- Email: [email protected]
Development and validation of a HPLC method for the determination of buprenorphine hydrochloride, naloxone hydrochloride and noroxymorphone in a tablet formulation
Talanta 77 (2009) 1415–1419
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Development and validation of a HPLC method for the determination of buprenorphine hydrochloride, naloxone hydrochloride and noroxymorphone in a tablet formulation Ali Mostafavi a , Ghazaleh Abedi b , Ahmad Jamshidi b , Daryoush Afzali a , Mohammad Talebi b,∗ a b
Department of Chemistry, Shahid Bahonar University of Kerman, Kerman, Iran Novel Drug Delivery Systems Department, Iran Polymer & Petrochemical Institute, P.O. Box 14185-458, Tehran, Iran
a r t i c l e
i n f o
Article history: Received 30 June 2008 Received in revised form 12 September 2008 Accepted 17 September 2008 Available online 24 September 2008 Keywords: Buprenorphine hydrochloride Naloxone hydrochloride Noroxymorphone High-performance liquid chromatography Tablet formulation
a b s t r a c t A simple isocratic reversed-phase high-performance liquid chromatographic method (RP-HPLC) was developed for the simultaneous determination of buprenorphine hydrochloride, naloxone hydrochloride dihydrate and its major impurity, noroxymorphone, in pharmaceutical tablets. The chromatographic separation was achieved with 10 mmol L−1 potassium phosphate buffer adjusted to pH 6.0 with orthophosphoric acid and acetonitrile (17:83, v/v) as mobile phase, a C-18 column, Perfectsil Target ODS3 (150 mm × 4.6 mm i.d., 5 m) kept at 35 ◦ C and UV detection at 210 nm. The compounds were eluted isocratically at a flow rate of 1.0 mL min−1 . The average retention times for naloxone, noroxymorphone and buprenorphine were 2.4, 3.8 and 8.1 min, respectively. The method was validated according to the ICH guidelines. The validation characteristics included accuracy, precision, linearity, range, specificity, limit of quantitation and robustness. The calibration curves were linear (r > 0.996) over the concentration range 0.22–220 g mL−1 for buprenorphine hydrochloride and 0.1–100 g mL−1 for naloxone hydrochloride dihydrate and noroxymorphone. The recoveries for all three compounds were above 96%. No spectral or chromatographic interferences from the tablet excipients were found. This method is rapid and simple, does not require any sample preparation and is suitable for routine quality control analyses. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Buprenorphine (Fig. 1a) is an oripavine derivative possessing partial mu agonist and kappa antagonist opioid activity with a potency of 20–40 times higher than that of morphine [1,2]. It has been used successfully by intramuscular, intravenous and sublingual routes for the treatment of moderate to severe pain at doses from 0.3 to 0.6 mg [3]. Clinical studies have shown that buprenorphine, like methadone, can also be used for the treatment of opioid addiction and withdrawal patients from heroin [4,5]. It is usually administered in sublingual formulation as either liquid or watersoluble tablets. As such, they can be diverted for illicit intravenous use. In an effort to circumvent this possibility, buprenorphine has now been coformulated in sublingual tablet with the mu antagonist, naloxone (Fig. 1b). Recently, sublingual tablets containing a fixed dose combination of buprenorphine hydrochloride (BH) and nalox-
∗ Corresponding author at: Novel Drug Delivery Systems Department, Iran Polymer & Petrochemical Institute, 15 KM Tehran-Karaj Highway, P.O. Box 14185-458, Tehran, Iran. Tel.: +98 21 44580000; fax: +98 21 44580021. E-mail address: talebi [email protected] (M. Talebi). 0039-9140/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.09.024
one hydrochloride dehydrate (NH), at a ratio of 4:1 with respect to the free bases, have been approved by the Food and Drug Administration (FDA) for treating opiate dependence [6]. They are available in two strengths: 2 mg/0.5 mg tablets containing 2.16 mg BH (equivalent to 2 mg buprenorphine base) and 0.61 mg NH (equivalent to 0.5 mg naloxone base); and 8 mg/2 mg containing 8.64 mg BH (equivalent to 8 mg buprenorphine base) and 2.44 mg NH (equivalent to 2 mg naloxone base). A literature survey reveals that there are a number of various analytical methods available for the quantitative individual determination of buprenorphine, or combination with other drugs mainly using chromatographic methods such as gas chromatography with electron-capture [7] or mass spectrometry [8,9] detection and HPLC with fluorescence [10], electrochemical [11] or mass spectrometry detection [12–14]. However, all of these methods involve an extraction and/or derivatization steps, which demand some well-known drawbacks including the possibility of incomplete derivatization, additional chromatographic interferences and increased method complexity and sample preparation time. The British Pharmacopeia (BP) provides two different potentiometric titration methods for assay of BH and NH. Regarding the related substances, however, two distinct RP-HPLC procedures
1416
A. Mostafavi et al. / Talanta 77 (2009) 1415–1419
Fig. 1. Chemical structures of buprenorphine (a), naloxone (b) and noroxymorphone (c).
