University of Groningen

Pharmacogenetic differences between , and Beinema, Maarten; Brouwers, Jacobus R. B. J.; Schalekamp, Tom; Wilffert, Bob

Published in: and Haemostasis

DOI: 10.1160/TH08-04-0116

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record

Publication date: 2008

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA): Beinema, M., Brouwers, J. R. B. J., Schalekamp, T., & Wilffert, B. (2008). Pharmacogenetic differences between warfarin, acenocoumarol and phenprocoumon. Thrombosis and Haemostasis, 100(6), 1052-1057. https://doi.org/10.1160/TH08-04-0116

Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license. More information can be found on the University of Groningen website: https://www.rug.nl/library/open-access/self-archiving-pure/taverne- amendment.

Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

Download date: 01-10-2021 ©2008 Schattauer GmbH,Stuttgart

Review Article

Pharmacogenetic differencesbetween warfarin,acenocoumarol andphenprocoumon

Maarten Beinema1;JacobusR.B.J.Brouwers2;Tom Schalekamp3;Bob Wilffert2 1ThrombosisCentre,Deventer Hospital,Deventer,The Netherlands; 2Department of Pharmacotherapy and Pharmaceutical Care,University of Groningen, Groningen, The Netherlands; 3Division of Pharmacoepidemiologyand Pharmacotherapy,Faculty of Science,UtrechtInstitute forPharmaceutical Sciences, Utrecht University,Utrecht, The Netherlands

Summary oral drugs have proven to be effective VKAs. TheCYP2C9 polymorphismisassociated with delayed forthe prevention of thromboembolic events.World-wide,war- stabilisationfor coumarin .Theeffects of CYP2C9 farinisthe most prescribeddrug.In Europe,acenocoumarol and polymorphisms on the and anticoagulant re- phenprocoumonare also administered.Ye tithas been proven sponse areleast pronounced in the case of phenprocoumon. In that variant allelesofthe VKORC1and CYP2C9genotypes in- the long term,patientsusing phenprocoumon have moreoften fluence the pharmacokineticsand pharmacodynamics of these international normalisedratio (INR) valuesinthe therapeutic drugs. Thecombination of thesetwo variant genotypesisa range,requiring fewermonitoringvisits.This leadsustocon- major cause of theinter-individual differencesincoumarin anti- clude that in theabsence of pharmacogenetic testing, phenpro- coagulantdrugdosage.Individualswho test positive forboth coumon seemspreferable foruse in long-term therapeutic anti- variantgenotypes areatincreasedriskofmajor bleeding.The . Pharmacogenetic testing before initiating coumarin impact of the CYP2C9 andVKORC1genotypeismost significant oral anticoagulants mayadd to the safety of all coumarin antico- duringthe initialperiod of coumarin anticoagulant therapy.The agulants especiallyinthe elderlyreceiving multiple drugs. effect ofVKORC1allelic variants is relativelysimilarfor all three

Keywords Clinical trials,oral anticoagulants,pharamacogenetics,phar- macodynamics ThrombHaemost 2008; 100: 1052–1057

Introduction The major side effect of VKAsismajor bleeding,with re- ported incidences of 1.5to5.0 per 100 patient-years (2,3). The Coumarinanticoagulants, also called vitamin Kantagonists incidenceofboth bleeding and thromboembolic eventsincreases (VKAs),are effective in the prevention of venous and arterial sharplywith advanced age. Therisk of over-and under-coagu- thromboembolism (1).VKAs suppress theregeneration of the lation in patients taking VKAsisassociatedwith drug-VKAs in- reduced formofvitamin Kbyinhibiting vitamin Kepoxide re- teractions, food-VKAs interactions, and disease-VKAs interac- ductase. The vitamin Kcycle regenerates reducedvitamin K1 tions.Alcoholconsumption, liverdisease, and other unknown from its epoxide.. Reduced vitamin Kisacofactor for post-trans- factorsalso influenceoptimal dailydosages. Other inter-individ- lational gamma-carboxylation of glutamic acidresiduesonsev- ual variationsthat affect predicting optimal dailydosagesin- eral proteins fornormal haemostasis. This resultsinnegatively clude pharmacogenetic predisposition, age and patient obesity, charged gamma-carboxyglutamateson factorsII, VII, IX, and X, whichcorresponds to the amount of coumarin anticoagulants whichbind to calcium cations and thentoplateletphospholipid requiredfor initiation of therapy(4–6). This explains whyoral membranes. Gamma-carboxylation is also requiredfor the de- anticoagulants have asmall therapeutic window, and the stability velopment of othertissues. VKAs also inhibit the gamma-car- of anticoagulant therapycan be easilydisturbed. boxylation of anticoagulant proteins C, S, Z, and osteocalcin. Both safety (primarily risk of bleeding)and effectiveness of VKAsinhibition of clotting factor activity is their main phar- VKAstherapyrelates to blood international normalised ratio macologic effect.This is responsible for VKAsability to inhibit (INR) values(7). Monitoring of INRand dose adjustments of clot formation. coumarin anticoagulants arefrequently required. Futhermore,

