jpc70014_425_436

7/29/2019 jpc70014_425_436

1/12

PRELIMINARY

COMMUNICATION

MicroRNA Expression Profiles AssociatedWith Prognosis and Therapeutic Outcomein Colon Adenocarcinoma

Aaron J. Schetter, PhD, MPH

Suet Yi Leung, MD

Jane J. Sohn, PhD

Krista A. Zanetti, PhD, MPHElise D. Bowman, MS

Nozomu Yanaihara, MD, PhD

Siu Tsan Yuen, MD

Tsun Leung Chan, MD

Dora L. W. Kwong, MD

Gordon K. H. Au, MD

Chang-Gong Liu, PhD

George A. Calin, MD, PhD

Carlo M. Croce, MD

Curtis C. Harris, MD

COLON ADENOCARCINOMA IS A

major cause of cancer mor-tality worldwide.1 Colorec-tal cancer is the third most

common and second leading cause ofcancer death in theUnited States.2 Eventhough 5-year mortality rates have mod-estly declined over the last 3 decades,3

there is still a need to identify new prog-nostic biomarkers and therapeutic tar-getsfor thisdisease. Currently, chemo-therapy has significant therapeutic

value, but surgery is the only curativeform of treatment.4

Ideal therapeutic targets should becausally associatedwith diseaseandame-nable to designing therapeutic interven-tions, whereas ideal biomarkers shouldbe easy to measure and have strong as-sociations with clinical outcomes. Mi-croRNAs could match both criteria.5-8

Author Affiliations: Laboratory of Human Carcino-genesis, Center for Cancer Research (Drs Schetter,Sohn, Zanetti, Yanaihara, and Harris and Ms Bow-man)and Cancer Prevention FellowshipProgram, Of-ficeof PreventiveOncology(Drs Schetter andZanetti),National CancerInstitute, National Institutes of Health,Bethesda, Maryland; Departments of Pathology (DrsLeung, Yuen, and Chan) and Clinical Oncology (DrsKwong and Au), University of Hong Kong, QueenMaryHospital, Pokfulam, HongKong; Department of

Molecular Virology, Immunology, and Medical Ge-netics, Ohio State University Comprehensive CancerCenter, Columbus, Ohio (Drs LiuandCroce); and De-partmentof Experimental Therapeutics,M. D. Ander-son Cancer Center, Houston, Texas (Dr Calin).Corresponding Author: Curtis C. Harris, MD, Labo-ratory of Human Carcinogenesis, Center for CancerResearch, National Cancer Institute, Bldg 37, Room3 0 6 8 A, B e th e sd a , M D 2 0 8 9 2 (Cu rt is_ Ha rr [email protected]).

Context MicroRNAs have potential as diagnostic biomarkers and therapeutic tar-gets in cancer. No study has evaluated the association between microRNA expressionpatterns and colon cancer prognosis or therapeutic outcome.

Objective To identify microRNA expression patterns associated with colon adeno-carcinomas, prognosis, or therapeutic outcome.

Design, Setting, and Patients MicroRNA microarray expression profiling of tu-mors and paired nontumorous tissues was performed on a US test cohort of 84 pa-tients with incident colon adenocarcinoma, recruited between 1993 and 2002. Weevaluated associations with tumor status, TNM staging, survival prognosis, and re-sponse to adjuvant chemotherapy. Associations were validated in a second, indepen-dent Chinese cohort of 113 patients recruited between 1991 and 2000, using quan-titative reverse transcription polymerase chain reaction assays. Thefinal date of follow-upwas December 31, 2004, for the Maryland cohort and August 16, 2004, for the HongKong cohort.

Main Outcome Measures MicroRNAs that were differentially expressed in tu-mors and microRNA expression patterns associated with survival using cancer-specific death as the end point.

Results Thirty-seven microRNAs were differentially expressed in tumors from thetest cohort. Selected for validation were miR-20a, miR-21, miR-106a, miR-181b, andmiR-203, and all 5 were enriched in tumors from the validation cohort (P.001). HighermiR-21 expression was present in adenomas (P = .006) and in tumors with more ad-vanced TNM staging (P.001). In situ hybridization demonstrated miR-21 to be ex-pressed at high levels in colonic carcinoma cells. The 5-year cancer-specific survivalrate was 57.5% for the Maryland cohort and was 49.5% for the Hong Kong cohort.High miR-21 expression was associated with poor survival in both the training (haz-ard ratio, 2.5; 95% confidence interval, 1.2-5.2) and validation cohorts (hazard ratio,2.4; 95% confidence interval, 1.4-3.9), independent of clinical covariates, includingTNM staging, and was associated with a poor therapeutic outcome.

Conclusions Expression patterns of microRNAs are systematically altered in colonadenocarcinomas. High miR-21 expression is associated with poor survival and poortherapeutic outcome.

JAMA. 2008;299(4):425-436 www.jama.com

2008 American Medical Association. All rights reserved. (Reprinted with Corrections) JAMA, January 30, 2008Vol 299, No. 4 425

wnloaded From: http://jama.jamanetwork.com/ on 09/06/2013

7/29/2019 jpc70014_425_436

2/12

MicroRNAs are 18- to 25-nucleo-tide, noncoding RNA molecules thatregulate the translation of many genes.9

Since their discovery,10,11 microRNAshave been found to regulate a variety ofcellular processes including apopto-

sis,

12-14

differentiation,

10,11,15

andcell pro-liferation.16 MicroRNAs may also have acausal role in carcinogenesis.5,17,18 Mi-croRNA expression levels are altered inmost tumor types,19,20 including colontu-mors.20-23 Experimental manipulationofspecific microRNAsmodulatestumorde-velopment in mouse-model sys-tems.16,24-26 The prognostic potential ofmicroRNAs has also been demon-strated for chronic lymphocytic leuke-mia,6 lung cancer,7 pancreatic cancer,27

and neuroblastomas.28

If aberrant microRNA expression is

causal to carcinogenesis, inhibiting spe-cific microRNAs may have therapeuticimplications. Modified antisense oligo-nucleotidescaneasilybedesignedtospe-cificallyinhibitmicroRNAfunction.29An-tagomirsare onetypeof antisenseoligo-nucleotide that has proven effective atinhibiting microRNA functionin vivoinmice.30 Theeaseofdesigningspecificin-hibitors of microRNA function makesthem candidates for therapeutictargets.

Given the therapeutic and prognos-tic potential for microRNAs in cancer,

we evaluated microRNA profiles of co-lontumorsandpairednontumorous tis-sue to study their potential role in tu-mor formation, diagnosis, and responseto chemotherapy in colon carcinoma.

METHODS

Tissue Collection

and RNA Isolation

Pairs of primary colon tumor and adja-centnontumoroustissuescamefrom 84patientsrecruitedfrom theUniversity ofMaryland Medical Center or Baltimore

Veterans Affairs MedicalCenterbetween1993and2002,andfrom113patientsre-cruited from Queen Mary Hospital inHong Kong between 1991 and 2000.Cases withfamilial adenomatouspolyp-osis or human nonpolyposis colorectalcancer were excluded from this study.

Tissues were flash frozen after sur-gery. Detailed backgrounds for each tis-

sue donor, including age, sex, clinicalstaging, tumor location, survival timesfrom diagnosis, and receipt of adjuvantchemotherapy havebeencollected. Thefinal date offollow-upwas December 31,2004 for the Maryland cohort and Au-

gust 16, 2004, for the Hong Kong co-hort. Tumor histopathology was classi-fied according to the World HealthOrganization Classification of Tumorsystem.1 Adenoma tissue was obtainedfrom the Cooperative Human TissueNetwork.This study wasapprovedbytheInstitutional Review Board of the Na-tional Institutes of Health, the Institu-tional ReviewBoardof the University ofHong Kong/Hospital Authority HongKong West Cluster, and the Institu-tional Review Board for Human SubjectResearch at the University of Maryland.