have been proposed, one for buprenorphine using UV detection at 288 nm, and another for naloxone impurities at 230 nm. BP specifies five substances as possible impurities for NH; one of them, noroxymorphone (NM, Fig. 1c), the starting material in NH synthesis, may cause gastric disturbance with nausea, vomiting and constipation and is required to be controlled [15]. In the United States Pharmacopoeia (USP), however; without specifying any other impurities than noroxymorphone hydrochloride, a TLC method is described for the determination of this compound and other NH impurities, yet monograph improvement has been requested [16,17]. It is worth noting the lack of official methods for assays of active ingredients and related substances in pharmaceutical preparations in pharmacopoeia, especially when it is currently possible to find a plethora of methods based on HPLC of these compounds. However, to the best of our knowledge, there are no chromatographic methods in the literature for the analysis of any buprenorphine dosage forms. The only reported analysis of buprenorphine in pharmaceutical products includes a spectrophotometric method based on colored ion-pair formation [18]. Hence, an attempt has been made to develop a simple, efficient and selective method for the determination of BH, NH and its major degradation impurity, NM, in sublingual tablets. In this study, HPLC instrumentation with UV detection, which is readily available in most analytical and pharmaceutical laboratories, was used. The method requires no extraction or derivatization steps reducing the total analysis run time to less than 10 min. 2. Experimental 2.1. Materials Reference standards were purchased from Macfarlan Smith (Edinburgh, UK) and were checked against European Pharmacopoeia CRS standards (Strasbourg, France). The common tablet excipients and commercial marketed products were provided by Darou Darman Pars (Tehran, Iran). All other reagents were spectrophotometric or HPLC grade obtained from Romil Chemicals (Loughborough, UK). Water purified via a Milli-Q system, Millipore Corp. (Bedford, USA) was used for all purposes. 2.2. Instrumentation and chromatographic conditions The HPLC system consisted of a Younglin ACME 9000 (Seoul, Korea) equipped with a quaternary pump, online degasser, column heater, autosampler and UV detector. Data collection and analyses were performed using Autochro 2000 software (Younglin). Separation was achieved on C-18 column, Perfectsil Target ODS3 (150 mm × 4.6 mm, 5 m) with a 10 mm × 4.0 mm, 5 m guard column (MZ-Analysentechnik, Mainz, Germany). The elution was isocratic with mobile phase of acetonitrile and 10 mmol L−1 potassium phosphate buffer adjusted to pH 6.0 with orthophosphoric
acid (83:17, v/v). The flow rate was 1.0 mL min−1 and yielded a backpressure of about 740 psi. The column temperature was maintained at 35 ◦ C, the detection was monitored at a wavelength of 210 nm and injection volume was 20 L. 2.3. Standard solutions and calibration graphs for chromatographic measurement Stock standard solutions of BH, NH and NM were separately prepared in methanol using individual EP CRS standards to obtain concentrations of 440, 200 and 200 g mL−1 , respectively. If haziness was observed in NM solution, addition of about 0.5%, 0.1 mol L−1 HCl was sufficient to clarify the solution. Calibration standards at eight levels were prepared by appropriately mixed and further diluted stock standard solutions in the concentration range of 0.022–220 g mL−1 for BH and 0.01–100 g mL−1 for NH and NM. Similarly, quality control (QC) standard solutions were prepared daily by diluting the corresponding stock standard solutions of individual reference standards for the final QC concentrations of 4.8, 48 and 240 g mL−1 for BH and 2.4, 24 and 120 g mL−1 for NH and NM, respectively. Samples in triplicates were made for each concentration and peak areas were plotted against the corresponding concentrations to obtain the calibration graphs. 2.4. Sample preparation A number of 20 accurately weighed tablets were ground into a fine powder using a glass mortar and pestle. A portion equivalent to about four tablets (typically each contains 2.16 mg BH and 0.61 mg NH) was accurately weighed and transferred to a 20 mL volumetric flask. Approximately 15 mL methanol and about 0.5%, 0.1 mol L−1 HCl were added to the flask and the contents were vortex-mixed for 10 min. The flask was adjusted to volume and mixed well. The resulting solution was filtered using 0.45 m PTFE filter into standard analytical glass vials and injected into the HPLC. Three such samples were prepared from each 20-tablet mixture according to the USP criteria and injected triplicate. 2.5. Method validation The method was validated according to the ICH guidelines [19]. The following validation characteristics were addressed: linearity, accuracy, precision, specificity, limits of detection and quantitation and robustness. 2.5.1. System suitability testing (SST) System suitability standard solution which contained 200 g mL−1 BH, 100 g mL−1 NH and 0.1 g mL−1 NM was prepared by appropriately diluting and mixing the corresponding stock standard solutions. System suitability was determined from six replicate injections of the system suitability standard before
A. Mostafavi et al. / Talanta 77 (2009) 1415–1419
sample analysis. According to the monograph, the acceptance criteria for NH were less than 2% R.S.D. and a signal-to-noise ratio of at least ten for the corresponding peak area. For NM, acceptance criteria were less than 2% R.S.D. for peak area and not less than four for the resolution between the peaks corresponding to NM and NH. Resolution was calculated as defined by the USP. 2.5.2. Linearity and range Standard calibration curves were prepared with seven calibrators over a concentration range of 0.22–220 g mL−1 for BH and 0.1–100 g mL−1 for both NH and NM. The data of peak area versus concentration were treated by linear least square regression analysis. The standard curves were evaluated for intra-day and inter-day linearity. 2.5.3. Accuracy To study the reliability and suitability of the developed method, recovery experiments were carried out. Placebo samples were spiked with different amount of BH and NH at 50, 100 and 150% in duplicate for each one (n = 6) over the theoretical values, and NM at 0.1% over the NH content. Measured values were compared with the theoretical concentration. Recovery for pharmaceutical formulations should be within the range 100 ± 5%. The R.S.D. percent of individual measurements was also determined. The results must be less than 5%. 2.5.4. Precision The precision of the developed method was assessed in terms of repeatability and intermediate precision by analyzing three replicate QC standard samples at 50, 100 and 150% levels that cover the calibration ranges for BH and NH, and at a single concentration of 0.1 g mL−1 for NM. The %R.S.D. values of the results corresponding to the peak area and retention time were expressed for intra-day precision and on 3 days for intermediate (inter-day) precision. 2.5.5. Specificity Injections of the extracted placebo were performed to demonstrate the absence of interference with the elution of the BH, NH and NM. Samples containing NH’s main impurities, the NM, were also injected. For determining selectivity of the method, a powder blend of typical tablet excipients containing lactose monohydrate, mannitol, maize starch, povidone K30, citric acid anhydrous granular, sodium citrate, natural lemon and lime flavour, acesulfame potassium and magnesium stearate was prepared and analyzed. All chromatograms were examined to determine if compounds of interest co-eluted with each other or with any additional excipient peaks. 2.5.6. Limits of detection and quantitation The limit of detection (LOD) and limit of quantitation (LOQ) for the procedure were performed on samples containing very low concentrations of analytes under the ICH guidelines. By applying the visual evaluation method, LOD was expressed by establishing the minimum level at which the analyte can be reliably detected. LOQs were considered as the lowest concentration of analytes in standards that can be reproducibly measured with acceptable accuracy and precision. 2.5.7. Robustness The robustness of the method was evaluated by analyzing the system suitability standards and evaluating system suitability parameter data after varying, individually, the HPLC pump flow rate (±10%), organic solvent content (±6%) and column compartment temperature (±14%).
