genetic variation predicting cisplatin cytotoxicity ... · genetic variation predicting cisplatin...

12
Predictive Biomarkers and Personalized Medicine Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based Chemotherapy Xiang-Lin Tan 1,2 , Ann M. Moyer 1,4 , Brooke L. Fridley 3 , Daniel J. Schaid 3 , Nifang Niu 1 , Anthony J. Batzler 3 , Gregory D. Jenkins 3 , Ryan P. Abo 1 , Liang Li 1 , Julie M. Cunningham 4 , Zhifu Sun 2 , Ping Yang 2 , and Liewei Wang 1 Abstract Purpose: Inherited variability in the prognosis of lung cancer patients treated with platinum-based chemotherapy has been widely investigated. However, the overall contribution of genetic variation to platinum response is not well established. To identify novel candidate single nucleotide polymorphisms (SNP)/genes, we carried out a genome-wide association study (GWAS) for cisplatin cytotoxicity by using lymphoblastoid cell lines (LCL), followed by an association study of selected SNPs from the GWAS with overall survival (OS) in lung cancer patients. Experimental Design: A GWAS for cisplatin was conducted with 283 ethnically diverse LCLs. A total of 168 top SNPs were genotyped in 222 small cell lung cancer (SCLC) and 961 non-SCLC (NSCLC) patients treated with platinum-based therapy. Association of the SNPs with OS was determined by using the Cox regression model. Selected candidate genes were functionally validated by siRNA knockdown in human lung cancer cells. Results: Among 157 successfully genotyped SNPs, 9 and 10 SNPs were top SNPs associated with OS for patients with NSCLC and SCLC, respectively, although they were not significant after adjusting for multiple testing. Fifteen genes, including 7 located within 200 kb up or downstream of the 4 top SNPs and 8 genes for which expression was correlated with 3 SNPs in LCLs were selected for siRNA screening. Knockdown of DAPK3 and METTL6, for which expression levels were correlated with the rs11169748 and rs2440915 SNPs, significantly decreased cisplatin sensitivity in lung cancer cells. Conclusions: This series of clinical and complementary laboratory-based functional studies identified several candidate genes/SNPs that might help predict treatment outcomes for platinum-based therapy of lung cancer. Clin Cancer Res; 17(17); 5801–11. Ó2011 AACR. Introduction Platinum compounds are the major chemotherapeutic agents used to treat both small cell lung cancer (SCLC) and non-SCLC (NSCLC), with platinum compounds as the base for combination chemotherapy. Meta-analyses from randomized trials of cisplatin-based chemotherapy versus best supportive care have shown that cisplatin-based che- motherapy was associated with a modest improvement in overall survival (OS; ref. 1). Unfortunately, although these agents have shown success in treating lung cancer, the use of platinum-based chemotherapy is limited by the devel- opment of chemoresistance and toxicity. In addition, response rates vary greatly among individuals for any given cisplatin-based chemotherapy regimen. Factors such as disease stage, tumor histology, sex, tobacco exposure, and patient age may influence platinum-based chemother- apy outcomes. Although tumor DNA is important for drug response, previous studies have shown clearly that germ- line genetic variations also play an important role in determining treatment outcome (2–4). Therefore, identi- fication and characterization of the role of genetic variation in differential response to platinum may help optimize individual treatment plans and improve platinum treat- ment outcomes. Chemoresistance and toxicity associated with platinum- based therapy are complex phenotypes that may involve many processes (5). These processes include alterations in Authors' Affiliations: 1 Division of Clinical Pharmacology, Department of Molecular Pharmacology and Experimental Therapeutics; Divisions of 2 Epidemiology and 3 Biomedical Statistics and Informatics, Department of Health Sciences Research; and 4 Department of Laboratory Medicine and Pathology, Mayo Clinic College of Medicine, Rochester, Minnesota Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/). Corresponding Author: Liewei Wang, Division of Clinical Pharmacology, Department of Molecular Pharmacology and Experimental Therapeutics, Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN 55905. Phone: 507-284-5264; Fax: 507-284-4455; E-mail: [email protected] or [email protected] doi: 10.1158/1078-0432.CCR-11-1133 Ó2011 American Association for Cancer Research. Clinical Cancer Research www.aacrjournals.org 5801 on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Upload: others

Post on 20-May-2020

9 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

Predictive Biomarkers and Personalized Medicine

Genetic Variation Predicting Cisplatin Cytotoxicity Associatedwith Overall Survival in Lung Cancer Patients ReceivingPlatinum-Based Chemotherapy

Xiang-Lin Tan1,2, Ann M. Moyer1,4, Brooke L. Fridley3, Daniel J. Schaid3, Nifang Niu1, Anthony J. Batzler3,Gregory D. Jenkins3, Ryan P. Abo1, Liang Li1, Julie M. Cunningham4, Zhifu Sun2,Ping Yang2, and Liewei Wang1

AbstractPurpose: Inherited variability in the prognosis of lung cancer patients treated with platinum-based

chemotherapy has been widely investigated. However, the overall contribution of genetic variation to

platinum response is not well established. To identify novel candidate single nucleotide polymorphisms

(SNP)/genes, we carried out a genome-wide association study (GWAS) for cisplatin cytotoxicity by using

lymphoblastoid cell lines (LCL), followed by an association study of selected SNPs from the GWAS with

overall survival (OS) in lung cancer patients.

Experimental Design: A GWAS for cisplatin was conducted with 283 ethnically diverse LCLs. A total of

168 top SNPs were genotyped in 222 small cell lung cancer (SCLC) and 961 non-SCLC (NSCLC) patients

treated with platinum-based therapy. Association of the SNPs with OS was determined by using the Cox

regression model. Selected candidate genes were functionally validated by siRNA knockdown in human

lung cancer cells.

Results: Among 157 successfully genotyped SNPs, 9 and 10 SNPs were top SNPs associated with OS for

patients with NSCLC and SCLC, respectively, although they were not significant after adjusting for multiple

testing. Fifteen genes, including 7 located within 200 kb up or downstream of the 4 top SNPs and 8 genes

for which expression was correlated with 3 SNPs in LCLs were selected for siRNA screening. Knockdown of

DAPK3 and METTL6, for which expression levels were correlated with the rs11169748 and rs2440915

SNPs, significantly decreased cisplatin sensitivity in lung cancer cells.

Conclusions: This series of clinical and complementary laboratory-based functional studies identified

several candidate genes/SNPs that might help predict treatment outcomes for platinum-based therapy of

lung cancer. Clin Cancer Res; 17(17); 5801–11. �2011 AACR.

Introduction

Platinum compounds are the major chemotherapeuticagents used to treat both small cell lung cancer (SCLC) andnon-SCLC (NSCLC), with platinum compounds as thebase for combination chemotherapy. Meta-analyses fromrandomized trials of cisplatin-based chemotherapy versus

best supportive care have shown that cisplatin-based che-motherapy was associated with a modest improvement inoverall survival (OS; ref. 1). Unfortunately, although theseagents have shown success in treating lung cancer, the useof platinum-based chemotherapy is limited by the devel-opment of chemoresistance and toxicity. In addition,response rates vary greatly among individuals for any givencisplatin-based chemotherapy regimen. Factors such asdisease stage, tumor histology, sex, tobacco exposure,and patient age may influence platinum-based chemother-apy outcomes. Although tumor DNA is important for drugresponse, previous studies have shown clearly that germ-line genetic variations also play an important role indetermining treatment outcome (2–4). Therefore, identi-fication and characterization of the role of genetic variationin differential response to platinum may help optimizeindividual treatment plans and improve platinum treat-ment outcomes.