Correspondence to: Received: April 4, 2008 MaartenBeinema, MD Accepted after major revision: September14, 2008 DeventerHospital Thrombosis Centre PO box5001, 7400GCDeventer,The Netherlands Prepublished online: November 13, 2008 Te l.: +31570 646286, Fax: +31 570 646287 doi:10.1160/TH08-04-0116 E-mail: [email protected]

1052 Beinema et al.Pharmacogeneticdifferencesbetween warfarin, acenocoumaroland phenprocoumon pharmacogenetics,the fieldofresearch describing the influence This complex recycles reduced , whichisessential for of variationsofDNA characteristics on drug response,plays an the post-translationalgamma-carboxylation of vitamin K–de- important roleinthe safety and effectiveness of VKAs. pendent clotting factorsII(prothrombin),VII, IX,and X, and of Warfarin is the VKAs drug of choiceinmost countries. Ye t proteins C, Sand Z. acenocoumarol and phenprocoumon are used in many European The impact of other enzymes polymorphisms, likeGGCX countries, includingThe Netherlands and Germany. Several re- (gamma-glutamyl carboxylase), on coumarin anticoagulant dose viewsand researchpapersonthe pharmacogenetic influences of finding has beenexamined,but thereare no significant results. warfarin are available(8, 9). Comparatively systematic informa- tion on the pharmacogenetic influences of acenocoumarol and Warfarin phenprocoumon is scarce. In this review,wefocus on the phar- macogeneticsofphenprocoumon and acenocoumarol, compar- Warfarin is aracemate, with S-warfarinbeing three timesaspo- ing these drugs with the pharmacogeneticsofwarfarin. We also tent as R-warfarin. TheS-enantiomerinwarfarinispredomi- exploredthese drugs’ relevance to therapeutic choices. nantlyresponsible for the anticoagulant effect.When adminis- We performed the literaturesearch in EMBASE and Medline tered orally, warfarin is completelyabsorbedand 99% is bound (PUBMED)from 1995 to June2008. We includedstudieswith to albumin in the plasma. The liverabsorbs the free warfarin originaldataonpharmacogeneticsofphenprocoumon and ace- where it exerts its anticoagulant effect andis metabolized by sev- nocoumarol. Data from warfarin reviewsand additional relevant eral CYP-enzymes. papers not included in the cited reviews were alsoincluded. Ye tfrom aclinical standpoint, R-warfarinisthe most active Search words and termsincluded: pharmacogeneticsand/or anticoagulant. This is becauseCYP2C9metabolisesS-warfarin pharmacogenomics; acenocoumarol;phenprocoumon; in the first-pass-effect very efficiently.CYP2C9convertsthe CYP2C9 and/or genotype and /or polymorphism;VKORC1; S-enantiomer to 6- and 7-hydroxy-warfarin, whichiseventually warfarin and review;and coumarin anticoagulants. excreted in the bile.Incomparison, CYP1A1, CYP1A2, and CYP3A4 metabolize R-enantiomer into an inactive alcohol- metabolite excreted in the urine. Cytochrome P4502C9 andVKORC1 and other Patients with eitheraCYP2C9*2 or CYP2C9*3 polymor- genetic variants phismhaveimpaired metabolismofthe more active enantiomer S-warfarinwhen comparedtopatients whoare homozygous for CytochromeP450 (CYP) is agroupofhepatic microsomal the wild-typeallele CYP2C9*1(11). The unbound S-warfarin enzymes thatact as monooxygenases of endobiotics (steroid hor- clearance is about two- to three-fold lowerinheterozygouscar- mones, fatty acids derivates, and vitamins) and xenobiotics riers of the CYP2C9*3 allele and about 10-fold lowerinhomozy- (drugs, pollutants, and ).The abbreviation CYP,fol- gous carriers (12). lowedbyanumber,acapitalletterand anothernumber (e.g. In vivo,these twoCYP2C9SNPshavebeen associatedwith CYP2D6, CYP3A4, CYP2C9), designates an individual enzyme increased responsiveness to warfarin (13). Aithaletal. (14)com- from this group of monooxygenases. Cytochromes transformli- pared controls requiring typical warfarin doses to patientswhose pophilic drugs into morehydrophilic metabolites whichfacili- therapeutic warfarin dose was10.5 mg/weekorless. Theyfound tatesfurther elimination and renal excretion. that patients requiring lowdoses of warfarin were moreproneto Thegene CYP2C9 encodes the enzyme CYP2C9, of which asupratherapeutic INRatthe timeofwarfarininduction; were al- about 30 variant alleles have beendescribed.The most fre- most fourtimesmore prone to bleed; and were sixtimesmore quentlyoccurring variant alleles in Caucasians are CYP2C9*2 likelytohavethe CYP2C9*2 or CYP2C9*3SNPs. Othersfound (CGT>TGT in exon 3) and CYP2C9*3 (ATT>CTT in exon 7) that CYP2C9*3 decreasedthe selectivity of CYP2C9 for S-war- (10). farinand that residue 359, the mutated amino acid,was acom- Thesefactors lead CYP2C9toplayarole in the metabolismof: ponent of the warfarin-binding site. –Coumarin anticoagulant drugs (warfarin, acenocoumarol, Patients with one or twoofthese SNPshavereduced warfarin phenprocoumon) requirementsand an elevatedrisk of an adverseevent by initial –Nonsteroidalanti-inflammatorydrugs (NSAIDs) (e.g. diclo- warfarin therapy. The CYP2C9SNPsare associated with atwo- fenac,, lornoxicam, celecoxib, flurbiprofen, na- to three-fold increasedrisk of bleeding during warfarin induction proxen) (15). This observation suggests thatpharmacogenetics-based –Sulfonylureas (e.g. tolbutamide, glipizide) warfarin therapywill mainlyaffect the initialwarfarindosage(s). –Phenytoin However, because aCYP2C9polymorphism is also associated with adecreased chance to achievestableanticoagulation (16),it Inhibitors of CYP2C9 include some SSRI type antidepressants is possible thatgenotype-steered dosing can alsoplayarole dur- (e.g. fluoxetine,fluvoxamine), amiodarone, benzbromarone, co- ing later stagesofwarfarintherapy. trimoxazol,cimetidine and antimycotic drugs (e.g.fluconazole, Recently CYP4F2 genetic variantswere associatedwith a miconazole, voriconazole). Inducers of CYP2C9 includerif- clinically relevant effect on warfarin requirement in the Cau- ampicinand carbamazepin. casianpopulation (17).Datafrom other ethnicgroups and the The gene VKORC1 encodes vitamin K-epoxide reductase roleCYP4F2inacenocoumarol and phenprocoumon dose (VKORC1),ofwhich several variant alleles have beende- requirement are lacking. scribed. Coumarin anticoagulant derivativestargetsVKORC1.