1417
3. Results and discussion 3.1. Method development and optimization NM is a major impurity of NH drug substance. The main target of the chromatographic assay method development was to separate the impurity co-eluted with naloxone. Typically, method development focuses on identifying buffer type, strength and pH, organic solvent and implementing small changes to optimize selectivity and enhance resolution. Initially, NM was found to be co-eluted with naloxone by using different stationary phases such as C-8 and C-18, with mobile phases containing buffers like phosphate, at different pH and temperature, and organic solvents like methanol and acetonitrile. At the first stage, a C-8 column chemistry and potassium phosphate buffer, pH 3.0, were used with methanol as the organic solvent. Though the column was base deactivated for improved peak shape of basic compounds, both peak symmetry and resolution between two compounds were poor. Subsequently, an acceptable peak shape and resolution were achieved by increasing the buffer pH to 6.0, approximately 2.5 and 2.0 pH units above the pKa1 of BH and NH, respectively; and using ternary solvent system consisting of 10% acetonitrile and 70% methanol as organic solvent. Nevertheless, the run time was prohibitively long at approximately 14 min. In order to better exploit the polarity differences between the desired compounds, while maintaining a short run time; two C-18 columns were evaluated. A typical silica-based monolithic column, which proves to dramatically reduce the analysis time, was examined at first. Although the retention time of buprenorphine reduced to about 6.0 min, applying the monolithic column at its best operating conditions (an organic content of 45% ACN at a flow rate of 2.0 mL min−1 and oven temperature of 40 ◦ C), produced neither complete resolution between the first peak (naloxone) and contaminants co-eluted at the beginning, nor better asymmetry factor of 0.7 for studied compounds. The excessive resolution at the beginning of the chromatogram and appropriate peak shape were finally achieved by switching to a base deactivated Perfectsil target ODS3 column. For optimum resolution, peak asymmetry and analysis time, the mobile phase consisting of 10 mmol L−1 potassium phosphate buffer adjusted to pH 6.0 with orthophosphoric acid and acetonitrile (17:83, v/v) was used. To improve repeatability of runs, shorten analysis time and reduce back pressure, which is important to extending the column life time, the column oven temperature was set at 35 ◦ C. In the optimized conditions, NM and the drug substances naloxone and buprenorphine were well separated with resolutions of more than 10 and 18, and asymmetry factors very close to 1.0. The optimal wavelength was established experimentally after measuring all spectra in mobile phase and testing the detector response of analytes using UV absorbance scanned over the range of 190–400 nm. Although the absorption maxima recommended by BP for BH and NH were 288 and 230 nm, respectively; it was shown that 210 nm is the optimal wavelength to maximize the sensitivity and has no interference with other components of the formulation. Under the experimental conditions investigated, the retention times for naloxone, noroxynorphone and buprenorphine were 2.40, 3.84 and 8.06 min, respectively (Fig. 2).
3.2. Method validation When a method has been optimized it must be validated before practical use. By following the ICH guidelines for analytical method validation, Q2 (R1), the SST was performed and the validation characteristics were addressed.
1418
A. Mostafavi et al. / Talanta 77 (2009) 1415–1419
Table 1 Linearity parameters for the simultaneous estimation of NH, NM and BH. Parameter −1
Linearity range (g mL ) Slope Intercept Correlation coefficient (r) Residual sum of squares
Naloxone hydrochloride
Noroxymorphone
Buprenorphine hydrochloride
0.10–100 67.70 ± 0.24 28.90 ± 7.01 0.9997 ± 0.0001 0.00028
0.10–100 75.75 ± 0.55 193.10 ± 0.59 0.9961 ± 0.0005 0.00378
0.22–220 79.80 ± 0.12 −35.20 ± 6.86 0.9997 ± 0.0001 0.00036
Values are reported as mean ± S.D. of three calibration curves generated on three consecutive days (n = 3). Seven concentrations in the linearity range were evenly distributed. Table 2 Method validation results for studied compounds. Naloxone hydrochloride
Noroxymorphone
Buprenorphine hydrochloride
SST Theoretical plates Asymmetry (As ) Resolution (Rs ) Repeatabilitya , tR (%R.S.D.) Repeatabilitya , A (%R.S.D.)
6485 1.0 3.5 0.3 0.6
7599 1.1 10.3 1.7 1.5
8668 1.0 18.8 1.3 0.6
Validation Precisiona , b (%R.S.D.) Accuracya , b (%R.S.D.) Accuracy (%recovery) Selectivity LOD (g mL−1 ) LOQ (g mL−1 )
3.1 4.1 102.8 No interference 0.01 0.10
2.9 5.0 96.2 No interference 0.01 0.10
4.2 3.3 102.3 No interference 0.02 0.22
a Intra-day precision (repeatability), inter-day precision and accuracy were tested by analyzing three replicate QC standard samples at 50, 100 and 150% levels for BH and NH and at a single concentration of 0.1 g mL−1 for NM under the guidelines of ICH. b Inter-day precision and accuracy were determined with three replicates on three consecutive days.
3.2.1. System suitability The system suitability test ensures the validity of the analytical procedure as well as confirms the resolution between different peaks of interest. All critical parameters tested met the acceptance criteria on all days. According to the monograph, however, the area of an individual secondary chromatographic peak, which appears in the test sample, should not exceed 0.5% of the naloxone peak area, and the total area of all secondary peaks, 1% naloxone peak area. In addition, the resolution between the peak corresponding to naloxone and NM in the chromatogram obtained with reference solution should not be less than 4 at S/N of at least 10 [15]. Adequate resolution of >10 between the naloxone and its impurity, NM, meets the acceptance criteria indicated in the monograph. As
shown in the chromatogram, all three analytes are eluted by forming symmetrical single peaks well separated from the solvent front (Fig. 2). 3.2.2. Linearity and range For the construction of calibration curves, seven calibration standard solutions were prepared over the concentration range of 0.22–220 g mL−1 for BH and 0.1–100 g mL−1 for NH and NM. The results, summarized in Table 1, show a good correlation between analytes peak area and concentration with r > 0.996 (n = 7). 3.2.3. Accuracy and precision Accuracy and precision were established across the analytical range for BH, NH and NM. The intra- and inter-day accuracy and precision were calculated from the QC samples (Table 2). Repeatability (intra-day precision) of the analytical method was found to be reliable based on %R.S.D. (<2%) corresponding to the peak areas and retention times. Intermediate precision (inter-day accuracy) was demonstrated on different days and evaluating the peak area data at three QC standards that cover the assay method range. The %R.S.D. values were less than 5% and illustrated the good precision for the analytical method. For determining accuracy, placebo solutions spiked with reference standards were used. The recovery was 100 ± 5% for all samples with %R.S.D. less than 5%. 3.2.4. Specificity Injections of the extracted placebo were performed to demonstrate the absence of interference with the elution of the drugs and impurity. These results demonstrate that there was no interference from other materials in the tablet formulation; therefore, confirm the specificity of the method (Fig. 2).
