Microsoft word - 52-319.doc

WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis Determination of amoxicillin in human plasma by LC-MS/MS
and its application to a bioequivalence study
SATISH G. PINGALE1, MADHUKARA. BADGUJAR1, KIRAN V. MANGAONKAR1, NIKOS E. MASTORAKIS2 1 Analytical Chemistry Research Laboratory, Mithibai College Vile Parle (W), Mumbai 400056 2 Military Institutes of University Education, Hellenic Naval Academy E-mail address: [email protected] Abstract: - An analytical method based on solid phase extraction has been developed and validated for analysis of amoxicillin in human plasma using gemifloxacin as an internal standard. A COSMOSIL 5C18-PAQ column provided chromatographic separation of analytes followed by detection with mass spectrometry. The method involves simple isocratic chromatographic conditions and mass spectrometric detection in the positive ionization mode using an API-3000 system. The proposed method has been validated with linear range of 100–15000 ng/mL for amoxicillin. The intra-run and inter-run precision values are within 3.53% and 5.63% respectively for amoxicillin at the LOQ level. Total elution time was as low as 2 min. This validated method was used successfully for analysis of plasma samples from a bioequivalence study. Key-Words: - Amoxicillin, Antibiotic, Automated solid phase extraction, Tandem mass spectrometry, Human plasma, Bioequivalence study. 1 Introduction
reaches Cmax (8mg/mL) about 2 hours after administration, exhibits low binding with plasma proteins (17%), is quickly distributed through the Amoxicillin is a semi synthetic antibiotic. It is an body, and has an elimination half-life of 1 hour [3]. analog of ampicillin, with a broad spectrum of bactericidal activity against many gram-positive commonly used antibiotic. To understand the and gram-negative microorganisms. IUPAC name pharmacokinetic behavior of Amoxicillin in of the amoxicillin is (2S,5R,6R)-6-[(R)-(-)-2- humans, a reliable quantitative method is needed. Several bioanalytical methods are reported to determine amoxicillin in body fluid by HPLC with 2-carboxylic acid trihydrate. The presence of a UV [4-10] and fluorescence [11] detection. benzyl ring in the side chain extends the Sensitive and selective methods based on antibacterial activity to gram-negative bacteria [1]. The action mechanism of these antibiotics has not developed for determination of amoxicillin in been unequivocally established, but it is thought bovine milk [12], pig tissues [13] and human they may interfere with peptidoglycan bacterial cell wall synthesis in the effected organisms [2]. Different methods of sample preparation Amoxicillin shows high absorption after have been applied prior to the chromatographic oral administration, and this is not altered by the analysis, mostly based on protein precipitation concomitant ingestion with food. Amoxicillin [4,16], liquid-liquid extraction [8,13,14], solid- E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis techniques like ultra filtration [5,9]. orthophosphoric acid and formic acid were The method reported by Khuroo et al. procured from Merck (Mumbai, India). Drug free describes solid phase extraction of amoxicillin in (blank) human plasma containing K3-EDTA was human plasma using LC-MS/MS. The lowest obtained in-house by enrolling healthy volunteers LOQ's reported by Khuroo et al., was 170 ng/mL and taking their consent before bleeding. The and the linear dynamic range was 170-17000 plasma thus obtained was stored at –20 °C prior to ng/mL [15]. As per guidelines LOQ should be five- use. Oasis HLB 30 mg/1 cc solid phase extraction fold (32 times) below the Cmax value, but (SPE) cartridges were procured from waters unfortunately the method reported by Khuroo et al. (Milford, Mass, USA). also failed to quantify LOQ of amoxicillin in human plasma which were observed in the present bioequivalence study. 2.2 Calibration curve and quality control