Race wasself-reportedas white or black.

RNA Isolation

and MicroRNA Profiling

RNA from frozen tissue samples was ex-tracted using standard TRIZOL (Invit-rogen, Carlsbad, California) methods.MicroRNAmicroarray profilingwasper-formed as previously described.31 Briefly,5 g of total RNA was labeled and hy-bridized to each microRNA microarray(Ohio State microRNA microarray ver-sion2.0,Columbus) containingquadru-

plicates of389humanmicroRNA probes.Tumor/nontumorous pairs of tissueswere profiled at the same time. Slideswere scanned using a PerkinElmerScanArray LX5Kscanner (Perkin Elmer,Waltham, Massachusetts).

Microarray Analysis

The data discussed in this publicationhave been deposited in National Cen-ter for Biotechnology Informations(NCBIs) Gene Expression Omnibus(NCBIGEO GSE7828). The data were

preprocessed by the statistical soft-ware R 2.5.0 (R Foundation for Statis-tical Computing, Vienna, Austria)to re-move probes with higher backgroundintensities than foreground and probeswith inconsistent measurements acrossthe quadruplicates. The data were nor-malized by locally weighted scatter plotsmoothing LOESS and imported into

Biometric Research Branch (BRB) ar-ray tools 3.5.0 (http://linus.nci.nih.gov/BRB-Arr ay Tools. html) for subse-quent microarray. Probes with valuesmissing from more than 20% of the ar-rays were removed from the analysis

leaving 230 probes. This filteringmethod was decided a priori to elimi-nate probes whose microRNAs expres-sion levels were thought to be unreli-able. Class comparison analysis usingpaired t tests identified microRNAs thatwere differentially expressed in tu-mors (P .001). Class prediction algo-rithms in BRB array tools were used todetermine whether microRNA micro-array expression patterns could accu-rately differentiate between tumor andpaired nontumor tissue. For theseanalyses, 3 nearest neighbors and near-

est centroid algorithms were arbi-trarily chosen and percent accuracy re-portsthe percentage of tissues that werecorrectly identified. These algorithmswere also used for quantitative reversetranscription polymerase chain reac-tion (RT-PCR) data in the Hong Kongvalidation cohort.

To initially search for microRNAsassociated with poor survival, tumor:nontumor (T:N) microRNA expres-sion ratios were analyzed in the Mary-land cohort using microarray data.

Tumor:nontumor expression ratios formicroRNAs were created by subtract-ing the log2 nontumor from the log2tumor expression values. MicroRNAsmissing more than 25% of T:N ratioswere filtered out leaving 208. Expres-sion data were dichotomized intoclearly defined high and low groups toexamine associations with microRNAexpression and survival. Tumor:nontumor expression ratios weredichotomized with the highest tertileclassified as high and the lower 2 ter-

tiles classified as low. This cutoff wasset based on associations within thetest cohort prior to analyzing thevalidation cohort. Once set, this high-low cutoff was used universallythroughout this study. To analyzeassociations with tumor expressionand nontumor expression with sur-vival using microarray data, the array

MICRORNA EXPRESSION PROFILES AND COLON ADENOCARCINOMA

426 JAMA, January 30, 2008Vol 299, No. 4 (Reprinted with Corrections) 2008 American Medical Association. All rights reserved.


7/29/2019 jpc70014_425_436

3/12

data had to be normalized based onthe day of microarray profiling toremove systematic bias introducedfr o m the day - to - day v ar i ab i l i tyobserved in the microarray dataacquisition. To do this, for each given

day, the highest one-third expressingvalues were labeled high and the low-est two-thirds were labeled low, con-sistent with the predetermined cutoffthat was used for this study.

Quantitative RT-PCR

Quantitative RT-PCR of microRNAswas performed using Taqman Mi-croRNA assays (Applied Biosystems,Foster City, California) according to themanufacturers instructions with the7500 real-time RT-PCR system (Ap-plied Biosystems, Foster City) using ex-

pression levels of the small nuclearRNA, U6B, as the normalization con-trol. All assays were performed in du-plicate (miR-20a, miR-203) or tripli-cate (miR-21, miR-106a, miR-181b).Quantitative RT-PCR formiR-21, miR-106a, and miR-181bwas performed byoneof the investigators(A.J.S.) whowasblinded to the survival outcomes andclinical data for members of the vali-dation cohort.

In Situ Hybridization

In situ hybridization was performedw i th pr o b es fo r hum an miR-21,scramble, and U6 (Exiqon, Woburn,Massachusetts) with a modified ver-sion of the manufacturers protocol forformalin-fixed paraffin-embedded tis-sue written by W. Kloosterman (http:// ww w. ex iq on .c om /u pl oa ds /L NA_52-_FFPE_miRNA_in_situ_protocol.pdf) on human colon tissue. Modifi-cations included the use of polyclonalrabbit anti-DIG/HRP-conjugated anti-body and DakoCytomation GenPoint

Tyramide Signal Amplification Sys-tem (DakoCytomation, Carpinteria,California), and VECTOR NovaRedsubstrate (Vector Laboratories, Burlin-game, California). Images were takenon an Olympus BX40 microscope usingthe Olympus DP70 digital camera andDP controller software (Olympus,Champaign, Illinois).

Statistical Analysis

Expression graphs and Wilcoxonmatched-pairs tests were used to ana-lyze differences in microRNA expres-sion between tumorsand paired nontu-morous tissue as well as differences

between adenoma and paired nonad-enoma tissue for all quantitative RT-PCR data using Graphpad Prism 4.0(Graphpad Software Inc,SanDiego, Cali-fornia). All trend tests reported are Cu-zick nonparametric test for trend acrossordered groups32 and were performedusing Stata 9.2 (StataCorp LP, CollegeStation, Texas). Associations withprog-nosis in the validation cohort were con-

sidered statistically significant only if theP value were less than .01 to adjust formultiple comparisons testing (5 testsusing a Bonferroni correction).

KaplanMeier analysis was per-formed with WINSTAT 2001 (R Fitch

Software, Bad Krozingen, Germany).Multivariate Cox regression analysiswasperformed using StataCorp 9.2.Forthese models, we dichotomized age as50 years or older vs younger than 50years becausethe recommended screen-ing age for colon cancer is at age 50years; TNM staging was dichotomizedbased on metastasic vs nonmetastasicdisease. One patient in theMaryland co-

Table 1. Characteristics of Study Population and Tumorsa

Maryland Test Cohort(n = 84)

Hong Kong Validation Cohort(n = 113)

Recruitment area Baltimore, Maryland Hong Kong, ChinaAge at enrollment, y

Mean (SD) 64.6 (10.7) 55.8 (15)

Range 32-87 32-84

Sex, No. (%)Men 66 (79) 56 (50)

Women 18 (21) 57 (50)

Race, No. (%)White 52 (62) 0

Black 32 (38) 0

Asian 0 113 (100)

Follow-up time, moMedian 68.0 84.6

Range 26.0-141.9 60.4-147.2

Specific mortality rates, %

1 year 82.1 87.65 year 57.5 49.6

Tumor location, No. (%) b

Distal 48 (59) 90 (80)

Proximal 34 (41) 23 (20)

Adenocarcinoma histology, No. (%)Adenocarcinoma 75 (89) 105 (93)

Mucinous adenocarcinoma 8 (10) 7 (6)

Adenosquamous carcinoma 1 (1) 0 (0)

Signet ring cell and mucinous 0 1 (1)

Adjuvant chemotherapy, No. (%) c

Received 22 (37) 40 (35)

Did not receive 37 (63) 73 (65)

TNM stage, No. (%) d

I 8 (10) 9 (8)

II 29 (34) 37 (33)III 36 (43) 48 (42)

IV 10 (12) 19 (17)a Percentages may not sum to 100 due to rounding.bDistal includes tumors located in or distal to the descending colon. Proximal tumors include tumors in or proximal to the

splenic flexure. Tumor location was available for 82 patients in the original cohort and all those in the validation cohort.c Detailed information pertainingto receipt of chemotherapy wasavailablefor 59 patients in thetest cohort andall those in

the validation cohort. Chemotherapy was primarily fluorouracil-based (in forms of either intravenous fluorouracil or oraldrugs including tegafur with uracil) with or without levamisole or leucovorin.

dFor 1 patient in the Maryland cohort, it was unclear whether that individual had stage III or IV cancer, so this patient wasnot included in the analysis.




