International Journal of Pharmaceutics 478 (2015) 540–552
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Pharmaceutical nanotechnology
Formulation, optimization and in vitro–in vivo evaluation of febuxostat nanosuspension Bhupesh K. Ahuja 1, Sunil K. Jena 1, Sharan K. Paidi, Surbhi Bagri, Sarasija Suresh * Department of Pharmaceutical Technology (Formulations), National Institute of Pharmaceutical Education and Research, S.A.S. Nagar, Punjab 160062, India
A R T I C L E I N F O
A B S T R A C T
Article history: Received 8 October 2014 Received in revised form 29 November 2014 Accepted 5 December 2014 Available online 6 December 2014
The purpose of the present study was to develop febuxostat nanosuspension and investigate its effect on febuxostat solubility, dissolution rate and oral bioavailability. The wet media milling technique was adopted with a combination of hydroxypropyl methylcellulose (HPMC E3) and D-a-tocopherol polyethylene glycol 1000 succinate (TPGS) as surface stabilizers for the generation of nanocrystals. Rotatable central composite design (CCD) was selected for nanosuspension optimization. The critical parameters were bead volume, milling time, polymer and surfactant concentrations; whereas particle size, polydispersity index (PDI) and zeta potential were taken as responses. The presence of crystallinity was confirmed by differential scanning calorimetry and powder X-ray diffraction. Scanning electron microscopy and transmission electron microscopy revealed small and uniform plate like morphology. A significant increase was observed in saturation solubility and dissolution rate of the optimized nanosuspension in all the pH conditions tested. Oral bioavailability of FXT and optimized FNC was evaluated in SD rats. The nanosuspension exhibited enhanced Cmax (26.48 2.71 vs. 19.85 2.96 mg/mL) and AUC01 (222.29 9.81 vs. 100.32 9.36 mg h/mL) with a 221.6% increase in relative bioavailability. Thus, FNC is a viable approach to enhance the bioavailability of FXT, a BCS Class II drug. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Febuxostat Wet media milling Nanosuspension Saturation solubility Stability Oral bioavailability
1. Introduction The large number of lead compounds springing up from drug discovery programs, exhibits poor water solubility (Kayaert et al., 2010). This phenomenon is a consequence of the new paradigm in high-throughput screening, which is based on good affinity and specific binding with target receptor. Candidates who emerge from these screening studies frequently have high log P value and molecular weight; both these factors contribute to poor water solubility (Rabinow, 2004). Such molecules are of two types, nonlipophilic hydrophobic drugs known as ‘brick dust’ or lipophilic hydrophobic drugs known as ‘grease balls’ (Müllertz et al., 2010). According to the biopharmaceutics classification system (BCS), drugs having low solubility and high permeability are classified as class II drugs (Frick et al., 1998). Numerous approaches have been adopted to increase the solubility of these molecules, including use of cosolvents (Miyako et al., 2010), surfactants (Torchilin, 2001; Kawakami et al., 2006), lipid based formulations (Chakraborty
* Corresponding author. Tel.: +91 172 2292055/9888069557; fax: +91 172 2214692. E-mail address: [email protected] (S. Suresh). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.ijpharm.2014.12.003 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
et al., 2009), inclusion complexation with cyclodextrin (Zirar et al., 2008), amorphous solid dispersions (Joshi et al., 2004) and nanotechnology (Ravichandran, 2009). However, complexation with cyclodextrin is limited by a suitable molecular size requirement while the amorphous systems are restricted due to their physical and chemical stability issues. When it comes to drugs which are insoluble in both aqueous and organic media (brick dust drugs), these approaches are often ineffective (Gao et al., 2012). Nanotechnology, in particular, nanocrystal (particle size reduction) technology can be adopted for enhancement of dissolution and saturation solubility of these drugs and ultimately, their bioavailability (Müller et al., 2001). Nanocrystals are crystalline nanoparticles with size ranging from 200 to 500 nm stabilized by surface stabilizers (MeriskoLiversidge et al., 2003). They increase the saturation solubility, dissolution rate and probably the mucoadhesion resulting in improved oral bioavailability of drugs exhibiting dissolution rate dependent bioavailability (Jacobs and Müller, 2002; Ponchel et al., 1997). In addition, nanocrystals exhibit the advantages of high drug loading, avoidance of organic solvents, enhanced stability and less toxicity, in comparison to liposomes, polymer nanoparticles, lipid nanoparticles and formulations with cosolvents (Kim et al., 2001; George and Ghosh, 2013). Nanosuspension formulations of several drugs are already marketed, exemplified by Rapamune1
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
(sirolimus), Emend1 (aprepitant), Tricor1 (fenofibrate), Cesamet1 (nabilone), Megace1 ES (megesterol acetate) and Invega Sustenna1 (paliperidone palmitate) (Junghanns Jens-Uwe and Müller, 2008). Quality by design (QbD) is a systematic, scientific, and risk based proactive approach to pharmaceutical development, which comprises of designing and developing formulations, and manufacturing processes with predefined product specifications (Lawrence, 2008). QbD encompasses the application of tools such as: design of experiments (DoE), risk assessment and process analytical technology (PAT) for the development of pharmaceuticals (Verma et al., 2009). Design of experiments (DoE) is an important tool of QbD, which allows understanding of the influence of formulation and process variables on the product quality by defining a ‘design space’ (DS) (Savic et al., 2012). DS provides the flexibility of operating within that space without further regulatory approvals. Febuxostat (FXT) (Fig. 1) is a novel, potent, non-purine, selective xanthine oxidase inhibitor (Maddileti et al., 2013). It is a weak acid (pKa 3.08) that is practically insoluble in water (solubility of 5 mg/ mL at 37 C was observed in our study). Its oral bioavailability is hampered by low aqueous solubility and its vulnerability to enzymatic degradation in both intestine and liver (Ernst and Fravel, 2009). Moreover, the presence of food decreases the maximum concentration of febuxostat in plasma (Cmax) by 38–49% (Khosravan et al., 2008). Among the methods employed for nanocrystal generation, wet media milling (Nanocrystal1) classified as a top–down approach, is a simple, effective and industrially applicable (Liversidge and Liversidge, 2011). Mechanical attrition of drug with milling media and impact with the walls of milling chamber results in the production of nanocrystals of desired size (Van Eerdenbrugh et al., 2008a,b,b; Liu et al., 2011). The objective of the present study was to formulate febuxostat nanocrystals (FNC) in order to improve its oral bioavailability by increasing its solubility and dissolution rate. FNC formulation was prepared by wet media milling, and the formulation and process were optimized by employing four factors and three level rotatable central composite design (CCD). The selected responses comprised of particle size, polydispersity index (PDI) and zeta potential. The
541
CCD is one of the most efficient designs of response surface experimental methodology to study the effects of process and formulation components on responses for exploring quadratic response surfaces and the second-order polynomial model. Data analysis using ANOVA and multifactor analysis is performed to assist in elucidating the interactions between different variables and provide a predictive model for the process. The physicochemical characterization of the optimized batch was systematically investigated by differential scanning calorimetry (DSC), powder X-ray diffraction (PXRD), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Further, saturation solubility, dissolution rate and relative oral bioavailability of FNC were investigated. 2. Theoretical prospective 2.1. Biopharmaceutical characterization Solubility, dissolution rate and permeability of the drug through the gastrointestinal membrane are the decisive parameters limiting oral drug absorption. As per biopharmaceutics classification system (BCS), these key parameters are characterized by three dimensionless numbers, i.e., dissolution number (Dn), absorption number (An), and dose number (Do) (Oh et al., 1995). Both physicochemical and physiological parameters of the drug were taken into account in calculating these dimensionless numbers. In addition, calculated dissolution time (Tdisso) and dose absorbable (Dabs), are also applied to define rate limiting step for oral absorption of drugs (Oh et al., 1995; Löbenberg and Amidon, 2000). 3. Materials and methods 3.1. Materials Febuxostat (FXT) was generously provided by Kemwell Biopharma Pvt., Ltd. (Bangalore, India) as a gift sample. Hydroxypropyl methylcellulose E3 Premium LV (HPMC E3) and E5 Premium LV (HPMC E5) were kindly gifted by Colorcon Asia Pvt., Ltd. (Goa, India). Phospholipon1 80H (P 80H) was generously provided by Lipoid (Ludwigshafen, Germany). D-a-Tocopherol polyethylene glycol 1000 succinate (TPGS) was purchased from Sigma–Aldrich (Bangalore, India). Poloxamer 188 (P 188) and poloxamer 407 (P407) were purchased from Signet Chemicals Corporation Pvt., Ltd. (Mumbai, India). Sodium lauryl sulphate (SLS) was purchased from Sisco Research Laboratories Pvt., Ltd. (Mumbai, India). Polyvinyl pyrolidone K30 (PVP K30) was procured from All Well Pharmaceuticals (Chandigarh, India). Potassium dihydrogen phosphate was purchased from Merck Specialities Pvt., Ltd. (Mumbai, India). Methanol (HPLC grade) was purchased from Sigma–Aldrich (Bangalore, India). All other chemicals and reagents were of analytical or chromatographic grade and used without further purification. Purified Milli-Q water (Millipore, Billerica, MA, USA), degassed and filtered through 0.45 mm hydrophilic PVDF filters (Millipore Millex-HV), was used in all experiments. 3.2. HPLC analysis of febuxostat
Fig. 1. Chemical structure of FXT.