1053 Beinema et al.Pharmacogeneticdifferencesbetween warfarin, acenocoumaroland phenprocoumon

Several geneticvariations of the VKORC1 gene have been VKORC1 SNPscovaried significantlywith warfarin dose, found to influencesensitivity to warfarin (18,19). Four different whichexplains 30% of dose variations(22). Cytochrome P450 heterozygous mutations in the VKORC1 gene were found in in- 2C9 (CYP2C9) explains 12% of variance in warfarin dose (23). dividualswith warfarin resistance. Riederetal. (18)investigated TheVKORC1 C1173 polymorphismisinstrong disequilibrium the genetic basis of the broadvariation amongpatients in re- with VKORC1 promotor1639. In Europeans, guanine (G)isex- sponse to warfarin therapy. Theydeterminedthe VKORC1 ha- pressed. In some other ethnic groups, adenine(A) is more com- plotype frequenciesinAfrican-American,European-American mon. Aheterozygousorhomozygous adenine (A)significantly and Asian-Americanpopulations, and theyfound VKORC1 reduces VKOR expression comparedwith G/G.The VKORC1 mRNA expression in human liversamples.Theyfound 28 variation p.Trp59Arg(24) and Asp36Tyr (25) have beende- VKORC1 SNPsinthe primarypopulation. Riederetal. ident- scribedtoexplain coumarin resistance. ified10common (>5%) noncoding VKORC1 single- Incorrect dosage,especially during the initial phase of treat- polymorphisms(SNPs) and inferred five major haplotypes, ment, carriesahigh risk for either severe bleeding or failureto from whichalow-dose haplotypegroup(A) and ahigh-doseha- prevent thromboembolism.Genotype-based dose predictions plotypegroup(B) were derived. mayenable personalizeddrugtreatment from the start of warfa- Initial variability in the INR response is morestrongly as- rintherapy(26). sociatedwith VKORC1 thanwith CYP2C9 (20). Themean maintenancedosagesofwarfarindiffers significantly amongthe Acenocoumarol threehaplotypegroupcombinations,atapproximately2.7 mg/ dayfor A/A, 4.9 mg/dayfor A/B,and 6.2 mg/dayfor B/B. LikeS-warfarin,S-acenocoumarol is morepotent than R-aceno- VKORC1 haplotypegroups Aand Bexplains approximately coumarol and is metabolizedbyCYP2C9(27–30). CYP1A2, 25% of the variance in warfarin dose. Moreover, one singlenu- 3A4, 2C9 and 2C19 metabolize R-acenocoumarol.An important cleotide polymorphism,such as C1173T in intron 1, appearsto pharmacokinetic difference betweenwarfarinand acenocouma- be as informative about coumarinanticoagulant sensitivity as all rol is that S- and R-warfarinhavehalf-lives of approximately 32 the five major haplotypes, whichrepresents96% to 99% of the and 43 hours, while S- and R-acenocoumarol have half livesof2 total haplotypes. Asian-Americans have ahigherproportion of and 8hours. As aresultofslowerelimination of the S-enanti- group Ahaplotypes,and African-Americans have ahigherpro- omer, R-acenocoumarol is largelyresponsible for the overall portion of group Bhaplotypes.Wadelius et al.(21) finds that anticoagulant response. Furthermore, R-acenocoumarol is clini-

CoumarinCYP T

R-Warfarin 1A2, 3A4, 2C8, 2C19 43

S-Warfarin 2C9, (2C8) 32

R-Acenocoumarol 1A2, 3A4, 2C9, 2C19 8

S-Acenocoumarol 2C9 2

R-Phenprocoumon 3A4, 2C9 172

S-Phenprocoumon 2C9, 2C8, 3A4 156

Figure1:The influence of None CYPisoenzymes on cou- (40%) marin anticoagulant hy- droxylation.

