A Sequence Polymorphism in miR-608 Predicts Recurrence … · Epstein–Barr virus (EBV) infection status, histologic grade, pathologic stage, and treatment. Immunoglobulin-A antibo-dies

Prevention and Epidemiology

A Sequence Polymorphism in miR-608 Predicts Recurrenceafter Radiotherapy for Nasopharyngeal Carcinoma

Jian Zheng1, Jieqiong Deng1, Mang Xiao4, Lei Yang5, Liyuan Zhang2, Yonghe You1, Min Hu3, Na Li1,Hongchun Wu1, Wei Li1, Jiachun Lu5, and Yifeng Zhou1

AbstractNasopharyngeal carcinoma is treated with radiotherapy and other modalities, but there is little infor-

mation on individual genetic factors to help predict and improve patient outcomes. Single-nucleotidepolymorphisms (SNP) in mature microRNA (miRNA) sequences have the potential to exert broad impact asmiRNAs target many mRNAs. The aim of this study was to evaluate the effects of SNPs in mature miRNAsequences on clinical outcome in patients with nasopharyngeal carcinoma receiving radiotherapy. Inparticular, we analyzed associations between seven SNPs and nasopharyngeal carcinoma locoregionalrecurrence (LRR) in 837 patients from eastern China, validating the findings in an additional 828 patientsfrom southern China. We found that miR-608 rs4919510C>G exhibited a consistent association with LRR inthe discovery set [HR, 2.05; 95% confidence interval (CI), 1.35–3.21], the validation set (HR, 2.24; 95% CI, 1.45–3.38), and the combined dataset (HR, 2.08; 95% CI, 1.41–3.26). Biochemical investigations showed thatrs4919510C>G affects expression of miR-608 target genes along with nasopharyngeal carcinoma cell growthafter irradiation in vivo and in vitro. Notably, X-ray radiation induced more chromatid breaks in lymphocytecells from rs4919510CC carriers than in those from subjects with other genotypes (P ¼ 0.0024). Our findingsreveal rs4919510C>G in miR-608 as a simple marker to predict LRR in patients with radiotherapy-treatednasopharyngeal carcinoma. Cancer Res; 73(16); 5151–62. �2013 AACR.

IntroductionNasopharyngeal carcinoma is one of the most common

head and neck malignancies in South China and SouthAsia, where the incidence is as high as 50 of 100,000, but itis rare in the Western world (1 of 100,000; refs. 1–3).Although significant progress has been made in the diag-nosis and treatment of nasopharyngeal carcinoma in recentdecades, it remains a highly frequent cause of cancer-related death in China. The worldwide 5-year overall sur-vival (OS) rate ranges from 32% to 62% among a series ofstudies involving more than 9,500 patients in all stages ofnasopharyngeal carcinoma (4). The main cause of deathamong patients with nasopharyngeal carcinoma is recur-

rence, and 80% of all recurrences occur during the first 3years after pathogenesis (4). Currently, radiotherapy is themain treatment modality for this malignancy; however,differences in individual sensitivity to radiotherapy havea great impact on the recurrence rate of nasopharyngealcarcinoma (5).

microRNAs (miRNA) comprise a group of endogenous,single-stranded, small, noncoding RNAs that have emerged askey regulators of fundamental biologic processes via theircontrol over the expression of more than 30% of human genes(6, 7). mRNAs are initially transcribed as primary miRNAs (pri-miRNA) with several hundred nucleotides that are furtherprocessed into hairpin-structured precursor miRNAs (pre-miRNA) and then intomaturemiRNAs (8–10). MaturemiRNAsconsist of approximately 22 to 25 nucleotides. To date, morethan 1,000 miRNAs have been detected in humans (11, 12).Physiologically, miRNAs act as negative gene regulators thatfine-tune translational output through targeted mRNA bind-ing. A variety of pathologic associations have been attributed toaltered miRNA networks, particularly in cancer, because miR-NAs can function as both oncogenic and tumor suppressorfactors (13–15).

Genome-wide association studies (GWAS) have identifiedseveral single-nucleotide polymorphisms (SNP) related tonasopharyngeal carcinoma susceptibility (1, 16). To date,candidate gene approaches remain the primary strategy usedin association studies of clinical outcomes. Genetic variants,such as SNPs in miRNAs, can affect their biogenesis, proces-sing, and target site binding in a variety of ways (10). A SNP in

Authors' Affiliations: 1Laboratory of Cancer Molecular Genetics, MedicalCollege of SoochowUniversity; Departments of 2Radiotherapy &Oncologyand 3Obstetrics and Gynecology, The Second Affiliated Hospital of Soo-chow University, Suzhou; 4Department of Otorhinolaryngology-Head andNeck Surgery, Sir Run Run Shaw Hospital, Zhejiang University, Hangzhou;and 5The Institute for Chemical Carcinogenesis, The State Key Lab ofRespiratory Disease, Guangzhou Medical University, Guangzhou, China

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

J. Zheng and J. Deng contributed equally to this work.

Corresponding Author: Yifeng Zhou, Medical College of Soochow Uni-versity, No.199 Ren-ai Rd., Suzhou 215123, China. Phone: 86-512-65884720; Fax: 86-512-65884720; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-13-0395

�2013 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 5151

on April 6, 2021. © 2013 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 24, 2013; DOI: 10.1158/0008-5472.CAN-13-0395

http://cancerres.aacrjournals.org/

themature sequence can alter target site interactions by eitherstrengthening or weakening hybridization kinetics, and SNPscan significantly transform the target library of the miRNAitself. Numerous studies have linked genetic variation inmature miRNA sequences to cancer risk and prognosis (17–19). Nonetheless, to the best of our knowledge, the importanceof mature miRNA sequence SNPs in the locoregional recur-rence (LRR; ref. 20) of nasopharyngeal carcinoma after radio-therapy remains unknown. In this study, we evaluated thefrequencies of mature miRNA sequence SNPs in patients withnasopharyngeal carcinoma and assessed their impact on LRRafter radiotherapy.

Materials and MethodsStudy population

A total of 1,665 patients with histologically confirmed naso-pharyngeal carcinoma [International Classification of Disease(ICD) 9:147, ICD10:C11] were recruited from Jiangsu Province

in eastern China [The First Affiliated Hospital of SoochowUniversity (Suzhou, China), The Second Affiliated Hospital ofSoochow University (Suzhou, China), The Third Hospital Affil-iated to Nantong University (Wuxi, China), and Huaian No.1Hospital (Huaian, China)] as well as Guangzhou City in south-ern China (The Tumor Hospitals affiliated to GuangzhouMedical College, Guangzhou, China) between 2000 and 2009and were followed up until 2012 (Table 1). All patients weretreated with definitive radiotherapy at urban hospitals, and nopatients underwent surgery. In our eastern Chinese popula-tion, 837 newly diagnosed patients were analyzed as a discov-ery set in this study. In the southern Chinese population, 828newly diagnosed patients were used as a validation set (21, 22).A self-administered questionnaire was used for all patients tocollect epidemiologic data including demographical charac-teristics, tobacco and alcohol use, family history of cancer,and medical history. All LRRs were diagnosed by endoscopyand biopsy and/or computed tomography (CT) scan of the

Table 1. Baseline demographic and clinical characteristics of study populations

Discovery set (eastern Chinese, N ¼ 837) Validation set (southern Chinese, N ¼ 828)

VariablesRecurrence(N ¼ 176)

No recurrence(N ¼ 661) P

Recurrence(N ¼ 218)

No recurrence(N ¼ 610) P

Age, mean (SEM) 53.91 (0.879) 52.85 (0.411) 0.856 50.81 (0.821) 51.54 (0.409) 0.413Gender, N (%) 0.239 0.476Male 116 (65.91) 466 (70.50) 162 (74.32) 438 (71.81)Female 60 (34.09) 195 (29.50) 56 (25.68) 172 (28.19)

