Scientific paper Improved HPLC Method with the Aid of Chemometric Strategy: Determination of Loxoprofen in Pharmaceutical Formulation P. Venkatesan,1* V. Sree Janardhanan,1 C. Muralidharan,2 and K. Valliappan1 1 Department of Pharmacy, Faculty of Engineering and Technology, Annamalai University, Annamalainagar, TN 608 002, India 2 Department of Manufacturing Engineering, Faculty of Engineering and Technology, Annamalai University, Annamalainagar, TN 608 002, India * Corresponding author: E-mail: venkatesan1978@gmail.com Tel.: +91 4144 239738; Fax: +91 4144 238145 Received: 15-05-2011 Abstract Loxoprofen belongs to a class of Nonsteroidal anti-inflammatory drug acts by inhibiting isoforms of cyclo-oxygenase 1 and 2. In this study an improved RP-HPLC method was developed for the quantification of loxoprofen in pharmaceutical dosage form. For that purpose an experimental design approach was employed. Factors-independent variables (organic modifier, pH of the mobile phase and flow rate) were extracted from the preliminary study and as dependent variables three responses (loxoprofen retention factor, resolution between loxoprofen probenecid and retention time of probenecid) were selected. For the improvement of method development and optimization step, Derringer's desirability function was applied to simultaneously optimize the chosen three responses. The procedure allowed deduction of optimal conditions and the predicted optimum was acetonitrile: water (53:47, v/v), pH of the mobile phase adjusted at to 2.9 with ortho phosphoric acid. The separation was achieved in less than 4minutes. The method was applied in the quality control of commercial tablets. The method showed good agreement between the experimental data and predictive value throughout the studied parameter space. The optimized assay condition was validated according to International conference on harmonisation guidelines to confirm specificity, linearity, accuracy and precision. Keywords: Central composite design, derringer's desirability function, loxoprofen, multiple response optimization, probenecid, reversed-phase HPLC 1. Introduction Nonsteroidal anti-inflammatory drugs (NSAIDs) are widely used for the treatment of pain and inflammation. NSAIDs produce their therapeutic effect by inhibiting the cyclooxygenase (COX) enzymes, which are involved in the biosynthesis of prostaglandins (PGs).1,2 Loxoprofen, 2-[4-(2-oxocyclopentylmethyl) phenyl]-propionate with two chiral centers, is marketed as an equal parts mixture of four stereoisomers. Loxoprofen sodium is an important non-steroidal anti-inflammatory drug (NSAID) of the 2-arylpropionic acid group used for the treatment of rheumatoid arthritis and osteoarthritis. Loxoprofen is a prodrug which produces effects after being absorbed from the gastrointestinal tract followed by conversion to an active metabolite. Loxoprofen has an activity to treat inflammatory rheumatoid diseases and relieve acute pain.3,4,5 It is effective against period pains, pain after surgery and fever. Loxoprofen available in pharmaceutical formulations as tablets and transdermal patches.6'7 A literature search revealed that seven methods are available for the determination of loxoprofen in pharmaceutical formulations and biological fluids. Hideo Naga-numa et al8 developed a simple and sensitive high-performance liquid chromatographic procedure to determine lo-xoprofen and its diastereomeric alcohol metabolites in Loxoprofen (LOX) Probenecid (SI) Fig.1. The chemical structures of analyte (LOX) and internal standard (IS) pharmaceutical or industrial environment, there is a need to optimize multiple responses (analysis time and resolu- tion) simultaneously.2 1 To achieve global optimiza- tion of multiple responses the Derringer's desirability function (Multi-Criteria decision making) has been applied.24,25,26,27,28 Hence there was a need to develop an improved HPLC method for determination of loxoprofen in pharmaceutical formulations applying a chemometric protocol. The aim of this work is to (i) develop