novel molecular proﬁles of endometrial cancer—new light

A

Io

sw

btT

ciotmt©

KR

BS

0d

Available online at www.sciencedirect.com

Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 221–229

Novel molecular profiles of endometrial cancer—new lightthrough old windows

A. Doll a, M. Abal a, M. Rigau a, M. Monge, M. Gonzalez a, S. Demajo a, E. Colas a,M. Llaurado, H. Alazzouzi a, J. Planaguma a, M.A. Lohmann e, J. Garcia c, S. Castellvi c,

J. Ramon y Cajal c, A. Gil-Moreno b, J. Xercavins b, F. Alameda d, J. Reventos a,∗a Unitat de Recerca Biomedica, Research Institute Vall d’Hebron University Hospital,

Barcelona, Spainb Department of Gynecology, Vall d’Hebron University Hospital,

Barcelona, Spainc Department of Pathology, Vall d’Hebron University Hospital,

Barcelona, Spaind Department of Pathology, Hospital del Mar, Barcelona, Spain

e Philipps-Universitat, Marburg, Germany

bstract

Endometrial carcinoma (EC) is the most common gynecological malignancy in the western world.A widely accepted dualistic model, which has been established on a morphological basis, differentiates EC into two broad categories: Type

oestrogen-dependent adenocarcinoma with an endometrioid morphology and Type II non-oestrogen-dependent EC with a serous papillaryr clear cell morphology.

Molecular genetic evidence indicates that endometrial carcinoma, as described in other malignancies, likely develops as the result of atepwise accumulation of alterations in cellular regulatory pathways, such as oncogene activation and tumor suppressor gene inactivation,hich lead to dysfunctional cell growth. These molecular alterations appear to be specific in Type I and Type II cancers.In type I endometrioid endometrial cancer, PTEN gene silencing in conjunction with defects in DNA mismatch repair genes, as evidenced

y the microsatellite instability phenotype, or mutations in the K-ras and/or �-catenin genes, are recognized major alterations, which definehe progression of the normal endometrium to hyperplasia, to endometrial intraepithelial neoplasia, and then on to carcinoma. In contrast,ype II cancers show mutations of TP53 and Her-2/neu and seem to arise from a background of atrophic endometrium.Nevertheless, despite the great effort made to establish a molecularly-based histological classification, the following issues must still be

larified: what triggers the tumor cells to invade the myometrium and what causes vascular or lymphatic dissemination, finally culminatingn metastasis? RUNX1, a transcription factor, was recently identified as one of the most highly over-expressed genes in a microarray studyf invasive endometrial carcinoma. Another candidate gene, which may be associated with an initial switch to myometrial infiltration, is the
ranscription factor ETV5/ERM. These studies, as well as those conducted for other genes possibly involved in the mitotic checkpoint as a
ajor mechanism of carcinogenesis in non-endometrioid endometrial cancer, could help in understanding the differences in the biology andhe clinical outcome among histological types.

2007 Elsevier Ltd. All rights reserved.

eywords: Oestrogen-dependent adenocarcinoma; Endometrioid morphology; NUNX1/AML1; ETV5/ERM

1

∗ Corresponding author at: INSTITUT DE RECERCA, Unitat de Recercaiomedica (URB), Pg. Vall d’Hebron, 119-129. 1a Planta, 08035 Barcelona,pain.

E-mail address: [email protected] (J. Reventos).

liN1

960-0760/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.oi:10.1016/j.jsbmb.2007.09.020

on-oestrogen-dependent; Endometrial carcinoma; Molecular pathology;

. Introduction

Endometrial cancer (EC) is the most common gyneco-
ogical malignancy encountered in western countries; thencidence is estimated at 15–20 per 100.000 women per year.inety percent of cases are sporadic, while the remaining0% arise from a genetic background.
mailto:[email protected]

dx.doi.org/10.1016/j.jsbmb.2007.09.020

222 A. Doll et al. / Journal of Steroid Biochemistry & Molecular Biology 108 (2008) 221–229

F ) of the(

imp

otippaoeMf

egcprlpt

sshbac

2e

2i

l

iiaot

dcgmhrdabciifoltobgbct1aic

2

m

ig. 1. (a) Histology of a grade I endometrioid endometrial carcinoma (EECc) Histology of a clear cell carcinoma of the uterus (NEEC).

Based on clinico-pathological and molecular character-stics, our current understanding of the biology of uterine

alignancies permits their dichotomy along two distinctathways [1].

The first is the Type-I or oestrogen-dependent endometri-id endometrial carcinomas (EECs) (Fig. 1a), representinghe majority of cases of sporadic endometrial cancer (approx-mately 80%) that arise in relatively younger pre- andost-menopausal women. They express oestrogen (ER) androgesterone (PR) receptors [2] and have a strong etiologicalssociation with either endogenous or exogenous, unopposedestrogen exposure [3]. They usually are low grade with anndometrioid morphology (with variants in all pathways ofullerian differentiation), and they are characterized by a

avorable prognosis.The second group, the Type II or non-endometrioid

ndometrial carcinomas (NEECs), is comprised of the high-rade papillary serous and clear cell carcinomas (Fig. 1b and). These arise in relatively older women and are not usuallyreceded by a history of unopposed estrogen exposure, butather from a background of atrophic endometrium. Theatter tumors have an aggressive clinical course, a greaterropensity for early spreading, and a worse prognosis thanhe more common endometrioid adenocarcinomas [4].

There is both epidemiological and molecular evidence touggest that endometrial hyperplasia represents a true precur-or lesion for endometrioid adenocarcinomas, with complexyperplasia relying on a histological and genetic continuumetween simple hyperplasia and carcinoma, whereas NEECsre frequently associated with endometrial intraepithelial car-inoma.

. Molecular genetics associated with endometrioidndometrial carcinoma

.1. DNA-mismatch repair genes and microsatellite
nstability
The progressive accumulation of alterations at microsatel-ite loci, so-called microsatellite instability (MSI), in

dgii

uterus. (b) Histology of a serous papillary carcinoma of the uterus (NEEC).

mportant regulatory genes may promote carcinogenesis ands known to play an important role in sporadic colon cancersnd in several non-colonic tumors. MSI is found in 17–25%f sporadic Type I ECs [5], but is rarely present in Type IIumors [6,7].