The sensitive and selective LC–MS/MS method was reported by Wen et al. involved Two separate stock solutions of amoxicillin were solvent precipitation procedure with methanol. prepared for bulk spiking of calibration curve and Separation was achieved on a Lichrospher C18 quality control samples for the method validation column (150 mm x 4.6 mm ID, 5 µm) using exercise as well as the subject sample analysis. The methanol (containing 0.2% of formic acid) and stock solutions of amoxicillin and gemifloxacin water (containing 0.2% of formic acid) as a mobile were prepared in methanol at free base phase by gradient elution at a flow rate of 1.0 concentration of 1000 µg/mL. Primary dilutions mL/min. Linear dynamic range was 5-20000 and working standard solutions were prepared from ng/mL [16]. This method involved precipitation stock solutions by dilution with water:methanol method with methanol which caused deposition of (40:60, v/v). These working standard solutions matrix on curtain plate of the mass spectrometer were used to prepare the calibration curve and and needed cleaning of curtain plate frequently. quality control samples. Blank human plasma was Hence, it is necessary to develop and screened prior to spiking to ensure it was free of validate a clean, rapid, selective and sensitive endogenous interference at retention times of method which can be successfully applied to a amoxicillin and internal standard gemifloxacin. A bioequivalence study. nine point standard curve and four quality control In the present paper we would like to samples were prepared by spiking the blank plasma present a simple solid phase extraction method for with an appropriate amount of amoxicillin. Calibration samples were made at concentrations of gemifloxacin as an internal standard using LC- 100, 200, 750, 1500, 3000, 6000, 9000, 12000 and MS/MS. The application of this validated method 15000 ng/mL and quality control samples were in analyzing samples from a bioequivalence study made at concentrations of 300, 3800, 7600 and involving amoxicillin is also presented. The 12900 ng/mL for amoxicillin. greatest advantage of this method over all those referenced is the simple and clean SPE procedure with short runtime. Linear dynamic range for 2.3 Sample preparation
amoxicillin was sufficient for application in A 0.5 mL aliquot of human plasma sample was bioavailability-bioequivalence studies. mixed with 20 µL of internal standard working solution (35 µg/mL of gemifloxacin). To this 0.5 mL of 20% orthophosphoric acid was mixed. 2 Experimental
Sample mixture was loaded onto Oasis 30 mg/1 cc extraction cartridge pre-conditioned with 1 mL methanol followed by 1 mL water. The extraction 2.1 Chemicals and reagents
cartridge was washed with 1 mL of methyl tertiary The reference standards of amoxicillin and butyl ether. Amoxicillin and gemifloxacin were eluted with 1 mL of the mobile phase. 5 µL of the Laboratories (Mumbai, India). High purity water eluant was injected into the LC–MS/MS system was prepared in-house using a Milli-Q A10 through the autosampler. gradient water purification system (Millipore, Bangalore, India). LC grade methanol and methyl tertiary butyl ether were purchased from J.T. Baker E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis 2.4 Liquid chromatography and mass
for selectivity, sensitivity, linearity, precision and spectrometric conditions
accuracy, recovery, dilution integrity, partial 2.4.1 Chromatography
volume, matrix effect, reinjection reproducibility Chromatographic separation was carried out on a and stability. Selectivity was performed by Shimadzu HPLC system (Kyoto, Japan) using a analyzing the human blank plasma samples from COSMOSIL 5C18-PAQ C six different sources (or donors) with two more 18 column (50×4.6 mm i.d., 5 µm) purchased from Chromatopak (Thermo lots, each of haemolyzed and lipemic to test for interference at the retention time of amoxicillin and conditions. The mobile phase was 0.2% formic acid gemifloxacin. The intrarun (within a day) and in water–0.2% formic acid in methanol (20:80, v/v) interrun (on three different days) accuracy was and set at a flow rate of 0.6 mL min-1. The column determined by replicate analysis of quality control was maintained at room temperature. The HPLC samples (n = 6) at the LOQ (limit of eluent was introduced directly into the positive quantification), LQC (low quality control), MQC electrospray ionization source and the total run (medium quality control), M1QC (middle quality time for each sample analysis was 2 min. control), HQC (high quality control) and ULOQ (upper limit of quantification) levels. The %CV should be less than 15% and accuracy (%RE) 2.4.2 Mass spectrometry
