See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/7794518

Endometrial cancer

Article  in  Cell and Tissue Research · November 2005

DOI: 10.1007/s00441-005-1109-5 · Source: PubMed

CITATIONS READS

84 1,265

4 authors, including:

Andrew Ryan Thomas W Jobling

Monash University (Australia) 22 PUBLICATIONS   1,067 CITATIONS   
73 PUBLICATIONS   610 CITATIONS   
SEE PROFILE
SEE PROFILE

Martin K Oehler
University of Adelaide
191 PUBLICATIONS   3,704 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

LACE Study View project

Robotic surgery in gynaecological oncology View project

All content following this page was uploaded by Andrew Ryan on 01 December 2014.

The user has requested enhancement of the downloaded file.

Cell Tissue Res (2005) 322: 53–61
DOI 10.1007/s00441-005-1109-5

REVIEW

Andrew J. Ryan . Beatrice Susil . Thomas W. Jobling .

Martin K. Oehler

Endometrial cancer

Received: 23 December 2004 / Accepted: 1 March 2005 / Published online: 10 June 2005
# Springer-Verlag 2005

Abstract Endometrial cancer is the most common gynae- The molecular pathogenesis of EC remains poorly un-
cological malignancy in the developed world. The majority derstood. However, as in other malignancies, such as co-
of cases can be divided into two broad categories based on lorectal cancer, the transition from normal endometrium to
clinico-pathological and molecular characteristics; Type I carcinoma is thought to involve a stepwise accumulation of
oestrogen-dependent with endometrioid morphology and alterations in genes favouring cell proliferation, the inhi-
Type II non-oestrogen-dependent with serous papillary or bition of apoptosis and angiogenesis (Enomoto et al. 1991).
clear cell morphology. As has been described for other This article reviews the molecular changes commonly as-
malignancies, such as colorectal carcinoma, the transition sociated with EC and discusses possible progression mod-
from normal endometrium to carcinoma is thought to in- els for the disease.
volve a stepwise accumulation of alterations in cellular
regulatory pathways leading to dysfunctional cell growth.
This article reviews the current knowledge of the molecular Clinico-pathological characteristics of endometrial
changes commonly associated with endometrial cancer and cancer
presents possible progression models.
Approximately 10% of EC cases are considered familial
Keywords Endometrial cancer . Microsatellite instability . with many of these being associated with hereditary non-
Oncogenes . Progression model . Tumour suppressor polyposis colorectal cancer (HNPCC), a dominantly inher-
genes . Human ited syndrome with germ-line abnormalities in one of five
DNA-mismatch repair genes with resultant micro-satellite
instability. Females with HNPCC have a ten-fold increased
Introduction lifetime risk of EC compared with that of the general
population and the lifetime risk of EC (42%) is higher than
Endometrial cancer (EC) is the most common gynaecolo- that for colorectal carcinoma (30%; Dunlop et al. 1997).
gical malignancy and the fourth most common malignancy Hereditary EC is more likely to occur at a younger age and
in women in the developed world after breast, colorectal is characterised by high FIGO stage and grade, cribriform
and lung cancer. The incidence is estimated at 15–20 per growth pattern, mucinous differentiation and necrosis (Parc
100,000 women per year. Despite the curability of EC et al. 2000).
being high, tumours with particular morphological var- Most EC cases (90%) are sporadic. Sporadic EC is mor-
iants, adverse histopathological features and/or advanced phologically heterogeneous with variants from all path-
stage are characterised by aggressive behaviour and poor ways of Mullerian differentiation, the most common having
prognosis. endometrioid, serous or clear cell morphology. However,
the clinico-pathological properties of endometrioid tu-
mours differ from those of serous and clear cell tumours.
A. J. Ryan . B. Susil These differences were first recognised over 20 years ago
Department of Anatomical Pathology,
Monash Medical Centre, by Bokhman (1983) who hypothesised that sporadic EC
Clayton, Victoria, Australia can be divided on clinical and prognostic grounds into two
main subgroups. The first group consists of oestrogen-
T. W. Jobling . M. K. Oehler (*) related tumours that occur in pre- and post-menopausal
Department of Gynaecological Oncology,
Monash Medical Centre, women, are usually low grade with endometrioid mor-
PO Box 72 East Bentleigh, Victoria, 3165, Australia phology, are frequently preceded by endometrial hyper-
e-mail: martin.oehler@phimr.monash.edu.au plasia and have a good prognosis. They are referred to as
54

This classification has also been justified at the molec-

ular level with Type 1 tumours being more commonly asso-
ciated with abnormalities of DNA-mismatch repair genes,
k-ras, PTEN and beta-catenin, and Type 2 tumours with
abnormalities of p53 and HER2/neu (Table 1). However,
these abnormalities are not present in all cases.

