mL of solvent. Milling the patches by a mortar and pestle for 10 min with 5 mL of phosphate buffer at pH 6.8 0.5 simulated chewing by the abusers. Smoking of the patch was simulated by heating the patch at 120 C for 20 min. The evaporated drug amount was determined by the difference of drug release amount from the patch before and after heating in 500 mL HCl solution at pH 1 0.5 by USP dissolution bath at 37 C and stirring rate at 50 rpm for 24 h. Oral indigestion is one of the most common and easiest ways for abusers. Fentanyl could be well extracted in stomach as it has higher solubility in low pH solutions (Roy and Flynn, 1989). The drug releases from patch in gastric and intestinal fluid conditions were simulated in a standardized USP apparatus II (Sotax AG, Switzerland) using 500 mL media at 37 C with stirring rate at 50 rpm. The patch sample containing approximate 2.1 mg fentanyl was fixed on a metal plate, which was then placed on the bottom of dissolution bath. Dissolution media were at pH 1 0.5 (0.1 M HCl), pH 6.8 0.5 (0.05 M KH2PO4, NaOH) or pH 1 0.5 with 40% ethanol to simulate the gastrointestinal condition or the gastric condition when patients digest the patch with alcohol drinks, respectively. Aliquots (2 mL) were withdrawn from the dissolution bath by a
sampling tube at the predetermined time intervals. Higuchi model was used to evaluate the release profiles and the fitting with R2 over 0.95 was considered as a good fitting. All tampering tests were conducted in triplicates and the concentration of fentanyl was measured using reverse-phase HPLC (Waters Corp., Milford, MA, USA) and a YMC-Triart C18 column (2.0 mm ID 12 mm, 3 mm, YMC, Japan). A PDA detector was utilized at a wavelength of 210 nm. The mobile phase was a mixture of acetonitrile and the solvent with 0.1 v/v% TFA, 72 v/v% water and 28 v/v% acetonitrile with a ratio of (=25:75, v/v). The mobile phase was degassed inline at a flow rate 0.3 mL/min. 3. Results and discussions 3.1. Evaluation of resistance to tampering and in vivo drug absorption from geopolymer granules The drug extraction from Patch C and geopolymer granules with particle size ranges of 315–710 mm was estimated in phosphate buffer and 50% ethanol solution at 37 C (Fig. 2). In both media, the drug releases from geopolymer granules with particle size range of 315–710 mm were slower than that from commercial patch: Patch C released all drug content within 1 h (Fig. 2). It suggested that geopolymer, as an inert and unswellable matrix, had better resistance to extraction than commercial patch. The results showed a slower fentanyl release in the buffer solution than in ethanol aqueous solution, which probably was due to the hydrophobicity and solubility of fentanyl (Roy and Flynn, 1988). The results of this in vitro drug release indicated the potential role of geopolymer material in reducing the risk of dose dumping. As a next step, the transdermal drug administration by the geopolymer granules and Patch C were evaluated in a pilot study on rats (Fig. 3). The plasma drug concentration vs. time profiles of the rats applied with the geopolymer granules were similar to those with the commercial patch. Although this is a pilot study and the number of animals is few, the results indicate that the geopolymer granules might be able to achieve similar bioavailability as the commercial patch. In general, both formulations resulted in a linear increase of plasma concentration in the first 6 h and maintained the concentration for 24 h. The responding plasma concentration had a larger deviation for the repeats using geopolymer granules than the ones using Patch C. We speculate that it could cause from different reasons e.g. inhomogeneous distribution of the granules and varied contact area of the attached granules with the skin. We believe that reducing the granule sizes could improve the consistency and adhesiveness of the patch, and
Fig. 2. In vitro fentanyl release from geopolymer granules and Patch C in pH 6.8 and 50 v/v% ethanol solution during the first 4 h.
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
105
Fig. 3. The absorption of fentanyl in geopolymer granules and commercial patch after applied to the shaved back of 5 rats. Plasma levels adjusted to dose around 1500 mg/kg.
thus the granules with smaller particle size, 50–100 mm, were integrated in the patch in the further study. 3.2. Evaluation of the resistance of transdermal formulations against tampering The cross-sections of both Patches A and B were observed under SEM. The images of Patch A showed that geopolymer granules were embedded in the adhesive layer (Fig. 4a and b). Some voids were found between granules and adhesives. The geopolymer particles showed similar microstructure as observations in the previous study (Jämstorp et al., 2010) (Fig. 4c). Tests simulating five common abuse methods were used to evaluate the tamper-resistance of Patches A, B and C (Fig. 5). Patch C represented a recognized standard dosage form, i.e. a commercially available patch. Patch B was the control sample, containing the same adhesive materials as Patch A. However, Patch B had a continuous matrix but Patch A had geopolymer particles replacing the space of some adhesives. Therefore, despite diffusing out from a thinner layer of adhesives, the drug molecules would experience other processes in Patch A than that in Patch B before release: i.e. the drug had to diffuse from geopolymer granules and thereafter a partition at the interface between geopolymer and adhesives occurred. Therefore, the difference of the extractability between
Fig. 5. (a) Fraction of released drug from Patches A, B and C after extraction tests. The error bars represent the confidence interval. (b) The in vitro drug release profile of the Patches A, B and C and the residual patches in pH1HCl solution after heating at 120 C for 20 min.
Patches A and B is believed to depend on the joint effect of the water and drug diffusion rate in geopolymer and adhesives and the partition rate at the interface under the chosen extraction condition. By extracting in heated water, Patch A showed a much better resistance than the other patches: Patch A released 12.5% while
Fig. 4. SEM images of the cross-sections of Patch A (a–c) and Patch B (d–f). The layers of release liner, matrix and backing layer are indicated by arrows in (a) and (d).