Chemoresistance and toxicity associated with platinum-based therapy are complex phenotypes that may involvemany processes (5). These processes include alterations in

Authors' Affiliations: 1Division of Clinical Pharmacology, Department ofMolecular Pharmacology and Experimental Therapeutics; Divisions of2Epidemiology and 3Biomedical Statistics and Informatics, Departmentof Health Sciences Research; and 4Department of Laboratory Medicineand Pathology, Mayo Clinic College of Medicine, Rochester, Minnesota

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Liewei Wang, Division of Clinical Pharmacology,Department of Molecular Pharmacology and Experimental Therapeutics,Mayo Clinic College of Medicine, 200 First Street SW, Rochester, MN55905. Phone: 507-284-5264; Fax: 507-284-4455; E-mail:[email protected] or [email protected]

doi: 10.1158/1078-0432.CCR-11-1133

�2011 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 5801

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 2: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

drug influx or efflux, detoxification through glutathioneconjugation, DNA repair capacity, and other cellular path-ways required for proper response to DNA damage (6). Thedevelopment of chemoresistance to platinum agents can bedue to germline genetic variation or acquired mutationthrough their effects on mRNA and protein levels in keypharmacokinetic or pharmacodynamic pathways. Candi-date gene and/or candidate pathway approaches indicatedthat single nucleotide polymorphisms (SNP) in genesresponsible for drug influx and efflux (7), metabolismand detoxification (8–10), and DNA damage repair path-ways (2–4, 11–17) were associated with treatment outcomeof lung cancerpatients.However, theoverall contributionofthese genetic biomarkers in the prediction of response toplatinum-based therapy is still not well established, suggest-ing that additional genes might be involved.

Therefore, in this study, we took a genome-wideapproach by using 283 lymphoblastoid cell lines (LCL)to identify SNPs/genes that might contribute to variationin cisplatin cytotoxicity as represented by IC50 values. Thisstep was followed by genotyping top candidate SNPsfrom the LCL genome-wide association study (GWAS)by using 1,183 DNA samples from lung cancer patientstreated with platinum-based therapy to identify SNPs thatmight be associated with OS. Finally, functional studies ofcandidate genes identified during the association studieswere carried out by using siRNA knockdown in humanlung cancer cells.

Materials and Methods

Cell linesThis study was reviewed and approved by the Mayo

Clinic Institutional Review Board. Human Variation PanelLCLs from sample sets HD100AA, HD100CAU,HD100CHI, corresponding to 100 African–American(AA), 100 Caucasian–American (CA), and 100 Han Chi-nese–American (HCA) unrelated subjects, were obtained

from the Coriell Cell Repository. The National Institute ofGeneral Medical Sciences had obtained and anonymizedthese cell lines before deposit, and all subjects had providedwritten informed consent for the use of their samples forresearch purposes. Human NSCLC cell lines H1437,H1299, and the SCLC cell line H196 were obtained fromthe American Type Culture Collection. LCLs were culturedin RPMI 1640 medium (Mediatech) supplemented with15% FBS (Mediatech). H1437, H1299, and H196 cell lineswere cultured in RPMI 1640 medium containing 10% FBS.

Patient samplesThe study group included 1,183 Caucasian patients with

pathologically confirmed primary lung cancer, 222 SCLCand 961 NSCLC, who were treated with platinum-basedchemotherapy at the Mayo Clinic between 1997 and 2008.Details with regard to clinical characteristics of patients,patient enrollment, diagnosis, and data collection proce-dures were described previously (18–20). Briefly, eachpatient was identified through the Mayo Clinic pathologicdatabase. After written informed consent had beenobtained, a blood sample was collected. The characteristicsof patients, including demographics, lung cancer pathol-ogy, anatomic site, and types and timing of treatment andchemotherapeutic agents, were abstracted from patientmedical records by a trained nurse. Vital status and causeof death were determined by reviewing the Mayo Clinicregistration database and medical records, correspondencefrom patients’ next-of-kin, death certificates, obituarydocuments, theMayo Clinic Tumor Registry, and the SocialSecurity Death Index website. Clinical staging and recur-rence or progression information was determined byresults from available chest radiography, computerizedtomography, bone scans, positron emission tomographyscans, and magnetic response imaging. All patients wereactively followed up during the initial 6 months afterdiagnosis, with subsequent annual follow-up by mailedquestionnaires and annual verification of the patients’ vitalstatus.

Cisplatin cytotoxicity assayCisplatin was obtained from Sigma-Aldrich and was

dissolved in dimethyl sulfoxide (DMSO) immediatelybefore use. Cells (4 � 104 cells per well) were plated into96-well plates and incubated with cisplatin for 72 hours in8 concentrations ranging from 0.1 to 80 mmol/L. DMSOalone was used as a control. Cisplatin cytotoxicity wasevaluated by determining the concentration of cisplatinrequired to inhibit growth and/or survival by 50% (IC50)by using the CellTiter 96@ Aqueous Non-Radioactive CellProliferation Assay (Promega). Experiments were con-ducted successfully for 283 LCLs (91 AA, 96 CA, and 96HCA).

Genome-wide SNP and expression arrays for LCLsThe genotyping and expression array data for all

LCLs were described previously (21–23) and are pub-lically available from NCBI Gene Expression Omnibus

Translational Relevance

Chemotherapy with platinum-based regimens is thestandard of care for both small cell and non–small celllung cancer, but large interindividual variations areobserved in platinum efficacy and toxicity. However,reliable genetic biomarkers for the prediction ofresponse to platinum-based therapy are still not wellestablished. Our study identified several genetic variantsthat are potentially associated with the efficacy of pla-tinum-based chemotherapy in patients with lung cancerby using genome-wide analysis of data generated withlymphoblastoid cell lines, followed by genotyping stu-dies using DNA samples from lung cancer patients andfunctional validation in lung cancer cells. These resultsprovide important insight into mechanisms that mightcontribute to variation in response to platinum-basedtherapy of lung cancer.

Tan et al.

Clin Cancer Res; 17(17) September 1, 2011 Clinical Cancer Research5802

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 3: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

(http://www.ncbi.nlm.nih.gov/geo) under SuperSeriesaccession No. GSE24277 and accession No. GSE23120.Briefly, Illumina HumanHap 550K and 510S BeadChipswere used for genotyping of DNA samples from all of theLCLs in the Genomics Shared Resource at Mayo Clinic.Publicly available Affymetrix SNP Array 6.0 Chip SNP datawere alsoobtained for the samecell lines. SNPs thatdeviatedfrom Hardy–Weinberg Equilibrium (HWE; ref. 24) basedon minimum P value from an exact test for HWE (25) andthe stratified test forHWE (ref. 26; P < 0.001); SNPswith callrates less than 95%; or SNPs with minor allele frequency(MAF) less than 5% were removed from the analysis (27,28). As a result, 1,348,798 SNPs that passed these qualitycontrolmeasureswere included in theGWAanalyses of SNPversus IC50 and SNP versus expression.

