Seminars in Diagnostic Pathology (2008) 25, 274-287
Molecular targets and biological modifiers
in gastric cancer
Fátima Carneiro, MD, PhD,a,b,c Carla Oliveira, PhD,a,b Marina Leite, PhD,a
Raquel Seruca, MD, PhDa,b
From the aInstitute of Molecular Pathology and Immunology of the University of Porto (IPATIMUP), Porto, Portugal;
b
Medical Faculty of the University of Porto, Porto, Portugal; and the
c
Department of Pathology, Hospital S. João, Porto, Portugal.
KEYWORDS
Gastric cancer;
Molecular targets;
Mutation;
E-cadherin (CDH1);
Microsatellite
instability
The overall survival of gastric cancer patients remains poor despite efforts and advances in its
prevention, diagnosis, and treatment. The development of new therapies is crucial for the effective
control of this disease. An increasing number of genetic and epigenetic alterations have been associated
with distinct histological types of gastric cancer. In this review, we will discuss the involvement of
E-cadherin, EGFR, ERBB2, MMR genes, KRAS, and PIK3CA in the development and progression of
gastric cancer and their role as biomarkers or as novel putative targets for therapy.
© 2008 Elsevier Inc. All rights reserved.
Gastric cancer (GC) is one of the most common forms of
cancer in Europe.1 In 2000, there were 192,000 new cases
with 158,000 deaths. In Southern Europe, Portugal shows
the highest incidence of GC (as estimated by GLOBOCAN
2002).2 Despite advances in prevention and diagnosis, the
overall outcome of the patients has remained poor. The
overall 5-year survival is around 23%. This is largely due to
the advanced stage of disease at presentation, so that most
patients only receive palliative treatment. Moreover, in the
various multimodal therapy regimens that are used to improve the patients’ prognosis, the majority of patients do not
respond to treatment. In summary, the prediction of therapy
response in GC is very limited, and the high prevalence of
incurable disease produces a large burden on patients, which
has a huge effect on health care resources. In this review, we
will focus on putative alternative therapeutic strategies
based on the identification of molecular targets and biological modifiers in GC.
Address reprint requests and correspondence: Fátima Carneiro,
MD, PhD, Rua Dr Roberto Frias s/n, 4200-465 Porto, Portugal.
E-mail address: fcarneiro@ipatimup.pt.
0740-2570/$ -see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1053/j.semdp.2008.07.004
Overview of GC
Carcinomas of the stomach are very heterogeneous from the
morphologic standpoint. This heterogeneity is amply reflected in the diversity of histopathological classifications
on record, which are based on different approaches: histologic profile, degree of differentiation, pattern of growth,
and histogenesis.3
Laurén’s classification4 individualizes two main types of
gastric carcinoma—intestinal and diffuse—which display
different clinicopathologic profiles and often occur in distinct epidemiologic settings. Intestinal carcinoma is more
prevalent in elderly persons of the masculine gender,
whereas diffuse carcinoma tends to occur in younger individuals, mainly females, and frequently depicts hereditary
conditioning. The incidence of intestinal carcinomas is
steadily decreasing in most countries, in contrast to diffuse
carcinomas, whose incidence is quite stable or even increasing.5 The classification proposed by Carneiro and coworkers6 highlights the heterogeneity of gastric carcinoma, individualizing mixed and solid carcinomas. By univariate
analysis, the survival of the patients with mixed carcinomas
Carneiro et al
Gastric Cancer
was shown to be significantly worse than that of patients
with pure histologic type of tumors. In a multivariate analysis using the Cox’s model, Carneiro’s classification kept its
independent prognostic significance, emerging as the second most important factor after TNM staging and before
venous invasion.6
About 90% of GC cases appear in a sporadic setting,
whereas familial clustering is observed in the remaining
10%.7 Of these, only 1% to 3% is hereditary.8
Among the cases with familial aggregation of GC,
several situations can be identified: cases, in which the
histopathology of the tumors is unknown, simply designated as familial GC (FGC) and cases in which it is
possible to have information on the histopathological
type of one or more GCs. The latter group encompasses
specific syndromes/diseases as follows: hereditary diffuse GC (HDGC), familial diffuse GC (FDGC), and
familial intestinal GC (FIGC).9
Histopathological features of HDGC
As fully described below, truncating germline E-cadherin
gene (CDH1) mutations were identified in 1998 as a causal
genetic defect for HDGC by Guilford and coworkers.10
Shortly afterward, the International Gastric Cancer Linkage
Consortium (IGCLC) defined the following criteria for
identification of HDGC families: (1) two or more documented cases of diffuse GC in first/second-degree relatives,
with at least one diagnosed before the age of 50; or (2) three
or more cases of documented diffuse GC in first/seconddegree relatives, independent of age.11 Furthermore, the
IGCLC recommended that carriers of CDH1 truncating
germline mutations should be offered the possibility of
275
being submitted to intensive screening and/or prophylactic
gastrectomy.
Current knowledge on the morphologic steps underlying
the development of HDGC stems from detailed studies
performed in stomachs that were totally mapped, encompassing prophylactic gastrectomy specimens, and total gastrectomies performed in patients referred from chromoendoscopic surveillance programs.12-18 In most specimens, at
least one focus of early invasive diffuse GC was identified.