Molecular pathogenesis of endometrial cancer

DNA-mismatch repair genes

Abnormalities of DNA-mismatch repair genes were first

detected in tumours from patients with HNPCC and were
shown to result in the progressive accumulation of al-
terations at microsatellite loci. This so-called microsatellite
instability (MSI) is now known to play a role in sporadic
colon cancers and in several non-colonic tumours. The
most common non-colonic tumour in HNPCC is EC. MSI
is found in 17%–25% of sporadic Type 1 EC (Gurin et al.
1999; Salvesen et al. 2000) but is rarely present in Type II
tumours (Tashiro et al. 1997b).
In HNPCC, MSI arises from germ-line and somatic mu-
tations of one of four DNA-mismatch repair genes (most
commonly MSH-2 and MLH-1; Watson et al. 1994). Gene
mutations are generally not found in sporadic Type 1 EC
but, instead, mismatch repair genes are inactivated or si-
Fig. 1 Uterus with an endometrioid carcinoma lenced by a process of gene promoter hypermethylation
particularly involving the MLH-1 gene (Risinger et al. 2003).
Similar hypermethylation has also been demonstrated in
Type I or oestrogen-dependent endometrioid carcinomas other genes associated with Type I tumours (Sasaki et al.
and comprise approximately 80% of sporadic cancers. The 2001). They are not seen in Type II tumours (Risinger et al.
second group are tumours that occur mainly in post- 2003).
menopausal women, are not related to oestrogen and are
usually not preceded by endometrial hyperplasia. They are
commonly high-grade tumours with serous or clear-cell Oncogenes
morphology and have a poor prognosis. These are referred
to as Type II or non-oestrogen-dependent ECs. Many of the K-ras
various other morphological variants can also be placed
into these categories but they occur at much lower rates; The ras genes are a family of proteins that have GTPase
thus, the classical picture for Type I is endometrioid, where- activity and act as switches in the signaling pathways
as Type II has serous papillary or clear cell morphology between cell surface receptors and the nucleus. In doing so,
(Figs. 1, 2a–c). they play a vital role in the control of cell growth and

Fig. 2 a Histology of a grade I

endometrioid carcinoma of the
uterus. b Histology of a grade III
endometrioid carcinoma of the
uterus. c Histology of a serous
papillary carcinoma of the uterus
 55
Table 1 Clinico-pathological Characteristic Type I Type II
characteristics and genetic ab-
normalities in Type 1 and Type 2
endometrial carcinomas Unopposed oestrogen Yes No
Background endometrium Hyperplastic Atrophic
Morphology Endometrioid Serous, clear cell
Genetic abnormalities MSI, PTEN, K-ras, β-catenin p53, HER2/neu

differentiation (Fig. 3a). Mutations of K-ras have been HER2/neu

identified in 19%–46% of EC, with most of these involving
point mutations at codon 12 (Enomoto et al. 1991; Lagarda The HER2/neu gene encodes a 185-kDa transmembrane
et al. 2001). Alterations of K-ras predominantly involve receptor tyrosine kinase that is similar to the epidermal
Type I tumours and have been reported in 26% of these growth factor receptor (EGF-R). HER2/neu functions as a
tumours but in only 2% with serous (Type II) differenti- preferred partner for heterodimerisation with members of
ation. K-ras mutations have also been found to be more the EGF-R family and therefore plays an important role in
common in MSI-positive tumours (Lax et al. 2000). Muta- coordinating the complex ErbB signaling network that is
tions are also detected in endometrial hyperplasia at a sim- responsible for regulating cell growth and differentiation
ilar rate to EC suggesting that mutation in the ras gene is an (Dougall et al. 1994).
early event in tumorigenesis of Type I EC (Sasaki et al. Over-expression of HER2/neu is present in 9%–30% of
1993). all ECs and has been linked to decreased overall survival

