novel slc22a16 polymorphisms and influence on doxorubicin...

RESEARCH REPORTFor reprint orders, please contact:[email protected]

Novel SLC22A16 polymorphisms and influence on doxorubicin pharmacokinetics in Asian breast cancer patients
Suman Lal1, Zee Wan Wong2, Srinivasa Rao Jada1, Xiaoqiang Xiang1, Xiao Chen Shu1, Peter Cher Siang Ang2, William D Figg3, Edmund JD Lee4 & Balram Chowbay1†
†Author for correspondence1National Cancer Centre, Division of Medical Sciences, 11 Hospital Drive, 169610 SingaporeTel.: +65 6436 8321;Fax: +65 6372 0161;E-mail: [email protected] Cancer Center, Department of Medical Oncology, Singapore3National Cancer Institute, Medical Oncology Branch, Center for Cancer Research, Bethesda, MD, USA4National University of Singapore, Department of Pharmacology, Singapore

part of

Keywords: Asian breast cancer patients, doxorubicinol, doxorubicin pharmacokinetics, hCT2, hOCT6, influx transporters, SLC22A16

10.2217/14622416.8.6.567 © 2

Objective: To identify novel polymorphisms in the solute carrier SLC22A16 gene and determine their influence on the pharmacokinetics of doxorubicin and doxorubicinol in Asian breast cancer patients. Methods: SLC22A16 coding regions were screened in a total of 400 healthy subjects belonging to three distinct Asian ethnic groups (Chinese [n = 100], Malays [n = 100] and Indians [n = 100]) and in the Caucasian population (n = 100). Pharmacokinetic parameters of doxorubicin and doxorubicinol were estimated in Asian breast cancer patients undergoing adjuvant chemotherapy to investigate genotype–phenotype correlations. Results: Four novel polymorphisms (c.146A>G [exon 2], c.312T>C, c.755T>C [exon 4] and c.1226T>C [exon 5]) were identified. The genotypic frequency of the homozygous c.146GG polymorphism was approximately twofold higher in the healthy Chinese (13%) & Malay (18%) populations compared with the Indian (7%) and Caucasian (9%) populations. The genotypic frequency of the c.1226T>C polymorphism was observed to be significantly higher among the Caucasian (11%) and Indian (8%) study subjects compared with the Chinese (1%) and Malay (1%) ethnic groups (p < 0.005 in each case). Breast cancer patients harboring the 146GG genotype showed a trend towards higher exposure levels to doxorubicin (AUC0–∞/dose/body surface area [BSA] [hm-5]: 21.6; range: 18.8–27.7) compared with patients with either the reference genotype (AUC0–∞/dose/BSA[hm-5]: 17.4; range: 8.2–26.3, p = 0.066) or heterozygotes (AUC0–∞/dose/BSA[hm-5]: 15.4; range: 6.2–38.0, p = 0.055). The exposure levels of doxorubicinol were also higher in patients harboring the variant 146GG genotype (AUC0–∞/dose/BSA[hm-5]: 13.3; range: 8.8–21.7) when compared with patients harboring the reference genotype (AUC0–∞/dose/BSA[hm-5]): 9.8; range: 6.1–24.3, p = 0.137) or heterozygotes (AUC0–∞/dose/BSA[hm-5]: 8.98; range: 3.7–20.6, p = 0.047). Conclusion: Among the four novel SLC22A16 polymorphisms identified, the c.146A>G and c.1226T>C polymorphisms exhibited interethnic variations in allele and genotype frequencies. This exploratory study suggests that the c.146A>G variation could contribute to the variations in the pharmacokinetics of doxorubicin and doxorubicinol in Asian cancer patients. Further in vitro studies are required to determine the functional impact of these novel polymorphisms on doxorubicin pharmacokinetics in cancer patients.

The therapeutic effects of drugs results from acomplex interplay of processes that modulatedrug disposition and response. These processescommonly involve the participation of drug-metabolizing enzymes that catalyze Phase I oxi-dation/reduction or hydrolysis reactions andPhase II conjugation reactions that utilize polarsubstrates such as glucuronic acid, glutathioneor sulfate for conjugation to the drug or itsPhase I derived metabolite to produce a morepolar metabolite that can be more readilyexcreted. Recently, drug transporters have alsobeen shown to play an important role in drugdisposition whereby they facilitate the inter-compartmental transfer of drug molecules inorgans involved in the detoxification processes.

These drug transporters are constitutivelyexpressed in several organs involved with thedetoxification of drugs, such as the liver, kidneysand intestine.

Since the sequencing of the human genome, itis estimated that approximately 500–1200 genesare involved in encoding drug transporters [1,2].The transport proteins are members of largefamilies of proteins responsible for gastro-intestinal [3], renal [4], hepatobiliary [5] andCNS [6] transport and include members of theATP-binding cassette (ABC) transporter super-family and the solute carrier (SLC) family ofproteins. The latter family includes the organicanion transporting polypeptides (OATPs,SLCO), organic anion transporters (OATs,

007 Future Medicine Ltd ISSN 1462-2416 Pharmacogenomics (2007) 8(6), 567–575 567

RESEARCH REPORT – Lal, Wong, Jada et al.

568

SLC22A), organic cation transporters (OCTs,SLC22A), organic cation/carnitine transporter(OCTN/SLC22A) and peptide transporters(PEPTs, SLC15A). The ABC transporters areinvolved with efflux of drugs, while theSLC families of proteins mainly regulateuptake activities.