In North American families, early invasive carcinoma was
not restricted to any topographic region in the stomach: foci
were identified from cardia to prepyloric region, without
evidence of antral clustering.12,13 In New Zealand Maori
families, a predilection was observed for the occurrence of
early invasive carcinomas for the distal stomach and the
body-antral transitional zone.15,16 Reasons for the different
anatomical localization of the cancer foci in the aforementioned studies remain to be clarified.
As precursors of the invasive cancers, two distinct types
of lesions were identified in prophylactic gastrectomies: (1)
in situ signet ring cell carcinoma, corresponding to the
presence of signet ring cells within basal membrane, generally with hyperchromatic and depolarized nuclei (Figure 1);
and (2) pagetoid spread of signet ring cells below the
preserved epithelium of glands/foveolae (Figure 1).14
Model of development of HDGC
On the basis of the findings in prophylactic gastrectomies, a model for the development of diffuse GC in
E-cadherin mutation carriers was proposed,14 encompassing the following lesions: mild nonatrophic gastritis,
in situ signet ring cell carcinoma, pagetoid spread of
Figure 1 Morphological steps in the development of HDGC encompassing in situ carcinoma, pagetoid spread, and early intramucosal
carcinoma (upper panel). Sequencing analysis of a germline mutation in exon 14 of CDH1 gene (lower panel, left). CDH1 promoter
methylation in a tumor from the same family, determined in bisulphite treated DNA by PCR and sequencing (lower panel, right).
276
Seminars in Diagnostic Pathology, Vol 25, No 4, November 2008
signet ring cells, and invasive carcinoma (Figure 1).
E-cadherin immunoexpression was shown to be reduced
or absent in early invasive gastric carcinomas, contrasting with the normal membranous E-cadherin expression
in adjacent nonneoplastic mucosa, in keeping with a
clonal origin of the cancer foci.13,14
It is accepted that the gastric mucosa in CDH1 germline mutation carriers is quite normal until the second
CDH1 allele is inactivated. It is postulated that this
downregulation occurs in multiple cells in the gastric
mucosa, accounting for the multifocal tumor lesions. The
role played by environmental factors (diet, Helicobacter
pylori infection, carcinogens), ulceration, and gastritis as
triggers of this downregulation event remains to be elucidated.
E-cadherin and diffuse GC
E-cadherin is a transmembrane glycoprotein critical for
establishing and maintaining polarized and differentiated
epithelia during development and in adult tissues. Adherens
junctions (AJs) cluster, via homophilic interactions, through
the extracellular domains of calcium-dependent E-cadherin
molecules on the surface of homotypic neighbor cells.19
Experimental evidence supports a role for the E-cadherin
complex, both in suppressing invasion of epithelial cells and
metastasis formation.20
Somatic mutations in CDH1 gene have been identified in
40% to 83% of sporadic diffuse-type gastric cancers but not
in sporadic intestinal-type gastric cancers.21
Germline mutations in CDH1 gene are the underlying
genetic defect responsible for HDGC (OMIM no. 137215),
an autosomal dominant syndrome characterized by high
susceptibility to early-onset diffuse gastric carcinomas.10
Since then, many studies reported different types of CDH1
mutations in HDGC families (Figure 1), and it was calculated that approximately 30% to 40% of the HDGC families
harbor CDH1 germline mutations.9,22
Heterozygous carriers of CDH1 germline mutations frequently remain asymptomatic at least until the second decade of life, when inactivation of the remaining wild type
allele of the CDH1 gene occurs.23,24 This two-hit inactivation mechanism is believed to determine the initiation of
diffuse GC, consistent with the two-hit inactivation model
proposed by Knudson.25 CDH1 germline mutations found
in HDGC families are spread along the whole sequence of
the gene, without preferential hot-spots. About 80% of these
mutations are nonsense, splice-site, and frameshift and are
predicted to produce premature termination codons. The
remaining 20% are missense mutations.9,22,26 Although little doubts exist about the inactivating function of truncating
mutations, the pathogenic role of missense mutations remains under debate.
CDH1 germline missense mutations
In contrast to CDH1 germline truncating mutations, for
which 80% disease-penetrance is estimated,27 missense mutations display a low-penetrance phenotype, with only few
mutation carriers affected within pedigrees. This, together
with the fact that these HDGC families are usually very
small so that very few individuals are available for testing,
has not allowed segregation analysis within E-cadherin
germline missense mutations families. Lack of this clinical
information is a limiting step to infer the pathogenic significance of missense mutations. In this regard, it was suggested by Fitzgerald and Caldas28 that the significance of
CDH1 missense mutations should be assessed in at least
four affected members in combination with functional and
transcript analysis (to look for activation of cryptic splice
sites). However, in some cases, this type of analysis is
impossible to accomplish. To circumvent this limitation, our
group developed an in vitro functional assay to infer the
deleterious nature of CDH1 germline missense variants.29
In this experimental model, cells negative for E-cadherin
expression are selected as model and engineered to stably
express either the wild-type E-cadherin protein or its mutants. Slow aggregation on soft agar and cell invasion into
matrigel or collagen matrix are used to assess the functional
effect of CDH1 germline missense variants. Most missense
mutations studied to date have a clear functional impact on
E-cadherin function, supporting their pathogenic nature.30
Aside from the critical importance of these studies for
the genetic counseling of HDGC families, the distinct stable
cell lines harboring the set of missense mutations now
available allow us to identify associated signaling pathways
and putative molecular targets for therapy.