SNP selection and genotyping in lung cancer patientsSNP selection was based on P values obtained during the

analysis of genome-wide SNP versus IC50, and SNP versusexpression, as well as expression versus IC50 by using LCLdata. We selected all SNPs with P < 10�5 from the SNPversus IC50 analysis. In addition, we defined a SNP peak asa locus that contained at least 2 SNPs associated withcisplatin IC50 with P < 10�4. SNPs within the SNP peakswere used to carry out an association study with 54,613expression probesets to identify SNPs that were associatedwith gene expression (P < 10�4). Finally an associationstudy was conducted with gene expression and cisplatinIC50 to identify SNPs that might be associated with cispla-tin IC50 through an influence on gene expression. There-fore, we also selected SNPs with SNP versus expression: P <10�4 and expression versus IC50: P < 10�4 to genotype thepatient samples (Supplementary Table S1).A total of 168 top hit SNPs were selected and genotyped

in 1,183 lung cancer patients who received platinum-basedchemotherapy. Genotyping was done in the Mayo ClinicGenomics Shared Resource by using a custom-designedIllumina GoldenGate panel. Quality control tests of thegenotyping results were done by assessing concordanceamong 3 control DNA samples that were present in dupli-cate, SNP call rates, sample call rates, MAFs of SNPs ordeparture of SNP genotypes from HWE. SNPs wereexcluded if they failed genotyping or displayed ambiguousclustering, monomorphic genotyping, MAFs of less than0.01, or significant departures fromHWE (P < 0.001). SNPshaving call rates more than 95% but which passed all otherquality control tests were included in the analyses. Of theSNPs with genotyping data, 157 SNPs passed the qualitycontrol tests and were included in the analyses.

Functional validation by siRNA knockdownHuman NSCLC cell lines, H1437 and H1299 and the

SCLC cell line, H196, were used in siRNA screening studies.siRNAs for the candidate genes and nontargeting negativecontrol siRNA pool were purchased from Dharmacon.Specifically, approximately 3,000 to 4,000 cells were seededinto 96-well plates,mixedwith an siRNAmixture consistingof 25 nmol/L of specific or negative control siRNAs and the

LipofectamineRNAiMAX reagent (Invitrogen). Twenty-fourhours after transfection, the cellswere treatedwith vehicle orincreasing concentrations of cisplatin for an additional 72hours, followed by MTS assay by using the CellTiter 96@Aqueous Non-Radioactive Cell Proliferation Assay.

Knockdown efficiency determination by real-time RT-PCR and Western blot analysis

Total RNA was isolated from cultured cells transfectedwith controls or specific siRNAs with the Mini RNA isola-tion kit (ZYMO Research), followed by quantitative reversetranscriptase PCR (qRT-PCR) conducted with the 1-step,Brilliant SYBR Green qRT-PCR master mix kit (Stratagene).Specifically, primers purchased fromQIAGEN were used toconducted qRT-PCR by using the ABI StepOne Real-TimePCR System (Applied Biosystems). Western blots weredone for DAPK3, METTL6, and RUFY1 by using goatpolyclonal antibodies purchased from Santa Cruz biotech-nology, Inc. All experiments were conducted in triplicatewith b-actin as an internal control.

Statistical analysisA detailed description for GWAS of the LCLs has been

described elsewhere (21–23). Briefly, the cisplatin cytotoxi-city IC50 phenotype was calculated for each cell line on thebasis of a logistic dose–response model by using the Rpackage "drc" (http://cran.r-project.org/web/packages/drc.pdf). Log-transformed IC50 values were then comparedbetween genders, batches of samples on the basis of timesince purchase and between the CA race and all othersamples by using independent samples t tests. An overallcomparison of transformed IC50 values among ethnicgroups was done by using an F test on the basis of ANOVA.Because we used LCLs from multiple races/ethnic groups,population stratification was adjusted by using the methoddeveloped by Price and colleagues (29), which uses an eigenanalysis for detecting and adjusting SNPs. The log-trans-formed IC50 values and GCRMA-normalized expressiondata were adjusted for race by using 5 eigen vectors. TheGWA analysis of the association of SNP versus IC50 or SNPversus expression was done with Pearson correlations byusing adjusted SNP, IC50, and expression values for eachindividual. For the siRNA knockdown experiments, groupmean values of IC50were compared by using Student’s t test.

The OS time was used as the primary endpoint, definedas the time from lung cancer diagnosis to either death orthe last known date alive. Patients known to be alivewere censored at the time of last contact. All patientsincluded in the analyses were Caucasians. To test forthe effect of SNP on OS, we used the Cox regressionmodel that included the effects of SNP genotype dosage(count of minor alleles). A total of 157 SNPs were includedin this analysis, and the association was done for NSCLCand SCLC separately because of significant differencesbetween the 2 diseases. To correct for multiple testing ofthe 157 SNPs assayed in the initial experiment, theBonferroni corrected P value threshold of 0.0003 wasused to determine statistically significant associations. To

Genetic Variation and Platinum Response in Lung Cancer

www.aacrjournals.org Clin Cancer Res; 17(17) September 1, 2011 5803

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 4: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

determine whether associations with SNPs should beadjusted for the clinical covariates of age at diagnosis,gender, smoking status, disease stage, and treatment, back-ward selection was done. The disease stage was included inthe final multivariate Cox regression model because it wassignificantly associated with OS of the lung cancer patients.The disease stage was divided into 5 categories: SCLC withlimited versus extensive stages; NSCLC with stages I and II,versus III, versus IV. We used 0.05 as a cutoff for P values(not adjusted for multiple testing) to select SNPs/genes forfurther functional validation.

Results

Correlation analysis of genome-wide SNP versuscisplatin IC50 in LCLs

Cytotoxicity assays were done for cisplatin to determinethe range of variation in drug response, and IC50 value wasused as a phenotype to indicate the drug sensitivity for eachcell line. The unadjusted average IC50 value for cisplatin inthese 283 cell lines was 1.79 � 1.64 (mean � SD) mmol/L.Gender had no significant effect on IC50 values (P ¼ 0.15).Caucasian-Americans (CA) were more sensitive to cispla-tin compared with the other 2 races (AAs and HCAs;P ¼ 0.0001). Time since the Coriell Institute acquired thecell lines had no significant effect on cisplatin cytotoxicity(P ¼ 0.64).

A total of 1,348,789 SNPS were used in the genome-wideSNP analysis for the 283 cell lines. Figure 1A depicts theassociation results graphically. A total of 364 SNPs wereassociated with cisplatin IC50 with P < 10�4 and 50 SNPshad P < 10�5. None of the top 50 SNPs were within the

coding region of a gene. Twenty-three, 16, and 11 SNPswere in introns, 50-upstream or 30-downstream flankingregions of genes, respectively. The top 3 SNPs from theGWAS were rs6633130 (P¼ 1.66� 10�7), rs6946197 (P¼1.99 � 10�7), and rs12304656 (P ¼ 3.64 � 10�7) at locicontaining the PPEF1 (intron), LOC100128030 (30-down-stream) and STYK1 (intron) genes, respectively.

Survival analysis for the selected SNPs and lungcancer patient samples

A total of 168 SNPs were selected and genotyped todetermine whether any of those SNPs might be associatedwith patient OS. Table 1 lists basic demographic andclinic data with regard to the patients studied. Of thesepatients, 656 (55.5%) were male and 527 (44.5%) werefemale, with a median age of 63.0 years. Most patientshad a history of smoking, with 47.3% (n ¼ 560) beingformer smokers and 36.5% (n ¼ 431) current smokers. Atotal of 550 (46.5%) tumors were classified as adenocar-cinoma, 194 (16.4%) as squamous cell carcinoma, and222 (18.8%) as small cell carcinoma. A total of 881(74.5%) patients received platinum drugs in combinationwith surgery and/or radiation and 302 (25.5%) patientsreceived only platinum drugs.