1054 Beinema et al.Pharmacogeneticdifferencesbetween warfarin, acenocoumaroland phenprocoumon callythe more important enantiomer because of the very short is essentiallyunchanged. About 40% of an oraldose of phenpro- half-life of S-acenocoumarol (31). coumon is excreted unchanged (48, 49), whereas warfarin and Patients with aCYP2C9variant allele have ahigherrisk of acenocoumarol are almostcompletelymetabolised(Table1) earlyacenocoumarol overanticoagulation (32,33). Patients with (50). Phenprocoumon therefore seemspreferable in poormeta- the CYP2C9*3, butnot the CYP2C9*2 variant,are knownto bolisers of coumarinanticoagulants. requirealower maintenancedose of racemic acenocoumarol Finally, likeS-warfarin and S-acenocoumarol,both S- and (34–37).The S-acenocoumarol clearanceinaheterozygouscar- R-phenprocoumon inhibitvitamin Kepoxide reductase (the rier of the CYP2C9*3allele is morethan 15-fold lowerthan in S-enantiomer being 2–5 times as potent as the R-enantiomer). wild-type subjects. This difference betweenCYP2C9*2 and *3 The effects of the CYP2C9 polymorphismsonthe phar- discerns acenocoumarol from warfarin.The S-acenocoumarol macokinetics and anticoagulant response areleastpronounced plasmahalf-lifeisprolonged in the CYP2C9*1/*3 genotypes. for phenprocoumon,yet the VKORC1 genotype can modifythe Theeffect is thatS-acenocoumarol pharmacokineticsisde- effect of the CYP2C9 genotype on phenprocoumon dose pendentonthe CYP2C9 genotype.Specifically, the presence of requirements. Greater variability in doserequirement is ob- the CYP2C9*3allele impairsoral clearance of the coumarin served by theVKORC1 genotype thanbythe CYP2C9 genotype. anticoagulants.Thetime needed to achievestability is mainlyas- Schalekamp et al. showedthat in patients without aVKORC1 sociatedwith the CYP2C9 genotype. variant allele, carriers of aCYP2C9variant need dosagesthat are Thegenotype VKORC1 modifiesthe effect of the CYP2C9 nearly30% lowerthan those for CYP2C9*1/*1 patients (51).In genotype on the anticoagulation status (38).Inseveral studies, patients with aVKORC1 variant allele, differences between car- the VKORC1 genotype,rather thanbythe CYP2C9 genotype, riers of aCYP2C9variant and CYP2C9*1/*1 are statistically in- explains alarger partofthe dose requirement. Severe, over anti- significant,suggesting that differencesbetween CYP2C9 geno- coagulation is mostlyassociatedwith the combination of variant types mainlyapplytopatients without aVKORC1 variant allele. alleles of CYP2C9 and VKORC1 (39–41). Carriers with acombination of CYP2C9 variant and VKORC1 variant alleles show asignificant increase in the risk of severe Phenprocoumon overanticoagulation, whereas delayed stabilization is mainlyas- sociatedwith the CYP2C9 genotype.Carriers of the *3 allele Phenprocoumon seemspreferable in poormetabolisers of cou- have ahigherrisk of bleeding (52).From aclinical perspective, marinanticoagulants (42).Phenprocoumon has along half-life: the bleeding risk of patients using phenprocoumon should be it is approximately 172 hours for the morepotent S-phenprocou- taken into consideration (53). mon enantiomer,and approximately 156 hours for R-phenpro- Patients usingphenprocoumon have morestableINR values coumon. The S-enantiomer is predominantlyresponsible for the thanpatients on acenocoumarol and requirefewer monitoring vi- anticoagulant effect in phenprocoumon. TheS-7-hydroxylation sits (54,55). Because of the long half-life time, overanticoagu- is the most important metabolising factor (43, 44).Ye tthe *2 and latedpatients usingphenprocoumon areatgreater risk of major *3 allele significantly compromisesthe S-7-hydroxylation in a bleedingsdue to the prolongedperiod of overanticoagulation. gene-dose-dependent manner. Phenprocoumon metabolismap- pears to be less influenced by the 2C9 genotypeswhen compared with other coumarin anticoagulants (45, 46). S- and R- phenpro- Coumarin anticoagulants andNSAIDs drug-drug coumon arepredominantlymetabolisedbyCYP2C9and 3A4 interactions (Fig. 1). The significant role of the non-polymorphic 3A4inphenpro- CYP2C9 metabolisesseveral drugs(seeabove), especially some coumon metabolism makes it apossibly saferdrugtouse over NSAIDs, making apharmacokinetic interaction with the couma- other coumarinanticoagulants. The oralclearance of S-aceno- rin anticoagulants conceivable(56–58). The effects of NSAIDs coumarol is morethan 15-fold lowerinacarrier of the heterozy- arewelldocumented. NSAIDs(excluding cyclooxygenase gous *3 genotypecompared with the *1 *1 genotype.The oral [COX]2-selectiveNSAIDs)inhibit plateletaggregation which clearanceofS-phenprocoumon is onlymarginally reduced in increases the risk of bleeding (59).This pharmacodynamic inter- variant allele carriers (47). Theclearance of R-phenprocoumon action is not reflectedbyachange in the prothrombin timeas

Ta ble1:The impact of pharmacogenetics on the different coumarindrugs. T½is thehalf-lifetime of acoumarin in theblood.The dose varianceisthe impact of thegenotype on thecoumarin dose required to getINR values withintargetrange.The administration of NSAIDs candelay themetabolismofsomecoumarins.

