Review Article Pharmacogenetics of Oral Antidiabetic DrugsPharmacogenetics of Oral Antidiabetic Drugs MatthijsL.Becker, 1,2 EwanR.Pearson, 3 andIvanTká I4,5 Department of Epidemiology,

Hindawi Publishing CorporationInternational Journal of EndocrinologyVolume 2013 Article ID 686315 10 pageshttpdxdoiorg1011552013686315

Review ArticlePharmacogenetics of Oral Antidiabetic Drugs

Matthijs L Becker12 Ewan R Pearson3 and Ivan TkaacuteI45

1 Department of Epidemiology Erasmus MC 3015 CE Rotterdam The Netherlands2 Pharmacy Foundation of Haarlem Hospitals 2035 RC Haarlem The Netherlands3Medical Research Institute University of Dundee Dundee DD1 9SY UK4Department of Internal Medicine 4 Faculty of Medicine P J Safarik University 041 80 Kosice Slovakia5 Department of Internal Medicine 4 L Pasteur University Hospital Rastislavova 43 041 90 Kosice Slovakia

Correspondence should be addressed to Ivan Tkac ivantkacupjssk

Received 15 February 2013 Revised 28 October 2013 Accepted 28 October 2013

Academic Editor Khalid Hussain

Copyright copy 2013 Matthijs L Becker et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Oral antidiabetic drugs (OADs) are used for more than a half-century in the treatment of type 2 diabetes Only in the last five yearsintensive research has been conducted in the pharmacogenetics of these drugs basedmainly on the retrospective register studies butonly a handful of associations detected in these studies were replicatedThe gene variants inCYP2C9ABCC8KCNJ11 and TCF7L2were associated with the effect of sulfonylureas CYP2C9 encodes sulfonylurea metabolizing cytochrome P450 isoenzyme 2C9ABCC8 and KCNJ11 genes encode proteins constituting ATP-sensitive K+ channel which is a therapeutic target for sulfonylureasandTCF7L2 is a genewith the strongest associationwith type 2 diabetes SLC22A1 SLC47A1 andATM gene variantswere repeatedlyassociated with the response to metformin SLC22A1 and SLC47A1 encodemetformin transporters OCT1 andMATE1 respectivelyThe function of a gene variant nearATM gene identified by a genome-wide association study is not elucidated so farThe first variantassociated with the response to gliptins is a polymorphism in the proximity of CTRB12 gene which encodes chymotrypsinogenEstablishment of diabetes pharmacogenetics consortia and reduction in costs of genomics might lead to some significant clinicalbreakthroughs in this field in a near future

1 IntroductionType 2 diabetes (T2D) affects more than 5 of populationof the developed countries and its prevalence increasesworldwide [1] The majority of patients with T2D starttheir treatment with oral antidiabetic drugs that influencetwo basic pathogenetic mechanisms in the developmentof T2Dmdashinsulin resistance or defects of insulin secretionBeside glucose stimulated insulin secretion more attentionin the recent years is devoted to incretin augmented insulinsecretion and new drugs were introduced in clinical practicewhich enhance the actions of incretins

There is a considerable variability in the effect of antidi-abetic drugs This variability is caused by nonbiological andbiological factors Among nonbiological factors psycholog-ical and social factors play an important role These includecompliance to medication variable access to health care andphysician prescribing practices which are dependent on bothinternational and national guidelines

Biological factors are related to pharmacokinetics andpharmacodynamics of drugs Biological factors might benongenetic such as the influence of intestinal hepatic andrenal functions or drug interactions predominantly on phar-macokinetics of drugs Pharmacogenetics focuses on thestudy of genetic factors which influence the effect and sideeffects of different drugs with the final aim of personalizingthe treatment of T2D While clear implications for clinicalpractice based on pharmacogenetic knowledge exist onlyfor some forms of monogenic diabetes [2] this review willfocus on the more recent work undertaken to elucidatepharmacogenetic mechanism in common type 2 diabetes

2 Methodological Aspects of PharmacogeneticStudies in T2D

The optimal design of pharmacogenetic studies requiresattention to some keymethodological issues Whilst the ideal

2 International Journal of Endocrinology

study is a prospective genotype-blind study design with ade-quate statistical power these are very costly time-consumingand often require participation ofmultiple centersThereforethe majority of studies published used retrospective dataretrieved from registers and databases In the retrospectivestudies it is possible to achieve reasonable sample sizes forpharmacogenetic studies but it can be difficult to adjustfor all confounding variables as these may not have beenmeasured However as most confounding should not begenotype dependent this should not be a major issue

With respect to the outcomes used in pharmacogeneticstudies there are two main types of endpoints Pathophys-iological endpoints reflect the effect of gene variants oninsulin resistance or insulin secretion Such studies may yieldnovel knowledge on the role of different gene variants in thepathogenesis of T2D but their clinical applicability mightbe limited Thus from the point of view of possible clinicalimplications endpoints such as reduction of HbA1c fastingplasma glucose and postprandial plasma glucose are of thehighest importance Among them both an achievement oftherapeutic target of HbA1clt7 defined in guidelines anda reduction in HbA1c seem to be the most appropriateendpoints for pharmacogenetic studies in T2D [3]

Different factors may confound the relationship betweenthe studied gene variants and endpoints When using achi-evement ofHbA1clt7or reduction inHbA1c as an endpointthe strongest confounder is the baseline level of HbA1c asit was shown in both the meta-analysis of clinical trials andalso in the individual pharmacogenetic studies [4 5] Thatmeanmdashif there is an imbalance in baseline HbA1c among thegenotypesmdashone could expect higher response in the patientswith higher baselineHbA1c values On the other hand higherbaseline HbA1c might also reflect the effect of the genevariant on previous diabetes control Thus it is correct toshow rough (unadjusted) data alongwith themodels adjustedfor confounding variables It is also important to take intoaccount whether the drug of interest is being used earlyon in the disease process or as add-on therapy later in thedisease where there is less likely to be a large therapeuticeffect Covariates that are usually taken into account such asage gender BMI and diabetes duration are rarely significantpredictors of the drug effect in pharmacogenetic studies Ifthe drug dose during the study is not constant it is reasonableto adjust for drug dosage Since the majority of the oralantidiabetic drugs are eliminated by the kidney adjustmentfor ameasure of renal function such asGFRor creatinine levelis also frequently needed Furthermore if retrievable fromdatabases medication adherence and period of drug use areuseful covariates

We selected studies discussed in this review taking intoaccount the above mentioned methodological aspects ofpharmacogenetic studies in patientswithT2DWeperformeda search in the MEDLINE and Web of Knowledge databasesusing ldquopharmacogenetics OR pharmacogenomicsrdquo keywordin combinations with the names of groups of antidiabeticdrugs or generic names of individual antidiabetic drugsIn the present review we will discuss the mechanisms ofassociations between gene variants and the drug effect in thehealthy subjects in cell lines and in the patients with T2D

The studies which examined the effect of gene variants on theindices of glucose control will be presented only when at least100 patients with T2D are included in the study

3 Pharmacogenetics of Insulin Secretagogues

Insulin secretagogues stimulate secretion of insulin from thepancreatic 120573-cells Sulfonylureas have been used for morethan a half-century while nonsulfonylurea secretagoguesmeglitinides (glinides) have been used for approximately thelast 15 years in the treatment of T2D The most commonside effects of both groups of drugs are hypoglycemia andweight gain While previously sulfonylureas were consideredas first-line treatment in less obese patients with T2D morerecent guidelines recommend both groups of drugs mainlyin combination therapy with metformin as a second-linetreatment [6]

31 Gene Variants Related to Pharmacokinetics ofInsulin Secretagogues

311 Cytochrome P450 Isoenzyme 2C9 Sulfonylureas aremetabolized in the liver primarily by cytochrome P450isoenzyme 2C9 encoded by 11986211988411987521198629 [7ndash10] The major alleleof this gene is 11986211988411987521198629lowast1 Two common nonsynonymousvariants Arg144Cys (11986211988411987521198629lowast2) and Ile359Leu (11986211988411987521198629lowast3)were identified [7] Studies on healthy subjects showed thatthe variant alleles are associated with increased plasmaconcentration and decreased clearance of sulfonylureas afteroral administration [7 10]

