5
High-performance liquid chromatographic determination of naltrexone in plasma of hemodialysis patients K. Kambia 1 , S. Bah 1 , T. Dine 1 *, R. Azar 2 , P. Odou 1 , B. Gressier 1 , M. Luyckx 1 , C. Brunet 1 , L. Ballester 1 , M. Cazin 1 and J. C. Cazin 1 1 Laboratoire de Pharmacologie, Pharmacocine ´ tique et Pharmacie clinique, Faculte ´ des Sciences Pharmaceutiques et Biologiques, 3 rue du Pr Laguesse, BP 83, 59006 Lille Cedex, France 2 Centre Hospitalier Ge ´ ne ´ ral de Dunkerque, De ´ partement de me ´ decine interne B, Service de ne ´ phrologie et he ´ modialyse, Avenue L. Herbeaux, BP 6-367 59385 Dunkerque Cedex 1, France Received 5 February 1999; revised 14 July 1999; accepted 23 July 1999 ABSTRACT: A simple, sensitive, selective and reliable reversed-phase high-performance liquid chromatographic (HPLC) method with UV detection is described for the determination of naltrexone in plasma samples. Naltrexone and the internal standard, naloxone, were isolated from plasma either with a liquid–liquid extraction method using ethyl acetate or with a solid-phase extraction method using Sep-Pack C 18 cartridge before chromatography. The extracts were dried under a stream of nitrogen and the samples were reconstituted in the mobile phase, then 20 mL were injected on a Waters Symmetry C 18 column (5 mm particle size, 4.6 150 mm). The mobile phase consisted of 0.06% triethylamine (pH 2.8)–acetonitrile (92:8, v/v) pumped at 1 mL/min. The peak–area ratio versus plasma concentration was linear over the range of 10–500 ng/mL and the detection limit was less than 8 ng/mL. Quantification was by ultra-violet detection at 204 nm. The present method was applied to the determination of the plasma concentration of naltrexone in dialyzed patients. Patients (n = 8) with severe generalized pruritus received 50 mg of naltrexone orally per day for 2 weeks. The variability in the therapeutic response in treated patients required plasma concentration investigations of this opioid antagonist. Copyright # 2000 John Wiley & Sons, Ltd. INTRODUCTION Pruritus is still one of the most vexing and disabling symptoms in chronic renal failure, afflicting up to 85% of patients on dialysis (Young et al., 1973). Despite two decades of clinical investigation into uraemic pruritus, little is known about the underlying pathogenetic mech- anisms, and effective therapeutic strategies are still elusive. The efficacy of a variety of agents including antihistamines has not been demonstrated convincingly (Wolfhagen et al., 1997). Naltrexone (NALT), an opiate receptor antagonist administered orally, is used to treat opiate addiction or dependance. Some studies have reported on the m- receptor antagonist NALT as an effective treatment modality in pruritus in patients undergoing dialysis, but the way by which this drug works in uraemic pruritus is not clear (Peer et al., 1996; Moreiras et al., 1998). It can be speculated that the m-receptor antagonist acts either centrally (by interfering with pruritus perception) or peripherally (by inhibition of histamine release or immunomodulation), as suggested by the authors. The lack of therapeutic response to NALT in some cases requires plasmatic concentration to be followed during the dialysis period (start, middle and the end). Many techniques have been described for the determination of NALT in biological materials, such as gas chromatogra- phy (GC) (Vereby et al., 1980), HPLC with ampero- metric (O’Connors et al., 1989) or coulometric detection (Zuccaro et al., 1991) or with UV detection (Asali et al., 1983). GC methods require extensive sample clean-up and derivatization procedures prior to assay, making these procedures time-consuming (Zuccaro et al., 1991). HPLC techniques with amperometric detection require stable conditions for good baseline stability, frequent electrode cleaning and, consequently, a thorough clean- up of sample (Zuccaro et al., 1991). Coulometric detectors are claimed to be more sensitive and selective than amperometric ones, and with shorter maintenance times, even though an amperometric cell, when con- taminated, can easily be dismantled and restored, whereas a coulometric cell usually needs to be replaced (Zuccaro et al., 1991). In the case of HPLC with UV detection, an endogenous peak was observed in the blood samples collected from an apnoeic, preterm infant (Asali et al., 1983). In addition, the methods described above BIOMEDICAL CHROMATOGRAPHY Biomed. Chromatogr. 14: 151–155 (2000) *Correspondence to: T. Dine, Laboratoire de Pharmacologie, Pharma- cocine ´tique et Pharmacie Clinique, Faculte ´ des Sciences Pharmaceu- tiques et Biologiques, 3 rue du Pr Laguesse, BP 83, 59006 Lille Cedex, France. Abbreviations used: L-L, Liquid–Liquid; NALT, naltrexone; S-P, Solid-Phase. Copyright 2000 John Wiley & Sons, Ltd. ORIGINAL RESEARCH