Family history of cancer, N (%) 0.859 0.466No 156 (88.64) 589 (89.11) 197 (90.37) 561 (91.96)Yes 20 (11.36) 72 (10.89) 21 (9.63) 49 (8.04)

Smoking status, N (%) 0.542 0.438Never 82 (46.59) 291 (44.02) 106 (48.62) 278 (45.57)Ever 94 (53.41) 370 (55.98) 112 (51.38) 332 (54.43)

Drinking status, N (%) 0.367 0.415Never 87 (49.43) 352 (53.25) 107 (49.08) 319 (52.30)Ever 89 (50.57) 309 (46.75) 111 (50.92) 291 (47.70)

BMI, N (%) 0.911 0.151�20 28 (15.91) 116 (17.55) 62 (28.44) 205 (33.61)20–28 133 (75.57) 481 (72.77) 148 (67.89) 387 (63.44)�28 15 (8.52) 64 (9.68) 8 (3.67) 18 (2.95)

EBV infection status, N (%) 0.024 0.007Positive 159 (90.34) 552 (83.51) 202 (92.66) 522 (85.58)Negative 17 (9.66) 109 (16.49) 16 (7.34) 88 (14.42)

Stage, N (%) 0.285 0.194I 8 (4.55) 28 (4.23) 9 (4.13) 26 (4.26)II 42 (23.86) 155 (23.45) 47 (21.55) 163 (26.72)III 68 (38.64) 316 (47.81) 92 (42.21) 245 (40.16)IV 58 (32.95) 162 (24.51) 70 (32.11) 176 (28.86)

Histologic grade, N (%) 0.119 0.434Undifferentiated 139 (78.97) 484 (73.22) 166 (76.15) 448 (73.44)Differentiated 37 (21.02) 177 (26.78) 52 (23.85) 162 (26.56)

Chemotherapy, N (%) 0.003 0.011Yes 143 (81.25) 593 (89.71) 175 (80.28) 533 (87.38)No 33 (18.75) 68 (10.29) 43 (19.72) 77 (12.62)

Zheng et al.

Cancer Res; 73(16) August 15, 2013 Cancer Research5152




nasopharynx and the skull base showing progressive boneerosion and/or soft tissue swelling (23). Diagnosis of a secondprimary tumor was based on a modification of the criteria ofWarren andGates (24), andwe did notfind any second primarytumor cases in our present study. Each patient saw his or herdoctor during the follow-up period for assessment of recur-rence once a month in the first year, every 2 months in years 2and 3, and every 6 months thereafter. We reviewed patients'medical records during the follow-up period to collect clinicalinformation including the date of diagnosis, recurrence status,Epstein–Barr virus (EBV) infection status, histologic grade,pathologic stage, and treatment. Immunoglobulin-A antibo-dies to EBV capsid antigen (EBV/IgA/VCA) and immunoglob-ulin-A antibodies to EBV early antigen were confirmed byserologic testing at the time of study enrollment. Thetumor, node, metastasis classification and tumor stagingwere evaluated according to the 2002 American Joint Com-mittee on Cancer staging system. At recruitment, informedconsent was obtained from each patient, and the study wasapproved by the Medical Ethics Committee of Soochow Uni-versity (SZUM2009061002) and the Institutional Review Boardof Guangzhou Medical College (GZMC2009060426).

Radiotherapy techniqueRadiotherapy was conducted largely using standard proce-

dures (5). For details, see Supplementary Materials and Meth-ods. The patients received about a month of radiotherapy, andtreatment was delivered once daily (5 fractions/week). The lagtime between date of diagnosis and date of first treatment iswithin 2 weeks.

Follow-up and endpoint selectionAll patients were evaluated weekly during the treatment

period, and after the completion of treatment, they werefollowed up every 3 months by telephone for the first 3 years,every 6 months in years 4 and 5, and annually thereafter.Overall, we recruited 2,073 patients with nasopharyngeal car-cinoma (1,075 from the eastern Chinese and 998 from thesouthern Chinese). A total of 1,665 patients (837 from theeastern Chinese and 828 from the southern Chinese) hadcomplete follow-ups and clinical information. Among theremaining 408 patients (238 from the eastern Chinese and170 from the southern Chinese) with incomplete follow-up orclinical information or both, 125 cases (6.03%) lacked stageand/or histology information, 132 cases (6.37%) had incorrecttelephone numbers, 49 cases (2.36%) refused to participate, 66cases (3.18%) had ambiguous death date and/or indirect deathbecause of nasopharyngeal carcinoma, and 36 cases (1.74%)moved or were unavailable for unknown reasons. However,there was no significant difference in the distributions ofdemographic characters (e.g., age and gender), smoking status,drinking status, body mass index (BMI), EBV infection status,stage, histologic grade, and chemotherapy between thepatients with nasopharyngeal carcinoma with and withoutfollow-up/clinical information (P ¼ 0.642 for age; P ¼ 0.365for gender; P¼ 0.516 for smoking status; P¼ 0.287 for drinkingstatus;P¼ 0.258 for BMI;P¼ 0.539 for EBV infection status;P¼0.225 for stage; P¼ 0.169 for histologic grade; and P¼ 0.258 for

chemotherapy). The endpoints of the current study includedthe time to recurrence (TTR) and OS. The TTR was calculatedas the time from the date of diagnosis of nasopharyngealcarcinoma to the date of the first observation of LRR, or untilthe last follow-up if the patient was recurrence-free at thattime. TTR was censored at the time of death or at the lastfollow-up if the patient remained recurrence-free at that time.The OS was defined as the time from pathologic diagnosis todeath fromany cause, or the last contact if the patientwas alive.

Tissue samplesTo determine the expression levels of selected genes, we

collected 35 nasopharyngeal carcinoma tissues from patientswho had undergone resection from The Second AffiliatedHospital of Soochow University. All cases were histopatholog-ically diagnosed as nasopharyngeal carcinoma by biopsy andwithout radio- or chemotherapy.

SNP selection and genotypingAccording to the bioinformatics analysis, SNPs located

in mature miRNA sequences with allelic frequencies inChinese populations were selected. First, based on themiRBase database (http://mirbase.org; up to January 1,2012), 205 polymorphic loci located in mature miRNAsequences (Supplementary Table S1) were screened. Second,based on the HapMap public database [HapMap Data Rel28 phase IIþIII, August 10, on National Center for Bio-technology Information (NCBI) B36 assembly, dbSNPb126; up to January 1, 2012], we found that there were onlyseven SNPs in 205 polymorphic loci that had allelic fre-quencies in Chinese populations (Supplementary Table S1).Finally, we selected these seven SNPs located in maturemiRNA sequences with the allelic frequencies in theChinese population for genotyping:miR-499 (rs3746444C>T),miR-608 (rs4919510C>G), miR-3152 (rs13299349A>G), miR-4513 (rs2168518C>T), miR-4520a (rs8078913C>T), miR-4741(rs7227168C>T), and miR-4762 (rs41524547C>G).

Genomic DNA was isolated from the peripheral bloodlymphocytes of all the study subjects. All subjects were geno-typed for the seven SNPs by using allele-specific matrix-assisted laser desorption/ionization–time-of-flight (MALDI-TOF) mass spectrometry (Sequenom; refs. 25, 26). The DNAisolation and genotyping were conducted in Suzhou Center(Suzhou, China; for the eastern Chinese population) andGuangzhou Center (Guangzhou, China; for the southern Chi-nese population), respectively. The cross-trained laboratorypersonnel conducting the genotyping were blinded to patientinformation. Approximately, 10% of the samples were alsorandomly selected for a blinded repeat of the genotypingwithout prior knowledge of the previous genotyping resultor the patient information, and the results were in 100%agreement.

Cell cultureHuman nasopharyngeal carcinoma cell lines (CNE-1 and

CNE-2) and 293T cells were purchased from the Cell Bankof Type Culture Collection of the Chinese Academy ofSciences, Shanghai Institute of Biochemistry and Cell Biology

Polymorphism in microRNA and Recurrence of Nasopharyngeal Carcinoma

www.aacrjournals.org Cancer Res; 73(16) August 15, 2013 5153




(Shanghai, China) and passaged for fewer than 6 months. Thecell lines were characterized by DNA fingerprinting analysisusing short-tandem repeat (STR) markers. Cells were main-tained according to the Cell Bank's protocols.