The first robust study to confirm association of11986211988411987521198629 variants with therapeutic response to sulfonylureaswas Genetics of Diabetes Audit and Research Tayside Study(GoDARTS) which included 1073 patients treated with sulfo-nylureas The 6 of the population who were carriers oftwo loss-of-function alleles (lowast2lowast2 or lowast2lowast3 or lowast3lowast3) had a05 greater reduction in HbA1c compared with wild-typehomozygotes (119875 = 0003) and were 34 times more likely toachieve on-treatment HbA1clt7 (119875 = 0009) [5]

Further studies in patients with type 2 diabetes used indi-rect endpoints of sulfonylurea effect such as prescribed doseof sulfonylurea Becker et al analyzed 475 elderly patientsincluded in the Rotterdam study who started sulfonylureatreatment In the largest subgroup of patients treated withtolbutamide (119899 = 172) there was a significant differencein daily prescribed dose observed between genotypes Whilethe prescribed dose of tolbutamide increased in the carriersof lowast1lowast1 lowast1lowast2 or lowast2lowast2 genotypes there was very littletolbutamide dose increase in patients with lowast1lowast3 or lowast2lowast3genotypes (119875 = 0009) [11] In a similar performed analysisfrom a different Dutch group which analyzed data from207 incident sulfonylurea users a trend towards lower stableglimepiride dose (119875 = 007) was observed in the carriers ofthe 11986211988411987521198629lowast3 allele [12]

Studies in healthy volunteers showed that variants inCYP2C9 influence also pharmacokinetics of nonsulfonylureasecretagogue nateglinide [13] while variant in CYP2C8 isassociated with pharmacokinetics of repaglinide [14] Phar-macokinetics of both of these glinide drugs is also influenced

International Journal of Endocrinology 3

by variant 521TgtC in SLCO1B1 the gene encoding organicanionic transporter B1 (OATPB1) [15]

32 Gene Variants Related Pharmacodynamics of InsulinSecretagogues The ATP-sensitive potassium (KATP) channelplays crucial role in glucose stimulated insulin secretion Inphysiologic condition ATP produced by glucose oxidationin mitochondria leads to closure of KATP channel withsubsequent depolarization of 120573-cell membrane increasedinflux of calcium ions and subsequent release of presyn-thesized insulin from the 120573-cell Insulin secretagogues (bothsulfonylureas and glinides) act by inducing KATP channelclosure by binding on its constituting proteins The innerpore of the KATP channel is constituted by four molecules ofpotassium inward rectifier 62 (Kir62) while the outside partis created by four molecules of sulfonylurea receptor 1 (SUR1)[16]

Two sulfonylurea binding sites were identified on theKATP channel The A-site is exclusively part of the SUR1 sub-unit whereas the B-site resides both on SUR1 and Kir62 sub-units [17] Sulfonylurea derivatives chlorpropamide tolbu-tamide and gliclazide as well as nateglinide and mitiglinidebind exclusively to the A-site Repaglinide binds exclusivelyto the B-site while glibenclamide glipizide and glimepirideare AB-site binding drugs [18 19]

Mutations in genes encoding KATP channel proteinsmdashKCNJ11 (encoding Kir62) and ABCC8 (encoding SUR1)mdashlead to neonatal diabetes mellitus Breakthrough pharmaco-genetic studies showed that sulfonylureas are able to correctthe defect caused by KCNJ11 and ABCC8mutations resultingin patients (believed previously to have type 1 diabetes) beingable to transition from long-term insulin therapy to sulfony-lurea treatment [20 21] Thus common variants in KCNJ11and ABCC8 were logical selection also for pharmacogeneticstudies in type 2 diabetes

321 Kir62 and SUR1 Nonsynonymous common variantsE23K in KCNJ11 and S1369A in ABCC8 were mostly studiedin pharmacogenetic studies with insulin secretagoguesThesetwo variants are in strong linkage disequilibrium so thatany association signal from these two polymorphisms isgenetically indistinguishable [22]

The most robust study which showed an association ofABCC8 S1369A polymorphism with glycemic control wasdone in Chinese population The study included 661 patientswhowere genotyped for 25 single nucleotide polymorphismsKCNJ11 rs5210 (different from E23K) and ABCC8 S1369Apolymorphisms were significantly associated with decreasein FPG Association analysis of ABCC8 S1369A with sulfony-lurea responsewas replicated in an independent cohort of 607patients In the combined analysis of both cohorts subjectswith a ABCC8 AA genotype had a significantly greaterdecrease in FPG (119875 lt 0001) and a 2-hour plasma glucose(119875 lt 0003) but only a borderline decrease in HbA1c (17versus 14 119875 = 006) in comparison with patients with SSgenotype [23] Lack of significance in HbA1c level reductioncan be explained by relatively short 8-week duration of studyduring which full effect of treatment on reduction in HbA1clevel was not observed Similar association was found by

another Chinese group which evaluated glycemic responseto 8-week gliclazide treatment In a group of 115 patientswith T2D a greater reduction in HbA1c in response to 8-week gliclazide treatment in the carriers of the A-allele (SAand AA) compared to homozygous carriers of the S-allele(160 plusmn 139 versus 076 plusmn 170 119875 = 0044) was observedafter adjustment for baseline HbA1c [24]

In Caucasian population the association betweenKCNJ11E23K and sulfonylurea efficacy was observed in the studyof Javorsky et al which included 101 patients treated withsulfonylurea after metformin monotherapy failure In thedominant model the carriers of the K-allele had higherreduction in HbA1c after 6-month therapy in comparisonwith EE homozygotes (104 plusmn 010 versus 079 plusmn 012 119875 =0036) [25]

Previously two studies did not find such association inCaucasian population In the United Kingdom ProspectiveDiabetes Study (UKPDS) population a group of 363 patientsprimarily assigned to sulfonylurea treatment was analyzedNo significant relationship was found between two KCNJ11polymorphisms (E23K and L270V) and the response tosulfonylurea The evaluated outcome was based on two mea-surements of FPG within the first year of treatment but noton HbA1c measurements [26] Since titration of sulfonylureadose was carried out in the UKPDS this may have con-founded the response phenotype Another study performedin a group of 525 Italian patients showed that the carriers ofK-allele had significantly higher probability of secondary sul-fonylurea failure In that study secondary sulfonylurea failurewas defined as not achieving FPGlt300mgdL (167mmolL)on-treatment with combination of sulfonylurea as the first-choice drug andmetformin as an add-on drug [27]Thus thisstudy reported the failure of the combination of sulfonylureawith metformin rather than the failure of sulfonylureatreatment itself

Another study in Chinese population examined the asso-ciation between KCNJ11 E23K and therapeutic response torepaglinide He et al found in a group of 100 patients treatedfor 24 weeks with repaglinide that the decrease in HbA1c washigher in patients with EK and KK genotypes than in EEhomozygotes (EE 152 plusmn 103 EK 233 plusmn 153 and KK265 plusmn 173 119875 = 0022) [28] The authors did not adjustthe decrease in HbA1c for the baseline levels thus it is notpossible to exclude that this effect was driven by an effect ofthe E23K variant on baseline HbA1c

Clinical studies showed higher efficacy of sulfonylureatreatment in risk allele carriers suggesting that in line withthe studies in neonatal diabetes sulfonylureas and possiblyalso glinides are able to correct the defect in insulin secretionby binding to the proteins of the KATP channel resulting in itsclosure Studies in cell lines transfected by cloned channelswith KCNJ11 K23-ABCC8 A1369 haplotype shed some lighton observed differences among different sulfonylureas andglinides to affect insulin secretion in carriers of risk haplo-type K23A1369 carriers were more sensitive to inhibitionof KATP channel by gliclazide and mitiglinide in comparisonwith E23S1369 haplotype carriers In contrast E23S1369carriers were more sensitive to glimepiride chlorpropamideand tolbutamide whilst there was no difference in KATP


channel sensitivity between the two haplotypes for gliben-clamide glipizide repaglinide and nateglinide [18 19]

