Hereditary nonpolyposis colorectal cancer in endometrial cancer patients

Sang Nam Yoon1,2, Ja-Lok Ku

3, Young-Kyoung Shin

1, Kyung-Hee Kim

1, Jin-Sung Choi

1, Eun-Ja Jang

1,

Hyoung-Chul Park1,2, Duck-Woo Kim1,2, Min A Kim4, Woo Ho Kim4, Taek Sang Lee5, Jae Weon Kim5,Noh-Hyun Park5, Yong-Sang Song5, Soon-Beom Kang5, Hyo-Pyo Lee5, Seung-Yong Jeong6 and Jae-Gahb Park1,2,3*

1Korean Hereditary Tumor Registry, Cancer Research Institute, Seoul National University, Seoul, Korea2Department of Surgery, Seoul National University College of Medicine, Seoul, Korea3Laboratory of Cell Biology, Cancer Research Institute, Seoul National University, Seoul, Korea4Department of Pathology, Seoul National University College of Medicine, Seoul, Korea5Department of Obstetrics and Gynecology, Seoul National University College of Medicine, Seoul, Korea6Center for Colorectal Cancer, Research Institute and Hospital, National Cancer Center, Goyang, Korea

Endometrial cancer is the second most common cancer in heredi-tary nonpolyposis colorectal cancer (HNPCC). It has often beenoverlooked to explore the possibility of HNPCC in endometrialcancer patients. Our study was to investigate how many HNPCCpatients existed among endometrial cancer patients. Amongpatients who underwent hysterectomy for endometrial cancer atSeoul National University Hospital from 1996 to 2004, 113 patientswere included, whose family history and clinical data could beobtained and tumor specimens were available for microsatelliteinstability (MSI) testing and immunohistochemical (IHC) stainingof MLH1, MSH2 and MSH6 proteins. There were 4 (3.5%) clini-cal HNPCC patients fulfilling the Amsterdam criteria II, and 2 (2/4, 50%) of them carried MSH2 germline mutations. There werealso 8 (7.1%) suspected HNPCC (s-HNPCC) patients fulfilling therevised criteria for s-HNPCC, and one (1/8, 12.5%) of themrevealed MLH1 germline mutation. In 101 patients, who were notclinical HNPCC or s-HNPCC, 11 patients showed both MSI-highand loss of expression of MLH1, MSH2 or MSH6 proteins, and 2(2/11, 18.2%) of them showed MSH6 germline mutations. In 113patients with endometrial cancer, we could find 5 (4.4%) HNPCCpatients with MMR germline mutation and 2 (1.8%) clinicalHNPCC patients without identified MMR gene mutation. Familyhistory was critical in detecting 3 HNPCC patients with MMRgermline mutation, and MSI testing with IHC staining for MLH1,MSH2 and MSH6 proteins was needed in the diagnosis of 2HNPCC patients who were not clinical HNPCC or s-HNPCC,especially for MSH6 germline mutation.' 2007 Wiley-Liss, Inc.

Key words: endometrial cancer; hereditary nonpolyposis colorectalcancer (HNPCC); microsatellite instability (MSI); mismatch repairgene (MMR); MLH1; MSH2; MSH6

The incidence of endometrial cancer has been increasing signif-icantly among general population in Korea over the past decades.1

According to data on cancer incidence between 1999 and 2001from the Korea Central Cancer Registry, the crude incidence rateper year is 3.11 cases per 100,000 Korean females.2 In a total of43,627 new cases of female cancer registered by the Korea CentralCancer Registry in 2002, 825 cases were endometrial cancers,accounting for 1.9% of all malignancies in female.3

Hereditary nonpolyposis colorectal cancer (HNPCC) is an auto-somal dominant inherited disease characterized by high incidenceof colorectal cancer (CRC), endometrial cancer and/or a variety ofother cancers.4 Defective DNA mismatch repair (MMR) process iscaused by germline mutations in the MMR genes, leading to thedevelopment of HNPCC. Germline mutations have been identifiedin five of these genes, MSH2, MLH1, PMS1, PMS2 and MSH6,and mutations in MSH2, MLH1 and MSH6 appear to account forthe MMR defects seen in the majority of HNPCC families.5

Endometrial cancer is the second most common cancer found inHNPCC families.6 The lifetime risk of endometrial cancer forwomen with HNPCC is reported 40–60%, which is the same orgreater than that of colorectal cancer.7 However, it has often beenoverlooked to explore the possibility of HNPCC in patients diag-nosed as endometrial cancer.