Abnormalities of DNA-mismatch repair genes were firstetected in tumors of patients with hereditary non-polyposisolorectal carcinoma (HNPCC), where MSI arose fromerm-line and somatic mutations in one of four DNA-ismatch repair genes. Most commonly, those genes are

MLH-1 and hMSH-2, which encode enzymes that areesponsible for repairing nucleotide mispairs, insertions oreletions produced during DNA replication [8–10]. Whilefew germ-line mutations of hMLH-2 and hMLH6 have

een described in EC [11,12], most mismatch repair defi-iencies underlying MSI in EC seem to involve an epigeneticnactivation or silencing mechanism, rather than mutationsn DNA repair genes, as these are encountered with lowrequency in sporadic cases of EC. Indeed, inactivationf MLH1 by the hypermethylation of normally unmethy-ated CpG islands in the promoter has been described ashe most common cause of MSI in sporadic endometri-id carcinomas [13,14]. Similar hypermethylation has alsoeen demonstrated in promoters of the APC and MGMTenes associated with Type I tumors. Microsatellite insta-ility is an uncommon feature in high-grade endometrialarcinomas [15]. Other commonly implicated genes includehe PTEN/MMAC1 tumor suppressor gene on chromosome0q23, the inactivation of which is found in up to 50% ofll endometrial cancers, and at an even higher frequencyf the subset of tumors with microsatellite instability isonsidered.

.2. Oncogenes

The K-ras gene encodes a small inner plasma, cellularembrane GTPase, which functions as a molecular switch

uring cell signaling and which is largely related to tumorrowth and differentiation. Mutations of K-ras have beendentified in 19–46% of ECs, with most of these cases involv-ng point mutations at codon 12 [16].

istry &

acatMotrra

rpimbp

2

ksHipna

rht[

2

htfRtEiiisiE

mtmrp

ifeatbt

ohcEstss

ewtrnp

mtcCtd

tdas[

oihebetsaomit[

A. Doll et al. / Journal of Steroid Biochem

Alterations of K-ras predominantly involve Type I tumorsnd have been reported in 10–30% of cases of endometrioidarcinoma, whereas they are almost absent in papillary serousnd clear-cell carcinomas [17]. Constitutive activating muta-ions in K-ras have also been found to be more frequent in

SI-positive tumors [17], suggesting that both events mayccur simultaneously before clonal expansion [18]. Muta-ions are also detected in endometrial hyperplasia at a similarate to that observed in EC, suggesting that mutations in theas gene may represent an early event in tumorigenesis withinsubset of Type I ECs [19].

Studies have shown that the frequency of K-ras mutationsises progressively from simple through to complex hyper-lasia and to carcinoma, and the presence of K-ras mutationsn pre-malignant biopsy samples has been suggested as a

arker of progression to malignancy. K-ras mutations haveeen shown to be associated with an adverse prognosis, inde-endent of age or stage [20,21].

.3. HER2/neu

The HER2/neu (also called erbB2) gene encodes a 185-Da transmembrane receptor, tyrosine kinase, which isimilar to the epidermal growth factor receptor (EGF-R).ER2/neu functions as a preferred partner for heterodimer-

zation with members of the EGF-R family and, therefore,lays an important role in coordinating the complex ErbB sig-aling network that is responsible for regulating cell growthnd differentiation [22].

The over-expression of Her-2/neu oncoprotein has beeneported in 9–30% of all endometrial adenocarcinomas, andas been associated in some studies with an adverse prognos-ic outcome and has been linked to decreased overall survival23,24].

.4. Other oncogenes

The involvement of the above-mentioned oncogenesas been well established in EC. However, many addi-ional proto-oncogenes are currently under investigationor their possible involvement in EC. C-myc, survivin,UNX1, ETV5, and human telomerase reverse transcrip-

ase (hTERT) have all been described in association withC. c-myc amplification and over-expression is present

n approximately 3–19% of ECs, and despite conflict-ng reports of links to tumor differentiation, myometrialnvasion and lymph node metastases, a recent report hashown nuclear and cytoplasmic c-myc immunohistochem-cal staining to be an independent prognostic factor inC [23,25,26].

The over-expression of hTERT is involved in EC develop-ent, particularly in conjunction with tamoxifen, by causing

he activation of telomerase and the subsequent telomereaintenance and potential cell immortalization; however, the

egulation of hTERT is known to be partly dependent on generoducts previously linked to EC (e.g., c-myc). Further stud-

2

c

Molecular Biology 108 (2008) 221–229 223

es are needed to establish its exact role [27]. The same goesor the inhibitor of the apoptosis protein survivin, whose over-xpression has been associated with a higher clinical gradend stage, but whose definitive role in tumorigenesis has yeto be established [28]. None of these three oncogenes haseen studied in relation to specific prevalence in type I orype II tumors.

ETV5/ERM (Ets-related protein) is a transcription factorf the ETS family and a divergent member of the wingedelix-turn-helix super-family. ETV5 binds to sequencesontaining the consensus pentanucleotide 5′-CGGA(AT)-3′.TV5 is a proto-oncogene that plays a role in the progres-ion of breast cancer, functions as an adaptor molecule inhe interactions of adhesion receptors and intracellular tyro-ine kinases, and is required for spermatogonial stem cellelf-renewal [29,30].

ETV5 has been proposed to play a role during the earlyvents of endometrial tumorigenesis and could be associatedith an initial switch to myometrial infiltration. Furthermore,

issue array immunohistochemistry has shown that this up-egulation correlated with the process of tumorigenesis fromormal atrophic endometrium to simple and complex hyper-lasia and then, on to carcinoma [31,32].

RUNX1/AML1 (runt-related transcription factor 1/acuteyeloid leukemia 1) belongs to the RUNX gene family of

ranscription factors that bind DNA as components of theore-binding factor (CBF) complex, in partnership with theBF� cofactor. This complex activates and represses the

ranscription of key regulators of the growth, survival andifferentiation pathways.

The RUNX genes have been found to function as bothumor suppressors and dominant oncogenes in a context-ependent manner. The RUNX genes are closely related andre essential for hematopoiesis, osteogenesis and neurogene-is, but are also important for other developmental processes34].

The possibility that RUNX1 over-expression plays anncogenic role outside the hematopoietic system has beenndicated by its recent identification as one of the mostighly over-expressed genes in a microarray study of invasivendometrial carcinoma [31]. A strong positive correlationetween RUNX1/AML1 and p21WAF1/CIP1 was found,specially in stage IC carcinomas, which infiltrated morehan 50% of the myometrium. Thus, it could be hypothe-ized that p21WAF1/CIP1, a target of p53-mediated growthrrest, and RUNX1/AML1 interact during the initial stepsf tumor dissemination in EEC through TGF�/SMAD-ediated myometrial infiltration and/or through a shift

n the balance of the cell growth/cell differentiationoward invasive, rather than proliferative, phenotypes35].