Plasmids, lentiviral production, and transductionLentiviral expression plasmids construction, lentiviral pro-

duction, and transduction were conducted by following theestablished and previously published procedures (27, 28). Fordetails, see Supplementary Materials and Methods. Finally,CNE-1-empty vector, CNE-1-miR-608-C, CNE-1-miR-608-G,CNE-2-empty vector, CNE-2-miR-608-C, and CNE-2-miR-608-G cells were stably selected with G418 at 500 mg/mL (Gibco),and the drug-resistant cell populations were used for subse-quent studies.

RNA extraction and microarray analysisTotal RNA was extracted from the cultured cells in both

experimental (CNE-1-miR-608-C and CNE-1-miR-608-G) andcontrol (CNE-1-empty vector) groups using RNeasy Mini Kits(Qiagen) according to the manufacturer's instructions. Geneexpression profiling was conducted using the Human OneAr-ray microarray (Phalanx Biotech; refs. 29, 30). Details are givenin Supplementary Materials andMethods. The array data havebeen deposited in Gene Expression Omnibus (GEO; accessionnumber: GSE46372).

Quantitative real-time PCR analysisTotal RNA was isolated from 35 nasopharyngeal carcinoma

tissue samples with TRIzol reagent (Molecular Research Cen-ter, Inc.). The relative gene expression for the selected geneswas quantified using the ABI Prism 7000 sequence detectionsystem (Applied Biosystems) based on the SYBR Green meth-od. The primers used for PCR amplification of the candidategenes are listed in Supplementary Table S2. The expression ofmiR-608 in nasopharyngeal carcinoma cells was calculatedrelative to theU6 small nuclear RNA (SupplementaryMaterialsand Methods).

Construction of FBXO32 30-UTR luciferase reporterplasmid

The reporter vector psiCHECK-2 (Promega)was prepared byamplifying a 683-bp FBXO32 30-untranslated region (30-UTR)region from a human genomic DNA sample, including theartificial XhoI and NotI enzyme restriction sites with theforward primer 50-TGTATTATGCTCGAGCCATAGTTCTC-30

and reverse primer 50-CGCTCTAAGTCTAAAGCGGCCGC-TAG-30. Construction of the FBXO32 30-UTR luciferase reporterplasmid was conducted according to a previously describedmethod (21). The resulting construct (psiCHECK-2-FBXO32-30-UTR) was verified by sequencing.

Transient transfections and luciferase assaysThe CNE-1 and CNE-2 cells were seeded at 1 � 105 cells per

well in 24-well plates (BD Biosciences). Sixteen hours afterplating, the cells were transfected with Lipofectamine 2000(Invitrogen) according to the manufacturer's instructions(Supplementary Materials and Methods).

Mutagen sensitivity assayTo further explore the differences in DNA repair ability

among individuals with different miR-608 rs4919510C>G gen-otypes, we evaluated X-ray radiation sensitivities in 135 addi-tional control subjects according to published protocols(Supplementary Materials and Methods; refs. 31, 32). Thevalues of chromatid breaks per cell (b/c) were used to indicatethe DNA repair capacity of the individual (31, 33).

Cell growth analysis in response to irradiationCNE-1-empty vector, CNE-1-miR-608-C, CNE-1-miR-608-G,

CNE-2-empty vector, CNE-2-miR-608-C, and CNE-2-miR-608-Gcells were plated in a 24-well culture plates (2.5 � 104/well).After incubation for 24 hours, the cells received 2 Gy ofirradiation. Cell growth was monitored by counting cell num-bers at various time intervals. Three independent experimentswere carried out in triplicate.

Animal modelThe animal experiments were carried out in accordance

with National guidelines and approved by the LaboratoryAnimal Center of Soochow University. CNE-1-empty vector,CNE-1-miR-608-C, CNE-1-miR-608-G, CNE-2-empty vector,CNE-2-miR-608-C, and CNE-2-miR-608-G cells were diluted toa concentration of 5� 107/mL in physiologic saline. Nudemicewere injected subcutaneously with 0.1 mL of the suspensioninto the back flank (6 mice/group). For details, see Supple-mentary Materials and Methods.

Radiation deliveryLocal irradiation of the implanted tumor was conducted

using a customized mouse jig with other parts of the bodyshielded with lead. Each mouse was confined to a customizedmouse jig with a circular window, through which the tumorbed was exposed to the radiation, and was irradiated locally.Mice were exposed to X-rays with 5-mm thick lead shieldswhen the tumor bed was gently extended into the radiationfield. Tumors were locally irradiated at a dose of 2, 10, or 20 Gyusing a RS2000 X-ray Biological Irradiator (Rad Source Tech-nologies) at a dose rate of 1 Gy/min through a 0.2-mm copperfilter beginning 1 week after transplantation.

Statistical analysisThe x2 test or Fisher exact test was applied separately to

compare the distribution of selected demographic and clinicalvariables according to recurrence status. The Cox proportionalhazards model was used to estimate HRs and their 95%confidence intervals (CIs) for multivariate analyses in thediscovery set, validation set, and combined dataset. The anal-yses were adjusted for age, gender, BMI, smoking status,drinking status, family history of cancer, EBV infection status,stage, histologic grade, and chemotherapy, as appropriate.Associations between genotypes and TTR and OS were esti-mated using the Kaplan–Meier method, and statistical signif-icance was determined using the log-rank test (34). For thefunctional analyses, data were presented by using mean �SEM; the comparison of mean between two groups was con-ducted by using Student t test; statistical comparisons of more

Zheng et al.





than two groups were conducted using the one-way ANOVA(35), and then least-significant difference (LSD) for multiplecomparisons. The statistical analyses were conducted usingSTATA software (version 10; STATA Corporation). All P valueswere two-sided and P < 0.05 was considered statisticallysignificant.

ResultsPatient characteristicsThe clinical information and demographic characteristics

of the 1,665 patients with nasopharyngeal carcinoma includ-ed in this study are shown in Table 1. For the easternChinese population, the median patient age was 54 years(range, 21–82 years) and the median follow-up time was 3.1years (range, 0.4–10.8 years); 176 patients (21.03%) showedtumor LRR after radiotherapy, resulting in a 3-year recur-rence probability of 0.21 � 0.04, and the median TTR was 1.9years (95% CI, 1.6–2.3 years); 180 patients (21.51%) have died(including 150 LRRs, 21 distant metastases, and 9 otherreasons), but the median OS had not been reached duringthe follow-up time. No significant differences were notedwith regard to sex, age, family history of cancer, smokingstatus, drinking status, BMI, stage, or histologic gradeaccording to recurrence status; however, significant differ-ences were found in EBV infection status and the number ofpatients who received chemotherapy (Table 1).For the southern Chinese population, the median patient

agewas 52 years (range, 19–87 years) and themedian follow-uptimewas 2.9 years (range, 0.5–10.2 years); 218 patients (26.33%)exhibited tumor recurrence after radiotherapy, resulting in a 3-year recurrence probability of 0.28� 0.04, and themedian TTRwas 2.2 years (95%CI, 2.1–2.5 years); 223 patients (26.93%) havedied (including 182 LRRs, 29 distant metastases, and 12 otherreasons), but median OS had not been reached during thefollow-up time.