Our study was to investigate how many HNPCC patientsexisted among the patients who underwent hysterectomy for endo-metrial cancer.

Material and methods

Patients

The study population was drawn from a consecutive series ofendometrial cancer patients who underwent hysterectomies atSeoul National University Hospital between January 1996 and De-cember 2004. We enrolled 113 patients who answered the ques-tions about their personal and family cancer histories, and whoseparaffin-embedded tumor and normal tissues were available formicrosatellite instability (MSI) analysis and immunohistochemical(IHC) staining. Patients were classified into clinical HNPCC whenthey fulfilled the Amsterdam criteria II,4 or into suspected HNPCC(s-HNPCC) when they fulfilled the revised criteria for s-HNPCCaccording to their personal and family cancer histories.8

Clinical data on pathologic result and patient demographicswere retrospectively reviewed. Approval for taking personal andfamily cancer histories, as well as performing MSI analysis, IHCstaining and mutational analysis was granted by the InstitutionalReview Board of the Seoul National University Hospital and theSeoul National University College of Medicine.

MSI analysis

For PCR-based MSI analysis, we utilized 2 mononucleotiderepeats (BAT25 and BAT26) and 3 dinucleotide repeats (D2S123,D5S346 and D17S250); the MSI criteria and primers were as pre-viously defined.9 The specific primer sequences of PCR amplifica-tion were obtained from http://www.gdb.org. Serial sections ofparaffin-embedded matched normal and neoplastic primary tissuesfrom each endometrial cancer patient were stained with H&E.Representative normal and tumor regions were identified by mi-croscopic examination and separately microdissected. The dis-sected tissues were incubated overnight at 56�C in a lysis buffercontaining proteinase K, the proteinase K was inactivated by heat-ing for 15 min at 70�C, and the samples were used as templateDNA for PCR analysis. The desired fragments were amplified inthe presence of [a-P32] dCTP, using a programmable thermalcycler (PCR System 9700, Applied Biosystems; Foster City, CA).The PCR conditions consisted of a denaturation step at 94�C for

The first three authors contributed equally to this work.Grant sponsors: Research Grant for National Cancer Center; BK21 Pro-

ject for Medicine, Dentistry, and Pharmacy.*Correspondence to: Korean Hereditary Tumor Registry, Cancer

Research Institute, Seoul National University, 28 Yongon-dong, Chongno-gu, 110-744, Seoul, Korea. Fax:182-2-742-4727.E-mail: jgpark@plaza.snu.ac.krReceived 23 January 2007; Accepted after revision 19 April 2007DOI 10.1002/ijc.22986Published online 31 October 2007 in Wiley InterScience (www.interscience.

wiley.com).

Int. J. Cancer: 122, 1077–1081 (2008)' 2007 Wiley-Liss, Inc.

Publication of the International Union Against Cancer

10 min followed by 35 cycles of 95�C for 30 sec, 50–55�C for 1min and 72�C for 1 min, and a final elongation at 72�C for 7 min.The PCR products were denatured and separated on 6 M urea/7%polyacrylamide gels run at 60 W. After electrophoresis, each gelwas transferred to Whatmann 3 M paper, dried, and subjected toautoradiography.