Fig. 3 a K-ras pathway. b PTEN pathway. c Beta-catenin pathway. d P53 pathway

56

(Niederacher et al. 1999; Saffari et al. 1995). Recent studies abnormalities of the PTEN gene and germ-line mutations
examining serous carcinomas have shown a wide variation are associated with Cowden’s disease and Bannayan–
of Her2/neu over-expression (18%–80%) but this is likely Zonona syndrome (Li et al. 1997). PTEN mutations have
to be attributable to differing sample numbers and staining also been demonstrated in ECs with rates ranging from
protocols (Prat et al. 1994; Santin et al. 2002). A recent 34% to 83% depending on sample selection (Mutter et al.
study of a large series of 68 serous carcinomas investigated 2000; Risinger et al. 1997). When analysed according to
by standard breast cancer HER2/neu immunohistochemis- histological type, these mutations are found almost ex-
try criteria found that only 12 (18%) of the tumours over- clusively in endometrioid (Type I) tumours (24, 25).
expressed HER2/neu. Of these, only two demonstrated PTEN mutations are seen at higher rates in MSI tumours
gene amplification. However, serous tumours with HER2/ and are also well documented in endometrial hyperplasia
neu over-expression have been shown to have decreased with and without atypia (Levine et al. 1998; Maxwell et al.
overall survival (Slomovitz et al. 2004). 1998). Given the role of endometrial hyperplasia as the
putative precursor of Type I tumours, PTEN mutations are
presumed to play an early role, although probably not the
Other oncogenes determining step, in tumorigenesis (Latta and Chapman
2002). Levine et al. (1998) have documented a case of syn-
The involvement of the aforementioned oncogenes is well chronous PTEN-positive/MSI-negative hyperplasia and P
established in EC. However, a large number of additional TEN-positive/MSI-negative carcinoma and proposed that
proto-oncogenes are under investigation for possible in- PTEN mutation could precede microsatellite instability;
volvement in EC. C-myc, survivin and human telomerase however, this remains controversial.
reverse transcriptase (hTERT) have all been described in
association with EC. C-myc amplification and over-expres-
sion is present in between 3% and 19% of ECs and, despite p53
conflicting reports of links to tumour differentiation, myo-
metrial invasion and lymph node metastases, a recent The TP53 tumour suppressor gene on chromosome 17
report has shown nuclear and cytoplasmic c-myc immuno- codes for a nuclear protein with an important role in pre-
histochemical staining to be an independent prognostic venting the propagation of cells with damaged DNA. After
factor in EC (Geisler et al. 2004; Niederacher et al. 1999; DNA damage, nuclear p53 accumulates and, through
Williams et al. 1999). Over-expression of hTERT is in- p21, causes cell cycle arrest by inhibiting cyclin-D1 phos-
volved in cancer cell pathogenesis by causing activation of phorylation of the Rb gene and by promoting apoptosis
telomerase and subsequent telomere maintenance and po- through proteins including Bax and Apaf-1 (Yin et al.
tential cell immortalisation (Horikawa and Barrett 2003). 1999; Fig. 3d). Abnormalities of TP53 have been well de-
This mechanism has been implicated in EC development scribed in various malignancies and germ-line mutations
(particularly in conjunction with tamoxifen); however, the are associated with Le Fraumeni syndrome. The mutant
regulation of hTERT is known to be partly dependent on p53 protein is non-functional but resists degradation and
gene products previously linked to EC (e.g. c-myc) and accumulates thereby acting as a dominant negative inhib-
further studies are needed to establish its exact role (Wang itor of the wild-type p53. The accumulated mutant protein
et al. 2002). Likewise for the inhibitor of apoptosis protein can be demonstrated immunohistochemically.
survivin, whose over-expression has been associated with a Mutations of the p53 gene are a frequent and character-
higher clinical grade and stage, but whose definitive role in istic finding in Type II serous tumours with positive im-
tumorigenesis is yet to be established (Takai et al. 2004). munohistochemistry reported in 71%–85% of tumours
None of these three oncogenes has been studied in relation (Kounelis et al. 2000; Moll et al. 1996; Zheng et al. 1996)
to specific prevalence in Type I or II tumours. and are considered to be an early event in tumorigenesis.
This is supported by concordant p53 staining in multiple
biopsies from different areas of individual tumours and by
Tumour suppressor genes the finding of positive immunohistochemical results in
endometrial intraepithelial carcinoma (EIC), the putative
PTEN precursor of serous carcinoma (Lecce et al. 2001; Weihua
et al. 2000).
The PTEN gene codes for a phosphatase that helps to Positive staining for p53 is less common in Type I EC
modulate cell signal transduction pathways by acting on and ranges from 16% to 40% (Erkanli et al. 2004; Lax et al.
phospholipid phosphatidylinositol-(3,4,5)-triphosphate (PIP3), 2000). When stratified by histological grade, grade 1 and 2
a second messenger produced after growth factors bind to tumours are positive at much lower rates than high grade
cell surface membrane receptors. Decreased activity of (grade 3) tumours and staining is rarely seen in atypical
PTEN and therefore increased PIP3 lead to increased cell endometrial hyperplasia, the putative precursor of Type I
proliferation and survival (Fig. 3b). The gene has been tumours. This suggests, in contrast to serous tumours, that
localised to chromosome 10 (10q23-24; Steck et al. 1997). p53 occurs as a late molecular event in Type I tumours (Lax
A number of tumours (e.g. glioblastoma multiforme, et al. 2000).
prostate carcinoma) that have been shown to have acquired
 57