Only disease stage was included in the analysis as aconfounder among all the others tested, including age atdiagnosis, gender, and smoking status. Nine and 10different SNPs were associated with OS for NSCLC andSCLC patients, respectively (P < 0.05; Table 2 andSupplementary Fig. S1). However none of the SNPs werestatistically significant after correcting for multiple testing.The most significant SNPs, rs1287276 (located within an

−Log

10(P

)

Chromosome location

1

6

5

4

3

2

1

0

2 3 4 5 6 7 8 9

10

11

13

15

X o

r Y17

19

22

19 SNPs associated with OS in clinical study (P < 0.05)

Gene expression level

>50 in LCLs (7 genes)Gene expression level

>50 in LCLs (8 genes)

15 Candidate genes selected for functional validation

Genes within 200 kb up or downstream

of the 12 SNPs (39 genes)SNP versus expression in LCLs with P <10–5

or P <10–4 with ≥ 2 probe sets (56 genes)

12 SNPs concordantly associated with IC50 and OS phenotype

A

B

Figure 1. Genome-wide SNP association with cisplatin IC50

values (A) and the strategies used to select candidate genes forsiRNA validation (B). Genome-wide SNP association analysis wasdone with cisplatin IC50 as the drug-response phenotype(n ¼ 283). The y-axis represents the �log10(P) for association ofSNPs and IC50 values. SNPs are plotted on the x-axis on the basisof their chromosomal locations. The cut-off P value of 10�5 ishighlightedwith a black line. Detailed strategies for the selection ofcandidate genes have been described in the text.

Tan et al.

Clin Cancer Res; 17(17) September 1, 2011 Clinical Cancer Research5804

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 5: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

intron of UGT3A2 on chromosome 5) and rs12669805(located 30-downstream of the LOC100128030 pseudogeneon chromosome 7) were associated withOS inNSCLC (P¼0.0004) and SCLC (P ¼ 0.0021), respectively. A HR > 1indicated that patients carrying the minor allele had worsesurvival. These SNPswould alsobe expected tobe associated

with higher IC50 values in LCLs, and the cells carrying theseSNPs would be predicted to be more resistant to cisplatin.Therefore, 12 of the 19 SNPs showed concordant asso-ciation directions between the 2 results (Table 2) whenwe compared the association direction of SNPswith clinicalOS with that of the same SNP with cisplatin IC50 valuesin LCLs.

Functional characterization of candidate geneAlthough none of the SNPs studied in the patient

samples reached statistical significance after multipletesting, to further pursue the possibility that the locusharboring these SNPs might include functional genes, wetook advantage of the expression data we had for the celllines (21–23) to help interpret SNP function and guidethe selection of genes for siRNA screening. The overallselection strategy is depicted in Figure 1B. First, wefocused on genes located within 200 kb up or down-stream of the 12 SNPs which showed a concordantassociation with the IC50 phenotype in LCLs and theOS phenotype in patients. A total of 39 genes were foundto be located within 200 kb up or downstream of theseSNPs, and 7 of the 39 genes close to 4 SNPs had expres-sion levels greater than 50 in LCLs after GCRMA normal-ization (Fig. 1B). Next, because SNPs might influencecisplatin response through the regulation of gene expres-sion by either cis- or trans-regulation, we also carried outassociation analysis by using the genotyping data forthese 12 SNPs and basal expression array data (54,613expression probesets) available for all of the LCLs. Eightof 12 SNPs were associated with the expression of 212probesets (154 unique genes) using a cutoff P < 10�4, and4 of 12 SNPs were associated with mRNA expressionlevels of 54 genes with P < 10�5 (Fig. 1B). No cis-regula-tion was identified. We selected these 54 genes, plus 2genes with multiple probesets associated with the SNPs tocheck expression levels in LCLs. However, only 8 of thesegenes had expression levels more than 50 after GCRMAnormalization. Finally, 15 genes, including 7 geneslocated within 200 kb up or downstream of 4 SNPs(rs11169748, rs2440915, rs5952066, and rs7620841)and 8 genes for which expression was correlated with 3SNPs (rs11169748, rs2440915, and rs5952066) wereselected for siRNA screening (Fig. 1B).

We conducted knockdown experiments for these 15genes in the NSCLC cell lines H1437 and H1299 and inthe SCLC cell line H196. Knockdown of DAPK3 andMETTL6 significantly decreased cisplatin sensitivity inall 3 lung cancer cell lines, whereas knockdown of RUFY1did not significantly alter cisplatin cytotoxicity (Fig. 2).The rs11169748 and rs2440915 SNPs that were asso-ciated with lung cancer OS were also correlated withcisplatin IC50 values in LCLs. In addition, these 2 SNPswere also associated with expression of DAPK3 andMETTL6 in LCLs (Fig. 3). However, only DAPK3 expres-sion level was associated with cisplatin IC50 (P ¼ 0.03) inLCLs, whereas METTL6 was not (P ¼ 0.32). To obtainungenotyped SNPs, we imputed SNPs that were not on

Table 1.Description of 1,183 patients whowerediagnosed with primary lung cancer andreceived platinum-based chemotherapy

Characteristics of diagnosisand treatment

Values andpercentages

Age at diagnosisMean (SD) 62.3 (10.34)Median (range) 63.0 (27.0–88.0)

Gender, n (%)Female 527 (44.5%)Male 656 (55.5%)

Cigarette smoking statusNever 192 (16.2%)Former smokers 560 (47.3%)Current smokers 431 (36.5%)

Lung cancer stagea

UnknownSCLCLimited 129 (10.9%)Extensive 93 (7.9%)

NSCLCI 123 (10.4%)II 82 (6.9%)III 403 (34.1%)IV 353 (29.8%)

Histologic cell typeAdenocarcinoma 550 (46.5%)Squamous cell carcinoma 194 (16.4%)Small cell carcinoma 222 (18.8%)Large cell carcinoma 39 (3.3%)Mixed and unspecified NSCLC 167 (14.1%)Carcinoid or salivary gland tumors 11 (0.9%)

Tumor differentiation gradeNongradable or unknown 80 (6.8%)Well differentiated 91 (7.7%)Moderately differentiated 392 (33.1%)Poor/undifferentiated 620 (52.4)

Treatment modalityOnly platinum drugs 302 (25.5%)Surgery and platinum drugs 192 (16.2%)Radiation and platinum drugs 455 (38.5%)Surgery, radiation, andplatinum drugs

234 (19.8%)

aStage for NSCLC is described on the basis of the tumor-node-metastasis classification. The small cell lung cancerstage is described as suggested by the American CancerSociety as either "extensive" or "limited."

Genetic Variation and Platinum Response in Lung Cancer

www.aacrjournals.org Clin Cancer Res; 17(17) September 1, 2011 5805

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 6: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

our genotyping platform and that were located within200 kb up or downstream of these 2 SNPs (rs11169748and rs2440915), using MACH 1.0 (30), with 1000 Gen-ome Project data (31) as the reference panel. We observedno stronger association with cisplatin IC50 for theimputed SNPs, compared with the observed SNPs(Fig. 3E and F).