Tumors were classified as MSI-H (microsatellite instability-high) when 2 or more of the 5 markers showed instability, asMSI-L (microsatellite instability-low) when only a marker showedinstability, and as MSS (microsatellite stable) when none of themarkers showed instability.9

Promoter hypermethylation in the MLH1 gene

Promoter hypermethylation of the MLH1 gene was determinedby the methylation-specific PCR (MS-PCR). The primers for MS-PCR could amplify the methylated and unmethylated DNA in thepromoter region of MLH1.10 Sodium bisulfite modification wasdone with a CpGenome DNA modification kit (Intergen, Oxford,UK). Amplified DNA fragments were fractionated in 2% agarosegel that was stained with ethidium bromide and visualized underUV. PCR reactions that demonstrated methylation were repeatedfor confirmation.

Tissue array analysis

Core tissue biopsies (2 mm in diameter) were taken from indi-vidual paraffin-embedded endometrial cancers (donor blocks) andarranged in a new recipient paraffin block (tissue array block)using a trephine apparatus (Superbiochips Laboratories, Seoul,Korea). Each tissue array block contained up to 50 cases, allowing3 array blocks to contain the total of 113 cases. Samples weredefined as adequate when the tumor occupied >10% of the corearea. Each block contained an internal control consisting of non-neoplastic endometrium. Sections (4 lm) were cut from each tis-sue array block, deparaffinized and dehydrated for use in IHCstaining.

IHC staining for MLH1, MSH2 and MSH6 protein expression

IHC staining for MLH1, MSH2 and MSH6 protein expressionwas performed as previously described,11 utilizing the followingantibodies: MLH1, clone G168-15 (1 mg/mL; Pharmingen, SanDiego, CA), a mouse MAb prepared with the full-length humanrecombinant MLH1 protein; MSH2, clone FE11 (210 mg/L; Zymed,San Francisco, CA), a mouse monoclonal antibody (MAb) generatedwith a carboxy-terminal fragment of the human MSH2 protein;MSH6, clone 44 (250 lg/mL; BD Transduction Laboratories, SanJose, CA), a mouse monoclonal antibody generated with an NH2-terminal fragment (codons 225–333) of the MSH6 protein. Normalmouse sera were used as negative controls. Infiltrating lymphocytesand normal epithelial cells adjacent to the tumor cells served as in-ternal positive controls. Tumors that demonstrated any evidence ofMLH1, MSH2 or MSH6 expression, even in small foci, were con-sidered positive for expression of that protein.

Mutational analysis of the MLH1, MSH2 and MSH6 genes

Mutational analysis was performed for patients who were iden-tified as clinical HNPCC or s-HNPCC based on personal and fam-ily cancer history, and also for patients whose tumors showed bothMSI-H and loss of MMR expression. Blood samples wereobtained from patients and peripheral blood lymphocytes wereisolated using Ficoll-Paque according to the manufacturer’sinstructions (Amersham Biosciences, Uppsala, Sweden). Totalgenomic DNA was extracted using the TRI reagent following themanufacturer’s instructions (Molecular Research Center, Cincin-nati, OH). When patient blood samples were unavailable, DNAsextracted from metastasis-free lymph nodes or microdissected nor-mal tissues were used for mutational analysis.

For identification of germline mutations in the MLH1, MSH2and MSH6 genes, polymerase chain reaction—direct sequencinganalysis was used to screen all gene coding regions, exon/intron

boundaries and core promoter sequences, as previously de-scribed.12 PCR products were bi-directionally sequenced using anABI Prism 3730 Genetic Analyzer (Applied Biosystems, FosterCity, CA). Potentially mutated sequences were cloned intopCR2.1 (Invitrogen, San Diego, CA) for separate analysis of eachallele, and to guard against sequencing errors.

Statistical analysis

Comparison of the variables was performed using Student’st-test, Pearson’s v2 test, or Fisher’s exact test, depending on thenature of the data. Significance was assigned at the p < 0.05 level.Results are reported as mean 6 standard deviation.