Associations between SNPs and recurrence riskGenotype distributions of all SNPs were found to be con-

sistent with Hardy–Weinberg equilibrium (HWE), with the

exception of rs7227168C>T and rs41524547C>G. No significantassociations were noted in the SNPs and baseline demo-graphic, clinical, or pathologic characteristics according torecurrence status. The selected seven candidate SNPs weregenotyped in a discovery set consisting of 837 patients withnasopharyngeal carcinoma from the eastern Chinese popula-tion. In the univariate analysis, patients carrying the miR-608rs4910510GG genotype had a median TTR of 6.2 years, com-pared with a median TTR of 8.9 years for patients withrs4910510CC genotype (HR, 2.02; 95% CI, 1.29–3.16; P ¼0.0008; Fig. 1A; Table 2). However, the other tested SNPs didnot show any statistically significant associations with naso-pharyngeal carcinoma LRR in the univariate analyses. In themultivariate analysis, a Cox proportional hazards model wasadjusted for age, gender, BMI, smoking status, drinking status,family history of cancer, EBV infection status, stage, histologicgrade, and chemotherapy, and the miR-608 rs4910510GGgenotype remained significantly associated with nasopharyn-geal carcinoma LRR (HR, 2.05; 95% CI, 1.35–3.21; P ¼0.0012; Table 2). These results were also confirmed in thesouthern Chinese population, in which rs4910510C>G dis-played a consistent association with recurrence in the valida-tion (HR, 2.24; 95% CI, 1.45–3.38; P ¼ 0.0006; Fig. 1B; Table 2)and combined (HR, 2.08; 95% CI, 1.41–3.26; P < 0.00001)datasets (Fig. 1C; Table 3).

About 85% of patients for both populations received thechemotherapy during the treatment; however, there were nosignificant differences in the associations with miR-608rs4919510C>G polymorphism by chemotherapy (P > 0.05), asshown in Supplementary Fig. S1.

Associations between miR-608 rs4919510C>G genotypesand OS

As shown in Supplementary Fig. S2A–S2C, we found thatmiR-608 rs4919510C>G exhibited a consistent association withthe OS in the discovery set (HR, 2.13; 95% CI, 1.39–3.28; P ¼0.0006), the validation set (HR, 1.89; 95% CI, 1.23–3.07; P ¼0.0023), and the combined dataset (HR, 1.95; 95% CI, 1.31–3.15;P < 0.0001). However, we also analyzed the time from

A100

90

80

70

60

50

40

30

20

10

100

90

80

70

60

50

40

30

20

10

100

90

80

70

60

50

40

30

20

10

0

B C

0 2 4Years since diagnosis of NPC

Log-rank test: P = 0.0012

rs4919510C>GCC (36/201) MTTR=8.9GC (78/420) MTTR=7.9GG (62/216) MTTR=6.2

CC (42/209) MTTR=9.0GC (110/439) MTTR=8.1GG (66/180) MTTR=5.2

CC (78/410) MTTR=8.9GC (188/859) MTTR=8.1GG (128/396) MTTR=5.3

rs4919510C>G rs4919510C>G

Log-rank test: P = 0.0006 Log-rank test: P < 0.00001

Est

imat

ed r

ecur

renc

e-fr

ee p

roba

bilit

y (%

)

Est

imat

ed r

ecur

renc

e-fr

ee p

roba

bilit

y (%

)

Est

imat

ed r

ecur

renc

e-fr

ee p

roba

bilit

y (%

)

Years since diagnosis of NPC Years since diagnosis of NPC6 8 10 12 0 2 4 6 8 10 12 0 2 4 6 8 10 12

Figure 1. Kaplan–Meier TTRcurves of patientswith nasopharyngeal carcinoma receiving radiotherapybygenotypesofmiR-608 rs4919510C>G.Discovery set(A), validation set (B), and combined dataset (C). MTTR, median time to recurrence.






Tab

le2.

Univa

riate

andmultiv

ariate

analysis

ofmaturemiRNAsse

que

ncepolym

orphism

san

dna

sopha

ryng

ealc

arcino

marecu

rren

cein

disco

very

setan

dva

lidationse

t

Disco

very

set(eas

tern

Chine

se,N

¼83

7)Validationse

t(southe

rnChine

se,N

¼82

8)

Univa

riatean

alys

isMultiva

riatean

alys

isUniva

riatean

alys

isMultiva

riatean

alys

is

Nd

MAFe

Probab

ility

aHR

(95%

CI)

Pb

HR

(95%

CI)

Pc

Nd

MAFe

Probab

ility

aHR

(95%

CI)

Pb

HR

(95%

CI)

Pc

miR-608

rs49

1951

00.50

90.00

080.00

120.48

20.00

010.00

06CC

201

0.17

�0.03

1.00

(Referen

ce)

1.00

(Referen

ce)

209

0.21

�0.04

1.00

(Referen

ce)

1.00

(Referen

ce)

GC

420

0.19

�0.03

1.41

(0.89–

2.04

)1.45

(0.92–

2.18

)43

90.27

�0.03

1.55

(1.03–

2.15

)1.62

(1.11–

2.19

)GG

216

0.26

�0.05

2.02

(1.29–

3.16

)2.05

(1.35–

3.21

)18

00.37

�0.04

2.21

(1.49–

3.24

)2.24

(1.45–

3.38

)GCþG

G63

60.22

�0.03

1.61

(1.13–

2.29

)1.68

(1.18–

2.61

)61

90.29

�0.03

1.65

(1.17–

2.39

)1.71

(1.23–

2.31

)miR-451

3rs21

6851

80.19

40.12

00.17

80.18

20.49

80.40

1CC

552

0.19

�0.02

1.00

(Referen

ce)

1.00

(Referen

ce)

560

0.26

�0.03

1.00

(Referen

ce)

1.00

(Referen

ce)

TC24

60.23

�0.03

1.18

(0.85–

1.62

)1.21

(0.87–

1.67

)23

40.28

�0.04

1.11

(0.71–

1.53

)1.15

(0.84–

1.58

)TT

3934

TCþT

T28

50.22

�0.05

1.14

(0.82–

1.68

)1.18

(0.84–

1.71

)26

80.29

�0.03

1.17

(0.78–

1.61

)1.23

(0.82–

1.65

)miR-452

0ars80

7891

30.34

70.73

70.70

50.33

50.68

30.78

2TT

368

0.18

�0.02

1.00

(Referen

ce)

1.00

(Referen

ce)

378

0.28

�0.03

1.00

(Referen

ce)

1.00

(Referen

ce)

CT

357

0.21

�0.03

1.14

(0.83–

1.57

)1.13

(0.82–

1.56

)34

60.27

�0.03

1.08

(0.87–

1.18

)1.04

(0.87–

1.21

)CC

112

0.22

�0.04

1.02

(0.64–

1.64

)0.91

(0.54–

1.51

)10

40.25

�0.05

0.92

(0.73–

1.14

)1.01

(0.78–

1.24

)CTþ

CC

469

0.22

�0.04

1.11

(0.82–

1.59

)1.09

(0.81–

1.52

)45

00.26

�0.04

0.96

(0.78–

1.17

)1.02

(0.85–

1.23

)miR-499

rs37

4644

40.14

80.64

70.52

30.15

00.46

90.21

3TT

611

0.19

�0.02

1.00

(Referen

ce)

1.00

(Referen

ce)

605

0.28

�0.02

1.00

(Referen

ce)

1.00

(Referen

ce)

TC20

40.20

�0.05

1.16

(0.82–

1.64

)1.13

(0.79–

1.60

)19

90.25

�0.03

0.87

(0.47–

1.64

)0.93

(0.45–

1.72

)CC

2224

TCþC

C22

60.21

�0.03

1.05

(0.72–

1.53

)1.06

(0.73–

1.54

)22

30.27

�0.02

0.85

(0.57–

1.17

)0.86

(0.53–

1.19

)miR-315

2rs13

2993

490.12

00.33

90.25

60.12

20.67

50.53

4GG

653

0.18

�0.02

1.00

(Referen

ce)

1.00

(Referen

ce)

639

0.27

�0.03

1.00

(Referen

ce)

1.00

(Referen

ce)

AG

167

0.21

�0.05

1.11

(0.65–

1.64

)1.13

(0.70–

1.67

)17

70.29

�0.05

1.15

(0.80–

1.61

)1.09

(0.77–

1.56

)AA

1712

AGþA

A18

40.24

�0.04

1.18

(0.75–

1.89

)1.19

(0.78–

1.93

)18

90.30

�0.04

1.07

(0.57–

1.69

)1.06

(0.76–

1.65

)

aProbab

ility

�SEof

3-ye

arrecu

rren

ce.

bOnthebas

isof

log-rank

test.

cOnthebas

isof

Waldtest

with

inCox

proportio

nalhaz

ardsmod

elwith

adjusted

fora

ge,g

ender,B

MI,sm

okingstatus

,drin

king

status

,fam

ilyhistoryof

canc

er,E

BVinfectionstatus

,stag

e,histolog

icgrad

e,an

dch

emothe

rapy.

dNum

ber

ofpatientswith

thegive

nge

notype.

eMinor

allele

freq

uenc

y.

