Meloxicam quantification in rabbit plasma by RP-HPLC: optimization and application to pharmacokinetic study

The goal of the proposed study was to validate a rapid, simple, an accurate, robust, and sensitive bioanalytical method for quantifying Meloxicam and Lornoxicam (as internal standard) in rabbit plasma. Limit of detection and limit of quantification for Meloxicam were found to be 0.0081 and 0.1035 μg mL−1, respectively. The bioanalysis was continued according to standard guidelines and successfully used for bioavailability studies of meloxicam after single dose administration of pure drug and the formulation in rabbit plasma. Finally, obtained results proved its simplicity and an efficiency to be applied for the therapeutic drug monitoring and bioequivalence studies. Therefore, the set RP-HPLC bioanalysis is simple, convenient, and acceptable to analyze meloxicam in bulk and pharmaceutical formulations in rabbit plasma.

Plentiful UV-VIS, HPLC, HPLC-MS/MS, LC-MS, LC-MS-MS, LC-ESI-MS/MS, LC-MS/TOF, and LC-MS analytical methods have been documented for quantifying the MLC from different formulation either alone or in combination with pharmacotherapeutic agents in biological fluid [1,[4][5][6][7][8][9][10][11][12][13][14][15][16][17][18]; a couple of them namely were employed liquid-liquid extraction or protein precipitation extraction (PPE) approach for quantification of MLC from biological sample [5,7,14]. In spite of that, the reported approaches are time-consuming; some of them used organic solvents and hazardous solvents for the extraction process, and some of which reported much less recovery which may be due to the drug loss during the transfer. Moreover, most of which followed solid-phase and liquid-liquid extraction techniques which required a more complex process [19]. In addition, the strong chemical bonds between plasma proteins and pharmacotherapeutic agents lead to a decrease in the efficiency of extraction, thus completely hampering the removal of pharmacotherapeutic agents, and consequently, a lower recovery [20]. Though solidphase extraction has been reported by Miyamoto et al. to quantify the MLC from biological sample, this method requires expensive equipment [1]. Couple of the bioanalytical methods used to quantify the MLC did not use the internal standard, which seemed to be a limitation of the reported method; some of which had been used but were complex in nature [8,11,12,15,17].
Therefore, it would be very beneficial to find a specific simple and efficient bioanalytical method for the rapid estimation of MLC in biological samples and also to reduce the above mentioned deficiencies. Furthermore, the binding ability of MLC to plasma proteins requires determining whether other related drugs mediate MLC binding, and vice versa [20]. Precipitation of proteins is the simplest way to extract most of the protein from the plasma and also involves minimal processing steps, which is why protein precipitation approach is used in pharmacokinetics analysis [21]. Biochemical analysis of MLC in plasma should be investigated in accordance with the FDA standards to ensure the safety and efficacy of marketed formulations [22].
The present research was carried out by considering the observations for estimating MLC in rabbit blood plasma using a simple extraction technique with specific and appropriate internal standards (Fig. 1b) to establish and validate a fast, simple bioanalytical reversed-phase HPLC-UV process along with randomized single-dose pilot study to test the bioavailability of MLC in a formulation.

Chemicals and reagents
HPLC grade methanol and acetonitrile (ACN) were purchased from MERCK India Pvt. Ltd. Meloxicam (MLC) and Lornoxicam (LRC) were obtained from Macleods Pharmaceuticals Ltd. (Mumbai-400093, Maharashtra, India), and all other chemicals were of analytical grade. Double-distilled water was used for preparing mobile phase solutions. The 0.45 μm nylon filters were purchased from Sigma-Aldrich, India.

Instrumentation
Systronics-HPLC (Isocratic mode) equipped with plus pump (Liquid Chromatography-100), UV/Vis detector (Liquid Chromatography-UV 100), and installed Shanghai Wufeng operating software. The separation was carried out on C8 Capcell pack DDS5 column along with 5 μm particle size and a length of 0.46 cm × 25 cm.

Chromatographic condition
Fifteen millimolar phosphate buffer having pH 4.7: methanol in the proportion of 40:60 v/v was prepared as a mobile phase, and then the mobile phase was degassed using sonicator for continuous 30 min. The final solution was filtered by using Millipore vacuum filter system with 0.45 μm membrane filter paper. The prepared mobile phase was run to maintain flow rate at 1.0 mL min −1 . UV 100 detector was set at 357.0 nm for the column effluence.