should be within 15% except the LOQ where it The plasma concentration of amoxicillin was should be within 20%. quantified using a Sciex API 3000 triple Accuracy is defined as the percent relative error (%RE) and was calculated using the formula Biosystems, MDS Sciex, Concord, Canada) %RE = (E - T) (100/T) where E is the equipped with a Turbo Ion Spray interface to experimentally determined concentration and T is generate the positively charged ions. Ion spray the theoretical concentration. Assay precision was voltage was 5500 V with a turbo gas temperature at calculated by using the formula %CV = (SD/M) 450°C. The operating conditions were optimized by (100) where M is the mean of the experimentally flow injection of a mixture of all analytes and were determined concentrations and SD is the standard as follows: nebulizing gas flow 13 L min-1; auxiliary gas flow 8 L min-1; collision activated The extraction efficiencies of amoxicillin dissociation (CAD) gas was set at 7 psi and curtain and gemifloxacin were determined by analysis of gas flow 15 L min-1. Compound-dependent six replicates at low, medium, middle and high parameters for amoxicillin and gemifloxacin were: quality control concentrations for amoxicillin and orifice voltage 50 eV, ring voltage 100 eV. at one concentration for the internal standard Quantitation was performed in multiple reaction gemifloxacin. The percent recovery was evaluated monitoring (MRM) mode employing collision by comparing the peak area of extracted analytes to energies of 16 eV for amoxicillin and 30 eV for the peak area of non extracted standards (sample gemifloxacin with dwell times of 250 ms for ion spiked in mobile phase). transitions m/z 366.2→ m/z 348.8 (amoxicillin) and The dilution integrity experiment was performed with an aim to validate the dilution test respectively. Quadrupoles, Q1 and Q3 were set to to be carried out on higher analyte concentrations unit resolution. Automated data acquisition and above upper limit of quantification (ULOQ), which analysis were performed using the Analyst software may be encountered during real subject samples (version 1.4.2). analysis. Dilution integrity experiment was carried For quantification peak area ratios of the out at 1.7 times the ULOQ concentration. Six target ions those of the internal standard were replicates each of 1/2 and 1/4 concentrations were compared with weighted (1/x2) least squares prepared and their concentrations were calculated calibration curves in which the peak area ratios of by applying the dilution factor 2 and 4 against the the calibration standards were plotted versus their freshly prepared calibration curve. The partial volume experiment was performed to validate the method, for application in case of insufficient volume of plasma in real 2.5 Validation
subject sample. Partial volume experiment was A through and complete method validation of performed on middle quality control (M1QC) amoxicillin in human plasma was done following concentration level. Six replicates each of half and USFDA guidelines [17]. The method was validated quarter volume of total volume of plasma required E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis Application
concentrations were calculated by applying the bioequivalence study
dilution factor 2 and 4 against the freshly prepared The validated method has been successfully applied calibration curve. to the analysis of amoxicillin concentrations in To study the effect of matrix on analyte twenty four healthy, adult, human male volunteers quantification with respect to consistency in signal which were enrolled in a bioequivalence study. One suppression, matrix effect was checked with six amoxicillin tablet (875 mg) was orally administered different lots of plasma. Two replicates each of to volunteers rang ranging in age from 18 to 45 LQC and HQC were prepared from six different years as per the inclusion criteria for the study lots of plasma (total 24 QC samples) and checked under fasting conditions. Volunteers who have used for %CV which should be less than 10% at LQC any prescribed medication during last 2 weeks or over the counter medicinal products during the last reproducibility, week preceding the first dose were excluded from initially LQC and HQC samples were injected and the study. Seven days of wash-out period were kept after analysis of these samples system hardware between the two periods of the study. was deactivated for two hours, after three hours The bioequivalence study was based on a again system hardware was activated and balanced, open label, analyst blind, single centre, equilibrated. Then same samples were reinjected. randomized, crossover, two treatment, two periods, Original values were compared with reinjected two sequences and single dose design. A test values with respect to %Change, which should be formulation containing Amoxicillin 875 mg with the reference formulation Augmentin® was As a part of the method validation, stability evaluated. The study was conducted according to was evaluated in stock solutions and in plasma current GCP guidelines after signed consent of