an improved HPLC method suitable for the routine quality control analysis of loxoprofen in a pharmaceutical laboratory and (ii) provide information on the sensitivity of chromato-graphic factors and their interaction effects on the separation characteristics. The chromatographic factors that had significant effects on analysis time were optimized using a central composite design and response surface methodology. biological specimens by fluorescence labeling. Hisomu Nagashima et al9 reported a column liquid chromato-graphy method for the simultaneous determination of the enantiomers of loxoprofen sodium and its metabolites in human urine. Hideko Kanazawa et alw established the chiral separation of loxoprofen in plasma by chiral column liquid chromatography with a circular dichroism-ba-sed detector. Kyo-Seok Choo et alu developed a simultaneous determination of loxoprofen and its diastereomeric alcohol metabolites in human plasma and urine by a simple HPLC-UV detection method. Hea-Young Cho et al12 reported a direct and simultaneous analysis of loxoprofen and its diastereomeric alcohol metabolites in human serum by on-line column switching liquid chromatography. Tomonori Murakami et al13 developed a method for identification of degradation products in loxoprofen sodium adhesive tapes by liquid chromatography-mass spectro-metry and dynamic pressurized liquid extraction-solid-phase extraction coupled to liquid chromatography-nuc-lear magnetic resonance spectroscopy. Although all the above methods have been applied for quantification of lo-xoprofen in biological matrices, due to their complexity they may not be suitable for routine analysis. The Japanese pharmacopeia14 describes an HPLC method for determination of loxoprofen in bulk drug. This HPLC method has reportedly a run time of about 7.0 min, but did not provide the details on the chromatographic variables, viz. capacity factor, resolution, asymmetric factor. None of the above methods reported above employed a systematic optimization procedure for the separation and quantification of loxoprofen, but a time-consuming trial-and error approach resulting in an apparent optimum only. Information on the sensitivity of the factors on the analytes separation and interaction between factors is not available. In this work a chemometric procedure is applied to realize the above objective15,16,17,18,19. However, since the HPLC method is intended to be applied for the 2. Experimental 2.1. Apparatus Chromatographic measurements were made on a Shimadzu (Tokyo, Japan) model which consisted of a LC10AD and LC10 ADvp solvent delivery module, SPD 20 PDA detector, a Rheodyne injector (model 7125, USA) valve fitted with a 20 ^l loop, and PDA detector (SPD-20). The system was controlled through a system controller (SCL-10A) and a personal computer using a Shimadzu chromatographic software (LC Solution, Release 1.11SP1) installed on it. The mobile phase was degassed using a Branson sonicator (Branson Ultrasonics Corporation, USA). Absorbance spectra were recorded using an UV-Visible spectrophotometer (Model UV-1601PC, Japan) employing a quartz cell with 1.00 cm of path length. 2. 2. Softwares Experimental design, data analysis and desirability function calculations were performed using Design-Expert® trial version 7.0.0. (Stat-Ease Inc., Minneapolis). 2. 3. Chemicals and Reagents Working standards of loxoprofen (99.79%) were donated by M/S Micro labs limited, Hosur. India. Probenecid (PRB) (>99%) was purchased from Fluka, Buchs, Switzerland. Acetonitrile (ACN) of HPLC grade and di-potassium hydrogen phosphate and phosphoric acid were of analytical-reagent grade supplied by M/S SD Fine chemicals, Mumbai, India. The HPLC grade water was prepared using a Milli-Q Academic system, Millipore, Bangalore, India. The pharmaceutical Loxomac® tablets (containing loxoprofen 60 mg) were purchased from Macleods Pharmaceuticals, Mumbai, India. 2. 