High-performance liquid chromatographic determination of naltrexone in plasma of hemodialysis patients

Embed Size (px)

Citation preview

Page 1: High-performance liquid chromatographic determination of naltrexone in plasma of hemodialysis patients

High-performance liquid chromatographic determinationof naltrexone in plasma of hemodialysis patients

K. Kambia1, S. Bah1, T. Dine1*, R. Azar2, P. Odou1, B. Gressier1, M. Luyckx1, C. Brunet1, L. Ballester1,M. Cazin1 and J. C. Cazin1

1Laboratoire de Pharmacologie, Pharmacocine tique et Pharmacie clinique, Faculte des Sciences Pharmaceutiques et Biologiques, 3 rue du PrLaguesse, BP 83, 59006 Lille Cedex, France

2Centre Hospitalier Ge ne ral de Dunkerque, De partement de me decine interne B, Service de ne phrologie et he modialyse, Avenue L.Herbeaux, BP 6-367 59385 Dunkerque Cedex 1, France

Received 5 February 1999; revised 14 July 1999; accepted 23 July 1999

ABSTRACT: A simple, sensitive, selective and reliable reversed-phase high-performance liquid chromatographic (HPLC) methodwith UV detection is described for the determination of naltrexone in plasma samples. Naltrexone and the internal standard, naloxone,were isolated from plasma either with a liquid–liquid extraction method using ethyl acetate or with a solid-phase extraction methodusing Sep-Pack C18 cartridge before chromatography. The extracts were dried under a stream of nitrogen and the samples werereconstituted in the mobile phase, then 20mL were injected on a Waters Symmetry C18 column (5mm particle size, 4.6� 150 mm).The mobile phase consisted of 0.06% triethylamine (pH 2.8)–acetonitrile (92:8, v/v) pumped at 1 mL/min. The peak–area ratio versusplasma concentration was linear over the range of 10–500 ng/mL and the detection limit was less than 8 ng/mL. Quantification was byultra-violet detection at 204 nm. The present method was applied to the determination of the plasma concentration of naltrexone indialyzed patients. Patients (n = 8) with severe generalized pruritus received 50 mg of naltrexone orally per day for 2 weeks. Thevariability in the therapeutic response in treated patients required plasma concentration investigations of this opioid antagonist.Copyright# 2000 John Wiley & Sons, Ltd.

INTRODUCTION

Pruritus is still one of the most vexing and disablingsymptoms in chronic renal failure, afflicting up to 85% ofpatients on dialysis (Younget al., 1973). Despite twodecades of clinical investigation into uraemic pruritus,little is known about the underlying pathogenetic mech-anisms, and effective therapeutic strategies are stillelusive. The efficacy of a variety of agents includingantihistamines has not been demonstrated convincingly(Wolfhagenet al., 1997).

Naltrexone (NALT), an opiate receptor antagonistadministered orally, is used to treat opiate addiction ordependance. Some studies have reported on them-receptor antagonist NALT as an effective treatmentmodality in pruritus in patients undergoing dialysis, butthe way by which this drug works in uraemic pruritus isnot clear (Peeret al., 1996; Moreiraset al., 1998). It canbe speculated that them-receptor antagonist acts eithercentrally (by interfering with pruritus perception) or

peripherally (by inhibition of histamine release orimmunomodulation), as suggested by the authors. Thelack of therapeutic response to NALT in some casesrequires plasmatic concentration to be followed duringthe dialysis period (start, middle and the end). Manytechniques have been described for the determination ofNALT in biological materials, such as gas chromatogra-phy (GC) (Verebyet al., 1980), HPLC with ampero-metric (O’Connorset al., 1989) or coulometric detection(Zuccaroet al., 1991) or with UV detection (Asaliet al.,1983).