Preparation of stock solution
One hundred milligrams of MLC and LRC were transferred into a separate 100 mL volumetric flask, and then the 20 mL mobile phase was added and mixed with continuous shaking. The clear solution was ultra-sonic for a further 15 min and the resulting solution was then diluted into the target with the same solvent in order to obtain a final solution with a concentration of 1000 μg mL −1 . Furthermore, concentration ranges of MLC from 0.5 to 25 mg mL −1 were prepared using stock solution of MLC further diluted with mobile phase and stored under standard condition. Additionally, the working standard solution of LRC (internal standard) was prepared a final concentration of 40 μg mL −1 with an addition of LRC stock solution.

Extraction procedure
One hundred microliteres of the plasma and 50 μL of respective concentrations (0.50, 1.00, 2.00, 5.00, 10.00, 15.00, 20.00, and 25.00 μg mL −1 ) of MLC solutions were dropped into eight labeled microcentrifuge tubes (1500 μL capacity), and then all prepared samples were vortexed for a minute. Subsequently, 50 μL of LRC (40 μg mL −1 ) as an internal standard was added into resultant each concentration solution. Repeatedly, the all resultant solutions were vortexed for a minute with following medium speed. Further, 100 μL of protein precipitating agent namely ACN was dropped in each labeled tube and the all resultant solution were vortex for 2 min under medium speed. Finally, the all prepared samples were precipitated at room temperature, followed by centrifugation at 5000 rpm for continuous 15 min, and the supernatant was filtered through a 0.45 μm filter paper. At last, 20 μL of all samples were assessed under selected chromatographic condition.

Methods
In order to estimating MLC in rabbit plasma with precise, accurate, and reproducible bioanalytical reversephase HPLC-UV method, various solvent combination for mobile phases, stationary phases, and MLC preparation techniques have been assessed; however, the proposed mobile phase, stationary phase, MLC preparation techniques, λmax, etc. have been found to be appropriate for quantitative determination of MLC and internal standard. Finally, the proposed optimized and developed method for MLC was validated for selected parameters according to standard guidelines [22,23].

System suitability test
Prior to the analysis of each batch of samples, tests were carried out to confirm the reproducibility of the chromatographic system according to USP24/NF19 [24]. However, selected criteria namely retention time (RT), theoretical plate number (N), and tailing factor (T) are based on the actual performance of the method, as specified during its validation [25]. Selected parameters for the MLC peak were assessed to determine the percent coefficient of variation (% RSD) for the system suitability test.

Linearity
The linearity was confirmed using eight concentrations that range from 0.50 to 25.00 μg mL −1 of MLC solution.
The linearity experiment was tested to detect the response of the detector in linear with different concentrations of MLC solutions spiked in rabbit plasma, i.e., 0.50, 1.00, 2.00, 5.00, 10.00, 15.00, 20.00, and 25.00 Lg mL −1 with fixed concentration of LRC solution as an internal standard using the abovementioned extraction method. Finally, the prepared concentrations were injecting into the HPLC system and the calibration curves were constructed by plotting peak area of MLC versus concentrations of MLC, and the correlation coefficient, slop, and intercept were obtained from regression equations. Signal-to-noise ratios have been used to determine the limit of detection (LOD) and limit of quantification (LOQ) of MLC [26].

LLOQ (sensitivity)
The LLOQ is the lowest sample analysis concentration, can be accurately and precisely quantified, can be considered the lowest calibration curve point, and should be considered at least 5 times the blank sample [27][28][29].

Precision
The quality control (QC) sample having a concentration of 0.50, 5.00, and 25.00 μg mL −1 are represented in the calibration curves of the MLC between the low, medium, and high ranges in order to determine the accuracy and precision of the system being developed. Six injections of each concentration (0.50, 5.00, and 25.00 μg mL −1 ) were tested for the intraday study on the same day, while nine injection of each concentration (0.50, 5.00, and 25.00 μg mL −1 ) were tested for an interday variation over a period of 15 day and finally observed results were used to calculate % RSD in precision, whereas accuracy was expressed as a percentage of the theoretical concentration [30].

% Recovery (accuracy)
In order to confirm the accuracy of the proposed method, three different concentrations of linearity curve covering the lower, medium, and higher ranges with a concentration of 0.50, 5, and 25 μg mL −1 respectively were assessed. Absolute and relative recovery of the same nominal concentration for MLC was calculated by adding known quantities of MLC to the MLC free rabbit plasma. At the same nominal concentrations, absolute recoveries were determined with the peak area of the extracted QC samples compared to the peak area of the recovery standards [30][31][32][33].