the under different conditions, simulating the same volunteers following approved by an authorized conditions, which occurred during study sample Ethics committee. In total, 20 time points after handling and analysis. Stock solution stability was blood collection per period were evaluated post- performed by comparing area response of stability dosing including the pre-dose sample. The blood sample of analyte and internal standard with the samples were collected in separate vacutainers area response of sample prepared from fresh stock containing K3-EDTA. The plasma from these solutions. Stability studies in plasma were samples was analyzed by LC-MS-MS. performed at LQC and HQC concentration level using six replicates at each level. Analyte was computed using the WinNonlin Software version considered stable if the %Change is less than 15% 5.2 (Pharsight Corporation, California, USA) and as per USFDA guidelines. Analyte was tested using 90% confidence interval was computed using SAS the quality control samples whenever appropriate. Software version 9.2 (SAS Institute, Cary, USA). The stability of spiked human plasma stored at room temperature (bench top stability) was evaluated for 12 h. The stability of spiked human 3 Results and discussion
plasma stored at -70 °C in coolant (coolant stability) was evaluated for 24 h. The autosampler sample stability was evaluated by comparing the 3.1 Method development
extracted plasma samples that were injected During method development different options were immediately (time 0), with the samples that were evaluated to optimize detection parameters, re-injected after keeping in the auto sampler at 10 chromatography and sample extraction. °C for 24 h. The freeze–thaw stability was conducted by comparing the stability samples that had been frozen at –20 °C and thawed three times, 3.1.1 Mass spectra
with freshly spiked quality control samples. Six The LC–MS/MS were tested to obtain sensitivity aliquots of each low and high concentration were and signal stability during infusion of the analyte in used for the freeze–thaw stability evaluation. For the continuous flow of mobile phase to electrospray long term stability evaluation the concentrations ion source operated both polarities at a flow rate of obtained after 15, 30, 45, 60 and 90 days intervals 10 µL/min. The amoxicillin gave better results in were compared with initial concentrations. positive ion mode. The predominant peaks in the E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis gemifloxacin corresponds to the MH+ ions at m/z amoxicillin and gemifloxacin scanned in 366.2 and 390.0 respectively. Product ions of
Fig.1 Parent ion mass spectrum of amoxicillin
Fig.2 Product ion mass spectrum of amoxicillin.
E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis Fig.3 Parent ion mass spectrum of gemifloxacin

Fig.4 Product ion mass spectrum of gemifloxacin
quadrupole 3 after a collision with nitrogen in The mobile phase containing 2mM ammonium quadrupole 2 had an m/z of 348.8 and 371.9 acetate:acetonitrile ammonium acetate:methanol (20:80v/v) exhibited better separation, but response was very low, which was insufficient to quantify LOQ. We had tried 3.1.2 Chromatography
mobile phase 0.2% formic acid in water solution in Initially, a mobile phase containing acetic acid combination with 0.2% formic acid in methanol solution and acetonitrile in varying combinations and 0.2% formic acid in acetonitrile in varying was tried in which poor peak shape was observed. combinations. Use of 2mM ammonium acetate E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis with 0.2% formic acid in water in combination with In the state of nonionic forms, the strong binding of methanol improves peak shape but response was analytes to the copolymer of the SPE cartridge low. Finally we received the best signal for enabled sufficient clean-up and suitable eluting amoxicillin and gemifloxacin using a mobile phase solution helped to elute analytes with more containing 0.2% formic acid in water solution in efficiency. So the method was optimized to achieve combination with 0.2% formic acid in methanol maximum extraction efficiency. The amoxicillin is (20:80 v/v) as there was almost a two fold increase approximately 17% protein-bound, so prior to in its area count as compared to the mobile phase extraction it is necessary to release the amoxicillin containing 0.2% formic acid in water solution in from protein binding. So orthophosphoric acid was combination with acetonitrile. Moreover, a marked added in the sample before extraction. for breaking