Beta-catenin step in oestrogen-dependent tumorigenesis. ERα mRNA

and protein expression are reported to decrease stepwise
Beta-catenin, a submembranous protein encoded by the from normal or grade I to grade 3 tumour lesions. In con-
CTNNB1 gene located at 3p21, has two independent func- trast, ERβ expression does not alter, suggesting a shift to a
tions: (1) as a cell adhesion factor helping to link cyto- decreased ERα/ERβ ratio (Jazaeri et al. 2001; Saegusa and
skeleton actin filaments to transmembrane E-cadherin and Okayasu 2000). The significance of the relative expression
(2) as a downstream transcriptional activator in the Wnt of both ER subtypes in EC remains to be clarified.
signal transduction pathway in which its activity is closely Variant proteins originating from transcriptional splic-
controlled by the APC tumour suppressor gene (Bullions ing errors have been described for ERα and ERβ. An ERα
and Levine 1998). In the latter role, beta-catenin has been exon 5 splice variant (Δ5 ERα) has not been detected in
implicated in a wide variety of malignancies including normal endometrium but is found at significantly increased
those of the endometrium (Fukuchi et al. 1998). levels in endometrial carcinomas compared with endome-
Excessive beta-catenin protein is usually removed by trial hyperplasia (Horvath et al. 2000). Indeed, Δ5 ERα has
ubiquitin-proteasome pathways (so-called ubiquitination) been shown to be able to activate constitutively transcrip-
but the mutated gene product resists degradation and can tion of ER-dependent genes in the absence of hormone
accumulate in the cytoplasm and the cell nucleus with (Bryant et al. 2004; Herynk and Fuqua 2004). This could
constitutive target gene activity (Fig. 3c). Alternately, ac- therefore provide endometrial tumour cells with a growth
cumulation of beta-catenin can be the result of abnormal- advantage, potentially leading to uncontrolled prolifera-
ities of tumour suppressor genes, such as the adenomatosis tion. Of particular interest is the ERβ exon 8 splice variant
polyposis coli (APC) gene. Accumulated beta-catenin can ERβcx (Ogawa et al. 1998), which has a dominant neg-
be demonstrated by immunohistochemistry and several ative effect on ERα function. Recent studies in breast
studies have analysed series of EC for nuclear accumula- cancer indicate that ERβcx influences prognosis in this
tion and for CTNNB1 gene mutations. Nuclear beta-ca- malignancy (Saji et al. 2002). ERβcx is expressed in both
tenin expression ranges from 16% to 38% in unclassified normal and neoplastic endometrium but its role in EC is
EC (Fukuchi et al. 1998; Scholten et al. 2003) and is sig- still unknown (Critchley et al. 2002; Skrzypczak et al.
nificantly more common in tumours of endometrioid mor- 2004).
phology (31%–47%) than of the non-endometrioid form PR exists in two distinct isoforms: PR-A and PR-B. The
(0%–3%; Moreno-Bueno et al. 2002; Schlosshauer et al. major role of PR-A in the endometrium is thought to be
2002). Rates of gene mutations of the CTNNB1 gene (exon to down-regulate oestrogen action by preventing ERα
3) are between 15% and 25% of endometrioid cancers and transactivation. In contrast, PR-B acts as an endometrial
none were observed in non-endometrioid cancers (Machin oestrogen-agonist. PR-A is therefore believed to be es-
et al. 2002). Discordance of gene mutation and nuclear sential for the inhibition of oestrogen-induced endometrial
accumulation rates has been suggested to be attributable to proliferation partly by limiting PR-B effects. The impor-
abnormalities in other Wnt proteins (e.g. APC, gamma- tance of the PR-A/PR-B ratio is indicated by a report on
catenin) but this remains unclear. Nuclear accumulation and aberrant ratios of PR isoforms in endometrial hyperplasia
exon 3 mutations of CTNNB1 have also been demon- and EC. Only one PR isoform is commonly found in
strated in atypical hyperplasia suggesting that beta-catenin endometrial cancers and the expression of a single PR
abnormalities arise relatively early in the development of isoform is associated with a higher clinical grade pointing
Type I EC (Saegusa et al. 2001). to a relationship between the loss of PR isoform expression
and poor prognosis. Disruption of relative PR isoform
expression has also been observed in complex atypical
Steroid receptor genes hyperplasia. Early alterations in the ratio of PR-A/PR-B
may therefore precede and/or be implicated in the devel-
Oestrogen (ER) and progesterone (PR) receptors are mem- opment of EC (Arnett-Mansfield et al. 2001). The impor-
bers of the nuclear receptor superfamily. They are ligand- tance of a balanced PR-A/PR-B ratio is further underlined
dependent transcription factors that, on activation, bind to by a recently described functional polymorphism in the
distinct DNA target sites to modulate the expression of promoter of PR; this polymorphism results in the increased
specific genes. In addition to this direct activation of target transcription of PR-B and an altered PR-A/PR-B ratio and
genes, indirect mechanisms acting through contacts with is associated with an increased risk for EC (De Vivo et al.
DNA-bound transcription factors such as AP-1 or NF-κB 2002).
have been described (Oehler et al. 2000).
Two distinct subtypes of ER (ERα and ERβ) are
known. Although the uterus is considered to be ERα-pre- Angiogenic factors
dominant, increased cell proliferation and the exaggerated
response to oestrogen in ERβ knockout mice suggest that Angiogenesis is a critical factor for the growth and spread
ERβ plays a role in modulating ERα function and that it of malignant tumours. Various studies have shown that
consequently has an anti-proliferative function (Lecce et al. high intratumour microvessel density in EC is associated
2001; Weihua et al. 2000). An imbalance in ERα and ERβ with advanced clinical stage and increased risk of recurrent
expression is therefore believed to be a possible critical disease and therefore with poor prognosis (Kirschner et al.
58

1996). Furthermore, a progressive increase in microvessel VEGF-A is highly expressed in EC, it is only rarely
density from benign endometrium through atypical com- identified in benign endometrium or atypical complex hy-
plex hyperplasia to invasive disease has been reported perplasia of postmenopausal women (Holland et al. 2003).
(Abulafia et al. 1995). VEGF-A has been reported to be an important indicator of
Angiogenesis is a highly regulated process resulting poor prognosis in patients with EC. Detection of VEGF/
from the increased production of stimulating factors and a KDR complexes is associated with an even worse outcome
concomitant decrease in inhibitors of angiogenesis. One of (Giatromanolaki et al. 1998). Another member of the VE
the major drivers of tumour angiogenesis is hypoxia, which GF family, VEGF-B, forms homodimers and can hetero-
is a common feature in malignancies in which the growth dimerise with VEGF-A. In contrast to VEGF-A, VEGF-B
of cells outstrips local neovascularisation, thereby creating expression in EC has been found to be significantly lower
areas of inadequate perfusion. The transcription factor than in benign endometrium (Holland et al. 2003). Loss of
hypoxia-inducible factor-1 (HIF-1) is an important reg- VEGF-B expression might contribute to endometrial tumor-
ulatory protein of cellular response to hypoxic stimuli. igenesis, although this remains to be clarified. Other studies
HIF-1 is a heterodimer composed of HIF-1α and HIF-1β/ have shown that the presence of VEGF-C or VEGF-D and
ARNT subunits and binds to hypoxia response elements in its receptor VEGFR-3 may predict myometrial invasion
the promoters of various hypoxia-regulated genes includ- and lymph node metastasis in EC and may prospectively
ing those of angiogenic factors. A recent study examining identify patients who are at increased risk for poor outcome
stage I endometrial carcinomas has found HIF-1α to be (Hirai et al. 2001; Yokoyama et al. 2003).
commonly up-regulated in these malignancies. HIF-1α Adrenomedullin (ADM) is another angiogenic factor
expression is related to an unfavourable prognosis, despite that has been implicated in EC pathogenesis. ADM, a 52-
its correlation with low tumour grade. HIF might there- amino-acid peptide belonging to the calcitonin gene-related
fore have potential as a new marker for the detection of peptide family, has been shown to be up-regulated in en-
heterogeneous behaviour patterns in groups of patients dometrium of women receiving tamoxifen therapy. Tam-
classified by conventional criteria (Sivridis et al. 2002). oxifen, the long-term endocrine treatment for selected
HIF-1 is the key regulator of the angiogenic factor VE patients with breast cancer, is known to induce proliferative
GF. The vascular endothelial growth factor family com- changes of the endometrium increasing the risk of the
prises four members: VEGF-A, VEGF-B, VEGF-C and development of EC. Xenograft experiments with endome-
VEGF-D. VEGF-A is a dimeric glycoprotein existing as trial cancer cells have shown that ADM is pro-tumorigen-
four main isoforms originating from alternative splicing ic, inducing angiogenesis and stimulating carcinoma cell
from a single gene. The VEGF-A isoforms bind to the ty- growth directly (Oehler et al. 2002). These properties pro-
rosine kinase receptors VEGFR-1 (flt-1) and VEGFR-2 vide support for the involvement of ADMs in the patho-
(KDR/flk-1). Increased VEGF-A expression is found in genesis of EC. Whether ADM expression in endometrial
many tumours including EC (Doldi et al. 1996). Whereas