HPLC method was used for the estimation of FXT in all samples, including the solubility and dissolution experiments. The HPLC system was comprised of a Waters 2695 separation module equipped with a quaternary pump, an auto sampler unit, and a Waters 2996 photodiode array (PDA) detector. ZORBAX XDB-C18 (5 mm; 250 mm 4.6 mm) (Agilent Technologies, Inc., USA) analytical column was used for the estimation. The mobile phase consisted of methanol (0.7% v/v triethylamine) and phosphate buffer (10 mM, pH 2.7) in 60:40 (% v/v) proportions. The flow rate
542
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
was maintained at 1 mL/min and the PDA detector was set at 315 nm. The Empower HPLC software was used for the analysis of the results. The procedure was performed in triplicate, and the average and standard deviations were calculated. 3.3. Generation of febuxostat nanocrystals (FNC) The FNC was generated by wet media milling method. Briefly, 40 mg (2% w/v) of FXT was dispersed in an aqueous solution containing 0.5% w/v primary stabilizer and 0.1% w/v secondary stabilizer. The resulting suspension was poured into a glass vial containing 4 mL of yttrium stabilized zirconium oxide beads (0.3–0.4 mm) and stirred on a magnetic stirrer (IKA1 RH basic 1, Bangalore, India) at 1500 rpm for 1 h at room temperature. 3.4. Particle size distribution FNC was evaluated for mean particle size and polydispersity index (PDI) by dynamic light scattering (DLS) using the Zetasizer (Nano ZS, Malvern Instruments, UK) at a detection angle of 173 at 25 C. Samples were suitably diluted with de-ionized water before analysis. Each sample was analyzed in triplicate. 3.5. Screening of surface stabilizers 3.5.1. Based on particle size distribution Surface stabilizers were selected by determining the particle size distribution of the resultant FNC. In this study, HPMC E3, HPMC E5 and PVP K30 as the primary stabilizers alone or in combination with Poloxamers 188 and 407, TPGS, SLS and Phospholipon1 80H as the secondary stabilizers were evaluated. The most suitable stabilizers were identified by optimum particle size and PDI. In these experiments, the drug concentration (2% w/v), primary stabilizer concentration (0.5% w/v), secondary stabilizer concentration (0.1% w/v), distilled water (2 mL), stirring rate (1500 rpm), bead volume (4 mL) and milling time (60 min) were kept constant. Each experiment was performed in triplicate. 3.5.2. Based on solubility study Solubility of FXT was determined in solutions of the surface stabilizers that resulted in particle size less than 300 nm. Briefly, excess amount of FXT was added to stabilizer solutions (5 mL) in sealed glass vials. The vials were agitated at 100 rpm for 72 h in the shaker water bath (EQUITRON, Medica Instrument Mfg., Co., India) at 37 C and centrifuged (OptimaTM LE-80K, Beckman Coulter, USA) at 40,000 rpm for 10 min at 4 C to remove the undissolved drug. The supernatant was suitably diluted with methanol and 10 mL aliquot was analyzed for FXT by HPLC at 315 nm. Each experiment was performed in triplicate. 3.6. Experimental design Initial screening studies were carried out for evaluating the effect of process parameters and formulation parameters on FNC formulation and its stability. The bead volume and milling time were identified as critical process parameters and polymer and surfactant concentrations as critical formulation parameters. The design of experiment (DoE) was employed systematically to evaluate and optimize the selected process and formulation parameters at three levels (1, 0, +1). Based on the number of factors and their levels, a rotatable central composite design (CCD) was selected to investigate their effects on critical quality attributes (CQAs) of nanosuspension. The batch size (2 mL), drug concentration (2% w/v), HPMC E3 as a primary stabilizer and TPGS as a secondary stabilizer, milling speed and solvent system (water) were kept constant in the experimental trials. Independent factors
Table 1 Independent factor and their coded levels of central composite design. Independent factors
Design level
Actual parameters
Coded
Bead volume (mL)
A
Milling time (min)
B
Polymer concentration (% w/v)
C
0.5 0.75 1.0
1 0 +1
Surfactant concentration (% w/v)
D
0.1 0.2 0.3
1 0 +1
Actual value 3.0 3.5 4.0 60 75 90
Coded level 1 0 +1 1 0 +1
and their levels used in this study are shown in Table 1. The design contains 30 experimental runs; i.e., sixteen (24) factorial points, eight (2 4) axial points and six centre points were generated as shown in Table 2 and analyzed by the statistical software package Design-Expert1 8.0.7.1 (Stat-Ease Inc., USA). 3.7. Preparation of physical mixture of FXT, HPMC E3 and TPGS (PM) PM of FXT, HPMC E3 and TPGS was prepared in same proportion as used for the preparation of optimized FNC. Briefly, FXT (2% w/v), HPMC E3 (0.5% w/v) and TPGS (0.14% w/v) were poured in a mortar and mixed for 10 min until a homogenous mixture was obtained. The mixture was passed through 40# mesh and stored in a desiccator. 3.8. Characterization of optimized febuxostat nanocrystals 3.8.1. Particle size distribution and zeta potential FNC mean particle size and PDI were determined as per the method described in Section 3.2.2. The surface charge of the particles was assessed by measurement of zeta potential in a Zetasizer (Nano ZS, Malvern Instruments, UK) at 25 C by applying fields of 20 V/cm. Samples were suitably diluted with de-ionized water before analysis. Each sample was analyzed in triplicate. 3.8.2. Surface morphology 3.8.2.1. Scanning electron microscopy (SEM). The morphology of FXT and optimized FNC was studied by scanning electron microscope (S-3400N, Hitachi Ltd., Tokyo, Japan) at an acceleration voltage of 10 kV. Samples were sputter coated (E-1010, Hitachi Ltd., Tokyo, Japan) with gold-palladium and observed at different magnifications. 3.8.2.2. Transmission electron microscopy (TEM). TEM was employed to study the morphology of FXT nanosuspension. FNC was diluted with deionized water (200 dilution 100 mg/mL) and visualized by TEM (FEI Tecnai G2F20, Netherlands). Briefly, a drop of the dispersion was placed on the carbon-coated copper grids and stained with phosphotungstic acid (1.5% w/v), and the excess was drained off with filter paper. The grid was air dried, and viewed under the TEM at different magnifications. 3.8.3. Differential scanning calorimetry (DSC) The DSC thermograms of FXT, HPMC E3, TPGS, PM and optimized FNC were recorded on a DSC (Mettler Toledo, DSC 821e, Switzerland) equipped with STARe SW 9.01 software. Each
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
543
Table 2 Effect of selected process and formulation parameters on FNC particle size distribution and zeta potential. Run
Critical factors Bead volume (mL)
Milling time (min)
Polymer concentration (% w/v)
Surfactant concentration (% w/v)