4. Standard Solutions Stock standard solutions of loxoprofen and probenecid (1 mg/ml) were prepared in mobile phase. The prepared stock solution was stored at 4 °C protected from light. Working standard solutions were freshly obtained by diluting the stock standard solutions with mobile phase during the analysis day. Calibration curves reporting peak area ratios of loxoprofen to that of the IS versus drug concentrations were established in the range of 1.5-15 pg/ml for loxoprofen, in the presence of probenecid (7.5 pg/ml) as an internal standard. Standard solutions prepared for the optimization procedure constituted loxoprofen and IS at 15 and 7.5 pg/ml, respectively. 2. 5. Sample Preparation Twenty tablets were weighed and finely powdered. In the case of capsule dosage, the contents of the capsule were mixed thoroughly. An amount of capsule/tablet powder equivalent to 15 mg loxoprofen was accurately weighed and transferred in a 10 ml volumetric flask; a suitable quantity of IS was added followed by 5 ml of mobile phase. This mixture was subjected to sonication for 10 min for complete extraction of the drug and the solution was made up to the mark with mobile phase to obtain a concentration of loxoprofen and IS of 15 and 7.5 pg/ml, respectively. The solution was centrifuged at 4000 rpm for 10 min; the clear supernatant was collected and filtered through a 0.2 pm membrane filter (Gelman Science, India) and 20 pl of this solution was injected for HPLC analysis. 2. 6. Chromatographic Procedure Chromatographic separations were carried out on a Phenomenex® C18 analytical column (150 mm x 4.6 mm 1.d., 5 pm) connected with a Phenomenex® C18 guard cartridge (4 mm x 3 mm i.d., 5 pm). The mobile phase consisted of acetonitrile: water; the pH of the mobile phase was adjusted to 2.9 with 10% ortho phosphoric acid. In order to increase the sensitivity for the less concentrated compound and to decrease the background of mobile phase, a wavelength of 220 nm was selected for detection. The injection volume of the sample was 20 pl. The HPLC system was used in an air-conditioned laboratory atmosphere (20 ± 2 °C). 2. 7. Validation Validation studies were conducted using the optimized assay conditions based on the principles of validation described in the ICH guidelines "Text on Validation of Analytical Procedures"29 and "Q2B, Validation of Analytical Procedures: Methodology".30 Key analytical parameters, including, specificity, accuracy, precision, linearity, detection limit and quantitation limit were evaluated. To study specificity, a placebo containing starch, lactose monohydrate, aerosil, hydroxy propyl methylcellulose, titanium dioxide and magnesium stearate was used. Calibration curves were constructed in the range of 0.05 to 1.0% of the target analyte concentration for the limit of detection and quantification.31 Also the robustness of the proposed method was assessed with respect to small alterations in the ACN concentration (53% ± 0.5) and the pH value (2.9 ± 0.2). 3. Results and Discussion 3.1. Optimization Design and Analysis The central composite design can be applied to optimize the separation and to assist in the development of a better understanding of the interaction of several chroma-tographic factors on separation quality.32 In this work, the important chromatographic factors were selected and optimized by a central composite design experiment. The selection of factors for optimization was based on preliminary experiments and prior knowledge from literature, as well as certain instrumental limitations. From preliminary experiments the key factors selected for optimization process were ACN concentration (A), pH of the mobile phase (B) and flow rate (C). Table 1 shows the set of conditions applied to these factors to optimize the determination of loxoprofen. As can be seen in this table, the conditions for each factor were limited: ACN concentration (45-55%), pH (2.5-3.0) and flow rate (1.0-1.5 ml/min). As response variables, the retention time of IS (tR2), capacity factor (k) and the resolution between two pairs, loxoprofen-IS (Rs12) were chosen. All experiments were performed in randomized