Fig. 2. Representative chromatograms obtained for the mobile phase, placebo and QC standard (corresponding to the 24, 24 and 48 g mL−1 NH, NM and BH, respectively).
3.2.5. Sensitivity The limit of detection and limit of quantitation decide about the sensitivity of the method. Tests for the procedure were performed on samples containing very low concentrations of analytes based
A. Mostafavi et al. / Talanta 77 (2009) 1415–1419
1419
Table 3 Method validation data for robustness study. Parameter altered
Optimized chromatographic conditionsa Increased organic solvent (16:84) Decreased organic solvent (18:82) Increased flow rate (1.1 mL min−1 ) Decreased flow rate (0.9 mL min−1 ) Decreased column temperature (30 ◦ C)
Retention time, tR (min)
Resolution (Rs )
Asymmetry (As )
N
NM
B
N
NM
B
N
NM
B
2.4 2.3 2.4 2.2 2.6 2.5
3.8 3.7 3.9 3.7 3.9 4.0
8.0 8.0 8.1 7.8 8.3 8.3
3.5 2.5 3.0 1.8 4.2 3.6
10.3 8.75 9.4 8.2 9.3 10.5
18.8 18.1 20.1 17.8 18.7 20.1
1.0 0.9 1.1 1.0 1.1 1.1
1.1 1.0 1.15 1.1 1.2 1.2
1.0 1.0 1.1 1.0 1.15 1.1
a Chromatographic conditions were 10 mmol L−1 potassium phosphate buffer, pH 6.0/ACN (17:83) maintained with flow rate of 1.0 mL min−1 , temperature at 35 ◦ C and detection at 210 nm.
amount of NM was found below the detection limit (0.01 g mL−1 ) and, therefore, should be less than 0.1% over the NH content. 4. Conclusion A simple isocratic reversed-phase HPLC method proposed was found to be accurate, precise, linear across the analytical range and robust. The method was specific for the simultaneous determination of BH, NH and its primary impurity NM in a tablet formulation. All the parameters for three substances met the criteria of the ICH guidelines for method validation. The method could therefore be recommended for routine quality control analysis of raw materials and various buprenorphine and naloxone dosage in tablet formulations by assaying for potency and accurately monitoring the NM impurity. Acknowledgement
Fig. 3. Representative chromatogram obtained for a commercially available BH/NH (4:1) drug product (N: naloxone; B: buprenorphine).
Kimia Sanat Jahan Co. is acknowledged for kindly furnishing the home edition of Autochro-2000 HPLC software. References
on the visual evaluation method. In this method, LOD is determined by the analysis of samples with known concentration of analyte and by establishing the minimum level at which the analyte can be reliably detected. Accordingly, the LOQ is determined by the analysis of samples with known concentration of analytes and by establishing the minimum level at which the analyte can be quantified with acceptable accuracy and precision (R.S.D. <2%). The LOD and LOQ values were found to be 0.022 and 0.22 g mL−1 for BH and 0.01 and 0.1 g mL−1 for both NH and NM. 3.2.6. Robustness To ensure the insensitivity of the developed HPLC method to minor changes in the experimental conditions, it is important to demonstrate its robustness. None of the alterations caused a significant change in resolution between naloxone and NM, peak area R.S.D., USP tailing factor and theoretical plates (Table 3). Although the changes in retention times were more significant, separation was sufficient and quantitation was still possible. 3.3. Analysis of the marketed product The validated method was used in the analysis of a marketed product in a 70 mg tablet dosage form with dose strength of 2.0 mg BH/0.5 mg NH. Representative chromatogram is shown in Fig. 3. The results for the drugs assay and the concentration of impurity NM were in good agreement with the label claims. BH content was between 91.4 and 93.5%, and for NH between 97.2 and 104.7%. The
[1] S.L. Walsh, K.L. Preston, G.E. Bigelow, M.L. Stitzer, J. Pharmacol. Exp. Ther. 247 (1995) 361. [2] G.B. Collins, M.S. McAllister, Clev. Clin. J. Med. 74 (2007) 514. [3] J.W. Lewis, in: A. Cowan, J.W. Lewis (Eds.), Buprenorphine: Combating Drug Abuse with a Unique Opioid, Wiley-Liss, New York, 1995, p. 151. [4] R.E. Johnson, J.H. Jaffe, P.J. Fudala, J. Am. Med. Assoc. 267 (1992) 2750. [5] W. Ling, D.R. Wesson, C. Charuvastra, C.J. Klett, Arch. Gen. Psychiatry 53 (1996) 401. [6] C.N. Chiang, R.L. Hawks, Drug Alcohol Depend. 70 (2003) S39. [7] E.T. Everhart, P. Cheung, P. Shwonek, K. Zabel, E.C. Tisdale, P. Jacob III, J. Mendelson, R.T. Jones, Clin. Chem. 43 (1997) 2292. [8] A.M. Lisi, R. Kazlauskas, G.J. Trout, J. Chromatogr. Biomed. Appl. 692 (1997) 67. [9] L. Debrabandere, M. Van Boven, L. Laruelle, P. Daenens, Anal. Chim. Acta 275 (1993) 295. [10] S.Y. Liu, K.S. Liu, C.H. Kuei, J.I. Tzeng, S.T. Ho, J.J. Wang, J. Chromatogr. B 818 (2005) 233. [11] M.A. García-Fernández, M.T. Fernández-Abedul, A. Costa-García, Chromatographia 53 (2001) 704. [12] D.E. Moody, M.H. Slawson, E.C. Strain, J.D. Laycock, A.C. Spanbauer, R.L. Foltz, Anal. Biochem. 306 (2002) 31. [13] S. Pirnay, F. Hervé, S. Bouchonnet, B. Perrin, F.J. Baud, I. Ricordel, J. Pharm. Biomed. Anal. 41 (2006) 1135. [14] M.E. Rodriguez-Rosas, M.R. Lofwall, E.C. Strain, D. Siluk, I.W. Wainer, J. Chromatogr. B 850 (2007) 538. [15] The British Pharmacopoeia 2007, The Stationary Office (TSO), Health Ministry, London, 2006. [16] The United State Pharmacopoeia, 27th ed., The United States Pharmacopoeial Convection, Rockville, MD, 2004. [17] http://www.usp.org/USPNF/submitMonograph/improveMon.html. [18] M. Amanlou, P. Khosravian, E. Souri, O.G. Dadrass, R. Dinarvand, M.M. Alimorad, H. Akbari, Bull. Korean Chem. Soc. 28 (2007) 183. [19] ICH guidelines Topic Q2 (R1) Validation of Analytical Procedures: Methodology, obtained from web site: www.emea.europa.eu/htms/human/ich/ ichquality.htm.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Development and validation of a HPLC method for the determination of buprenorphine hydrochloride, naloxone hydrochloride and noroxymorphone in a tablet formulation Ali Mostafavi a , Ghazaleh Abedi b , Ahmad Jamshidi b , Daryoush Afzali a , Mohammad Talebi b,∗ a b