Dose variance (%) NSAID T½(hours) VKORC1 CYP2C9 Delayed metabolism Warfarin 32–4325(14–36) 13 (4–21) 1.32 fold Acenocoumarol 2–8 29 (21–36) 10 (5–14)1.28 fold Phenprocoumon 156–178 29 71.11 fold

1055 Beinema et al.Pharmacogeneticdifferencesbetween warfarin, acenocoumaroland phenprocoumon

foundthat it mainlyalters the pharmacokineticsofS-phenpro- What is knownabout thistopic? coumon (61). − Coumarinanticoagulants aremetabolized predominantly by CYP2C9.Their target on the enzyme vitamin Kepox- idase is VKORC1. Conclusion − CYP2C9 and VKORC1 genetic variantsare associated Knowingabout the presenceofboth CYP2C9 and VKORC1 in with variability in coumarinanticoagulant dose require- the patient’sgenotype broadens our insight into the activity of the ment, overanticoagulation, stabilization and bleeding risk. different coumarinanticoagulants in the individual patient con- tributing to asafer pharmacotherapy(62–66). What doesthispaper add? The effect of VKORC1 allelic variantsisrelativelysimilar − In this paperacomparison is made betweenthe role of for all three VKAsand explains 25% of the variance of the re- CYP2C9 and VKORC1 genetic variantsfor the coumarin sponse to coumarinanticoagulant therapy. The effect of the anticoagulants warfarin,acenocoumarol and phenprocou- CYP2C9*2 and *3 polymorphismsismost prominent for warfa- mon. rin. Foracenocoumarol,onlyCYP2C9*3 has an effect. Forphen- − We concludethat the impact of genetic variantsof procoumon, we observe amarginal effect of the CYP2C9 poly- CYP2C9 and VKORC1 on warfarin,acenocoumarol and morphisms, whichisprobably limitedtopatients without a phenprocoumon is quite different. VKORC1 variant allele. Thecombination of variant VKORC1 and CYP2C9 alleles explains amajor partofthe inter-individual differencesinVKAs dosages (67). The combination of variant VKORC1 and measured by the INR.Kohl et al. introducedamodel to approach CYP2C9 allelesincreases the risk of major bleedings. The im- the interaction of NSAIDswith coumarinanticoagulants based pact of the combination of the polymorphisms is the greatestin on CYP2C9 and compareditwith studies in vivo.Theyfound an the initialweeks of starting coumarinanticoagulant therapy. De- increase in plasma concentrations of coumarin anticoagulants: layedstabilisation is associatedwith having aCYP2C9polymor- 1.32 fold for racemic warfarin;1.28 fold for racemic acenocou- phism.The CYP4F2 genetic variant alters the required warfarin marol; 1.11 fold for racemic phenprocoumon. ForS-warfarin the dose, datafor acenocoumarol and phenprocoumon arelacking. increase is 1.58-fold; for S-phenprocoumon it was1.13 fold. Race-based differences in warfarin maintenancedose seems In patients treatedwith acenocoumarol,the risk of overanti- mainlydependentonthe linked VKORC1 variants. Although coagulation when using CYP2C9 metabolised NSAIDswas speculative, onecan expectthe same patternfor acenocoumarol modifiedbyallelic variantsofCYP2C9, especially CYP2C9*3. and phenprocoumon. This wasnot observedwith phenprocoumon. Possibly there were From aclinical perspective, phenprocoumon seemsprefer- notenough patients included in this study for afirmconclusion. able for therapeutic anticoagulation in the absence of pharmaco- Patients whohaveavariant CYP2C9 enzyme and useaceno- genetic testing.Phenprocoumon therapyshows the least risk of coumarol are at greater risk of INRvalues above 4.9 when con- delayeddestabilisation compared to warfarin or acenocoumarol. comitant NSAIDs are used (60). Analysis of the variant groups TheeffectsofCYP2C9polymorphismsonthe pharmacokinetics for NSAID usersshows that thereisnodifference within the and anticoagulant response areleastpronounced in the caseof wild-typesubgroups, because these patients have the fullcapac- phenprocoumon. Patients on phenprocoumonhaveINR values ity for eliminating drugs.Anincrease in the measuredINRsis moreoften in the therapeutic windowand require less control seen only in the subgroup of patients with CYP2C9 variantstype than patients on acenocoumarol. The drug-drug pharmacoki- *2 or *3 whoalso use NSAIDs.The capacity of the CYP2C9 netic interactions with NSAIDsare the leastpronounced for variant polymorphisms to hydroxylate substrate drugs is dimin- phenprocoumon. ished,and the administration of twosubstrate drugs mayhave Pharmacogenetic testing mayadd to the safety of coumarin pharmacokinetic effects on the elimination on one or both drugs. anticoagulant therapy. Pharmacoeconomic evaluations of phar- Masche et al. investigated the effect of the co-administration macogenetic testing suggest this is cost-effectivefor acenocou- of the NSAID lornoxicam on S- and R- phenprocoumon and marol and warfarin (68,69).