improvement in the peak shape of amoxicillin and protein binding. Different types of cartridges such gemifloxacin was also observed using this mobile as Phenomenex (Strata-X, Strata-X-C, Strata- phase combination. XCW, Strata-XAW) and Oasis HLB were tried. Short length columns such as, COSMOSIL Finally we choose an Oasis HLB 30 mg/1 cc 5C18-PAQ (50 mm x 4.6 mm, 5 µm), Symmetry cartridge. It was difficult to find a compound which Shield RP18 (50 mm x 2.1 mm, 3.5 µm), Inertsil could ideally mirror the analytes to serve as a good C18 (50 mm x 4.6 mm, 5 µm), HyPURITY C18 IS. Several compounds were investigated to find a (50 mm x 4.6 mm, 5 µm) and HyPURITY Advance suitable IS, and finally gemifloxacin of the same (50 mm x 4.0 mm, 5 µm) were tried during the class of compound was found most appropriate for method development. In Symmetry Shield RP18 the present purpose. There was no significant effect columns poor chromatography was observed. of IS on analyte recovery, sensitivity or ion Inertsil C18, HyPURITY Advance and HyPURITY suppression. The results of method validation using C18 columns gave a relatively good peak shape but gemifloxacin as the IS were acceptable in this study the response was low. The best signal was obtained based on FDA guidelines. The high recovery and using the COSMOSIL 5C18-PAQ (50 mm x 4.6 selectivity was observed in the solid phase mm, 5 µm) column. It gave satisfactory peak extraction method which was used in the present shapes for all the analytes and a flow rate of 0.6 mL/min reduced the run time to 2.0 min. These optimized detection parameters, Introducing such a high flow directly in to the ionization source affects evaporation of solvents procedure resulted in reduced analysis time with which further causes improper ionization and accurate and precise detection of amoxicillin in reduces response, hence splitter was utilized to control direct flow in the ionization source. 3.1.3 Extraction
3.2 Method validation
Prior to LC injection, the co-extracted proteins should be removed from the prepared solution. Several organic solvents were employed to extract 3.2.1 Selectivity and sensitivity
analytes from the plasma sample. In the tested Representative chromatograms obtained from blank solvents (ethyl acetate, chloroform, hexane, plasma, plasma spiked with internal standard and dichloromethane, methyl tertiary butyl ether and plasma spiked with LOQ standard for amoxicillin acetonitrile) samples were not clean and poor and gemifloxacin is presented in Fig. 5. chromatography observed. E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis
Fig.5 Representative chromatograms of amoxicillin (left) and gemifloxacin (right) in human plasma. (A)
Blank plasma (B) Plasma spiked with internal standard and (C) LLOQ

The mean %interference observed at the retention
the present method. All the values obtained below time of analytes between eight different lots of 100 ng/mL for amoxicillin were excluded from human plasma including haemolyzed and lipemic statistical analysis as they were below the LOQ plasma containing K3-EDTA as the anti-coagulant values validated for amoxicillin. calculated was 0.00% for amoxicillin and gemifloxacin, which was within acceptance criteria. Six replicates of extracted samples at the Linearity,
precision
accuracy,
LOQ level in one of the plasma samples having recovery
least interference at the retention time of The peak area ratios of calibration standards were amoxicillin was prepared and analyzed. The %CV proportional to the concentration of amoxicillin in of the area ratios of these six replicates of samples each assay over the nominal concentration range of was 6.68% for amoxicillin confirming that 100–15000 ng/mL. The calibration curves appeared interference does not affect the quantification at linear and were well described by least-squares LOQ level. Utilization of selected product ions for linear regression lines. As compared to the 1/x each compound enhanced mass spectrometric weighing factor, a weighing factor of 1/x2 properly selectivity. The product ions of m/z 366.2 and achieved the homogeneity of variance and was 390.0 were hence concluded to be specific for chosen to achieve homogeneity of variance. The amoxicillin and gemifloxacin. correlation coefficient was ≥0.9900 for amoxicillin. The LOQ for amoxicillin was 100 ng/mL. The observed mean back-calculated concentration The intra-run precision and intra-run accuracy with accuracy (%RE) and precision (%CV) of four (%RE) of the LOQ plasma samples containing linearities analyzed during method validation are amoxicillin was 3.53 and -1.81, respectively. We given in table 1. The deviation of the back obtained the mean Cmax of amoxicillin, for test and calculated values from the nominal standard reference formulations were 8521.98 and 8696.75 concentrations were less than 15% except LOQ ng/mL, respectively. As per guidelines LOQ should where it should be less than 20%. This validated be below 32 times (five-fold) of Cmax. So for linearity range justify the concentration observed amoxicillin, the LOQ was easily, quantified using during real sample analysis. E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis Table 1 Summary of calibration curve standards