7/29/2019 jpc70014_425_436

4/12

hort died onthe day ofsurgery and wasincluded in Kaplan-Meier analysis butremoved for Cox regression analysis.Analyses involving response to adju-

vant therapy included only TNM stageII and III cases because treatment instageIV is palliative care andTNM stageI cases have excellent survival progno-

sisregardlessof therapy. Univariate Coxregression was performed on eachclini-cal covariate to examine influence ofeach on patient survival. Final multi-

Table 2. MicroRNAs That Are Differentially Expressed in Tumors Compared With Nontumorous Tissue

Probe From Microarray Mature miR P Valuea False Discovery Rate, %b Fold Change Chromosomal Location

MicroRNAS With Higher Expression in Tumors

hsa-mir-21No1 miR-21 1 107 0.01 1.7 17q23.2

hsa-mir-021-prec-17No1 miR-21 1 107 0.01 1.8 17q23.2

hsa-mir-092-prec-13092-1No2 miR-92 1 107 0.01 1.4 13q31.3

hsa-mir-222-precNo2 miR-222 1 106 0.01 1.2 Xp11.3

hsa-mir-181b-2No1 miR-181b 2 106 0.01 1.2 9q33.3

hsa-mir-210-prec miR-210 1 105 0.03 1.2 11p15.5

hsa-mir-020-prec miR-20a 3 105 0.06 1.5 13q31.3

hsa-mir-106-prec-X miR-106a 3 105 0.06 1.4 X26.2

hsa-mir-106aNo1 miR-106a 4 105 0.06 1.4 X26.2

hsa-mir-093-prec-7.1093-1 miR-93 4 105 0.06 1.2 7q22.1

hsa-mir-335No2 miR-335 4 105 0.06 1.2 7q32.2

hsa-mir-222-precNo1 miR-222 4

105

0.07 1.2 Xp11.3hsa-mir-338No1 miR-338 6 105 0.07 1.1 17q25.3

hsa-mir-133bNo2 miR-133b 7 105 0.08 1.1 6p12.2

hsa-mir-092-prec-X092-2 miR-92 8 105 0.08 1.4 Xq26.2

hsa-mir-346No1 miR-346 8 105 0.08 1.2 10q23.2

hsa-mir-106bNo1 miR-106b .0002 0.2 1.2 7q22.1

hsa-mir-135-2-prec miR-153a .0002 0.2 1.1 12q23.1

hsa-mir-219-1No2 miR-219 .0003 0.2 1.3 9q34.11

hsa-mir-34aNo1 miR-34a .0003 0.2 1.1 1p36.22

hsa-mir-099b-prec-19No1 miR-99b .0004 0.3 1.1 19q13.41

hsa-mir-185-precNo2 miR-185 .0004 0.3 1.2 22q11.21

hsa-mir-223-prec miR-223 .0004 0.3 1.4 Xq12


hsa-mir-135-1-prec miR-135a .0005 0.3 1.1 3p21.1

hsa-mir-127-prec miR-127 .0005 0.3 1.1 14q32.31


hsa-mir-212-precNo1 miR-212 .0006 0.4 1.1 17p13.3

hsa-mir-095-prec-4 miR-95 .0007 0.4 1.2 4p16.1

hsa-mir-017-precNo2 miR-17-5p .0007 0.4 1.3 13q31.3

MicroRNAs With Reduced Expression in Tumors

hsa-mir-342No2 miR-342 4 106 0.02 0.9 14q32.2

hsa-mir-192-2/3No1 miR-192 9 106 0.03 0.7 11q13.1

hsa-mir-1-2No2 miR-1 2 105 0.06 0.9 18q11.2

hsa-mir-34bNo2 miR-34b 5 105 0.07 0.8 11q23.1

hsa-mir-215-precNo1 miR-215 5 105 0.07 0.7 1q41

hsa-mir-192No1 miR-192 7 105 0.08 0.7 11q13.1

hsa-mir-301No2 miR-301 7 105 0.08 0.7 17q23.2

hsa-miR-324-5pNo2 miR-324-5p .0001 0.1 0.9 17p13.1hsa-mir-030a-precNo2 miR-30a-3p .0002 0.1 0.9 6q13

hsa-mir-1-1No2 miR-1 .0003 0.2 0.9 20q13.33

hsa-mir-34cNo2 miR-34c .0007 0.4 0.9 11q23.1

hsa-mir-331No2 miR-331 .0009 0.5 0.9 12q22

hsa-mir-148bNo2 miR-148b .0009 0.5 0.9 12q13.13aPvalues reported are the result of paired class comparison analysis of microRNA expression patterns from 84 pairs of colon adenocarcinomas and nontumorous tissue using

Biometric Research Branch (BRB) array Tools 3.5.0.b False discovery rate is calculated by BRB array tools. The false discovery rate of 0.5% predicts that this list is 99.5% accurate.




7/29/2019 jpc70014_425_436

5/12

variate models were based on step-wise addition and removal of clinicalcovariates found to be associated withpoor survival in univariate models(P.10). A Wald statistic ofP .05wasused as the criterion for inclusion in fi-

nal multivariate models. All stepwiseaddition models gave the same finalmodels as stepwise removal models. AllP values reported are 2-sided. All uni-variate and multivariate Cox regres-sion models were tested for propor-tional hazards assumptions based onSchoenfeld residuals, and no model vio-lated these assumptions.

RESULTS

MicroRNA Expression Patterns

in Colon Tumors

The characteristics of the patients with

incident colon adenocarcinoma in thetest cohort (from Baltimore) and thevalidationcohort (from Hong Kong) areshown in TABLE 1. The median fol-low-up time was 68.0 months for theBaltimore cohort and 84.6 months forthe Hong Kong cohort. The 2 cohortswere similar in TNMstaging, tumor his-tology, and cancer-specific mortalityrates. The 5-year survival rate was57.5% for the US cohort and 49.5% fortheHong Kong cohort andwere notsig-nificantly different from one another

(P=.49, Kaplan-Meier test). In addi-tion to the racial, geographic, and cul-tural differences between these 2 co-horts, the Baltimore cohort wasconsiderably older (average 64.6 yearsvs 55.8 years) had a higher percentageof men (79% vs 50%).

We compared microRNA profiles of84 pairs of colon tumor and adjacentnontumorous tissues in the Baltimorecohort using microRNA microarrays.31

Tumor microRNA profiles were dis-tinctly different from nontumor pro-

files. Using class comparison analysisinBRB array tools, 37 independent mi-croRNAs were found to be differen-tiallyexpressed in tumors(P.001witha false-discovery rate0.5%;TABLE2).Twenty-six microRNAs were ex-pressed at higher levels in tumors withmiR-21 enriched the most at 1.8-fold.Global microRNA profiles distinguish

between tumor and paired nontumor-ous tissue with 89% accuracy usingeither the 3 nearest neighbors or near-est centroid class prediction algo-rithms within BRB array tools (10-foldcross validation repeated 100 times),suggesting a systematic change in mi-croRNA expression patterns during tu-mor formation.