Department of Chemistry, Shahid Bahonar University of Kerman, Kerman, Iran Novel Drug Delivery Systems Department, Iran Polymer & Petrochemical Institute, P.O. Box 14185-458, Tehran, Iran
a r t i c l e
i n f o
Article history: Received 30 June 2008 Received in revised form 12 September 2008 Accepted 17 September 2008 Available online 24 September 2008 Keywords: Buprenorphine hydrochloride Naloxone hydrochloride Noroxymorphone High-performance liquid chromatography Tablet formulation
a b s t r a c t A simple isocratic reversed-phase high-performance liquid chromatographic method (RP-HPLC) was developed for the simultaneous determination of buprenorphine hydrochloride, naloxone hydrochloride dihydrate and its major impurity, noroxymorphone, in pharmaceutical tablets. The chromatographic separation was achieved with 10 mmol L−1 potassium phosphate buffer adjusted to pH 6.0 with orthophosphoric acid and acetonitrile (17:83, v/v) as mobile phase, a C-18 column, Perfectsil Target ODS3 (150 mm × 4.6 mm i.d., 5 m) kept at 35 ◦ C and UV detection at 210 nm. The compounds were eluted isocratically at a flow rate of 1.0 mL min−1 . The average retention times for naloxone, noroxymorphone and buprenorphine were 2.4, 3.8 and 8.1 min, respectively. The method was validated according to the ICH guidelines. The validation characteristics included accuracy, precision, linearity, range, specificity, limit of quantitation and robustness. The calibration curves were linear (r > 0.996) over the concentration range 0.22–220 g mL−1 for buprenorphine hydrochloride and 0.1–100 g mL−1 for naloxone hydrochloride dihydrate and noroxymorphone. The recoveries for all three compounds were above 96%. No spectral or chromatographic interferences from the tablet excipients were found. This method is rapid and simple, does not require any sample preparation and is suitable for routine quality control analyses. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Buprenorphine (Fig. 1a) is an oripavine derivative possessing partial mu agonist and kappa antagonist opioid activity with a potency of 20–40 times higher than that of morphine [1,2]. It has been used successfully by intramuscular, intravenous and sublingual routes for the treatment of moderate to severe pain at doses from 0.3 to 0.6 mg [3]. Clinical studies have shown that buprenorphine, like methadone, can also be used for the treatment of opioid addiction and withdrawal patients from heroin [4,5]. It is usually administered in sublingual formulation as either liquid or watersoluble tablets. As such, they can be diverted for illicit intravenous use. In an effort to circumvent this possibility, buprenorphine has now been coformulated in sublingual tablet with the mu antagonist, naloxone (Fig. 1b). Recently, sublingual tablets containing a fixed dose combination of buprenorphine hydrochloride (BH) and nalox-
∗ Corresponding author at: Novel Drug Delivery Systems Department, Iran Polymer & Petrochemical Institute, 15 KM Tehran-Karaj Highway, P.O. Box 14185-458, Tehran, Iran. Tel.: +98 21 44580000; fax: +98 21 44580021. E-mail address: talebi [email protected] (M. Talebi). 0039-9140/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.09.024
one hydrochloride dehydrate (NH), at a ratio of 4:1 with respect to the free bases, have been approved by the Food and Drug Administration (FDA) for treating opiate dependence [6]. They are available in two strengths: 2 mg/0.5 mg tablets containing 2.16 mg BH (equivalent to 2 mg buprenorphine base) and 0.61 mg NH (equivalent to 0.5 mg naloxone base); and 8 mg/2 mg containing 8.64 mg BH (equivalent to 8 mg buprenorphine base) and 2.44 mg NH (equivalent to 2 mg naloxone base). A literature survey reveals that there are a number of various analytical methods available for the quantitative individual determination of buprenorphine, or combination with other drugs mainly using chromatographic methods such as gas chromatography with electron-capture [7] or mass spectrometry [8,9] detection and HPLC with fluorescence [10], electrochemical [11] or mass spectrometry detection [12–14]. However, all of these methods involve an extraction and/or derivatization steps, which demand some well-known drawbacks including the possibility of incomplete derivatization, additional chromatographic interferences and increased method complexity and sample preparation time. The British Pharmacopeia (BP) provides two different potentiometric titration methods for assay of BH and NH. Regarding the related substances, however, two distinct RP-HPLC procedures
1416
A. Mostafavi et al. / Talanta 77 (2009) 1415–1419
Fig. 1. Chemical structures of buprenorphine (a), naloxone (b) and noroxymorphone (c).