Fig. 4 Endometrial cancer pro-

gression models
 59

malignancies has an impact on the clinical course of the et al. (1997a) has clarified this further by demonstrating
disease is unknown. higher rates of loss of heterozygosity in serous carcinoma
(100%) compared with EIC (43%) and suggesting that loss
of the wild-type p53 allele can result in EIC, whereas
Endometrial cancer progression models serous carcinoma develops after the loss of the second
allele. The timing of the appearance of HER2/neu muta-
Much is known about these common molecular changes tions in Type II EC pathogenesis is unknown.
that we have described but a challenge still lies in the
development of a progression model for EC to match the
adenoma-carcinoma model of colorectal carcinoma de- Future perspective
veloped by Vogelstein et al. (1988). Two pathways of tu-
morigenesis provide a basic outline for such a model, the With targeted research involving the latest technology, e.g.
details of which are shown in Fig. 4. proteomics, further molecular pathways with relevance for
EC tumourigenesis are going to be discovered. The ac-
quired knowledge of the molecular pathology of EC has the
Type I (endometrioid) potential for the identification and development of:
a) Diagnostics–with the identification of pre-morpholog-
Hypermethylation plays a significant role in several areas
ical disease (early detection)
of Type I tumorigenesis with some suggestion that it may
b) Prognostics–with improved disease classification and
be an initial event, as it is suspected of being in colorectal
treatment planning
cancer (Breivik and Gaudernack 1999). Demonstration of
c) Therapeutics–with specific targeting of molecular
MLH1 hypermethylation in microsatellite stable atypical
pathways
endometrial hyperplasia supports these claims and suggests
that MLH1 hypermethylation leads to the so-called mutator
phenotype associated with microsatellite instability. As a
References
tumour progresses, mutations are thought to accumulate in
numerous susceptible genes (BAX, IGFRII, Caspase-5, Abulafia O, Triest WE, Sherer DM, Hansen CC, Ghezzi F (1995)
etc.; Catasus et al. 2000). This is supported by the hetero- Angiogenesis in endometrial hyperplasia and stage I endometrial
geneous distribution of these mutations in studied tumours. carcinoma. Obstet Gynecol 86:479–485
Methylation also plays a role in K-ras gene mutations Arnett-Mansfield RL, deFazio A, Wain GV, Jaworski RC, Byth K,
Mote PA, Clarke CL (2001) Relative expression of progesterone
with most mutations thought to result from methylation- receptors A and B in endometrioid cancers of the endometrium.
related transitions. K-ras mutations are frequently present Cancer Res 61:4576–4582
in tumours known to express MSI, suggesting that they Bokhman JV (1983) Two pathogenetic types of endometrial carci-
are often acquired after the development of the mutator noma. Gynecol Oncol 15:10–17
phenotype (Lagarda et al. 2001). The association between Breivik J, Gaudernack G (1999) Genomic instability, DNA methyl-
ation, and natural selection in colorectal carcinogenesis. Semin
PTEN mutations and microsatellite instability is also well Cancer Biol 9:245–254
documented and short coding mononucleotide repeats in Bryant W, Snowhite AE, Rice LW, Shupnik MA (2004) The estrogen
the PTEN gene are thought to be susceptible to mutation in receptor-{alpha} (ER{alpha}) variant {Delta}5 exhibits domi-
the setting of microsatellite instability secondary to the nant positive activity on ER-regulated promoters in endometrial
carcinoma cells. Endocrinology 146:751-759
hypermethylation of MLH1 (Salvesen et al. 2004). A small Bullions LC, Levine AJ (1998) The role of beta-catenin in cell
percentage of PTEN abnormalities are independent of adhesion, signal transduction, and cancer. Curr Opin Oncol
MLH1 hypermethylation and are the result of direct pro- 10:81–87
moter methylation (Salvesen et al. 2004). Catasus L, Matias-Guiu X, Machin P, Zannoni GF, Scambia G,
Benedetti-Panici P, Prat J (2000) Frameshift mutations at coding
Abnormalities of the CTNNB1 gene resulting in accu- mononucleotide repeat microsatellites in endometrial carcinoma
mulation of nuclear beta-catenin are thought to be indepen- with microsatellite instability. Cancer 88:2290–2297
dent of MSI status (Mirabelli-Primdahl et al. 1999). Some Critchley HO, Henderson TA, Kelly RW, Scobie GS, Evans LR,
evidence suggests a role in the transition from benign Groome NP, Saunders PT (2002) Wild-type estrogen receptor