E-cadherin second-hit inactivation
mechanisms in GC
In sporadic diffuse gastric carcinomas, somatic CDH1 mutations are identified in a high proportion of cases, but
tumors do not commonly show loss of heterozigosity (LOH)
at the CDH1 locus.31 Later, it was shown that CDH1 promoter methylation is the second-hit inactivation mechanism
in more than half of the sporadic diffuse gastric carcinoma
cases harboring CDH1 mutations.32 We analyzed a series of
23 sporadic gastric carcinomas for the presence of CDH1
mutations, CDH1 promoter methylation, and LOH. CDH1
mutations were detected in 56.3% of diffuse gastric carcinomas and in none of the seven intestinal gastric carcinomas. CDH1 promoter methylation was observed in 66.7%
of the cases harboring CDH1 mutations. In diffuse gastric
carcinomas harboring CDH1 mutations, LOH was observed
in a single case. These results show that CDH1 promoter
methylation is the second-hit inactivation mechanism in
more than half of the sporadic diffuse gastric carcinoma
cases harboring CDH1 mutations.
Carneiro et al
Gastric Cancer
In the setting of HDGC, few tumors were analyzed for
second-hit inactivation mechanisms. Somatic structural alterations of CDH1 are uncommon mechanisms of CDH1
inactivation, and promoter hypermethylation (Figure 1) was
shown to be the most common mechanism.23,24,33 Recently,
we extended the characterization of the second hit of CDH1
inactivation to 14 HDGC families carrying different CDH1
germline mutations (unpublished data). The combined analysis of the results of this study and those from previous
studies23,24,33 showed that CDH1 promoter hypermethylation was found as a sole event in approximately 40% of the
HDGC tumors analyzed, and LOH was observed to be a
rather more frequent event than was initially postulated.
Moreover, LOH was found concomitantly with promoter
hypermethylation in one-fifth of the cases and, importantly,
metastatic lesions displayed preferentially LOH.
The thorough characterization of the second-hit mechanisms of CDH1 inactivation in HDGC tumors is critical to
distinguish those that are irreversible (LOH, somatic mutations) from those that are reversible, such as promoter
hypermethylation, the latter putatively targeted for therapy
with “epigenetic” drugs (demethylating agents). This information will be pivotal for the therapeutic management of
HDGC patients.
E-cadherin and non-sense-mediated
decay (NMD)
NMD is a conserved mRNA surveillance mechanism that
typically degrades mRNAs harboring premature termination
codons (PTCs), using the exon junction complex (EJC), a
set of proteins recruited to mRNAs near exon– exon junctions after RNA splicing.34 Evidence suggests that NMD
modulates the phenotype of numerous diseases.35,36
Because in HDGC the vast majority CDH1 germline
mutations (80%) are truncating and generate PTCs,22,26 we
decided to examine the impact of NMD in the phenotype of
HDGC patients caused by such mutations.37 To test the role
of NMD in the downregulation of truncating CDH1 transcripts, we performed allele-specific expression (ASE) analysis on normal gastric mucosa from asymptomatic individuals carrying five different truncating CDH1 mutations.
Four of these mutations generate a PTC that has at least one
exon– exon junction downstream (in the “NMD-competent”
region), and the fifth mutation generates a PTC in the last
exon (in the “NMD-incompetent” region). We found that
the mRNAs harboring PTCs located at the predicted NMD
competent region are, in fact, expressed at a lower abundance compared with the wild type, although presenting
different degradation rates. In contrast, the mRNA carrying
the mutant PTC at the predicted NMD-incompetent region
was not degraded but, on the contrary, was four- to fivefold
more abundant than the wild type. These results suggest that
NMD may promote disease in carriers of CDH1 mutations
that generate PTCs in the “NMD-competent” region. To
277
assess this hypothesis, we analyzed medical data from a
large set of published HDGC families (a total of 264 patients) carrying truncating germline mutations22,26 in both
NMD-competent and -incompetent regions of the CDH1
gene. This analysis revealed that there is a significant association (P ⫽ 0.03) between the ability to induce NMD and
the age of onset of GC. Patients carrying PTCs in the
NMD-competent region developed cancer earlier than patients carrying PTCs in the NMD-incompetent region. This
notion is supported by a previous study of 4 families carrying the 2398delC mutation (located in the NMD-incompetent region) with a cumulative risk of acquiring GC of
only 40% for males and 63% for females, which was
considerably lower than for HDGC patients in general: 67%
for males and 83% for females.26
The association of predicted NMD activity with earlier
age of onset of GC, as well as the observation of lower
cumulative risk in patients harboring a CDH1 mutation that
escapes NMD, suggests that NMD may have a detrimental
effect on the clinical progression of HDGC. The simplest
explanation for this is that NMD decreases the level of
transcripts encoding mutated but still functional E-cadherin
proteins that retain tumor suppressor activity. In agreement
with this, it has previously been shown that truncated Ecadherin molecules lacking the COOH-terminal region retain partial function.38
In the future, it will be important to determine whether
truncated E-cadherin proteins generated by PTCs found in
CDH1 mutation carriers retain partial tumor suppressor
function. This information will be critical to determine the
value of inhibiting NMD in HDGC patients as a means to
treat or reduce the risk of diffuse gastric carcinoma.