have been proposed, one for buprenorphine using UV detection at 288 nm, and another for naloxone impurities at 230 nm. BP specifies five substances as possible impurities for NH; one of them, noroxymorphone (NM, Fig. 1c), the starting material in NH synthesis, may cause gastric disturbance with nausea, vomiting and constipation and is required to be controlled [15]. In the United States Pharmacopoeia (USP), however; without specifying any other impurities than noroxymorphone hydrochloride, a TLC method is described for the determination of this compound and other NH impurities, yet monograph improvement has been requested [16,17]. It is worth noting the lack of official methods for assays of active ingredients and related substances in pharmaceutical preparations in pharmacopoeia, especially when it is currently possible to find a plethora of methods based on HPLC of these compounds. However, to the best of our knowledge, there are no chromatographic methods in the literature for the analysis of any buprenorphine dosage forms. The only reported analysis of buprenorphine in pharmaceutical products includes a spectrophotometric method based on colored ion-pair formation [18]. Hence, an attempt has been made to develop a simple, efficient and selective method for the determination of BH, NH and its major degradation impurity, NM, in sublingual tablets. In this study, HPLC instrumentation with UV detection, which is readily available in most analytical and pharmaceutical laboratories, was used. The method requires no extraction or derivatization steps reducing the total analysis run time to less than 10 min. 2. Experimental 2.1. Materials Reference standards were purchased from Macfarlan Smith (Edinburgh, UK) and were checked against European Pharmacopoeia CRS standards (Strasbourg, France). The common tablet excipients and commercial marketed products were provided by Darou Darman Pars (Tehran, Iran). All other reagents were spectrophotometric or HPLC grade obtained from Romil Chemicals (Loughborough, UK). Water purified via a Milli-Q system, Millipore Corp. (Bedford, USA) was used for all purposes. 2.2. Instrumentation and chromatographic conditions The HPLC system consisted of a Younglin ACME 9000 (Seoul, Korea) equipped with a quaternary pump, online degasser, column heater, autosampler and UV detector. Data collection and analyses were performed using Autochro 2000 software (Younglin). Separation was achieved on C-18 column, Perfectsil Target ODS3 (150 mm × 4.6 mm, 5 m) with a 10 mm × 4.0 mm, 5 m guard column (MZ-Analysentechnik, Mainz, Germany). The elution was isocratic with mobile phase of acetonitrile and 10 mmol L−1 potassium phosphate buffer adjusted to pH 6.0 with orthophosphoric
acid (83:17, v/v). The flow rate was 1.0 mL min−1 and yielded a backpressure of about 740 psi. The column temperature was maintained at 35 ◦ C, the detection was monitored at a wavelength of 210 nm and injection volume was 20 L. 2.3. Standard solutions and calibration graphs for chromatographic measurement Stock standard solutions of BH, NH and NM were separately prepared in methanol using individual EP CRS standards to obtain concentrations of 440, 200 and 200 g mL−1 , respectively. If haziness was observed in NM solution, addition of about 0.5%, 0.1 mol L−1 HCl was sufficient to clarify the solution. Calibration standards at eight levels were prepared by appropriately mixed and further diluted stock standard solutions in the concentration range of 0.022–220 g mL−1 for BH and 0.01–100 g mL−1 for NH and NM. Similarly, quality control (QC) standard solutions were prepared daily by diluting the corresponding stock standard solutions of individual reference standards for the final QC concentrations of 4.8, 48 and 240 g mL−1 for BH and 2.4, 24 and 120 g mL−1 for NH and NM, respectively. Samples in triplicates were made for each concentration and peak areas were plotted against the corresponding concentrations to obtain the calibration graphs. 2.4. Sample preparation A number of 20 accurately weighed tablets were ground into a fine powder using a glass mortar and pestle. A portion equivalent to about four tablets (typically each contains 2.16 mg BH and 0.61 mg NH) was accurately weighed and transferred to a 20 mL volumetric flask. Approximately 15 mL methanol and about 0.5%, 0.1 mol L−1 HCl were added to the flask and the contents were vortex-mixed for 10 min. The flask was adjusted to volume and mixed well. The resulting solution was filtered using 0.45 m PTFE filter into standard analytical glass vials and injected into the HPLC. Three such samples were prepared from each 20-tablet mixture according to the USP criteria and injected triplicate. 2.5. Method validation The method was validated according to the ICH guidelines [19]. The following validation characteristics were addressed: linearity, accuracy, precision, specificity, limits of detection and quantitation and robustness. 2.5.1. System suitability testing (SST) System suitability standard solution which contained 200 g mL−1 BH, 100 g mL−1 NH and 0.1 g mL−1 NM was prepared by appropriately diluting and mixing the corresponding stock standard solutions. System suitability was determined from six replicate injections of the system suitability standard before
A. Mostafavi et al. / Talanta 77 (2009) 1415–1419
sample analysis. According to the monograph, the acceptance criteria for NH were less than 2% R.S.D. and a signal-to-noise ratio of at least ten for the corresponding peak area. For NM, acceptance criteria were less than 2% R.S.D. for peak area and not less than four for the resolution between the peaks corresponding to NM and NH. Resolution was calculated as defined by the USP. 2.5.2. Linearity and range Standard calibration curves were prepared with seven calibrators over a concentration range of 0.22–220 g mL−1 for BH and 0.1–100 g mL−1 for both NH and NM. The data of peak area versus concentration were treated by linear least square regression analysis. The standard curves were evaluated for intra-day and inter-day linearity. 2.5.3. Accuracy To study the reliability and suitability of the developed method, recovery experiments were carried out. Placebo samples were spiked with different amount of BH and NH at 50, 100 and 150% in duplicate for each one (n = 6) over the theoretical values, and NM at 0.1% over the NH content. Measured values were compared with the theoretical concentration. Recovery for pharmaceutical formulations should be within the range 100 ± 5%. The R.S.D. percent of individual measurements was also determined. The results must be less than 5%. 2.5.4. Precision The precision of the developed method was assessed in terms of repeatability and intermediate precision by analyzing three replicate QC standard samples at 50, 100 and 150% levels that cover the calibration ranges for BH and NH, and at a single concentration of 0.1 g mL−1 for NM. The %R.S.D. values of the results corresponding to the peak area and retention time were expressed for intra-day precision and on 3 days for intermediate (inter-day) precision. 2.5.5. Specificity Injections of the extracted placebo were performed to demonstrate the absence of interference with the elution of the BH, NH and NM. Samples containing NH’s main impurities, the NM, were also injected. For determining selectivity of the method, a powder blend of typical tablet excipients containing lactose monohydrate, mannitol, maize starch, povidone K30, citric acid anhydrous granular, sodium citrate, natural lemon and lime flavour, acesulfame potassium and magnesium stearate was prepared and analyzed. All chromatograms were examined to determine if compounds of interest co-eluted with each other or with any additional excipient peaks. 2.5.6. Limits of detection and quantitation The limit of detection (LOD) and limit of quantitation (LOQ) for the procedure were performed on samples containing very low concentrations of analytes under the ICH guidelines. By applying the visual evaluation method, LOD was expressed by establishing the minimum level at which the analyte can be reliably detected. LOQs were considered as the lowest concentration of analytes in standards that can be reproducibly measured with acceptable accuracy and precision. 2.5.7. Robustness The robustness of the method was evaluated by analyzing the system suitability standards and evaluating system suitability parameter data after varying, individually, the HPLC pump flow rate (±10%), organic solvent content (±6%) and column compartment temperature (±14%).