322 Transcription Factor 7-Like 2 Among more than 50gene variants associated with T2D the variants in tran-scription factor 7-like 2 (TCF7L2) gene are the strongestpredictors of increased risk of developing type 2 diabetes a40 increased risk per allele [29 30] TCF7L2 is a nuclearfactorwhich binds120573-cateninmediatesWnt-signaling relatesto normal development of pancreas during embryogenesisand affects secretion of glucagon-like peptide 1 (GLP-1) by L-cells in the small intestine [31] Several studies observed thatcarriers of the risk T-allele of TCF7L2 rs7903146 polymor-phism have reduced insulin secretion [32 33] Mechanismswhereby altered TCF7L2 production andor function maycontribute to the development of type 2 diabetes are notfully understood but likely include a decrease in 120573-cell massimpaired insulin processing or release and impaired incretinsignaling in 120573-cells [34 35]

GoDARTSwas the first pharmacogenetic study to addressthe relationship between the TCF7L2 rs12255372 (GT) andrs7903146 (CT) gene variants and response to sulphony-lurea therapy in type 2 diabetic patients In that study in901 Scottish patients with type 2 diabetes subjects withthe rs12255372TT genotype had approximately two timeshigher probability for early sulphonylurea treatment failure(HbA1cgt7 within the period of 3ndash12 months after startingsulphonylurea therapy) compared to those with the CC geno-type In a complementary approach a linear regressionmodelwas used with minimal HbA1c during treatment within ayear after sulfonylurea initiation as the dependent variableThe predicted HbA1c on-treatment was for rs1225372 GGgenotype 70 while for TT genotype it was 733 Similarresults were observed for rs7903146 [36]

Results of this study were replicated by two study groupsJavorsky et al found in a group of 101 Slovakian patients thatthe reduction in HbA1c after six months of sulphonylureatherapy (ΔHbA1c) was significantly lower in the CT+TTgenotype group in comparison with the CC homozygotesfor rs7903146 genotype The absolute difference in ΔHbA1cbetween the two groups was 035 (119875 = 0006) [37] thatis very similar to the value observed in GoDARTS studyIn another study in a German population Holstein et alincluded 179 patients treated with sulfonylureas and analyzedsulfonylurea treatment failure after 6 months accordingto rs7903146 genotypes They found more than twice theprobability of sulfonylurea failure in TT homozygotes incomparison with CC homozygotes (OR 209 95 CI 102ndash427 119875 = 0043) In this study the adjustment for the baselineHbA1c values was not performed [38]

4 Pharmacogenetics of Metformin

Metformin has been used for the treatment of type 2 dia-betes mellitus since 1959 and is still the cornerstone in thetreatment of this disease At physiological pH metforminis positively charged and therefore very hydrophilic givingit some interesting pharmacokinetic properties [39] Firstmetformin is not metabolized in the body but very efficiently

excreted in the urineTherefore the glucose lowering effect ofmetformin is not affected by genetic variation inmetabolizingenzymes Second metformin cannot diffuse through mem-branes passively but it is dependent on drug transporters forthe absorption distribution and elimination of metformin[39] The initial pharmacogenetic research focused on therole ofmetformin transporterswhile themost extensive studythus far is the genome-wide association study (GWAS)

41 Genome-Wide Association Study In GWAS Zhou et algenotyped more than 700000 polymorphisms in 1024 met-formin users from the GoDARTs study [40] They identified14 polymorphisms in a locus containing the ATM genethat were associated with the ability to reach the treatmentgoal of an HbA1clt7 The strongest association was withthe rs11212617 polymorphism Participants had a 164 timeshigher change of reaching the treatment goal for each minorallele with a 119875 value of 19 times 10minus7 The authors subse-quently genotyped this polymorphism in two independentpopulations (another GoDARTS population and the UKPDSpopulation) for replication And in both populations asignificant association was found with treatment responseThe combined effect was 135 times higher change of reachingthe treatment goal (119875 = 29 times 10minus9) for each minor alleleThe secondary analysis was the association between thispolymorphism and the reduction in HbA1c and a 011larger reduction in HbA1c per minor allele (119875 = 66 times 10minus7)was found in the three cohorts combined In the replicationstudy by van Leeuwen et al the rs11212617 polymorphismwas studied in three independent populations the DiabetesCare System (DCS)West-Friesland (119899 = 929) the RotterdamStudy (119899 = 182) and the CARDS Trial (119899 = 254) Theyused the same endpoint as was used in the initial GWAS byZhou et al In the three populations the combined odds ratiowas 124 with a 119875 value of 0016 For the secondary endpointthe reduction in HbA1c per minor allele no significantassociation was found [41]

The rs11212617 polymorphism has also been genotypedin the at-risk population of the aforementioned DPP trialHowever Florez et al describe that no association was foundbetween this polymorphism and the incidence of diabetes inthe participants randomized to metformin therapy [42]

42 Genes Related to Pharmacokinetics of Metformin

421 Organic Cation Transporters Shu et al were the firstto study the effect of genetic variation in the SLC22A1 geneencoding the OCT1 transporter and the glucose loweringeffect of metformin both in animal model and in healthy vol-unteers They identified four polymorphisms in the SLC22A1gene coding for a change in amino-acid sequence (R61CG401S 420del or G465R) and studied the association withthe glucose lowering effect In the subjects carrying one ormore variant alleles themetformin plasma levels were higherand the glucose lowering effect during a glucose tolerance testwas impaired [43 44] InCaucasians R61C and 420del are themost important genetic variants because the variant allelesoccur frequently and decrease transporter activity [45] ForR61C it has been shown that this variant strongly reduces


OCT1 protein expression [46] Tzvetkov et al identified thatOCT1 beside the liver is also expressed on the apical side oftubules and that healthy volunteers carrying 420del alleles inOCT1 had an increased renal metformin clearance due to adecrease in reabsorbance [47]

Strikingly in the largest study in subjects with T2Dperformed by Zhou et al no association was found [48] Inthis study 1531 type 2 diabetes mellitus patients participatingin the GoDARTS were included Neither the R61C nor the420del variant had a significant association with variousendpoints including the maximum HbA1c reduction in the18 months after start of metformin therapy and the ability toreach a treatment target of HbA1clt7

Christensen et al studied the effect of eleven polymo-rphisms in various transporters including OCT1 on thethrough plasma levels and the glucose lowering effect in 151diabetes mellitus patients in whom metformin was addedto the previous insulin therapy Patients with one or twovariant alleles of 420del had lower trough plasma levels Inthis study two other polymorphisms G401S and rs461473 inan intronic region not coding for an amino-acid change weresignificantly associated with the glucose lowering effect afterstart with metformin therapy [49]

Two other polymorphisms that have been described inthe OCT1 transporter are the rs622342 and the M408V poly-morphisms Becker et al screened the SLC22A1 gene usingtagging polymorphisms and found that the rs622342 poly-morphism was significantly associated with the glucoselowering effect in 102 incident metformin users [50] Thisresult has not been replicated thus farTheM408V variant hasbeen associated with gastrointestinal side effects by Tarasovaet al in 246metformin users [51]However Shu et al reportedthat this genetic variation is not associated with reduction inmetformin transporter activity [43]

In Asians the frequencies of R61C and 420del variantsare low and no polymorphisms in the SLC22A1 gene havebeen described that occur frequently and have a reducedtransporter activity [45] Several studies have suggested thatgenetic variation in OCT2 has a more important role inAsians than OCT1 OCT2 encoded by the SLC22A2 gene isexpressed in the basolateral membrane of the renal epithelialcells In three studies in Asian populations an associationbetween A270S and renal metformin clearance or plasmalactate concentrations has been described [52ndash54] An asso-ciation between A270S and renal clearance was also found ina study performed in the USA [55]

422 Multidrug and Toxin Extrusion Transporters Metfo-rmin is also a substrate for the multidrug and toxin extrusion1 (MATE1) transporter This transporter encoded by theSLC47A1 gene is strongly expressed in the brush bordermembrane of the kidney and the bile canaliculi in theliver It is believed to facilitate the excretion of compoundssuch as metformin in the urine and bile Becker et alwere the first to identify that the tagging polymorphismrs2289669 (GA) was associated with the HbA1c loweringeffect in incident metformin users [56] They subsequentlydescribed an interaction with the rs622342 polymorphismin the SLC22A1 gene identified earlier by this group [57]