We next performed a preliminary

analysistoidentifywhetheranyofthe37differentiallyexpressed(P.001)microR-NAswereassociatedwithcancersurvival.We analyzedindividual microRNAT:Nexpression ratios for associations withpoor prognosis in the Maryland test co-hort. Tumor:nontumor microRNA ex-pression ratios were classified as highbasedonhighesttertile.WesearchedforanymicroRNAforwhichhighT:Nratioswere associated with cancer survival(P.05)using Coxregression.Fivemi-croRNAssatisfiedthesecriteria. Highex-

pressionofmiR-20a

(hazard ratio [HR],2.2;95%confidenceinterval[CI],1.1-4.6;P=.03),miR-21(HR,2.5;95%CI,1.2-5.0;P=.01),miR-106a (HR,2.3;95%CI,1.1-4.5;P=.02),miR-181b (HR,2.0;95%CI,1.0-3.9; P=.04), and miR-203 (HR, 3.1;95% CI, 1.5-6.4;P=.003) were each as-sociatedwithpoor survivaland were se-lected for further analysis.

To validate overall differences in mi-croRNA expression between tumor andnontumorous tissue, we measured theexpression levels of these 5 microRNAswith quantitativeRT-PCR in tumor andpaired nontumorous tissue in the inde-pendent validation cohort of 113 pa-tients with incident colon cancer re-cruited from Hong Kong, China(Table 1). MiR-20a (2.3-fold), miR-21

(2.8-fold), miR106a (2.4-fold), miR-181b (1.4-fold), andmiR-203 (1.8-fold)were all expressedat higher levelsin tu-mors (P .001, Wilcoxon matched-pairs test; TABLE 3). Most tumors (89%for miR-20a, 87% for miR-21, 90% formiR-106a, 71% for miR-181b, and 74%for miR-203) had higher expression ofthese microRNAs than paired nontu-morous tissue. Expression patterns forthese 5 microRNAsdistinguish tumor vspaired nontumor statuswith96% or 98%accuracy based on 3 nearest neighbors

or nearest centroid algorithms, respec-tively (10-fold cross validation, re-peated 100 times).

High miR-21 Expression

and Prognosis

Colon adenocarcinomas from 89% to93% of the patients in this study were ofa typical histology. A minority of tu-

Table 3. Expression of MicroRNAs in Colon Adenocarcinoma Tumors and Colon Adenomas

microRNAAverage Differencein Threshold Cyclea

SD (Differencein Threshold Cycle) Fold Change b

P

Valuec

MicroRNA Expression in Tumors vs Paired Nontumorous TissueFrom the Hong Kong Validation Cohortd

miR-20a 1.18 0.97 2.3 .001

miR-21 1.47 1.20 2.8 .001

miR-106a 1.25 0.94 2.4 .001

miR-181b 0.47 1.03 1.4 .001

miR-203 0.83 1.40 1.8 .001

MicroRNA Expression in Adenoma vs Paired Nonadenoma Tissuee

miR-20a 0.11 0.97 0.9 .82

miR-21 0.64 0.90 1.6 .006

miR-106a 0.28 1.22 1.2 .19

miR-181b 0.30 1.24 1.2 .27

miR-203 0.77 1.98 1.7 .14aThreshold cycle is the unit of measurement in quantitative reverse transcription polymerase chain reaction (RT-PCR) to

measure relative gene expression. Average (tumor change in threshold cycle minus paired nontumor change in thresh-old cycle) or average (adenoma change in threshold cycle minus paired nonadenoma change in threshold cycle) fromquantitative RT-PCR. Positive values indicate higher expression in tumor tissue.

bCalcluatedby 2average difference in threshold cycles.c Wilcoxon matched pairs test.d

Forthe tumor/nontumor comparisons, 113pairsof tissueswere usedformiR-20a andmiR-203 while 111pairsof tissuewere used for miR-21, miR-106a, and miR-181b.e For all adenoma/nonadenoma comparisons, 18 pairs of tissue were used.




7/29/2019 jpc70014_425_436

6/12

mors were of mucinous adenocarci-noma (8 of 84 patients [10%] in the UScohort; 7 of 113 patients [6%] in theHong Kong cohort), adenosquamouscarcinoma (1 of 84 patients [1%] in theUS cohort), or signet ring cell carci-

noma (1 of 113 patients [1%] in theHong Kong cohort)histologies (Table1).Different subtypes of adenocarcinomascan be associated with different clinicaloutcomes, including survival progno-sis.33 To remove potential confoundingassociated with histology, we excludedall patients (9 in the US cohort; 8 in theHong Kong cohort) with mucinous ad-enocarcinomas, adenosquamous carci-

nomas, and signet ring cell carcinomasfrom the initial analysis.

We found high T:N expression ratiosfor miR-20a, miR-21, miR-106a, miR-181b,andmiR-203 to be associated withpoor survival in the Maryland test co-

hort. These associationscould be duetomicroRNA expression levels in the tu-mor tissue, the surrounding nontumor-ous tissue, or a combination of both. Todistinguish these possibilities, we ana-lyzed the association of microRNA ex-pression in tumors and paired nontu-morous tissues separately. Highexpression levels in tumors (based onhighesttertile) formiR-20a (HR,2.7; 95%

CI, 1.3-5.8; P=.01), miR-21 (HR, 2.5;95%CI,1.2-5.2;P=.01), miR-106a (HR,2.4; 95% CI, 1.2-5.1; P=.02), miR-181b(HR, 3.2; 95% CI,1.6-6.7;P=.002), andmiR-203 (HR, 3.3; 95% CI, 1.5-7.1;P=.001)were each associatedwith a poor

survivalin theMarylandtest cohort. Nosignificant association with microRNAexpression in nontumorous tissue wasobserved for any of the 5 microRNAs.

The associations with microRNA ex-pression and survival in the test cohortwere made in the context of a microar-rayexperimentin which we were evalu-ating the dichotomized expressionof 37microRNAs. To validate these findings,we used quantitative RT-PCR to mea-sure tumor- and nontumor-expressionlevels for these5 microRNAs intheHongKong validation cohort andanalyzed as-

sociations with prognosis. We dichoto-mizedhigh andlow expression for eachmicroRNA based on highesttertiles,con-sistent with our methods in the test co-hort.HighmiR-21 tumorexpression wasassociated with poor prognosis in theHong Kong validation cohort (P=.001,Kaplan-Meier log-rank test) while ex-pression in nontumoroustissue was not(FIGURE 1), consistent with associa-tions in the Maryland test cohort. Wedid not find statistically significant as-sociations with prognosis and expres-

sion ofmiR-20a, miR-106a, miR-181b,or miR-203 in this cohort.