1417
3. Results and discussion 3.1. Method development and optimization NM is a major impurity of NH drug substance. The main target of the chromatographic assay method development was to separate the impurity co-eluted with naloxone. Typically, method development focuses on identifying buffer type, strength and pH, organic solvent and implementing small changes to optimize selectivity and enhance resolution. Initially, NM was found to be co-eluted with naloxone by using different stationary phases such as C-8 and C-18, with mobile phases containing buffers like phosphate, at different pH and temperature, and organic solvents like methanol and acetonitrile. At the first stage, a C-8 column chemistry and potassium phosphate buffer, pH 3.0, were used with methanol as the organic solvent. Though the column was base deactivated for improved peak shape of basic compounds, both peak symmetry and resolution between two compounds were poor. Subsequently, an acceptable peak shape and resolution were achieved by increasing the buffer pH to 6.0, approximately 2.5 and 2.0 pH units above the pKa1 of BH and NH, respectively; and using ternary solvent system consisting of 10% acetonitrile and 70% methanol as organic solvent. Nevertheless, the run time was prohibitively long at approximately 14 min. In order to better exploit the polarity differences between the desired compounds, while maintaining a short run time; two C-18 columns were evaluated. A typical silica-based monolithic column, which proves to dramatically reduce the analysis time, was examined at first. Although the retention time of buprenorphine reduced to about 6.0 min, applying the monolithic column at its best operating conditions (an organic content of 45% ACN at a flow rate of 2.0 mL min−1 and oven temperature of 40 ◦ C), produced neither complete resolution between the first peak (naloxone) and contaminants co-eluted at the beginning, nor better asymmetry factor of 0.7 for studied compounds. The excessive resolution at the beginning of the chromatogram and appropriate peak shape were finally achieved by switching to a base deactivated Perfectsil target ODS3 column. For optimum resolution, peak asymmetry and analysis time, the mobile phase consisting of 10 mmol L−1 potassium phosphate buffer adjusted to pH 6.0 with orthophosphoric acid and acetonitrile (17:83, v/v) was used. To improve repeatability of runs, shorten analysis time and reduce back pressure, which is important to extending the column life time, the column oven temperature was set at 35 ◦ C. In the optimized conditions, NM and the drug substances naloxone and buprenorphine were well separated with resolutions of more than 10 and 18, and asymmetry factors very close to 1.0. The optimal wavelength was established experimentally after measuring all spectra in mobile phase and testing the detector response of analytes using UV absorbance scanned over the range of 190–400 nm. Although the absorption maxima recommended by BP for BH and NH were 288 and 230 nm, respectively; it was shown that 210 nm is the optimal wavelength to maximize the sensitivity and has no interference with other components of the formulation. Under the experimental conditions investigated, the retention times for naloxone, noroxynorphone and buprenorphine were 2.40, 3.84 and 8.06 min, respectively (Fig. 2).
3.2. Method validation When a method has been optimized it must be validated before practical use. By following the ICH guidelines for analytical method validation, Q2 (R1), the SST was performed and the validation characteristics were addressed.
1418
A. Mostafavi et al. / Talanta 77 (2009) 1415–1419
Table 1 Linearity parameters for the simultaneous estimation of NH, NM and BH. Parameter −1
Linearity range (g mL ) Slope Intercept Correlation coefficient (r) Residual sum of squares
Naloxone hydrochloride
Noroxymorphone
Buprenorphine hydrochloride
0.10–100 67.70 ± 0.24 28.90 ± 7.01 0.9997 ± 0.0001 0.00028
0.10–100 75.75 ± 0.55 193.10 ± 0.59 0.9961 ± 0.0005 0.00378
0.22–220 79.80 ± 0.12 −35.20 ± 6.86 0.9997 ± 0.0001 0.00036
Values are reported as mean ± S.D. of three calibration curves generated on three consecutive days (n = 3). Seven concentrations in the linearity range were evenly distributed. Table 2 Method validation results for studied compounds. Naloxone hydrochloride
Noroxymorphone
Buprenorphine hydrochloride
SST Theoretical plates Asymmetry (As ) Resolution (Rs ) Repeatabilitya , tR (%R.S.D.) Repeatabilitya , A (%R.S.D.)
6485 1.0 3.5 0.3 0.6
7599 1.1 10.3 1.7 1.5
8668 1.0 18.8 1.3 0.6
Validation Precisiona , b (%R.S.D.) Accuracya , b (%R.S.D.) Accuracy (%recovery) Selectivity LOD (g mL−1 ) LOQ (g mL−1 )
3.1 4.1 102.8 No interference 0.01 0.10
2.9 5.0 96.2 No interference 0.01 0.10
4.2 3.3 102.3 No interference 0.02 0.22
a Intra-day precision (repeatability), inter-day precision and accuracy were tested by analyzing three replicate QC standard samples at 50, 100 and 150% levels for BH and NH and at a single concentration of 0.1 g mL−1 for NM under the guidelines of ICH. b Inter-day precision and accuracy were determined with three replicates on three consecutive days.
3.2.1. System suitability The system suitability test ensures the validity of the analytical procedure as well as confirms the resolution between different peaks of interest. All critical parameters tested met the acceptance criteria on all days. According to the monograph, however, the area of an individual secondary chromatographic peak, which appears in the test sample, should not exceed 0.5% of the naloxone peak area, and the total area of all secondary peaks, 1% naloxone peak area. In addition, the resolution between the peak corresponding to naloxone and NM in the chromatogram obtained with reference solution should not be less than 4 at S/N of at least 10 [15]. Adequate resolution of >10 between the naloxone and its impurity, NM, meets the acceptance criteria indicated in the monograph. As
shown in the chromatogram, all three analytes are eluted by forming symmetrical single peaks well separated from the solvent front (Fig. 2). 3.2.2. Linearity and range For the construction of calibration curves, seven calibration standard solutions were prepared over the concentration range of 0.22–220 g mL−1 for BH and 0.1–100 g mL−1 for NH and NM. The results, summarized in Table 1, show a good correlation between analytes peak area and concentration with r > 0.996 (n = 7). 3.2.3. Accuracy and precision Accuracy and precision were established across the analytical range for BH, NH and NM. The intra- and inter-day accuracy and precision were calculated from the QC samples (Table 2). Repeatability (intra-day precision) of the analytical method was found to be reliable based on %R.S.D. (<2%) corresponding to the peak areas and retention times. Intermediate precision (inter-day accuracy) was demonstrated on different days and evaluating the peak area data at three QC standards that cover the assay method range. The %R.S.D. values were less than 5% and illustrated the good precision for the analytical method. For determining accuracy, placebo solutions spiked with reference standards were used. The recovery was 100 ± 5% for all samples with %R.S.D. less than 5%. 3.2.4. Specificity Injections of the extracted placebo were performed to demonstrate the absence of interference with the elution of the drugs and impurity. These results demonstrate that there was no interference from other materials in the tablet formulation; therefore, confirm the specificity of the method (Fig. 2).
Fig. 2. Representative chromatograms obtained for the mobile phase, placebo and QC standard (corresponding to the 24, 24 and 48 g mL−1 NH, NM and BH, respectively).