Particle size (nm)
PDI
Zeta potential (mV)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
3 3.5 3 4 3.5 3 4 3.5 4 3.5 3.5 4 3 3.5 4.5 4 4 3.5 3 3.5 3 3 3.5 2.5 3.5 4 3 3.5 4 3.5
90 75 90 60 75 60 90 75 90 75 75 90 60 75 75 60 90 75 90 75 60 90 75 75 75 60 60 45 60 105
1.00 0.75 0.50 1.00 0.75 1.00 0.50 0.75 1.00 1.50 0.75 0.50 0.50 0.00 0.75 0.50 1.00 0.75 1.00 0.75 0.50 0.50 0.75 0.75 0.75 0.50 1.00 0.75 1.00 0.75
0.1 0.2 0.3 0.1 0.0 0.1 0.3 0.2 0.3 0.2 0.2 0.1 0.1 0.2 0.2 0.3 0.1 0.2 0.3 0.2 0.3 0.1 0.2 0.2 0.4 0.1 0.3 0.2 0.3 0.2
279.5 265.2 288.6 289.1 303.5 324.2 282.3 267.7 302.4 334.9 269.4 258.7 304.6 281.0 279.7 298.3 266.7 276.9 309.3 269.4 335.7 264.4 272.6 414.3 335.6 250.0 359.9 317.4 309.1 293.7
0.134 0.129 0.156 0.139 0.166 0.162 0.112 0.133 0.107 0.160 0.126 0.089 0.149 0.143 0.109 0.132 0.097 0.134 0.154 0.128 0.217 0.131 0.134 0.206 0.182 0.085 0.278 0.217 0.162 0.112
18.5 19.5 20.2 18.2 15.1 17.9 20.1 18.8 19.1 18.7 19.3 18.8 19.1 20.2 19.3 19.5 17.3 19.4 20.4 19 20 19 19.2 19.5 21.2 19.4 19.7 19.6 20 19.2
Responses
sample (2–4 mg) was heated in an aluminium pan at a scanning rate of 10 C/min in an atmosphere of nitrogen gas (40 mL/min) in the range of 25–240 C. The DSC was calibrated for baseline with empty pans, and for temperature and enthalpy with indium. 3.8.4. X-ray powder diffraction (PXRD) analysis The PXRD patterns of FXT, HPMC E3, TPGS, PM and optimized FNC were recorded on X-ray diffractometer (Bruker D8 Advance Diffractometer, Germany). The X-ray source was a Cu Ka 1 tube (wavelength 1.5406 Å) operated at 40 kV and 40 mA. The samples were scanned from 3 to 40 2u at a scan rate of 0.1 2u min1. 3.8.5. Saturation solubility study Saturation solubility of FXT, PM and optimized FNC were determined individually in three different buffers. Briefly, excess of each FXT, PM and optimized FNC was added to each 5 mL of HCl (pH 1.2), phthalate buffer (pH 4.5) and phosphate buffer (pH 6.8) in sealed glass vials at 37 C. The vials were agitated at 100 rpm for 72 h in the shaker water bath (EQUITRON, Medica Instrument Mfg., Co., India). Then the dispersion was centrifuged (OptimaTM LE-80K, Beckman Coulter, USA) at 40,000 rpm for 10 min at 4 C. The supernatant was suitably diluted with methanol and quantified by HPLC at 315 nm. Each experiment was performed in triplicate. 3.8.6. Drug content and in vitro drug release study The amount of FXT in the optimized FNC was determined by HPLC (Waters 2695 Separations Module). Briefly, FNC equivalent to 10 mg of FXT was dissolved in methanol and sonicated for 10 min before analysis. Dissolution of FXT, PM and FNC were studied in USP 37 type II automated dissolution test apparatus (ElectrolabTDT-08L, India) in three different media viz HCl (pH 1.2), phthalate buffer (pH 4.5) and phosphate buffer (pH 6.8). The samples, equivalent to 40 mg of FXT, were placed in the dissolution vessel containing 900 mL buffer maintained at 37 0.5 C and stirred at 75 rpm. Samples were
collected at 5, 15, 30 and 60 min and replaced with a fresh dissolution medium to maintain the sink condition. The samples were filtered (0.45 mm PVDF syringe filter, Millipore Millex-HV) and centrifuged (OptimaTM LE-80K, Beckman Coulter, USA) at 40,000 rpm for 10 min at 4 C to remove the undissolved drug. The supernatant was injected into the HPLC system for the estimation of FXT. 3.9. Stability study of optimized FNC The physical stability of optimized FNC was evaluated at 4 C and 25 C for a period of three months (Kakran et al., 2012). Aliquots of the FNC were periodically withdrawn, diluted and analyzed for particle size distribution and zeta potential by Zetasizer (Nano ZS, Malvern Instruments, UK). Each sample was analyzed in triplicate. 3.10. In vivo oral pharmacokinetic study 3.10.1. Animals and dosing Female Sprague Dawley (SD) rats of 180–200 g were obtained from the central animal facility (CAF), NIPER, S.A.S. Nagar, India. The study protocol was approved by the Institutional Animal Ethics Committee (IAEC), National Institute of Pharmaceutical Education and Research (NIPER), S.A.S Nagar, India. The animals were maintained at 25 2 C and 50–60% relative humidity (RH) under 12 h light/dark cycles for seven days before the experiment. They were randomly distributed into two groups, each containing five animals. Before the study, both groups were fasted overnight for 12 h with free access to water. One group was administered FNC orally, dose equivalent to 10 mg/kg body weight of FXT while the other group was administered FXT suspended in water, at a constant dose. Blood samples were collected from the retro-orbital plexus under mild anaesthesia into micro centrifuge tubes containing
544
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
heparin (40 IU/mL blood) at 0.25, 0.5, 1, 2, 4, 6, 8, 12 and 24 h time points. Plasma was separated by centrifuging the blood samples at 12,000 rpm for 10 min at 4 C and stored at 20 C prior to analysis. 3.10.2. Quantification of FXT in plasma samples FXT was quantified by the HPLC (Waters 2487). Briefly, an aliquot (100 mL) of plasma was mixed with 50 mL of internal standard solution (telmisartan 5 mg/mL) and 50 mL of drug solution. After vortexing for 1 min, chilled methanol and acetonitrile in 1:1 proportion (300 mL) was added as a protein precipitating agent and vortexed for 3 min. The mixture was centrifuged at 12,000 rpm for 10 min. The supernatant was filtered (0.45 mm PVDF syringe filter, Millipore Millex-HV) and the filtrate (80 mL) was analyzed by HPLC system. ZORBAX XDB-C18 (5 mm; 250 mm 4.6 mm) (Agilent Technologies, Inc., USA) analytical column was employed for the estimation. The mobile phase consisted of acetonitrile and ammonium acetate buffer (15 mM, pH 6.0) in 44:56 (% v/v) proportions, maintained at 1 mL/min flow rate. FXT was quantified at 315 nm using a UV detector (Waters 2487 dual l absorbance detector). Calibration curve was obtained from 5 to 5000 ng/mL with r2 value of 0.998. 3.10.3. Pharmacokinetic data analysis The pharmacokinetic parameters of FNC were calculated by Kinetica software version 5.0 (Thermo Fisher Scientific Inc., USA), and compared with pure FXT. Maximum concentration (Cmax) and time to reach maximum concentration (Tmax) are the values obtained directly from concentration–time curve. Area under the curve (AUC024 h or AUC01), elimination half life (t1/2), mean residence time (MRT) was determined. The relative bioavailability (F) of the FNC to the FXT was calculated by the following equation: AUCtest (1) 100 Relative bioavailabilityðFÞ ¼ AUCreference
3.11. Statistical analysis The means and standard deviation of all values were calculated. Statistical analysis was performed by paired ‘t’ test using Microcal (TM) Origin1 software version 6.0 (Microcal software, Inc., Northampton, USA). Significant difference was regarded as p < 0.05.