Nominal concentration
Mean back calculated concentration (ng/mL) CV: coefficient of variation; RE: relative error. individual assay results of replicate (n = 6) quality The inter-run precision and accuracy were control of two separate batch runs analyzed on the determined by pooling all individual assay results same day. The intra-run precision (%CV) and intra- of replicate (n = 6) quality control over three run accuracy (%RE) was ≤3.53% and ≤-3.99% separate batch runs analyzed on three different respectively for amoxicillin. All these data days. The inter-run precision and inter-run presented in Table 2 indicate that the method is accuracy (%RE) was ≤5.63% and ≤-8.43% precise and accurate. respectively for amoxicillin. The intra-run precision and accuracy were determined by pooling all
Table 2 Intra-run and inter-run precision and accuracy (n = 6) of amoxicillin in human plasma

Run
Concentration Mean concentration added (ng/mL) found (ng/mL) Intra 3754.37 3643.88 1.72 -2.94 12475.94 1.63 -2.26 102.12 96.63 5.63 -5.38 _ CV: coefficient of variation; RE: relative error. E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis Table 3 Absolute recovery for amoxicillin and gemifloxacin

Analytes Level
HQC 895241 333162 Gemifloxacin M1QC 749219 A: Area of unextracted sample; B: Area of extracted sample; Mean recovery was found to be 37.25% for amoxicillin and 48.83 for gemifloxacin; CV: coefficient of variation. and HQC were 4.45% and 2.47% respectively for Six aqueous replicates (samples spiked in mobile phase) at low, medium, middle and high amoxicillin. Hence minor suppression of analyte quality control concentration levels for amoxicillin signal due to endogenous matrix interferences did were prepared for recovery determination and the not affect the quantification of amoxicillin. areas obtained were compared versus the areas Reinjection reproducibility exercise was obtained for extracted samples ( shown in Table 3) performed to check weather the instrument of the same concentration levels from a precision performance remains unchanged after hardware and accuracy batch run on the same day. The mean deactivation due to any instrument failure during recovery for amoxicillin was 37.25% with a real subject sample analysis. %Change is less than precision of 4.85%. The mean recovery for 4.13% for LQC and HQC level concentration; gemifloxacin was 48.83% with a precision of hence batch can be reinjected in case of instrument 4.81%. This indicates that the extraction efficiency failure during real subject sample analysis. for the amoxicillin as well as gemifloxacin was Stock solution stability was performed to consistence and reproducible. check stability of amoxicillin and gemifloxacin in stock solutions prepared in water : methanol (80:20) and stored at 2-8 °C in refrigerator. The 3.2.3 Dilution integrity and partial volume
freshly prepared stock solutions were compared The mean back calculated concentrations for 1/2 with stock solutions prepared before 30 days. The and 1/4 dilution samples were within 85-115% of %Change for amoxicillin and gemifloxacin were their nominal. The %CV for 1/2 and 1/4 dilution 2.22% and 1.99% respectively indicates that stock samples were 1.52% and 1.53% respectively. The solutions were stable at least for 30 days. mean back calculated concentrations for half and Bench top, coolant and autosampler quarter partial volume samples were within 85- stability for amoxicillin was investigated at LQC 115% of their nominal. The %CV for half and and HQC levels. The results revealed that quarter partial volume samples were 2.16% and amoxicillin was stable in plasma for at least 12 h at 1.31% respectively. room temperature, 24 h in coolant at -70 °C and 24 h in an autosampler at 10 °C. It was confirmed that repeated freezing and thawing (three cycles) of 3.2.4 Matrix effect, reinjection reproducibility
plasma samples spiked with amoxicillin at LQC and stabilities.