.5. Tumor suppressor genes

The PTEN (phosphatase and tensin homolog deleted onhromosome ten) gene is a tumor suppressor gene localized

2 istry &

oih

wcpmmmtPa

e(hPra[

hcgdm[ppduham[

mteenaM

m(ie

eotpbk

sr[v[fwppa

Accl�aactbtmfuhpnhtE

1rDtpsicpmitiiisl7il

24 A. Doll et al. / Journal of Steroid Biochem

n chromosome 10 (10q23-24) whose name derives fromts preserved tyrosine phosphatase domain and its sequenceomology with the matrix protein tensin [36].

The PTEN gene encodes a 403 amino acid protein,hich acts as a lipid phosphatase that helps to modulate

ell signal transduction pathways by acting on phospholipidhosphatidylinositol-(3,4,5)-triphosphate (PIP3), a secondessenger produced after growth factors bind to cell surfaceembrane receptors. Decreased activity or loss-of-functionutations of PTEN and constitutive activation lead to consti-

utive activation of multiple signaling pathways, including theI3K/Akt pathway, which affects cell proliferation, apoptosisnd migration [37,38].

PTEN has also been shown to control p53 protein lev-ls and transcriptional activity [39]. A number of tumorse.g., glioblastoma multiforme, prostate carcinoma, etc.)ave been shown to have acquire abnormalities of theTEN gene. Up to 80% of cases of endometrioid carcinomaeveal a loss of expression, mainly due to mutations [40],nd to a lesser extent, to a loss of heterozygosity (LOH)41].

Interestingly, higher rates of mutations in the PTEN geneave been described to occur in MSI tumors (60–86%)ompared to tumors without MSI (24–35%). This sug-ests that PTEN could be a target for mutations in aeficient DNA repair context [42]. They are also well docu-ented in endometrial hyperplasia with and without atypia

43,44]. Given the role of endometrial hyperplasia as theutative precursor of Type I tumors, PTEN mutations areresumed to play an early role, although probably not aetermining step, in tumorigenesis. Recent studies have doc-mented a case of synchronous PTEN-positive/MSI-negativeyperplasia and PTEN-positive/MSI-negative carcinoma,nd they have proposed that PTEN mutations could precedeicrosatellite instability; however, this remains controversial

43,45].The association of PTEN inactivation with MSI and

ethylation and the low expression of MLH1 seemo lay the foundations for the initial steps towardsndometrioid endometrial tumorigenesis [46]. Moreover,ndometrial pre-cancers (e.g., endometrial intraepithelialeoplasia) have been postulated to share common geneticlterations with EEC, including PTEN mutations andSI.The catenins are a family of structurally related cytoplas-

ic proteins, which have been classified as alpha (�), beta�), and gamma (�), according to their electrophoretic mobil-ty. The �-catenin gene is located on chromosome 3p21 andncodes an 88 kDa protein [47,48].

�-Catenin is a multi-functional protein that plays anssential role in the E-cadherin-mediated anchoring andrganization of the cytoskeleton and acts as a downstream
ranscriptional activator in the Wnt signaling transductionathway. Cellular �-catenin levels are tightly regulatedy a multi-protein complex comprised of serine/threonineinase GSK3�, the APC (adenomatous polyposis coli) tumor
oWs(

Molecular Biology 108 (2008) 221–229

uppressor gene product, and axin, which facilitates phospho-ylation and subsequent degradation of the �-catenin protein49]. In the latter role, �-catenin has been implicated in a wideariety of malignancies including those of the endometrium50]. Excessive �-catenin protein is usually removed rapidlyrom the cytoplasmic pool by the ubiquitin-proteasome path-ay (so-called ubiquitination). However, the mutated generoduct resists degradation and can accumulate in the cyto-lasm and the cell nucleus with constitutive target genectivity.

Abnormalities of tumor suppressor genes, such as thePC gene or mutations in �-catenin, have been shown to

ause dysregulation of �-catenin degradation, leading to aytoplasmic accumulation of the protein, which is then fol-owed by translocation to the nucleus. The accumulation of-catenin has been demonstrated by immunohistochemistry,nd several studies have analyzed series of EC for nuclearccumulation, which has been found to be significantly moreommon in tumors of endometrioid morphology (31–47%)han in the non-endometrioid forms (0–3%); [51,52]. It haseen suggested that the discordance of gene mutation andhe nuclear accumulation rates could be attributable to abnor-

alities in other Wnt proteins (e.g., APC, �-catenin), but theunction of �-catenin in endometrioid tumorigenesis remainsnclear. No correlations to MSI, K-Ras or PTEN mutationsave been found, suggesting that the Wnt pathway couldlay an independent role in endometrial cancer [53]. Becauseuclear accumulation has also been demonstrated in atypicalyperplasia, it has been suggested that �-catenin abnormali-ies could arise relatively early in the development of Type IC [54,55].

The TP53 tumor suppressor gene located on chromosome7 codes for a nuclear protein, which plays an importantole in preventing the propagation of cells with damagedNA. After DNA damage, nuclear p53 accumulates and,

hrough p21, causes cell cycle arrest by inhibiting cyclin-D1hosphorylation of the Rb gene and by promoting apopto-is [56]. Abnormalities of TP53 have been well describedn various malignancies, and germ-line mutations are asso-iated with the Li-Fraumeni syndrome. The mutant p53rotein is non-functional but resists degradation, and it accu-ulates in the cell, thereby acting as a dominant negative

nhibitor of the wild-type p53. The accumulated mutant pro-ein can be demonstrated immunohistochemically. Mutationsn the p53 gene are a frequent and characteristic findingn Type II serous tumors with positive immunohistochem-stry reported in 71–85% of tumors [57,58]. It has beenhown that p53 abrogation occurs relatively early in the evo-ution of these tumors, and over-expression is present in5% of endometrial intraepithelial carcinomas (EICs), whichs considered to be the putative precursor lesions of papil-ary serous cancers [59,60]. In contrast, p53 is mutated in
nly about a third of endometrioid adenocarcinomas [61,62].hen stratified by histological grade, grades 1 and 2 tumors
how positive staining at much lower rates than high gradegrade 3) tumors. Only a minor proportion of endometrial

istry &

histt

2

tnticttttw[

cEEt[ct

dlacr

epdncbs[ct

ieioreita

bvebtiahfgpePcoiPdTPw

rlbtvtcrwma

3

recognized over 20 years ago [72], and in general, thediscrimination between endometrioid adenocarcinomas andhigh-grade papillary serous and clear cell carcinomas hasbeen well established on a morphological basis [73,74].