Chemotherapy failure in cancer patients iscaused in part by the phenomenon of multidrugresistance resulting from either overexpression ofefflux transporters or the downregulation ofuptake transporters in tumor cells that eventuallyaffects the attainment of effective cytotoxic intra-cellular concentrations [7,8]. These transportersoften have overlapping substrate capabilities andthe intracellular tumoricidal concentrationachieved is the result of a dynamic interplaybetween uptake and efflux transporters. The roleof efflux transporters affecting disposition ofchemotherapeutic agents are well studied.Recent studies have emphasized the importantrole of influx transporters in influencing thepharmacokinetics and pharmacodynamics ofchemotherapeutic agents and endogenouscompounds [9–12].

Amongst the various influx transporters iden-tified to date, the organic cation transporterSLC22A16, also known as the organic cationtransporter 6 (hOCT6), carnitine transporter 2(hCT2), Fly-like putative transporter 2(hFLIPT2), or organic cation-binding protein 1(OKB1) [10–12], is constitutively expressed inleukemic as well as a variety of cancer celllines [16], brain, kidney and the Sertoli and epi-thelial cells of the epididymal ducts in thetestis [15]. SLC22A16 is known to transportL-carnitine, which is an essential cofactor ofmetabolism with a definitive role in diseasestates [14,17–20]. Recent in vitro studies in Xenopusoocytes expressing the SLC22A16 protein dem-onstrated a dose-dependent and saturable kinet-ics with regards to influx of doxorubicin [16].Studies with Jurkat cells overexpressingSLC22A16 showed them to be increasingly sus-ceptible to the cytotoxic effects of doxorubicin,and this was postulated to result from theincreased influx of the drug. These studies sug-gest that alterations in the expression and activityof SLC22A16 may have profound implicationson pharmacokinetics and pharmacodynamics ofdoxorubicin in cancer patients.

Functional polymorphic variants in genesencoding proteins involved in drug disposition areknown to cause inherited differences in theobserved therapeutic efficacy or toxicities of drugs.

The genotypic and allelic frequencies of func-tional genetic variants vary in different ethnicgroups and often account for the observed differ-ences in the pharmacokinetics and pharmaco-dynamics of chemotherapeutic agents in cancerpatients belonging to different ethnic origins.Understanding the clinical relevance of geneticpolymorphisms in newly identified transportergenes such as SLC22A16 may provide importantinsights on the mechanistic basis of differences inthe pharmacokinetics of candidate drug substratessuch as doxorubicin in cancer patients belongingto ethnically distinct populations. In the presentexploratory study, we characterized thepharmacogenetic profile of the SLC22A16 gene inthree distinct healthy Asian ethnic populations,namely, the Chinese, Malay and Indian popula-tions, as well as in healthy Caucasians. Thegenotypic–phenotypic effects of novel SLC22A16polymorphisms on the disposition of doxorubicinand its metabolite, doxorubicinol, wereinvestigated in a cohort of Asian breast cancerpatients receiving adjuvant chemotherapy withdoxorubicin and cyclophosphamide.

Materials & methodsHealthy subjectsThe healthy subjects comprised three ethnicgroups predominant in the Asian population(Chinese [n = 100], Malay [n = 100] and Indi-ans [n = 100]) and Caucasian subjects (n = 100).The ethnicity of the study subjects was con-firmed by careful screening and verified againstNational Registry Identification Cards. All par-ticipants provided approved informed consentfor the study. The study was approved by theethics review committee of the National CancerCenter, Singapore.

Breast cancer patientsPatients who had histologically confirmed invasivebreast cancer (n = 62) and were deemed fit toreceive adjuvant chemotherapy with doxorubicinwere recruited for the study. Informed consent wasobtained from all patients, and the institutionalethics committee at the National Cancer Centre,Singapore, approved the study. Inclusion criteriaincluded: patients to have adequate bone marrow(absolute neutrophil count < 1500/µl, plateletcount < 100,000/µl), hepatic (aspartate ami-notransferase and alanine aminotransferaselevels ≤ 2.5 and total bilirubin < 2.0 times upperlimit of normal) and renal functions (serumcreatinine < 140 mmol/l) documented at the timeof enrollment. Patients with serious comorbidities,

Pharmacogenomics (2007) 8(6) future science groupfuture science group

SLC22A16 polymorphisms and doxorubicin pharmacokinetics – RESEARCH REPORT

future science groupfuture science group

including poorly controlled diabetes mellitus,ischemic heart disease, uncontrolled hypertension,active infection or a performance status score ofmore than two on the Eastern Cooperative Oncol-ogy Group (ECOG) scale, were excluded from thestudy. Pregnancy and the use of growth factorsduring the cycle of chemotherapy were also con-sidered as criteria for exclusion. No forms of endo-crine therapy, immunotherapy or biologicalresponse modifiers were allowed during the periodof chemotherapy.