106
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
Patches B and C released 30% and 24.6%, respectively. Since the heated polymer in the patch matrix generally get more flexible and expanded than ceramics, the drug molecules could diffuse faster in the polymer matrix after heating. Consequently, the drug release from Patch B was influenced by the temperature to a greater extent than that of Patch A. It could also be applied to the results of simulated smoking: the release was 13.4% and 27.6% from Patches B and C, respectively, while only 2% of the drug released from Patch A. After the heating, the drug release profiles of residual patches were compared to the respectively non-tampered patches at pH1 (Fig. 5b). The drug release from the heated patches was all slower compared to the ones that had not been heated, indicating that heating influenced the drug release from the residual patches. Despite the release percentage, Patch C showed similar drug release profiles before and after heating, while Patches A and B had a slower release in the first 2–3 h after heating. In 40% ethanol solution, Patch A performed similarly as Patch B, while Patch C released the highest fraction of drug content. As discussed previously, it was probably due to the joint effect of the different rate in diffusion and partition experienced by the drug molecules in these patches. Moreover, Patches A and B showed similar resistance to the simulated chewing and the drug release was 5.8% and 4.4%, respectively. Patch C, however, released 40.5% of the drug content. In pH1 solution, Patch C, yet again, had the highest fraction of drug release, 40.4%. However, extraction at low pH, Patch A had a higher released drug fraction than Patch B: Patch A released 19% while the Patch B released 11%. The low pH solution opens up the geopolymer pores and increases the overall all drug diffusivity, resulting in a faster drug release from Patch A (Jämstorp et al., 2010). Polymers, such as methyl acrylate-methacrylic acid copolymers and methyl methacrylate-methacrylic acid copolymers, could be blended into the geopolymer to improve the resistance to acid solutions for future development. It is worth to notice that although Patch A released a slightly higher fraction in low pH solution than Patch B, the resistance of Patch A to tampering was better than Patch B in the other two extraction conditions and better than Patch C in all tested extractions. Fig. 6 shows the drug release profiles of the three patches in the dissolution media simulating stomach and intestinal pH. At both pH, the drug release from Patch C was fastest among the three patches, released over 90% of the drug content within 6 h. It indicates that Patch C had much less resistance to these in vitro conditions than the other two patches, which could lead to dose dumping and lethal side effects. The release was slightly faster at pH 6.8 than at pH 1 from Patch C, but it was the opposite for the
Table 1 The R2 of the linear approximation of the drug release profile of the Patches A, B and C to square root of time. Samples
R2
Patch Patch Patch Patch Patch Patch
0.993 0.999 0.999 0.981 0.996 0.979
A in pH1 B in pH1 C in pH1 A in pH6.8 B in pH6.8 C in pH6.8
other two patches. On the other hand, Patches A and B released fentanyl in a more controlled manner. Although Patch A released fentanyl slightly faster than Patch B in pH 1 solution, it released slightly slower in pH 6.8 solution. But both patches released similar amount of drug at the end of the measurements, i.e. 24 h. The dissolution results were different to what was observed in the extraction test at pH 1. It is possible that, even though the solution at low pH could open up the geopolymer pores, the large amount of dissolution buffer in the drug release test could increase the drug release in the polymer matrix as well. As a result, the overall effect of drug diffusion and partition in Patches A and B could be similar. The release profiles from all three patches in the tested dissolution media had the linear relationship to square root of time (Table 1), which could indicate that the releases were controlled by a diffusion-based release mechanism. However, all patches released much faster in pH1 with 40% ethanol: most of the drug content was released in 3 h, probably owing to the hydrophobicity of fentanyl. Patch A, again, showed a slightly better resistance to this media in the first hour, indicating it could have lower risk to cause dose-dumping if the patient digest the patch with alcohol drinks. An in vivo bioavailability study of Patch A could be the next step of this study. As reported previously, drug depot would form in the upper layer of skin in the first few hours after applying Patch C and the depot is responsible for the subsequent absorption (Margetts and Sawyer, 2007; Van Nimmen and Veulemans, 2007). The drug release might be controlled by the release rate from Patch A matrix to a greater extent. Therefore, the slower release of Patch A might provide a safer therapy regardless of the skin condition of the patient. With the increased usage of transdermal opioid formulations, the amount of the related abuse cases rises considerably as well. In this study, we showed a possibility to improve the resistance to tampering by integrating ceramic granules in the adhesive matrix of fentanyl transdermal patch. Incorporating geopolymer granules in the matrix could increase the difficulty to compromise the controlled drug release under the tested extraction conditions, especially at elevated temperature. Compared to the commercial patch, this novel patch could hinder the non-medical use of opioid transdermal patches and thus possibly alleviate the situation of opioid abuse. Further improvements on the resistance in low pH and ethanol solution could be explored by addition of polymer excipients that have low solubility at acidic and/or ethanol solution into the geopolymer (Jämstorp et al., 2012). However, the influence of these excipients to the efficiency of the transdermal delivery and skin irritation is pending to be evaluated. 4. Conclusion
Fig. 6. Drug release profile from the Patches A, B and C in pH1, pH6.8 and pH1 with 40% ethanol solution. The error bar represents the confidence interval.