EGFR signaling activation by loss of
E-cadherin function
For a long time, the functional consequences of E-cadherin
loss of function have been seen as a structural cell– cell
adhesion disruption rather than as a loss of E-cadherindependent regulatory events. E-cadherin acts indeed as a
cell membrane receptor,20,39 and many signaling molecules
have been reported to interact with E-cadherin, namely the
receptor tyrosine kinases (RTKs), which have been found to
localize at the basolateral membrane of epithelial cells.40,41
RTK activity in resting, normal cells is tightly controlled,
but RTKs can become potent oncoproteins. Of notice, the
epidermal growth factor receptor (EGFR) has been reported
to be involved in a bidirectional cross-talk with E-cadherin.42 Using a set of deletion constructs along the Ecadherin gene [E-cad Db-catenin, Ecad (AAA(764) and
E-cad DNT], Qian and coworkers42 suggested that the interaction of E-cadherin with EGFR would require an intact
extracellular E-cadherin domain. Recently, this result was
confirmed for naturally disease occurring CDH1 mutations,
which have been identified in patients with HDGC.43
278
Seminars in Diagnostic Pathology, Vol 25, No 4, November 2008
EGFR/E-cadherin heterodimer stability was disturbed by
the presence of single point mutations in the extracellular
domain (T340A and A634V) but not of intracellular Ecadherin alterations (P799R and V832M). Interestingly, this
reduced stability of the heterodimers correlated with increased EGFR activation on ligand (EGF) stimulation,
which indicates that E-cadherin exerts an inhibitory function on EGFR activity.43 EGFR activation results in the
phosphorylation of specific tyrosine residues within its cytoplasmic tail, which serve as docking sites for various
adaptor proteins and signaling molecules involved in the
regulation of intracellular signaling pathways that promote
cell proliferation and migration, among others.44 In particular, EGFR activation has been reported to be involved in
cell migration by triggering the cytoskeleton reorganization.45 On the other hand, the Rho family of small GTPases,
and in particular RhoA, Rac1, and Cdc42, play a pivotal role
in the reorganization of the actin cytoskeleton, thereby
regulating cell migration.46,47 Our group showed very recently a correlation between increased EGFR activation,
increased RhoA activation, and increased cell migration in
cells expressing extracellular CDH1 missense mutations
identified in HDGC families.43 On pharmacological inhibition of EGFR, E-cadherin mutant-expressing cells showed a
reversion of the motile phenotype, accompanied by a decrease in RhoA activation, supporting the idea that Rho-like
proteins are downstream effectors of EGFR activation.
From these experiments, we suggested that, although Ecadherin normally binds to EGFR blocking its availability
to EGF and consequently its activation, in cells with extracellular mutations, this interaction is disrupted and EGFR
becomes abnormally activated, stimulating a series of pathways that lead to aberrant chaotic movements. In patients
carrying such CDH1 extracellular mutations, this mechanism is likely to contribute to tumor progression.
Of notice, most somatic CDH1 mutations identified in
sporadic diffuse GC are found to cluster between exons 7
and 10 encoding part of the extracellular E-cadherin domain,31 supporting the idea that loss of function of Ecadherin may also then contribute to the frequent liganddependent activation of EGFR in tumors and to EGFR
signaling activation.
E-cadherin and tumor cell survival
Cell survival depends on signals from the environment,
such as those provided by adhesion molecules that mediate
contacts between cells or between cells and the extracellular
matrix.48 If these interactions are disrupted, the cells are
programmed to die.49,50 E-cadherin, aside from its role in
cell-to-cell adhesion, was also shown to be involved in
programmed cell death.51,52
Recently, Ferreira and coworkers53 demonstrated that, in
vitro, loss of functional E-cadherin interferes with cell-tocell adhesion and renders cells more resistant to apoptotic
stimuli. Cell lines stably expressing mutant E-cadherins are
not only capable of invading but are able also to survive and
grow in the absence of contact with other cells. The apoptotic stimulus used in this in vitro study was Paclitaxel
(Taxol®; Brystol–Myers Squibb Company, New York,
NY), a drug widely used in the treatment of advanced
cancers, including epithelial tumors harboring E-cadherin
dysfunction.53 Importantly, these results question the effectiveness of such treatment in these types of tumors and
highlight the need for further research on the subject.
The existence of an interplay between E-cadherin and the
antiapoptotic factor bcl-2 was also demonstrated,53 supporting the hypothesis that E-cadherin is more than a simple
mediator of cell contact and adhesion, but is also involved
in the control of programmed cell death and thus has a dual
role in cancer development.