GC methods require extensive sample clean-up andderivatization procedures prior to assay, making theseprocedures time-consuming (Zuccaroet al., 1991).HPLC techniques with amperometric detection requirestable conditions for good baseline stability, frequentelectrode cleaning and, consequently, a thorough clean-up of sample (Zuccaroet al., 1991). Coulometricdetectors are claimed to be more sensitive and selectivethan amperometric ones, and with shorter maintenancetimes, even though an amperometric cell, when con-taminated, can easily be dismantled and restored,whereas a coulometric cell usually needs to be replaced(Zuccaroet al., 1991). In the case of HPLC with UVdetection, an endogenous peak was observed in the bloodsamples collected from an apnoeic, preterm infant (Asaliet al., 1983). In addition, the methods described above

BIOMEDICAL CHROMATOGRAPHYBiomed. Chromatogr.14: 151–155 (2000)

*Correspondence to: T. Dine, Laboratoire de Pharmacologie, Pharma-cocinetique et Pharmacie Clinique, Faculte´ des Sciences Pharmaceu-tiques et Biologiques, 3 rue du Pr Laguesse, BP 83, 59006 Lille Cedex,France.

Abbreviations used: L-L, Liquid–Liquid; NALT, naltrexone; S-P,Solid-Phase.

Copyright 2000 John Wiley & Sons, Ltd.

ORIGINAL RESEARCH

Page 2: High-performance liquid chromatographic determination of naltrexone in plasma of hemodialysis patients

did not useboth liquid–liquid andsolid-phaseextractionmethodssuccessfully.

This paperdescribesa simpleandreliablemethodforthe determination of NALT in plasma of dialyzedpatientsby HPLC with UV detection,usingsimpleandsensitiveliquid–liquid or solid-phaseextractionmethods.

EXPERIMENTAL

Chemicals. Naltrexone (NALT) and the internal standard,naloxone, were obtained from Sigma-Aldrich (Saint-Quentin-Fallavier, France).HPLC-gradeacetonitrile and methanolwerepurchasedfrom Aldrich (Marchiennes,France).Analytical gradeethyl acetate, sodium hydroxide, phosphoric acid, potassiumphosphateand triethylaminewere obtainedfrom Prolabo(Paris,France).The waterusedto prepareaqueousbuffersanddilutionswas de-ionizedand purified by distillation (Milli-Q, Millipore,Saint-QuentinYvelines,France).All theotherreagentsusedwereanalyticalgrade.

Preparation of standard solutions and plasma samples.Stocksolutionsof NALT andnaloxonewerepreparedin mobilephaseat 1 mg/mL and storedat 4°C. The working solutionsforplasma spiking (NALT at 0.1, 0.5, 1, 2.5 and 5mg/mL andnaloxoneat1mg/mL) werefreshlyprepared.Thestandardsamplesfor NALT were preparedby spiking blank plasma(1 mL) with100mL of theworkingsolutions.Calibrationgraphswereobtainedby analyzing1 mL plasmasamplesspikedwith 10, 50, 100, 250and500ng of NALT and100ng of naloxone.

HPLC. Chromatographicanalyseswere performedwith an HP1090 high-performanceliquid chromatograph(Hewlett-Packard,Orsay, France) equipped with a variable-volume injector, anautomaticsamplingsystemanda Hewlett-PackardModel 79994Adiode-arrayUV detectoroperatingat 204nm.Theoutputfrom thedetectorwas connectedto a Hewlett-Packard9000 Model 300integratorand the datawere recordedon a HP Thinkjet printer.Separationwas achievedusing a 5mm WatersSymmetry1 C18

column (4.6� 150mm) (Waters, Milford, MA). During assaydevelopment,NALT waselutedisocraticallywith a mobilephaseconsistingof acetonitrileandaqueousbuffer (triethylamine0.06%ajustedto pH 2.8with phosphoricacid1M) mixture(8:92,v/v) ataflow-rateof 1.0mL/min. Themobile phasewasfiltered througha0.45mm membraneand degassedundera helium streambeforeuse.The run-timewas7 min.