dissolution rate limited absorption owing to the higher Tdisso value than small intestinal transit time (Tsi) and lower Dabs than the dose of FXT. The Tdisso value was decreased considerably for the nanosized FXT indicating that the nanonization of FXT has the potential to improve its dissolution limited absorption. 4.2. Screening of surface stabilizers Despite all the advantages associated with nanocrystals, they exhibit thermodynamic instability. This could be explained by the fact that as the particle size decreases, the surface area of the particle increases, which leads to an increase in the Gibbs-free energy. System with such a high free energy is thermodynamically unstable. In order to reduce this free energy, agglomeration of smaller particles occurs (Lindfors et al., 2006; Wu et al., 2011). Surface stabilizers are added to reduce the Gibbs-free energy of the system and hence the aggregation. They act by inhibiting crystal growth and particle aggregation through the mechanism of electrostatic and steric stabilization. The efficiency of a stabilizer depends on its potential for interaction with the drug molecule (Ghosh et al., 2012). 4.2.1. Based on particle size distribution Fig. 2 depicts the particle size distribution of FNC generated by different surface stabilizers. Generation of FNC with primary stabilizers resulted in higher particle size and less stability when stored at room temperature for seven days. This might be due to an inadequate amount of the surface stabilizer necessary for complete coverage of the drug particle surface which is required to provide steric repulsion between the nanoparticles in the suspension (Ghosh et al., 2012). Conversely, an excess of the stabilizer like HPMC E3 could lead to an increase in viscosity, resulting in a decrease in milling efficiency (Singare et al., 2010). However, the minimum particle size distribution was obtained with HPMC E3 and TPGS combination and the particles remained stable even after seven days when stored at room temperature. Adsorption of HPMC E3 onto the drug crystals due to the interaction of its hydrophobic (methoxyl) group with the drug molecule provides better surface coverage (Ghosh et al., 2011). On the other hand, the hydrophobic portion of the TPGS may interact with a hydrophobic portion of FXT, thus stabilizing the FNC. The unique properties of the TPGS as solubilizer, permeability enhancer and stabilizer lead to its selection in the nanosuspension formulation (Ghosh et al., 2012,b; Van Eerdenbrugh et al., 2008a,b).
4. Results and discussions 4.1. Biopharmaceutical characterization Theoretical calculations of An, Dn and Do of FXT indicates that it belongs to BCS class II (An > 1, Dn < 1, Do > 1) drugs having both solubility and dissolution rate limited oral bioavailability. Table 3 lists the biopharmaceutical parameters that help in estimating the absorption rate limiting factors for the drug. Calculated Tdisso and Dabs values also showed that FXT exhibits both solubility and
4.2.2. Based on solubility study Fig. 3 presents the apparent solubility profiles of FXT in presence of different stabilizers. FXT exhibits higher solubility in SLS containing surface stabilizers. This could be due to the increased solubility of FXT in SLS micelle (Verma et al., 2011). The increased solubility can lead to Ostwald ripening, resulting in an increased particle size during the storage. In contrast, minimum solubility of FXT was observed in the presence of HPMC E3 and TPGS. Hence, the combination of HPMC E3 and TPGS was selected
Table 3 Biopharmaceutical parameters of FXT. Parameters
Values
Comments
Tdisso unmicronized (min) Tdisso micronized (min) Tdisso nanosized (min) Dose absorbable (Dabs) (mg) Absorption number (An) Dose number (Do) Dissolution number (Dn)
3465 173 8.7 23.48 3.91 64 0.03
Tdisso 199 Tdisso 50 Tdisso 50 Dabs dose (80) Permeability is not a problem for bioavailability (An > 1) Solubility limited bioavailability (Do > 1) Dissolution limited bioavailability (Dn < 1)
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
545
Fig. 2. The effect of surface stabilizers on FNC particle size and PDI.
as the surface stabilizer. Further, the less stability of the ionic stabilizers in the gastrointestinal fluid favors the use of steric stabilizers (Singare et al., 2010; Peltonen and Hirvonen, 2010). 4.3. Optimization of FNC 4.3.1. Experimental design The regression analysis of the data obtained from the experimental runs generated the following polynomial equations in which the model F ratios were statistically significant at a < 0.05 with Adj-R2 value in the range of 0.8–1 with a statistically non-significant lack of fit at a > 0.05. Particle size ¼ 270:20 9:26 A 12:19 B þ 12:59 C þ 16:39 D þ 8:76 AB þ 2:19 AC þ 0:8 AD 2:9 BC 2:99 BD þ 0:67 CD þ 4:31 A2 þ 7:11 B2 þ 8:96 C 2 þ 7:20 D2
(2)
PDI ¼ 0:13 0:018 A 0:026 B þ 0:0058 C þ 0:024 D 0:0017 AB þ 0:0023 AC 0:000046 AD 0:0062 BC 0:014 BD 0:0018 CD þ 0:0023 A2 þ 0:0085 B2 þ 0:0053 C 2 þ 0:0013 D2
(3)
Zeta potential ¼ þ19:31 0:053 A 0:11 B 0:39 C þ 0:81 D 0:30 AB 0:18 AC þ 0:046 AD þ 0:054 BC þ 0:004 BD þ 0:15 CD
(4)
where A,B, C and D are bead volume, milling time, percentage concentrations of polymer and surfactant, respectively. Table 4 presents the results of statistical analysis, which suggests a less than 0.01% chance that the model F value of the model Eqs. (2)–(4) occurred due to noise. The p values of the lack of fit test of the three models were not significant, indicating that these model equations fitted the data well. Therefore, the two quadratic models (particle size, PDI) and a two-factor interaction (2FI) model (zeta potential) could describe adequately the data and could be employed to navigate the design space. A positive sign represents a synergistic effect, while a negative sign indicates an antagonistic effect. In Eqs. (2) and (3), the negative coefficient of A and B in the model refer to a decrease in particle size and PDI with increasing bead volume and milling time, respectively. The positive coefficient of C and D indicates an increase in particle size with increasing amount of polymer and surfactant, respectively. In case of zeta potential, the negative coefficients of A–C in the model refer to decrease in zeta potential with increasing bead volume, milling time and percent concentration of polymer, respectively. The positive coefficient
Table 4 Results of statistical analysis of the experimental design. Responses
Particle size PDI Zeta potential Fig. 3. Solubility of FXT in different surface stabilizers.