Doxorubicin administration & pharmacokinetic analysisDoxorubicin was administered at a dose of60 mg/m2 intravenously (IV) over 10 min, onceevery 3-week cycle after standard premedicationswith IV dexamethasone 10 mg, diphen-hydramine 50 mg, cimetidine 300 mg or raniti-dine 50 mg. The adjuvant regimen also includedIV cyclophosphamide 600 mg/m2 administeredover 30 min once every 3 weeks. Hematologicalparameters including hemoglobin (Hb), totalleukocyte count, platelet count and the absoluteneutrophil count (ANC) after the start of treat-ment (day 7 and 14) and prior to the start ofeach cycle were measured. New treatment cycleswere started only if the ANC count was greaterthan 1500/µl and the platelet count was greaterthan 100,000/µl. Blood samples for pharmaco-kinetic analysis were drawn at the following timepoints after the start of infusion on the first dayof the first cycle of chemotherapy: at predose, 5,15 and 30 min and at 1, 4, 8 and 24 h. Bloodsamples were collected in plain ethylenediamine-tetraacetic-acid-containing vacutainer glass tubesand immediately centrifuged at 1000 g for10 min; the plasma fraction was collected andstored at -20°C until analysis.

Plasma concentrations of doxorubicin and itsmajor metabolite doxorubicinol were estimatedby reversed-phase high-performance liquidchromatography with fluorescence detection.Briefly, following a single protein precipitationstep, chromatographic separation was accom-plished using a C-18 column with a mobilephase consisting of 50 mM sodium phosphatebuffer-acetonitrile-1-propanol (65:25:2, v/v),pH 2.0. The analytes were measured by fluores-cence detection with an excitation wavelength of480 nm and emission wavelength of 560 nm.The lower limits of quantitation were 10 ng/mlfor doxorubicin, and 5 ng/ml for doxorubicinol.The calibration curves were linear over a concen-tration range of 10–2500 ng/ml for doxorubicin,

and 5–1250 ng/ml for doxorubicinol. The aver-age recoveries were greater than 89% for all ana-lytes and within-day and between-daycoefficients of variation were less than 13%.

Pharmacokinetic parameters were determinedusing a nonlinear regression program WinNon-Lin version 2.1 (Pharsight Inc, CA, USA). Thearea under the plasma concentration–time curve(AUC) was calculated from time zero to the time(t) of the last detectable concentration (AUC0→t)using the trapezoidal rule. The area was extra-polated to infinity (AUC0–∞) by adding Ct/λz toAUC0→t, where Ct was the last detectable plasmaconcentration and λz is the elimination rate con-stant. Peak plasma concentrations (Cmax) weredirectly identified from individual subjectconcentration–time curves.

Identification of SLC22A16 variants by sequence analysisPurified genomic DNA was isolated fromperipheral blood samples (5 ml) using red celllysis buffer and Proteinase K digestion. Table 1

lists the primers used for amplification of exons1–8 of the SLC22A16 gene. Amplification of theexonic regions of SLC22A16 gene (GenBankaccession number: NM033125) was performedusing PCR and the products electrophoresed on2% agarose gels. PCR products were purifiedusing the Qiagen PCR purification kit (QiagenGmbh, Germany) and subjected to directsequencing using the CEQ Dye terminator cyclesequencing (DTCS) quick start kit. The prod-ucts were analyzed on a CEQ 2000XL DNAanalysis system (Beckman Coulter, CA, USA)according to manufacturer’s instructions.

Statistical analysisThe Exact test was used to determinewhether the polymorphisms conformed toHardy–Weinberg proportions. Fisher’s exact testwas used to assess the difference in genotype andallele distributions among the different groups inhealthy subjects and cancer patients. All pharma-cokinetic parameters are expressed as median;range, unless otherwise stated. The possible con-tribution of subject characteristics (age, heightand body surface area [BSA]) on pharmaco-kinetics of doxorubicin and doxorubicinol weredetermined using univariate linear regressionanalysis. The minimum level of statistical signif-icance was set at p = 0.05 for pairwise compari-sons. All statistical analyses were performedusing STATA (STATA Statistical Software release7.0, Stata Corporation, TX, USA).

569www.futuremedicine.com


570

Table 1. Primers use

Exon Primer seque

1 F: 5´-AGAATACR: 5´-CGGGGT

2 F: 5´-AGATAAAR: 5´-CATAATT

3 F: 5´-AAAGGCR: 5´-CGACAC

4 F: 5´-GGAGGAR: 5´-GCCATTT

5 F: 5´-CATTTCAR: 5´-GTCATTG

6 F: 5´-CACACTGR: 5´-GGCCAA

7 F: 5´-GCCTGACR: 5´-AGCTGTT

8 F: 5´-TCAAATCR: 5´-TGGGAG

bp: Base pairs; F: Forward

ResultsDemographics of study subjectsMost of the breast cancer patients belonged to theChinese ethnic group (76%), followed by Malays(18%) and Indians (5%). The median age, heightand BSA of the patients were 51 years(range: 29–73 years), 153 cm (range: 144–168 cm)and 1.52 m2 (range: 1.23–1.95 m2), respectively.

SLC22A16 polymorphisms in healthy subjects and breast cancer patientsFour novel polymorphisms were identified bydirect sequencing of the coding regions of theSLC22A16 gene (Table 2). Three polymorphismswere nonsynonymous transitions: c.146A>G(exon 2), c.755T>C (exon 4) and c.1226T>C(exon 5) polymorphisms resulting in the p.H49R,p.V252A and p.M409T amino-acid changes,respectively. The locations of amino-acid substitu-tions were assigned using GenBank sequenceNM033125 as reference. The c.312T>C poly-morphism was synonymous and was in completelinkage to the c.146A>G polymorphism.