As the abuse problems related to transdermal opioid formulation increases, new solution to increase resistance to tampering is needed. In this paper, a patch technology with drug loaded geopolymer granules incorporated into the adhesive matrix is presented. To our knowledge, this is the first attempt to blend
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
ceramic materials into the adhesive to improve the resistance to tampering. The geopolymer granules showed a better resistance to extraction and gave a similar in vivo response as a commercial patch. Bench testing showed that the geopolymer-integrated patch had better resistance to the extraction in some common solvents than the commercial fentanyl patch. However, the resistance to extraction of geopolymer-integrated patch in big volumes of alcohol containing solution was less pronounced. The novel patch could release drug in a more controlled manner at pH 1 and pH 6.8, while the commercial patch released most of the drug content within 6 h. With better resistance to tampering than the commercial patch, this novel patch could hinder the non-medical use of opioid patches and possible alleviate the abuse situation of opioids. Acknowledgements This work has received support from Orexo AB for kind contribution of the materials and Sweden’s Innovation Agency (VINNOVA) and the Swedish Research Council (2011-3399 and 2011-4444) for financial contributions. Karin Söderkvist, Anders Sågström, Cecilia Coupland, Sari Öbrink, Ulrica Roos are acknowledged for some of the experimental works. References Anonymous, 2010a. The DAWN Report: Trends in Emergency Department Visits Involving Nonmedical Use of Narcotic Pain Relievers. Drug Abuse Warning Network (DAWN), Substance Abuse and Mental Health Services Administration, Office of Applied Studies, Rockville. Anonymous, 2010b. Guidance for Industry Assessment of Abuse Potential of Drugs. In: F.a.D.A., U.S. Department of Health and Human Services (Ed.), Center for Drug Evaluation and Research (CDER), Rockville. Anonymous, 2013. Guidance for Industry Abuse-Deterrent Opioids – Evaluation and Labeling, Draft Guidance. U.S. Department of Health and Human Services; Food and Drug Administration; Center for Drug Evaluation and Research (CDER). Butler, S.F., Black, R.A., Cassidy, T.A., Dailey, T.M., Budman, S.H., 2011. Abuse risks and routes of administration of different prescription opioid compounds and formulations. Harm Reduct. J. 8, 29. Cai, B., Engqvist, H., Bredenberg, S., 2014. Evaluation of the resistance of a geopolymer-based drug delivery system to tampering. Int. J. Pharm. 465, 169–174. Carson, H.J., Knight, L.D., Dudley, M.H., Garg, U., 2010. A fatality involving an unusual route of fentanyl delivery: chewing and aspirating the transdermal patch. Legal Med. (Tokyo, Japan) 12, 157–159.
107
Goliber, P.A., 2005. Benchtop Evaluations of Tampering with Pharmaceutical Dosage Forms. Opioid abuse resistance conference. Hart, J.R., Anderson, S.C., Pohl, M.A., Bosl, E.R., 2007. Abuse-resistant transdermal dosage form, US 7182955 B2. Howard, S.A., Reidenberg, B., 2004. Tamper resistant transdermal dosage form, US 20050002997 A1. Howard, S.A., Reidenberg, B., 2005. Tamper-resistant transdermal dosage form comprising an active agent component and an adverse agent component at the distal site of the active agent layer, WO 2004098567 A3. Jämstorp, E., Forsgren, J., Bredenberg, S., Engqvist, H., Strømme, M., 2010. Mechanically strong geopolymers offer new possibilities in treatment of chronic pain. J. Controlled Release 146, 370–377. Jämstorp, E., Strømme, M., Frenning, G., 2011. Modeling structure–function relationships for diffusive drug transport in inert porous geopolymer matrices. J. Pharm. Sci. 100, 4338–4348. Jämstorp, E., Yarra, T., Cai, B., Engqvist, H., Bredenberg, S., Strømme, M., 2012. Polymer excipients enable sustained drug release in low pH from mechanically strong inorganic geopolymers. Results Pharm. Sci. 2, 23–28. Katz, N.P., Buse, D.C., Budman, S.H., Wing Venuti, S., Fernandez, K.C., Benoit, C., Bianchi, R., Cooper, D., Jasinski, D.R., Smith, D.E., Butler, S.F., 2006. Development and preliminary experience with an ease of extractability rating system for prescription opioids. Drug Dev. Ind. Pharm. 32, 727–746. Kugelmann, H., Bartholomaeus, J., 2006. Abuse-resistant transdermal system, US20070065365 A1. Margetts, L., Sawyer, R., 2007. Transdermal drug delivery: principles and opioid therapy. Contin. Educ. Anaesth. Crit. Care Pain 7, 171–176. Mastropietro, D.J., Omidian, H., 2013. Current approaches in tamper-resistant and abuse-deterrent formulations. Drug Dev. Ind. Pharm. 39, 611–624. Moon, J.M., Chun, B.J., 2011. Fentanyl intoxication caused by abuse of transdermal fentanyl. J. Emerg. Med. 40, 37–40. Prosser, J.M., Jones, B.E., Nelson, L., 2010. Complications of oral exposure to fentanyl transdermal delivery system patches. J. Med. Toxicol.: Off. J. Am. Coll. Med. Toxicol. 6, 443–447. Roy, S.D., Flynn, G.L., 1988. Solubility and related physicochemical properties of narcotic analgesics. Pharm. Res. 5, 580–586. Roy, S., Flynn, G., 1989. Solubility behavior of narcotic analgesics in aqueous media: solubilities and dissociation constants of morphine, fentanyl, and sufentanil. Pharm. Res. 6, 147–151. Stinchcomb, A.L., Golinski, M.J., Hammell, D.C., Howard, J.L., 2009. Transdermally deliverable opioid prodrugs, abuse-resistant compositions and methods of using opioid prodrugs, US 20090156814 A1. Tavares, L., Reidenberg, B.E., Sackler, R.S., Wright, C., Alfonso, M.A., Oshlack, B., Cassidy, J.P., Carpanzano, A.E., Gullapalli, R.P., Shevchuk, I., 2011. Abuse resistant opioid containing transdermal systems, EP Patent 2062573. Van Nimmen, N.F., Veulemans, H.A., 2007. Validated GC–MS analysis for the determination of residual fentanyl in applied Durogesic reservoir and Durogesic D-Trans matrix transdermal fentanyl patches. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 846, 264–272. Woodall, K.L., Martin, T.L., McLellan, B.A., 2008. Oral abuse of fentanyl patches (Duragesic): seven case reports. J. Forensic Sci. 53, 222–225.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Development and evaluation of a tampering resistant transdermal fentanyl patch Bing Cai, Håkan Engqvist * , Susanne Bredenberg Division for Applied Materials Science, Department of Engineering Sciences, The Ångström Laboratory, Uppsala University, Box 534, SE-751 21 Uppsala, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 January 2015 Received in revised form 17 April 2015 Accepted 20 April 2015 Available online 22 April 2015
With the increasing number of misuse and abuse of opioids, the resistance to tampering becomes an important attribute for transdermal opioid patches. In this study, drug-containing geopolymer granules were integrated into an adhesive matrix to improve the resistance of fast drug release against some common abuse techniques. Bench testing showed that fentanyl loaded geopolymer granules had better resistance to tampering compared to a commercial fentanyl patch. Moreover, in a pilot in vivo study on a few rats, the granules showed potential to give similar drug plasma concentrations as the commercial fentanyl patch. After integrating geopolymer granules into an adhesive matrix, the new patch showed a better resistance against the investigated tampering tests compared with the commercially available patch. In this study, we showed that incorporating drug loaded geopolymer granules into a patch adhesive has potential to improve the resistance of the fentanyl patch against tampering without compromising the drug release. ã 2015 Elsevier B.V. All rights reserved.