MultivariateCoxproportionalhazardsanalysiswas used to further evaluate theassociationofmiR-21expressionintumorswithprognosisinboththeMarylandtestcohortandtheHongKongvalidationco-hort (TABLE 4) to evaluate the potentialformiR-21expressionasaprognosticbio-marker.Thedichotomized miR-21expres-sionvalueswerenotassociatedwithage,sex, race, or tumor location (Fisher ex-act test). In univariate analysis for the

Marylandtest cohort,highexpression ofmiR-21 in tumors(HR,2.5; 95% CI,1.2-5.2; P=.01) and TNM staging (HR, 3.5;95%CI,1.6-7.9;P=.002)wasassociatedwith prognosis while age, sex, race, andtumorlocationwerenot.Inthefinalmul-tivariate model, which includedmiR-21expression and TNM staging, highmiR-21 expression in tumors was associated

Figure 1. High miR-21 Expression in Tumors and Poor Survival in Patients With TypicalAdenocarcinoma Histology

1.0

0.6

0.8

0.4

0.2

0

No. at RiskmiR-21 expression

miR-21 expression

High

Low

24

Log-rankP=.45

45

20 40 60 80

18 13 8 4

32 20 14 6

Months

Nontumor expression

ProbabilityofSurvival

Maryland test cohortA

1.0

0.6

0.8

0.4

0.2

0

26

Log-rankP=.005

46

20 40 60 80

16 10 7 3

37 26 17 9

Months

Tumor expression


1.0

0.6

0.8

0.4

0.2

0

No. at Risk

HighLow

35

Log-rankP=.47

68

20 40 60 80

25 18 16 1459 42 36 19

Months


Hong Kong validation cohortB

1.0

0.6

0.8

0.4

0.2

0

34

Log-rankP=.001

69

20 40 60 80

24 11 10 660 47 42 27

Months


High

miR-21 expression

Low

Thisanalysis excludes patients witheither mucinous adenocarcinoma or adenosquamous carcinoma histology. A,MicroRNA microarrays were used to measure microRNA expression levels of tumors and nontumorous tissues.Tissues with undetectable expression of miR-21 based on microarray data were excluded. High miR-21 expres-sion was classified according to the highest tertile (2.6-fold to 7.9-fold higher than nontumor). B, The associationof high miR-21 expression in tumors with poor prognosis is validated in an independent cohort. Expression levelsof miR-21 were measured by quantitative reverse transcription polymerasechain reaction. Highexpressionis basedon thehighest tertile (3.3-foldto 8.7-foldhigher thannontumor). Log-rankP values are fromKaplan-Meieranalysis.




7/29/2019 jpc70014_425_436

7/12

witha poorsurvival prognosis indepen-dent of tumor staging (HR, 2.7; 95% CI,1.3-5.5; P=.008; Table 4).

In the Hong Kong validation cohort,the dichotomized values formiR-21 ex-pressionintumorswerenotsignificantly

associatedwithage,sex,tumorhistology,ortumorlocation(Fisherexacttest).HighmiR-21 expression in tumors (HR, 2.4;95%CI, 1.4-3.9;P=.002)andTNMstag-ing (HR, 4.7; 95% CI, 2.4-9.5;P .001)weresignificantlyassociatedwithsurvivalin univariate models while age, sex, andtumorlocationwerenot(Table4).InthefinalmultivariateCoxregression model,including miR-21 expression and TNMstaging,highmiR-21expressionintumorswas associated with poor survival prog-nosis (HR,2.4;95%CI, 1.4-4.1;P=.002)independentof other clinicalcovariates,

consistent with findings intheMarylandtest cohort.

miR-21 Expression

in Colon Adenomas

Adenomasrepresentaprecursorstageforcolonadenocarcinomas.34WetestedmiR-20a, miR-21, miR-106a, miR-181b, andmiR-203expressionlevelsbyquantitativeRT-PCR in 18 pairs of adenoma and ad-jacent nonadenoma tissue that were ac-quiredfromtheCooperativeHumanTis-sue Network. Using only 18 pairs of

tissues reduced the power to detect dif-ferencesin these analysesand mayresultin false-negatives. However,miR-21wassignificantly enriched at 1.6-foldhigher(P= .006, Wilcoxon matched-pairstest;Table3).Adenomatissueexpressedhigherlevels of in 15 of 18 matched pairs.

Tumor Stages

and miR-21 Expression

IfmiR-21 expressioniscausaltothepro-gression of colon cancer, expression ofmiR-21may beassociatedwith more ad-

vancedstagesofthedisease.Patientswerestratified based on the diagnosis of ad-enoma and TNM staging, in which ad-enomawasconsideredtheleastadvancedand TNM stage IV was most advanced.Adenomasexpressedlowerlevels ofmiR-21thantumorsfromthevalidationcohort(P.001,Mann-Whitneytest).Moread-vancedtumorsexpressedhigherlevelsof

miR-21usingeithermicroarraydatafromtheMarylandtestcohort(P=.04,testfortrend) or the quantitative RT-PCR datafrom the Hong Kong validation cohort(test for trend,P.001; FIGURE 2).

miR-21 Expression

in Colonic Epithelial Cells

Although we found that high expres-sion ofmiR-21 in tumors was associ-ated with a worsesurvival outcome,these

experiments did not identify the cellswithin a tumor that expressed miR-21.To identify these cells, we used in situhybridization to visualizemiR-21 expres-sionintumor andadjacent nontumor tis-sue (FIGURE 3). The miR-21 is ex-

pressed at higher levels in colonicepithelial cells in human tumor tis-sue, consistent for a role for miR-21overexpression within tumor cells dur-ing colon carcinogenesis.

Table 4. Univariate and Multivariate Cox Regression Analysis of miR-21 Expression Levelsand Overall Cancer Survival in Subjects With Colon Adenocarcinoma a

Characteristic

Univariate Analysis Multivariate Analys isb

HR (95% CI) P Value HR (95% CI) P Value

Maryland Test CohortmiR-21 expression (n=71)c

Low 1.0 [Reference].01

1.0 [Reference].008

High 2.5 (1.2-5.2) 2.7 (1.3-5.5)

TNM stageI-II 1.0 [Reference]

.0021.0 [Reference]

.002III-IV 3.5 (1.6-7.9) 3.7 (1.6-8.3)

Age at enrollment, y50 1.0 [Reference]

.5250 0.7 (0.2-2.3)

SexWomen 1.0 [Reference]

.57Men 1.4 (0.5-3.9)

RaceWhite 1.0 [Reference]

.97Black 1.0 (0.5-2.1)

Tumor location(proximal/distal)

Distal 1.0 [Reference].26

Proximal 0.6 (0.3-1.4)

Hong Kong Validation CohortmiR-21 expression (n=103)c

Low 1.0 [Reference].002

1.0 [Reference].002

High 2.4 (1.4-3.9) 2.4 (1.4-4.1)

TNM stageI-II 1.0 [Reference]

.0011.0 [Reference]

.001III-IV 4.7 (2.4-9.5) 4.7 (2.4-9.5)


.1450 1.5 (0.9-2.6)

SexWomen 1.0 [Reference]

.29Men 1.4 (0.8-2.3)

Tumor locationDistal 1.0 [Reference]

.27Proximal 0.7 (0.3-1.4)

Abbreviations: CI, confidence interval; HR, hazard ratio.a Cases with mucinous adenocarcinoma, adenosquamous carcinoma, or signet ring cell carcinomas were excluded from

this analysis.bMultivariate analysisused stepwiseadditionand removalof clinical covariates found to be associatedwith survivalin uni-

variate models (P .10) and final models include only those covariates that were significantly associated with survival(Wald statistic, P .05). For both final models, only miR-21 expression and TNM staging were included.

c High expressionin tumors for allmiRNAs wasdefined based on the highest tertile. MicroRNA expressionwas measuredwith miRNA microarrays for the Maryland cohort and with quantitative reverse transcription polymerase chain reactionwith the Hong Kong cohort.