Zheng et al.





recurrence to death among patients with nasopharyngealcarcinoma (Supplementary Fig. S2D) and found that there wasno significant difference among patients carrying the differentmiR-608 rs4919510C>G genotypes (ANOVA test, P > 0.05).

Effects of the rs4919510C>G genotypes on the miR-608target genes expressionThe nasopharyngeal carcinoma cell line CNE-1 was infected

withmiR-608-C-allele lentivirus,miR-608-G-allele lentivirus, orcontrol lentivirus, and the infection efficiency exceeded 90% forall lentiviruses, as shown in Supplementary Fig. S3. Next, theexpression levels of miR-608 target genes were detected bymicroarray analysis (Fig. 2A). We compared RNA transcriptionlevels between the CNE-1-miR-608-C and CNE-1-miR-608-Ggroups. Overall, 2,242 genes were differentially expressed witha P< 0.01. Of these genes, 801were upregulated and 1,441 geneswere downregulated (GEO; accession number: GSE46372); thegenes with altered expression after infection induced immu-nity and defense genes, DNA repair genes, cell growth–relatedgenes, tumor invasion and metastasis-related genes, cancerstem cell–related genes, and cell death–related genes. Wealso compared gene expression levels between the CNE-1-miR-608-C and CNE-1-miR-608-G groups and the controlgroup (CNE-1-empty vector). We identified 108 genes thatwere downregulated with a P < 0.01 (GEO; accession number:GSE46372). Interestingly, four differentially expressedgenes were present in both comparisons: FBXO32, RHOD,TRIM31, and TSC22D3 (Fig. 2A). Differentially expressedgenes from the microarray experiments were analyzed usingSTRING, a database of known and predicted protein–proteininteractions. Figure 2B summarizes the network of predictedassociations for differentially expressed gene-encoded pro-teins. The results indicated that FBXO32 was the key gene ofthis protein interaction net. This gene was linked to TNF andKITLG, and these genes were linked to many downstreamgenes. All of these genes were interrelated, thereby forming a

large network. However, the other 3 genes (RHOD, TRIM31,and TSC22D3) were not linked to other genes (outside thenetwork).

On the basis of the microarray array results, the expres-sion levels of 20 selected genes (9 upregulated and 11downregulated) were evaluated using the quantitative PCR(qPCR) analysis, and all these selected genes with alteredexpression after infection induced immunity and defensegenes, DNA repair genes, cell growth–related genes, tumorinvasion and metastasis-related genes, cancer stem cell–related genes, and cell death–related genes. The results forthe 20 selected genes had shown that the direction ofexpression changes were consistent with those found bymicroarray analysis (Fig. 3A).

The rs4919510C>G genotypes affect FBXO32 expressionby inhibiting the binding of miR-608 in vitro

According to a bioinformatics analysis software program(MIRanda Java Interface v1.0), the FBXO32 30-UTR waspredicted to bear a miR-608–binding site. CNE-1 cells weretransiently cotransfected with twomiR-608mimics (contain-ing different rs4919510C>G alleles) and the reporter con-structs and then assessed for luciferase activity. Comparedwith the rs4919510C allele, the rs4919510G allele was asso-ciated with significantly reduced luciferase activity in aconcentration-dependent manner (Fig. 3B). The sameexperiments were repeated using CNE-2 cells with similarresults (Fig. 3C).

Effects of themiR-608 rs4919510C>G genotypes on X-rayradiation-induced chromatid breaks in lymphocytes

We investigated the phenotype of X-ray radiation-inducedchromatid breaks in lymphocyte cells from 125 controlsubjects and the rs4919510C>G genotype–phenotype asso-ciation in these individuals. The exact b/c value was definedas the b/c value of the treatment group minus the

Table 3. Univariate and multivariate analysis of mature miRNAs sequence polymorphisms andnasopharyngeal carcinoma recurrence in combined dataset

Combined dataset (eastern and southern Chinese, N ¼ 1,665)

Univariate analysis Multivariate analysis

Nd MAFe Probabilitya HR (95% CI) Pb HR (95% CI) Pc

miR-608 rs4919510 0.496

spontaneous b/c value of the untreated group. As shownin Fig. 3D, we found that the mean� SEM b/c value in the 28rs4919510CC carriers was 0.225 � 0.008, which was signifi-cantly higher than those of 39 individuals with rs4919510GGgenotype (0.182 � 0.007) and 58 individuals with thers4919510GC genotype (0.204 � 0.007; ANOVA test, P ¼0.0024).

Effects of the miR-608 rs4919510C>G genotypes on thecell growth in response to irradiation

CNE-1-empty vector, CNE-1-miR-608-C, and CNE-1-miR-608-G cells were subjected to 2 Gy of radiation to examinethe effect on cell growth. As shown in Fig. 4A, the cell growthdelay after irradiation was shorter for CNE-1-miR-608-G cellsthan for CNE-1-miR-608-C cells. The same experiments were

A Lentivirus

Lentivirus

Lentivirus

miRNA-empty vector

Infect target cells

CNE-1-miR-608-C

miR-608-Cvs.

miR-608-G

miR-608-C+

miR-608-Gvs.

empty vector

108 genesdownregulated

FBXO32RHOD

TRIM31TSC22D3

801 genesupregulated1,441 genes

downregulated

CNE-1-miR-608-G

CNE-1-empty vector

miR-608-C

miR-608-G

B

Figure 2. A, the CNE-1 cells were infected with miR-608-C-allele lentivirus, miR-608-G-allele lentivirus, or control lentivirus. Next, the expression levelsofmiR-608 target genes were detected by microarray analysis. B, differentially expressed genes between CNE-1-miR-608-C group and CNE-1-miR-608-Ggroup were analyzed using the STRING database. The network nodes represent the proteins encoded by the differentially expressed genes. Differentline colors represent the types of evidence for the association.

Zheng et al.





repeated using CNE-2-empty vector, CNE-2-miR-608-C, andCNE-2-miR-608-G cells with similar results (Fig. 4A).

Effects of themiR-608 rs4919510C>Ggenotypes on tumorgrowthWe evaluated the effects of radiation on tumor growth in

different animal models. During a 1-week period, all micewere injected with CNE-1-miR-608-C, CNE-1-miR-608-G, orCNE-1-empty vector cells developed tumors, after whichmice were assigned to receive 2, 10, or 20 Gy of localradiation. As shown in Fig. 4B, no significant tumor growthinhibition was observed in CNE-1-empty vector, CNE-1-miR-608-C, or CNE-1-miR-608-G xenografts that were locallyirradiated with 2Gy of radiation (P ¼ 0.625). In contrast, itwas showed that the CNE-1-miR-608-G xenografts grewfaster than the CNE-1-miR-608-C xenografts after 10-Gyirradiation (average volumes �SEM; 593 � 59.49 mm3 vs.891 � 93.44 mm3; P ¼ 0.009). Unfortunately, approximately70% of the mice died after being locally irradiated at a doseof 20 Gy (Fig. 4B). The same experiments were repeatedusing the CNE-2-empty vector, CNE-2-miR-608-C, and CNE-2-miR-608-G cell xenografts with similar results (Fig. 4B).