Table 1Clinico-pathological characteristics and genetic abnormalities in Type I andType II ECs

Characteristic Type I (EEC) Type II (NEEC)

Unopposed oestrogen Yes NoBackground endometrium Hyperplastic AtrophicMorphology Endometrioid Serous, clear cellMicrosatellite instability 20–40% 0–5%p53 mutations 10–20% 90%


yperplasias, with complex hyperplasias show positive stain-ng, demonstrating a higher frequency of abnormalities thanimple hyperplasias. This suggests, in contrast to serousumors, that p53 occurs as a late molecular event in Type Iumors [17].

.6. Steroid receptor genes

Oestrogen receptors ER�, ERß and progesterone recep-ors (PR-A, PR-B) belong to the steroid/thyroid hormoneuclear receptor super-family. They are ligand-activatedranscription factors involved in hormone-mediated signal-ng, hormone-mediated inhibition of gene expression, andellular proliferation and differentiation in various targetissues. They are translocated from the plasma membraneo the nucleus and are mediated by MAP kinase activa-ion. They bind as a homo- or heterodimer with ESR2o oestrogen-responsive elements (ERE), or they interactith other transcription factors such as AP-1 or NF-�B

63].The predominant subtype in the uterus is ER�. Increased

ell proliferation and exaggerated response to oestrogen inR� was found in an ER� knockout model, suggesting thatR� could play a role in modulating ER� function and

hat it consequently could have an anti-proliferative function59,60]. An imbalance in ER� and ER� expression, therefore,ould signify the crucial, critical step in oestrogen-dependentumorigenesis.

ER� mRNA and protein expression are reported toecrease in stages from normal or grade I to grade 3 tumoresions. In contrast, ER� expression is not altered, suggesting

shift to a decreased ER�/ER� ratio [64,65]. The signifi-ance of the relative expression of both ER subtypes in ECemains to be clarified.

Variant proteins originating from transcriptional splicingrrors have been described for ER� and ER�. For exam-le, an ER� exon 5 splice variant (�5 ER�) has not beenetected in normal endometrium but has been found at sig-ificantly increased levels in endometrial carcinomas, whenompared with endometrial hyperplasias [66]. �5 ER� haseen shown to be able to constitutively activate the tran-cription of ER-dependent genes in the absence of hormone67,68]. This could, therefore, provide endometrial tumorells with a growth advantage, potentially leading to uncon-rolled proliferation.

Progesterone acting through the PR is the physiolog-cal negative regulator of the oestrogen action in thendometrium. The expression of PR in endometrial glandss under the control of oestrogen and progesterone, whereestrogen induces PR synthesis and progesterone down-egulates the expression of its own receptor. The receptor
xists in two distinct isoforms: PR-A and PR-B, which aredentical except for an additional 164 amino acids in the N-erminus of PR-B. PR-A acts as a transcriptional repressornd PR-B as an activator [69].
�

KPH


The major role of PR-A in the endometrium is thought toe the down-regulation of the oestrogen action through pre-ention of ER� transactivation. In contrast, PR-B acts as anndometrial oestrogen-agonist. PR-A is therefore believed toe essential for the inhibition of oestrogen-induced endome-rial proliferation, partly by limiting PR-B effects. Themportance of the PR-A/PR-B ratio has been indicated inreport on the aberrant ratios of PR isoforms in endometrialyperplasias and EC. Only one PR isoform is commonlyound in endometrial cancers, and the expression of a sin-le PR isoform is associated with a higher clinical grade,ointing to a relationship between the loss of PR isoformxpression and a poor prognosis. The disruption of relativeR isoform expression has also been observed in cases ofomplex atypical hyperplasia. Early alterations in the ratiof PR-A/PR-B may therefore precede and/or be implicatedn the development of EC [70]. The importance of a balancedR-A/PR-B ratio has been further underlined by a recentlyescribed functional polymorphism in the promoter of PR.his polymorphism resulted in the increased transcription ofR-B and an altered PR-A/PR-B ratio and was associatedith an increased risk for EC [71].Progesterone exposure diminishes endometrial cancer

isk, due to the interruption of continued oestrogenic stimu-ation of the endometrium. Endometrial hyperplasia, causedy unopposed oestrogenic stimulation, can be reversed byreatment with progestins. Progestin treatment also pro-ides protection against the stimulatory effects of oestrogenicreatments; i.e., hormone replacement therapy (HRT), usingombinations of oestrogens and progestins, yields a lowerisk of endometrial carcinoma than treatments associatedith oestrogen alone. Progestational agents are the most com-on type of hormone treatment for endometrial hyperplasia

nd endometrial cancer.

. Conclusions

The dualistic model for endometrial cancer was first

-Catenin mutations 31–47% 0–3%-ras mutations 15–30% 0–5%TEN inactivation 35–50% 10%ER2/neu No information 18–80%

2 istry &

eaiaam[we(toppaapccicocpcebceeomefpMg

NtmTptc

artdahsaagcdamisa

A

fdl2


In summary, a progression model for the development ofndometrioid carcinoma has been proposed, based on geneticlterations already present in atypical hyperplasias, on thencrease in the nature and prevalence of these genetic alter-tions in well-differentiated carcinomas when compared totypical hyperplasias, and on a higher number of chromoso-al aberrations in endometrioid lesions than in hyperplasias

75] (see Table 1). First, a genetic background can provide usith information about endometrial carcinoma susceptibility,

specially with information on those high-penetrance genesi.e., DNA mismatch repair genes), and also with informa-ion on the status of low-penetrance genes associated with theestrogen metabolism [76–78]. This hypothesis proposes therogression from simple to complex hyperplasia as a reactiverocess due to hyper-oestrogenism [79], while monoclonalityssociated with PTEN and K-Ras mutations, as well as theppearance of microsatellite instability, seem to define therogression from atypical hyperplasia to endometrioid car-inoma [80]. It has been demonstrated that the detection oflonality in endometrial biopsy samples obtained by pipelles a useful application for the early diagnosis of endometrialancer [81]. Both alterations, together with the methylationf the MLH1 promoter, coexist in atypical hyperplasia adja-ent to endometrioid carcinoma [43,82–84]. Mutations in53 and the amplification and over-expression of HER2/neuharacterize late events during progression and the dediffer-ntiation of endometrioid carcinoma [17,85]. Alternatively,ased on findings from mixed endometrioid and serous car-inomas [86], these later molecular alterations could definearly events in de novo, occurring in poorly-differentiatedndometrioid carcinomas and those serous carcinomas devel-ping from endometrioid carcinomas. Despite the great effortade to unravel the molecular alterations associated with

ndometrial tumorigenesis and the clearly demonstrated use-
ulness of these alterations in understanding the molecularathogenesis of endometrial carcinomas, tumors lacking theSI phenotype or mutations in any of the above-mentioned
enes suggest the existence of still unrecognized pathways.

rMt0


evertheless, not all endometrial carcinomas fit into thesewo pathways; some tumors show mixed or overlapping

orphological and immunohistochemical features of bothype-I and Type-II endometrial carcinomas. Thus, it has beenroposed that occasional serous carcinomas could develophrough the dedifferentiation of pre-existing endometrioidarcinomas [86].