Jablonski et al studied the effect of genetic variation in 40candidate genes in 2994 participants of the Diabetes Preven-tion Program (DPP) study that were at risk for developingdiabetes mellitus [58] They randomized the participants toplacebo metformin or lifestyle intervention and observedthe finding that the rs8065082 polymorphism is associatedwith metformin response in the at-risk population treatedwith metformin Since this polymorphism is in tight linkagedisequilibrium with rs2289669 (1199032 asymp 08) and the effect wasconsistent in both studies this is considered as a replicationof the previous findings of Becker et al For the otherpolymorphisms no significant associations were found aftercorrection for multiple testing

Also Tkac et al identified in 148 incident metforminusers that the rs2289669 polymorphism is associated withHbA1c reduction The homozygous carriers of the minor Aallele had twofold reduction in HbA1c in comparison withthe G-allele carriers (110 plusmn 018 versus 055 plusmn 009)[59] Stocker et al studied the gminus66TgtC polymorphismin the promoter region of the SLC47A1 gene and foundthat in healthy volunteers the variant allele resulted inlower glucose levels and in 145 incident metformin usersit resulted in an increased HbA1c lowering effect [60] In asubsequent study they identified that this polymorphismwasnot associated with metformin disposition and that the effectof the gminus66TgtC polymorphism was larger in patients withnormal functioningOCT1 alleles [60]Most likely the variantallele associatedwith reduced expression of the SLC47A1 generesults in higher hepatic plasma levels due to reduced effluxtransporter activity In patients with a reduced OCT1 influxtransporter activity the hepatic plasma levels will be lowermasking the effect of the SLC47A1 efflux polymorphism

Also the MATE2 transporter encoded by the SLC47A2gene is expressed at the brush border membrane of thekidneyThis transporter has two functional isoformsMATE2and MATE2-K The gminus130GgtA polymorphism is situated inthe basal promoter region of MATE2-K and results in anincrease in promoter activity In the study by Choi et al thispolymorphismwas associatedwith theHbA1c lowering effectin 248 incident metformin users [61] In the study by Stockeret al the renal clearance of metformin was diminished inhealthy volunteers carrying a variant allele and the glucoselowering effect increased [60] The effect of the gminus130GgtApolymorphism was influenced by the previously describedgminus66TgtC polymorphism in the SLC47A1 gene BothMATE1and MATE2 are coexpressed at the apical membrane of thekidney possibly explaining their mutual effect [60]

5 Pharmacogenetics of Thiazolidinediones

The group of thiazolidinediones including troglitazone pio-glitazone and rosiglitazonewere introduced in the late 1990sThese drugs are agonists of the peroxisome proliferator-acti-vated receptor (PPAR-120574) Activation of this receptor regulatesthe transcription of hundreds of genes involved in lipid andglucose metabolism Effects associated with PPAR-120574 acti-vation include decreased insulin resistance decreased leptinlevels and increased adiponectin levels Troglitazone hasbeen withdrawn from the market worldwide due to liver


injury and rosiglitazone has beenwithdrawn from themarketin Europe due to an increased risk of cardiovascular eventsand put under restrictions in the USA

51 Genes Related to Pharmacokinetics of Thiazolidinediones

511 Cytochrome P450 2C8 Both pioglitazone and rosigli-tazone are metabolized by the cytochrome P450 2C8 isoen-zyme The CYP2C8lowast3 and lowast11 polymorphisms coding for areduced functioning CYP2C8 enzyme have been studied inrelation to the pharmacokinetics of these drugsThree studiesassociated these polymorphisms with increased pioglitazoneand rosiglitazone plasma levels [62ndash64] and one study did notfind such as association [65] All four studies were performedin healthy volunteers Only in the study including patientswith type 2 diabetes treated by rosiglitazone from the SouthDanish Diabetes Study cohort (119899 = 187) the associationbetween 11986211988411987521198628lowast3 variant and change HbA1c was studiedThe authors found reduced therapeutic response and a lowerrisk for developing edemas in 11986211988411987521198628lowast3 carriers In thisstudy rosiglitazone was added to the previous insulin treat-ment and it was allowed to change the insulin dose during thetreatment which could have influenced the observed studyresults [65]

52 Genes Related to Pharmacodynamics ofThiazolidinediones

521 PPAR-120574 The obvious first choice in the pharmacoge-netic research after thiazolidinediones was the PPAR-120574 geneOne polymorphism in this gene the P12A polymorphism(rs1801282) was identified in 1997 and studied extensively inthe relation to the incidence of type 2 diabetes mellitus [66]In ameta-analysis published in 2010 66 studieswere includedand the risk of developing type 2 diabetes mellitus was 14 per-cent lower for each minor allele of this polymorphism [67]Three studies examined the association of this polymorphismwith the glucose lowering response of thiazolidinediones Inthe study by Hsieh et al in 250 diabetes patients and thestudy by Kang et al in 198 diabetes patients the alanine (A)allele was associated with a stronger reduction in HbA1c andfasting plasma glucose levels in pioglitazone and rosiglitazoneusers respectively [68 69] On the contrary other studiesdid not find an association between this polymorphism andthiazolidinedione therapy In the largest study thus far byFlorez et al no association was found with insulin sensitivityin 340 participants of the DPP study using troglitazone [70]Bluher et al (131 pioglitazone users) and Namvaran et al(101 pioglitazone users) did not find an association betweenthis polymorphism and HbA1c reduction insulin sensitivityand therapeutic response respectively [71 72] Whetherthere is a true association between the P12A polymorphismand response to thiazolidinedione therapy remains to bedetermined

522 PGC-1120572 The proliferator-activated receptor-120574 coacti-vator-1120572 (PGC-1120572) is a regulator of PPAR-120574 and it is associ-ated with the risk of developing diabetes Both the T394T andthe G482S variants in this gene were related to response to

rosiglitazone therapy in the study byKang et al in 241 patients[69] But in the study byHsieh et al no associationwas foundwith theG482S polymorphism in 250 pioglitazone users [68]

523 Adiponectin The binding of thiazolidinediones toPPAR-120574 regulates numerous other genes and some ofthese genes have been subject of pharmacogenetic researchAdiponectin is the best studied gene that is regulated byPPAR-120574 Three polymorphisms in this gene minus 11377CgtG+45TgtG and +276GgtT were previously associated withobesity development and the incidence of type 2 diabetesmellitus and the effect of these polymorphisms on thiazo-lidinedione response was studied In the study by Eun et althe +45TgtG and +276GgtT polymorphisms were associatedwith the change in HbA1c levels in 166 rosiglitazone users[73]The association with the minus11377CgtGpolymorphismwasfound in the study by Li et al They found an association withthe change inHbA1c level after start of pioglitazone treatmentin 113 users but no association was found with the +45TgtGpolymorphism [74] In the study by Namvaran et al noassociation was found between the +45TgtG polymorphismand response to pioglitazone in 101 users either [75]

6 Pharmacogenetics of Incretin Mimetics

The incretin effect is mediated by stimulation of insulinsecretion from pancreatic 120573-cells after ingestion of glucose ormixed meals by intestinal hormones glucagon-like peptide-1(GLP-1) and glucose-dependent insulinotropic peptide (GIP)[76] In a pilot study including 88 healthy individuals itwas shown that two GLP-1 receptor (GLP1R) variants wereassociated with altered 120573-cell sensitivity to GLP-1 infusion[77] Variation in the GIP receptor (GIPR) locus was reportedto influence the responses of the glucose and insulin to an oralglucose challenge in a meta-analysis of nine GWAS whichincluded 15234 nondiabetic subjects [78]

Besides the gene variants encoding GLP-1 and GIP recep-tors several T2D related gene variants were reported to beassociated with the incretin effect Gene variants in TCF7L2[79ndash81] voltage-gated potassium channel KQT-like subfam-ily member 1 (KCNQ1) [82] and wolframin 1 (WFS1) [83]were associated with decreased incretin secretion decreasedsensitivity of GLP-1 or GIP receptors or with decreasedsuppression of glucagon secretion [84]

Since endogenous and exogenously applied GLP-1 is rapi-dly degraded by dipeptidylpeptidase-4 two pharmacologicapproaches were developed to bypass this limitation andto enhance the incretin effect GLP-1 receptor antagonistsare resistant to DPP-4 and applied subcutaneously Anotherpossibility is to inhibit DPP-4 by orally applied DPP-4inhibitors (gliptins) [76]