Receptor tyrosine kinases (EGFR and ERBB2)
activation in GC
RTKs, which are found to be aberrantly activated or overexpressed in a variety of tumors, represent promising targets
for therapeutical intervention. EGFR and ERBB are members of the RTK superfamily of ERBB receptors, which are
glycoproteins that consist of an extracellular domain where
the binding of ligands takes place, a short lipophilic transmembrane domain, and an intracellular domain carrying the
tyrosine kinase activity.54,55
EGFR gene is located on chromosome 7p12 and codes
for a 170-kDa receptor. Aside from the above-mentioned
mechanism dependent of E-cadherin loss, EGFR activation
includes amplified copy number, structural rearrangements
of the receptor, and activating mutations.56 EGFR somatic
mutations cluster in the kinase domain of EGFR (exons
18-21) and cause ligand-independent activation of the receptor. Mutations and amplification occur in about 20% of
the patients with nonsmall cell lung cancer (NSCLC) and
correlate with the clinical response to tyrosine kinase inhibitors.57,58 In GC, EGFR mutations were recently reported by
our group and occur in about 3% of the cases.59 In our
study, mutation and increased copy number of EGFR
showed a significant association with tumor size, and all
cases with alterations in EGFR were spreading into the
gastric wall (T2-T4), suggesting that alterations of this gene
may confer an invasive behavior to neoplastic cells.
The ERBB2 gene maps to 17q12-q21 and encodes
a 185-kDa transmembranar tyrosine kinase receptor
(p185).60,61 In breast carcinomas, ERBB2 functions as an
oncogene62 and is a marker of poor prognosis.63 Apart from
breast cancer, ERBB2 amplification has been identified in
several types of cancer64-66 as well as in gastric carcinomas.67,68 The evaluation of ERBB2 status in gastric carcinoma, using immunohistochemistry (IHC) and fluorescence
in situ hybridization (FISH) analyses showed a frequency of
overexpression ranging from 8.2% to 22.6%, whereas am-
Carneiro et al
Gastric Cancer
plification ranged from 3.8% to 14.8%.69-72 ERBB2 amplification/overexpression has been shown to correlate with
features of biological aggressiveness, such as venous invasion.67 However, the clinical significance of ERBB2 amplification/overexpression has so far not been fully clarified,
because most studies did not have follow-up data or lacked
statistical power for that purpose.67-69,72,73
Two major classes of anti-RTK therapies are available:
ectodomain-binding antibodies and small-molecule TK inhibitors that compete with ATP in the TK domain. Many of
these therapies are either in clinical use or in advanced
clinical trials. As an example, Erlotinib (Roche Pharmaceuticals, Basel, Switzerland) selectively and reversibly inhibits
the TK activity of EGFR, competing with adenosine
triphosphate for binding in the receptor’s TK domain. Erlotinib has been approved for the treatment of chemotherapyresistant advanced NSCLC patients, and in combination
with Gemcitabine (Gemzar®; Eli Lilly and Company, Indianapolis, IN) for the treatment of advanced pancreatic
cancer in patients who have not received previous chemotherapy.74 However, the clinical response to tyrosine kinase
inhibitors in GC treatment is yet unknown.
In breast cancer, ERBB2 amplification is a predictive
marker for targeted therapy with the monoclonal antibody
Trastuzumab (Herceptin®, Genentech, Inc., San Francisco,
California, USA),75 which is currently used as adjuvant
treatment in breast carcinoma patients with ERBB2 gene
amplification/overexpression.76,77
Recently it was shown that the GC cell line N87 presenting ERBB2 amplification is sensitive to Trastuzumab.78
Furthermore, Rebischung and coworkers79 reported a case
of a GC patient with ERBB2 overexpression who responded
to a combination of chemotherapy and Trastuzumab. An
international large-scale clinical study is currently being
performed to compare the response to chemotherapy combined with Trastuzumab versus chemotherapy alone in gastric carcinoma patients with ERBB2 amplification.
Mismatch repair genes and microsatellite
instability (MSI) in GC
Germline mutations in genes responsible for the maintenance of mismatch repair (MMR) are linked to cancer
predisposition in patients from the hereditary non-polyposis
colorectal cancer (HNPCC) syndrome.80 MSI phenotype or
mutator phenotype is caused by defects in the MMR system
responsible for the correction of mismatches that occur
during DNA replication.81 MSI has been reported in many
human sporadic cancers, and in sporadic gastric carcinoma this phenotype is found in about 20% of the cases
(Figure 2).82-93 The recognition of GC as a component of
HNPCC syndrome94 raised the hypothesis of MMR genes
being involved in GC development. To date, no germline
mutations have been described in MMR genes in familial
GC. On the other hand, somatic mutations in MMR genes
279
are rare in sporadic gastric carcinomas. More recently, our
group investigated the role of MMR defects in MSI sporadic
GC.95 Twenty-nine sporadic GCs with high level of MSI
were screened for somatic mutations in MLH1, MSH2,
MSH6, MLH3, and MBD4, and only five truncating mutations (3 in MSH6, 1 in MLH3, and 1 in MBD4) and one
missense mutation (MLH1) were identified.95 All truncating
mutations were found in the coding poly-A tracts, thus
suggesting that they result from the MSI phenotype rather
than causing it.95 In sporadic gastric carcinomas with MSI,
hypermethylation of the promoter region of MLH1 was
demonstrated to be very frequent (75% in the experience of
our group) and associated with decreased/absent protein
expression of the MLH1 protein.88,92,93,96,97
In summary, somatic mutations in MMR genes are rare
in sporadic GCs, and MLH1 hypermethylation (Figure 2)
explains the development of most of cases belonging to this
subtype of gastric carcinomas.