order to minimize the effects of uncontrolled variables that may introduce a bias on the measurements. Replicates (n = 6) of the central points were performed to estimate the experimental error. For an experimental design with three factors, the model including linear, quadratic, and cross terms can be expressed as (1) where Y is the response to be modeled, P is the regression coefficient and X1, X2 and X3 represent factors A, B and C, respectively. To obtain a simple and yet realistic model, the insignificant terms (P > 0.05) are eliminated from the model through a 'backward elimination' process. The statistical parameters obtained from the ANOVA for the reduced models are given in Table 2. Since R2 always decreases when a regressor variable is eliminated from a regression model, in statistical modeling the adjusted R2 which takes the number of regressor variables into account, is usually selected.33 In the present study, the adjusted R2 values were Table 1. Central composite rotatable design arrangement and responses3 Design Factor levels Responses points A B C k1 tR2 (%v/v) (pH) (ml/min) Rs 1,2 1 52.97 2.60 1.42 2.00 3.15 4.60 2.01 3.15 4.61 2 50.00 2.75 1.30 2.43 4.21 5.87 2.43 4.21 5.88 3 50.00 2.50 1.30 3.05 5.49 6.99 3.05 5.49 6.99 4 52.97 2.90 1.42 2.06 3.85 4.65 2.08 3.82 4.66 5 45.00 2.75 1.30 3.40 5.28 7.87 3.41 5.28 7.88 6 47.03 2.60 1.18 2.88 5.61 7.10 2.86 5.62 7.11 7 50.00 2.75 1.30 2.43 4.21 5.87 2.43 4.21 5.88 8 47.03 2.90 1.42 2.97 4.41 6.65 2.95 4.41 6.60 9 50.00 2.75 1.30 2.43 4.21 5.87 2.43 4.21 5.88 10 50.00 2.75 1.30 2.43 4.21 5.87 2.43 4.21 5.88 11 55.00 2.75 1.30 1.81 3.23 4.29 1.79 3.24 4.23 12 52.97 2.60 1.18 1.99 4.02 5.00 2.00 4.01 5.04 13 50.00 3.00 1.30 2.11 3.79 5.07 2.11 3.64 4.86 14 47.03 2.90 1.18 2.99 5.60 7.10 2.97 5.61 7.11 15 50.00 2.75 1.10 2.44 5.24 6.28 16 50.00 2.75 1.30 2.43 4.21 5.87 17 50.00 2.75 1.30 2.43 4.21 5.87 18 50.00 2.75 1.50 2.45 3.50 5.49 19 47.03 2.60 1.42 2.86 4.42 6.64 20 52.97 2.90 1.18 2.06 3.15 4.56 a Randomized well within the acceptable limits of R2 > 0.8034 which revealed that the experimental data show a good fit with the second-order polynomial equations. For all the reduced models, P values < 0.05 are obtained, implying that these models are significant. The adequate precision value is a measure of the "signal (response) to noise (deviation) ratio". A ratio greater than 4 is desirable35. In this study, the ratio was found to be in the range of 15.61 to 29.31, which indicates an adequate signal and therefore the model is significant for the separation process. The coefficient of variation (CV) is a measure of reproducibility of the model and as a general rule a model can be considered reasonably reproducible if it is less than 10%.35 The CV for all the models was found to be less than 10%. As can be seen in table 2, the interaction term with the largest absolute coefficients among the fitted models is A (+0.28) of the tR2 model. The positive interaction between A and C is statistically significant (P = 0.0001) for t-R2. The study reveals that changing the fraction of ACN from low to high results in a rapid decline in the retention time of loxoprofen, both at low and high flow rate. This interaction is synergistic as it led to a decrease in run time. The existence of such interactions emphasizes the necessity to carry out active multifactor experiments for optimization of the chromatographic separation. Perturbation r 0.000 T o.saa Deviation from reference point (coded unit) Fig 2. Perturbation plot showing the effect of each of the independent variables on tR2 while keeping other variables at their respective midpoint levels Table 2. Reduced response modelsa and statistical parameters obtained from ANOVA (after backward elimination) Response Regression model Adjusted Model %C.V Adequate R2 P-value precision k, 2.49 - 0.46A 0.8439 0.000 6.67 29.31 tR2 4.30 - 0.68A- 0.23B - 0.40C + 0.28AC 0.8009 0.000 8.19 15.61 Rs1,2 5.87 - 1.08A -0.28B -0.18C 0.9185 0.000 4.83 28.65 a Only significant coefficients with P < 0.05 are included. Factors are in coded levels. In order to gain a better understanding of the results, the predicted models are presented in Fig. 2 in a perturbation plot.36 This graph shows how the response changes as each factor moves from a chosen reference point, with all other factors held constant at the reference value. A steep slope or curvature in a factor indicates that the response is sensitive to that factor. Hence, the plot shows that factor A mostly affected the analysis time (tR2), followed by factor C and then B. 3. 