Procedure for samples extraction. NALT quantificationrequiresa plasmaextraction procedurebefore chromatographicanalysis.Two methodshave beendevelopedfor NALT plasmaextraction.

Solid-phaseextraction: The plasma samples (1 mL) weretreated in glass centrifuge tubes with 100mL of the internalstandardsolution and were applied to an octadecyl extractioncolumn (C18 Sep-Pakcartridge), which had been previouslywashedwith 2 mL of methanoland2� 2 mL of 0.01M potassiumphosphatebuffer (pH 7), respectively.The cartridge was thenwashedtwice with 2 mL of the mixture water/acetonitrile(80:20,v/v) and then twice with 2 mL of water. Finally, NALT and

naloxonewereelutedwith 3 mL of acidifiedmethanol(pH 3) intoglasstubes.The tubecontentwasevaporatedto drynessin water-bathat 40°C undernitrogen.Theresiduewasdissolvedin 100mLof mobilephaseandaftercentrifugation,20mL of thesupernatantwereinjectedinto the column.

Liquid–liquid extraction: eachsample(1 mL) wastreatedwith100mL of theinternalstandardsolutionin theglasstubes,followedby sodiumhydroxide1M (300mL) andethyl acetate(5 mL). Themixturewasvortexed(1 min), centrifuged(1620g for 5 min) andthe separatedorganic layer (fraction 1) was transferredto cleanconicaltubes.Theaqueousphasewasextractedagainwith 5 mL ofethyl acetateandthemixturewastreatedasabove.Theseparatedorganicphase(fraction 2) wascombinedwith fraction 1 and thetotal organicphasewasevaporatedto drynessin a water-bathat40°C undernitrogen.Thesampleswerereconstitutedin theHPLCmobilephase(100mL) andaliquots(20mL) wereinjectedinto thecolumn.

RESULTS AND DISCUSSION

Chromatography

In thepresentstudy,weusedbothliquid–liquid (L-L) andsolide-phase(S-P) extraction methods, because weobtaineda satisfactoryaveragerecoveryof drugs(S-P:NALT, 94%; naloxone, 92%; L-L: NALT, 93%;naloxone, 95%). However, the L-L extraction usingdouble extraction proceduremakesthis method time-consuming.

Figure 1 illustrates chromatogramsobtained fromextractedplasmasamplesusingS-Pextractionprocedure.Figure 1(A) showsa representativechromatogramof aprocessedblank, indicating that no endogenouscom-poundsexist at the retentiontime of NALT or internalstandard.Figure1(B) is obtainedfrom a plasmasamplecontaining 500ng/mL of NALT and 100ng/mL ofnaloxoneastheinternalstandard.Identificationof NALTand naloxone was achievedby comparisonwith theauthenticspecimens.The retentiontimeswere3.95minfor naloxoneand5.90min for NALT. Thecapacityfactork' valueswere2.3for naloxoneand4 for NALT, showinga satisfactoryseparationand a relatively short analysistime.

Linearity, precision and accuracy

Analytical reproducibility from plasmacontaining 10,50, 100, 250 and 500ng addedNALT was assessedinfive replicatessamples(Table1). The calibrationcurveswereobtainedfor the peak–arearatio (NALT/naloxone)vs drug concentration.They were linear over the range10–500ng/mL. The mean linear regressionequationsobtained from five replicates were y = 7.9 xÿ 0.09(r = 0.9985) for the S-P extraction procedure andy = 6.82 x� 0.0236 (r = 0.9987) for L-L extraction (y,peak-area ratio; x, drug concentrationng/mL). No

Copyright 2000JohnWiley & Sons,Ltd. Biomed.Chromatogr.14: 151–155(2000)

152 ORIGINALRESEARCH K. Kambiaet al.

Page 3: High-performance liquid chromatographic determination of naltrexone in plasma of hemodialysis patients

significantdifferenceswere observedbetweenequationparameters.Standardcurveswerenot forcedthroughtheorigin.

The precisionof the assaywasestimatedby perform-ing simultaneouslythe analysisof replicates(5�) ofspikedplasmasamples.The precisionof both methodswas good,as shownby the lower valuesof intra-assayandinter-assaycoefficientsof variation(Table1). Intra-

assayreproducibility and accuracywere determinedbysimultaneouslyassayingreplicatesof plasmaspikedwithNALT. The intra-assayreproducibilitywaswithin 5.4%for S-Pextractionand 4.5% for L-L extractionand theaccuracy,which wasdefinedas(measuredconcentrationÿ theoretical concentration)/theoreticalconcentration,was10%and7.1%or better,respectively.