Results

In 113 study patients, there were 4 (3.5%) clinical HNPCCpatients fulfilling the Amsterdam criteria II. Two (1.8%) patients(SNU-H24 and SNU-H33) had already been registered as clinicalHNPCC fulfilling the Amsterdam criteria II in Korean HereditaryTumor Registry before hysterectomy.13 SNU-H33 carried MSH2germline mutation, but not only SNU-H24 but also her youngersister with both endometrial cancer and colon cancer had no muta-tion of MMR genes.13 Another 2 (1.8%) (SNU-HE1 and SNU-HE2) of 113 study patients were newly found to be clinicalHNPCC during our study. Mutational screening revealed germlinemutation of the MMR genes in 2 (SNU-H33 and SNU-HE1) ofthese 4 clinical HNPCC patients. Both had germline mutations inthe MSH2 gene, and showed MSI-H and loss of MMR expression(Table I, Fig. 1).

Eight (7.1%) patients met the revised criteria for s-HNPCCaccording to family history and mutational screening for mismatchrepair (MMR) genes revealed germline mutation of the MLH1gene in one (1/8, 12.5%) patient (SNU-HE1006). Another s-HNPCC patient had a single nucleotide polymorphism in whichcytosine was replaced by thymine (Leu to Phe) at codon 390 ofexon7 in the MSH2 gene.

In MSI testing, MSI-H phenotype was seen in 29 (25.7%) of113 patients, MSI-L in 21 (18.6%), and MSS in 63 (55.8%). Therewere no difference in mean age, distribution of tumor stage, histo-logic grade and cellular type of cancer (data not shown). In IHCstaining for MMR proteins, tumors from 26 (23.0%) patientsshowed loss of MMR protein expression. Loss of MLH1 expres-sion was detected in 23 patients, loss of both MSH2 and MSH6expression in 2, and loss of MSH6 expression in 1. No germlinemutation was found in cases where the endometrial tumor sampleexhibited MSI-L or MSS in MSI testing and retained MMR pro-tein expression in IHC staining. Representative examples of MSItesting and IHC staining results are shown in Figures 2 and 3,respectively.

Sixteen patients had MSI-H phenotype and concurrentlyshowed loss of MMR protein expression. In these 16 patients, 3patients were clinical HNPCC, 2 were s-HNPCC and the remain-ing 11 patients were not clinical HNPCC or s-HNPCC. Amongthese 11 patients, 2 (18.2%) carried germline mutations in MSH6gene; one patient showed loss of both MSH2 and MSH6 expres-sion, and the other exhibited loss of MSH6 expression. The other9 patients, all of which showed loss of MLH1 expression, had nodetectable germline mutations. Analysis for MLH1 promoter hy-permethylation revealed that 7 of these samples showed MLH1promoter hypermethylation, while the remaining 2 samples couldnot be analyzed due to the failure of the PCR amplification. Flowdiagram for overall study results is illustrated at Figure 1.

Discussion

Recent studies reported that at least 1.8–2.1% of newly diag-nosed endometrial cancer patients had germline mutations in theMMR genes.7,14 Five (4.4%) of 113 patients with endometrial can-cer had germline mutation of the MMR genes in our series. In our

1078 YOON ET AL.

study, we used the revised criteria for s-HNPCC8 for mutationscreening of MMR genes additional to the Amsterdam criteria II,and one mutation carrier could be found from 8 patients who metthe revised criteria for s-HNPCC.

MSI-H phenotype was seen in 29 (25.7%) of 113 patients in ourstudy and this is consistent with the reported rate of 25–27%.15,16

Two previous studies reported that MSI-H phenotype were foundin 18 and 45% of their endometrial cancer patients respectively,but these studies used different markers and criteria for MSIanalysis, complicating the direct comparison of our results.11,12

TABLE

I–CHARACTERISTICS

OF

5HNPCC

PATIENTS

WITH

MMR

GERMLIN

EMUTATIO

NIN

THIS

STUDY

PatientID

Ageatdiagnosis

(years)

Fam

ilyhistory

MSItesting

Loss

ofMMR

protein

inIH

Cstaining

Mutatedgenes

Mutationalprofiles

Consequences

SNU-H

331

43

ClinicalHNPCC2

MSI-H

MLH1,MSH2(w

eakpositive)