Discussion

In this study, we took advantage of the extensivegenomic data that we have already obtained for 283human LCLs, together with DNA samples obtained fromlung cancer patients treated with platinum-based therapyto determine whether SNPs identified during a GWASconducted with LCLs might contribute to our understand-ing of the drug actions of platinum compounds andthe response to platinum compounds in the treatmentof lung cancer. Therefore, we first conducted a GWAS forcisplatin in 283 LCLs to identify top candidate SNPs thatwere associated with platinum cytotoxicity in vitro (Fig. 1and Supplementary Table S1). We then conducted anassociation study with selected top candidate SNPs byusing DNA samples obtained from NSCLC and SCLC

patients (Table 2), followed by functional validation ofcandidate genes by using human lung cancer celllines (Fig. 2 and Table 3). We found that 19 SNPswere associated with OS with P < 0.05 for either NSCLCor SCLC patients (Table 2) and that knockdown ofDAPK3 and METTL6, for which expression levels werecorrelated with the rs11169748 and rs2440915 SNPs(Fig. 3), significantly decreased cisplatin sensitivity inthe lung cancer cell lines H1437, H1299, and H196(Fig. 2).

Previous candidate gene and candidate pathway–basedpharmacogenomic analyses have shown that genetic varia-tion may play a role in determining response to platinum-based chemotherapy. Much of the focus has been on SNPswithin the DNA damage and repair pathway (2–4, 11–17,32). However, those results remain controversial. Forexample, Zhou and colleagues reported that the ERCC18092C>A variant allele was associated with a 50%increased risk of death in stage III (AþB) and IV NSCLCpatients (32), whereasWu and colleagues observed that thevariant A allele was significantly associated with a shorterOS (HR ¼ 0.68, 95% CI: 0.48–0.95; ref. 3). These previousstudies indicate that the effects of each individual SNP aremodest and suggest that additional SNPs/genes and/or

Table 2. The 19 SNPs with the lowest unadjusted P values that were associated with the OS of lung cancerpatients treated with platinum compounds

LCL GWAS Clinical Study

SNP Chr. SNP position MAF Ra P MAF HR (95% CI)b P

Among NSCLC patientsrs1287276 5 36083458 0.242 �0.247 4.72E-05 0.095 1.36 (1.15–1.61) 0.0004rs9405302c 6 8228049 0.394 0.267 1.08E-05 0.211 1.22 (1.08–1.37) 0.001rs8064630 17 3789240 0.152 0.251 3.61E-05 0.187 0.82 (0.71–0.94) 0.004rs5970160c 23 150844654 0.17 0.259 2.25E-05 0.126 1.17 (1.05–1.31) 0.006rs1946518 11 111540668 0.384 0.249 5.55E-05 0.396 0.87 (0.78–0.96) 0.008rs2440915c 10 61343778 0.058 0.249 4.10E-05 0.029 1.41 (1.08–1.83) 0.012rs5952066c 23 146692840 0.178 0.242 6.92E-05 0.200 1.14 (1.03–1.26) 0.014rs901893c 5 8228049 0.353 0.239 8.53E-05 0.337 1.14 (1.02–1.27) 0.018rs11169748c 12 49865438 0.062 0.237 9.60E-05 0.011 1.75 (1.03–2.97) 0.039Among SCLC patientsrs12669805c 7 79509055 0.188 0.253 3.32E-05 0.024 3.42 (1.56–7.48) 0.002rs6092092c 20 53291685 0.076 0.249 4.28E-05 0.006 7.74 (1.85–32.30) 0.005rs2833537 21 32135333 0.074 0.254 3.03E-05 0.036 0.40 (0.20–0.79) 0.008rs7620841c 3 187425014 0.06 0.242 6.77E-05 0.052 1.80 (1.12–2.89) 0.015rs1921626c 2 77025574 0.205 0.238 9.00E-05 0.084 1.57 (1.09–2.26) 0.015rs17043088 12 76103718 0.104 �0.238 8.88E-05 0.026 2.21 (1.16–4.22) 0.016rs753921c 4 35579894 0.231 0.239 8.69E-05 0.274 1.31 (1.05–1.64) 0.018rs7618373 3 177819713 0.446 �0.255 3.17E-05 0.378 1.27 (1.02–1.57) 0.030rs5916072 23 5406196 0.343 0.247 5.04E-05 0.442 0.84 (0.70–1.00) 0.046rs10517364c 4 35562853 0.246 0.258 2.12E-05 0.310 1.23 (1.00–1.52) 0.050

aR values represent correlation coefficients for the associations between SNPs and cisplatin IC50.bIf HR > 1, patients carrying the minor allele had worse OS.cSNP showed a concordant association with the IC50 and OS phenotypes.

Tan et al.

Clin Cancer Res; 17(17) September 1, 2011 Clinical Cancer Research5806

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 7: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

other possible genetic mechanisms such as CpG methyla-tion or copy number variation might influence response toplatinum-based therapy.With advances in genotyping and gene expression

technologies, several studies have taken an unbiasedapproach toward understanding genetic factors influen-cing response to platinum-based chemotherapy. In LCLsobtained from pedigrees of Centr d’Etudes du poly-

morphism Human (CEPH) individuals of Europeanbackground, it was estimated that 30% to 40% of thevariation in cisplatin-induced cytotoxicity was because ofheritable factors (33, 34). By using this cell line modelsystem, those investigators subsequently identified sev-eral genetic variants that contributed to cisplatin andcarboplatin cytotoxicity (24, 34–36). However, the resultsof those studies have not been replicated in cancer

A B C

DAPK352 kDa

METTL630 kDa

RUFY180 kDa

β-Actin43 kDa

Specific siRNA Negative siRNA

E

H1437

H1299

H196H1437

H1299

H196

–0.5 0.0 0.5 1.0 1.5 2.00.0

0.2

0.4

0.6

0.8

1.0

1.2

DAPK3 siRNAIC50 (95% CI): 16.0 (15.3–16.6)

Negative siRNAIC50 (95% CI): 12.3 (11.0–13.8)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

–0.5 0.0 0.5 1.0 1.5 2.00.0

0.2

0.4

0.6

0.8

1.0

1.2

METTL6 siRNAIC50 (95% CI): 14.9 (13.5–16.5)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

–0.5 0.0 0.5 1.0 1.5 2.0–0.1

0.1

0.3

0.5

0.7

0.9

1.1

DAPK3 siRNAIC50 (95% CI): 26.7 (23.8–29.8)

Negative siRNAIC50 (95% CI): 15.9 (13.6–18.7)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

–0.5 0.0 0.5 1.0 1.5 2.0–0.1

0.1

0.3

0.5

0.7

0.9

1.1

METTL6 siRNAIC50 (95% CI): 33.9 (26.3–43.7)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

–0.5 0.0 0.5 1.0 1.5 2.0–0.1

0.1

0.3

0.5

0.7

0.9

1.1

RUFY1 siRNAIC50 (95% CI): 15.9 (13.7–18.5)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

–0.5 0.0 0.5 1.0 1.5 2.00.0

0.2

0.4

0.6

0.8

1.0

1.2DAPK3 siRNAIC50 (95% CI): 8.1 (7.4–8.8)

Negative siRNAIC50 (95% CI): 6.1 (5.6–6.7)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

–0.5 0.0 0.5 1.0 1.5 2.00.00

0.25

0.50

0.75

1.00

METTL6 siRNAIC50 (95% CI): 8.4 (7.9–9.0)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