3.2.5. Sensitivity The limit of detection and limit of quantitation decide about the sensitivity of the method. Tests for the procedure were performed on samples containing very low concentrations of analytes based
A. Mostafavi et al. / Talanta 77 (2009) 1415–1419
1419
Table 3 Method validation data for robustness study. Parameter altered
Optimized chromatographic conditionsa Increased organic solvent (16:84) Decreased organic solvent (18:82) Increased flow rate (1.1 mL min−1 ) Decreased flow rate (0.9 mL min−1 ) Decreased column temperature (30 ◦ C)
Retention time, tR (min)
Resolution (Rs )
Asymmetry (As )
N
NM
B
N
NM
B
N
NM
B
2.4 2.3 2.4 2.2 2.6 2.5
3.8 3.7 3.9 3.7 3.9 4.0
8.0 8.0 8.1 7.8 8.3 8.3
3.5 2.5 3.0 1.8 4.2 3.6
10.3 8.75 9.4 8.2 9.3 10.5
18.8 18.1 20.1 17.8 18.7 20.1
1.0 0.9 1.1 1.0 1.1 1.1
1.1 1.0 1.15 1.1 1.2 1.2
1.0 1.0 1.1 1.0 1.15 1.1
a Chromatographic conditions were 10 mmol L−1 potassium phosphate buffer, pH 6.0/ACN (17:83) maintained with flow rate of 1.0 mL min−1 , temperature at 35 ◦ C and detection at 210 nm.
amount of NM was found below the detection limit (0.01 g mL−1 ) and, therefore, should be less than 0.1% over the NH content. 4. Conclusion A simple isocratic reversed-phase HPLC method proposed was found to be accurate, precise, linear across the analytical range and robust. The method was specific for the simultaneous determination of BH, NH and its primary impurity NM in a tablet formulation. All the parameters for three substances met the criteria of the ICH guidelines for method validation. The method could therefore be recommended for routine quality control analysis of raw materials and various buprenorphine and naloxone dosage in tablet formulations by assaying for potency and accurately monitoring the NM impurity. Acknowledgement
Fig. 3. Representative chromatogram obtained for a commercially available BH/NH (4:1) drug product (N: naloxone; B: buprenorphine).
Kimia Sanat Jahan Co. is acknowledged for kindly furnishing the home edition of Autochro-2000 HPLC software. References
on the visual evaluation method. In this method, LOD is determined by the analysis of samples with known concentration of analyte and by establishing the minimum level at which the analyte can be reliably detected. Accordingly, the LOQ is determined by the analysis of samples with known concentration of analytes and by establishing the minimum level at which the analyte can be quantified with acceptable accuracy and precision (R.S.D. <2%). The LOD and LOQ values were found to be 0.022 and 0.22 g mL−1 for BH and 0.01 and 0.1 g mL−1 for both NH and NM. 3.2.6. Robustness To ensure the insensitivity of the developed HPLC method to minor changes in the experimental conditions, it is important to demonstrate its robustness. None of the alterations caused a significant change in resolution between naloxone and NM, peak area R.S.D., USP tailing factor and theoretical plates (Table 3). Although the changes in retention times were more significant, separation was sufficient and quantitation was still possible. 3.3. Analysis of the marketed product The validated method was used in the analysis of a marketed product in a 70 mg tablet dosage form with dose strength of 2.0 mg BH/0.5 mg NH. Representative chromatogram is shown in Fig. 3. The results for the drugs assay and the concentration of impurity NM were in good agreement with the label claims. BH content was between 91.4 and 93.5%, and for NH between 97.2 and 104.7%. The
[1] S.L. Walsh, K.L. Preston, G.E. Bigelow, M.L. Stitzer, J. Pharmacol. Exp. Ther. 247 (1995) 361. [2] G.B. Collins, M.S. McAllister, Clev. Clin. J. Med. 74 (2007) 514. [3] J.W. Lewis, in: A. Cowan, J.W. Lewis (Eds.), Buprenorphine: Combating Drug Abuse with a Unique Opioid, Wiley-Liss, New York, 1995, p. 151. [4] R.E. Johnson, J.H. Jaffe, P.J. Fudala, J. Am. Med. Assoc. 267 (1992) 2750. [5] W. Ling, D.R. Wesson, C. Charuvastra, C.J. Klett, Arch. Gen. Psychiatry 53 (1996) 401. [6] C.N. Chiang, R.L. Hawks, Drug Alcohol Depend. 70 (2003) S39. [7] E.T. Everhart, P. Cheung, P. Shwonek, K. Zabel, E.C. Tisdale, P. Jacob III, J. Mendelson, R.T. Jones, Clin. Chem. 43 (1997) 2292. [8] A.M. Lisi, R. Kazlauskas, G.J. Trout, J. Chromatogr. Biomed. Appl. 692 (1997) 67. [9] L. Debrabandere, M. Van Boven, L. Laruelle, P. Daenens, Anal. Chim. Acta 275 (1993) 295. [10] S.Y. Liu, K.S. Liu, C.H. Kuei, J.I. Tzeng, S.T. Ho, J.J. Wang, J. Chromatogr. B 818 (2005) 233. [11] M.A. García-Fernández, M.T. Fernández-Abedul, A. Costa-García, Chromatographia 53 (2001) 704. [12] D.E. Moody, M.H. Slawson, E.C. Strain, J.D. Laycock, A.C. Spanbauer, R.L. Foltz, Anal. Biochem. 306 (2002) 31. [13] S. Pirnay, F. Hervé, S. Bouchonnet, B. Perrin, F.J. Baud, I. Ricordel, J. Pharm. Biomed. Anal. 41 (2006) 1135. [14] M.E. Rodriguez-Rosas, M.R. Lofwall, E.C. Strain, D. Siluk, I.W. Wainer, J. Chromatogr. B 850 (2007) 538. [15] The British Pharmacopoeia 2007, The Stationary Office (TSO), Health Ministry, London, 2006. [16] The United State Pharmacopoeia, 27th ed., The United States Pharmacopoeial Convection, Rockville, MD, 2004. [17] http://www.usp.org/USPNF/submitMonograph/improveMon.html. [18] M. Amanlou, P. Khosravian, E. Souri, O.G. Dadrass, R. Dinarvand, M.M. Alimorad, H. Akbari, Bull. Korean Chem. Soc. 28 (2007) 183. [19] ICH guidelines Topic Q2 (R1) Validation of Analytical Procedures: Methodology, obtained from web site: www.emea.europa.eu/htms/human/ich/ ichquality.htm.