Keywords: Transdermal patch Tamper-resistance Geopolymer Fentanyl Abuse Drug delivery
1. Introduction The misuse and abuse of prescription opioid products have been medical concerns for many years. Fentanyl is one of the most abused opioids: a survey published in 2010 revealed that the number of emergency visits related to the non-medical use of the fentanyl has increased 105% during 2004–2008 in United States (Anonymous, 2010a). Although the transdermal fentanyl gained a lot of patient compliance for its non-invasive and around-the-clock treatment, the number of abuse cases on these patches increase significantly, causing dose dumping and thus serious side effects (Carson et al., 2010; Moon and Chun, 2011; Prosser et al., 2010; Woodall et al., 2008). Commonly, abusers override the controlled release mechanisms of the patches in order to obtain fast on-set euphoria by oral ingestion, injection and inhalation (Butler et al., 2011; Mastropietro and Omidian, 2013). To address the growing problem of the nonmedical use of fentanyl, there is a need for transdermal formulations that could reduce abuse potential and the risk of dose dumping (Howard and Reidenberg, 2004; Kugelmann and Bartholomaeus, 2006; Tavares et al., 2011). Several tamper-resistant patches have been designed to solve this problem. Patent documents WO2004098568 A2,
* Corresponding author. Tel.: +46 18 471 7130; fax: +46 18 471 3572. E-mail address: [email protected] (H. Engqvist). http://dx.doi.org/10.1016/j.ijpharm.2015.04.061 0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
US7182955 B2 and US8790689 B2 describe transdermal dosage forms with a separated compartment containing antagonist/aversive agent (Hart et al., 2007; Howard and Reidenberg, 2004, 2005). The antagonists or aversive agents are not liberated if the patch is correctly used but will release along with the opioids if the patch formulation is tampered. US7511054 B2 illustrates a dosage form that contains opioid pro-drugs and a form of antagonist poorly absorbed through the skin (Stinchcomb et al., 2009). The antagonist would be minimally delivered transdermally but would take effect when the dosage form is tampered with. As the tamper-resistance of the product will never be absolute, researchers are endeavoring to find better formulations that will further reduce the abuse potential without compromising the efficacy of drug administration. Geopolymers, a type of ceramic materials, are composed of three-dimensional networks of SiO4 and AlO4. Previous studies have suggested that geopolymers could be used as a drug carrier for controlled-release for oral formulation with better tamper-resistance than the compared commercial tablet (Cai et al., 2014; Jämstorp et al., 2010). For the geopolymer-based drug carrier, diffusion is the main rate-limiting step of drug release (Jämstorp et al., 2010, 2011). The physical properties of geopolymer, such as porosity and mechanical strength, could be adjusted by changing its composition and synthesis condition. Our previous study showed that these geopolymer-based formulations could maintain controlled drug release even after milled into fine
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
granules (Cai et al., 2014). Moreover, in comparison to the commercial tablet based on a polymer matrix, these formulations had better resistance against the extraction in heated water. In Cai et al. (2014), geopolymer showed its ability to increase the resistance of oral dosage forms against some common tampering methods and reduce the risk of dose dumping. This study aims to evaluate geopolymer-integrated transdermal patch in its resistance to tampering. To our knowledge, this is the first attempt to integrate ceramics into the matrix of transdermal patches to reduce their abuse potential. The transdermal patch formulation that contains geopolymer granules in the matrix layer was expected to have better tamper-resistance against some common abuse methods without compromising the efficiency of drug delivery, as schematically illustrated in Fig. 1.