Sources Model p value
Adj–R2
Lack of fit test p value
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Pharmaceutical nanotechnology
Formulation, optimization and in vitro–in vivo evaluation of febuxostat nanosuspension Bhupesh K. Ahuja 1, Sunil K. Jena 1, Sharan K. Paidi, Surbhi Bagri, Sarasija Suresh * Department of Pharmaceutical Technology (Formulations), National Institute of Pharmaceutical Education and Research, S.A.S. Nagar, Punjab 160062, India
A R T I C L E I N F O
A B S T R A C T
Article history: Received 8 October 2014 Received in revised form 29 November 2014 Accepted 5 December 2014 Available online 6 December 2014
The purpose of the present study was to develop febuxostat nanosuspension and investigate its effect on febuxostat solubility, dissolution rate and oral bioavailability. The wet media milling technique was adopted with a combination of hydroxypropyl methylcellulose (HPMC E3) and D-a-tocopherol polyethylene glycol 1000 succinate (TPGS) as surface stabilizers for the generation of nanocrystals. Rotatable central composite design (CCD) was selected for nanosuspension optimization. The critical parameters were bead volume, milling time, polymer and surfactant concentrations; whereas particle size, polydispersity index (PDI) and zeta potential were taken as responses. The presence of crystallinity was confirmed by differential scanning calorimetry and powder X-ray diffraction. Scanning electron microscopy and transmission electron microscopy revealed small and uniform plate like morphology. A significant increase was observed in saturation solubility and dissolution rate of the optimized nanosuspension in all the pH conditions tested. Oral bioavailability of FXT and optimized FNC was evaluated in SD rats. The nanosuspension exhibited enhanced Cmax (26.48 2.71 vs. 19.85 2.96 mg/mL) and AUC01 (222.29 9.81 vs. 100.32 9.36 mg h/mL) with a 221.6% increase in relative bioavailability. Thus, FNC is a viable approach to enhance the bioavailability of FXT, a BCS Class II drug. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Febuxostat Wet media milling Nanosuspension Saturation solubility Stability Oral bioavailability
1. Introduction The large number of lead compounds springing up from drug discovery programs, exhibits poor water solubility (Kayaert et al., 2010). This phenomenon is a consequence of the new paradigm in high-throughput screening, which is based on good affinity and specific binding with target receptor. Candidates who emerge from these screening studies frequently have high log P value and molecular weight; both these factors contribute to poor water solubility (Rabinow, 2004). Such molecules are of two types, nonlipophilic hydrophobic drugs known as ‘brick dust’ or lipophilic hydrophobic drugs known as ‘grease balls’ (Müllertz et al., 2010). According to the biopharmaceutics classification system (BCS), drugs having low solubility and high permeability are classified as class II drugs (Frick et al., 1998). Numerous approaches have been adopted to increase the solubility of these molecules, including use of cosolvents (Miyako et al., 2010), surfactants (Torchilin, 2001; Kawakami et al., 2006), lipid based formulations (Chakraborty
* Corresponding author. Tel.: +91 172 2292055/9888069557; fax: +91 172 2214692. E-mail address: [email protected] (S. Suresh). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.ijpharm.2014.12.003 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
et al., 2009), inclusion complexation with cyclodextrin (Zirar et al., 2008), amorphous solid dispersions (Joshi et al., 2004) and nanotechnology (Ravichandran, 2009). However, complexation with cyclodextrin is limited by a suitable molecular size requirement while the amorphous systems are restricted due to their physical and chemical stability issues. When it comes to drugs which are insoluble in both aqueous and organic media (brick dust drugs), these approaches are often ineffective (Gao et al., 2012). Nanotechnology, in particular, nanocrystal (particle size reduction) technology can be adopted for enhancement of dissolution and saturation solubility of these drugs and ultimately, their bioavailability (Müller et al., 2001). Nanocrystals are crystalline nanoparticles with size ranging from 200 to 500 nm stabilized by surface stabilizers (MeriskoLiversidge et al., 2003). They increase the saturation solubility, dissolution rate and probably the mucoadhesion resulting in improved oral bioavailability of drugs exhibiting dissolution rate dependent bioavailability (Jacobs and Müller, 2002; Ponchel et al., 1997). In addition, nanocrystals exhibit the advantages of high drug loading, avoidance of organic solvents, enhanced stability and less toxicity, in comparison to liposomes, polymer nanoparticles, lipid nanoparticles and formulations with cosolvents (Kim et al., 2001; George and Ghosh, 2013). Nanosuspension formulations of several drugs are already marketed, exemplified by Rapamune1
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
(sirolimus), Emend1 (aprepitant), Tricor1 (fenofibrate), Cesamet1 (nabilone), Megace1 ES (megesterol acetate) and Invega Sustenna1 (paliperidone palmitate) (Junghanns Jens-Uwe and Müller, 2008). Quality by design (QbD) is a systematic, scientific, and risk based proactive approach to pharmaceutical development, which comprises of designing and developing formulations, and manufacturing processes with predefined product specifications (Lawrence, 2008). QbD encompasses the application of tools such as: design of experiments (DoE), risk assessment and process analytical technology (PAT) for the development of pharmaceuticals (Verma et al., 2009). Design of experiments (DoE) is an important tool of QbD, which allows understanding of the influence of formulation and process variables on the product quality by defining a ‘design space’ (DS) (Savic et al., 2012). DS provides the flexibility of operating within that space without further regulatory approvals. Febuxostat (FXT) (Fig. 1) is a novel, potent, non-purine, selective xanthine oxidase inhibitor (Maddileti et al., 2013). It is a weak acid (pKa 3.08) that is practically insoluble in water (solubility of 5 mg/ mL at 37 C was observed in our study). Its oral bioavailability is hampered by low aqueous solubility and its vulnerability to enzymatic degradation in both intestine and liver (Ernst and Fravel, 2009). Moreover, the presence of food decreases the maximum concentration of febuxostat in plasma (Cmax) by 38–49% (Khosravan et al., 2008). Among the methods employed for nanocrystal generation, wet media milling (Nanocrystal1) classified as a top–down approach, is a simple, effective and industrially applicable (Liversidge and Liversidge, 2011). Mechanical attrition of drug with milling media and impact with the walls of milling chamber results in the production of nanocrystals of desired size (Van Eerdenbrugh et al., 2008a,b,b; Liu et al., 2011). The objective of the present study was to formulate febuxostat nanocrystals (FNC) in order to improve its oral bioavailability by increasing its solubility and dissolution rate. FNC formulation was prepared by wet media milling, and the formulation and process were optimized by employing four factors and three level rotatable central composite design (CCD). The selected responses comprised of particle size, polydispersity index (PDI) and zeta potential. The
541
CCD is one of the most efficient designs of response surface experimental methodology to study the effects of process and formulation components on responses for exploring quadratic response surfaces and the second-order polynomial model. Data analysis using ANOVA and multifactor analysis is performed to assist in elucidating the interactions between different variables and provide a predictive model for the process. The physicochemical characterization of the optimized batch was systematically investigated by differential scanning calorimetry (DSC), powder X-ray diffraction (PXRD), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Further, saturation solubility, dissolution rate and relative oral bioavailability of FNC were investigated. 2. Theoretical prospective 2.1. Biopharmaceutical characterization Solubility, dissolution rate and permeability of the drug through the gastrointestinal membrane are the decisive parameters limiting oral drug absorption. As per biopharmaceutics classification system (BCS), these key parameters are characterized by three dimensionless numbers, i.e., dissolution number (Dn), absorption number (An), and dose number (Do) (Oh et al., 1995). Both physicochemical and physiological parameters of the drug were taken into account in calculating these dimensionless numbers. In addition, calculated dissolution time (Tdisso) and dose absorbable (Dabs), are also applied to define rate limiting step for oral absorption of drugs (Oh et al., 1995; Löbenberg and Amidon, 2000). 3. Materials and methods 3.1. Materials Febuxostat (FXT) was generously provided by Kemwell Biopharma Pvt., Ltd. (Bangalore, India) as a gift sample. Hydroxypropyl methylcellulose E3 Premium LV (HPMC E3) and E5 Premium LV (HPMC E5) were kindly gifted by Colorcon Asia Pvt., Ltd. (Goa, India). Phospholipon1 80H (P 80H) was generously provided by Lipoid (Ludwigshafen, Germany). D-a-Tocopherol polyethylene glycol 1000 succinate (TPGS) was purchased from Sigma–Aldrich (Bangalore, India). Poloxamer 188 (P 188) and poloxamer 407 (P407) were purchased from Signet Chemicals Corporation Pvt., Ltd. (Mumbai, India). Sodium lauryl sulphate (SLS) was purchased from Sisco Research Laboratories Pvt., Ltd. (Mumbai, India). Polyvinyl pyrolidone K30 (PVP K30) was procured from All Well Pharmaceuticals (Chandigarh, India). Potassium dihydrogen phosphate was purchased from Merck Specialities Pvt., Ltd. (Mumbai, India). Methanol (HPLC grade) was purchased from Sigma–Aldrich (Bangalore, India). All other chemicals and reagents were of analytical or chromatographic grade and used without further purification. Purified Milli-Q water (Millipore, Billerica, MA, USA), degassed and filtered through 0.45 mm hydrophilic PVDF filters (Millipore Millex-HV), was used in all experiments. 3.2. HPLC analysis of febuxostat
Fig. 1. Chemical structure of FXT.