–0.5 0.0 0.5 1.0 1.5 2.00.0

0.2

0.4

0.6

0.8

1.0

RUFY1 siRNAIC50 (95% CI): 6.0 (5.6–6.4)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

–0.5 0.0 0.5 1.0 1.5 2.00.0

0.2

0.4

0.6

0.8

1.0

RUFY1 siRNAIC50 (95% CI): 11.6 (10.9–12.5)

Cisplatin concentration, Log (µmol/L)

Cell

surv

ival

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Negative

siRNA

DAPK3

siRNA

METTL6

siRNA

RUFY1

siRNA

Re

lative

mR

NA

exp

ressio

n

H1437H1299H196

D

Figure 2. Functional characterizations of candidate genes with siRNA knockdown of DAPK3 and METTL6. Knockdown of DAPK3 and METTL6 in humanNSCLC cell lines, H1437 (A), H1299 (B), and the SCLC cell line, H196 (C), showed increased resistance to cisplatin. A control was also includedin which knockdown of RUFY1 did not significantly change the response to cisplatin as determined by MTS assay. Quantitative RT-PCR (D) and Western blot(E) were done to determine the knockdown efficiency for the 3 genes. RT-PCR results are expressed as % of control. Error bars represent SEM valuesfor 3 independent experiments.

Genetic Variation and Platinum Response in Lung Cancer

www.aacrjournals.org Clin Cancer Res; 17(17) September 1, 2011 5807

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 8: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

patients treated with platinum-based chemotherapeutics.Our genome-wide pharmacogenomics analysis did notconfirm those variants associated with response to cis-platin (24). The difference could be due to differences

between populations from which these LCLs are derived.In Huang’s study, the LCLs derived from CEU, Europeandescent and YRI, African descent, whereas our LCLsderived from CA, AA, and HCA. It could also be because

A B

C D

E F

Ad

juste

d I

C50

2.0

1.5

1.0

0.5

0.0

–0.5

–1.0

Ad

juste

d I

C50

2.0

1.5

1.0

0.5

0.0

–0.5

–1.0

Ad

juste

d D

AP

K3

mR

NA

exp

ressio

n (

Log)

15

10

5

0

Rs1169748 AA AG GG AA AG GG AA AG GG AA AG GGNumber 63 25 3 94 2 0 94 2 0 251 29 3Ethnicity AA CA HCA Total

Ad

juste

d M

ET

TL6

mR

NA

exp

ressio

n (

Log)

12

10

8

6

4

2

0

Rs2440915 AA AG GG AA AG GG AA AG GG AA AG GGNumber 63 25 3 94 2 0 96 0 0 253 27 3Ethnicity AA CA HCA Total

Rs1169748 AA AG GG AA AG GG AA AG GG AA AG GGNumber 63 25 3 94 2 0 94 2 0 251 29 3Ethnicity AA CA HCA Total

Imputed SNPGenotyped SNP

-Lo

g10

(P)

5

4

3

2

1

0

5

4

3

2

1

0

-Lo

g10

(P)

49700 49800 49900 50000Position (kb)

61200 61300 61400 615500Position (kb)

Imputed SNPGenotyped SNPrs1169748 rs2440915

Rs2440915 AA AG GG AA AG GG AA AG GG AA AG GGNumber 63 25 3 94 2 0 96 0 0 253 27 3Ethnicity AA CA HCA Total

Figure 3. Association between the rs1169748 and rs2440915 SNPs and cisplatin IC50 values and expression of DAPK3 and METTL6 in LCLs, as wellas imputation analyses for these 2 SNPs. Genotypes for each SNP are plotted against cisplatin IC50 values as well as gene expression levels ofDAPK3 and METTL6, respectively. A and C, SNP rs1169748 association with cisplatin IC50 values and DAPK3 expression; B and D, SNP rs2440915association with cisplatin IC50 values and METTL6 expression; E and F, imputation analysis for the rs1169748 and rs2440915 SNPs. Black circlesindicate SNPs observed by genotyping, whereas black triangles indicate imputed SNPs. The y-axis represents �log10(P) for the association of each SNPwith cisplatin IC50, and the x-axis represents the chromosomal locations of the SNPs. No stronger association with cisplatin IC50 were observed for theimputed SNPs, compared with the genotyped SNPs. AA, African-Americans; CA, Caucasian-Americans; and HCA, Han Chinese-Americans.

Tan et al.

Clin Cancer Res; 17(17) September 1, 2011 Clinical Cancer Research5808

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 9: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

of confounding factors associated with the use of LCLs,such as cell growth rate and ATP levels (37). Therefore, wecarried out siRNA screening for selected candidate genesto pursue the underlying biology. Several previous studieshave suggested an important contribution of trans-actingvariants to individual variation in human gene expression(38, 39). Among the genes tested, knockdown of DAPK3and METTL6, for which expression levels correlated withthe rs11169748 and rs2440915 SNPs, significantlydecreased cisplatin sensitivity in 3 lung cancer cell lines,althoughMETTL6 expression levels were not significantlyassociated with cisplatin IC50 in LCLs, suggesting apotential difference between LCLs and lung cancer celllines.DAPK3, also known as zipper interacting protein kinase

(ZIPK) or DAPK-like kinase (DLK), is a member of theDAPK family. DAPK was initially identified as beingencoded by a gene whose reduced expression, mediatedby antisense cDNA transfection, protected HeLa cells fromgamma interferon-induced cell death. DAPK3 containsseveral putative nuclear localization signal sequencesand a leucine zipper domain, required for homo-oligo-merization, interaction with other leucine zipper-contain-ing proteins, and for its death-promoting effects (40, 41).DAPK3 has been shown to play distinct roles in tumorsuppression and the regulation of apoptosis (42–44). Wehave shown that knockdown of DAPK3 significantlydecreased cisplatin sensitivity in lung cancer cell lines

and might be related to OS of NSCLC patients treatedwith platinum-based therapy.

METTL6 belongs to the methyltransferase superfamily. Ithas been suggested that chemoresistance and toxicity asso-ciated with platinum compound treatment might bemediated through methylation-dependent gene silencing.For example, methylation of FANCF, a gene associated withFanconi anemia, confers cisplatin sensitivity in ovariancancer cell lines (45). In addition, Ramirez and colleaguesobserved that 14-3-3s, a cell-cycle checkpoint gene, ismethylated in one-third of NSCLC patients and thatmethy-lation is related to better medianOS for these patients (46).It is possible that METTL6 might influence DNA methyla-tion and further influence response to platinum-basedtherapy.

In summary, we have identified several genetic variantsthat are potentially associated with the efficacy of plati-num-based chemotherapy in patients with lung cancer byusing genome-wide interrogation of LCLs together withgenotyping studies by using DNA samples from lungcancer patients, followed by functional validation in lungcancer cells. Our results provide important insight intonovel genes and mechanisms that may contribute tovariation in response to platinum drug therapy. Mostof our patients with advanced NSCLC were treated withcisplatin combination therapy, often with the addition ofother chemotherapeutics, radiation, or surgery. Eventhough the SNPs that we tested were selected on the

Table 3. Fifteen candidate genes selected for functional validation

Basis for selectionSNP versus expression

in LCLs

Gene SNP

Distance fromthe SNP(< 200 kb)

Expressionassociated withthe SNP in LCLs

R P

DAZAP2 rs11169748 Yes NAa NATFCP2 rs11169748 Yes NA NALETMD1 rs11169748 Yes NA NAGALNT6 rs11169748 Yes NA NADAPK3 rs11169748 Yes 0.293 1.05E-06EML3 rs11169748 Yes 0.310 2.22E-07LRIG1 rs11169748 Yes 0.273 5.66E-06NENF rs11169748 Yes 0.292 1.15E-06RUFY1 rs11169748 Yes 0.301 4.92E-07CCDC6 rs2440915 Yes NA NAMETTL6 rs2440915 Yes 0.267 9.64E-06FMR1 rs5952066 Yes NA NAELOVL5 rs5952066 Yes 0.278 3.75E-06METTL9 rs5952066 Yes 0.250b 3.57E-05

0.261 1.53E-05ETV5 rs7620841 Yes NA NA

aThe gene is locatedwithin 200 kb up or downstreamof the SNP, and no cis-regulation was identified. NA, no association between thegene and SNP.bTwo probesets were associated with this SNP.