103
the geopolymer-integrated patch (Patch A) was 0.314 mg/cm2. A control patch (Patch B) that had the same adhesive matrix and backing layer as Patch A but contained fentanyl base without geopolymer carrier was used for comparison. The drug concentration in the control patch was 0.304 mg/cm2. The drug concentration in Patches A and B were calculated based on patch area, thickness, fentanyl loading and an assumed adhesive layer density of 1. Patch C, a commercial fentanyl patch (Durogesic1), was used for comparison with Patches A and B in their tamperresistance. The adhesive used in Patches A and B is DuroTak1 87-4098, which is an acrylates copolymer containing vinyl acetate similar to the adhesive used in Patch C, DuroTak1 87-4287. 2.3. Evaluation of the resistance to tampering and in vivo drug availability of the geopolymer granules and commercial patches
2. Materials and methods 2.1. Materials Kaolin (Al2Si2O5(OH)4), fumed silica (SiO2, 7 nm particle size), reagent grade sodium hydroxide (NaOH), monopotassium phosphate (KH2PO4), 37% fuming hydrochloric acid (HCl) and 99.5% ethanol were purchased from Sigma–Aldrich. Fentanyl base (MacFarlan and Smith, Edinburgh, UK) was donated by Orexo AB, Sweden. DuroTak1 87-4098 were obtained from National Starch Chemicals (Bridgewater, U.S.). A commercially available fentanyl patch, Durogesic1 (Janssen-Cilag, Belgium) was used for comparison. 2.2. Synthesis The synthesis procedure of geopolymer drug carrier is described in detail in Jämstorp et al. (2010). The geopolymer precursor used in this study had the following composition: Si/Al 2.129, H2O /Al2O3 14.95 and Na2O/Al2O3 1.472 (in molar ratio). The cured geopolymer with 14.9 wt% of fentanyl was ground manually using a mortar and pestle and the residual particles were separated into two particle-size ranges: 315–710 mm and 50–100 mm, respectively, using Retsch1 sieves (F. Kurt Retsch GmbH, Germany). The geopolymer granules with particle size 50–100 mm were integrated into matrix layer to fabricate a transdermal patch intended to resistant against tampering. The drug containing geopolymer carriers with diameter between 50 and 100 mm or fentanyl base were dispersed in IPA (propan-2-ol) within 3 min before mixed with adhesive (DuroTak1 87-4098). The mixture was with the ratio of 1 g geopolymer or 149 mg of fentanyl base to 2.8 g of IPA and 9.98 g of adhesive. The vial contents were mixed via rotation for 3 min until the powder dispersion appeared homogenous. The mixture was immediately laminated on release liner at a 500 mm setting. The casted film was dried in steps: 15 min at room temperature, 5 min at 50 C and then 5 min at 90 C. The dried film was cooled and covered with a backing layer. The finished patches were stored in the ambient temperature with relative humidity <20% before testing. The drug concentration in
As a first step, a preliminary drug release study was performed in order to investigate the tamper resistance of geopolymer granules with the size range 315–710 mm. A single dose of Patch C (Durogesic1 12 mcg/h) was used as comparison to the geopolymer granules with the corresponding dose in terms of extractability in the selected media. The test was carried out in a USP dissolution apparatus II (Sotax AT7 Smart, Sotax AG, Switzerland) equipped with mini vessels and paddles. The measurements were performed using a paddle speed of 50 rpm at 37 C in 200 mL of phosphate buffer at pH 6.8 0.5 or in 200 mL of 50% v/v ethanol solution. The amount of fentanyl was analyzed using isocratic reversed-phase HPLC (Dionex Summit1, Dionex UK Ltd.) with a XTerra MS C18 column (2.1 mm ID 50 mm, 3.5 mm, Waters Corp., Milford, MA, USA). A PDA detector was utilized at a wavelength of 210 nm. The mobile phase was a mixture of acetonitrile and the solvent with 0.1 v/v% TFA, 72 v/v% water and 28 v/v% acetonitrile. The mobile phase was degassed inline at a flow rate 0.25 mL/min. As a second step, a pilot in vivo study on the plasma concentration response of one forth of Patch C (Durogesic1 12 mcg/h) and geopolymer granules (size range of 315–710 mm) with corresponding dose was investigated on male Sprague Dawley rats (Taconic, Denmark). The animals were housed in ventilated cabinet with 12 h lighting arrangement. A 5-week acclimatization period was allowed before the test commenced. The fur on the back of the rats was shaved and treated with depilatory agent the day before the experiment. The granules were attached to the skin on the back of the rats using a circular-shaped placebo elastic textile patch (Cederroth AB, Sweden) with 1.2 cm in diameter in order to hold the granules in place. Due to the size variance between rats, the applied dose was adjusted to around 1500 mg/kg for comparison. Blood samples were taken after 0.25, 0.5, 1, 2, 4, 6 and 24 h and stored at +4 C for maximum 30 min until centrifugation at 6000 rpm for 5 min. Plasma was stored initially at –20 C for 1–3 days and then at –80 C for 4 days until analyzed. Before analysis, the plasma samples were thawed and mixed on a shaker. A 50 mL of plasma was transferred into a microwell plate and precipitated with 200 mL of acetonitile. Internal standard solution was added into the plate and the mixture was mixed on a shaker for 30 min. The plasma samples were then centrifuged at
Fig. 1. Illustration of the geopolymer-integrated fentanyl transdermal patch.
104
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
4000 rpm for 15 min. The supernatant was transferred to another microwell plate and the acetonitile was evaporated under nitrogen at 45 C until approximately 50–100 mL was left. The residue was diluted in 50 mL acetonitile:water (50:50) with 0.1% acetic acid. The content was transferred to 96-deepwell plate and mixed on a shaker. The concentration of fentanyl in plasma was determined using reversed-phase UHPLC (Surveyor MS, Thermo Scientific, Bremen, Germany) with an Eclipse Plus C18 column (2.1 mm ID 50 mm, 3.5 mm, Agilent Technologies, Palo Alto, CA, USA) and Heated Electrospray Tandem Mass Spectrometry (HESI-MSMS, Thermo TSQ Quantum Ultra, Thermo Scientific, Bremen, Germany). The mobile phase was a mixture of 80 v/v% acetonitirle, water 20 v/v% and 0.1% acetic acid. The mobile phase was degassed inline at a flow rate 0.4 mL/min. All animal procedures were approved by the local ethical committee in Uppsala (permit number: C C279/10) and followed the guidelines of European Communities Council Directive (86/609/EEC). 