References 1. Cranenburg ECM, SchurgersLJ, Vermeer C. Vit- 5. TanakaE.Invivoage-relatedchanges in hepatic 9. RettieAE,Tai G. The pharmocogenomics of warfa- aminK:the coagulationvitamin that became omnip- drug-oxidizingcapacityinhumans.JClin Pharm Ther rin: closing in on personalized medicine. Mol Interv otent.Thromb Haemost2007;98: 120-125. 1998;23: 247–255. 2006; 6: 223–227. 2. Vander Meer FJ,Rosendaal FR, Vandenbroucke JP 6. SotaniemiEA, ArrantoAJ, PelkonenO,etal. Age 10. Xie HG,Prasad HC,Kim RB, et al. CYP2C9 allelic et al.Assessment of ableeding risk indexintwo cohorts and cytochrome P450linkeddrugmetabolisminhu- variants: ethnicdistributionand functionalsignifi- of patients treated withoral anticoagulants.Throm mans :ananalysis of 226 subjects withequalhistopath- cance. AdvDrugDelivRev 2002; 54: 1257–1270. Haemost1996;76: 12–16. ologic conditions.Clin PharmacolTher 1997; 61: 11. Daly AK,KingBP. Pharamcogeneticsoforal anti- 3. Torn M, Bollen WL, vander Meer FJ,etal. Risks of 331–339. coagulants. Pharmacogenetics 2003; 13: 247–252. anticoagulant therapywith increasingage. Arch Intern 7. Oden A, Fahlén M.Oral anticoagulationand risk of 12. Takahashi H, Echizen H. Pharmacogenetics of war- Med 2005;165: 1527–1532. death:amedical recordlinkagestudy.BrMed J2002; farinelimination and its clinical implications.Clin 4. Zu Schwabedissen CM,Mevissen V, SchmitzFet 325:1073-1075. Pharmacokinet 2001; 40: 587–603. al. Obesity is associated withaslow response to initial 8. Kamali F. Geneticinfluences on theresponseto 13. TabriziAR, ZehnbauerBA, BoreckiIB, et al. The phenprocoumon therapywhereas CYP2C9 genotypes warfarin.CurrOpin Hematol2006;13: 357–361. frequencyand effects of cytochrome P450(CYP)2C9 arenot. Eur JClin Pharmacol2006;62: 713–720.