2. Multi-Criteria Decision Making In the present study, to optimize three responses with different targets, Derringer's desirability function, was used.23 The Derringer's desirability function, D, is defined as the geometric mean, weighted, or otherwise, of the individual desirability functions. The expression that defines the Derringer's desirability function is: P 3 (2) Where pi is the weight of the response, n the number of responses and di is the individual desirability function of each response obtained from the transformation of the individual response of each experiment. The scale of the individual desirability function ranges between di = 0, for a completely undesired response, to di = 1 for a fully desired response. Weights can range from 0.1 to 10. Weights lower than 1 give less emphasis to the goal, whereas weights greater than 1 give more emphasis to the goal (in both cases, di varies in a non linear way while approaching to the desired value). With a weight of 1, di varies in a linear way. In the present report we chose weights equal to 1 for all the six responses. A value of D different to zero implies that all responses are in a desirable range simultaneously and consequently, for a value of D close to 1, the combination of the different criteria is globally optimal, so as the response values are near target values. Table 3. Criteria for the optimization of the individual responses. Res- Lower Upper Criteria ponse limit limit Goal Importance ki 1.8 3.413 Target = 2 4 «2 3.156 5.621 Minimize 5 Rsl,2 4.24 7.88 Minimize 4 The criteria for the optimization of each individual response are shown in Table 3, and it is proposed for selecting an optimum experimental condition for analyzing routine quality control samples. As can be seen in Table 3, two responses (tR2 and Rs1,2) were minimized in order to shorten the analysis time. On the other hand, k1 was targe- ted at 2.00. The importance can range from 1 (the least important) to 5 (the most important), which gives emphasis to a target value. For instance, a high importance value of 5 was assigned to the tR2 response as a short analysis time is usually preferred for routine analysis. Following the conditions and restrictions above, the optimization procedure was carried out. The response surface obtained for the global desirability function is presented in Fig. 3. The coordinates producing the maximum desirability value (D = 0.971) were: ACN concentration of 53%, pH of the mobile phase 2.9 and flow rate of 1.42 ml/min. The predicted response values corresponding to the latter value of D were: k1 = 2.0, tR2 = 3.25 min, Rs12 = 4.32. The prediction efficiency of the model was confirmed by performing the experiment under the optimal condition and the corresponding chromatogram is shown in Fig. 4.The agreement between experimental and predicted responses for the predicted optimum are shown in Table 4. The errors for retention factor, retention time and resolution were found to be in good agreement:37 2.04, 0.61 and 1.88%, respectively. 3. 