HPLC method was used for the estimation of FXT in all samples, including the solubility and dissolution experiments. The HPLC system was comprised of a Waters 2695 separation module equipped with a quaternary pump, an auto sampler unit, and a Waters 2996 photodiode array (PDA) detector. ZORBAX XDB-C18 (5 mm; 250 mm 4.6 mm) (Agilent Technologies, Inc., USA) analytical column was used for the estimation. The mobile phase consisted of methanol (0.7% v/v triethylamine) and phosphate buffer (10 mM, pH 2.7) in 60:40 (% v/v) proportions. The flow rate
542
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
was maintained at 1 mL/min and the PDA detector was set at 315 nm. The Empower HPLC software was used for the analysis of the results. The procedure was performed in triplicate, and the average and standard deviations were calculated. 3.3. Generation of febuxostat nanocrystals (FNC) The FNC was generated by wet media milling method. Briefly, 40 mg (2% w/v) of FXT was dispersed in an aqueous solution containing 0.5% w/v primary stabilizer and 0.1% w/v secondary stabilizer. The resulting suspension was poured into a glass vial containing 4 mL of yttrium stabilized zirconium oxide beads (0.3–0.4 mm) and stirred on a magnetic stirrer (IKA1 RH basic 1, Bangalore, India) at 1500 rpm for 1 h at room temperature. 3.4. Particle size distribution FNC was evaluated for mean particle size and polydispersity index (PDI) by dynamic light scattering (DLS) using the Zetasizer (Nano ZS, Malvern Instruments, UK) at a detection angle of 173 at 25 C. Samples were suitably diluted with de-ionized water before analysis. Each sample was analyzed in triplicate. 3.5. Screening of surface stabilizers 3.5.1. Based on particle size distribution Surface stabilizers were selected by determining the particle size distribution of the resultant FNC. In this study, HPMC E3, HPMC E5 and PVP K30 as the primary stabilizers alone or in combination with Poloxamers 188 and 407, TPGS, SLS and Phospholipon1 80H as the secondary stabilizers were evaluated. The most suitable stabilizers were identified by optimum particle size and PDI. In these experiments, the drug concentration (2% w/v), primary stabilizer concentration (0.5% w/v), secondary stabilizer concentration (0.1% w/v), distilled water (2 mL), stirring rate (1500 rpm), bead volume (4 mL) and milling time (60 min) were kept constant. Each experiment was performed in triplicate. 3.5.2. Based on solubility study Solubility of FXT was determined in solutions of the surface stabilizers that resulted in particle size less than 300 nm. Briefly, excess amount of FXT was added to stabilizer solutions (5 mL) in sealed glass vials. The vials were agitated at 100 rpm for 72 h in the shaker water bath (EQUITRON, Medica Instrument Mfg., Co., India) at 37 C and centrifuged (OptimaTM LE-80K, Beckman Coulter, USA) at 40,000 rpm for 10 min at 4 C to remove the undissolved drug. The supernatant was suitably diluted with methanol and 10 mL aliquot was analyzed for FXT by HPLC at 315 nm. Each experiment was performed in triplicate. 3.6. Experimental design Initial screening studies were carried out for evaluating the effect of process parameters and formulation parameters on FNC formulation and its stability. The bead volume and milling time were identified as critical process parameters and polymer and surfactant concentrations as critical formulation parameters. The design of experiment (DoE) was employed systematically to evaluate and optimize the selected process and formulation parameters at three levels (1, 0, +1). Based on the number of factors and their levels, a rotatable central composite design (CCD) was selected to investigate their effects on critical quality attributes (CQAs) of nanosuspension. The batch size (2 mL), drug concentration (2% w/v), HPMC E3 as a primary stabilizer and TPGS as a secondary stabilizer, milling speed and solvent system (water) were kept constant in the experimental trials. Independent factors
Table 1 Independent factor and their coded levels of central composite design. Independent factors
Design level
Actual parameters
Coded
Bead volume (mL)
A
Milling time (min)
B
Polymer concentration (% w/v)
C
0.5 0.75 1.0
1 0 +1
Surfactant concentration (% w/v)
D
0.1 0.2 0.3
1 0 +1
Actual value 3.0 3.5 4.0 60 75 90
Coded level 1 0 +1 1 0 +1
and their levels used in this study are shown in Table 1. The design contains 30 experimental runs; i.e., sixteen (24) factorial points, eight (2 4) axial points and six centre points were generated as shown in Table 2 and analyzed by the statistical software package Design-Expert1 8.0.7.1 (Stat-Ease Inc., USA). 3.7. Preparation of physical mixture of FXT, HPMC E3 and TPGS (PM) PM of FXT, HPMC E3 and TPGS was prepared in same proportion as used for the preparation of optimized FNC. Briefly, FXT (2% w/v), HPMC E3 (0.5% w/v) and TPGS (0.14% w/v) were poured in a mortar and mixed for 10 min until a homogenous mixture was obtained. The mixture was passed through 40# mesh and stored in a desiccator. 3.8. Characterization of optimized febuxostat nanocrystals 3.8.1. Particle size distribution and zeta potential FNC mean particle size and PDI were determined as per the method described in Section 3.2.2. The surface charge of the particles was assessed by measurement of zeta potential in a Zetasizer (Nano ZS, Malvern Instruments, UK) at 25 C by applying fields of 20 V/cm. Samples were suitably diluted with de-ionized water before analysis. Each sample was analyzed in triplicate. 3.8.2. Surface morphology 3.8.2.1. Scanning electron microscopy (SEM). The morphology of FXT and optimized FNC was studied by scanning electron microscope (S-3400N, Hitachi Ltd., Tokyo, Japan) at an acceleration voltage of 10 kV. Samples were sputter coated (E-1010, Hitachi Ltd., Tokyo, Japan) with gold-palladium and observed at different magnifications. 3.8.2.2. Transmission electron microscopy (TEM). TEM was employed to study the morphology of FXT nanosuspension. FNC was diluted with deionized water (200 dilution 100 mg/mL) and visualized by TEM (FEI Tecnai G2F20, Netherlands). Briefly, a drop of the dispersion was placed on the carbon-coated copper grids and stained with phosphotungstic acid (1.5% w/v), and the excess was drained off with filter paper. The grid was air dried, and viewed under the TEM at different magnifications. 3.8.3. Differential scanning calorimetry (DSC) The DSC thermograms of FXT, HPMC E3, TPGS, PM and optimized FNC were recorded on a DSC (Mettler Toledo, DSC 821e, Switzerland) equipped with STARe SW 9.01 software. Each
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
543
Table 2 Effect of selected process and formulation parameters on FNC particle size distribution and zeta potential. Run
Critical factors Bead volume (mL)
Milling time (min)
Polymer concentration (% w/v)
Surfactant concentration (% w/v)