61 Pharmacogenetics of Gliptins

611 Chymotrypsinogen Until recently no pharmacogeneticstudy with incretin mimetics was published In their recentlypublished study lsquot Haart et al used Metabochip (includingtests for approximately 186000 SNPs that previously wereassociated with metabolic or cardiovascular disease) and


showed that three genetic loci (TMEM114 CHST3 andCTRB12) had large effects on GLP-1 stimulated insulinsecretion during hyperinsulinemic clamp [85] A pharmaco-genetic study in two independent populationsmdashDutch DCSWest Friesland and GoDARTSmdashwas performed This studyincluded 173 patients treated by GLP-1 receptor antagonistsin which no association of glycemic control after treatmentwith any of the three gene variants was observed Among 354patients treated with gliptins there was a significant associa-tion between the rs7202877 TG variant in the proximity toCTRB12 gene encoding chymotrypsinogen and the responseto gliptin treatment 10 patients carrying the minor G-alleleshowed a significantly smaller decrease in HbA1c (in averageby 05) in comparison to TT homozygotes [85]

7 Conclusions

There are a few areas where genetics is used to guide the-rapy This is becoming mainstream in the field of cancertherapy where somatic mutations can determine the choiceof treatment Out with the cancer setting there are cur-rently limited examples of clinical utility of genetics withprobably the most accepted use being HLA genotypingprior to commencing the antiretroviral abacavir The use ofgenotyping abolishes hypersensitivity reactions to the drug[86] Diabetes is one of the few other disease areas wheregenotype is used to guide therapy albeit in rare monogenicsubtypes [20] In this review we have not focused on theserare subtypes but addressed the evidence for the role ofgenetic variants in the pharmacogenetics of common type 2diabetes This field is progressing rapidly and with increasingrigor The most recent and promising advances have beenin the field of metformin therapy Whilst at present thereis no convincing clinical role for genotype led prescribingsome of the organic cation transporter variants do offersuch promise and the evidence is starting to accumulateto sufficient level to justify a genotype led clinical trial Weshould recognize that pharmacogenetics can also be a usefultool to understand drug mechanism and this is highlightedby the recent genome-wide association study that points toa novel biological mechanism of action of metformin Withincreased collaboration between groups establishment ofdiabetes pharmacogenetics consortia and with reduction incosts of genomics we anticipate that the next five years shouldlead to some significant clinical breakthroughs in this field

Conflict of Interests

The authors declare there is no conflict of interests with res-pect to the present paper

Acknowledgments

The paper was supported by research Grants VEGA 1011211and VEGA 1034012 from the Ministry of EducationResearch Science and Sport of Slovak Republic and APVV0134-11 from the Slovakian Agency for Research and Devel-opment

References

[1] OECD Health at a Glance Europe 2012 OECD 2012

[2] A Hattersley J Bruining J Shield P Njolstad and K C Dona-ghue ldquoThe diagnosis and management of monogenic diabetesin children and adolescentsrdquo Pediatric Diabetes vol 10 supple-ment 12 pp 33ndash42 2009

[3] A Vella ldquoPharmacogenetics for type 2 diabetes practical con-siderations for study designrdquo Journal of Diabetes Science andTechnology vol 3 no 4 pp 705ndash709 2009

[4] D Sherifali K Nerenberg E Pullenayegum J E Cheng andHC Gerstein ldquoThe effect of oral antidiabetic agents onA1C levelsa systematic review and meta-analysisrdquo Diabetes Care vol 33no 8 pp 1859ndash1864 2010

[5] K Zhou LDonnelly L Burch et al ldquoLoss-of-functionCYP2C9variants improve therapeutic response to sulfonylureas in type2 diabetes a go-DARTS studyrdquo Clinical Pharmacology andTherapeutics vol 87 no 1 pp 52ndash56 2010

[6] D M Nathan J B Buse M B Davidson et al ldquoMedical mana-gement of hyperglycemia in type 2 diabetes a consensus algo-rithm for the initiation and adjustment of therapyrdquo DiabetesCare vol 32 no 1 pp 193ndash203 2009

[7] J Kirchheiner J Brockmoller I Meineke et al ldquoImpact ofCYP2C9 amino acid polymorphisms on glyburide kinetics andon the insulin and glucose response in healthy volunteersrdquoClinical Pharmacology andTherapeutics vol 71 no 4 pp 286ndash296 2002

[8] D J Elliot S Suharjono B C Lewis et al ldquoIdentification ofthe human cytochromes P450 catalysing the rate-limiting path-ways of gliclazide eliminationrdquo British Journal of Clinical Phar-macology vol 64 no 4 pp 450ndash457 2007

[9] R Wang K Chen S-Y Wen J Li and S-Q Wang ldquoPhar-macokinetics of glimepiride and cytochrome P450 2C9 geneticpolymorphismsrdquo Clinical Pharmacology and Therapeutics vol78 no 1 pp 90ndash92 2005

[10] J Kirchheiner S Bauer I Meineke et al ldquoImpact of CYP2C9and CYP2C19 polymorphisms on tolbutamide kinetics and theinsulin and glucose response in healthy volunteersrdquo Pharmaco-genetics vol 12 no 2 pp 101ndash109 2002

[11] M L Becker L E Visser P H Trienekens A Hofman R HN Van Schaik and B H C Stricker ldquoCytochrome P450 2C9lowast2 and lowast3 polymorphisms and the dose and effect of sulfo-nylurea in type II diabetes mellitusrdquo Clinical Pharmacology andTherapeutics vol 83 no 2 pp 288ndash292 2008

[12] J J Swen J A M Wessels A Krabben W J J Assendelftand H-J Guchelaar ldquoEffect of CYP2C9 polymorphisms onprescribed dose and time-to-stable dose of sulfonylureas inprimary care patients with type 2 diabetes mellitusrdquo Pharma-cogenomics vol 11 no 11 pp 1517ndash1523 2010

[13] J Kirchheiner I Meineke G Muller et al ldquoInfluence ofCYP2C9 and CYP2D6 polymorphisms on the pharmacokinet-ics of nateglinide in genotyped healthy volunteersrdquo ClinicalPharmacokinetics vol 43 no 4 pp 267ndash278 2004

[14] T B Bidstrup I Bjoslashrnsdottir U G Sidelmann M SThomsenand K T Hansen ldquoCYP2C8 and CYP3A4 are the principalenzymes involved in the human in vitro biotransformation ofthe insulin secretagogue repagliniderdquo British Journal of ClinicalPharmacology vol 56 no 3 pp 305ndash314 2003


[15] A Kalliokoski M Neuvonen P J Neuvonen and M NiemildquoDifferent effects of SLCO1B1 polymorphism on the pharma-cokinetics and pharmacodynamics of repaglinide and nateglin-iderdquo Journal of Clinical Pharmacology vol 48 no 3 pp 311ndash3212008

[16] S-L Shyng and C G Nichols ldquoOctameric stoichiometry of theK(ATP) channel complexrdquo Journal of General Physiology vol110 no 6 pp 655ndash664 1997

[17] M Winkler D Stephan S Bieger P Kuhner F Wolff and UQuast ldquoTesting the bipartite model of the sulfonylurea receptorbinding site binding of A- B- and a + B-site ligandsrdquo Journalof Pharmacology and Experimental Therapeutics vol 322 no 2pp 701ndash708 2007

[18] K S CHammingD Soliman L CMatemisz et al ldquoCoexpres-sion of the type 2 diabetes susceptibility gene variants KCNJ11E23K and ABCC8 S1369A alter the ATP and sulfonylureasensitivities of the ATP-sensitive K+ channelrdquo Diabetes vol 58no 10 pp 2419ndash2424 2009

[19] V Y Lang M Fatehi and P E Light ldquoPharmacogenomicanalysis of ATP-sensitive potassium channels coexpressing thecommon type 2 diabetes risk variants E23K and S1369Ardquo Pha-rmacogenetics and Genomics vol 22 no 3 pp 206ndash214 2012

[20] E R Pearson I Flechtner P R Njoslashlstad et al ldquoSwitching frominsulin to oral sulfonylureas in patients with diabetes due toKir62 mutationsrdquo The New England Journal of Medicine vol355 no 5 pp 467ndash477 2006

[21] M Rafiq S E Flanagan A-M Patch et al ldquoEffective treatmentwith oral sulfonylureas in patients with diabetes due to sulfony-lurea receptor 1 (SUR1) mutationsrdquo Diabetes Care vol 31 no 2pp 204ndash209 2008