The genotype and allele frequencies of thefour novel polymorphisms identified in theAsian and Caucasian healthy ethnic groups andthe breast cancer patients are listed in Table 3.There were no significant deviations fromHardy–Weinberg equilibrium in the genotypefrequencies in both healthy subjects and cancerpatients. The genotypic frequency of thehomozygous c.146GG polymorphism wasalmost twofold higher in the healthy Chinese(13%) & Malay (18%) populations compared

with the Indian (7%) and Caucasian (9%) popu-lations. Genotypic frequencies were significantlydifferent only between Malays and Indians(p = 0.019). The allele frequency distribution ofthe c.146A>G polymorphism did not differsignificantly between ethnic groups. Thec.755T>C polymorphism in exon 4 was mono-morphic in 2% of Chinese and absent in the restof the Asian ethnic groups and the Caucasianpopulation. The genotypic frequency of thec.1226T>C polymorphism in exon 5 wasobserved to be similar in Caucasian (11%) andIndian (8%) study subjects and significantlyhigher compared with the Chinese (1%) andMalay (1%) ethnic groups (Chinese versus Cauca-sians: p < 0.001; Chinese versus Indians:p = 0.001; Malays versus Indians: p = 0.004;Malays versus Caucasians: p = 0.001). The allelefrequency of all polymorphisms among cancerpatients conformed to the frequency found amongthe local population from which they wererecruited, suggesting lack of selection bias.

SLC22A16 polymorphisms & doxorubicin pharmacokineticsTable 4 summarizes pharmacokinetic parametersof doxorubicin and its metabolite doxorubicinolin Asian breast cancer patients. Plasma samplesfor pharmacokinetic analysis were available in43 of the 62 patients. The clearance of doxo-rubicin was 24.1 l/hr/m2 and was within therange reported in earlier studies [21]. Approxi-mately fourfold (range: 15.04–55.66 l/hr/m2)and sixfold (range: 122.0–703.7 l/hr/m2) varia-tions in clearance and volume of distribution atsteady state of doxorubicin were observed amongthe cancer patients, respectively. Covariate analy-sis showed significant correlations of t1/2 of doxo-rubicin with the age (p = 0.017) and peak plasmaconcentration (Cmax) with BSA (p = 0.039).

The influence of SLC22A16 c.146A>G andc.755T>C polymorphisms on the pharmaco-kinetic parameters of doxorubicin and doxo-rubicinol are depicted in Table 5. The impact ofthe c.312T>C polymorphism on the pharmaco-kinetic parameters of doxorubicin and doxo-rubicinol was identical to that of the c.146A>Gpolymorphism, as they were in complete linkagedisequilibrium (r2 = 1). Breast cancer patients har-boring the 146GG genotype showed a trendtowards higher exposure levels to doxorubicincompared with patients with either the referencegenotype (GG vs AA, AUC0–∞/dose/BSA [hm-5]:21.6; range: 18.8–27.7 vs 17.4; range: 8.2–26.3,p = 0.066) or heterozygotes (GG vs AG,

d in amplification of SLC22A16 gene exons.

nces Fragment size (bp)

AGCCTCGCGCTGGT-3´GAGGAGGAAGTGGAG-3´

317

CATGATGGAGACCC-3ĆCTTGTGTCCCAAG-3´

651

CTCTCCACTGAACA-3ÁGAGGCAAACACAC-3´

546

TGAGGGAAGATGAA-3´TTCCATGAACTGG-3´

952

CAGGAAACGCTCA-3ĆACCCAGGTTAGG-3´

577

AAACTTTCCTCCACA-3ÁTACAATGAAATGACA-3´

393

CACCCTTATCCTT-3´GGTAATGGGGTTG-3´

361

CATTACCCTCTGAAAA-3ÁTGAGAATAAGATTCCT-3´

397

primer; R: Reverse primer; SLC: Solute carrier.




AUC0–∞/dose/BSA [hm-5]: 21.6; range: 18.8–27.7vs 15.4; range: 6.2–38.0, p = 0.055). The exposurelevels of doxorubicinol was also higher in patientsharboring the variant 146GG genotype whencompared with patients harboring the referencegenotype (GG vs AA, AUC0–∞/dose/BSA [hm-5]:13.3; range: 8.8–21.7 vs 9.8; range 6.1–24.3,p = 0.137) or heterozygotes (GG vs AG,AUC0–∞/dose/BSA [hm-5]: 13.3; range: 8.8–21.7vs 8.98; range 3.7–20.6, p = 0.047). No signifi-cant difference in exposure levels of doxorubicinand doxorubicinol were observed for the twopatients who harbored the heterozygote geno-type for the c.755T>C polymorphism whencompared with the patients harboring the refer-ence genotype. The c.1226T>C polymorphismwas not detected in any of the cancer patients.

DiscussionGenetic polymorphisms in candidate genesinvolved in the biochemical pathway of anti-cancer agents are becoming increasingly impor-tant in understanding their disposition profiles.The majority of the polymorphic genes encodeproteins that regulate the metabolism and trans-port of drugs and are responsible for theobserved interpatient variability in thepharmacokinetics and pharmacodynamics ofchemotherapeutic agents [22]. Polymorphisms inthe coding or regulatory regions of candidategenes may also influence the tissue-specificexpression and activity of drug-metabolizingenzymes and transporter proteins resulting inaltered phenotypes. The present study wasundertaken to explore the pharmacogenetic pro-file of the SLC22A16 gene and their impact onthe pharmacokinetics of doxorubicin and doxo-rubicinol in Asian breast cancer patients. To thebest of our knowledge, this study is the first toreport the presence of novel polymorphisms inthe SLC22A16 gene and its impact on thepharmacokinetics of doxorubicin.