Particle size (nm)
PDI
Zeta potential (mV)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
3 3.5 3 4 3.5 3 4 3.5 4 3.5 3.5 4 3 3.5 4.5 4 4 3.5 3 3.5 3 3 3.5 2.5 3.5 4 3 3.5 4 3.5
90 75 90 60 75 60 90 75 90 75 75 90 60 75 75 60 90 75 90 75 60 90 75 75 75 60 60 45 60 105
1.00 0.75 0.50 1.00 0.75 1.00 0.50 0.75 1.00 1.50 0.75 0.50 0.50 0.00 0.75 0.50 1.00 0.75 1.00 0.75 0.50 0.50 0.75 0.75 0.75 0.50 1.00 0.75 1.00 0.75
0.1 0.2 0.3 0.1 0.0 0.1 0.3 0.2 0.3 0.2 0.2 0.1 0.1 0.2 0.2 0.3 0.1 0.2 0.3 0.2 0.3 0.1 0.2 0.2 0.4 0.1 0.3 0.2 0.3 0.2
279.5 265.2 288.6 289.1 303.5 324.2 282.3 267.7 302.4 334.9 269.4 258.7 304.6 281.0 279.7 298.3 266.7 276.9 309.3 269.4 335.7 264.4 272.6 414.3 335.6 250.0 359.9 317.4 309.1 293.7
0.134 0.129 0.156 0.139 0.166 0.162 0.112 0.133 0.107 0.160 0.126 0.089 0.149 0.143 0.109 0.132 0.097 0.134 0.154 0.128 0.217 0.131 0.134 0.206 0.182 0.085 0.278 0.217 0.162 0.112
18.5 19.5 20.2 18.2 15.1 17.9 20.1 18.8 19.1 18.7 19.3 18.8 19.1 20.2 19.3 19.5 17.3 19.4 20.4 19 20 19 19.2 19.5 21.2 19.4 19.7 19.6 20 19.2
Responses
sample (2–4 mg) was heated in an aluminium pan at a scanning rate of 10 C/min in an atmosphere of nitrogen gas (40 mL/min) in the range of 25–240 C. The DSC was calibrated for baseline with empty pans, and for temperature and enthalpy with indium. 3.8.4. X-ray powder diffraction (PXRD) analysis The PXRD patterns of FXT, HPMC E3, TPGS, PM and optimized FNC were recorded on X-ray diffractometer (Bruker D8 Advance Diffractometer, Germany). The X-ray source was a Cu Ka 1 tube (wavelength 1.5406 Å) operated at 40 kV and 40 mA. The samples were scanned from 3 to 40 2u at a scan rate of 0.1 2u min1. 3.8.5. Saturation solubility study Saturation solubility of FXT, PM and optimized FNC were determined individually in three different buffers. Briefly, excess of each FXT, PM and optimized FNC was added to each 5 mL of HCl (pH 1.2), phthalate buffer (pH 4.5) and phosphate buffer (pH 6.8) in sealed glass vials at 37 C. The vials were agitated at 100 rpm for 72 h in the shaker water bath (EQUITRON, Medica Instrument Mfg., Co., India). Then the dispersion was centrifuged (OptimaTM LE-80K, Beckman Coulter, USA) at 40,000 rpm for 10 min at 4 C. The supernatant was suitably diluted with methanol and quantified by HPLC at 315 nm. Each experiment was performed in triplicate. 3.8.6. Drug content and in vitro drug release study The amount of FXT in the optimized FNC was determined by HPLC (Waters 2695 Separations Module). Briefly, FNC equivalent to 10 mg of FXT was dissolved in methanol and sonicated for 10 min before analysis. Dissolution of FXT, PM and FNC were studied in USP 37 type II automated dissolution test apparatus (ElectrolabTDT-08L, India) in three different media viz HCl (pH 1.2), phthalate buffer (pH 4.5) and phosphate buffer (pH 6.8). The samples, equivalent to 40 mg of FXT, were placed in the dissolution vessel containing 900 mL buffer maintained at 37 0.5 C and stirred at 75 rpm. Samples were
collected at 5, 15, 30 and 60 min and replaced with a fresh dissolution medium to maintain the sink condition. The samples were filtered (0.45 mm PVDF syringe filter, Millipore Millex-HV) and centrifuged (OptimaTM LE-80K, Beckman Coulter, USA) at 40,000 rpm for 10 min at 4 C to remove the undissolved drug. The supernatant was injected into the HPLC system for the estimation of FXT. 3.9. Stability study of optimized FNC The physical stability of optimized FNC was evaluated at 4 C and 25 C for a period of three months (Kakran et al., 2012). Aliquots of the FNC were periodically withdrawn, diluted and analyzed for particle size distribution and zeta potential by Zetasizer (Nano ZS, Malvern Instruments, UK). Each sample was analyzed in triplicate. 3.10. In vivo oral pharmacokinetic study 3.10.1. Animals and dosing Female Sprague Dawley (SD) rats of 180–200 g were obtained from the central animal facility (CAF), NIPER, S.A.S. Nagar, India. The study protocol was approved by the Institutional Animal Ethics Committee (IAEC), National Institute of Pharmaceutical Education and Research (NIPER), S.A.S Nagar, India. The animals were maintained at 25 2 C and 50–60% relative humidity (RH) under 12 h light/dark cycles for seven days before the experiment. They were randomly distributed into two groups, each containing five animals. Before the study, both groups were fasted overnight for 12 h with free access to water. One group was administered FNC orally, dose equivalent to 10 mg/kg body weight of FXT while the other group was administered FXT suspended in water, at a constant dose. Blood samples were collected from the retro-orbital plexus under mild anaesthesia into micro centrifuge tubes containing
544
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
heparin (40 IU/mL blood) at 0.25, 0.5, 1, 2, 4, 6, 8, 12 and 24 h time points. Plasma was separated by centrifuging the blood samples at 12,000 rpm for 10 min at 4 C and stored at 20 C prior to analysis. 3.10.2. Quantification of FXT in plasma samples FXT was quantified by the HPLC (Waters 2487). Briefly, an aliquot (100 mL) of plasma was mixed with 50 mL of internal standard solution (telmisartan 5 mg/mL) and 50 mL of drug solution. After vortexing for 1 min, chilled methanol and acetonitrile in 1:1 proportion (300 mL) was added as a protein precipitating agent and vortexed for 3 min. The mixture was centrifuged at 12,000 rpm for 10 min. The supernatant was filtered (0.45 mm PVDF syringe filter, Millipore Millex-HV) and the filtrate (80 mL) was analyzed by HPLC system. ZORBAX XDB-C18 (5 mm; 250 mm 4.6 mm) (Agilent Technologies, Inc., USA) analytical column was employed for the estimation. The mobile phase consisted of acetonitrile and ammonium acetate buffer (15 mM, pH 6.0) in 44:56 (% v/v) proportions, maintained at 1 mL/min flow rate. FXT was quantified at 315 nm using a UV detector (Waters 2487 dual l absorbance detector). Calibration curve was obtained from 5 to 5000 ng/mL with r2 value of 0.998. 3.10.3. Pharmacokinetic data analysis The pharmacokinetic parameters of FNC were calculated by Kinetica software version 5.0 (Thermo Fisher Scientific Inc., USA), and compared with pure FXT. Maximum concentration (Cmax) and time to reach maximum concentration (Tmax) are the values obtained directly from concentration–time curve. Area under the curve (AUC024 h or AUC01), elimination half life (t1/2), mean residence time (MRT) was determined. The relative bioavailability (F) of the FNC to the FXT was calculated by the following equation: AUCtest (1) 100 Relative bioavailabilityðFÞ ¼ AUCreference
3.11. Statistical analysis The means and standard deviation of all values were calculated. Statistical analysis was performed by paired ‘t’ test using Microcal (TM) Origin1 software version 6.0 (Microcal software, Inc., Northampton, USA). Significant difference was regarded as p < 0.05.