[22] J C Florez N Burtt P I W de Bakker et al ldquoHaplotype struc-ture and genotype-phenotype correlations of the sulfonylureareceptor and the islet ATP-sensitive potassium channel generegionrdquo Diabetes vol 53 no 5 pp 1360ndash1368 2004

[23] Y Feng G Mao X Ren et al ldquoSer 1369Ala variant in sul-fonylurea receptor gene ABCC8 Is associated with antidiabeticefficacy of gliclazide in Chinese type 2 diabetic patientsrdquo Dia-betes Care vol 31 no 10 pp 1939ndash1944 2008

[24] H Zhang X Liu H Kuang R Yi and H Xing ldquoAssociationof sulfonylurea receptor 1 genotype with therapeutic responseto gliclazide in type 2 diabetesrdquo Diabetes Research and ClinicalPractice vol 77 no 1 pp 58ndash61 2007

[25] M Javorsky L Klimcakova Z Schroner et al ldquoKCNJ11 geneE23K variant and therapeutic response to sulfonylureasrdquo Euro-pean Journal of Internal Medicine vol 23 no 3 pp 245ndash2492012

[26] A L Gloyn Y Hashim S J H Ashcroft R Ashfield S Wil-tshire and R C Turner ldquoAssociation studies of variants in pro-moter and coding regions of beta-cell ATP sensitive K-channelgenes SUR1 and Kir62 with type 2 diabetes mellitus (UKPDS53)rdquo Diabetic Medicine vol 18 no 3 pp 206ndash212 2001

[27] G Sesti E Laratta M Cardellini et al ldquoThe E23K vari-ant of KCNJ11 encoding the pancreatic 120573-cell adenosine 51015840-triphosphate-sensitive potassium channel subunit Kir62 isassociated with an increased risk of secondary failure tosulfonylurea in patients with type 2 diabetesrdquo Journal of ClinicalEndocrinology and Metabolism vol 91 no 6 pp 2334ndash23392006

[28] Y-Y He R Zhang X-Y Shao et al ldquoAssociation of KCNJ11 andABCC8 genetic polymorphisms with response to repaglinide inChinese diabetic patientsrdquo Acta Pharmacologica Sinica vol 29no 8 pp 983ndash989 2008

[29] S F A Grant G Thorleifsson I Reynisdottir et al ldquoVariant oftranscription factor 7-like 2 (TCF7L2) gene confers risk of type2 diabetesrdquo Nature Genetics vol 38 no 3 pp 320ndash323 2006

[30] Y Tong Y Lin Y Zhang et al ldquoAssociation between TCF7L2gene polymorphisms and susceptibility to type 2 diabetesmellitus a large HumanGenome Epidemiology (HuGE) reviewand meta-analysisrdquo BMC Medical Genetics vol 10 article 152009

[31] F Yi P L Brubaker and T Jin ldquoTCF-4 mediates cell type-specific regulation of proglucagon gene expression by 120573-cateninand glycogen synthase kinase-3120573rdquo The Journal of BiologicalChemistry vol 280 no 2 pp 1457ndash1464 2005

[32] J C Florez K A Jablonski N Bayley et al ldquoTCF7L2 polymor-phisms and progression to diabetes in the Diabetes PreventionProgramrdquoThe New England Journal of Medicine vol 355 no 3pp 241ndash250 2006

[33] R Saxena L Gianniny N P Burtt et al ldquoCommon singlenucleotide polymorphisms in TCF7L2 are reproducibly asso-ciated with type 2 diabetes and reduce the insulin response toglucose in nondiabetic individualsrdquoDiabetes vol 55 no 10 pp2890ndash2895 2006

[34] E R Pearson ldquoTranslating TCF7L2 from gene to functionrdquoDiabetologia vol 52 no 7 pp 1227ndash1230 2009

[35] D T Villareal H Robertson G I Bell et al ldquoTCF7L2 variantrs7903146 affects the risk of type 2 diabetes by modulatingincretin actionrdquo Diabetes vol 59 no 2 pp 479ndash485 2010

[36] E R Pearson L A Donnelly C Kimber et al ldquoVariationin TCF7L2 influences therapeutic response to sulfonylureas aGoDARTs Studyrdquo Diabetes vol 56 no 8 pp 2178ndash2182 2007

[37] M Javorsky E Babjakova L Klimcakova et al ldquoAssocia-tion between TCF7L2 genotype and glycemic control in dia-betic patients treated with gliclaziderdquo International Journal ofEndocrinology vol 2013 Article ID 374858 5 pages 2013

[38] A Holstein M Hahn A Korner M Stumvoll and P KovacsldquoTCF7L2 and therapeutic response to sulfonylureas in patientswith type 2 diabetesrdquo BMC Medical Genetics vol 12 article 302011

[39] G G Graham J Punt M Arora et al ldquoClinical pharmacoki-netics of metforminrdquo Clinical Pharmacokinetics vol 50 no 2pp 81ndash98 2011

[40] K Zhou C Bellenguez C C A Spencer et al ldquoCommonvariants near ATM are associated with glycemic response tometformin in type 2 diabetesrdquo Nature Genetics vol 43 no 2pp 117ndash120 2011

[41] N van Leeuwen G Nijpels M L Becker et al ldquoA gene variantnear ATM is significantly associated with metformin treatmentresponse in type 2 diabetes a replication and meta-analysis offive cohortsrdquo Diabetologia vol 55 no 7 pp 1971ndash1977 2012

[42] J C Florez K A Jablonski A Taylor et al ldquoThe C allele ofATM rs11212617 does not associate with metformin response inthe Diabetes Prevention Programrdquo Diabetes Care vol 35 no 9pp 1864ndash1867 2012

[43] Y Shu S A Sheardown C Brown et al ldquoEffect of genetic vari-ation in the organic cation transporter 1 (OCT1) on metforminactionrdquo The Journal of Clinical Investigation vol 117 no 5 pp1422ndash1431 2007

[44] Y Shu C Brown R A Castro et al ldquoEffect of genetic variationin the organic cation transporter 1 OCT1 on metforminpharmacokineticsrdquo Clinical Pharmacology and Therapeuticsvol 83 no 2 pp 273ndash280 2008


[45] H Takane E Shikata K Otsubo S Higuchi and I Ieiri ldquoPoly-morphism in human organic cation transporters and met-formin actionrdquo Pharmacogenomics vol 9 no 4 pp 415ndash4222008

[46] A T Nies H Koepsell S Winter et al ldquoExpression of organiccation transporters OCT1 (SLC22A1) and OCT3 (SLC22A3)is affected by genetic factors and cholestasis in human liverrdquoHepatology vol 50 no 4 pp 1227ndash1240 2009

[47] M V Tzvetkov S V Vormfelde D Balen et al ldquoThe effectsof genetic polymorphisms in the organic cation transportersOCT1 OCT2 and OCT3 on the renal clearance of metforminrdquoClinical Pharmacology andTherapeutics vol 86 no 3 pp 299ndash306 2009

[48] K Zhou L A Donnelly C H Kimber et al ldquoReduced-functionSLC22A1 polymorphisms encoding organic cation transporter1 and glycemic response to metformin a GoDARTS studyrdquoDiabetes vol 58 no 6 pp 1434ndash1439 2009

[49] M M H Christensen C Brasch-Andersen H Green et alldquoThe pharmacogenetics of metformin and its impact on plasmametformin steady-state levels and glycosylated hemoglobinA1crdquo Pharmacogenetics and Genomics vol 21 no 12 pp 837ndash850 2011

[50] M L Becker L E Visser R H N van Schaik A HofmanA G Uitterlinden and B H C Stricker ldquoGenetic variation inthe organic cation transporter 1 is associated with metforminresponse in patients with diabetes mellitusrdquo PharmacogenomicsJournal vol 9 no 4 pp 242ndash247 2009

[51] L Tarasova I Kalnina K Geldnere et al ldquoAssociation of gen-etic variation in the organic cation transporters OCT1 OCT2and multidrug and toxin extrusion 1 transporter protein geneswith the gastrointestinal side effects and lower BMI in metfo-rmin-treated type 2 diabetes patientsrdquo Pharmacogenetics andGenomics vol 22 no 9 pp 659ndash666 2012