MSH

2c.2634_263411delGg

Splice

defect

SNU-H

E13

44

ClinicalHNPCC2

MSI-H

MSH2,MSH6

MSH

2c.2089T>

Cp.Cys697Arg

SNU-H

E1006

56

SuspectedHNPCC4

MSS

MLH1

MLH1

c.1757_1758insC

p.M

et587HisfsX6

SNU-H

E5001

48

None

MSI-H

MSH2,MSH6

MSH

6c.3823G>

A,

c.3821_3824dupAATG

p.Glu1274Lys,p.Cys1275X

SNU-H

E5002

72

None

MSI-H

MSH6

MSH

6c.3206G>

Ap.Gly1069Glu

MSI,microsatelliteinstability;IH

Cstaining,im

munohistochem

ical

staining;MMR,mismatch

repair;HNPCC,hereditarynonpolyposiscolorectal

cancer;MSI–H,microsatellite-high;MSS,

microsatellitestable.

1SNU-H

33had

alreadybeenregisteredas

clinicalHNPCCin

theKorean

HereditaryTumorRegistrybefore

hysterectomy.–

2ClinicalHNPCCmeansthatthepatientfulfilled

theAmsterdam

cri-

teriaII.–3SNU-H

E1was

new

lydiagnosedas

clinicalHNPCCin

thepresentstudy.–4SuspectedHNPCCmeansthatthepatientfulfilled

therevised

criteriaforsuspectedHNPCC.

FIGURE 1 – Flow diagram for overall study results. In 113 patients,information of personal and family cancer histories was available andmicrosatellite instability and immunohistochemistry were assessed.Two known clinical HNPCC patients, and 2 newly found clinicalHNPCC and 8 suspected HNPCC patients, and 11 patients showingMSI-H and loss of MMR protein expression without definite personalor family cancer history for clinical or suspected HNPCC were identi-fied, and germline mutations of them are as illustrated. 1MMR proteinis the abbreviation of mismatch repair protein; MLH1, MSH2 andMSH6 proteins.

FIGURE 2 – Representative examples of microsatellite instability.(a) BAT26; (b) D2S123. Arrows indicate microsatellite instability.

1079HNPCC IN KOREAN ENDOMETRIAL CANCER PATIENTS

However, the rate of MSI-H in Korean endometrial cancer patientsis higher than the 9.3% reported in primary sporadic CRC inKorea,17 and similarly, our finding that 23% of endometrial tumorsamples showed loss of MMR protein expression indicates thatthis rate is higher than the 8.6% seen in primary sporadic CRC inKorean patients.18

We performed mutational analyses of patient samples from theidentified clinical HNPCC, s-HNPCC and selected patients whoshowed both MSI-H and loss of MMR protein expression. Overall,the mutation rate in clinical HNPCC patients was 50% (2/4), andthat in the s-HNPCC patients was 12.5% (1/8). Although our sam-ple size was relatively small, these mutation rates are comparableto the 41.5% (22/53) and 19.8% (22/111) respectively found inclinical HNPCC and s-HNPCC patients identified through Korean

CRC probands and registered in the Korean Hereditary TumorRegistry.13 We did not test the two mutation-negative clinicalHNPCC patients (SNU-H24, SNU-HE2) for possible largegenomic rearrangements of DNA mismatch repair genes. Furthertest seems to be required to identify the role of large genomic rear-rangements in the pathogenesis of our study patients.

A previous report suggested that a combination of MSI and IHCanalysis could be used to predict MMR gene germline mutationsin patients with HNPCC.19 Consistent with this hypothesis, wefound that the loss of MSH2 or MSH6 expression combined withMSI-H accurately predicted germline mutation in our patients,regardless of personal and family cancer history.