1056 Beinema et al.Pharmacogeneticdifferencesbetween warfarin, acenocoumaroland phenprocoumon polymorphismsinpatientsreceiving warfarin. JAm rome P450 CYP2C9. Blood Coagul 2003; monanticoagulationstatus: interactionbetween both Coll Surg 2002; 194: 267–273. 14: 761–764. genotypes affects dose requirement.Clin Pharmacol 14. Aithal GP,Day CP,KestevenPJ, et al.Association 34. Thijssen HH, VerkooijenIW, Frank HL. Thepos- Ther 2007; 81: 185-193. of polymorphisms in theCytochrome P450 CYP2C9 session of theCYP2C9*3 allele is associated with low 52. Hummers-Pradier E, HessS,Adham IM et al.De- with warfarindose requirement and risk of bleeding dose requirement of acenocoumarol. Pharmacogen- terminationofbleeding risk using genetic markers in Complications. Lancet 1999; 353: 717–719. etics2000; 10: 757-760. patientstaking phenprocoumon. Eur JClin Pharmacol 15. Margaglione M, Colaizzo D, Brancaccio V, et al. 35. Thijssen HH, Drittij MJ,VervoortLM, et al.Alter- 2003; 59: 213–219. Genetic modulationoforal anticoagulationwith warfa- ed pharmacokineticsofR-and S-acenocoumarol in a 53. Visser LE, vanSchaikRH, vanVlietM,etal. The rin. Thromb Haemost2000; 84: 775–778. subject heterozygous for CYP2C9*3. Clin Pharmacol risk of bleeding complications in patientswith cytoch- 16. Higashi MK,Veenstra DL, Kondo LM et al.Associ- Ther 2001; 70: 292–298. rome P450 CYP2C9*2 or CYP2C9*3 allelesonaceno- ationBetween CYP2C9 Genetic variants and anti- 36. HermidaJ,Zarza J, Alberca I, et al.Differentialef- coumarol or phenprocoumon. ThrombHaemost 2004; coagulation-related outcomes during warfarintherapy. fects of 2C9*3 and 2C9*2 variants of cytochrome 92: 61–66. JAmMed Assoc 2002;287: 1690–1698. P-450 CYP2C9 on sensitivity to acenocoumarol. Blood 54. Fihn SD,Gadisseur AA, PasterkampE,etal. Com- 17. CaldwellMD, Awad T, JohnsonJA, et al.CYP4F2 2002; 99: 4237–4239. parison of control and stability of oral anticoagulant genetic variantalters required warfarindose. Blood 37. Visser LE, vanVlietM,van SchaikRH, et al.The therapyusing acenocoumarol versus phenprocoumon. 2008; 111: 4106-4112. risk of overanticoagulationinpatientswith cytochrome Thromb Haemost2003; 90: 260–266. 18. Rieder MJ,Reiner AP,Gage BF,etal. Effectof P450 CYP2C9*2 or CYP2C9*3 allelesonacenocou- 55. GadisseurAP,Van der Meer FJ,Adriaansen HJ et al. VKORC1haplotypes on transcriptional regulationand marol or phenprocoumon.Pharmacogenetics2004; 14: Therapeutic quality control of oral anticoagulant ther- warfarindose. NEngl JMed 2005; 352: 2285–2293 27–33. apycomparing theshort-actingacenocoumarol and the 19. OldenburgJ,Bevans CG,FreginA,etal..Current 38. Montes R, RuizdeGaona E, Martinez-Gonzalez long-acting phenprocoumon. Br JHaematol 2002; 117: pharmacogenetic developmentsinoral anticoagulation MA, et al.The c.-1639G >Apolymorphism of the 940–946. therapy: theinfluence of variantVKORC1 and VKORC1gene is amajor determinant of theresponse 56. Kohl C, Steinkellner M. Predictionofpharmacoki- CYP2C9 alleles.Thromb Haemost 2007; 98: 570-578. to acenocoumarol in anticoagulated patients. Br JHae- netic drug-drug interactions from In vitrodata: interac- 20. Schwarz UI,RitchieMD, BradfordYetal..Genetic matol2006; 133: 183–187. tions of thenonsteroidal anti-inflammatory drug lorn- determinantsofresponsetowarfarin during initialanti- 39. BodinL,Verstuyft C, Tregouet DA,etal. Cytoch- oxicam with oral anticoagulants. Drug Metab Dispos coagulation. NEngl JMed 2008; 358: 999–1008. rome P450 2C9 (CYP2C9) and vitamin Kepoxide re- 2000; 28: 161–168. 21. WadeliusM,Chen LY,Lindh JD et al.The largest ductase (VKORC1) genotypes as determinantsoface- 57. Visser LE, vanSchaikRH, vanVlietM,etal. Al- prospectivewarfarin-treated cohortsupports genetic nocoumarol sensitivity.Blood 2005; 106: 135–140. lelic variants of cytochrome P450 2C9 modify thein- forecasting. Blood 2008; prepub online 40. SchalekampT,Brassé BP,Roijers JF,etal. teractionbetween nonsteroidal anti-inflammatory doi:10.1182/blood-2008–04–149070. VKORC1and CYP2C9 genotypes and acenocoumarol drugs and coumarin anticoagulants. Clin Pharmacol 22. Daly AK, Aithal GP.Genetic regulationofwarfarin anticoagulationstatus: interactionbetween bothgeno- Ther 2005; 77: 479–485. metabolismand response. Semin Vasc Med 2003;3: types affects overanticoagulation. Clin PharmacolTher 58. VanDijk KN,PlatAW, vanDijk AA, et al.Potential 231–238. 2006; 80: 13–22. interactionbetween acenocoumarol and diclofenac, na- 23. Carlquist JF,Horne BD,MuhlesteinJB, et al.Geno- 41. RettieAE, FarinFM, BeriNG, et al.Acase study of proxenand ibuprofen and role of CYP2C9 genotype. types of thecytochrome p450 isoform, CYP2C9, and acenocoumarol sensitivity and genotype-phenotype ThrombHaemost 2004; 91: 95–101. thevitamin Kepoxide reductase complex subunit 1 discordancyexplained by combinations of polymor- 59. Zarza J. Major bleeding during combined treatment conjointly determinestablewarfarin dose: aprospec- phisms inVKORC1and CYP2C9. Br JClin Pharmacol with indomethacinand lowdoses of acenocoumarol in tive study. JThrombThrombolysis 2006; 22: 191–197. 