JOURNAL OF
PHARMACEUTICALAND BIOMEDiCAL
ANALYSIS
ELSEVIERJournal of Pharmaceutical and Biomedical Analysis
28 (2002) 999-1004www.elsevier.com/locate/jpba
~Short Communication
1
!!
LC determination and pharmacokinetics of meloxicam
Bhavesh Dasandi a, Shivaprakash a,*, H. Saroj a, K.M. Bhat b
a Synchron Research Services Pvt. Ltd., Sarkhej Gandhinagar Highway, Bodakdev, Ahmedabad 380 054, Indiab K.B. Institute of Pharmaceutical Education and Research, GH-6, Sector-23, Gandhinagar 382 023, India
Received 30 May 2001; received in revised form 12 November 2001; accepted 25 November 2001
Abstract
A simple and rapid HPLC assay method for the estimation of meloxicam in plasma was developed. The methodtotally eliminated the solvent extraction procedure. The plasma proteins were precipitated using perchloric acid (70%)and acetonitrile mixture n:l v/v) and the supernatant was directly injected to the HPLC system. The separation wasachieved on a Lichrospher CI8 511(125 x 4.0 mm) analytical column with a mobile phase of sodium acetate buffer (pH3.3, 170 mmol):acetonitrile (62:38 v/v) mixture. Detection was by UV detector at 355 nm. The retention time observedfor meloxicam and piroxicam (internal standard) were at 6.0 and 4.0 min, respectively. The response was linear overa range of 50-1500 ng ml- I in human plasma. The method was simple, specific, precise and accurate. The methodwas also used for the bioequivalence study of meloxicam formulation in healthy, human, Indian, male volunteers.
@ 2002 Elsevier Science B.V. All rights reserved. lKeywords: Meloxicam; Piroxicam; One step sample preparation; ODS column; UV detection; Pharmacokinetics; Indian malevolunteers .
} 1. Introduction
~i
Meloxicam is a new NSAID of the enol carba-mide class. Animal studies have shown that inaddition to high anti-inflammatory efficacymeloxicam appears to have low ulcerogenic po-tency [1] and exhibits less gastric irritation andlocal tissue irritation in comparison to otherNSAIDs [2]. This good tolerability profile may beexplained by the ability of meloxicam to preferen-tially inhibit the inducible cyclo oxygenase present
~I
* Corresponding author. Tel.: +91-79-6853-419; fax: +91-79-6853-415.
E-mail address:[email protected] (Shivaprakash).
in inflamed tissue (COX-2) over COX-I that has ahouse keeping function in prostaglandin formula-tion [3].
Previous single dose pharmacokinetic studies [4]in healthy fasting volunteers have shown thatmeloxicam has prolonged absorption after oraladministration. This avoids high initial drug con-centration and is suitable for once daily dosage[5]. The plasma protein binding of meloxicam ismore than 99.5% [6]. Few LC assay methods formeloxicam quantitation in biological fluids havebeen reported. The method reported by Busch etal. involves long extraction procedures and re-quires sophisticated equipments like columnswitching [7].The method reported by Velpandian
0731-7085/02/$ - see front matter ~ 2002 Elsevier Science B.y. All rights reserved.
'PII: S0731-7085(02)00064-X
.
1000 B. Dasandi et al. / J. Pharm. Biomed. Anal. 28 (2002) 999-1004
et al. [8] is simple, but involves the conventionalextraction procedure. The method reported herebypasses the whole process of extraction, separa-tion and evaporation. The plasma is treated withperchloric acid and acetonitrile mixture and thesupernatant was directly injected to HPLC. An-other method reported by Schmed et al. [9] alsoinvolves a complicated process of precolumn en-richment of the sample with ammonium formateand gradient elution procedure.
This paper describes a sensitive, specific andsimple method involving one-step rapid samplepreparation technique. Human plasma samplescontaining meloxicam and internal standard were.precipitated using perchloric acid (70%):acetonitrile (1:1 vjv) with more than 85% recov-ery. Determination of meloxicam was performedon reversed phase CI8 column.