Nevertheless, despite the great effort made to establishmolecularly-based histological classification, some points

emain to be clarified. What triggers the tumor cells to invadehe myometrium, and what causes vascular or lymphaticissemination, finally culminating in metastasis? RUNX1,transcription factor recently identified as one of the most

ighly over-expressed genes in a microarray study of inva-ive endometrial carcinoma, and ETV5/ERM, which may bessociated with an initial switch to myometrial infiltrationnd whose up-regulation correlated to the process of tumori-enesis from normal atrophic endometrium to simple andomplex hyperplasia and then, on to carcinoma, are good can-idate genes meriting further characterization. These studies,s well as those conducted on other genes involved in theitotic checkpoint as a major mechanism of carcinogenesis

n non-endometrioid endometrial cancer, could help to under-tand the differences in the biology and the clinical outcomemong the histotypes.

cknowledgements

Andreas Doll is the recipient of a postdoctoral fellowshiprom the Institut de Recerca del Hospital Universitari Vall’Hebron. Eva Colas is the recipient of a predoctoral fel-owship del Ministerio de Educacion y Ciencia (grant BES006 14152). The authors also want to thank the support
eceived for this research from: Marato TV3 (grant 050431),
inisterio de Educacion y Ciencia grant (SAF 2005-06711),he Fondo de Investigaciones Sanitarias (grants 99/0897,1/10076, 02/0733), the Departament de Universitats, de

istry &

RCCtm

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[


ecerca i de Societat de la Informacio de la Generalitat deatalunya (grants SGR00231, 00391), the Instituto de Saludarlos III (Red de Genomica del Cancer y Genotipado de

umores C03/10) and the Institut Catala de la Salut. Theanuscript revision by Lisa Piccione is also acknowledged.

eferences

[1] S.F. Lax, R.J. Kurman, A dualistic model for endometrial carcinogen-esis based on immunohistochemical and molecular genetic analyses,Verh. Dtsch. Ges. Pathol. 81 (1997) 228–232.

[2] S.F. Lax, E.S. Pizer, B.M. Ronnett, R.J. Kurman, Comparison ofestrogen and progesterone receptor, Ki-67, and p53 immunoreactivityin uterine endometrioid carcinoma and endometrioid carcinoma withsquamous, mucinous, secretory, and ciliated cell differentiation, Hum.Pathol. 29 (1998) 924–931.

[3] M.E. Sherman, S. Sturgeon, L.A. Brinton, N. Potischman, R.J. Kurman,M.L. Berman, R. Mortel, L.B. Twiggs, R.J. Barrett, G.D. Wilbanks,Risk factors and hormone levels in patients with serous and endometri-oid uterine carcinomas, Mod. Pathol. 10 (1997) 963–968.

[4] M.K. Oehler, A. Brand, G.V. Wain, Molecular genetics and endometrialcancer, J. Br. Menopause Soc. 9 (2003) 27–31.

[5] H.B. Salvesen, N. MacDonald, A. Ryan, O.E. Iversen, I.J. Jacobs,L.A. Akslen, S. Das, Methylation of hMLH1 in a population-basedseries of endometrial carcinomas, Clin. Cancer Res. 6 (2000) 3607–3613.

[6] L. Catasus, P. Machin, X. Matias-Guiu, J. Prat, Microsatellite instabilityin endometrial carcinomas: clinicopathologic correlations in a series of42 cases, Hum. Pathol. 29 (1998) 1160–1164.

[7] H. Tashiro, M.S. Blazes, R. Wu, K.R. Cho, S. Bose, S.I. Wang, J. Li,R. Parsons, L.H. Ellenson, Mutations in PTEN are frequent in endome-trial carcinoma but rare in other common gynecological malignancies,Cancer Res. 57 (1997) 3935–3940.

[8] D. Hardisson, G. Moreno-Bueno, L. Sanchez, D. Sarrio, A. Suarez, F.Calero, J. Palacios, Tissue microarray immunohistochemical expres-sion analysis of mismatch repair (hMLH1 and hMSH2 genes) inendometrial carcinoma and atypical endometrial hyperplasia: relation-ship with microsatellite instability, Mod. Pathol. 16 (2003) 1148–1158.

[9] A. Koul, R. Willen, P.O. Bendahl, M. Nilbert, A. Borg, Distinct sets ofgene alterations in endometrial carcinoma implicate alternate modes oftumorigenesis, Cancer 94 (2002) 2369–2379.

10] H.B. Salvesen, N. MacDonald, A. Ryan, I.J. Jacobs, E.D. Lynch, L.A.Akslen, S. Das, PTEN methylation is associated with advanced stageand microsatellite instability in endometrial carcinoma, Int. J. Cancer91 (2001) 22–26.

11] N.P. Taylor, M.A. Powell, R.K. Gibb, J.S. Rader, P.C. Huettner, S.N.Thibodeau, D.G. Mutch, P.J. Goodfellow, MLH3 mutation in endome-trial cancer, Cancer Res. 66 (2006) 7502–7508.

12] F. Bianchi, S. Rosati, L. Belvederesi, C. Loretelli, R. Catalani, A. Man-dolesi, R. Bracci, I. Bearzi, E. Porfiri, R. Cellerino, MSH2 splice sitemutation and endometrial cancer, Int. J. Gynecol. Cancer 16 (2006)1419–1423.

13] M. Esteller, R. Levine, S.B. Baylin, L.H. Ellenson, J.G. Herman, MLH1promoter hypermethylation is associated with the microsatellite insta-bility phenotype in sporadic endometrial carcinomas, Oncogene 17(1998) 2413–2417.

14] N.D. Macdonald, H.B. Salvesen, A. Ryan, S. Malatos, I. Stefansson,O.E. Iversen, L.A. Akslen, S. Das, I.J. Jacobs, Molecular differencesbetween RER+ and RER- sporadic endometrial carcinomas in a largepopulation-based series, Int. J. Gynecol. Cancer 14 (2004) 957–965.