Robustness
In proposed study, robustness is used to determine the effect of a small but deliberate variance in the flow rate of the pump at 0.9 and 1.1 mL min −1 ; pH of the phosphate buffer at 4.6 and 4.7: and percent of the phosphate buffer on 38 and 42 % in the mobile phase and were noted in % RSD.

Specificity
Checking any interference of an endogenous compound or corresponding placebo in the retention time of the MLC and LRC.

Stability
Stock solutions stability study was evaluated in three ways viz: (1) the freshly prepared plasma rabbit sample was accessed and then held overnight at a temperature of 15-20°C for 24 h and then re-injected into the HPLC device; (2) rabbit plasma samples were stored overnight in the freezer at − 20°C, collected at room temperature, and injected within 1 h after thawing thoroughly; and (3) standard stock solutions stored for 1 week at − 20°C in freezer, brought to room temperature, and injected within 1 h after thawing.

Application to bioavailability studies
In a single dose pilot study with randomized, single dose, crossover protocol, bioavailability and bioequivalence of bulk MLC (as a reference) and its tablet formulation ( [34]. The collected rabbit blood samples were then placed in Vacutainer Heparin Tubes and followed the centrifugation for 10 min at 4°C at a speed of 10,000 rpm, and then the supernatant was collected and stored in brown tubes at − 30°C until analysis time. The pharmacokinetic parameters namely maximum peak plasma concentration (C max ), T max , area under curve (AUC), mean residence time (MRT), half-life at the terminal phase (T 1/2 ), and elimination rate constant (Ke) were assessed under non-compartmental moment analysis model.

Statistical analysis
AUC 0-t , C max , and AUC 0-∞ were considered as crucial variables in bioequivalence study of MLC (in bulk and formulation) and were checked with the means of analysis of variance (ANOVA). 90% CI was calculated using Log transformed data [35,36].

System suitability test
The theoretical plate number (N) and tailing factors for MLC (15 μg mL −1 ) were found to be 2251.50 ± 1.9992 and 1.69 ± 1.7553 respectively. The chromatographic conditions described ensured adequate retention and asymmetry for drug compounds [26,37]. Therefore, the retention time of MLC was noted at 6.2398 ± 0.8608 min ( Table 1). The theoretical plate were observed in the range of 2210 to 2339, whereas tailing factor of the MLC were noted in the range of 1.65 to 1.74. However, it was found that MLC retention time was within 6.1769 to 6.2914 min.

Specificity
According to the chromatogram results, no interference from impurity, excipients, or additives were found and the all present additives in tablets were practically insoluble in mobile phase (Fig. 2).

Linearity
Linear calibration curves was observed after plotting peak area versus concentration of MLC over the concentration range of 0.5-25 μg mL −1 and the correlation coefficients were found to be 0.9940, whereas intercept and slope were found to be − 576.45 and 1324.25 respectively for MLC ( Table 2). Limit of detection and limit of quantification for MLC in rabbit plasma samples were found to be 0.0081 and 0.1035 μg mL -1 respectively and proved the sensitivity of the proposed bioanalytical methods.

LLOQ (sensitivity)
Diluting known concentrations of MLC in rabbit plasma was used to check the LLOQ; however, the proposed assay method offered an LLOQ of 0.50 μg mL -1 for MLC in rabbit plasma. % RSD for the intraday precision and accuracy was noted to be 1.85% (in RSD) and 98.56% respectively for MLC. Whereas, an interdays precision of the method was found to be 1.95% in RSD and the accuracy of the method was found to be 95.98% for MLC.

Accuracy (% recovery)
The absolute recovery of MLC was found to be 91.35 ± 1.55, 93.31 ± 1.66, and 94.71 ± 1.43 for the concentration of 0.50, 5.00, and 20.00 μg mL −1 respectively, while the LRC was noted to be 89.85 ± 1.18 (Table 3). Moreover, relative recovery of the MLC was determined by comparing the strength of the drug-spiked plasma with the actual added concentration. Relative percent recoveries for 0.50, 5.00, and 20.00 μg mL −1 MLC solution were found to be 101.00 ± 1.97, 100.36 ± 1.85, and 100.62 ± 0.88 respectively (Table 3).

Precision
The intraday precision of the proposed method was noted to be 1.97, 1.85, and 0.88 % RSD for MLC concentration of 0.50, 5.00, and 20.00 μg mL −1 respectively, whereas interday precision was found to 1.09, 1.63, and 1.60 μg mL −1 % RSD for 0.50, 5.00, and 20.00 μg mL −1 concentration of MLC, respectively. The above results indicated that the method is precise ( .00, and 20.00 μg mL −1 respectively. The accuracy results also indicate that the proposed method is accurate.