The developed method was applied to bioequiv-alence study of two oral dosage forms of me1oxi-cam (test and reference). The open randomized,cross over study performed on a group of 12healthy, Indian male \!olunteers confirmed thebioequivalence of both the formulations.
2. Experimental
2.1. MaterialsI
.'
Meloxicam and piroxicam were obtained fromCadila Pharma (Ahmedabad, India) as gift sam-ple. Sodium acetate, glacial acetic acid, perchloricacid (70% vjv) of analytical grade and acetonitrileand water of HPLC grade were obtained fromMerck, India (Mumbai).
2.2. Apparatus and conditions
Analysis was performed using HPLC systemconsisting of a pump (L-71l0, Merck Hitachi),UV-visible Detector (L-7400, Merck Hitachi)and auto sampler (L-7200, Merck Hitachi). Thesystem was connected with help of D-7000 inter-face to HSM software in a computer system fordata collection and processing. .
The analytical column used was LichrospherCI8 (5Jl, 4.0 x 125 mm, Merck, Germany). The
protein precipitating solution was a mixture ofacetonitrile and perchloric acid (70%) in the ratioof 1:1 vjv. The mobile phase contained 62% ofbuffer and 38% of acetonitrile and delivered at arate of 1 ml min - I. The buffer was 170 mmol ofsodium acetate in water with pH adjusted to 3.3with glacial acetic acid. The eluent was monitoredat 355 nm. Under these conditions the retentiontimes observed for me10xicam and piroxicam were6.0 and 4.0 min, respectively.
!/
/J . .
2.3. Pharmacokinetics
Two formulations of meloxicam (test and refer- :)~ence) were administered to 12 healthy, Indianmale volunteers in a double blind, randomized,cross over design. The washout period was 7 days.The volunteers were selected on a pre set inclu-sion-exclusion criteria. The volunteers werescreened for vital signs, blood and urine analysisbefore enrolment. Oral dose of 30 mg me10xicamwas administered with 240 m1 of water. Blood-samples were withdrawn at 0 h and 1,2,3,4,6,8,12, 24, 48, 72 h post dose. The samples werestored at - 20°C pending analysis and analyzedby the above method. A concentration time curvewas plotted and AUC calculated by trapezoidalrule (AUCo-n)' AUCo-oc was also calculated.Time to achieve the maximum concentration(CmaJ tmaxwas obtained directly from the concen-tration time curve without interpolation. All the
pharmacokinetic data are calculated using)'QUICKCALC',in house software. ~
2.4. Preparation of standard curve
One hundred microlitres of meloxicam solutionprepared using drug free plasma, of appropriateconcentration and 100 Jll of piroxicam of (5 Jlgml-I) were added to 900 JlI of drug free plasmacontained in a clean borosilicate glass tube andvortexed for 10 s. To this 100 Jll of proteinprecipitating reagent was added and vortexed for1 min. After centrifugation at 3000 rpm for 20min, 100 Jll of the supernatant was injected to theHPLC system.
3. Results and discussions
3.1. Specificity
B. Dasandi et al. / J. Phdrm. Biomed. Anal. 28 (2002) 999-1004 1001
human blank plasma, LQC and HQC were pre-sented in Figs. 1-3, respectively. The chro-matograms shown are the true scale, unmodifiedones. Though the peaks look slightly broad, theyare quiet reproducible and did not effect theRepresentative' chromatograms of processed
5~.;
4~..
:j
! 3-':.~...c.4JC...
2-:..
..:!Jj
1-:1
~0 ~
~
~-1 t
i0
If.\1\ I
I' I
~I~
,... .-;. .A~
. .~.J,
t~"ji...,.~~
., I" ,. I' " 'I' " I"" I' '.' . I'" 'I"" I'" """ I' "'I"" I""
1 -2 3 4 5 6
JleteDtica 'lime (8in)
Fig. 1. Representative chromatogram of processed blank human plasma.
II
~
n1
\"I .".
f .I -I ~.i ,\.I \ \
\I
j'
...,......
<G...
...
co...on
',"",' '1"
1 2. , '1'" 'I'" 'I' "'I'" 'I"" I'" "'" 'I""
3 4 5 6
Jlet.eatica 'lime (Jain)
Fig. 2. Representative chromatogram of LQC.