DiscussionIn the present study, we investigated the effects of SNPs in

mature miRNA sequences on LRR in patients with naso-pharyngeal carcinoma after radiotherapy. We found thatcarrying at least one G allele (GC, GG) of the miR-608rs4919510C>G SNP significantly increased the risk of LRRcompared with that in patients carrying a homozygous

C allele. Importantly, these results remained significantafter adjustment for other potential predictors of patientoutcome in this patient cohort study, and our functionalresults were also consistent with these findings. This studyrepresents the first finding that the miR-608 rs4919510C>GSNP may serve as a predictive marker for the LRR ofnasopharyngeal carcinoma in patients who were treatedwith radiotherapy.

Nasopharyngeal carcinoma is highly sensitive to radiationthat alone for early nasopharyngeal carcinoma can achieve arelatively high cure rate; however, its efficacy is disappointingfor locally advanced disease. It is well known that recurrence inthe nasopharynx is one of the important causes of treatmentfailure (36, 37); therefore, assessments of recurrence are crucialfor the selection of appropriate treatment. Previous reportsindicated that miRNAs were aberrantly expressed in nasopha-ryngeal carcinoma compared with that in normal epithelialtissue, and this aberrant expression promoted an aggressivetumor phenotype by changing the expression of mRNA targets(38–40). Therefore, miRNA-related SNPs may be used individ-ually and jointly to predict the risk of recurrence of early-stagehead and neck cancer (41). Liu and colleagues suggested thatsome important miRNAs had a significant value for determin-ing the survival prognosis in addition to nasopharyngealcarcinoma development and progression (42). As the impor-tant roles of miRNAs in cancer are gradually being revealed,their potential applications as predictive markers and treat-ment targets have generated great interest for cancer diagno-sis, classification, prognosis, risk factor evaluation, and therapystrategies.

Figure 3. A, the 20 selected genes mRNA expression levels in tissue samples from patients with nasopharyngeal carcinoma as function of miR-608rs4919510C>Ggenotypes (7 rs4919510CC, 17 rs4919510CG, and 11 rs4919510GG); data aremean�SEM, normalized tob-actin. Relative luciferase activityof the psiCHECK-2-FBXO32-30-UTR construct cotransfected with miR-608 containing different rs4919510C>G alleles in CNE-1 (B) and CNE-2 cells (C).Renilla luciferase activitywasmeasured andnormalized to firefly luciferase. Six replicateswere carried out for eachgroup, and the experimentwas repeated atleast three times. Data are mean� SEM. D, effect of themiR-608 rs4919510C>GSNP on X-ray–induced chromatid b/c in peripheral blood lymphocytes from125 control subjects. The levels of chromatid breaks in controls with different genotypes of rs4919510C>G were analyzed with ANOVA test.






A series of epidemiologic studies revealed that the miR-608rs4919510C>G SNP has a key role in cancer progression (17),and evidence of its influence on prognosis has also accumu-lated recently (18, 34, 43). However, the importance of themiR-608 rs4919510C>G SNP in the LRR of nasopharyngeal carci-noma after radiotherapy remains unclear, and the biologicfunctions of this SNP have not yet been elucidated. Thers4919510C>G SNP is located within the mature sequence ofmiR-608 and at the joint of the stemwith the canonical hairpinloop (18). Because this rigid secondary structure is a requisitefor recognition, and thus processing, of pre-miRNA by theRNase Drosha, structural disruptions at this critical point mayaffect recognition or subsequent processing. Each miRNA hashundreds of targets, and thus, a singular change in a mature

miRNA sequence could have an exponentially large effect onprotein output, perhaps an effect sufficient to skew, evenslightly, the clinical outcome of cancer. Our results indicatedthat the rs4919510C>G SNP might influence the expression ofmiR-608 target genes, which include immunity and defensegenes, DNA repair genes, cell growth–related genes, tumorinvasion and metastasis-related genes, cancer stem cell–relat-ed genes, and cell death–related genes. These genes could haveconsequences directly related to cancer cell survival and tumorgrowth, thereby influencing the LRR of nasopharyngeal carci-noma after radiotherapy. Moreover, the X-ray radiationinduced more chromatid breaks in lymphocyte cells fromrs4919510CC carriers than in those from subjects with othergenotypes. As we know, medical radiation can cause DNA

A

0 1 2 3 4 5 6

Day

Subcutaneous injection Radiation

2 Gy 2 Gy

10 Gy 10 Gy

20 Gy 20 Gy

0 3 7 11 15 19

0 1 2 3

Time (d)

Time (d)

Time (d)

Tum

or

volu

me

(mm

3 )

Tum

or

volu

me

(mm

3 )

Tum

or

volu

me

(mm

3 )

Tum

or

volu

me

(mm

3 )

Tum

or

volu

me

(mm

3 )

Tum

or

volu

me

(mm

3 )

0

2,000

1,500

1,000

500

0

2,000

1,500

1,000

500

0

2,000

1,500

1,000

500

0

2,000

1,500

1,000

500

0

600

400

200

0

800

600

400

200

0

1 3 5 7 9 11 13 15 17 19

0 1 3 5 7 9 11 13 15 17 19

0 1 3 5 7 9 11 13 15 17 19 0 1 3 5 7 9 11 13 15 17 19

0 1 3 5 7 9 11 13 15 17 19

0 1 3 5 7 9 11 13 15 17 19Time (d)

Time (d) Time (d)

Time (d) Time (d)

Cel

l nu

mb

er (

10E

4)C

ell n

um

ber

(10

E4)

50

40

30

20

10

0

50

40

30

20

10

04 5 6

B

CNE-1-miR-608-C

CNE-1-miR-608-G

CNE-1-empty vector

CNE-1-miR-608-CCNE-1-miR-608-GCNE-1-empty vector

CNE-2-miR-608-CCNE-2-miR-608-GCNE-2-empty vector

CNE-2-miR-608-C

CNE-2-miR-608-G

CNE-2-empty vector

Figure 4. A, CNE-1-empty vector, CNE-1-miR-608-C, CNE-1-miR-608-G, CNE-2-empty vector, CNE-2-miR-608-C, and CNE-2-miR-608-G cells weresubjected to 2 Gy of radiation to examine the effect on cell growth. Cells were seeded in 24-well culture plate 24 hours before irradiation. Cellnumbers were counted at different times after irradiation. Cell multiplication of CNE-1-empty vector and CNE-2-empty vector cells were used ascontrols. Three experiments were carried out; points, mean; bars, SEM. B, subcutaneously implanted CNE-1-empty vector, CNE-1-miR-608-C, CNE-1-miR-608-G, CNE-2-empty vector, CNE-2-miR-608-C, and CNE-2-miR-608-G cells xenografted tumors were established. During a 1-week period, all micedeveloped tumors, after which mice were assigned to receive 2, 10, or 20 Gy of local radiation. Each point represents the mean tumor volume. Bars,SEM. Mean tumor volume from 6 nude mice of each group are shown.

Zheng et al.