15] J.I. Risinger, G.L. Maxwell, A. Berchuck, J.C. Barrett, Promoterhypermethylation as an epigenetic component in Type I and Type IIendometrial cancers, Ann. N.Y. Acad. Sci. 983 (2003) 208–212.

16] M. Esteller, A. Garcia, J.M. Martinez-Palones, J. Xercavins, J. Reven-tos, The clinicopathological significance of K-RAS point mutation and

[

[


gene amplification in endometrial cancer, Eur. J. Cancer 33 (1997)1572–1577.

17] S.F. Lax, B. Kendall, H. Tashiro, R.J. Slebos, L. Hedrick, The frequencyof p53, K-ras mutations, and microsatellite instability differs in uter-ine endometrioid and serous carcinoma: evidence of distinct moleculargenetic pathways, Cancer 88 (2000) 814–824.

18] H. Lagarda, L. Catasus, R. Arguelles, X. Matias-Guiu, J. Prat, K-rasmutations in endometrial carcinomas with microsatellite instability, J.Pathol. 193 (2001) 193–199.

19] H. Sasaki, H. Nishii, H. Takahashi, A. Tada, M. Furusato, Y. Terashima,G.P. Siegal, S.L. Parker, M.F. Kohler, A. Berchuck, et al., Mutationof the Ki-ras protooncogene in human endometrial hyperplasia andcarcinoma, Cancer Res. 53 (1993) 1906–1910.

20] I.N. Mammas, A. Zafiropoulos, D.A. Spandidos, Involvement of the rasgenes in female genital tract cancer, Int. J. Oncol. 26 (2005) 1241–1255.

21] M. Abal, J. Planaguma, A. Gil-Moreno, M. Monge, M. Gonzalez, T.Baro, A. Garcia, J. Castellvi, Y.C.S. Ramon, J. Xercavins, et al., Molec-ular pathology of endometrial carcinoma: transcriptional signature inendometrioid tumors, Histol. Histopathol. 21 (2006) 197–204.

22] W.C. Dougall, X. Qian, N.C. Peterson, M.J. Miller, A. Samanta,M.I. Greene, The neu-oncogene: signal transduction pathways, trans-formation mechanisms and evolving therapies, Oncogene 9 (1994)2109–2123.

23] D. Niederacher, H.X. An, Y.J. Cho, P. Hantschmann, H.G. Ben-der, M.W. Beckmann, Mutations and amplification of oncogenes inendometrial cancer, Oncology 56 (1999) 59–65.

24] B. Saffari, L.A. Jones, A. el-Naggar, J.C. Felix, J. George, M.F. Press,Amplification and overexpression of HER-2/neu (c-erbB2) in endome-trial cancers: correlation with overall survival, Cancer Res. 55 (1995)5693–5698.

25] J.A. Williams Jr., Z.R. Wang, R.S. Parrish, L.J. Hazlett, S.T. Smith,S.R. Young, Fluorescence in situ hybridization analysis of HER-2/neu,c-myc, and p53 in endometrial cancer, Exp. Mol. Pathol. 67 (1999)135–143.

26] J.P. Geisler, H.E. Geisler, K.J. Manahan, G.A. Miller, M.C. Wie-mann, Z. Zhou, W. Crabtree, Nuclear and cytoplasmic c-myc stainingin endometrial carcinoma and their relationship to survival, Int. J.Gynecol. Cancer 14 (2004) 133–137.

27] Z. Wang, S. Kyo, Y. Maida, M. Takakura, M. Tanaka, N. Yatabe, T.Kanaya, M. Nakamura, K. Koike, K. Hisamoto, et al., Tamoxifen regu-lates human telomerase reverse transcriptase (hTERT) gene expressiondifferently in breast and endometrial cancer cells, Oncogene 21 (2002)3517–3524.

28] N. Takai, T. Ueda, M. Nishida, K. Nasu, I. Miyakawa, The relation-ship between oncogene expression and clinical outcome in endometrialcarcinoma, Curr. Cancer Drug Targets 4 (2004) 511–520.

29] D. Monte, L. Coutte, F. Dewitte, P.A. Defossez, M. Le Coniat, D. Ste-helin, R. Berger, Y. de Launoit, Genomic organization of the humanERM (ETV5) gene, a PEA3 group member of ETS transcription factors,Genomics 35 (1996) 236–240.

30] C. Chen, W. Ouyang, V. Grigura, Q. Zhou, K. Carnes, H. Lim, G.Q.Zhao, S. Arber, N. Kurpios, T.L. Murphy, et al., ERM is required fortranscriptional control of the spermatogonial stem cell niche, Nature436 (2005) 1030–1034.

31] J. Planaguma, M. Diaz-Fuertes, A. Gil-Moreno, M. Abal, M. Monge,A. Garcia, T. Baro, T.M. Thomson, J. Xercavins, F. Alameda, J. Reven-tos, A differential gene expression profile reveals overexpression ofRUNX1/AML1 in invasive endometrioid carcinoma, Cancer Res. 64(2004) 8846–8853.

32] J. Planaguma, M. Abal, A. Gil-Moreno, M. Diaz-Fuertes, M. Monge, A.Garcia, T. Baro, J. Xercavins, J. Reventos, F. Alameda, Up-regulationof ERM/ETV5 correlates with the degree of myometrial infiltration in
endometrioid endometrial carcinoma, J. Pathol. 207 (2005) 422–429.
34] K. Blyth, E.R. Cameron, J.C. Neil, The RUNX genes: gain or loss offunction in cancer, Nat. Rev. Cancer 5 (2005) 376–387.

35] J. Planaguma, M. Gonzalez, A. Doll, M. Monge, A. Gil-Moreno, T.Baro, A. Garcia, J. Xercavins, F. Alameda, M. Abal, J. Reventos, The

2 istry &

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[


up-regulation profiles of p21WAF1/CIP1 and RUNX1/AML1 correlatewith myometrial infiltration in endometrioid endometrial carcinoma,Hum. Pathol. 37 (2006) 1050–1057.

36] P.A. Steck, M.A. Pershouse, S.A. Jasser, W.K. Yung, H. Lin, A.H.Ligon, L.A. Langford, M.L. Baumgard, T. Hattier, T. Davis, et al.,Identification of a candidate tumour suppressor gene, MMAC1, at chro-mosome 10q23.3 that is mutated in multiple advanced cancers, Nat.Genet. 15 (1997) 356–362.

37] J. Downward, PI 3-kinase, Akt and cell survival, Semin. Cell Dev. Biol.15 (2004) 177–182.

38] M.M. Hill, B.A. Hemmings, Inhibition of protein kinase B/Akt. impli-cations for cancer therapy, Pharmacol. Ther. 93 (2002) 243–251.