[52] Q Li F Liu T-S Zheng J-L Tang H-J Lu and W-P JialdquoSLC22A2 gene 808 GT variant is related to plasma lactateconcentration in Chinese type 2 diabetics treated with metfo-rminrdquo Acta Pharmacologica Sinica vol 31 no 2 pp 184ndash1902010

[53] I S Song H J Shin E J Shim et al ldquoGenetic variants of theorganic cation transporter 2 influence the disposition of metfo-rminrdquo Clinical Pharmacology and Therapeutics vol 84 no 5pp 559ndash562 2008

[54] Z-J Wang O Q P Yin B Tomlinson and M S S ChowldquoOCT2 polymorphisms and in-vivo renal functional conse-quence studies with metformin and cimetidinerdquo Pharmacoge-netics and Genomics vol 18 no 7 pp 637ndash645 2008

[55] Y Chen S Li C Brown et al ldquoEffect of genetic variation inthe organic cation transporter 2 on the renal elimination ofmetforminrdquo Pharmacogenetics and Genomics vol 19 no 7 pp497ndash504 2009

[56] M L Becker L E Visser R H N van Schaik A HofmanA G Uitterlinden and B H C Stricker ldquoGenetic variationin the multidrug and toxin extrusion 1 transporter proteininfluences the glucose-lowering effect of metformin in patientswith diabetes a preliminary studyrdquo Diabetes vol 58 no 3 pp745ndash749 2009

[57] M L Becker L E Visser R H N van Schaik A Hofman A GUitterlinden and B H C Stricker ldquoInteraction between poly-morphisms in the OCT1 and MATE1 transporter and met-formin responserdquo Pharmacogenetics and Genomics vol 20 no1 pp 38ndash44 2010

[58] K A Jablonski J B McAteer P I W de Bakker et alldquoCommon variants in 40 genes assessed for diabetes incidenceand response tometformin and lifestyle intervention in the dia-betes prevention programrdquo Diabetes vol 59 no 10 pp 2672ndash2681 2010

[59] I Tkac L Klimcakova M Javorsky et al ldquoPharmacogenomicassociation between a variant in SLC47A1 gene and therapeuticresponse to metformin in type 2 diabetesrdquoDiabetes Obesity andMetabolism vol 15 no 2 pp 189ndash191 2012

[60] S L Stocker K M Morrissey S W Yee et al ldquoThe effect ofnovel promoter variants in MATE1 and MATE2 on the phar-macokinetics and pharmacodynamics of metforminrdquo ClinicalPharmacology andTherapeutics vol 93 no 2 pp 186ndash194 2013

[61] J H Choi S W Yee A H Ramirez et al ldquoA common 51015840-UTR variant in MATE2-K is associated with poor response tometforminrdquo Clinical Pharmacology and Therapeutics vol 90no 5 pp 674ndash684 2011

[62] J Kirchheiner S Thomas S Bauer et al ldquoPharmacokineticsand pharmacodynamics of rosiglitazone in relation to CYP2C8genotyperdquo Clinical Pharmacology and Therapeutics vol 80 no6 pp 657ndash667 2006

[63] A Tornio M Niemi P J Neuvonen and J T BackmanldquoTrimethoprim and the CYP2C8lowast3 allele have opposite effectson the pharmacokinetics of pioglitazonerdquoDrugMetabolism andDisposition vol 36 no 1 pp 73ndash80 2008

[64] C-W Yeo S-J Lee S S Lee et al ldquoDiscovery of a novel allelicvariant of CYP2C8 CYP2C8lowast11 in asian populations and itsclinical effect on the rosiglitazone disposition in vivordquo DrugMetabolism and Disposition vol 39 no 4 pp 711ndash716 2011

[65] T B StageMMH Christensen S Federsen H Beck-Nielsenand K Broslashsen ldquoThe role of genetic variants in CYP2C8 LPIN1PPARGCA and PPAR120574on the trough steady state plasma con-centrations of rosiglitazone and on glycosylated haemoglobinA1c in type 2 diabetesrdquo Pharmacogenetics and Genomics vol 22no 4 pp 219ndash227 2013

[66] Y Chung-Jen B A Beamer CNegri et al ldquoMolecular scanningof the human peroxisome proliferator activated receptor 120574(hPPAR120574) gene in diabetic Caucasians identification of aPro12Ala PPAR1205742 missense mutationrdquo Biochemical and Bio-physical Research Communications vol 241 no 2 pp 270ndash2741997

[67] H N Gouda G S Sagoo A-H Harding J YatesM S Sandhuand J P T Higgins ldquoThe association between the peroxisomeproliferator-activated receptor-1205742 (PPARG2) Pro12Ala genevariant and type 2 diabetes mellitus a HuGE review and meta-analysisrdquo American Journal of Epidemiology vol 171 no 6 pp645ndash655 2010

[68] M-C Hsieh K-D Lin K-J Tien et al ldquoCommon poly-morphisms of the peroxisome proliferator-activated receptor-120574 (Pro12Ala) and peroxisome proliferator-activated receptor-120574coactivator-1 (Gly482Ser) and the response to pioglitazone inChinese patients with type 2 diabetes mellitusrdquoMetabolism vol59 no 8 pp 1139ndash1144 2010

[69] E S Kang S Y Park H J Kim et al ldquoEffects of Pro12Ala poly-morphism of peroxisome proliferator-activated receptor 1205742gene on rosiglitazone response in type 2 diabetesrdquo ClinicalPharmacology and Therapeutics vol 78 no 2 pp 202ndash2082005

[70] J C Florez K A Jablonski M W Sun et al ldquoEffects of thetype 2 diabetes-associated PPARG P12A polymorphism on pro-gression to diabetes and response to troglitazonerdquo Journal of


Clinical Endocrinology and Metabolism vol 92 no 4 pp 1502ndash1509 2007

[71] M Bluher G Lubben and R Paschke ldquoAnalysis of the relation-ship between the Pro12Ala variant in the PPAR-1205742 gene and theresponse rate to therapy with pioglitazone in patients with type2 diabetesrdquo Diabetes Care vol 26 no 3 pp 825ndash831 2003

[72] F Namvaran N Azarpira P Rahimi-Moghaddam and M HDabbaghmanesh ldquoPolymorphism of peroxisome proliferator-activated receptor 120574 (PPAR120574) Pro12Ala in the Iranian popula-tion relation with insulin resistance and response to treatmentwith pioglitazone in type 2 diabetesrdquo European Journal ofPharmacology vol 671 no 1ndash3 pp 1ndash6 2011

[73] S K Eun Y P So J K Hyeong et al ldquoThe influence of adipo-nectin gene polymorphism on the rosiglitazone response inpatients with type 2 diabetesrdquo Diabetes Care vol 28 no 5 pp1139ndash1144 2005

[74] Z Li X Peng Y Wu Y Xia X Liu and Q Zhang ldquoThe influ-ence of adiponectin gene polymorphism on the pioglitazoneresponse in the Chinese with type 2 diabetesrdquo Diabetes Obesityand Metabolism vol 10 no 9 pp 794ndash802 2008

[75] F Namvaran P Rahimi-Moghaddam N Azarpira and MH Dabbaghmanesh ldquoPolymorphism of adiponectin (45TG)and adiponectin receptor-2 (795GA) in an Iranian populationrelation with insulin resistance and response to treatment withpioglitazone in patientswith type 2 diabetesmellitusrdquoMolecularBiology Reports vol 39 no 5 pp 5511ndash5518 2012

[76] D J Drucker andM A Nauck ldquoThe incretin system glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4inhibitors in type 2 diabetesrdquoThe Lancet vol 368 no 9548 pp1696ndash1705 2006

[77] A Sathananthan C Dalla Man F Micheletto et al ldquoCommongenetic variation in GLP1R and insulin secretion in responseto exogenous GLP-1 in nondiabetic subjects a pilot studyrdquoDiabetes Care vol 33 no 9 pp 2074ndash2076 2010

[78] R Saxena M F Hivert C Langenberg et al ldquoGenetic variationin GIPR influences the glucose and insulin responses to an oralglucose challengerdquo Nature Genetics vol 42 no 2 pp 142ndash1482010