hyperplasia to premalignant disease, as CTNNBI mutations (ERbeta1) and the splice variant (ERbetacx/beta2) are both
expressed within the human endometrium throughout the normal
are only observed in atypical hyperplasia and are seen uni- menstrual cycle. J Clin Endocrinol Metab 87:5265–5273
formly in low grade endometrioid tumours (Machin et al. De Vivo I, Huggins GS, Hankinson SE, Lescault PJ, Boezen M,
2002). Abnormalities of p53 are a late event in Type I Colditz GA, Hunter DJ (2002) A functional polymorphism in the
tumours and are thought to be important in the progression promoter of the progesterone receptor gene associated with
endometrial cancer risk. Proc Natl Acad Sci USA 99:12263–
to high grade tumours (Lax et al. 2000). 12268
Doldi N, Bassan M, Gulisano M, Broccoli V, Boncinelli E, Ferrari A
(1996) Vascular endothelial growth factor messenger ribonucleic
Type II (serous) acid expression in human ovarian and endometrial cancer.
Gynecol Endocrinol 10:375–382
Abnormalities of p53 are well documented in Type II
tumours and its putative precursor EIC. A study by Tashiro
60
Dougall WC, Qian X, Peterson NC, Miller MJ, Samanta A, Greene Levine RL, Cargile CB, Blazes MS, Rees B van, Kurman RJ,
MI (1994) The neu-oncogene: signal transduction pathways, Ellenson LH (1998) PTEN mutations and microsatellite insta-
transformation mechanisms and evolving therapies. Oncogene bility in complex atypical hyperplasia, a precursor lesion to
9:2109–2123 uterine endometrioid carcinoma. Cancer Res 58:3254–3258
Dunlop MG, Farrington SM, Carothers AD, Wyllie AH, Sharp L, Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, Puc J,
Burn J, Liu B, Kinzler KW, Vogelstein B (1997) Cancer risk Miliaresis C, Rodgers L, McCombie R, Bigner SH, Giovanella
associated with germline DNA mismatch repair gene mutations. BC, Ittmann M, Tycko B, Hibshoosh H, Wigler MH, Parsons R
Hum Mol Genet 6:105–110 (1997) PTEN, a putative protein tyrosine phosphatase gene
Enomoto T, Inoue M, Perantoni AO, Buzard GS, Miki H, Tanizawa mutated in human brain, breast, and prostate cancer. Science
O, Rice JM (1991) K-ras activation in premalignant and ma- 275:1943–1947
lignant epithelial lesions of the human uterus. Cancer Res 51: Machin P, Catasus L, Pons C, Munoz J, Matias-Guiu X, Prat J (2002)
5308–5314 CTNNB1 mutations and beta-catenin expression in endometrial
Erkanli S, Eren F, Pekin S, Bagis T (2004) BCL-2 and P53 expression carcinomas. Human Pathol 33:206–212
in endometrial carcinoma. J Exp Clin Cancer Res 23:97–103 Maxwell GL, Risinger JI, Gumbs C, Shaw H, Bentley RC, Barrett JC,
Fukuchi T, Sakamoto M, Tsuda H, Maruyama K, Nozawa S, Berchuck A, Futreal PA (1998) Mutation of the PTEN tumor
Hirohashi S (1998) Beta-catenin mutation in carcinoma of the suppressor gene in endometrial hyperplasias. Cancer Res 58:
uterine endometrium. Cancer Res 58:3526–3528 2500–2503
Geisler JP, Geisler HE, Manahan KJ, Miller GA, Wiemann MC, Zhou Mirabelli-Primdahl L, Gryfe R, Kim H, Millar A, Luceri C, Dale D,
Z, Crabtree W (2004) Nuclear and cytoplasmic c-myc staining in Holowaty E, Bapat B, Gallinger S, Redston M (1999) Beta-
endometrial carcinoma and relationship to survival. Int J Gy- catenin mutations are specific for colorectal carcinomas with
necol Cancer 14:133–137 microsatellite instability but occur in endometrial carcinomas
Giatromanolaki A, Sivridis E, Koukourakis MI, Harris AL, Gatter irrespective of mutator pathway. Cancer Res 59:3346–3351
KC (1998) Bcl-2 and p53 expression in stage I endometrial Moll UM, Chalas E, Auguste M, Meaney D, Chumas J (1996) Uterine
carcinoma. Anticancer Res 18:3689–3693 papillary serous carcinoma evolves via a p53-driven pathway.
Gurin CC, Federici MG, Kang L, Boyd J (1999) Causes and con- Human Pathol 27:1295–1300
sequences of microsatellite instability in endometrial carcinoma. Moreno-Bueno G, Hardisson D, Sanchez C, Sarrio D, Cassia R,
Cancer Res 59:462–466 Garcia-Rostan G, Prat J, Guo M, Herman JG, Matias-Guiu X,
Herynk MH, Fuqua SA (2004) Estrogen receptor mutations in human Esteller M, Palacios J (2002) Abnormalities of the APC/beta-
disease. Endocr Rev 25:869-898 catenin pathway in endometrial cancer. Oncogene 21:7981–
Hirai M, Nakagawara A, Oosaki T, Hayashi Y, Hirono M, Yoshihara 7990
T (2001) Expression of vascular endothelial growth factors Mutter GL, Lin MC, Fitzgerald JT, Kum JB, Baak JP, Lees JA, Weng
(VEGF-A/VEGF-1 and VEGF-C/VEGF-2) in postmenopausal LP, Eng C (2000) Altered PTEN expression as a diagnostic
uterine endometrial carcinoma. Gynecol Oncol 80:181–188 marker for the earliest endometrial precancers. J Natl Cancer Inst