2.4. Evaluation of the resistance of transdermal formulations against tampering The cross-section morphology of the house-made patches (Patches A and B) was obtained by the images from a Leo 1550 FEG scanning electron microscope (SEM, Zeiss, Germany). The patches were firstly coated with a thin layer of gold/palladium prior to observation to avoid the charging effect of the samples. One eighth of Patch C (Durogesic1 25 mcg/h) and Patches A and B with corresponding dose were evaluated by in vitro tampering tests aiming to evaluate the difficulty level of by passing or compromising controlled release mechanisms of the transdermal patch to achieve rapid or immediate drug release. These test conditions were selected and designed according to the previous published studies and regulations (Anonymous, 2010b, 2013; Goliber, 2005), which simulate some common abuse methods: extraction, chewing, smoking and swallowing (Butler et al., 2011; Katz et al., 2006). The patches were extracted under the following conditions in two-hour course: 40% ethanol aqueous solution (ambient temperature), distillated water (70 C) and HCl solution at pH 1 0.5 (ambient temperature). Drug release in the solvents with 40% ethanol and 70 C water simulated the extraction in the commercial alcohol drinks and in water at elevated temperature, respectively. The solution at pH 1 represented the low-pH solvent that can extract fentanyl effectively. All extractions were performed with the same drug-to-solvent ratio of 1 mg fentanyl to 19 mL of solvent. Milling the patches by a mortar and pestle for 10 min with 5 mL of phosphate buffer at pH 6.8 0.5 simulated chewing by the abusers. Smoking of the patch was simulated by heating the patch at 120 C for 20 min. The evaporated drug amount was determined by the difference of drug release amount from the patch before and after heating in 500 mL HCl solution at pH 1 0.5 by USP dissolution bath at 37 C and stirring rate at 50 rpm for 24 h. Oral indigestion is one of the most common and easiest ways for abusers. Fentanyl could be well extracted in stomach as it has higher solubility in low pH solutions (Roy and Flynn, 1989). The drug releases from patch in gastric and intestinal fluid conditions were simulated in a standardized USP apparatus II (Sotax AG, Switzerland) using 500 mL media at 37 C with stirring rate at 50 rpm. The patch sample containing approximate 2.1 mg fentanyl was fixed on a metal plate, which was then placed on the bottom of dissolution bath. Dissolution media were at pH 1 0.5 (0.1 M HCl), pH 6.8 0.5 (0.05 M KH2PO4, NaOH) or pH 1 0.5 with 40% ethanol to simulate the gastrointestinal condition or the gastric condition when patients digest the patch with alcohol drinks, respectively. Aliquots (2 mL) were withdrawn from the dissolution bath by a
sampling tube at the predetermined time intervals. Higuchi model was used to evaluate the release profiles and the fitting with R2 over 0.95 was considered as a good fitting. All tampering tests were conducted in triplicates and the concentration of fentanyl was measured using reverse-phase HPLC (Waters Corp., Milford, MA, USA) and a YMC-Triart C18 column (2.0 mm ID 12 mm, 3 mm, YMC, Japan). A PDA detector was utilized at a wavelength of 210 nm. The mobile phase was a mixture of acetonitrile and the solvent with 0.1 v/v% TFA, 72 v/v% water and 28 v/v% acetonitrile with a ratio of (=25:75, v/v). The mobile phase was degassed inline at a flow rate 0.3 mL/min. 3. Results and discussions 3.1. Evaluation of resistance to tampering and in vivo drug absorption from geopolymer granules The drug extraction from Patch C and geopolymer granules with particle size ranges of 315–710 mm was estimated in phosphate buffer and 50% ethanol solution at 37 C (Fig. 2). In both media, the drug releases from geopolymer granules with particle size range of 315–710 mm were slower than that from commercial patch: Patch C released all drug content within 1 h (Fig. 2). It suggested that geopolymer, as an inert and unswellable matrix, had better resistance to extraction than commercial patch. The results showed a slower fentanyl release in the buffer solution than in ethanol aqueous solution, which probably was due to the hydrophobicity and solubility of fentanyl (Roy and Flynn, 1988). The results of this in vitro drug release indicated the potential role of geopolymer material in reducing the risk of dose dumping. As a next step, the transdermal drug administration by the geopolymer granules and Patch C were evaluated in a pilot study on rats (Fig. 3). The plasma drug concentration vs. time profiles of the rats applied with the geopolymer granules were similar to those with the commercial patch. Although this is a pilot study and the number of animals is few, the results indicate that the geopolymer granules might be able to achieve similar bioavailability as the commercial patch. In general, both formulations resulted in a linear increase of plasma concentration in the first 6 h and maintained the concentration for 24 h. The responding plasma concentration had a larger deviation for the repeats using geopolymer granules than the ones using Patch C. We speculate that it could cause from different reasons e.g. inhomogeneous distribution of the granules and varied contact area of the attached granules with the skin. We believe that reducing the granule sizes could improve the consistency and adhesiveness of the patch, and
Fig. 2. In vitro fentanyl release from geopolymer granules and Patch C in pH 6.8 and 50 v/v% ethanol solution during the first 4 h.
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
105
Fig. 3. The absorption of fentanyl in geopolymer granules and commercial patch after applied to the shaved back of 5 rats. Plasma levels adjusted to dose around 1500 mg/kg.
thus the granules with smaller particle size, 50–100 mm, were integrated in the patch in the further study. 3.2. Evaluation of the resistance of transdermal formulations against tampering The cross-sections of both Patches A and B were observed under SEM. The images of Patch A showed that geopolymer granules were embedded in the adhesive layer (Fig. 4a and b). Some voids were found between granules and adhesives. The geopolymer particles showed similar microstructure as observations in the previous study (Jämstorp et al., 2010) (Fig. 4c). Tests simulating five common abuse methods were used to evaluate the tamper-resistance of Patches A, B and C (Fig. 5). Patch C represented a recognized standard dosage form, i.e. a commercially available patch. Patch B was the control sample, containing the same adhesive materials as Patch A. However, Patch B had a continuous matrix but Patch A had geopolymer particles replacing the space of some adhesives. Therefore, despite diffusing out from a thinner layer of adhesives, the drug molecules would experience other processes in Patch A than that in Patch B before release: i.e. the drug had to diffuse from geopolymer granules and thereafter a partition at the interface between geopolymer and adhesives occurred. Therefore, the difference of the extractability between
Fig. 5. (a) Fraction of released drug from Patches A, B and C after extraction tests. The error bars represent the confidence interval. (b) The in vitro drug release profile of the Patches A, B and C and the residual patches in pH1HCl solution after heating at 120 C for 20 min.
Patches A and B is believed to depend on the joint effect of the water and drug diffusion rate in geopolymer and adhesives and the partition rate at the interface under the chosen extraction condition. By extracting in heated water, Patch A showed a much better resistance than the other patches: Patch A released 12.5% while
Fig. 4. SEM images of the cross-sections of Patch A (a–c) and Patch B (d–f). The layers of release liner, matrix and backing layer are indicated by arrows in (a) and (d).