[79] S A Schafer O Tschritter F Machicao et al ldquoImpaired gluca-gon-like peptide-1-induced insulin secretion in carriers oftranscription factor 7-like 2 (TCF7L2) gene polymorphismsrdquoDiabetologia vol 50 no 12 pp 2443ndash2450 2007

[80] K Faeligrch K Pilgaard F K Knop et al ldquoIncretin and pan-creatic hormone secretion in Caucasian non-diabetic carriersof the TCF7L2 rs7904146 risk T allelerdquo Diabetes Obesity andMetabolism vol 15 no 1 pp 91ndash95 2013


[82] K Mussig H Staiger F Machicao et al ldquoAssociation of type 2diabetes candidate polymorphisms in KCNQ1with incretin andinsulin secretionrdquo Diabetes vol 58 no 7 pp 1715ndash1720 2009

[83] S A Schafer K Mussig H Staiger et al ldquoA common geneticvariant in WFS1 determines impaired glucagon-like peptide-1-induced insulin secretionrdquoDiabetologia vol 52 no 6 pp 1075ndash1082 2009

[84] G Smushkin M Sathananthan A Sathananthan et al ldquoDia-betes-associated common genetic variation and its associationwith GLP-1 concentrations and response to exogenous GLP-1rdquoDiabetes vol 61 no 5 pp 1082ndash1089 2012

[85] L M rsquot Haart A Fritsche G Nijpels et al ldquoThe CTRB12 locusaffects diabetes susceptibility and treatment via the incretinpathwayrdquo Diabetes vol 62 no 9 pp 3275ndash3281 2013

[86] M A Martin T E Klein B J Dong M Pirmohamed DW Haas and D L Kroetz ldquoClinical pharmacogenetics imple-mentation consortium guidelines for HLA-B genotype andabacavir dosingrdquo Clinical Pharmacology and Therapeutics vol91 no 4 pp 734ndash738 2012

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom


study is a prospective genotype-blind study design with ade-quate statistical power these are very costly time-consumingand often require participation ofmultiple centersThereforethe majority of studies published used retrospective dataretrieved from registers and databases In the retrospectivestudies it is possible to achieve reasonable sample sizes forpharmacogenetic studies but it can be difficult to adjustfor all confounding variables as these may not have beenmeasured However as most confounding should not begenotype dependent this should not be a major issue

With respect to the outcomes used in pharmacogeneticstudies there are two main types of endpoints Pathophys-iological endpoints reflect the effect of gene variants oninsulin resistance or insulin secretion Such studies may yieldnovel knowledge on the role of different gene variants in thepathogenesis of T2D but their clinical applicability mightbe limited Thus from the point of view of possible clinicalimplications endpoints such as reduction of HbA1c fastingplasma glucose and postprandial plasma glucose are of thehighest importance Among them both an achievement oftherapeutic target of HbA1clt7 defined in guidelines anda reduction in HbA1c seem to be the most appropriateendpoints for pharmacogenetic studies in T2D [3]

Different factors may confound the relationship betweenthe studied gene variants and endpoints When using achi-evement ofHbA1clt7or reduction inHbA1c as an endpointthe strongest confounder is the baseline level of HbA1c asit was shown in both the meta-analysis of clinical trials andalso in the individual pharmacogenetic studies [4 5] Thatmeanmdashif there is an imbalance in baseline HbA1c among thegenotypesmdashone could expect higher response in the patientswith higher baselineHbA1c values On the other hand higherbaseline HbA1c might also reflect the effect of the genevariant on previous diabetes control Thus it is correct toshow rough (unadjusted) data alongwith themodels adjustedfor confounding variables It is also important to take intoaccount whether the drug of interest is being used earlyon in the disease process or as add-on therapy later in thedisease where there is less likely to be a large therapeuticeffect Covariates that are usually taken into account such asage gender BMI and diabetes duration are rarely significantpredictors of the drug effect in pharmacogenetic studies Ifthe drug dose during the study is not constant it is reasonableto adjust for drug dosage Since the majority of the oralantidiabetic drugs are eliminated by the kidney adjustmentfor ameasure of renal function such asGFRor creatinine levelis also frequently needed Furthermore if retrievable fromdatabases medication adherence and period of drug use areuseful covariates

We selected studies discussed in this review taking intoaccount the above mentioned methodological aspects ofpharmacogenetic studies in patientswithT2DWeperformeda search in the MEDLINE and Web of Knowledge databasesusing ldquopharmacogenetics OR pharmacogenomicsrdquo keywordin combinations with the names of groups of antidiabeticdrugs or generic names of individual antidiabetic drugsIn the present review we will discuss the mechanisms ofassociations between gene variants and the drug effect in thehealthy subjects in cell lines and in the patients with T2D

The studies which examined the effect of gene variants on theindices of glucose control will be presented only when at least100 patients with T2D are included in the study

3 Pharmacogenetics of Insulin Secretagogues

Insulin secretagogues stimulate secretion of insulin from thepancreatic 120573-cells Sulfonylureas have been used for morethan a half-century while nonsulfonylurea secretagoguesmeglitinides (glinides) have been used for approximately thelast 15 years in the treatment of T2D The most commonside effects of both groups of drugs are hypoglycemia andweight gain While previously sulfonylureas were consideredas first-line treatment in less obese patients with T2D morerecent guidelines recommend both groups of drugs mainlyin combination therapy with metformin as a second-linetreatment [6]

31 Gene Variants Related to Pharmacokinetics ofInsulin Secretagogues

311 Cytochrome P450 Isoenzyme 2C9 Sulfonylureas aremetabolized in the liver primarily by cytochrome P450isoenzyme 2C9 encoded by 11986211988411987521198629 [7ndash10] The major alleleof this gene is 11986211988411987521198629lowast1 Two common nonsynonymousvariants Arg144Cys (11986211988411987521198629lowast2) and Ile359Leu (11986211988411987521198629lowast3)were identified [7] Studies on healthy subjects showed thatthe variant alleles are associated with increased plasmaconcentration and decreased clearance of sulfonylureas afteroral administration [7 10]

The first robust study to confirm association of11986211988411987521198629 variants with therapeutic response to sulfonylureaswas Genetics of Diabetes Audit and Research Tayside Study(GoDARTS) which included 1073 patients treated with sulfo-nylureas The 6 of the population who were carriers oftwo loss-of-function alleles (lowast2lowast2 or lowast2lowast3 or lowast3lowast3) had a05 greater reduction in HbA1c compared with wild-typehomozygotes (119875 = 0003) and were 34 times more likely toachieve on-treatment HbA1clt7 (119875 = 0009) [5]

Further studies in patients with type 2 diabetes used indi-rect endpoints of sulfonylurea effect such as prescribed doseof sulfonylurea Becker et al analyzed 475 elderly patientsincluded in the Rotterdam study who started sulfonylureatreatment In the largest subgroup of patients treated withtolbutamide (119899 = 172) there was a significant differencein daily prescribed dose observed between genotypes Whilethe prescribed dose of tolbutamide increased in the carriersof lowast1lowast1 lowast1lowast2 or lowast2lowast2 genotypes there was very littletolbutamide dose increase in patients with lowast1lowast3 or lowast2lowast3genotypes (119875 = 0009) [11] In a similar performed analysisfrom a different Dutch group which analyzed data from207 incident sulfonylurea users a trend towards lower stableglimepiride dose (119875 = 007) was observed in the carriers ofthe 11986211988411987521198629lowast3 allele [12]

Studies in healthy volunteers showed that variants inCYP2C9 influence also pharmacokinetics of nonsulfonylureasecretagogue nateglinide [13] while variant in CYP2C8 isassociated with pharmacokinetics of repaglinide [14] Phar-macokinetics of both of these glinide drugs is also influenced






















































7 Conclusions




Acknowledgments


References
































































































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





































































7 Conclusions




Acknowledgments


References
































































































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

























































7 Conclusions




Acknowledgments


References
































































































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of













































7 Conclusions




Acknowledgments


References
































































































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

































7 Conclusions




Acknowledgments


References
































































































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of


















7 Conclusions




Acknowledgments


References
































































































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

































































































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of


































































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of







































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





















of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
















Review Article Pharmacogenetics of Oral Antidiabetic DrugsPharmacogenetics of Oral Antidiabetic Drugs MatthijsL.Becker, 1,2 EwanR.Pearson, 3 andIvanTká I4,5 Department of Epidemiology,

Documents