Holland CM, Day K, Evans A, Smith SK (2003) Expression of the 92:924–930
VEGF and angiopoietin genes in endometrial atypical hyper- Niederacher D, An HX, Cho YJ, Hantschmann P, Bender HG,
plasia and endometrial cancer. Br J Cancer 89:891–898 Beckmann MW (1999) Mutations and amplification of onco-
Horikawa I, Barrett JC (2003) Transcriptional regulation of the genes in endometrial cancer. Oncology 56:59–65
telomerase hTERT gene as a target cellular and viral oncogenic Oehler MK, Rees MC, Bicknell R (2000) Steroids and the
mechanisms. Carcinogenesis 24:1167–1176 endometrium. Curr Med Chem 7:543–560
Horvath G, Leser G, Hahlin M, Henriksson M (2000) Exon deletions Oehler MK, Hague S, Rees MC, Bicknell R (2002) Adrenomedullin
and variants of human estrogen receptor mRNA in endometrial promotes formation of xenografted endometrial tumors by
hyperplasia and adenocarcinoma. Int J Gynecol Cancer 10:128– stimulation of autocrine growth and angiogenesis. Oncogene
136 21:2815–2821
Jazaeri AA, Nunes KJ, Dalton MS, Xu M, Shupnik MA, Rice LW Ogawa S, Inoue S, Watanabe T, Orimo A, Hosoi T, Ouchi Y,
(2001) Well-differentiated endometrial adenocarcinomas and Muramatsu M (1998) Molecular cloning and characterization of
poorly differentiated mixed Mullerian tumors have altered ER human estrogen receptor betacx: a potential inhibitor of estrogen
and PR isoform expression. Oncogene 20:6965–6969 action in human. Nucleic Acids Res 26:3505–3512
Kirschner CV, Alanis-Amezcua JM, Martin VG, Luna N, Morgan E, Parc YR, Halling KC, Burgart LJ, McDonnell SK, Schaid DJ,
Yang JJ, Yordan EL (1996) Angiogenesis factor in endometrial Thibodeau SN, Halling AC (2000) Microsatellite instability and
carcinoma: a new prognostic indicator? Am J Obstet Gynecol hMLH1/hMSH2 expression in young endometrial carcinoma
174:1879–188 (discussion 1882–1884) patients: associations with family history and histopathology. Int
Kounelis S, Kapranos N, Kouri E, Coppola D, Papadaki H, Jones J Cancer 86:60–66
MW (2000) Immunohistochemical profile of endometrial ade- Prat J, Oliva E, Lerma E, Vaquero M, Matias-Guiu X (1994) Uterine
nocarcinoma: a study of 61 cases and review of the literature. papillary serous adenocarcinoma. A 10-case study of p53 and c-
Mod Path 13:379–388 erbB-2 expression and DNA content. Cancer 74:1778–1783
Lagarda H, Catasus L, Arguelles R, Matias-Guiu X, Prat J (2001) K- Risinger JI, Hayes AK, Berchuck A, Barrett JC (1997) PTEN/
ras mutations in endometrial carcinomas with microsatellite MMAC1 mutations in endometrial cancers. Cancer Res 57:
instability. J Pathol 193:193–199 4736–4738
Latta E, Chapman WB (2002) PTEN mutations and evolving Risinger JI, Maxwell GL, Berchuck A, Barrett JC (2003) Promoter
concepts in endometrial neoplasia. Curr Opin Obstet Gynecol hypermethylation as an epigenetic component in Type I and
14:59–65 Type II endometrial cancers. Ann N Y Acad Sci 983:208–212
Lax SF, Kendall B, Tashiro H, Slebos RJ, Hedrick L (2000) The Saegusa M, Okayasu I (2000) Changes in expression of estrogen
frequency of p53, K-ras mutations, and microsatellite instability receptors alpha and beta in relation to progesterone receptor and
differs in uterine endometrioid and serous carcinoma: evidence pS2 status in normal and malignant endometrium. Jpn J Cancer
of distinct molecular genetic pathways. Cancer 88:814–824 Res 91:510–518
Lecce G, Meduri G, Ancelin M, Bergeron C, Perrot-Applanat M Saegusa M, Hashimura M, Yoshida T, Okayasu I (2001) Beta-catenin
(2001) Presence of estrogen receptor beta in the human endo- mutations and aberrant nuclear expression during endometrial
metrium through the cycle: expression in glandular, stromal, and tumorigenesis. Br J Cancer 84:209–217
vascular cells. J Clin Endocrinol Metab 86:1379–1386
 61
Saffari B, Jones LA, el-Naggar A, Felix JC, George J, Press MF Steck PA, Pershouse MA, Jasser SA, Yung WK, Lin H, Ligon AH,
(1995) Amplification and overexpression of HER-2/neu (c- Langford LA, Baumgard ML, Hattier T, Davis T, Frye C, Hu R,
erbB2) in endometrial cancers: correlation with overall survival. Swedlund B, Teng DH, Tavtigian SV (1997) Identification of a
Cancer Res 55:5693–5698 candidate tumour suppressor gene, MMAC1, at 10q23.3 that is
Saji S, Omoto Y, Shimizu C, Warner M, Hayashi Y, Horiguchi S, mutated in multiple advanced cancers. Nat Genet 15:356–362
Watanabe T, Hayashi S, Gustafsson JA, Toi M (2002) Expres- Takai N, Ueda T, Nishida M, Nasu K, Miyakawa I (2004) The