and HQC levels did not affect their stability. The The assessment of matrix effect constitutes an long term stability results also indicated that important and integral part of validation for amoxicillin was stable in matrix for up to 90 days at a storage temperature of –20 °C. The results pharmacokinetics studies. The results found were obtained from all these stability studies are well within the acceptable limits as the %CV of the tabulated in Tables 4. area ratios of post spiked recovery samples at LQC E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis Table 4 Stability results for montelukast

Stability
LQC 294.70 4.86 274.82 3.32 -6.74 HQC 12084.80 1.95 11791.98 1.72 -2.42 LQC 294.70 4.86 286.95 6.02 -2.63 (12 h at Room temp.) HQC 12084.80 1.95 11781.571 1.99 -2.51 LQC 294.70 4.86 284.47 4.70 -3.47 HQC 12084.80 1.95 648.66 1.57 -4.60 LQC 313.07 4.85 321.67 4.85 2.75 (2 h, ambient temp.) HQC 12526.17 0.92 13042.88 1.79 4.13 3rd freeze-thaw cycle LQC 306.96 4.54 302.62 6.46 -1.41 HQC 12043.18 2.92 12009.97 2.85 -0.28 LQC 297.07 3.50 341.46 0.64 14.94 (90 days, -20 °C) HQC 12390.67 1.32 12702.09 4.27 2.51 A: comparison sample concentration (ng/mL); B: stability sample concentration (ng/mL); CV: coefficient of variation; h: hours, temp.: temperature. extrapolated to infinity, calculated using the formula AUC0-t + Ct/Kel, where Ct was the last 3.2.5 Application of the Analytical Method to
measurable drug concentration), Tmax (time to Pharmacokinetic Studies
observe maximum drug concentration), Kel Statistical data analysis included determination of Cmax (maximum observed drug concentration calculated from a semi-log plot of the plasma during the study), AUC0–t (area under the plasma concentration versus time curve, using the method concentration- time curve measured to the last of least square regression) and T1/2 (terminal half- quantifiable concentration, using the trapezoidal life as determined by quotient 0.693/Kel). rule), AUC0–inf (AUC0–t plus additional area
Table 5 Pharmacokinetic parameters of amoxicillin using non-compartmental analyses

Parameters
8521.98 ± 3371.87 8696.75 ± 3053.90 AUC0–t (ng x h/mL) 37330.73 ± 12634.43 37915.47 ± 12659.06 AUC0–inf (ng x h/mL) 37926.25 ± 12619.85 38390.10 ± 12654.11 E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis Fig.6 Mean concentration versus time profile of amoxicillin in human plasma from 24 subjects receiving
a single oral dose of 875 mg amoxicillin tablet as test and reference
These variables calculated post-study are tabulated References:
in Table 5. The mean Cmax data obtained justified [1] Tomas A., From penicillin-binding proteins to the selected LOQ levels as they were at least less the lysis and death of bacteria, Rev Infect Dis, 1: than five half-lives of the obtained Cmax values. The mean concentration versus time profile [2] Neu, HC., Antimicrobial activity and human of amoxicillin in human plasma is shown in Fig.6. pharmacology of amoxicillin, J Infect Dis, 129: The 90% confidence intervals based on the ratios of means Cmax, AUC0–t and AUC0–inf fell [3] Simon C. Stille W., Antibiotikatherapie in Klinik und Praxis, Schattauer, NY, USA, 1985. demonstrating the bioequivalence of the two [4] Numan AA. Noman M. Alkadi H. Tayeb AI. formulations of amoxicillin. Alsolwi A. Zawia N. Alhakami A., Validation and application of a reversed-phase HPLC method for the determination of amoxicillin trihydrate in 4 Conclusions