In the present study, screening the codingregions of the SLC22A16 gene in 400 healthysubjects of distinct ethnic groups identified four

novel polymorphisms, of which three were transi-tions (c.146A>G [exon 2], c.755T>C [exon 4]and c.1226T>C [exon 5]) that resulted in aminoacid changes. The frequencies of the c.146A>Gand c.1226C>T polymorphisms were alsoobserved to vary among the Asian ethnic groupscompared with the Caucasian population. TheChinese and the Malay ethnic groups had ahigher percentage of subjects who werehomozygous for the variant c.146A>G poly-morphism compared with the Indians and Cauca-sians. A significantly higher proportion of subjectsheterozygous for the c.1226C>T polymorphismwas present among the Indian and Caucasian eth-nic groups compared with the Chinese andMalays. The differences in allele and genotype fre-quencies observed between the Asian ethnicgroups and Caucasian population suggest thatpharmacogenetic-based analysis of SLC22A16could provide additional insights into the mecha-nisms affecting interethnic and interindividualvariations in the disposition of doxorubicin, aswell as other candidate drug substrates.

SLC22A16 has 12 predicted α-helical trans-membrane domains and one large extracellularloop between domains 1 and 2 [23]. In vitro muta-genesis studies in transporters of the SLC22Afamily have shown that domains 4, 8, 10 and 11,and possibly other additional domains, deter-mine the structure of substrate-bindingpockets [24,25]. Among the three nonsynonymouspolymorphisms identified in this study, thec.146A>G polymorphism is located in the firsttransmembrane domain and the c.755T>C andc.1226T>C polymorphisms are located in thefifth and eighth transmembrane domains, respec-tively. These are highly conserved regions amongthe organic ion transporter family, and the occur-rence of polymorphic variants in these conservedregions of the transmembrane domain may beassociated with altered SLC22A16 transportingactivity that warrants further investigations [13,15].

In vitro studies by Okabe and colleagues [16] onSLC22A16-mediated uptake of [14C]doxorubicinin Xenopus oocytes showed sodium-independent,

Table 2. SLC22A16 exon polymorphisms.

Polymorphism Exon Amino acid substitution Flanking sequence (5´–3´)

c.146A>G 2 His→Arg (p.H49R) GTCACCCCTCA/GTCATGTCTGC

c.312T>C 2 Synonymous GTAGCAGGAAT/CAAGAGGGAGA

c.755T>C 4 Val→Ala (p.V252A) TTTTTTGCAGT/CTGGAACCCTG

c.1226T>C 5 Met→Thr (p.M409T) TGCATCGCCAT/CGGACAAGGTC

SLC: Solute carrier.



572

Table 3. Genotype acancer patients.

Populations

Healthy subjects

• Chinese

• Malays

• Indians

• Caucasians

Asian cancer patients

Healthy subjects

• Chinese

• Malays

• Indians

• Caucasians


Healthy subjects

• Chinese

• Malays

• Indians

• Caucasians


Healthy subjects

• Chinese

• Malays

• Indians

• Caucasians


n: Number of subjects; SLC

saturable and dose-dependent kinetics with anapparent Km value of 5.2 ± 0.4 µM. The investi-gators also demonstrated an increase in apop-totic potential of doxorubicin in SLC22A16-overexpressing Jurkat cells, which probablyresulted from increased influx of doxorubicin.In the present study, the mean maximum con-centration of doxorubicin attained at the doseof 60 mg/m2 in the breast cancer patients was5.1 ± 2.5 µM and is within the range of intra-cellular concentrations responsible for thepharmacodynamic activity of doxorubicin [16].

This is probably suggestive of a role ofSLC22A16 in influencing the intracellularaccumulation of doxorubicin.

In the present study, patients harboring thevariant allele for the c.146A>G polymorphismhad 24 and 35% higher exposure levels of doxo-rubicin and its major metabolite doxorubicinol,respectively, in comparison with patients belong-ing to the reference genotype. The exposure lev-els of doxorubicin and doxorubicinol were also40 and 48% higher, respectively, when com-pared with the patients who were heterozygous

nd allele frequency of SLC22A16 polymorphisms in healthy subjects and Asian breast

N Genotype frequencies, n (%) Allele frequencies

c.146A>G A G

AA AG GG

100 44 (44) 43 (43) 13 (13) 0.66 0.34

100 39 (39) 43 (43) 18 (18) 0.61 0.39

100 48 (50) 45 (45) 7 (7) 0.71 0.29

100 48 (48) 43 (43) 9 (9) 0.70 0.30

62 24 (38.7) 33 (53.2) 5 (8.1) 0.65 0.35

c.312T>C T C

TT TC CC

100 43 (43) 44 (44) 13 (13) 0.65 0.35

100 39 (39) 43 (43) 18 (18) 0.61 0.39

100 48 (48) 45 (45) 7 (7) 0.71 0.29

100 45 (45) 46 (46) 9 (9) 0.68 0.32

62 24 (38.7) 33 (53.2) 5 (8.1) 0.65 0.35

c.755T>C T C

TT TC CC

100 88 (88) 10 (10) 2 (2) 0.93 0.07

100 86 (86) 14 (14) 0 (0) 0.93 0.07

100 85 (85) 15 (15) 0 (0) 0.93 0.07

100 87 (87) 13 (13) 0 (0) 0.93 0.07

62 54 (87.1) 8 (12.9) 0 (0) 0.94 0.06

c.1226T>C T C

TT TC CC

100 99 (99) 0 (0) 1 (1) 0.99 0.01

100 98 (98) 2 (2) 0 (0) 0.99 0.01

100 85 (85) 14 (14) 1 (1) 0.92 0.08

100 81 (81) 15 (15) 3 (3) 0.89 0.11

62 59 (95.2) 3 (4.8) 0 (0) 0.98 0.02

: Solute carrier.