7/29/2019 jpc70014_425_436

8/12

miR-21 Expression Levels

and Therapeutic Outcome

We analyzed associations withmiR-21expression and therapeutic outcomesin stage II andIII cancer patients treatedwith adjuvant chemotherapy. Informa-tion on the administration of adjuvantchemotherapy was available for 47 of65 stage II or III patients in the Mary-land test cohort and for all patients inthe Hong Kong validation cohort. In

both cohorts chemotherapy regimenswere primarily fluorouracil based (informs of either intravenous 5-fluoro-uracil or oral drugs including tegafurwith uracil [UFT]) with or without ge-nerics. Only patients with typical ad-enocarcinoma histology were used forthis analysis, leaving 20 of 42 stage IIor III patients who had received che-motherapy in the Maryland cohort. Forthose whoreceived chemotherapy, highmiR-21 expression in tumors pre-dicted worse overall survival (HR, 4.3;

95% CI, 1.1-16.4;P

= .03) giving pre-liminary support that high miR-21is associated with poor therapeuticoutcome.

For the Hong Kong validation co-hort, 77 individuals with stage II or IIIcancer with typical adenocarcinomahistology were used for this analysis.Among the36 patientswho received ad-

juvant chemotherapy, highmiR-21 ex-pression in tumors was associated witha poor response to therapy (HR, 3.5;95% CI, 1.1-11.6; P= .04), consistentwith observations in the Maryland co-hort. Additionally, among the 25 pa-tients with stage III cancer who re-ceived adjuvant chemotherapy, highmiR-21 expression was associated withpoor survival (HR, 3.9; 95% CI, 1.2-12.9;P=.03). Analyses using cancer re-

lapse as an end point resulted in simi-lar associations with high miR-21expression in tumorspredicting a morerapid disease recurrence in patients withTNM stage III cancer who received ad-juvant chemotherapy (HR, 3.5;95% CI,1.0-11.5; P=.04).

Both cohorts showed similar asso-ciations of high miR-21 expression intumors with therapeutic outcomes,although neither cohort was largeenough to perform a fully stratifiedanalysis. Although these are 2 differ-

ent populations, the similarity of theassociations with miR-21 expression,prognosis, and therapeutic outcomeallowed for pooled analysis usingboth cohorts. Kaplan-Meier analysisof the pooled cohorts demonstratedthat high miR-21 expression wasassociated with a poor prognosis ineither stage II (P=.02) or stage III

(P=.004) patients (FIGURE 4) furtherindicating its potential as a prognos-tic biomarker. Receipt of adjuvantchemotherapy was beneficial forpatients with either stage II or stageIII cancer, although this was onlysignificant for stage III patients. Forindividuals who received adjuvanttherapy, high miR-21 expression wasassociated with a poor therapeuticoutcome in patients with stage II or

III cancer (P=.003, Kaplan-Meier logrank) or in patients with stage IIIcancer alone (P=.007, Kaplan-Meierlog rank; Figure 4). Multivariate Coxregression demonstrated that highmiR-21 expression predicted poorprognosis (HR, 3.0; 95% CI, 1.7-5.4;P .001) and treatment with adju-vant chemotherapy was associatedwith better survival (HR, 0.4; 95%CI, 0.2-0.8; P=.004) independent ofother clinical covariates (TABLE 5).

COMMENT

We per formed the lar gest study todate analyzing microRNA profiles incolon cancer tissues and the first, toour knowledge, using 2 independentcohorts. Thirty-seven microRNAswere differentially expressed intumor tissues by microRNA microar-ray analysis in the Maryland test

Figure 2. miR-21 Expressed at Higher Levels in Colon Adenocarcinomas With Increasing Expression in More Advanced Tumors

4

2

Pfor trend=.04 Pfor trend

7/29/2019 jpc70014_425_436

9/12

cohort. Expression patterns of all 5tested microRNAs were validated inthe Hong Kong cohort. The discrimi-natory power of 5 microRNAs todifferentiate between tumor and non-tumorous tissue suggests that pre-

dictable and systematic changes ofmicroRNA expression patterns mayoccur during tumorigenesis and maybe representative of sporadic colonadenocarcinomas.miR-20a, miR-21, miR-106a, miR-

181b, and miR-203 were all found tobe expressed at higher levels in colontumors, although it is uncertainwhether these changes in microRNAexpression patterns are merely asso-ciated with colon cancer or causal tothe histological progression to can-cer. Our data are consistent with

published studies that provide evi-dence for changes in microRNAexpression promoting tumor forma-tion, especially for miR-20a and miR-21. miR-20a is part of the miR-17-92polycistronic microRNA cluster.35

O v e r e x p r e s s i o n o f t h i s c l u s t e renhances cell proliferation in vitro36

and accelerates tumor formation inanimal models.16 Enforced expres-sion of the miR-17-92 cluster causesincreased tumor size and tumor vas-cularization in mice by negatively

r e g u l a t i n g t h e a n t i - a n g io g e n i cthrombospondin 1 (Tsp1) protein.24

Experimental evidence also suggeststhat increased miR-21 expressionpromotes tumor development.miR-21 is expressed at high levels in

most solid tumors.20,37 Overexpres-sion ofmiR-21 acts as an antiapoptoticfactor in human glioblastoma cells.13

Inhibition ofmiR-21 inhibits cell growthin vitro and inhibits tumor growth inxenograft mouse models through anindirect down-regulation of the anti-

apoptotic factor, B-cell lymphoma 2(Bcl-2).38 Studies in human cell lineshave shown miR-21 can also target thetumor suppressor genes, phosphataseand tensin homolog (PTEN)39 andtropomyosin 1 (TPM1).40 These data,taken together, support a causal role foraltered microRNA expression duringtumorigenesis.

Adenomas represent a precursorstage of adenocarcinoma. Adenomasexpress high levels of miR-21 . Ifincreased miR-21 expression pro-motes colon tumor progression,increased expression in adenomas

may be an early cellular event in theprogression to cancer. InhibitingmiR-21 activity may help preventtumor promotion in populations at

high risk of colon cancer, such asindividuals with familial adenoma-tous polyposis.41

In this study, we also demonstrateda n a s s o c ia t i o n w i t h m i c r o RN Aexpression patterns with colon cancer

prognosis and therapeutic outcome. Arobust association of high miR-21expression in tumors with poor sur-vival was observed in the Maryland

Figure 3. In Situ Hybridization of miR-21 in Colon Tumors

miR-21 probe

Magnification x40

Magnification x400

Scramble control probe

A B

C D

Tumor

Nontumor

Tumor

Nontumor

E F

500 m 500 m

500 m500 m

Tumor

Nontumor

Tumor

Nontumor

In situ hybridization formiR-21 was optimized to distinguish high (brown) and low expression of miR-21. The3 DIG-labeled probe was hybridized and detected with a polyclonal anti-DIG antibody (DakoCytomation)using amplification with the GenPoint Tyramide Signal Amplification System (DakoCytomation) using VectorNovaRed (Vector Laboratories) as the substrate. The slide was counterstained with Mayers hematoxylin. A,Colonic epithelial cells in human tumor express higher levels of miR-21 compared with adjacent nontumoroustissue. C, Colonic epithelial cells in tumor tissue express significant amounts of miR-21, at high magnification.E, Nontumor tissue shows nosignificant expression of miR-21 at thesame magnification. B, D, F, Thescramblecontrol probe shows no significant staining at low or high magnification in serial sections of tumor and non-tumor tissue, as expected.