5 .4
.! 3 -
2...c.4J .;c 1...
0 .
-1 JI I
0
1002 B. Dasandi et al. / J. Pharm. Biomed. Anal. 28 (2002) 999-1004
:j
1ioj~
~::J
-1 .-1,.. ... , . 'I' , ., I'" 'I' " 'I' ,., I' " , I'" 'I' '" I'" """ I' "'I'" 'I'" ,
0 1 2 3 4 S 6
ReteDtico ti8a (IIin)
s""'3
~ ~
ft
.,
~\ ..,.-~-'
-1'\!
4...3:.3 .3
3"':;
~
"'3
21>-~.....c.~c~
CDM0/1
....r-M
..
...M
()..
Fig. 3. Representative chromatogram of HQC.
results in any manner. No interfering peaks wereobserved in blank at retention time correspondingto the drug and internal standard. This-shows thatthe assay procedure is specific to meloxicam. Allhuman plasma samples used for calibration stan-dards and QC were free from interfering peaks.
3.2. Linearity of calibration curves "
A standard curve of seven points was plotted(n = 5). A straight-line equation with weightagefactor of IjX2 was derived for the observed re-sponse. Calibration curve data and calibrationcurve parameters for meloxicam and piroxicam(internal standard) in human plasma demonstratethat calibration curves were linear in the concen-tration range from 50 to 1500 ng ml-'. Thecorrelation coefficient was found to be 0.9991 :t0.0005 (Table 1).
I'I!
3.3. Precision and accuracy
The limit of detection (LOD) for this assay waslOng ml - '. The limit of quantitation (LOQ) ofthe method is 50 ng ml- i. The interday andintraday accuracy and precision was assessed byreplicate analysis of three QC samples. The mean
RSD values for intraday and interday assay re-o producibility (n = 5) were 7.23 and 6.41%, respec-
tive'ly (Tables 2 and 3, respectively). Datapresented in the above tables are the coefficient ofvariation (%CV) for each sample processed.
3.4. Recovery
Absolute recoveries were determined by com-paring the ratio of peak height of meloxicam to
J
~ 2000(;)z
6 1600HI-a:
g: 1200ZWu
6 800U:La: 400uHXg 0w 0:L
,~je TEST
.. REFERENCE
10 20 30 40 50 60
TIME IN HOURS
70 80
Fig. 4. Concentration vs. time graph of meloxicam for test andreference formulation.
B. Dasandi et al. / J. Pharm. Biamed. Anal. 28 (2002) 999-1004
internal standard for standard preparationsagainst those of the same preparations in pro-cessed samples. Recoveries were performed atthree QCs were greater than 85%. Evaluation ofshort-term storage of extracted plasma samples on
Table 2Interday variabilityof the assay of quality controLsamples
1003
the standard curve characteristics and chromato-
graphic behavior of meloxicam and piroxicam(internal standard) were also performed. Stabilitydata of the extracted sample inside the autosam-pIer at 20°C is given in Table 4. These sampleswere stable even after 8 h inside the autosampler.Regression analysis of the standard curve datagave correlation coefficients and values for theslope and y-intercept within the same order ofmagnitude following storage of samples in theauto sampler. Freeze thaw stability results aregiven in Table 5.