Table 4. Summary opharmacokinetic papatients (n = 43).

Pharmacokinetic par

Doxorubicin

AUC0-∞/dose/BSA (hm-5)

Cmax/dose/BSA (m-5)

t1/2 (h)

CL/BSA (Lh-1m-2)

Vss/BSA (Lm-2)

Doxorubicinol


Cmax/dose/BSA (10–2m-5

t1/2 (h)

AUC0-∞/dose/BSA: Area und

infinity normalized by doseCL/BSA: Plasma clearance no

concentration normalized bt1/2: Half-life; Vss/BSA: Volum

surface area.

for the c.146A>G polymorphism. In concord-ance with the low frequency of the c.755T>Cand the c.1226T>C polymorphisms in thehealthy Asian subjects, only two patients carriedthe c.755T>C polymorphism, whereas thec.1226T>C polymorphism was absent in theAsian cancer patients. These results seem to sug-gest that the c.755T>C and c.1226T>C poly-morphic variants may have limited impact ondoxorubicin pharmacokinetics in Asian cancerpatients. This postulation may not be true forthe c.1226T>C polymorphism in the Caucasianpopulation, as the allele frequency of c.1226C isalmost 11-fold higher compared with theChinese and Malay populations.

The findings from this study that homo-zygosity for the SLC22A16 c.146G allele couldpredispose to increase in exposure levels of doxo-rubicin and doxorubicinol may have implicationson the antitumor activity, as well as adverse effectssuch as myelosuppression, which have establishedcorrelations with exposure levels to doxorubicin.The functional impact of the novel poly-morphisms on expression levels of SLC22A16 inorgans involved with the detoxification of doxo-rubicin or in different tumor types are notknown, which may further influence the pharma-cokinetics and pharmacodynamics of doxoru-bicin. Further correlative studies aimed atcharacterizing the functional influence of thesenovel polymorphisms in different tumors, as well

as pre- and post-treatment tumor expression ofSLC22A16 transporter proteins to correlate withefficacy and toxicity, are warranted. Studies oftransport mechanisms have shown that doxo-rubicin may permeate membranes in ways thatinclude both diffusion and carrier-mediatedmechanisms. The pharmacokinetics of doxo-rubicin could be influenced by variations in theexpression or activity of the ATP-binding cassettefamily of efflux transporters ABCB1, ABCG2,ABCB5 and also the Ral-interacting proteinRLIP76, which counts doxorubicin among itsvariety of substrates [26–30]. The polygenic influ-ence of these transporters could have contributedto the observed relatively minimal impact ofSLC22A16 polymorphisms on doxorubicin expo-sure levels in the present study. The modest sam-ple size of cancer patients recruited in the studyshould also be considered a potential limitation invalidating the influence of SLC22A16 c.146A>Gpolymorphism on doxorubicin exposure levels.

In conclusion, this study identified four novelpolymorphisms in the SLC22A16 gene in dis-tinct healthy populations of Asian and Cauca-sian origin. Exploratory investigations in Asianbreast cancer patients showed that theSLC22A16 c.146A>G variant may influence thepharmacokinetics of doxorubicin and doxorubi-cinol. Two of the four novel polymorphisms(c.755T>C and c.1226T>C) were found athigher frequencies in the Caucasian population,and their influence on the pharmacokinetics ofdoxorubicin and doxorubicinol should be inves-tigated in this population. In vitro functionalcharacterization of these novel polymorphismswith regards to SLC22A16-mediated transportof doxorubicin would be useful in understandingthe role of this influx transporter in influencingthe pharmacokinetics and pharmacodynamicsof doxorubicin and its metabolites, as well asother potential drug substrates. These in vitrofunctional studies are currently ongoing.

Future perspectiveEvolutionarily designed as protective mecha-nisms, drug transporters pose serious obstacles inoptimizing drug efficacy, overcoming chemo-resistance and modulating toxicity in cancerchemotherapy. In the future, it is envisaged thatpharmacogenetic studies of drug transporters,combined with metabolic and proteomic pro-files, will be important in elucidating the mecha-nistic basis of drug resistance and drug sensitivityand will play important roles in predicting thephenotypic outcome of treatment.

f doxorubicin and doxorubicinol rameters in Asian breast cancer

ameters Median (range)

17.0 (6.2–38.03)

32.9 (4.6–97.7)

15.6 (11.3–24.6)

24.1 (15.0–55.7)

242.3 (122.0–703.7)

9.3 (3.7–24.3)

) 0.33 (0.15–1.59)

27.3 (13.81–112.4)

er plasma concentration–time curve from time zero to

and body surface area; BSA: Body surface area; rmalized by body surface area; Cmax/dose/BSA: Peak plasma

y dose and body surface area; N: Number of subjects; e of distribution at steady-state normalized by body



574

Table 5. SLC22A16 gcancer patients (n =

Pharmacokinetic parametersc.146A>G*

Doxorubicin


t1/2 (h)

Cmax/dose/BSA (m-5)

CL/BSA (Lh-1m-2)

Vss/BSA (Lm-2)

Doxorubicinol


t1/2 (h)