Inter-assayreproducibility and accuracywere deter-minedfrom spikedNALT plasmawith differentstandardcurves. The inter-assayreproducibility obtainedusingS-P or L-L extraction(Table 1) was within 8.7% and7.07%,respectively.Theabsoluteaccuracywas10%forS-Pextractionand14%for L-L extraction.

Extractionrecoveriesof drugandits internalstandardfrom plasmawereon averagebetterthan90% for bothS-P or L-L extractions.The detection limit was theminimumquantifiableconcentrationfor NALT, thatis tosay,8 ng/mL at a signal-to-noiseratio of 2.

Clinical applications

No significantdifferencewasshownbetweenextractionrecovery and precision of S-P and L-L extraction

Figure 1. Typical chromatogramsobtainedafter solid-phaseextractionfrom plasmasamples:(A) blank plasma;(B) blankplasmaspikedwith 500ng of NALT and 100ng of internalstandard(naloxone).

Table 1. Reproducibility and accuracyof NALT determinations (n = 5)

Amount Amount Inter-assay

Methods added(ng/mL) found(ng/mL) CV (%) Accuracy(%)

Solid-phaseextraction 10 9.8� 0.85 8.7 250 55� 3.90 7.09 10

100 98� 3.80 3.87 2250 249� 7.00 2.81 0.4500 505� 6.00 1.2 1

Liquid–liquid extraction 10 8.6� 0.30 3.50 1450 53� 2.50 4.70 6

100 99� 7.00 7.07 1250 249� 8.00 3.20 0.8500 502� 1.35 0.27 0.4

CV: coefficientof variation.

Figure 2. Chromatogramobtainedfrom plasmaof a NALT-treateddialysedpatient.

Copyright 2000JohnWiley & Sons,Ltd. Biomed.Chromatogr.14: 151–155(2000)

Determinationof naltrexone ORIGINALRESEARCH 153

Page 4: High-performance liquid chromatographic determination of naltrexone in plasma of hemodialysis patients

methods. Consequently, the solid-phase extractionmethod using Sep-Pak C18 cartridge was used forthe determination of NALT plasma concentrationsin dialyzed patients, becauseof its handling abilityand rapidity. Figure 2 showsa chromatogramobtainedfrom plasma of a dialyzed patient after NALTtreatment.

Uraemicpruritus is a commonand usually disablingsymptomin patientswith chronic renal failure. Endo-genousopioids (b-endorphin)may be involved in thispruritus (Peer et al., 1996; Kivity et al., 1997). Thecondition involves treatment with opiate antagonists(Bergasaet al., 1995;Carsonet al., 1996).Theprobableactionmechanismof NALT in relievinguraemicpruritusin hemodialysispatientsis through inhibition of hista-mine release(Peeret al., 1996).It is unclearwhy somepatientsrespondsatisfactorilyto NALT treatmentwhileothersdo not, so the aim of this study was to followNALT plasmaconcentrationsduringdialysissessions(atthestart,middle andend).

After approval by the local ethical committee andobtaining informed consent,eight adults patients onregular hemodialysiswith severeresistantgeneralizedpruritus were consideredsuitablefor the study.Duringthe treatmentperiod,patientsreceived50mg of NALT(onetabletof Nalorex* 50mg) orally oncedaily. Eightdaysafterthebeginningof treatment,bloodsampleswerecollecteddirectly from thehemodialysisline at thestart,middle andthe endof a 4 h dialysissession.They wereimmediatelycentrifugedand plasmawas collectedandstoredatÿ20°C until HPLC analysis.

Figure3 showstheplasmatimecourseof NALT levelsduringdialysisin four treatedpatients.Followinganoraldose of 50mg, the plasma concentrationof NALTshowedlargedifferencesbetweenpatientsstudiedduringdialysissession.Major variationsof theplasmalevelsofthedrugwereobservedat thebeginning,themiddleandthe end of dialysis session.In four naltrexone-treatedpatients,plasmaconcentrationsof NALT rangedfrom

0.55 to 0.08mg/mL betweenthe start and the end ofdialysissession.