Germline MSH6 mutations were detected in 2 of 11 patientswho did not meet the Amsterdam criteria II or the revised criteria

FIGURE 3 – Representative examples of immunohistochemical staining. (a) MLH1 negative; (b) MLH1 positive; (c) MSH2 negative; (d)MSH2 positive; (e) MSH6 negative; (f) MSH6 positive. Negative and positive mean the loss of expression and the retained expression, respec-tively. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

1080 YOON ET AL.

for s-HNPCC and showed both MSI-H and loss of MMR expres-sion; one with loss of both MSH2 and MSH6 expression, and theother with loss of MSH6 expression. A previous study reportedthat germline MSH6 mutations have a relatively low penetrancewith regard to neoplasm.20 However, endometrial cancer was alsoreported to be the most common cancer type among female car-riers of MSH6 germline mutations.21

Among 9 patients who did not meet the Amsterdam criteria IIor the revised criteria for s-HNPCC and showed both MSI-H andloss of MLH1 expression, we did not identify any germline MMRgene mutation. This prompted us to speculate the involvement ofMLH1 promoter hypermethylation, since previous studies of ran-domly selected CRC samples revealed that MLH1 inactivation ismost frequently due to MLH1 promoter hypermethylation.22 Fur-

thermore, up to 70% of MSI-positive endometrial cancers havebeen associated with MLH1 promoter hypermethylation.15 Con-sistent with this, 7 of these 9 patients with both MSI-H and loss ofMLH1 expression showed MLH1 promoter hypermethylation.

In conclusion, we found 3 mutation carriers in MLH1 or MSH2genes among 2 clinical HNPCC and 8 s-HNPCC patients identi-fied by family history. Additional 2 mutation carriers in MSH6gene were found by using MSI analysis and IHC staining. Thus,personal and family cancer histories should be obtained from allnewly diagnosed endometrial cancer patients to identify HNPCCpatients. In addition, MSI analysis and IHC staining for MMR pro-teins, especially for MSH6, might be helpful to find a small frac-tion of mutation carriers of MMR gene among patients with endo-metrial cancer.

References

1. Lee SE, Kim JW, Park NH, Song YS, Kang SB, Lee HP. Contempo-rary trends of endometrial cancer in Korean women. Korean J Gyne-col Oncol 2005;16:215–20.

2. Park JG, ed. Cancer incidence in Korea 1999–2001. Goyang NationalCancer Center, 2005.

3. Korea Central Cancer Registry. 2002 Annual report of the Korea Cen-tral Cancer Registry, based-on registered data from 139 hospitals.Goyang National Cancer Center, 2003.

4. Vasen HF, Watson P, Mecklin JP, Lynch HT. New clinical criteria forhereditary nonpolyposis colorectal cancer (HNPCC. Lynch syndrome)proposed by the. International. Collaborative. Group on HNPCC. Gas-troenterology 1999;116:1453–6.

5. Peltomaki P, Vasen H. Mutations associated with HNPCC predisposi-tion—update of ICG-HNPCC/INSiGHT mutation database. DisMarkers 2004;20:269–76.

6. Aarnio M, Sankila R, Pukkala E, Salovaara R, Aaltonen LA, de laChapelle A, Peltomaki P, Mecklin JP, Jarvinen HJ. Cancer risk inmutation carriers of DNA-mismatch-repair genes. Int J Cancer1999;81:214–18.

7. Hampel H, Frankel W, Panescu J, Lockman J, Sotamaa K, Fix D,Comeras I, La Jeunesse J, Nakagawa H, Westman JA, Prior TW,Clendenning M, et al. Screening for Lynch syndrome (hereditary non-polyposis colorectal cancer) among endometrial cancer patients. Can-cer Res 2006;66:7810–17.

8. Park JG, Vasen HF, Park YJ, Park KJ, Peltomaki P, de Leon MP,Rodriguez-Bigas MA, Lubinski J, Beck NE, Bisgaard ML, Miyaki M,Wijnen JT, et al. Suspected HNPCC and Amsterdam criteria. II. Eval-uation of mutation detection rate, an international collaborative study.Int J Colorectal Dis 2002;17:109–14.

9. Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman JR,Burt RW, Meltzer SJ, Rodriguez-Bigas MA, Fodde R, Ranzani GN,Srivastava S. A National Cancer Institute Workshop on MicrosatelliteInstability for cancer detection and familial predisposition: develop-ment of international criteria for the determination of microsatelliteinstability in colorectal cancer Cancer Res 1998;58:5248–57.