Hypermethylation and GC
It is well known that methylation of regulatory regions of
genes acts as an important alternative to structural alteration
for gene inactivation. Methylation of cytosines within CpG
islands is also observed in physiological conditions as a
common epigenetic event, such as chromosome X inactivation and ageing.98,99 In neoplasia, genome-wide epigenetic
disturbances result in altered DNA methylation patterns.
Genome-wide hypomethylation and selective hypermethylation of DNA sequences is a common event in several
cancers and became recognized as a hallmark of human
cancers.100 CpG island hypermethylation of normally unmethylated promoter regions correlates with loss of transcription.101 Hypermethylation of several gene promoters
has been described in sporadic gastric carcinomas, namely in genes of the repair pathway, such as hMLH1 and
MGMT, cell cycle regulators, such as the cyclin-dependent kinase inhibitorOp16, the mediator of epithelial
cell growth COX2, and cell adhesion molecules, such as
CDH1.32,100,102
Moreover, it was shown that hypermethylation of gene
promoters increases along the pathway that evolves from
chronic gastritis, intestinal metaplasia, and adenomas to
carcinomas of the stomach.103,104 In our group, we searched
for the promoter methylation status of the cancer-related
genes hMLH1, CDH1, MGMT, and COX2 in sporadic
GC.105 Hypermethylation was detected in all genes: about
30% for COX2 and hMLH1, and 50% to 60% for CDH1 and
MGMT gene promoters. The high frequency of hypermethylation in CDH1 and MGMT genes suggests that the inactivation of these tumor-related genes may play a pivotal role
in gastric tumorigenesis. The lower frequency of promoter
hypermethylation detected in COX2 and hMLH1 in comparison to CDH1 and MGMT is likely to be related to the
fact that, in primary GC, COX2 and hMLH1 hypermethyl-
280
Seminars in Diagnostic Pathology, Vol 25, No 4, November 2008
Figure 2 MSI status, hMLH1 promoter methylation and oncogenic mutations in sporadic gastric carcinoma. Examples of MSS and MSI
gastric carcinomas; the evaluation of tumor microsatellite instability was performed using five quasimonomorphic mononucleotide repeats
and pentaplex PCR (upper panel). Examples of hMLH1 promoter methylation by direct sequencing using flanking markers in bisulphite
treated DNA (middle panel). Examples of oncogenic KRAS and PIK3CA mutations in MSI gastric carcinoma (lower panel).
ation is associated only to a specific subset of gastric tumors. In primary gastric carcinomas, aberrant methylation
of COX2 was described in 12% of the cases and linked with
the loss of expression of COX2 mRNA.102,106 Regarding
hMLH1, it is known that hypermethylation of its promoter
region leads to MSI, as previously described in this review.93,97,104,107 The percentage of CDH1 promoter hypermethylation (51%) detected in this study is in accordance
with previous results in sporadic gastric carcinomas.108
CDH1 promoter was found to occur frequently in diffuse
gastric carcinoma (sporadic and hereditary) as second-hit
inactivation mechanism but also in a percentage of cases in
all histological types of GC.32,108,109 MGMT is a repair
protein responsible for the removal of O(6)-alkyl adducts
produced by several carcinogens, including N-nitrosomethylurea. In primary gastric carcinomas, as well as in
many other types of malignancies, epigenetic silencing of
the MGMT gene by promoter hypermethylation has been
demonstrated.110-111 When MGMT is silenced, O(6)-alkyl
adducts are not removed from the guanine nucleotides,
leading the DNA polymerase to misread guanines as adenines. This will promote G-to-A transitions,112 increasing
the mutation rate and leading to chromosomal instability.112
Through the analysis of different tumor tissues, Esteller
and collaborators100 showed that, for each human cancer,
there exists a unique profile of promoter hypermethylation,
in which some gene changes are shared, whereas others are
cancer type-specific. However, in our study, we found no
preferential combination of hypermethylated loci, and no
particular profile was noticed.
Altogether, the results of our group and those from
literature show that hypermethylation of several promoter
regions occurs frequently in GC, reinforcing the idea that
hypermethylation plays an important role in gastric carcinogenesis and inactivation of different gene promoters by
hypermethylation is likely to lead to specific characteristics
of the tumors.
Carneiro et al
Gastric Cancer
MSI and clinicopathologic features of gastric
carcinomas
The clinicopathologic features of gastric carcinoma with
MSI are distinct from those observed in microsatellite stable
(MSS) gastric carcinomas.83,84,86,113 MSI gastric carcinomas tend to occur as large and expanding tumors of the
distal stomach in relatively old patients, frequently displaying an intestinal (glandular) or atypical (solid) histotype; the
frequency of lymph node metastases is low, regardless of
the degree of wall invasion (usually occur with low pTNM
stages); the survival curve of patients with MSI carcinomas
is significantly better than that of patients with MSS carcinomas. The significant association between MSI phenotype
of gastric carcinomas and the good outcome of patients may
be ascribed, partly at least, to the close relationship between
MSI status and staging, namely low pTNM staging of the
tumors. The clinical significance of MSI was recently accessed in a large series of gastric carcinomas (n ⫽ 510).113
MSI was detected in 16% of the cancers and significantly
correlated with long survival (P ⬍ 0.001). In multivariate
analysis, MSI phenotype was identified as an independent
factor (P ⫽ 0.005), adding prognostic information to TNM
stage. The relative risk of death for MSI cancer patients was
0.6 (95% CI 0.4-0.8). Moreover, when grouped according to
stage, stage II cancers showed a significant effect of MSI
status on survival (P ⫽ 0.011; hazard ratio ⫽ 0.3; 95% CI
0.1-0.8).