dissolution rate limited absorption owing to the higher Tdisso value than small intestinal transit time (Tsi) and lower Dabs than the dose of FXT. The Tdisso value was decreased considerably for the nanosized FXT indicating that the nanonization of FXT has the potential to improve its dissolution limited absorption. 4.2. Screening of surface stabilizers Despite all the advantages associated with nanocrystals, they exhibit thermodynamic instability. This could be explained by the fact that as the particle size decreases, the surface area of the particle increases, which leads to an increase in the Gibbs-free energy. System with such a high free energy is thermodynamically unstable. In order to reduce this free energy, agglomeration of smaller particles occurs (Lindfors et al., 2006; Wu et al., 2011). Surface stabilizers are added to reduce the Gibbs-free energy of the system and hence the aggregation. They act by inhibiting crystal growth and particle aggregation through the mechanism of electrostatic and steric stabilization. The efficiency of a stabilizer depends on its potential for interaction with the drug molecule (Ghosh et al., 2012). 4.2.1. Based on particle size distribution Fig. 2 depicts the particle size distribution of FNC generated by different surface stabilizers. Generation of FNC with primary stabilizers resulted in higher particle size and less stability when stored at room temperature for seven days. This might be due to an inadequate amount of the surface stabilizer necessary for complete coverage of the drug particle surface which is required to provide steric repulsion between the nanoparticles in the suspension (Ghosh et al., 2012). Conversely, an excess of the stabilizer like HPMC E3 could lead to an increase in viscosity, resulting in a decrease in milling efficiency (Singare et al., 2010). However, the minimum particle size distribution was obtained with HPMC E3 and TPGS combination and the particles remained stable even after seven days when stored at room temperature. Adsorption of HPMC E3 onto the drug crystals due to the interaction of its hydrophobic (methoxyl) group with the drug molecule provides better surface coverage (Ghosh et al., 2011). On the other hand, the hydrophobic portion of the TPGS may interact with a hydrophobic portion of FXT, thus stabilizing the FNC. The unique properties of the TPGS as solubilizer, permeability enhancer and stabilizer lead to its selection in the nanosuspension formulation (Ghosh et al., 2012,b; Van Eerdenbrugh et al., 2008a,b).
4. Results and discussions 4.1. Biopharmaceutical characterization Theoretical calculations of An, Dn and Do of FXT indicates that it belongs to BCS class II (An > 1, Dn < 1, Do > 1) drugs having both solubility and dissolution rate limited oral bioavailability. Table 3 lists the biopharmaceutical parameters that help in estimating the absorption rate limiting factors for the drug. Calculated Tdisso and Dabs values also showed that FXT exhibits both solubility and
4.2.2. Based on solubility study Fig. 3 presents the apparent solubility profiles of FXT in presence of different stabilizers. FXT exhibits higher solubility in SLS containing surface stabilizers. This could be due to the increased solubility of FXT in SLS micelle (Verma et al., 2011). The increased solubility can lead to Ostwald ripening, resulting in an increased particle size during the storage. In contrast, minimum solubility of FXT was observed in the presence of HPMC E3 and TPGS. Hence, the combination of HPMC E3 and TPGS was selected
Table 3 Biopharmaceutical parameters of FXT. Parameters
Values
Comments
Tdisso unmicronized (min) Tdisso micronized (min) Tdisso nanosized (min) Dose absorbable (Dabs) (mg) Absorption number (An) Dose number (Do) Dissolution number (Dn)
3465 173 8.7 23.48 3.91 64 0.03
Tdisso 199 Tdisso 50 Tdisso 50 Dabs dose (80) Permeability is not a problem for bioavailability (An > 1) Solubility limited bioavailability (Do > 1) Dissolution limited bioavailability (Dn < 1)
B.K. Ahuja et al. / International Journal of Pharmaceutics 478 (2015) 540–552
545
Fig. 2. The effect of surface stabilizers on FNC particle size and PDI.
as the surface stabilizer. Further, the less stability of the ionic stabilizers in the gastrointestinal fluid favors the use of steric stabilizers (Singare et al., 2010; Peltonen and Hirvonen, 2010). 4.3. Optimization of FNC 4.3.1. Experimental design The regression analysis of the data obtained from the experimental runs generated the following polynomial equations in which the model F ratios were statistically significant at a < 0.05 with Adj-R2 value in the range of 0.8–1 with a statistically non-significant lack of fit at a > 0.05. Particle size ¼ 270:20 9:26 A 12:19 B þ 12:59 C þ 16:39 D þ 8:76 AB þ 2:19 AC þ 0:8 AD 2:9 BC 2:99 BD þ 0:67 CD þ 4:31 A2 þ 7:11 B2 þ 8:96 C 2 þ 7:20 D2
(2)
PDI ¼ 0:13 0:018 A 0:026 B þ 0:0058 C þ 0:024 D 0:0017 AB þ 0:0023 AC 0:000046 AD 0:0062 BC 0:014 BD 0:0018 CD þ 0:0023 A2 þ 0:0085 B2 þ 0:0053 C 2 þ 0:0013 D2
(3)
Zeta potential ¼ þ19:31 0:053 A 0:11 B 0:39 C þ 0:81 D 0:30 AB 0:18 AC þ 0:046 AD þ 0:054 BC þ 0:004 BD þ 0:15 CD
(4)
where A,B, C and D are bead volume, milling time, percentage concentrations of polymer and surfactant, respectively. Table 4 presents the results of statistical analysis, which suggests a less than 0.01% chance that the model F value of the model Eqs. (2)–(4) occurred due to noise. The p values of the lack of fit test of the three models were not significant, indicating that these model equations fitted the data well. Therefore, the two quadratic models (particle size, PDI) and a two-factor interaction (2FI) model (zeta potential) could describe adequately the data and could be employed to navigate the design space. A positive sign represents a synergistic effect, while a negative sign indicates an antagonistic effect. In Eqs. (2) and (3), the negative coefficient of A and B in the model refer to a decrease in particle size and PDI with increasing bead volume and milling time, respectively. The positive coefficient of C and D indicates an increase in particle size with increasing amount of polymer and surfactant, respectively. In case of zeta potential, the negative coefficients of A–C in the model refer to decrease in zeta potential with increasing bead volume, milling time and percent concentration of polymer, respectively. The positive coefficient
Table 4 Results of statistical analysis of the experimental design. Responses
Particle size PDI Zeta potential Fig. 3. Solubility of FXT in different surface stabilizers.
Sources Model p value
Adj–R2
Lack of fit test p value