double-strand breaks in cancer cells and suppress cancer cellgrowth (44), which is consistent with the rs4919510CCvariant genotype having low DNA repair genes expression anddeficient DNA repair capacity, and the association betweenrs4919510CC variant genotype and medical radiation ondecreasing LRR risk of nasopharyngeal carcinoma is bio-logically plausible. Our results also indicated that miR-608rs4919510C>G exhibited an association with the OS;however, there was no significant difference in time fromrecurrence to death among patients carrying different miR-608 rs4919510C>G genotypes. All these results suggested thatmiR-608 rs4919510C>G was associated with higher risk of LRRand then the TTR in advance, which finally may lead to thehigher risk of all-cause mortality. Therefore, our present studydetected the importance of mature miRNA sequence SNPs inthe LRR of nasopharyngeal carcinoma after radiotherapy.FBXO32 (also known as atrogin-1) is a member of the F-box

protein family and constitutes 1 of 4 subunits of the ubiquitinprotein ligase complex (45, 46). FBXO32 has been reported toplay a role in muscle atrophy (47), and recent findings havesuggested that FBXO32 is a novel apoptosis regulator that isnegatively regulated by a prosurvival signal (48, 49). Interest-ingly, Tan and colleagues also showed that FBXO32 wastranscriptionally silenced by epigenetic mechanisms in cancercells (48). Furthermore, Chou and colleagues found thatFBXO32 might be a novel tumor suppressor gene that wasassociated with poor prognosis in human ovarian cancer (50).Our reporter gene assays suggested that the rs4919510C alleleaffected FBXO32 expression by inhibiting the binding ofmiR-608 in nasopharyngeal carcinoma cell lines, and theseresults are consistent with previous findings on the contribu-tions of FBXO32 to tumor suppression.In summary, our preliminary study provides the first

evidence that miR-608 rs4919510C>G may be a predictivemarker to identify patients with a high risk of nasopharyn-geal carcinoma recurrence, and these data may help topredict response in a subgroup of patients treated withradiotherapy. In addition, this may help to select subgroups

of patients with nasopharyngeal carcinoma who may benefitfrom newly developed gene-therapy strategies. Nevertheless,a median follow-up time of approximately 3 years for bothpopulations included in the present studies may not besufficient to identify all of the recurrences that would occur,and further validation of our hypothesis-generating findingsin prospective biomarker-embedded clinical trials is needed.Therefore, large patient cohort studies are warranted tofurther confirm our results.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: J. Zheng, J. Lu, Y. ZhouDevelopment of methodology: J. Zheng, L. Zhang, M. Hu, H. Wu, W. Li, J. Lu,Y. ZhouAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): J. Zheng, J. Deng, M. Xiao, L. Yang, L. Zhang, Y. You,W. Li, J. Lu, Y. ZhouAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): J. Zheng, J. Deng, Y. You, N. Li, J. Lu, Y. ZhouWriting, review, and/or revision of the manuscript: J. Zheng, J. Deng,H. Wu, J. Lu, Y. ZhouAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): J. Zheng, J. Deng, L. Zhang, M. Hu,N. Li, J. Lu, Y. ZhouStudy supervision: J. Lu, Y. Zhou

Grant SupportThis study was supported by the National Scientific Foundation of China

grants 81001278, 81171895, and 81072366; a project funded by the PriorityAcademic Program Development of Jiangsu Higher Education Institutions,Jiangsu Provincial Natural Science Foundation (no. BK2011297); Jiangsu Prov-ince Science and Technology Support Program (no. BE2012648); the ScientificResearch Foundation for the Returned Overseas Chinese Scholars, State Edu-cation Ministry (no. 20101561), Suzhou Key Laboratory for Molecular CancerGenetics (no. SZS201209); and Zhejiang Provincial Natural Science Foundationof China (Y2100449 and Y2110410).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicate thisfact.

Received February 10, 2013; revised May 22, 2013; accepted June 2, 2013;published OnlineFirst June 24, 2013.

References1. Bei JX, Li Y, Jia WH, Feng BJ, Zhou G, Chen LZ, et al. A genome-wide

association study of nasopharyngeal carcinoma identifies three newsusceptibility loci. Nat Genet 2010;42:599–603.

2. Chang CM, Yu KJ, Mbulaiteye SM, Hildesheim A, Bhatia K. Theextent of genetic diversity of Epstein–Barr virus and its geographicand disease patterns: a need for reappraisal. Virus Res 2009;143:209–21.

3. Cho WC. Nasopharyngeal carcinoma: molecular biomarker discoveryand progress. Mol Cancer 2007;6:1.

4. Farias TP, Dias FL, Lima RA, Kligerman J, de Sa GM, Barbosa MM,et al. Prognostic factors and outcome for nasopharyngeal carcinoma.Arch Otolaryngol Head Neck Surg 2003;129:794–9.

5. Wang J, Shi M, Hsia Y, Luo S, Zhao L, Xu M, et al. Failure patterns andsurvival in patients with nasopharyngeal carcinoma treated with inten-sity modulated radiation in Northwest China: a pilot study. RadiatOncol 2012;7:2.

6. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and func-tion. Cell 2004;116:281–97.

7. Croce CM. Causes and consequences of microRNA dysregulation incancer. Nat Rev Genet 2009;10:704–14.

8. Lee Y, Jeon K, Lee JT, Kim S, Kim VN. MicroRNA maturation:stepwise processing and subcellular localization. EMBO J 2002;21:4663–70.

9. Lee Y, Ahn C, Han J, Choi H, Kim J, Yim J, et al. The nuclear RNase IIIDrosha initiates microRNA processing. Nature 2003;425:415–9.

10. Ryan BM, Robles AI, Harris CC. Genetic variation in microRNA net-works: the implications for cancer research. Nat Rev Cancer 2010;10:389–402.

11. Bartel DP. MicroRNAs: target recognition and regulatory functions.Cell 2009;136:215–33.

12. Griffiths-Jones S, Grocock RJ, van Dongen S, Bateman A, Enright AJ.miRBase: microRNA sequences, targets and gene nomenclature.Nucleic Acids Res 2006;34:D140–4.

13. Guo H, Ingolia NT, Weissman JS, Bartel DP. Mammalian microRNAspredominantly act to decrease target mRNA levels. Nature 2010;466:835–40.

14. Thai TH, Christiansen PA, Tsokos GC. Is there a link between dysre-gulatedmiRNAexpression anddisease?DiscovMed2010;10:184–94.

15. Iorio MV, Croce CM. MicroRNAs in cancer: small molecules with ahuge impact. J Clin Oncol 2009;27:5848–56.






16. Tse KP, SuWH, Chang KP, Tsang NM, Yu CJ, Tang P, et al. Genome-wide association study reveals multiple nasopharyngeal carcinoma-associated loci within the HLA region at chromosome 6p21.3. Am JHum Genet 2009;85:194–203.

17. HuangAJ, YuKD, Li J, FanL, ShaoZM.Polymorphism rs4919510:C>Ginmature sequence of humanmicroRNA-608 contributes to the risk ofHER2-positive breast cancer but not other subtypes. PLoS ONE2012;7:e35252.

18. Ryan BM, McClary AC, Valeri N, Robinson D, Paone A, Bowman ED,et al. rs4919510 in hsa-mir-608 is associatedwith outcomebut not riskof colorectal cancer. PLoS ONE 2012;7:e36306.

19. ChristensenBC, Avissar-WhitingM,Ouellet LG, Butler RA, NelsonHH,McClean MD, et al. Mature microRNA sequence polymorphism inMIR196A2 is associated with risk and prognosis of head and neckcancer. Clin Cancer Res 2010;16:3713–20.

20. Langendijk JA, Leemans CR, Buter J, Berkhof J, Slotman BJ. Theadditional value of chemotherapy to radiotherapy in locally advancednasopharyngeal carcinoma: a meta-analysis of the published litera-ture. J Clin Oncol 2004;22:4604–12.

21. Zheng J, Jiang L, Zhang L, Yang L, Deng J, You Y, et al. Functionalgenetic variations in the IL-23 receptor gene are associatedwith risk ofbreast, lung and nasopharyngeal cancer in Chinese populations.Carcinogenesis 2012;33:2409–16.

22. Zheng J, Liu B, Zhang L, Jiang L, Huang B, You Y, et al. The protectiverole of polymorphism MKK4-1304 T>G in nasopharyngeal carcinomais modulated by Epstein–Barr virus' infection status. Int J Cancer2012;130:1981–90.

23. Ma J, Mai HQ, Hong MH, Min HQ, Mao ZD, Cui NJ, et al. Results of aprospective randomized trial comparing neoadjuvant chemotherapyplus radiotherapy with radiotherapy alone in patients with locoregion-ally advanced nasopharyngeal carcinoma. J Clin Oncol 2001;19:1350–7.

24. Shin DM, Lee JS, Lippman SM, Lee JJ, Tu ZN, Choi G, et al. p53expressions: predicting recurrence andsecondprimary tumors in headand neck squamous cell carcinoma. J Natl Cancer Inst 1996;88:519–29.