Summing up, several studies clearly demonstrate that
microsatellite analysis in GC has clinical utility for prognostic evaluation of the patients. It is tempting to speculate
that MSI tumors show a specific set of genes that, whenever
mutated, somehow prevent or delay tumor dissemination
and metastatization.
Target gene mutations in GC
MSI carcinomas accumulate hundreds of thousands of mutations within repetitive sequences in noncoding as well as
in coding repeat sequences due to the high background of
genetic instability that characterizes these tumors. It has
been difficult to distinguish whether these alterations occur
in genes whose inactivation contributes to tumor development (real target genes) or in genes that are inactivated by
chance due to the hypermutability pathway (bystander
genes) in MSI tumors. The real contribution of a large
number of target genes has to rely on mutation frequencies
and the functional role played by these genes, due to the
lack of in vitro or in vivo functional suppressor studies.
We contributed to define the molecular signatures of
target genes in MSI gastric tumors. Initially, we performed
a mutation screening in genes with coding mononucleotide
repeats: TGFRII, IGFII R, BAX, and TCF-4.87,114 Mutations were almost exclusively detected in MSI tumors, thus
confirming that the involvement of such genes is most likely
281
due to a mismatch repair deficiency. The higher frequency
of mutations was verified in TGFRII gene (70%), suggesting that the alterations of the TGFRII gene occur as an
earlier event than those of IGFIIR, BAX, or TCF-4 genes in
gastric carcinogenesis. IGFIIR and BAX mutations were
found in 25% and 30% of the MSI sporadic gastric carcinomas, respectively.87 The frequency of MSI tumors harboring TCF-4 gene mutations was very low (15%) when
compared with the frequency of mutations in the other
target genes and, moreover, never occurred as a single
mutation. These findings suggest that TCF-4 is most likely
to be a “bystander” gene in gastric carcinogenesis rather
than a real target gene, as described in colorectal cancer.114
HPRT and CDH1 repetitive coding regions did not harbor
mutations in MSI tumors (unpublished data). Mutation frequencies at 25 genes containing coding repeats were determined in MSI gastric carcinoma and compared with
colorectal and endometrial MSI tumors.114 From these,
TGFRII, BAX, IGFIIR, hMSH3, hMSH6, GRB-14, and
RAD-50 genes all appeared to be real targets for instability
in both gastric and colon MSI carcinomas. Recently, frameshift EPHB2 mutations were identified in 39% of MSI
gastric carcinomas, showing that EPHB2 is selectively targeted in MSI GCs.115
Clinicopathologic features associated to
target gene mutations in MSI GC
An indirect way to elucidate the putative contribution of
target genes to cancer development is to correlate the mutations in these genes with the clinicopathologic features of
tumors and patients.
Data from our group and others showed that TGFRII
poly(A)10 tract is a real mutation target in MSI gastric
carcinoma, and mutations in the TGFRII gene were significantly associated with the intestinal histotype. The resulting inactivating frameshift mutations are predicted to
truncate the TGF receptor, altering the TGF pathway and
providing a means for tumor cells to escape TGF-mediated
growth restriction.116 TGF conveys signals via two serine/
threonine kinase receptors, type I and type II, both of which
are necessary for signal transduction.117,118 In this complex,
the serine/threonine kinase domain of the type II receptor
phosphorylates serine residues in a serine- and glycine-rich
region adjacent to the amino-terminal boundary of the kinase domain (GS domain) of the type I receptor. This
phosphorylation step activates the kinase domain of the type
I receptor, which can subsequently propagate the signal to
downstream effectors.119,120 Mutations of the type II receptor can interrupt the signal transduction pathway by dominant-negative mechanisms,119,121 leading to growth stimulation rather than growth restriction. The significant
association found between TGFRII mutations and the MSI
phenotype of the tumors somehow reflects the result of
disruption of the TGF pathway, considering that MSI
282
Seminars in Diagnostic Pathology, Vol 25, No 4, November 2008
tumors are larger and more expanding when compared with
MSS tumors. Our finding of a close relationship between
TGFRII mutations and the intestinal histotype might be
supported by the high percentage of TGFRII mutations
found in colon cancer,122 if one believes that the intestinal
type of gastric carcinoma mimics the morphologic appearance of colon cancer, but further studies are needed to
confirm this finding.
IGFIIR poly(G)8 tract was shown to be mutated in about
30% of MSI gastric tumors, and a trend (P ⫽ 0.06) was
found between IGFIIR mutations and lower prevalence of
lymph node metastases in the setting of MSI cases.87
IGFIIR mutations are predicted to generate inactive receptors.