106
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
Patches B and C released 30% and 24.6%, respectively. Since the heated polymer in the patch matrix generally get more flexible and expanded than ceramics, the drug molecules could diffuse faster in the polymer matrix after heating. Consequently, the drug release from Patch B was influenced by the temperature to a greater extent than that of Patch A. It could also be applied to the results of simulated smoking: the release was 13.4% and 27.6% from Patches B and C, respectively, while only 2% of the drug released from Patch A. After the heating, the drug release profiles of residual patches were compared to the respectively non-tampered patches at pH1 (Fig. 5b). The drug release from the heated patches was all slower compared to the ones that had not been heated, indicating that heating influenced the drug release from the residual patches. Despite the release percentage, Patch C showed similar drug release profiles before and after heating, while Patches A and B had a slower release in the first 2–3 h after heating. In 40% ethanol solution, Patch A performed similarly as Patch B, while Patch C released the highest fraction of drug content. As discussed previously, it was probably due to the joint effect of the different rate in diffusion and partition experienced by the drug molecules in these patches. Moreover, Patches A and B showed similar resistance to the simulated chewing and the drug release was 5.8% and 4.4%, respectively. Patch C, however, released 40.5% of the drug content. In pH1 solution, Patch C, yet again, had the highest fraction of drug release, 40.4%. However, extraction at low pH, Patch A had a higher released drug fraction than Patch B: Patch A released 19% while the Patch B released 11%. The low pH solution opens up the geopolymer pores and increases the overall all drug diffusivity, resulting in a faster drug release from Patch A (Jämstorp et al., 2010). Polymers, such as methyl acrylate-methacrylic acid copolymers and methyl methacrylate-methacrylic acid copolymers, could be blended into the geopolymer to improve the resistance to acid solutions for future development. It is worth to notice that although Patch A released a slightly higher fraction in low pH solution than Patch B, the resistance of Patch A to tampering was better than Patch B in the other two extraction conditions and better than Patch C in all tested extractions. Fig. 6 shows the drug release profiles of the three patches in the dissolution media simulating stomach and intestinal pH. At both pH, the drug release from Patch C was fastest among the three patches, released over 90% of the drug content within 6 h. It indicates that Patch C had much less resistance to these in vitro conditions than the other two patches, which could lead to dose dumping and lethal side effects. The release was slightly faster at pH 6.8 than at pH 1 from Patch C, but it was the opposite for the
Table 1 The R2 of the linear approximation of the drug release profile of the Patches A, B and C to square root of time. Samples
R2
Patch Patch Patch Patch Patch Patch
0.993 0.999 0.999 0.981 0.996 0.979
A in pH1 B in pH1 C in pH1 A in pH6.8 B in pH6.8 C in pH6.8
other two patches. On the other hand, Patches A and B released fentanyl in a more controlled manner. Although Patch A released fentanyl slightly faster than Patch B in pH 1 solution, it released slightly slower in pH 6.8 solution. But both patches released similar amount of drug at the end of the measurements, i.e. 24 h. The dissolution results were different to what was observed in the extraction test at pH 1. It is possible that, even though the solution at low pH could open up the geopolymer pores, the large amount of dissolution buffer in the drug release test could increase the drug release in the polymer matrix as well. As a result, the overall effect of drug diffusion and partition in Patches A and B could be similar. The release profiles from all three patches in the tested dissolution media had the linear relationship to square root of time (Table 1), which could indicate that the releases were controlled by a diffusion-based release mechanism. However, all patches released much faster in pH1 with 40% ethanol: most of the drug content was released in 3 h, probably owing to the hydrophobicity of fentanyl. Patch A, again, showed a slightly better resistance to this media in the first hour, indicating it could have lower risk to cause dose-dumping if the patient digest the patch with alcohol drinks. An in vivo bioavailability study of Patch A could be the next step of this study. As reported previously, drug depot would form in the upper layer of skin in the first few hours after applying Patch C and the depot is responsible for the subsequent absorption (Margetts and Sawyer, 2007; Van Nimmen and Veulemans, 2007). The drug release might be controlled by the release rate from Patch A matrix to a greater extent. Therefore, the slower release of Patch A might provide a safer therapy regardless of the skin condition of the patient. With the increased usage of transdermal opioid formulations, the amount of the related abuse cases rises considerably as well. In this study, we showed a possibility to improve the resistance to tampering by integrating ceramic granules in the adhesive matrix of fentanyl transdermal patch. Incorporating geopolymer granules in the matrix could increase the difficulty to compromise the controlled drug release under the tested extraction conditions, especially at elevated temperature. Compared to the commercial patch, this novel patch could hinder the non-medical use of opioid transdermal patches and thus possibly alleviate the situation of opioid abuse. Further improvements on the resistance in low pH and ethanol solution could be explored by addition of polymer excipients that have low solubility at acidic and/or ethanol solution into the geopolymer (Jämstorp et al., 2012). However, the influence of these excipients to the efficiency of the transdermal delivery and skin irritation is pending to be evaluated. 4. Conclusion
Fig. 6. Drug release profile from the Patches A, B and C in pH1, pH6.8 and pH1 with 40% ethanol solution. The error bar represents the confidence interval.