2006; 62: 617–620. ahomozygous patient for 2C9*3 variantofcytochrome 24. Wilms EB,Touw DJ,Conemans JM,etal. Anew 42. Ufer M. Comparative pharmacokineticsofvitamin p-450 CYP2C9.ThrombHaemost 2003; 90: 161–162. VKORC1allelic variant–p.Trp59Arg-inapatient with Kantagonists: warfarin, phenprocoumon andaceno- 60. BeinemaMJ, de Jong PH,Salden HJ,etal. Thein- partialresistance to acenocoumarol and phenprocou- coumarol. Clin Pharmacokinet 2005; 44: 1227–1246. fluence of NSAIDs on coumarin sensitivity in patients mon. JThromb Haemost 2008; 6: 1224–1226. 43. Ufer M, Svensson JO,Krausz KW,etal. Identifica- with CYP2C9 polymorphism after totalhip replace- 25. LoebsteinR,DvoskinI,HalkinH,etal. Acoding tionofcytochromes P450 2C9 and 3A4asthe major ment surgery. Mol Diagn Ther 2007; 11: 123–128. VKORC1Asp36Tyr polymorphism predisposes to catalysts of phenprocoumon hydroxylationinvitro. 61. Masche UP,Rentsch KM,von FeltenA,etal. Op- warfarinresistance. Blood 2007; 6: 2477–2480. Eur JClin Pharmacol 2004; 60: 173–182. positeeffectsoflornoxicam co-administrationon 26. Lal S, Jada SR, Xiang X, et al.Pharmacogeneticsof 44. Kamerer B, KahlichR,Ufer M, et al.Stereospecific phenprocoumon pharmacokineticsand pharmacody- target genes acrossthe warfarinpharmacological path- pharmacokinetic characterisation of phenprocoumon namics. Eur JClin Pharmacol 1999; 54: 857–864. way. ClinPharmacokinet 2006; 45: 1189-1200. metabolites, andmass-spectrometic identificationof 62. Gage BF,LeskoLL. Pharmacogenetics of warfarin: 27. Thijssen HH, FlinoisJP, BeaunePH. Cytochrome twonovel metabolitesinhuman plasma and regulatory,scientific,and clinical issues. JThromb P4502C9 is theprincipal catalyst of racemicacenocou- microsomes.Anal Bioanal Chem 2005; 383: 909–917. 2008; 25: 45–51. marol hydroxylationreactions in human livermicro- 45. Ufer M, Kammerer B, KahlichRetal. Genetic 63. Daly AK, KingBP. Pharmacogeneticsoforal anti- somes. Drug Metab Dispos 2000; 28: 1284–1290. polymorphisms of cytochrome P450 2C9 causing re- coagulants. Pharmacogenetics2003; 13: 247–252. 28. Thijssen HH, Ritzen B. Acenocoumarol phar- duced phenprocoumon (S)-7-hydroxylationinvitro 64. Schwarz UI,Stein CM.Genetic determinantsof macokineticsinrelationtocytochrome P450 2C9 ge- and in vivo.Xenobiotica2004; 34: 847–859. dose and clinical outcomes in patientsreceiving oral notype. Clin Pharmacol Ther 2003; 74: 61–68. 46. He M, KorzekwaKR, Jones JP,etal. Structural anticoagulants. Clin Pharmacol Ther 2006; 80: 7–12. 29. Morin S, BodinL,Loriot MA, et al.Pharmacogen- forms of phenprocoumon and warfarinthat are meta- 65. Shurin SB,Nabel EG.Pharmacogenomics– eticsofacenocoumarol pharmacodynamics. ClinPhar- bolized at theactivesite of CYP2C9. Arch Biochem Ready for primetime?NEngl JMed 2008; 358: macol Ther 2004; 75: 403-414. Biophys 1999; 372: 16–28. 1061–1063. 30. Takahashi H, Wilkinson GR, Padrini R, et al. 47. Ufer M. Effects of CYP2C9 polymorphisms on 66. Voora D, EbyC,Linder MW,etal. Prospectivedos- CYP2C9 and oral anticoagulationtherapywith aceno- phenprocoumon anticoaglationstatus(letter). Clin ingofwarfarin based on cytochrome P-450 2C9 geno- coumarol and warfarin: similaritiesyet differences. PharmacolTher 2005; 77: 335. type. ThrombHaemost 2005; 93: 700–705. Clin Pharmacol Ther 2004; 75: 376–380. 48. Toon S, Heimark LD,Trager WF,etal. Metabolic 67. YinT,MiyataT.Warfarin dose and thephar- 31. Thijssen HH, Drittij MJ,VervoortLM, et al.Alter- fate of phenprocoumon in humans. JPharmSci 1985; macogenomicsofCYP2C9 and VKORC1–rationale ed pharmacokineticsofR-and S-acenocoumarol in a 74: 1037–1040. and perspectives. ThrombRes 2007; 120: 1–10. subject heterozygous for CYP2C9*3. Clin Pharmacol 49. Kirchheiner J, Ufer M, WalterEC, et al.. Effects of 68. Vegter S, Boersma C, Rozenbaum M, et al.Phar- Ther 2001; 70: 292–298. CYP2C9 polymorphisms on thepharmacokineticsof macoeconomic evaluations of pharmacogenetic and 32. Verstuyft C, MorinS,RobertA,etal. Earlyaceno- R- andS-phenprocoumon in healthyvolunteers.Phar- genomic screeningprogrammes: asystematic review coumarol overanticoagulationamong cytochrome macogenetics2004; 14: 19–26. on content and advherance to guidelines. Pharmacoe- P450 2C9 poor metabolizers.Pharmacogenetics2001; 50. He M, KorzekwaKR, Jones JP,etal. Structural conomics2008; 26: 569–587. 11: 735–737. forms of phenprocoumon and warfarinthat are meta- 69. Yo uJHS, Chan FWh, Wong RSM, et al. Thepoten- 33. Andre-KerneisE,Leroy-Matheron C, Gouault- bolized at theactivesite of CYP2C9. Arch Biochem tialclinical and economic outcomesofpharmacogen- Heilmann M. Earlyoveranticoagulationwith aceno- Biophys 1999; 372: 16–28. etics-orientated management of warfarin therapy–a coumarol due to agenetic polymorphism of cytoch- 51. SchalekampT,Brassé BP,Roijers JF,etal. decision analysis.ThrombHaemost 2004; 92: 590–597. VKORC1and CYP2C9 genotypes and phenprocou-

1057