Robustness
The intentional changes in the method have not been much affected in the selected parameter namely tailing factor, theoretical plates, and the % assay. Tailing factor, theoretical plate, and percent assay of the MLC were found to be 1.    (Table 5).

Pharmacokinetic study
The developed protein precipitation extraction method was used to determine the bioavailability parameter after oral administration of the MLC bulk drug (reference) and MELFLAM tab (test) sample. Rabbit plasma (blank rabbit plasma), MELFLAM tab (after 1 h), and bulk drug suspension (after 1 h) were shown in Fig. 3a-c respectively.
Concentration time profiles of MLC in rabbit plasma following oral administration of the test and reference sample at interval 0, 0.5, 1, 2, 3, 4, 6, 8, 12, 24, 36, and 48 h were presented in Fig. 4. However, all pharmacokinetic parameters namely C max , AUC 0−t , AUC 0→∞ , MRT, T 1/2 , and Ke for MLC (test and formulation) were determined after administration of MELFLAM tab formulation and MLC bulk drug using a non-compartmental timing analysis model. C max , AUC 0-t , and AUC 0-∞ were found to be 0.6577 ± 0.01, 17.3070 ± 0.29, and 25.2047 ± 0.67 μg mL −1 respectively for reference sample, whereas for standard sample, C max , AUC 0-t , and AUC 0-∞ were noted to be 0.7792 ± 0.01, 19.7973 ± 1.06, and 25.1649 ± 0.89 μg mL −1 respectively. In addition, T max , T 1/2 , and MRT of the reference sample were observed to be 6.1 ± 0.36, 21.5261 ± 0.57, and 31.2453 ± 0.69 h respectively, while 4.0 ± 0.46, 20.3629 ± 0.20, and 29.3837 ± 0.55 h T max , T 1/2 , and MRT were found for the test sample. However, Ke for reference and test sample was observed to be 0.0320 ± 0.01 and 0.0340 ± 0.01, respectively (Table 6), and the two-tailed unpaired student t test was also used to measure the statistical significance for MLC between groups. Geometric mean ratio of T/R for 90% CI of C max , AUC 0-t , and AUC 0-∞ were noted to be 118.50 (115. 21   test formulation was quite similar to the reference formulation for pharmacokinetic parameters in MLC. Accordingly, the test and reference formulations were considered to be bioequivalent. All the findings are consistent with previous MLC interaction reports [1,4,5,7,10,12,14,18]. In the final analysis, the results obtained for applying the proposed bioanalytical RP-HPLC approach have shown its efficacy in applying therapeutic drug monitoring and bioequivalence studies.

Optimization of extraction method
The simple extraction method is important for bioanalysis if you want to progress in extraction efficiency. So, we studied various factors and got them together and the names are as follows, i.e., ion strength, dispersing solvents, pH, sample volume, and applied time. Solvents are very critical in the extraction process, so when selecting the solvent, it must be taken into account high densities, water immiscibility, and high solubility for analytes [38]; thus, the solvent is chosen according to the above. As a result, acetonitrile, diethyl ether, and methanol solvents have therefore been employed for detection of MLC and internal standard from rabbit plasma. In the ACN, the recovery of the MLC and internal standard was better than that of the solvents; in addition, the volume of ACN in the range of 0.05 to 0.5 mL should also be checked for better recovery. The best recoveries of the MLC and internal standard have been recorded with 0.1 mL of acetonitrile extraction solvent; henceforth, 0.1 mL ACN volume was continued for the experiment. As a result, an average recovery has been shown to be 98.83 ± 1.32 and 93.62 ± 1.47%, respectively, for MLC and LRC.