39] D.J. Freeman, A.G. Li, G. Wei, H.H. Li, N. Kertesz, R. Lesche, A.D.Whale, H. Martinez-Diaz, N. Rozengurt, R.D. Cardiff, et al., PTENtumor suppressor regulates p53 protein levels and activity throughphosphatase-dependent and -independent mechanisms, Cancer Cell 3(2003) 117–130.

40] G.L. Mutter, M.C. Lin, J.T. Fitzgerald, J.B. Kum, J.P. Baak, J.A. Lees,L.P. Weng, C. Eng, Altered PTEN expression as a diagnostic markerfor the earliest endometrial precancers, J. Natl. Cancer Inst. 92 (2000)924–930.

41] S.B. Simpkins, S. Peiffer-Schneider, D.G. Mutch, D. Gersell, P.J.Goodfellow, PTEN mutations in endometrial cancers with 10q LOH:additional evidence for the involvement of multiple tumor suppressors,Gynecol. Oncol. 71 (1998) 391–395.

42] C. Bilbao, G. Rodriguez, R. Ramirez, O. Falcon, L. Leon, R. Chirino,J.F. Rivero, O. Falcon Jr., B.N. Diaz-Chico, J.C. Diaz-Chico, M. Peru-cho, The relationship between microsatellite instability and PTEN genemutations in endometrial cancer, Int. J. Cancer 119 (2006) 563–570.

43] R.L. Levine, C.B. Cargile, M.S. Blazes, B. van Rees, R.J. Kurman,L.H. Ellenson, PTEN mutations and microsatellite instability in com-plex atypical hyperplasia, a precursor lesion to uterine endometrioidcarcinoma, Cancer Res. 58 (1998) 3254–3258.

44] G.L. Maxwell, J.I. Risinger, C. Gumbs, H. Shaw, R.C. Bentley, J.C. Bar-rett, A. Berchuck, P.A. Futreal, Mutation of the PTEN tumor suppressorgene in endometrial hyperplasias, Cancer Res. 58 (1998) 2500–2503.

45] E. Latta, W.B. Chapman, PTEN mutations and evolving concepts inendometrial neoplasia, Curr. Opin. Obstet. Gynecol. 14 (2002) 59–65.

46] H.B. Salvesen, I. Stefansson, E.I. Kretzschmar, P. Gruber, N.D. Mac-Donald, A. Ryan, I.J. Jacobs, L.A. Akslen, S. Das, Significance ofPTEN alterations in endometrial carcinoma: a population-based studyof mutations, promoter methylation and PTEN protein expression, Int.J. Oncol. 25 (2004) 1615–1623.

47] M.E. Bracke, F.M. Van Roy, M.M. Mareel, The E-cadherin/catenincomplex in invasion and metastasis, Curr. Top. Microbiol. Immunol.213 (Pt 1) (1996) 123–161.

48] M. Ozawa, H. Baribault, R. Kemler, The cytoplasmic domain of thecell adhesion molecule uvomorulin associates with three independentproteins structurally related in different species, EMBO J. 8 (1989)1711–1717.

49] L.C. Bullions, A.J. Levine, The role of beta-catenin in cell adhesion,signal transduction, and cancer, Curr. Opin. Oncol. 10 (1998) 81–87.

50] T. Fukuchi, M. Sakamoto, H. Tsuda, K. Maruyama, S. Nozawa, S. Hiro-hashi, Beta-catenin mutation in carcinoma of the uterine endometrium,Cancer Res. 58 (1998) 3526–3528.

51] G. Moreno-Bueno, D. Hardisson, C. Sanchez, D. Sarrio, R. Cassia, G.Garcia-Rostan, J. Prat, M. Guo, J.G. Herman, X. Matias-Guiu, et al.,Abnormalities of the APC/beta-catenin pathway in endometrial cancer,Oncogene 21 (2002) 7981–7990.

52] P.W. Schlosshauer, L.H. Ellenson, R.A. Soslow, Beta-catenin and E-cadherin expression patterns in high-grade endometrial carcinomaare associated with histological subtype, Mod. Pathol. 15 (2002)
1032–1037.
53] J. Palacios, L. Catasus, G. Moreno-Bueno, X. Matias-Guiu, J. Prat, C.Gamallo, Beta- and gamma-catenin expression in endometrial carci-noma. Relationship with clinicopathological features and microsatelliteinstability, Virchows Arch. 438 (2001) 464–469.

[

[


54] M. Saegusa, M. Hashimura, T. Yoshida, I. Okayasu, beta- Cateninmutations and aberrant nuclear expression during endometrial tumori-genesis, Br J Cancer 84 (2001) 209–217.

55] M. Saegusa, I. Okayasu, Frequent nuclear beta-catenin accumulationand associated mutations in endometrioid-type endometrial and ovariancarcinomas with squamous differentiation, J. Pathol. 194 (2001) 59–67.

56] Y. Yin, G. Solomon, C. Deng, J.C. Barrett, Differential regulation of p21by p53 and Rb in cellular response to oxidative stress, Mol. Carcinog.24 (1999) 15–24.

57] U.M. Moll, E. Chalas, M. Auguste, D. Meaney, J. Chumas, Uterinepapillary serous carcinoma evolves via a p53-driven pathway, Hum.Pathol. 27 (1996) 1295–1300.

58] W. Zheng, P. Cao, M. Zheng, E.E. Kramer, T.A. Godwin, p53 overex-pression and bcl-2 persistence in endometrial carcinoma: comparisonof papillary serous and endometrioid subtypes, Gynecol. Oncol. 61(1996) 167–174.

59] G. Lecce, G. Meduri, M. Ancelin, C. Bergeron, M. Perrot-Applanat,Presence of estrogen receptor beta in the human endometrium throughthe cycle: expression in glandular, stromal, and vascular cells, J. Clin.Endocrinol. Metab. 86 (2001) 1379–1386.

60] Z. Weihua, S. Saji, S. Makinen, G. Cheng, E.V. Jensen, M. Warner,J.A. Gustafsson, Estrogen receptor (ER) beta, a modulator of ERal-pha in the uterus, Proc. Natl. Acad. Sci. U.S.A. 97 (2000) 5936–5941.

61] S. Erkanli, F. Eren, S. Pekin, T. Bagis, BCL-2 and P53 expression inendometrial carcinoma, J. Exp. Clin. Cancer Res. 23 (2004) 97–103.