3. Assay Method Validation The optimised assay method is specific in relation to the placebo used in this study because there was no excipient peak co-eluted with the analytes and IS (Fig. 4). An excellent linearity was established at five levels in the ran- Fig. 3. Graphical representation of the overall desirability function D. ACN concentration (A) is plotted against flow rate (C) with factor B held constant at 2.75 pH ge of 1.5-15 ug/ml for loxoprofen with an R2 better than 0.9989. The slope and intercept for the calibration curve of loxoprofen were 0.322 and -0.005, respectively. The LOD and LOQ for loxoprofen were estimated as 1.74 and 5.89 LOX PRB K A LOX PRB B C T—I—i—I—I—I—1—»—i—'—1—1—)—1—i—1—i—i—i—r 0 2 4 6 8 10 Tone (min) Fig. 4. Chromatograms corresponding to (A) a real sample of Lo-xomac tablets containing Lox( 10.08 ^g/ml), IS (7.48 ng/ml): (B) a synthetic mixture of lox (10.06 ng/ml), IS (7.48 ng/ml): (C) mobile phase solution under optimum condition ng/ml, respectively. Accuracy (n = 9), assessed by spike recovery, was found to be 99.86 and within acceptable ranges of 100 ± 2%.38 The intra and inter-assay precision (n = 6) was confirmed since, the %C.V. were well within the target criterion of <2.39 The robustness study reveals that small changes did not alter the retention times, retention factor and resolutions more than 3% and therefore it would be concluded that the method conditions are robust. 3. 4. Application of the Method The proposed RP-HPLC method was applied to the quantitative analysis of real samples (Loxomac® tablets) containing loxoprofen. Representative chromatograms are presented in Fig. 4. The results achieved when analyzing Loxomac® tablets was 861.08 (0.09) mg of Loxoprofen, with the values in parenthesis representing the %C.V. of the six replicates. Good agreement was found between the assay results and the label claim of the product. The %C.V. tablets was <2, indicating the precision of the analytical methodology. 4. Conclusion Statistically-based experimental designs proved to be a valuable approach in optimizing selectivity-controlling parameters for the determination of loxoprofen in commercial tablets. The significant factors were optimized by applying the central composite design and surface response methodology. The objective responses, resolution and the analysis time, were then simultaneously optimized by applying the Derringer's desirability function, a multi-criteria decision making tool. The improved method showed higher sensitivity and shorter analysis time than the existing methods making it viable to be implemented for routine quality control analysis in a pharmaceutical laboratory. The method was validated; the validation study supported the selection of the assay conditions by confirming that the assay was specific, accurate, linear, precise, and robust. 5. References 1.J. R. Vane, Nature New boil, 1971 pp 231-235. 2. J. R. Vane, Y. S. Bakhle, R. M. Botting. Annu. Rev. Pharmacol. Toxicol., 1998 38, 97-120. 3. K. Matsuda, K. Ohnishi, T. Sha, M. Yamazaki, Y. Tanaka, K. Tanaka, Jpn. J Inflam., 1982, 2, 263. 4. T. Yamaguti, T. Kojima, K. Kobayashi, Y. Endo, Y. Misawa, E. Nakajima, E. Misaka, K. Tanaka, Jpn. J. Inflam., 1983, 3, 63. 5.H. Naganuma, Y. Mochizuki, Y. Kawahara, J Clin Therap Med., 1986, 2, 1219. 6. Tohru Araki, Teruhiko Yokoyama, Motoo Araki and Seiji Fu-ruya. J Acta Med Okayama. 2008, 62, 373-378. 7. Shinichi Yoshikawa, Ryo Murata, Shigenari Shida, Koji Uwai, Tsuneyoshi Suzuki, Shunji Katsumata and Mitsuhiro Takeshita. J. Chem Pharm Bull., 2010, 58, 34-37. Table 4. The comparison of experimental and predictive values of different objective functions under optimal conditions. Optimum ACN pH Flow tR2 k1 Rs1>2 conditions (%) (ml/min) I Desirability value (D) = 0.971 52.97 2.90 1.42 Experimental 3.23 1.96 4.24 Predictive 3.25 2.0 4.32 Error 0.61 2.04 1.88 8. Hideo Naganuma and Yukinori Kawahara. J. Chromatography., 1990, 530, 387-396. 9.Hisomu Nagashima, Yorihisa Tanaka and Ryozo Hayashi. J Chromatography., 1985, 345, 373-379. 10. Hideko Kanazawa, Akane Tsubayashi, Yoshiko Nagata, Yos-hikazu Matsushima, Chiharu Mori, Junko Kizu, Megumu Higaki. J. Chromatogr A., 2002, 948, 303-308. 11.Kyo-Seok Choo, In-Wha Kim, Jae-Kyung Jung, Young-Ger Suh, Suk-Jae Chung, Min-Hwa Lee, Chang-Koo Shim. J. Pharm Biomed Anal., 2001, 25, 639-650. 12. Hea-Young Cho, Chan-Ho Park, Yong-Bok Lee. J Chromatogr B., 2006, 835, 27-34. 