Cmax/dose/BSA (m-5)

*c.312T>C influence on dop < 0.05 considered statistiBonferroni correction. AUC0-∞/dose/BSA: Area und

surface area; CL/BSA: Plasm

surface area; N: Number o

Executive summary

• Human genome sequinvolved in the mechachemotherapeutic ag

• The overlapping substdynamic interplay wit

• Functional variations iaccount for the interin

• In vitro studies on thedoxorubicin with an a

• Four novel polymorphthe SLC22A16 gene aof these polymorphism

The identification of population-specifichaplotypes will better define and consolidate theobserved genetic heterogeneity of drug transport-ers, thus increasing their clinical relevance. Inthis context, it is foreseeable that involvement ofdrug transporters and their interspecies, inteindi-vidual and interethnic variability will have a sig-nificant impact on optimizing pharmacokineticprofiles and understanding drug interactions

during drug discovery. Novel model systems thatincorporate the above-mentioned factors influ-encing drug disposition might be vital during thedrug-development pipeline.

AcknowledgementsThis study was supported by grants from Singhealth ResearchFund (SRF-SU110/2004) and Singapore Cancer Syndicate(SCS-PS0023).

enotypes and pharmacokinetics of doxorubicin and doxorubicinol in Asian breast 43, median; range).

Genotypes Pairwise p-values Overallp-value AA (n = 17) AG (n = 21) GG (n = 5) AA vs AG AA vs GG AG vs GG

17.36 (8.2–26.3)

15.4 (6.2–38.0)

21.6 (18.8–27.7)

0.649 0.066 0.055 0.123

16.88 (11.8–24.6)

14.77 (11.3–20.6)

15.12 (13.2–19.0)

0.146 0.411 0.922 0.329

36.78 (13.6–65.4)

28.21 (4.5–97.7)

40.96 (10.4–49.7)

0.355 0.906 0.626 0.638

23.66 (16.4–55.7)

26.55 (15.0–49.4)

20.43 (19.6–29.2)

0.355 0.906 0.283 0.486

276.7 (122.0–703.7)

238.0 (130.4–487.6)

227.3 (162.4–410.1)

0.692 0.557 0.626 0.461

9.84 (6.1–24.3)

8.98 (3.68–20.6)

13.31 (8.8–21.7)

0.118 0.137 0.047 0.058

26.38 (13.8–68.9)

28.57 (14.1–112.4)

29.82 (20.3–49.9)

0.783 1.000 0.922 0.965

0.29 (0.18 –0.73)

0.33 (0.15–1.09)

0.42 (0.29–1.49)

0.803 0.196 0.182 0.365

xorubicin and doxorubicinol pharmacokinetic parameters was similar to c.146A>G polymorphism. cally significant for pairwise comparisons and p < 0.017 considered statistically significant for overall comparisons after

er plasma concentration-time curve from time zero to infinity normalized by dose and body surface area; BSA: Body

a clearance normalized by body surface area; Cmax/dose/BSA: Peak plasma concentration normalized by dose and body

f subjects; SLC: Solute carrier; Vss/BSA: Volume of distribution at steady-state normalized by body surface area.

ence analysis reveals that approximately 500–1200 genes encode drug-transporter proteins, potentially nisms underlying variations in efficacy and incidence of adverse reactions of several clinically important ents.

rate affinities of the various families of cationic and anionic influx and efflux transporters suggests a h other passive and facilitated mechanisms to determine effective intracellular drug concentrations.

n genes encoding regions of drug transporters influence drug disposition in different ethnic groups and dividual variations in pharmacokinetics and pharmacodynamics.

recently characterized organic cation transporter SLC22A16 demonstrated its role in transporting pparent Km value of 5.2 ± 0.4 µM.

isms, of which three were nonsynonymous transitions, were identified by screening the exonic regions of nd occurred in relatively conserved regions, suggesting a potential functional impact. The allele frequencies s were found to differ between Asian ethnic groups and the Caucasian population.




• Higher exposure levelhomozygous SLC22Avariations in pharmac

• Future studies shouldpolymorphisms with r

Executive summary

s of doxorubicin and its major metabolite doxorubicinol were observed in patients harboring the 16 c.146GG genotype, suggesting that the polymorphism could contribute to the observed interindividual okinetics of doxorubicin.

investigate SLC22A16 polymorphisms in other ethnic groups and characterize the functional impact of the egards to doxorubicin pharmacokinetics and pharmacodynamics.

(cont.)

12. Kivisto KT, Niemi M: Influence of drug 21. Speth PA, van Hoesel QG, Haanen C:
BibliographyPapers of special note have been highlighted as either of interest (•) or of considerable interest (••) to readers.1. Venter JC, Adams MD, Myers EW et al.:
The sequence of the human genome. Science 291, 1304–1351 (2001).

2. Lander ES, Linton LM, Birren B et al.: Initial sequencing and analysis of the human genome. Nature 409, 860–921 (2001).

3. Lennernas H: Intestinal drug absorbtion and bioavailability: beyond involvement of single transport function. J. Pharm. Pharmacol. 55, 429–433 (2003).

4. Lee W, Kim RB: Transporters and renal drug elimination. Ann. Rev. Pharmacol. Toxicol. 44, 137–166 (2004).

5. Chandra P, Brouwer KL: The complexities of hepatic drug transport: current knowledge and emerging concepts. Pharm. Res. 21, 719–735 (2004).