7/29/2019 jpc70014_425_436

10/12

Figure 4. Combined Analysis of Maryland Test Cohort and Hong Kong Validation Cohort Examining Associations Between miR-21 Expressionin Tumors and Receipt of Adjuvant Chemotherapy With Prognosis

With

Adjuvant chemotherapy

Without

High

miR-21 expression

Low High

With chemotherapy by

miR-21 expression

Low

High

Without chemotherapy by

miR-21 expression

Low

1.0

0.6

0.8

0.4

0.2

0

No. at Risk

WithWithout

0

55

P=.17

64

20 40 60 80

48 37 30 1753 34 28 21

Months

Adjuvant chemotherapy

Probability

Stage II and III CasesA

1.0

0.6

0.8

0.4

0.2

0

No. at RiskChemotherapy miR-21 expression

HighLow

0

38

P

7/29/2019 jpc70014_425_436

11/12

test cohort and the Hong Kong valida-tion cohort, separately. In eachcohort, these associations were inde-pendent of other clinical covariatesindicating that miR-21 expression maybe a useful prognostic indicator, in

addition to TNM staging and otherclinical parameters, to help identifypatients at a higher risk of terminalcancer. These observations were madein 2 independent cohorts with differ-ent racial and geographical composi-tions. Therefore, our observationsmay be broadly applicable to otherpopulations.

High miR-21 expression in tumorswas associated with a poor therapeu-tic outcome in both cohorts. Thisassociation may help predict the ben-efits of therapy in individuals whose

miR-21 expression status is knownand identify patients who are candi-dates for more aggressive initialtherapies. But, if high miR-21 expres-sion is causal to the poor therapeuticoutcome, antagomirs30, 42 or otherantisense therapeutics that targetmiR-21 may prove to have therapeu-tic benefits in patients with tumorswith high expression ofmiR-21.

Additional studies are required todemonstrate a causal link with miR-21and the progression of colon cancer to

determine the potential ofmiR-21 aseither a biomarker or therapeutic tar-get. Although a causal role is stilluncertain, many of Hills criteria forcausation43 have already been met.There is a temporal relationship suchthat a high expression ofmiR-21 intumors precedes the progression,therapeutic response, and subsequentdeath due to cancer. The strength andconsistency of these associations hasbeen found in 2 independent cohorts.There is a dose-response relationship

such that more advanced tumorsexpress higher levels ofmiR-21 inboth cohorts. Published research hasdemonstrated biological plausibility ofmiR-21 causing the progression oftumors using in vitro and in vivomodels. All of these are consistentwith a role for miR-21 in colon carci-nogenesis.

In conclusion, we found systematicdifferences in microRNA expression pat-terns between colon tumors and pairednontumorous tissue. Tumors with high

expression ofmiR-21

wasassociated withpoor survival outcome and poor re-sponse to adjuvant chemotherapy in 2independent cohorts, independent ofstaging andother clinical covariates sug-gesting that miR-21 may be a useful di-agnostic biomarker for colon adenocar-cinomas andsurvival prognosisincludingresponse to therapy.

Author Contributions: DrsSchetter andHarrishad fullaccessto all ofthe datain the study and takerespon-sibility for the integrity of the data and the accuracyof the data analysis.Studyconcept and design:Schetter, Leung, Yanaihara,Yuen, Harris.

Acquisition of data: Schetter, Leung, Sohn, Bowman,Yuen, Chan, Kwong, Au, Liu, Calin, Croce, Harris.Analysis and interpretation of data: Schetter, Leung,Sohn, Zanetti, Liu, Croce, Harris.Drafting of the manuscript: Schetter, Sohn, Croce,Harris.Critical revision of the manuscript for importantintellectual content:Schetter,Leung, Zanetti, Bowman,Yanaihara, Yuen, Chan, Kwong, Au, Liu, Calin,Harris.Statistical analysis: Schetter, Zanetti, Kwong.Obtained funding: Leung, Yuen, Chan.

Administrative, technical,or material support:Leung,Sohn, Bowman, Yuen, Chan, Kwong, Au, Liu.Study supervision: Harris.Financial Disclosures: None reported.Funding/Support: Thisresearchwas supported by theIntramural Research Program of the National Cancer

Institute, Center for Cancer Research, National Insti-tutesof Health. DrsLeung,Chan,andYuenweresup-ported by the Research Grants Council of the HongKong Special Administrative Region (HKU 7498/06M). Drs Schetterand Zanetti weresupported by theCancer Prevention FellowshipProgram, Office of Pre-ventive Oncology, National Cancer Institute.Role of the Sponsors: The funding agencies had norole in the design and conduct of the study; collec-tion,management, analysis, and interpretation of thedata; andpreparation, review, or approvalof themanu-script.Additional Contributions: We thank Xin Wang,PhD, Stefan Ambs, PhD, Kensuke Kumamoto, MD,PhD, Ewy Mathe, PhD, and Rosemary Braun, PhD,for helpful discussion; Megan Hames, BS, and JinJen, MD, PhD, for advice on in situ hybridizationtechniques; Judith Welsh, BS, for technical assis-tance; and Dorothea Dudek-Creaven, MSC, for edi-

torial assistance, all of the Center for CancerResearch, National Cancer Institute. We thankKathleen Meyer, BA, Center for Information Tech-nology, National Institutes of Health, for uploadingmicroarray data files to Gene Omnibus Profiles. Wealso than Raymond Jones, PhD, Audrey Salabes, BS,and John Cottrell, MS, of the University of Mary-land Medical Center, the clinicians of the depart-ments of surgery and pathology in Queen MaryHospital, at University of Maryland Hospital, and atthe Baltimore Veterans Administration MedicalCenter for contributions to this study.

Table 5. Univariate and Multivariate Cox Regression Analysis of miR-21 Expression, Receiptof Adjuvant Chemotherapy, and Cancer Survival in Patients With Stage II or III WithAdenocarcinoma in both Maryland and Hong Kong Cohortsa

Characteristics

Univariate Analysis Multivariate Analys isb

HR (95 CI) P Value HR (95% CI) P Value

miR-21 expression (n=119)c

Low 1.0 [Reference] .001 1.0 [Reference] .001High 2.6 (1.5-4.5) 3.0 (1.7-5.4)

Adjuvant chemotherapyDid not receive 1.0 [Reference]

.211.0 [Reference]

.004Received 0.7 (0.4-1.2) 0.4 (0.2-0.8)

TNM stageII 1.0 [Reference]

.0011.0 [Reference]

.001III 3.2 (1.7-6.1) 5.2 (2.6-11)

Tumor locationDistal 1.0 [Reference]

.021.0 [Reference]

.007Proximal 0.4 (0.2-0.8) 0.3 (0.1-0.7)


.3250 1.4 (0.7-2.5)

Sex

Women 1.0 [Reference] .44Men 1.3 (0.7-2.2)

Abbreviations: CI, confidence interval; HR, hazard ratio.a Patients with TNM stage II or III cancer with typical adenocarcinoma histology were included in this analysis.bMultivariate analysisused stepwiseadditionand removalof clinical covariates found to be associatedwith survivalin uni-

variate models (P .10) and final models include only those covariates that were significantly associated with survival(Wald statistic, P .05). miR-21 expression, receipt of adjuvant therapy, TNM staging, and tumor location were in-cluded in final multivariate model.

c High expression in tumors for all miRNAs was defined based on the highest tertile. Race was not associated with poorprognosis. MicroRNA expression was measured with miRNA microarrays for the Maryland cohort and with quantitativereverse transcription polymerase chain reaction with the Hong Kong cohort.