Genetic Variation and Platinum Response in Lung Cancer

www.aacrjournals.org Clin Cancer Res; 17(17) September 1, 2011 5809

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 10: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

basis of their association with cisplatin cytotoxicity inLCLs, we cannot exclude the possibility of potentialinteraction of genetic variation with other treatmentsand their effect on cisplatin treatment outcome inpatients with lung cancer. Therefore, it is particularlyimportant that our functional studies confirmed thefunctional effects of these genes in cisplatin responsein vitro. Obviously, additional replication studies in inde-pendent sample sets of patients with lung cancer treatedwith platinum-based chemotherapy would be required tofurther confirm these findings.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Luanne Wussow for her assistance with the preparation of thismanuscript.

Grant Support

This study was supported by U.S. NIH research grants, R01 CA80127 (to P.Yang), R01 CA84354 (to P. Yang), K22 CA130828 (to L. Wang), R01 CA138461 (toL. Wang), R25T CA92049 (to X-L. Tan) and U19 GM61388 (The Pharmacoge-nomics Research Network), Mayo Foundation funds (to P. Yang), ASPET-Astellas Award (to L. Wang) and a PhRMA Foundation "Center of Excellencein Clinical Pharmacology" Award (to L. Wang).

The costs of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received April 29, 2011; revised July 11, 2011; accepted July 13, 2011;published OnlineFirst July 20, 2011.

References1. Chemotherapy in non-small cell lung cancer: a meta-analysis using

updated data on individual patients from 52 randomised clinical trials.Non-small Cell Lung Cancer Collaborative Group. BMJ 1995;311:899–909.

2. Gurubhagavatula S, Liu G, Park S, Zhou W, Su L, Wain JC, et al. XPDand XRCC1 genetic polymorphisms are prognostic factors inadvanced non-small-cell lung cancer patients treated with platinumchemotherapy. J Clin Oncol 2004;22:2594–601.

3. Wu X, Lu C, Ye Y, Chang J, Yang H, Lin J, et al. Germline geneticvariations in drug action pathways predict clinical outcomes inadvanced lung cancer treated with platinum-based chemotherapy.Pharmacogenet Genomics 2008;18:955–65.

4. Wei SZ, Zhan P, Shi MQ, Shi Y, Qian Q, Yu LK, et al. Predictive value ofERCC1 and XPD polymorphism in patients with advanced non-smallcell lung cancer receiving platinum-based chemotherapy: a systema-tic review and meta-analysis. Med Oncol 2010;28:315–21.

5. Huang RS, Ratain MJ. Pharmacogenetics and pharmacogenomics ofanticancer agents. CA Cancer J Clin 2009;59:42–55.

6. Marsh S, McLeod H, Dolan E, Shukla SJ, Rabik CA, Gong L, et al.Platinum pathway. Pharmacogenet Genomics 2009;19:563–4.

7. Rabik CA, Maryon EB, Kasza K, Shafer JT, Bartnik CM, Dolan ME.Role of copper transporters in resistance to platinating agents. CancerChemother Pharmacol 2009;64:133–42.

8. Booton R, Ward T, Heighway J, Ashcroft L, Morris J, Thatcher N.Glutathione-S-transferase P1 isoenzyme polymorphisms, platinum-based chemotherapy, and non-small cell lung cancer. J Thorac Oncol2006;1:679–83.

9. Lu C, Spitz MR, Zhao H, Dong Q, Truong M, Chang JY, et al.Association between glutathione S-transferase pi polymorphismsand survival in patients with advanced nonsmall cell lung carcinoma.Cancer 2006;106:441–7.

10. Moyer AM, Sun Z, Batzler AJ, Li L, Schaid DJ, Yang P, et al. Glu-tathione pathway genetic polymorphisms and lung cancer survivalafter platinum-based chemotherapy. Cancer Epidemiol BiomarkersPrev 2010;19:811–21.

11. Kim HT, Lee JE, Shin ES, Yoo YK, Cho JH, Yun MH, et al. Effect ofBRCA1 haplotype on survival of non-small-cell lung cancer patientstreated with platinum-based chemotherapy. J Clin Oncol 2008;26:5972–9.

12. Li F, Sun X, Sun N, Qin S, Cheng H, Feng J, et al. Association betweenpolymorphisms of ERCC1 and XPD and clinical response to platinum-based chemotherapy in advanced non-small cell lung cancer. Am JClin Oncol 2010;33:489–94.

13. Liu L, Yuan P, Wu C, Zhang X, Wang F, Guo H, et al. Assessment ofXPD Lys751Gln and XRCC1 T-77C polymorphisms in advanced non-small-cell lung cancer patients treated with platinum-based che-motherapy. Lung Cancer 2011;73:110–5.

14. Matakidou A, el Galta R, Webb EL, Rudd MF, Bridle H, Eisen T, et al.Genetic variation in the DNA repair genes is predictive of outcome inlung cancer. Hum Mol Genet 2007;16:2333–40.

15. Shiraishi K, Kohno T, Tanai C, Goto Y, Kuchiba A, Yamamoto S, et al.Association of DNA repair gene polymorphisms with response toplatinum-based doublet chemotherapy in patients with non-small-cell lung cancer. J Clin Oncol 2010;28:4945–52.

16. Sun X, Li F, Sun N, Shukui Q, Baoan C, Jifeng F, et al. Polymorphismsin XRCC1 and XPG and response to platinum-based chemotherapy inadvanced non-small cell lung cancer patients. Lung Cancer2009;65:230–6.

17. Tibaldi C, Giovannetti E, Vasile E, Mey V, Laan AC, Nannizzi S, et al.Correlation of CDA, ERCC1, and XPD polymorphisms withresponse and survival in gemcitabine/cisplatin-treated advancednon-small cell lung cancer patients. Clin Cancer Res 2008;14:1797–803.

18. Yang P, Wentzlaff KA, Katzmann JA, Marks RS, Allen MS, LesnickTG, et al. Alpha1-antitrypsin deficiency allele carriers amonglung cancer patients. Cancer Epidemiol Biomarkers Prev 1999;8:461–5.

19. Yang P, Bamlet WR, Sun Z, Ebbert JO, Aubry MC, Krowka MJ, et al.Alpha1-antitrypsin and neutrophil elastase imbalance and lung cancerrisk. Chest 2005;128:445–52.

20. Yang P, Sun Z, KrowkaMJ, AubryMC, BamletWR,Wampfler JA, et al.Alpha1-antitrypsin deficiency carriers, tobacco smoke, chronicobstructive pulmonary disease, and lung cancer risk. Arch InternMed 2008;168:1097–103.