62] S.F. Lax, R.J. Kurman, E.S. Pizer, L. Wu, B.M. Ronnett, A binaryarchitectural grading system for uterine endometrial endometrioid car-cinoma has superior reproducibility compared with FIGO grading andidentifies subsets of advance-stage tumors with favorable and unfavor-able prognosis, Am. J. Surg. Pathol. 24 (2000) 1201–1208.

63] M.K. Oehler, M.C. Rees, R. Bicknell, Steroids and the endometrium,Curr. Med. Chem. 7 (2000) 543–560.

64] A.A. Jazaeri, K.J. Nunes, M.S. Dalton, M. Xu, M.A. Shupnik, L.W.Rice, Well-differentiated endometrial adenocarcinomas and poorly dif-ferentiated mixed mullerian tumors have altered ER and PR isoformexpression, Oncogene 20 (2001) 6965–6969.

65] M. Saegusa, I. Okayasu, Changes in expression of estrogen receptorsalpha and beta in relation to progesterone receptor and pS2 status innormal and malignant endometrium, Jpn. J. Cancer Res. 91 (2000)510–518.

66] G. Horvath, G. Leser, M. Hahlin, M. Henriksson, Exon deletions andvariants of human estrogen receptor mRNA in endometrial hyperplasiaand adenocarcinoma, Int. J. Gynecol. Cancer 10 (2000) 128–136.

67] W. Bryant, A.E. Snowhite, L.W. Rice, M.A. Shupnik, The estrogenreceptor (ER)alpha variant Delta5 exhibits dominant positive activity onER-regulated promoters in endometrial carcinoma cells, Endocrinology146 (2005) 751–759.

68] M.H. Herynk, S.A. Fuqua, Estrogen receptor mutations in human dis-ease, Endocr. Rev. 25 (2004) 869–898.

69] K.W. Cheng, C.K. Cheng, P.C. Leung, Differential role of PR-A and-B isoforms in transcription regulation of human GnRH receptor gene,Mol. Endocrinol. 15 (2001) 2078–2092.

70] R.L. Arnett-Mansfield, A. deFazio, G.V. Wain, R.C. Jaworski, K. Byth,P.A. Mote, C.L. Clarke, Relative expression of progesterone receptorsA and B in endometrioid cancers of the endometrium, Cancer Res. 61(2001) 4576–4582.

71] I. De Vivo, G.S. Huggins, S.E. Hankinson, P.J. Lescault, M. Boezen,G.A. Colditz, D.J. Hunter, A functional polymorphism in the promoterof the progesterone receptor gene associated with endometrial cancerrisk, Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 12263–12268.

72] J.V. Bokhman, Two pathogenetic types of endometrial carcinoma,Gynecol. Oncol. 15 (1983) 10–17.

73] M. Esteller, J. Xercavins, J. Reventos, Advances in the moleculargenetics of endometrial cancer (Review), Oncol. Rep. 6 (1999) 1377–1382.

istry &

[

[

[

[

[

[

[

[

[

[

[

[

carcinoma, Int. J. Gynecol. Pathol. 18 (1999) 138–143.


74] M. Esteller, A. Garcia, J.M. Martinez-Palones, J. Xercavins, J. Reven-tos, Clinicopathologic features and genetic alterations in endometrioidcarcinoma of the uterus with villoglandular differentiation, Am. J. Clin.Pathol. 111 (1999) 336–342.

75] S.F. Lax, Molecular genetic pathways in various types of endometrialcarcinoma: from a phenotypical to a molecular-based classification,Virchows Arch. 444 (2004) 213–223.

76] J. Schneider, M.M. Huh, H.L. Bradlow, J. Fishman, Antiestrogen actionof 2-hydroxyestrone on MCF-7 human breast cancer cells, J. Biol.Chem. 259 (1984) 4840–4845.

77] M. Esteller, A. Garcia, J.M. Martinez-Palones, J. Xercavins, J. Reven-tos, Susceptibility to endometrial cancer: influence of allelism at p53,glutathione S-transferase (GSTM1 and GSTT1) and cytochrome P-450(CYP1A1) loci, Br. J. Cancer 75 (1997) 1385–1388.

78] M. Esteller, A. Garcia, J.M. Martinez-Palones, J. Xercavins, J. Reven-tos, Germ line polymorphisms in cytochrome-P450 1A1 (C4887CYP1A1) and methylenetetrahydrofolate reductase (MTHFR) genesand endometrial cancer susceptibility, Carcinogenesis 18 (1997)2307–2311.

79] G.L. Mutter, J.P. Baak, C.P. Crum, R.M. Richart, A. Ferenczy,
W.C. Faquin, Endometrial precancer diagnosis by histopathology,clonal analysis, and computerized morphometry, J. Pathol. 190 (2000)462–469.
80] H. Sun, T. Enomoto, K.R. Shroyer, K. Ozaki, M. Fujita, Y. Ueda, R.Nakashima, C. Kuragaki, G. Ueda, Y. Murata, Clonal analysis and

[


mutations in the PTEN and the K-ras genes in endometrial hyperplasia,Diagn. Mol. Pathol. 11 (2002) 204–211.

81] M. Esteller, A. Garcia, J.M. Martinez-Palones, J. Xercavins, J. Reven-tos, Detection of clonality and genetic alterations in endometrial pipellebiopsy and its surgical specimen counterpart, Lab. Invest. 76 (1997)109–116.

82] T. Kanaya, S. Kyo, Y. Maida, N. Yatabe, M. Tanaka, M. Nakamura,M. Inoue, Frequent hypermethylation of MLH1 promoter in normalendometrium of patients with endometrial cancers, Oncogene 22 (2003)2352–2360.

83] M. Esteller, L. Catasus, X. Matias-Guiu, G.L. Mutter, J. Prat, S.B.Baylin, J.G. Herman, hMLH1 promoter hypermethylation is an earlyevent in human endometrial tumorigenesis, Am. J. Pathol. 155 (1999)1767–1772.

84] B.D. Duggan, J.C. Felix, L.I. Muderspach, J.L. Tsao, D.K. Shibata,Early mutational activation of the c-Ki-ras oncogene in endometrialcarcinoma, Cancer Res. 54 (1994) 1604–1607.

85] C.D. Rolitsky, K.S. Theil, V.R. McGaughy, L.J. Copeland, T.H. Nie-mann, HER-2/neu amplification and overexpression in endometrial

86] X. Matias-Guiu, L. Catasus, E. Bussaglia, H. Lagarda, A. Garcia, C.Pons, J. Munoz, R. Arguelles, P. Machin, J. Prat, Molecular pathologyof endometrial hyperplasia and carcinoma, Hum. Pathol. 32 (2001)569–577.

novel molecular proﬁles of endometrial cancer—new light

Documents