Optimization chromatographic conditions
To achieve the greatest selectivity in separating and evaluating target products, optimization of chromatographic  conditions is necessary [20]. A variety of chromatographic conditions namely flow rate, pH of the buffer, ratio of the mobile phase, and different solvents of the mobile phase have been tried to improve the resolution and shape of the peak for MLC and LRC. Nonetheless, various combinations of water, phosphate buffer, methanol, ACN, etc. were tried to set the mobile phase, among them the combination of methanol and phosphate buffer had a good peak shape and also an increased sensitivity over other, but as their volume ratio changed their runtime and resolution was changing. The ratio of methanol and phosphate buffer was therefore maintained at 60:40, so that the runtime and resolution of MLC and LRC would be better. Moreover, the extraction solvent and the LC mobile phase have difference in nature and strength. ACN is a very strong eluting solvent when injected into a LC system with the methanol and buffer as a mobile phase which have potentially impacted on the chromatogram and tailing factor of MLC.
In addition, variability of the phosphate buffer pH led to poor peak shape (especially at bottom). Finding the right pH for the phosphate buffer was difficult because pH of the buffer increased, broadening at MLC and LRC peak base were to be observed. The pH of the phosphate buffer has therefore been changed to precisely 4.7 by using o-phosphoric acid.
Moreover, flow rate parameter was assessed at interval 0.2 mL min −1 from 0.4 to 1.6 mL min −1 to get proper resolution and shape of MLC and LRC peak; consequently, MLC and LRC peak merging was observed at a higher range and also high expansion with high resolution of MLC and LRC peak were noted at a lower range; henceforth, 1.0 mL min −1 flow rate was part of optimized chromatographic parameter having better shape of peak. Eventually, the optimized mobile phase consisted of 15 mM phosphate buffer having pH 4.7: methanol (40:60 v/v), with 357.0 nm λ max of UV  detector. During all analytical processes, the retention period for MLC was continuously observed at about 6.2498 ± 1.4001 min (Fig. 5). The linear regression of the MLC showed a very significant non-zero intercept. The representative chromatogram rabbit plasma with LRC is depicted in Fig. 3d. Whereas, Fig. 5a-h illustrates the chromatogram of rabbit blood plasma spiked with MLC and internal standard in the strength of 0.50, 1.00, 2.00, 5.00, 10.00, 15.00, 20.00, and 25.00 μg mL -1 respectively.

Pharmacokinetic study
The proposed approach was effectively implemented in the pharmacokinetic and bioequivalence studies following a single oral dose of MLC bulk drugs and the formulation of MELFLAM tab (15 mg) tablets in rabbit. MELFLAM tab (15 mg) is the product of zydus cadila also prescribed regularly in the region of Karad taluka, Maharashtra-415110. We have served in our region, and most of the doctors prefer the MELFLAM tab to their patients, and therefore we have chosen the said brand for the pharmacokinetic study of the developed method. Different parameters were verified for the bioequivalance study of bulk drug as reference and MELFLAM tab formulation as test sample of MLC. However, oral suspension of the formulation of MELFLAM tab and bulk drugs were well tolerated by the respective rabbit group and no adverse reactions were observed. Subsequently, rabbit blood plasma of respective groups was tested in HPLC until 48 h for the pharmacokinetic approach. Experimental pharmacokinetic parameters, test/reference ratio (T/R), confidence intervals (90 CIs), and their significant statistical numbers for MLC following oral administration of 15 mg of MELFLAM tablets ranged from 80 to 125% in accordance with the FDA Bioequivalence Guideline for AUC o-t , AUC 0-∞ , and C max [39,40].

Figure of merits
Miyamoto et al. [1] have developed a solid-phase extraction approach for the estimation of MLC from biological sample, which have required extra assembly for the extraction process, while Bae et al. [5], Hye et al. [7], and Velpandian et al. [14] have developed a biological method for the estimation of MLC using a liquid-liquid extraction procedure due to the high plasma protein binding characters of MLC that make its complete extraction from plasma suspicious. Moreover Dasandi et al. [4], Pairis-Garcia et al. [10], Shukla et al. [12], and Yuan et al. [18] have used protein precipitation method which is a cost-effective method than abovementioned (Table 7). However, Pairis-Garcia et al. [10] and Shukla et al. [12] have not clearly mentioned the linearity range of meloxicam, while Yuan et al. [18] have used limited validation parameter (excluding LOD, LOQ, etc.). Moreover, Dasandi et al.'s method have followed all validation parameter but recovery of MLC is less than the proposed method (Table 7). Nonetheless, the results of recovery for MLC using 0.1 mL ACN proved that the protein precipitation approach is suitable for MLC quantification and also has high extraction efficiency. Method reproducibility at short times is of major concern in rabbit plasma determination. The process is non-tedious and substantially time consuming compared to the other method, which takes approximately 13 min to extract the MLC and internal standard with high reproducibility absolutely. Therefore, the expense of testing diminished only was 100 μL of rabbit plasma and ACN solvent. While the extraction processes have been performed in a simple step, manipulation in a short time is still easy. The validation and bioequivalence results proved the usefulness of the developed biological method in analysis of MLC with LRC.

Conclusion
In conclusion, an optimized and established bioanalytical method for quantification of Meloxicam is selective, precise, accurate, fast, and sensitive. Furthermore, Meloxicam analyte was extracted from biological sample of rabbit with simple extraction process and proved the bioequivalence of bulk drug sample and selected tablet formulation which could aid in therapeutic drug monitoring and bioavailability studies.