10. Chung WB, Hong SH, Kim JA, Sohn YK, Kim BW, Kim JW. Hyper-methylation of tumor-related genes in genitourinary cancer cell lines JKorean Med Sci 2001;16:756–61.

11. Kang GH, Yoon GS, Lee HK, Kwon YM, Ro JY. Clinicopathologiccharacteristics of replication error-positive gastric carcinoma. ModPathol 1999;12:15–20.

12. Shin KH, Shin JH, Kim JH, Park JG. Mutational analysis of promotersof mismatch repair genes hMSH2 and hMLH1 in hereditary nonpoly-posis colorectal cancer and early onset colorectal cancer patients:

identification of three novel germ-line mutations in promoter of thehMSH2 gene. Cancer Res 2002;62:38–42.

13. Shin YK, Heo SC, Shin JH, Hong SH, Ku JL, Yoo BC, Kim IJ, ParkJG. Germline mutations in MLH1, MSH2 and MSH6 in Korean he-reditary non-polyposis colorectal cancer families. Hum Mutat 2004;24:351.

14. Ollikainen M, Abdel-Rahman WM, Moisio AL, Lindroos A, KariolaR, Jarvela I, Poyhonen M, Butzow R, Peltomaki P. Molecular analysisof familial endometrial carcinoma: a manifestation of hereditary non-polyposis colorectal cancer or a separate syndrome? J Clin Oncol2005;23:4609–16.

15. Buttin BM, Powell MA, Mutch DG, Rader JS, Herzog TJ, Gibb RK,Huettner P, Edmonston TB, Goodfellow PJ. Increased risk for heredi-tary nonpolyposis colorectal cancer-associated synchronous and meta-chronous malignancies in patients with microsatellite instability-posi-tive endometrial carcinoma lacking MLH1 promoter methylation.Clin Cancer Res 2004;10:481–90.

16. Gurin CC, Federici MG, Kang L, Boyd J. Causes and consequencesof microsatellite instability in endometrial carcinoma. Cancer Res1999;59:462–6.

17. Lim SB, Jeong SY, Lee MR, Ku JL, Shin YK, Kim WH, Park JG.Prognostic significance of microsatellite instability in sporadic color-ectal cancer. Int J Colorectal Dis 2004;19:533–7.

18. Jeong SY, Shin KH, Shin JH, Ku JL, Shin YK, Park SY, Kim WH,Park JG. Microsatellite instability and mutations in DNA mismatchrepair genes in sporadic colorectal cancers. Dis Colon Rectum2003;46:1069–77.

19. de Leeuw WJ, Dierssen J, Vasen HF, Wijnen JT, Kenter GG, Meijers-Heijboer H, Brocker-Vriends A, Stormorken A, Moller P, Menko F,Cornelisse CJ, et al. Prediction of a mismatch repair gene defect bymicrosatellite instability and immunohistochemical analysis in endo-metrial tumours from HNPCC patients. J Pathol 2000;192:328–35.

20. Wagner A, Hendriks Y, Meijers-Heijboer EJ, de Leeuw WJ, Morreau H,Hofstra R, Tops C, Bik E, Brocker-Vriends AH, van Der Meer C, Lindh-out D, Vasen HF, et al. Atypical HNPCC owing toMSH6 germline muta-tions: analysis of a large Dutch pedigree. J Med Genet 2001;38:318–22.

21. Wijnen J, de Leeuw W, Vasen H, van der Klift H, Moller P, Stor-morken A, Meijers-Heijboer H, Lindhout D, Menko F, Vossen S,Moslein G, Tops C, et al. Familial endometrial cancer in female car-riers of MSH6 germline mutations Nat Genet 1999;23:142–4.

22. Cunningham JM, Christensen ER, Tester DJ, Kim CY, Roche PC,Burgart LJ, Thibodeau SN. Hypermethylation of the hMLH1 pro-moter in colon cancer with microsatellite instability. Cancer Res1998;58:3455–60.

1081HNPCC IN KOREAN ENDOMETRIAL CANCER PATIENTS