7/29/2019 jpc70014_425_436

12/12

REFERENCES

1. Aaltonen LA, Hamilton SR; World Health Organi-zation Classification of Tumors. Pathology and Ge-netics of Tumours of the Digestive System. Lyon,France: International Agency for Research on CancerPress; 2000.2. Jemal A, Siegel R, Ward E, Murray T, Xu J, ThunMJ. Cancer statistics, 2007. CA Cancer J Clin. 2007;57(1):43-66.3. Goldman E, Fisher JL. Discrepanciesin cancer mor-tality estimates. Arch Med Res. 2006;37(4):548-551.4. Rodriguez-Bigas MA, Hoff P, Crane CH. Carci-noma of thecolon andrectum. In:Kufe DW,Bast RC,HaitWN, etal, eds. Holland-Frei Cancer Medicine 7.7thed. Hamilton, Ontario: BC Decker Inc; 2006:1369-1391.5. Calin GA, Croce CM. MicroRNA signatures in hu-man cancers. Nat Rev Cancer. 2006;6(11):857-866.6. Calin GA, Ferracin M, Cimmino A, et al. A Mi-croRNAsignatureassociatedwith prognosis and pro-gression in chronic lymphocytic leukemia. NEnglJMed.2005;353(17):1793-1801.7. Yanaihara N, Caplen N, Bowman E, et al. UniquemicroRNA molecular profiles in lung cancer diagno-sis and prognosis. Cancer Cell. 2006;9(3):189-198.8. Waldman SA, Terzic A. TranslatingMicroRNA dis-covery intoclinical biomarkers in cancer.JAMA. 2007;297(17):1923-1925.9. Bartel DP. MicroRNAs: genomics, biogenesis,mechanism, and function. Cell. 2004;116(2):281-297.10. LeeRC, Feinbaum RL,Ambros V. The C. elegansheterochronic gene lin-4encodessmall RNAswith an-tisense complementarity to lin-14. Cell. 1993;75(5):843-854.11. Wightman B, Ha I, Ruvkun G. Posttranscrip-tional regulation of the heterochronic gene lin-14 bylin-4 mediates temporal pattern formation in C.elegans. Cell. 1993;75(5):855-862.12. BrenneckeJ, Hipfner DR,StarkA, Russell RB,Co-hen SM. Bantam encodes a developmentally regu-latedmicroRNAthat controls cellproliferationand regu-lates the proapoptotic gene hid in Drosophila. Cell.2003;113(1):25-36.13. Chan JA, Krichevsky AM, Kosik KS. MicroRNA-21 is an antiapoptotic factor in human glioblastoma

cells. Cancer Res. 2005;65(14):6029-6033.14. Xu P, Vernooy SY, Guo M, Hay BA. The Dro-sophila microRNA Mir-14 suppresses cell death and

is requiredfor normal fat metabolism. CurrBiol. 2003;13(9):790-795.15. Chen CZ, Li L, Lodish HF, Bartel DP. MicroRNAsmodulate hematopoietic lineage differentiation.Science. 2004;303(5654):83-86.16. He L, Thomson JM, Hemann MT, et al. A mi-croRNA polycistron as a potential human oncogene.Nature. 2005;435(7043):828-833.17. Esquela-Kerscher A, Slack FJ. Oncomirs

microRNAswith a role in cancer. NatRev Cancer. 2006;6(4):259-269.18. Calin GA, Dumitru CD, Shimizu M, et al. Fre-quent deletions and down-regulation of micro-RNAgenes miR15 andmiR16at 13q14in chronic lympho-cytic leukemia. Proc Natl Acad Sci U S A. 2002;99(24):15524-15529.19. Lu J, Getz G, Miska EA, et al. MicroRNA expres-sion profiles classify human cancers. Nature. 2005;435(7043):834-838.20. Volinia S, Calin GA, Liu CG, et al. A microRNAexpression signature of human solid tumors definescancer gene targets. Proc Natl Acad Sci U S A. 2006;103(7):2257-2261.21. Cummins JM, He Y, Leary RJ, et al. The colorec-tal microRNAome. Proc Natl Acad Sci U S A. 2006;103(10):3687-3692.22. Bandres E, C ubedo E, Agirre X, et al. Identifica-tion by Real-time PCR of 13 mature microRNAs dif-ferentially expressed in colorectal cancer and non-tumoral tissues. Mol Cancer. 2006;5:29.23. Michael MZ,O ConnorSM, vanHolstPellekaanNG, Young GP, James RJ. Reduced accumulation ofspecific microRNAs in colorectal neoplasia. Mol Can-cer Res. 2003;1(12):882-891.24. Dews M, Homayouni A, Yu D, et al. Augmenta-tion of tumor angiogenesis by a Myc-activated mi-croRNA cluster. Nat Genet. 2006;38(9):1060-1065.25. Wang CL, Wang BB, Bartha G, et al. Activationof an oncogenic microRNA cistron by provirusintegration. Proc Natl Acad Sci U S A. 2006;103(49):18680-18684.26. Georgantas RW III, Hildreth R, Morisot S, et al.CD34 hematopoietic stem-progenitor cell mi-croRNA expression and function: a circuit diagram ofdifferentiation control.ProcNatlAcad Sci US A. 2007;104(8):2750-2755.27. Bloomston M, Frankel WL, Petrocca F, et al. Mi-croRNA expression patterns to differentiate pancre-

aticadenocarcinomafrom normal pancreas andchronicpancreatitis. JAMA. 2007;297(17):1901-1908.28. Chen Y, Stallings RL. Differential patterns of mi-

croRNA expression in neuroblastoma are correlatedwith prognosis, differentiation, and apoptosis. Can-cer Res. 2007;67(3):976-983.29. WurdingerT, Costa FF.Molecular therapy in themicroRNA era. Pharmacogenomics J. 2006;7(5):297-304.30. Krutzfeldt J, Rajewsky N, Braich R, et al. Silenc-ing of microRNAs in vivo with antagomirs. Nature.2005;438(7068):685-689.

31. Liu CG, Calin GA, Meloon B, et al. An oligo-nucleotide microchip for genome-wide microRNApro-filing in human andmousetissues. Proc Natl Acad SciU S A. 2004;101(26):9740-9744.32. Cuzick J. Wilcoxon-type test for trend.Stat Med.1985;4(1):87-90.33. Kang H, OConnell JB, Maggard MA, Sack J, KoCY. A 10-year outcomes evaluation of mucinous andsignet-ring cell carcinoma of the colon and rectum.Dis Colon Rectum. 2005;48(6):1161-1168.34. FearonER, VogelsteinB. A geneticmodel forco-lorectal tumorigenesis. Cell. 1990;61(5):759-767.35. TanzerA, Stadler PF.Molecularevolution ofa mi-croRNA cluster. J Mol Biol. 2004;339(2):327-335.36. HayashitaY, OsadaH, Tatematsu Y, et al.A poly-cistronic microRNA cluster, miR-17-92, is overex-pressed in human lung cancers and enhances cellproliferation. Cancer Res. 2005;65(21):9628-9632.37. Iorio MV, Ferracin M, Liu CG, et al. MicroRNAgeneexpressionderegulation in human breast cancer.Cancer Res. 2005;65(16):7065-7070.38. Si ML, Zhu S, Wu H, Lu Z, Wu F, Mo YY. miR-21-mediated tumor growth. Oncogene. 2007;26(19):2799-2803.39. Meng F, Henson R, Lang M, et al. Involvementof human micro-RNA in growthand response to che-motherapy in human cholangiocarcinoma cell lines.Gastroenterology. 2006;130(7):2113-2129.40. Zhu S, Si ML, Wu H, Mo YY. MicroRNA-21 tar-getsthe tumorsuppressor genetropomyosin 1 (TPM1).J Biol Chem. 2007;282(19):14328-14336.41. Brosens LA, vanHattem WA, JansenM, de LengWW, Giardiello FM, Offerhaus GJ. Gastrointestinal pol-yposis syndromes. Curr Mol Med. 2007;7(1):29-46.42. Mattes J, Yang M, Foster PS. Regulation of mi-croRNA by antagomirs: a newclassof pharmacologi-cal antagonists for the specific regulation of genefunction? Am J Respir Cell Mol Biol. 2007;36(1):

8-12.43. Hill AB.StatisticalEvidence and Inference.9thed.New York, NY: Oxford University Press; 1971.



jpc70014_425_436

Documents