Uraemicprurituswasratedby patientsusinga visualanalog scale ranging from 0 (no pruritus) to 10(maximum intensity of pruritus). The scoresrecordedby eachpatientduring the study period were averageddaily. For two of the four patients treated, the totalpruritusscorewasreduced(7/10and8/10, respectively)andthe prurituswasrelievedsignificantlywhenplasmaNALT concentrationsdecreased(Fig. 3, patients2 and3). During the same period, the other two patientspresentedregular and stable concentrationsof NALTwithout clinical improvement (the pruritus score re-mainedunchanged,Fig. 3, patients1 and4).

When drug concentrationdecreased,pruritus wassignificantly relieved. These results support the find-ings of Ferrari et al. (1998). Indeed, attention mustfocus on the part played in clinical efficacy by themain metabolite,6 b-naltrexol, which exerts minimalbut long-lasting potent opioid antagonistactivity. Inpharmacokineticsstudies,plasmalevels of 6b-naltrexolwere usually higher than the parent drug (Wolfhagenet al., 1997; Ferrari et al., 1998). However, the de-creasedhepatic first-pass metabolism of NALT canexplainthe largevariability observedin plasmaconcen-trations,which is answerablefor a decreaseof clinicalefficacy.

In view of thesedata,dialysisseemsto haveno effecton NALT blood levels,howeverit shouldbe interestingto studythe pharmacokineticsof this drug betweentwodialysis sessions.Further studies with more dialyzedpatientsare neededto determineboth NALT and 6 b-naltrexol concentrationsin plasmaof patients and tocomparetheclinical response.

Acknowledgements

Thetechnicalassistanceof M.C. Pironis acknowledged.

Figure 3. Time-courseof NALT plasmaconcentrationsduringa dialysissession.

Copyright 2000JohnWiley & Sons,Ltd. Biomed.Chromatogr.14: 151–155(2000)

154 ORIGINALRESEARCH K. Kambiaet al.

Page 5: High-performance liquid chromatographic determination of naltrexone in plasma of hemodialysis patients

REFERENCES

Asali, L. A., Nation, R. L. and Brown, K. F. 1983. Journal ofChromatography278:329.

Bergasa,N. V., Alling, D. W. and Talbot, T. L. 1995. Annals ofInternal Medicine123:161.

Carson,K. L., Tran,T. T. andCotton,P. 1996.AmericanJournal ofGastroenterology91:1022.

Ferrari,A., Bertolotti, M., Dell’Utri, A., Avico, U. and Sternieri,E.1998.Drug andAlcohol Dependency52:211.

Kivity, S.,Fireman,E.,Schwarz,Y., Peer,G.,Agami,O. andIaina,A.1997.InflammationResearch46:404.

Moreiras,M., Goyanes,G.R.,Cuina,L., Pineiro,A. G.andPerez,A. J.1998.Nephrology,DialysisandTransplantation13:232.

O’Connors,E. F., Chen,S.W. T. andNorth, W. G. 1989.Journal ofChromatography, 491:240.

Peer,G., Kivity, S.,Agami,O., Fireman,E., Silverberg,D., Blum, M.andIaina,A. 1996.TheLancet, 348:1552.

Vereby,K., De Pace,A., Jukofsky,D., Volavka,J. V. andMule, S. J.1980.JournalAnalytical Toxicology4:33.

Wolfhagen,F. H. J.,Sternier,E., Hop,W. C. J.,Vitale, G., Bertolotti,M. andVan Buuren,H. R. 1997.Gastroenterology, 113:1264.

Young,A. W., Sweeney,E.W., David,D. S.,Cheigh,J.,Hochgelerent,E.L., Sakai,S.,Stenzel,K. H. andRubin,A. L. 1973.NYSt.Journalof Medicine, 73:2670.

Zuccaro,P.,Altieri, I., Betto,P.,Pacifici,R. andRicciarello,G. 1991.Journalof Chromatography, 567:485.

Copyright 2000JohnWiley & Sons,Ltd. Biomed.Chromatogr.14: 151–155(2000)

Determinationof naltrexone ORIGINALRESEARCH 155