3.5. Pharmacokinetics
Overlay graph of mean concentration vs. timeof the two formulations (Test and Reference) isshown in Fig. 4. The area under the curve from 0to 72 h was determined with the help of lineartrapezoidal rule. The extrapolation to infinity that
Bias ('Yo)Concentration added (ng ml-1) Concentration analyzed (ng ml-l) (mean:t S.D.) CV ('Yo)
50600
1000
50.63 :t 3.83
563.37 :t 33.84
863.08 :t 48.88
7.576.015.66
1.19-6.66
-14.21
Table 3
Intraday variability of the assay of quality control samples
Bias ('Yo);Concentration added (ng mI-l) Concentration analyzed (ng ml-l) (mean:t S.D.) CV ('Yo)
50600
1000
46.23 :t 4.11
573.06 :t 57.36
897.01 :t 25.26
8.8810.012.82
-7.59-5.06
-10.83
Table 4
Auto injector stability
Concentration (ng ml-1)
50.0600.0
1000.0
Table 1
Calibration curve data of me10xicam in human plasma
Day Slope Intercept r2
1 0.001247 -0.009165 0.99952 0.002564 -0.03727 0.99823 0.002341 -0.009043 0.99954 0.002051 -0.009120 0.99915 0.001996 -0.009060 0.9994Mean :t S.D. 0.002039 .NA 0.9991
:t 0.0005 :t 0.00055R.S.D. ('Yo) 24.46 NA 0.055
Concentration determined after Mean :t S.D. CV ('Yo)
2h 4h 8h
. 52.921 48.521 49.586 586 :t 2.95 5.95575.513 545.090 568.724 563.109:t 15.97 2.84876.985 910.411 845.920 877. 772 :t 32.25 3.67
1004 B. Dasandi et al. / J. Pharm. Biomed. Anal. 28 (2002) 999-1004
Table 5Freeze thaw cycle
Concentration (ng ml-l) Recovery (%) CV (%)
50.0600.0
1000.0
80.0282.2480.78
2.412.524.45
~
Table 6Pharmacokinetic data of meloxicam
Parameter Test
AUCo-n (ng h-1 ml-1)AUCo-a: (ng h-' ml-I)Cmax(ng ml-l)
35027.5846051.08
1126.75
is necessary for AUCo-oc evaluation was calcu-lated using a linear regression model from the lastthree data points in the elimination phase that hasbeen log transformed, Maximum concentrationachieved (CmaJ was obtained directly from themeasured concentration without interPolation.
Assuming the multiplicative models expectedmedians of these parameters of the test and refer-ence formulations were computed and presentedin Table 6 as their ratios. The confidence inter-vals, suitable for bioequivalence testing .'~erefound well within the bioequivalence range of0.8-1.25 adopted in a recent cGMP guidelines onbioequivalence studies [10].
4. Conclusions
. The HPLC assay described here is simple, selec-tive, precise and accurate for quantitation ofmeloxicam in human plasma. The sensitivity, sim-plicity and the rapidity of the method were themain advantages that can be applied to routinetherapeutic monitoring of the drug and have
proved useful in evaluating the pharmacokineticin human volunteers.
. References
[I] G. Engelhardt,D. Homma, K. Schlegel,C.H.R. Schnit-zler, Inflamm. Res. 44 (1995) 423.
[2] P. Stei, B. Kross, J. Wiegleb, V. Trach, Boehringer Ingel-heim, 1993, Data on File.
[3] G. Engelhardt, Presented at Ninth International Confer-ence on 'Prostaglandins and Related Compounds' June
6-10, 1994, Florence, Italy.
[4] D. Turck, U. Busch, G. Heinzel, H. Narjes, G. Nehmiz,Clin. Drug Invest. 9 (1995) 270.
[5] U. Busch, G. Heinzel, G. Bozler, Abstracts of XVIIthILAR Conference of Rheumatology, Brazilian Society 01
Rheumatology, Campinas, Brazil, 1989, p. 190.[6] D. Turck, U. Busch, G. Heinzel, H. Narjes, Eur. J.
Rheumatol. Inflamm. IS (1995) 23.[7] U. Busch, Boerhinger Ingelheim, 1989, Data on file.
[8] T. Velpandian, J. Jaiswal, R.K. Bhardwaj, S.K. Gupta, J.
Chromatogr. B. Biomed. Sci.Applic. 738 (2) (2000) 431.[9] J. Schmud, W. Roth, in: D.J. Benford, J.W. Bridges,
G.G. Gibson (Eds.), Drug Metabolism: From Moleculeto Man, Taylor and Frances, London, 1987.
[10] M. Gibaldi, Biopharmaceutics and Clinical Pharmacoki-netics, fourth ed., Lea and Febiger, Philadelphia, 1991.
Mean :t S.D.
2 3
76.26 78.34 78.21 :t 1.8886.08 88.66 83.66:t 2.1188.23 84.54 84.54 :t 3.77
Reference Ratio 90% confidence interval . "
33580.3 1.169 98.11-103.11
42155.20 1.209 99.46-102.531111.70 1.051 83.15-117.18