sion of estrogen receptor (ER) (beta)cx protein in ER(alpha)- relationship between oncogene expression and clinical outcome
positive breast cancer: specific correlation with progesterone in endometrial carcinoma. Curr Cancer Drug Targets 4:511–520
receptor. Cancer Res 62:4849–4853 Tashiro H, Isacson C, Levine R, Kurman RJ, Cho KR, Hedrick L
Salvesen HB, MacDonald N, Ryan A, Iversen OE, Jacobs IJ, Akslen (1997a) p53 gene mutations are common in uterine serous
LA, Das S (2000) Methylation of hMLH1 in a population-based carcinoma and occur early in their pathogenesis. Am J Pathol
series of endometrial carcinomas. Clin Cancer Res 6:3607–3613 150:177–185
Salvesen HB, Stefansson I, Kretzschmar EI, Gruber P, Macdonald Tashiro H, Lax SF, Gaudin PB, Isacson C, Cho KR, Hedrick L
ND, Ryan A, Jacobs IJ, Akslen LA, Das S (2004) Significance of (1997b) Microsatellite instability is uncommon in uterine serous
PTEN alterations in endometrial carcinoma: a population-based carcinoma. Am J Pathol 150:75–79
study of mutations, promoter methylation and PTEN protein Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC,
expression. Int J Oncol 25:1615–1623 Leppert M, Nakamura Y, White R, Smits AM, Bos JL (1988)
Santin AD, Bellone S, Gokden M, Palmieri M, Dunn D, Agha J, Genetic alterations during colorectal-tumor development. N
Roman JJ, Hutchins L, Pecorelli S, O’Brien T, Cannon MJ, Engl J Med 319:525–532
Parham GP (2002) Overexpression of HER-2/neu in uterine Wang Z, Kyo S, Maida Y, Takakura M, Tanaka M, Yatabe N, Kanaya
serous papillary cancer. Clin Cancer Res 8:1271–1279 T, Nakamura M, Koike K, Hisamoto K, Ohmichi M, Inoue M
Sasaki H, Nishii H, Takahashi H, Tada A, Furusato M, Terashima Y, (2002) Tamoxifen regulates human telomerase reverse tran-
Siegal GP, Parker SL, Kohler MF, Berchuck A, et al (1993) scriptase (hTERT) gene expression differently in breast and
Mutation of the Ki-ras protooncogene in human endometrial endometrial cancer cells. Oncogene 21:3517–3524
hyperplasia and carcinoma. Cancer Res 53:1906–1910 Watson P, Vasen HF, Mecklin JP, Jarvinen H, Lynch HT (1994) The
Sasaki M, Kotcherguina L, Dharia A, Fujimoto S, Dahiya R (2001) risk of endometrial cancer in hereditary nonpolyposis colorectal
Cytosine-phosphoguanine methylation of estrogen receptors in cancer. Am J Med 96:516–520
endometrial cancer. Cancer Res 61:3262–3266 Weihua Z, Saji S, Makinen S, Cheng G, Jensen EV, Warner M,
Schlosshauer PW, Ellenson LH, Soslow RA (2002) Beta-catenin and Gustafsson JA (2000) Estrogen receptor (ER) beta, a modulator
E-cadherin expression patterns in high-grade endometrial carci- of ERalpha in the uterus. Proc Natl Acad Sci USA 97:5936–
noma are associated with histological subtype. Mod Path 5941
15:1032–1037 Williams JA Jr, Wang ZR, Parrish RS, Hazlett LJ, Smith ST, Young
Scholten AN, Creutzberg CL, Broek LJ van den, Noordijk EM, Smit SR (1999) Fluorescence in situ hybridization analysis of HER-2/
VT (2003) Nuclear beta-catenin is a molecular feature of type I neu, c-myc, and p53 in endometrial cancer. Exp Mol Pathol
endometrial carcinoma. J Pathol 201:460–465 67:135–143
Sivridis E, Giatromanolaki A, Gatter KC, Harris AL, Koukourakis Yin Y, Solomon G, Deng C, Barrett JC (1999) Differential regulation
MI (2002) Association of hypoxia-inducible factors 1alpha and of p21 by p53 and Rb in cellular response to oxidative stress.
2alpha with activated angiogenic pathways and prognosis in Mol Carcinog 24:15–24
patients with endometrial carcinoma. Cancer 95:1055–1063 Yokoyama Y, Charnock-Jones DS, Licence D, Yanaihara A, Hastings
Skrzypczak M, Bieche I, Szymczak S, Tozlu S, Lewandowski S, JM, Holland CM, Emoto M, Sakamoto A, Sakamoto T,
Girault I, Radwanska K, Szczylik C, Jakowicki JA, Lidereau R, Maruyama H, Sato S, Mizunuma H, Smith SK (2003) Expres-
Kaczmarek L (2004) Evaluation of mRNA expression of sion of vascular endothelial growth factor (VEGF)-D and its
estrogen receptor beta and its isoforms in human normal and receptor, VEGF receptor 3, as a prognostic factor in endometrial
neoplastic endometrium. Int J Cancer 110:783–787 carcinoma. Clin Cancer Res 9:1361–1369
Slomovitz BM, Broaddus RR, Burke TW, Sneige N, Soliman PT, Wu Zheng W, Cao P, Zheng M, Kramer EE, Godwin TA (1996) p53
W, Sun CC, Munsell MF, Gershenson DM, Lu KH (2004) Her-2/ overexpression and bcl-2 persistence in endometrial carcinoma:
neu overexpression and amplification in uterine papillary serous comparison of papillary serous and endometrioid subtypes.
carcinoma. J Clin Oncol 22:3126–3132 Gynecol Oncol 61:167–174

View publication stats