21. Li L, Fridley B, Kalari K, Jenkins G, Batzler A, Safgren S, et al.Gemcitabine and cytosine arabinoside cytotoxicity: association withlymphoblastoid cell expression. Cancer Res 2008;68:7050–8.

22. Li L, Fridley BL, Kalari K, Jenkins G, Batzler A, WeinshilboumRM, et al.Gemcitabine and arabinosylcytosin pharmacogenomics: genome-wide association and drug response biomarkers. PLoS One2009;4:e7765.

23. Niu N, Qin Y, Fridley BL, Hou J, Kalari KR, Zhu M, et al. Radiationpharmacogenomics: a genome-wide association approach to identifyradiation response biomarkers using human lymphoblastoid cell lines.Genome Res 2010;20:1482–92.

24. Huang RS, Duan S, Shukla SJ, Kistner EO, Clark TA, Chen TX, et al.Identification of genetic variants contributing to cisplatin-inducedcytotoxicity by use of a genomewide approach. Am J Hum Genet2007;81:427–37.

25. Wigginton JE, Cutler DJ, Abecasis GR. A note on exact tests of Hardy-Weinberg equilibrium. Am J Hum Genet 2005;76:887–93.

26. Schaid DJ, Batzler AJ, Jenkins GD, Hildebrandt MA. Exact tests ofHardy-Weinberg equilibrium and homogeneity of disequilibriumacross strata. Am J Hum Genet 2006;79:1071–80.

Tan et al.

Clin Cancer Res; 17(17) September 1, 2011 Clinical Cancer Research5810

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 11: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

27. Wellcome Trust Case Control Consortium. Genome-wide associationstudy of 14,000 cases of seven common diseases and 3,000 sharedcontrols. Nature 2007;447:661–78.

28. Pongpanich M, Sullivan PF, Tzeng JY. A quality control algorithm forfiltering SNPs in genome-wide association studies. Bioinformatics2010;26:1731–7.

29. Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, ReichD. Principal components analysis corrects for stratification in gen-ome-wide association studies. Nat Genet 2006;38:904–9.

30. Li Y,Willer CJ, Ding J, Scheet P, Abecasis GR.MaCH: using sequenceand genotype data to estimate haplotypes and unobserved geno-types. Genet Epidemiol 2010;34:816–34.

31. Durbin RM, Abecasis GR, Altshuler DL, Auton A, Brooks LD, GibbsRA, et al. A map of human genome variation from population-scalesequencing. Nature 2010;467:1061–73.

32. Zhou W, Gurubhagavatula S, Liu G, Park S, Neuberg DS, Wain JC,et al. Excision repair cross-complementation group 1 polymorphismpredicts overall survival in advanced non-small cell lung cancerpatients treated with platinum-based chemotherapy. Clin CancerRes 2004;10:4939–43.

33. Dolan ME, Newbold KG, Nagasubramanian R, Wu X, Ratain MJ, CookEH Jr, et al. Heritability and linkage analysis of sensitivity to cisplatin-induced cytotoxicity. Cancer Res 2004;64:4353–6.

34. Shukla SJ, Duan S, Badner JA, Wu X, Dolan ME. Susceptibility lociinvolved in cisplatin-induced cytotoxicity and apoptosis. Pharmaco-genet Genomics 2008;18:253–62.

35. Huang RS, Duan S, Kistner EO, Hartford CM, Dolan ME. Geneticvariants associated with carboplatin-induced cytotoxicity in cell linesderived from Africans. Mol Cancer Ther 2008;7:3038–46.

36. Shukla SJ, Duan S, Wu X, Badner JA, Kasza K, Dolan ME. Whole-genome approach implicates CD44 in cellular resistance to carbo-platin. Hum Genomics 2009;3:128–42.

37. Choy E, Yelensky R, Bonakdar S, Plenge RM, Saxena R, De Jager PL,et al. Genetic analysis of human traits in vitro: drug response andgene expression in lymphoblastoid cell lines. PLoS Genet 2008;4:e1000287.

38. MorleyM,Molony CM,Weber TM, Devlin JL, Ewens KG, Spielman RS,et al. Genetic analysis of genome-wide variation in human geneexpression. Nature 2004;430:743–7.

39. Emilsson V, Thorleifsson G, Zhang B, Leonardson AS, Zink F, Zhu J,et al. Genetics of gene expression and its effect on disease. Nature2008;452:423–8.

40. Kawai T, Matsumoto M, Takeda K, Sanjo H, Akira S. ZIP kinase, anovel serine/threonine kinase which mediates apoptosis. Mol Cell Biol1998;18:1642–51.

41. Shoval Y, Pietrokovski S, Kimchi A. ZIPK: a unique case of murine-specific divergence of a conserved vertebrate gene. PLoS Genet2007;3:1884–93.

42. Raveh T, Kimchi A. DAP kinase-a proapoptotic gene that functions asa tumor suppressor. Exp Cell Res 2001;264:185–92.

43. WangWJ, Kuo JC, Yao CC, Chen RH. DAP-kinase induces apoptosisby suppressing integrin activity and disrupting matrix survival signals.J Cell Biol 2002;159:169–79.

44. Bialik S, Kimchi A. DAP-kinase as a target for drug design in cancerand diseases associated with accelerated cell death. Semin CancerBiol 2004;14:283–94.

45. Taniguchi T, Tischkowitz M, Ameziane N, Hodgson SV, Mathew CG,Joenje H, et al. Disruption of the Fanconi anemia-BRCA pathway incisplatin-sensitive ovarian tumors. Nat Med 2003;9:568–74.

46. Ramirez JL, Rosell R, Taron M, Sanchez-Ronco M, Alberola V, de LasPenas R, et al. 14-3-3sigma methylation in pretreatment serumcirculating DNA of cisplatin-plus-gemcitabine-treated advancednon-small-cell lung cancer patients predicts survival: The SpanishLung Cancer Group. J Clin Oncol 2005;23:9105–12.

Genetic Variation and Platinum Response in Lung Cancer

www.aacrjournals.org Clin Cancer Res; 17(17) September 1, 2011 5811

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133

Page 12: Genetic Variation Predicting Cisplatin Cytotoxicity ... · Genetic Variation Predicting Cisplatin Cytotoxicity Associated with Overall Survival in Lung Cancer Patients Receiving Platinum-Based

2011;17:5801-5811. Published OnlineFirst July 20, 2011.Clin Cancer Res   Xiang-Lin Tan, Ann M. Moyer, Brooke L. Fridley, et al.   Chemotherapy

Platinum-BasedOverall Survival in Lung Cancer Patients Receiving Genetic Variation Predicting Cisplatin Cytotoxicity Associated with

  Updated version

  10.1158/1078-0432.CCR-11-1133doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://clincancerres.aacrjournals.org/content/suppl/2011/07/20/1078-0432.CCR-11-1133.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://clincancerres.aacrjournals.org/content/17/17/5801.full#ref-list-1

This article cites 46 articles, 15 of which you can access for free at:

  Citing articles

  http://clincancerres.aacrjournals.org/content/17/17/5801.full#related-urls

This article has been cited by 6 HighWire-hosted articles. Access the articles at:

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/17/17/5801To request permission to re-use all or part of this article, use this link

on May 20, 2020. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 20, 2011; DOI: 10.1158/1078-0432.CCR-11-1133