6. Lee G, Dallas S, Hong M et al.: Drug transporters in the central nervous system: brain barriers and brain parenchyma considerations. Pharmacol. Rev. 53, 569–595 (2001).

7. Ulrike S, Wolfgang W: Reversal of ABC transporter-dependent multidrug resistance in cancer: a realistic option? Am. J. Cancer 5, 285–297 (2006).

8. Andreadis C, Gimotty PA, Wahl P et al.: Members of the glutathione and ABC-transporter families are associated with clinical outcome in patients with diffuse large B-cell lymphoma. Blood 109(8), 3409–3416 (2007).

9. Xiang X, Rao SJ, Li HH et al.: Pharmacogenetics of SLCO1B1 gene and the impact of *1b and *15 haplotypes on irinotecan disposition in Asian cancer patients. Pharmacogenet. Genomics 16, 683–691 (2006).

10. Cui Y, Konig J, Leier I et al.: Hepatic uptake of bilirubin and its conjugates by the human organic anion transporter SLC21A6. J. Biol. Chem. 276, 9626–9630 (2001).

11. Thomas J, Wang L, Clark RE et al.: Active transport of imatinib into and out of cells: implications for drug resistance. Blood 104, 3739–3745 (2004).

transporter polymorphisms on pravastatin pharmacokinetics in humans. Pharm. Res. 24, 239–247 (2007).

13. Gong S, Lu X, Xu Y et al.: Identification of OCT6 as a novel organic cation transporter preferentially expressed in hematopoietic cells and leukemias. Exp. Hematol. 30, 1162–1169 (2002).

14. Enomoto A, Wempe MF, Tsuchida H et al.: Molecular identification of a novel carnitine transporter specific to human testis. Insights into the mechanism of carnitine recognition. J. Biol. Chem. 277, 36262–36271 (2002).

15. Eraly SA, Nigam SK: Novel human cDNAs homologous to Drosophila Orct and mammalian carnitine transporters. Biochem. Biophys. Res. Commun. 297, 1159–1166 (2002).

16. Okabe M, Unno M, Harigae H et al.: Characterization of the organic cation transporter SLC22A16: a doxorubicin importer. Biochem. Biophys. Res. Commun. 333, 754–762 (2005).

• The authors demonstrate that the SLC22A16 transporter mediates the uptake of doxorubicin and determines doxorubicin transport kinetic parameters in in vitro experiments.

17. Treem WR, Stanley CA, Finegold DN et al.: Primary carnitine deficiency due to a failure of carnitine transport in kidney, muscle, and fibroblasts. N. Engl. J. Med. 319, 1331–1336 (1988).

18. Waber LJ, Valle D, Neill C et al.: Carnitine deficiency presenting as familial cardiomyopathy: a treatable defect in carnitine transport. J. Pediatr. 101, 700–705 (1982).

19. Chapoy PR, Angelini C, Brown WJ et al.: Systemic carnitine deficiency – a treatable inherited lipid-storage disease presenting as Reye’s syndrome. N. Engl. J. Med. 303, 1389–1394 (1980).

20. Elisaf M, Bairaktari E, Katopodis K et al.: Effect of L-carnitine supplementation on lipid parameters in hemodialysis patients. Am. J. Nephrol. 18, 416–421 (1998).

Clinical pharmacokinetics of doxorubicin. Clin. Pharmacokinet. 15, 5–31 (1988).

22. Kirchheiner J, Fuhr U, Brockmoller J: Pharmacogenetics based therapeutic recommendations – ready for clinical practice? Nat. Rev. Drug Discov. 4, 639–647 (2005).

23. Koepsell H, Endou H: The SLC22 drug transporter family. Pflugers Arch. 447, 666–676 (2004).

• Provides an overview on the classification and function of the solute carrier family 22 transporters.

24. Popp C, Gorboulev V, Muller TD et al.: Amino acids critical for substrate affinity of rat organic cation transporter 1 line the substrate binding region in a model derived from the tertiary structure of lactose permease. Mol. Pharmacol. 67, 1600–1611 (2005).

25. Feng B, Dresser MJ, Shu Y, Johns SJ, Giacomini KM: Arginine 454 and lysine 370 are essential for the anion specificity of the organic anion transporter, rOAT3. Biochemistry 40, 5511–5520 (2001)

26. Wu H, Hait WN, Yang J: Small interfering RNA-induced suppression of MDR1 (P-glycoprotein) restores sensitivity to multidrug-resistant cancer cells. Cancer Res. 63, 1515–1519 (2003).

27. Frank NY, Margaryan A, Huang Y et al.: ABCB5-mediated doxorubicin transport and chemoresistance in human malignant melanoma. Cancer Res. 65, 4320–4333 (2005).

28. Pratt S, Shepard RL, Kandasamy RA et al.: The multidrug resistance protein 5 (ABCC5) confers resistance to 5-fluorouracil and transports its monophosphorylated metabolites. Mol. Cancer Ther. 5, 855–863 (2005).

29. Wang X, Furukawa T, Nitanda T et al.: Breast cancer resistance protein (BCRP/ABCG2) induces cellular resistance to HIV-1 nucleoside reverse transcriptase inhibitors. Mol. Pharmacol. 63, 65–72 (2003).

30. Singhal SS, Singhal J, Nair MP et al.: Doxorubicin transport by RALBP1 and ABCG2 in lung and breast cancer. Int. J. Oncol. 30, 717–725 (2007).


novel slc22a16 polymorphisms and influence on doxorubicin...

Documents