As the abuse problems related to transdermal opioid formulation increases, new solution to increase resistance to tampering is needed. In this paper, a patch technology with drug loaded geopolymer granules incorporated into the adhesive matrix is presented. To our knowledge, this is the first attempt to blend
B. Cai et al. / International Journal of Pharmaceutics 488 (2015) 102–107
ceramic materials into the adhesive to improve the resistance to tampering. The geopolymer granules showed a better resistance to extraction and gave a similar in vivo response as a commercial patch. Bench testing showed that the geopolymer-integrated patch had better resistance to the extraction in some common solvents than the commercial fentanyl patch. However, the resistance to extraction of geopolymer-integrated patch in big volumes of alcohol containing solution was less pronounced. The novel patch could release drug in a more controlled manner at pH 1 and pH 6.8, while the commercial patch released most of the drug content within 6 h. With better resistance to tampering than the commercial patch, this novel patch could hinder the non-medical use of opioid patches and possible alleviate the abuse situation of opioids. Acknowledgements This work has received support from Orexo AB for kind contribution of the materials and Sweden’s Innovation Agency (VINNOVA) and the Swedish Research Council (2011-3399 and 2011-4444) for financial contributions. Karin Söderkvist, Anders Sågström, Cecilia Coupland, Sari Öbrink, Ulrica Roos are acknowledged for some of the experimental works. References Anonymous, 2010a. The DAWN Report: Trends in Emergency Department Visits Involving Nonmedical Use of Narcotic Pain Relievers. Drug Abuse Warning Network (DAWN), Substance Abuse and Mental Health Services Administration, Office of Applied Studies, Rockville. Anonymous, 2010b. Guidance for Industry Assessment of Abuse Potential of Drugs. In: F.a.D.A., U.S. Department of Health and Human Services (Ed.), Center for Drug Evaluation and Research (CDER), Rockville. Anonymous, 2013. Guidance for Industry Abuse-Deterrent Opioids – Evaluation and Labeling, Draft Guidance. U.S. Department of Health and Human Services; Food and Drug Administration; Center for Drug Evaluation and Research (CDER). Butler, S.F., Black, R.A., Cassidy, T.A., Dailey, T.M., Budman, S.H., 2011. Abuse risks and routes of administration of different prescription opioid compounds and formulations. Harm Reduct. J. 8, 29. Cai, B., Engqvist, H., Bredenberg, S., 2014. Evaluation of the resistance of a geopolymer-based drug delivery system to tampering. Int. J. Pharm. 465, 169–174. Carson, H.J., Knight, L.D., Dudley, M.H., Garg, U., 2010. A fatality involving an unusual route of fentanyl delivery: chewing and aspirating the transdermal patch. Legal Med. (Tokyo, Japan) 12, 157–159.
107
Goliber, P.A., 2005. Benchtop Evaluations of Tampering with Pharmaceutical Dosage Forms. Opioid abuse resistance conference. Hart, J.R., Anderson, S.C., Pohl, M.A., Bosl, E.R., 2007. Abuse-resistant transdermal dosage form, US 7182955 B2. Howard, S.A., Reidenberg, B., 2004. Tamper resistant transdermal dosage form, US 20050002997 A1. Howard, S.A., Reidenberg, B., 2005. Tamper-resistant transdermal dosage form comprising an active agent component and an adverse agent component at the distal site of the active agent layer, WO 2004098567 A3. Jämstorp, E., Forsgren, J., Bredenberg, S., Engqvist, H., Strømme, M., 2010. Mechanically strong geopolymers offer new possibilities in treatment of chronic pain. J. Controlled Release 146, 370–377. Jämstorp, E., Strømme, M., Frenning, G., 2011. Modeling structure–function relationships for diffusive drug transport in inert porous geopolymer matrices. J. Pharm. Sci. 100, 4338–4348. Jämstorp, E., Yarra, T., Cai, B., Engqvist, H., Bredenberg, S., Strømme, M., 2012. Polymer excipients enable sustained drug release in low pH from mechanically strong inorganic geopolymers. Results Pharm. Sci. 2, 23–28. Katz, N.P., Buse, D.C., Budman, S.H., Wing Venuti, S., Fernandez, K.C., Benoit, C., Bianchi, R., Cooper, D., Jasinski, D.R., Smith, D.E., Butler, S.F., 2006. Development and preliminary experience with an ease of extractability rating system for prescription opioids. Drug Dev. Ind. Pharm. 32, 727–746. Kugelmann, H., Bartholomaeus, J., 2006. Abuse-resistant transdermal system, US20070065365 A1. Margetts, L., Sawyer, R., 2007. Transdermal drug delivery: principles and opioid therapy. Contin. Educ. Anaesth. Crit. Care Pain 7, 171–176. Mastropietro, D.J., Omidian, H., 2013. Current approaches in tamper-resistant and abuse-deterrent formulations. Drug Dev. Ind. Pharm. 39, 611–624. Moon, J.M., Chun, B.J., 2011. Fentanyl intoxication caused by abuse of transdermal fentanyl. J. Emerg. Med. 40, 37–40. Prosser, J.M., Jones, B.E., Nelson, L., 2010. Complications of oral exposure to fentanyl transdermal delivery system patches. J. Med. Toxicol.: Off. J. Am. Coll. Med. Toxicol. 6, 443–447. Roy, S.D., Flynn, G.L., 1988. Solubility and related physicochemical properties of narcotic analgesics. Pharm. Res. 5, 580–586. Roy, S., Flynn, G., 1989. Solubility behavior of narcotic analgesics in aqueous media: solubilities and dissociation constants of morphine, fentanyl, and sufentanil. Pharm. Res. 6, 147–151. Stinchcomb, A.L., Golinski, M.J., Hammell, D.C., Howard, J.L., 2009. Transdermally deliverable opioid prodrugs, abuse-resistant compositions and methods of using opioid prodrugs, US 20090156814 A1. Tavares, L., Reidenberg, B.E., Sackler, R.S., Wright, C., Alfonso, M.A., Oshlack, B., Cassidy, J.P., Carpanzano, A.E., Gullapalli, R.P., Shevchuk, I., 2011. Abuse resistant opioid containing transdermal systems, EP Patent 2062573. Van Nimmen, N.F., Veulemans, H.A., 2007. Validated GC–MS analysis for the determination of residual fentanyl in applied Durogesic reservoir and Durogesic D-Trans matrix transdermal fentanyl patches. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 846, 264–272. Woodall, K.L., Martin, T.L., McLellan, B.A., 2008. Oral abuse of fentanyl patches (Duragesic): seven case reports. J. Forensic Sci. 53, 222–225.