13.Tomonori Murakami, Takao Kawasaki, Akira Takemura, Naoto Fukutsu, Naoyuki Kishi, Fumiyo Kusu. J Chromatogr A., 2008, 1208, 164-174. 14. The Japanese pharmacopoeia 2006, 15, 828-829. 15. J. A. Lewis, L. R. Snyder, J. W. Dolan, J Chromatogr A., 1996, 721, 15-29. 16. K. Valliappan, K. Kannan, R. Manavalan, C. Muralidharan, Indian J Chem., 2002, 41A, 1334-1340. 17. R. H. Myers, D. Montgomery, Response Surface Methodology, Wiley, New York. 1995. 18. N. Matthijs, D. L. Massart, M. Maftouh, Y. V. Heyden, Che-mometrics and Chromatography, 15th International Symposium on Pharmaceutical and Biomedical Analysis, PBA 2004, Florence, Italy, 2-6 May 2004 19. E. Morgan, Chemometrics - Experimental Design, Wiley, Chichester 1991. 20. S. N. Deming, J. Chromatogr A., 1991, 550, 15-25. 21. M. R. Hadjmohammadi, F. Safa, J Sep Sci 2004, 27, 997-1004. 22. E. C. Harrington, Ind Quality Control 1965, 21, 494-498. 23. G. Derringer, R. Suich, J Qual Technol., 1980, 12, 214-219. 24. T. Sivakumar, R. Manavalan, C. Muralidharan, K. Valliappan, J Pharm Biomed Anal., 2007, 43, 1842-1848. 25. T. Sivakumar, R. Manavalan, C. Muralidharan, K. Valliappan, J Sep Sci., 2007, 30, 3143-3153. 26. Khalil Farhadi, Morteza Bahram, Donya Shokatynia, Floria Salehiyan, Talanta. 2008, 76, 320-326. 27. Ljiljana Zivanovic, Ana Protic, Mira Zecevic, Biljana Jocic, Mirjana Kostic, J Pharm Biomed Anal., 2009, 50, 640-648. 28. C. Marcia, Breitkreitz, Isabel CSF Jardim, Roy E, Bruns, J Chromatogr A., 2009, 1216, 1439-1449. 29. International conference on harmonization (ICH), Q2A: text on validation of analytical procedures: definitions and terminology US FDA federal register, 1995. 30. International conference on harmonization (ICH), Q2B: validation of analytical procedures: methodology US FDA federal register, 1997. 31. J. B. Crowther, in: S. Ahuja, S. Scypinski. (Eds.), Handbook of Modern Pharmaceutical Analysis, Academic Press, New York, 2001 pp. 415-443. 32. Y. Wang, M. Harrison, B. J. Clark, J Chromatogr A., 2006, 1105, 199-207. 33. J. C. Parajo, J. L. Alonso, M. A. Lage, D. Vazquez, Biopro-cess Eng., 1992, 8, 129-136. 34. T. Lundstedt, E. Seifert, L. Abramo, B. Thelin, M. Nystr, J. Pettersen, R. Bergman, R., Chemom. Intell. Lab. Syst. 1998, 42, 3-40. 35. Q. Beg, V. Sahai, R. Gupta, Process Biochem., 2003, 39, 203-209. 36. T. N. Decaestecker, W. E. Lambert, C. H. Van Peteghem, D. Deforce, J. F. Van Bocxlaer, J Chromatogr A., 2004, 1056, 57-65. 37. P. Wester, J. Gottfries, K. Johansson, F. Klintebck, B. Winb-lad, J Chromatogr B., 1987, 415, 261-274. 38. T. Sivakumar, R. Manavalan, K. Valliappan, Acta Chromato-gr., 2007, 18, 130-142. 39. G. Kleinschmidt, in: J. Ermer, J. H. M. Miller (Eds.), Method Validation in Pharmaceutical Analysis. A Guide to Best Practice, Wiley-VCH, Weinheim, 2005, pp. 195-226 Povzetek Loksoprofen spada v skupino nesteroidnih protivnetnih zdravil. Deluje preko inhibicije izoform ciklooksigenaze 1 in 2. V pričujoči študiji smo razvili izboljšano RP-HPLC metodo za kvantifikacijo loksoprofena v obliki farmacevtskega pripravka. Za ta namen smo uporabili eksperimentalni načrt. Iz preliminarne študije smo določili faktorje - neodvisne spremenljivke (organsko topilo, pH mobilne faze, pretok), kot odvisne spremenljivke pa smo izbrali tri tipe odziva (re-tencijski faktor loksoprofena, ločljivost med loksoprofenom in probenecidom ter retencijski čas probenecida). Za izboljšavo razvoja metode in njene optimizacije smo uporabili Derringerjevo funkcijo zaželenosti in tako hkrati optimizirali izbrane tri odzive. Postopek je dopuščal določitev optimalnih pogojev. Predvideni optimum je bil: razmerje acetoni-tril : voda (53:47, v/v), pH mobilne faze uravnan na 2, 9 z ortofosforno kislino. Ločba je potekla v manj kot 4 min. Metodo smo uporabili za kontrolo kakovosti komercialnih tablet. Pri metodi smo ugotovili skladnost med eksperimentalnimi podatki in predvidenimi vrednostmi v celotnem izbranem območju parametrov. Optimizirani postopek smo validira-li glede na smernice Mednarodne konference za harmonizacijo (ICH) in s tem potrdili specifičnost, linearnost, točnost in natančnost.