human plasma, J Applied Sciences Research, 12, The developed LC-MS/MS assay for amoxicillin is measurement of subject samples. In the present chromatographic determination of amoxicillin in study solid phase extraction method was reported human plasma and its clinical application, J for extraction of amoxicillin in human plasma Pharma Sciences, 25: 112, 2009. using LC-MS/MS. The solid phase extraction [6] Kuo TF. Wang YK. Sheu SY. Chang MH. method provided excellent sample clean up and Wang JH., Detection of residues by use of solid- high extraction efficiency has been successfully phase extraction and reversed phase high- applied to the bioequivalence study of a tablet performance liquid chromatography after oral containing 875 mg amoxicillin as an oral dose in 24 administration of amoxicillin in bass muscle, healthy human volunteers under fasting conditions. Taiwan Shouyixue Zazhi, 35:66, 2009. [7] Samanidou VF. Giannakis DE. Papadaki, A., Development and validation of an HPLC method for the determination of seven penicillin antibiotics E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012 WSEAS TRANSACTIONS on BIOLOGY Satish G. Pingale, Madhukar A. Badgujar, Kiran V. Mangaonkar, Nikos E. Mastorakis in veterinary drugs and bovine blood plasma, J Biomed Analysis, 48: 829, 2008. Separation Science, 32:1302, 2009. [17] Guidance for industry: bioanalytical method [8] Alegre MR. Broch SC. Romero JE., validation, U.S. Department of health and human Development and validation of a method to services, food and drug administration centre for determine amoxicillin in physiological fluids using drug evaluation and research (CDER), centre for micellar liquid chromatography, J Separation veterinary medicine (CVM), May 2001. Science, 31: 2813, 2008. [9] Simultaneous determination of amoxicillin and clavulanic acid in human plasma by isocratic reversed-phase Foroutan, Seyed Mohsen; Zarghi, Afshin; Shafaati, Alireza; Khoddam, Arash; Movahed, Hooman, Journal of Pharmaceutical and Biomedical Analysis, Vol.45, No.3, 2007, pp. 531-534. [10] Luis RA. Rodrigo AO, HPLC determination of amoxicillin comparative bioavailability in healthy volunteers after a single dose administration, J Pharm Pharmaceut Sci, 6: 223, 2003. [11] Fernandez TR. Consentino MO. Lopez MB. Mochon MC., Simultaneous determination of 11 antibiotics and their main metabolites from four different performance liquid chromatography-diode array-fluorescence (HPLC-DAD-FLD) in human urine samples, Talanta, 81: 871, 2010. [12] Kantiani L. Farre M. Sibum M. Postigo C. Lopez M. Barcelo D., Fully automated analysis of β -lactams in bovine milk by online solid phase extraction-liquid tandem mass spectrometry, Analytical Chemistry (Washington, DC, United States), 81: 4285, 2009. [13] Reyns T. Cherlet M. De S. De P. Croubels S., Rapid method for the quantification of amoxicillin and its major metabolites in pig tissues by liquid chromatography-tandem mass spectrometry with emphasis on stability issues, J Chromatography B, 861: 108, 2008. [14] Thenmozhi A. Sridharan D. Rajamanickam V. Palanivelu M., Thenmozhi A. Sridharan D. Rajamanickam V. Palanivelu M. Ajitha A., Rapid and sensitive LC-MS/MS method for the simultaneous estimation of amoxicillin and clavulanic acid in human plasma, Asian J Pharm and Clinical Research, 3: 106, 2010. [15] Khuroo AH. Monif T. Verma PP. Gurule S., Simple, economical, and reproducible LC-MS method for the determination of amoxicillin in human pharmacokinetic study, J Chromatogr Science, 46: 854, 2008. [16] Wen A. Hang T. Chen S. Wang Z. Ding L. Tian Y. Zhang M. Xu X., Simultaneous determination of amoxicillin and ambroxol in human plasma by LC-MS/MS: validation and application to pharmacokinetic study, J Pharma E-ISSN: 2224-2902 Issue 1, Volume 9, January 2012

Source: http://www.wseas.org/multimedia/journals/biology/2012/52-319.pdf