25. Jiang L, Zhang C, Li Y, Yu X, Zheng J, Zou P, et al. A non-synonymouspolymorphism Thr115Met in the EpCAM gene is associated with anincreased risk of breast cancer in Chinese population. Breast CancerRes Treat 2011;126:487–95.

26. Jiang L, Deng J, Zhu X, Zheng J, You Y, Li N, et al. CD44 rs13347 C>Tpolymorphism predicts breast cancer risk and prognosis in Chinesepopulations. Breast Cancer Res 2012;14:R105.

27. Hwang-Verslues WW, Chang PH, Wei PC, Yang CY, Huang CK, KuoWH, et al. miR-495 is upregulated by E12/E47 in breast cancer stemcells, and promotes oncogenesis and hypoxia resistance via down-regulation of E-cadherin and REDD1. Oncogene 2011;30:2463–74.

28. Zhou J, Wang W. Analysis of microRNA expression profiling identifiesmicroRNA-503 regulates metastatic function in hepatocellular cancercell. J Surg Oncol 2011;104:278–83.

29. Qi HH, Sarkissian M, Hu GQ, Wang Z, Bhattacharjee A, Gordon DB,et al. Histone H4K20/H3K9 demethylase PHF8 regulates zebrafishbrain and craniofacial development. Nature 2010;466:503–7.

30. QuY,Wang J, RayPS,GuoH,Huang J, Shin-SimM, et al. Thioredoxin-like 2 regulates human cancer cell growth and metastasis via redoxhomeostasis and NF-kappaB signaling. J Clin Invest 2011;121:212–25.

31. LuJ,WangLE, XiongP,SturgisEM,SpitzMR,WeiQ. 172G>Tvariant inthe 50 untranslated region of DNA repair gene RAD51 reduces risk ofsquamous cell carcinoma of the head and neck and interacts with aP53 codon 72 variant. Carcinogenesis 2007;28:988–94.

32. Yang L, Li Y, Cheng M, Huang D, Zheng J, Liu B, et al. A functionalpolymorphism at microRNA-629-binding site in the 30-untranslatedregion of NBS1 gene confers an increased risk of lung cancer inSouthern and Eastern Chinese population. Carcinogenesis 2012;33:338–47.

33. BolzanAD,BianchiMS,GimenezEM, FlaqueMC,Ciancio VR.Analysisof spontaneous and bleomycin-induced chromosome damage inperipheral lymphocytes of long-haul aircrewmembers from Argentina.Mutat Res 2008;639:64–79.

34. Lin M, Gu J, Eng C, Ellis LM, Hildebrandt MA, Lin J, et al. Geneticpolymorphisms in microRNA-related genes as predictors of clinicaloutcomes in colorectal adenocarcinoma patients. Clin Cancer Res2012;18:3982–91.

35. Dossus L,McKay JD, Canzian F,Wilkening S, Rinaldi S, BiessyC, et al.Polymorphisms of genes coding for ghrelin and its receptor in relationto anthropometry, circulating levels of IGF-I and IGFBP-3, and breastcancer risk: a case–control study nested within the European Pro-spective Investigation into Cancer and Nutrition (EPIC). Carcinogen-esis 2008;29:1360–6.

36. Yi JL, Gao L, Huang XD, Li SY, Luo JW, Cai WM, et al. Nasopha-ryngeal carcinoma treated by radical radiotherapy alone: ten-yearexperience of a single institution. Int J Radiat Oncol Biol Phys2006;65:161–8.

37. Chua DT, Sham JS, Kwong DL, Au GK. Treatment outcome afterradiotherapy alone for patients with stage I–II nasopharyngeal carci-noma. Cancer 2003;98:74–80.

38. Sengupta S, den Boon JA, Chen IH, NewtonMA, Stanhope SA, ChengYJ, et al. MicroRNA 29c is down-regulated in nasopharyngeal carci-nomas, up-regulating mRNAs encoding extracellular matrix proteins.Proc Natl Acad Sci U S A 2008;105:5874–8.

39. Chen HC, Chen GH, Chen YH, Liao WL, Liu CY, Chang KP, et al.MicroRNA deregulation and pathway alterations in nasopharyngealcarcinoma. Br J Cancer 2009;100:1002–11.

40. Li T, Chen JX, Fu XP, Yang S, Zhang Z, Chen Kh H, et al. microRNAexpression profiling of nasopharyngeal carcinoma. Oncol Rep2011;25:1353–63.

41. Lin J, Horikawa Y, Tamboli P, Clague J, Wood CG, Wu X. Geneticvariations in microRNA-related genes are associated with survival andrecurrence in patients with renal cell carcinoma. Carcinogenesis2010;31:1805–12.

42. Liu N, Chen NY, Cui RX, LiWF, Li Y,Wei RR, et al. Prognostic value of amicroRNA signature in nasopharyngeal carcinoma: a microRNAexpression analysis. Lancet Oncol 2012;13:633–41.

43. Xing J, Wan S, Zhou F, Qu F, Li B, Myers RE, et al. Genetic poly-morphisms in pre-microRNA genes as prognostic markers of colo-rectal cancer. Cancer Epidemiol Biomarkers Prev 2012;21:217–27.

44. Harper JV, Anderson JA, O'Neill P. Radiation induced DNA DSBs:contribution from stalled replication forks? DNA Repair 2010;9:907–13.

45. Bodine SC, Latres E, Baumhueter S, Lai VK, Nunez L, Clarke BA, et al.Identification of ubiquitin ligases required for skeletal muscle atrophy.Science 2001;294:1704–8.

46. Li HH, Kedar V, Zhang C, McDonough H, Arya R, Wang DZ, et al.Atrogin-1/muscle atrophy F-box inhibits calcineurin-dependent car-diac hypertrophy by participating in an SCF ubiquitin ligase complex.J Clin Invest 2004;114:1058–71.

47. Hanai J, CaoP, Tanksale P, Imamura S, Koshimizu E, Zhao J, et al. Themuscle-specific ubiquitin ligase atrogin-1/MAFbx mediates statin-induced muscle toxicity. J Clin Invest 2007;117:3940–51.

48. TanJ,YangX,ZhuangL, JiangX,ChenW,LeePL, et al. Pharmacologicdisruption of Polycomb-repressive complex 2-mediated gene repres-sion selectively induces apoptosis in cancer cells. Genes Dev2007;21:1050–63.

49. Stitt TN, Drujan D, Clarke BA, Panaro F, Timofeyva Y, Kline WO, et al.The IGF-1/PI3K/Akt pathway prevents expression of muscle atrophy-induced ubiquitin ligases by inhibiting FOXO transcription factors. MolCell 2004;14:395–403.

50. Chou JL, Su HY, Chen LY, Liao YP, Hartman-Frey C, Lai YH, et al.Promoter hypermethylation of FBXO32, a novel TGF-beta/SMAD4target gene and tumor suppressor, is associated with poor prognosisin human ovarian cancer. Lab Invest 2010;90:414–25.

Zheng et al.





2013;73:5151-5162. Published OnlineFirst June 24, 2013.Cancer Res Jian Zheng, Jieqiong Deng, Mang Xiao, et al. Radiotherapy for Nasopharyngeal Carcinoma

Predicts Recurrence aftermiR-608A Sequence Polymorphism in

Updated version

10.1158/0008-5472.CAN-13-0395doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2013/06/25/0008-5472.CAN-13-0395.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/73/16/5151.full#ref-list-1

This article cites 50 articles, 10 of which you can access for free 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://cancerres.aacrjournals.org/content/73/16/5151To request permission to re-use all or part of this article, use this link


http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-0395http://cancerres.aacrjournals.org/content/suppl/2013/06/25/0008-5472.CAN-13-0395.DC1http://cancerres.aacrjournals.org/content/73/16/5151.full#ref-list-1http://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/73/16/5151http://cancerres.aacrjournals.org/

Documents

A Sequence Polymorphism in miR-608 Predicts Recurrence … · Epstein–Barr virus (EBV) infection status, histologic grade, pathologic stage, and treatment. Immunoglobulin-A antibo-dies