The biological role of IGF II is mediated through both IGFI
and II receptors, with higher affinity to IGFIR.123 IGFIIR
is a multifunctional protein that plays a role in lysosomal enzyme trafficking, endocytosis, and activation of
TGF.124,125 The IGFIIR also inhibits cell proliferation
mediated by the IGFII ligand, itself a potent growth stimulant, by internalizing and degrading this protein.126 Thus,
IGFIIR, by antagonizing the growth stimulatory effect of
IGFII and activating the growth inhibitory effect of TGF,
serves as a growth suppressor gene.123 We observed a trend
toward an association between IGFIIR mutations and low
prevalence of lymph node metastases. Recently, IGFIIR
mutations were described to be inversely associated to serosa invasion.127 Altogether, these results are in accordance
with the role on cell motility and metastatization advanced
for IGFIIR.128 Basic and clinical studies are necessary to
confirm whether or not there is a relationship between
IGFIIR mutations and decreased nodal metastatic ability of
gastric carcinoma.
KRAS, BRAF mutations in GC
Our group has been focusing on determining the frequency
and spectrum of mutations of diverse oncogenes in MSI
carcinomas, specifically on KRAS, BRAF, and PI3KCA
genes (Figure 2). We and others verified that HNPCC,
sporadic MSI gastric, and colorectal carcinoma progress
through RAS/RAF/MAPK pathway, though targeting distinct
kinases and with specific profiles of activating mutations.
Recently, a high frequency of activating mutations in
BRAF, a gene from the RAS regulated kinase-encoding RAF
family, was found in colorectal tumors and associated to
MSI tumors.129-131 Further, data on BRAF showed a hotspot
mutation within exon 15 (V600E), which is the most frequent somatic substitution ever identified in MSI colon
cancers. In addition, these mutations were inversely associated to KRAS mutations, showing that KRAS and BRAF
genes are alternative genetic events in colorectal cancer and
reinforcing the idea that both MSI and MSS colorectal
tumors progress through the same RAS/RAF/MAPK pathway.130 However, it was not clear whether MSI colon and
gastric tumors share the alterations of the RAS/RAF/MAPK
genes found in MSI colon tumors, nor if KRAS and BRAF
genes are also alternative genetic events in GC. We investigated whether KRAS gene was mutated at a similar low
frequency in MSI and MSS gastric tumors. Mutations were
found in 28% of MSI carcinomas and absent in MSS carcinomas.132 The significant association between KRAS and
MSI (P ⫽ 0.0005) is exclusive of GC because, in colon
carcinomas, KRAS mutations occur in MSS and MSI carcinomas. In contrast to KRAS, BRAF mutations do not occur
in MSI GC and are also extremely rare in MSS GC. In fact,
we only detected a BRAFV600E mutation in a MSS carcinoma (1/124, 0.8%).133 Altogether, these results suggest
that KRAS but not BRAF mutations contribute to the tumorigenesis of MSI GC and that KRAS and BRAF mutations are
not alternative genetic events in GC.
PIK3CA and GC
In GC, PIK3CA mutations were identified in 4% to 11% of
the cases.134-136 A higher frequency (25%) was found by
Samuels and coworkers,137 but this study was performed in
a very small series of tumors. In gastric carcinoma, the
MMR status of the tumors is an important parameter to
consider when determining the frequency of PIK3CA mutations. The mutational distribution of PIK3CA in GC is
similar to the one observed for KRAS. As for KRAS, we
were unable to find PIK3CA mutations in MSS gastric
carcinoma.135 Another group reported a single mutation in a
series of 73 MSS gastric carcinomas.136 It is likely that
PIK3CA mutations do not represent an important oncogenic
event in MSS gastric carcinogenesis analogously to what is
observed for KRAS and BRAF mutations in MSS gastric
carcinoma.132,133 In MSI gastric carcinoma, we found
PIK3CA mutations in 19.2% of the cases135 and verified that
PIK3CA and KRAS mutations were mutually exclusive
events.135 These findings are not surprising because PI3K
may function as a downstream effector of the RAS pathway.138,139 It was suggested that genes involved in the same
signaling pathway may manifest mutations in cancer cells in
a mutually exclusive manner, presumably due to the lack of
selective growth advantage in having a second hit in the
already altered pathway.136 In GC, PIK3CA alterations were
found in both early and advanced specimens,134 which
could indicate an important role in the development and
progression of this tumor type.
Genomic amplification of PIK3CA was found to occur in
36.4% of gastric tumors, strongly associated with increased
expression of PIK3CA transcript and elevated levels of
phosphor-AKT.140 Moreover, PIK3CA amplification was
predominantly detected in tumors with no PTEN alterations,
suggesting that loss of PTEN and activation of PIK3CA
genes are mutually exclusive events in gastric tumorigenesis.140
PIK3CA constitutes an attractive molecular marker for
early detection or for the monitoring of tumor progression in
Carneiro et al
Gastric Cancer
diverse types of cancer. More importantly, PIK3CA is a
promising therapeutic target for the development of specific
inhibitors of the p110␣ subunit.
Some inhibitors of PI3K pathway have been extensively used in vitro and in vivo, such as wortmannin and
LY294002, which inhibit the catalytic activity of p110
subunits of PI3K.141-143 These compounds were shown to
inhibit cell proliferation and/or induce apoptosis in cancer
cells via inhibition of the PI3K–AKT pathway. They also
play a role in enhancing the effectiveness of radio- or
chemotherapy.143-147 These inhibitors may also be applied
to mutant forms of PIK3CA, considering that the transforming potential of PIK3CA mutants can still be inhibited by
broadly acting PI3K inhibitors, such as LY294002.148 Despite their efficacy toward inhibition of active PI3K pathway, such inhibitors have the disadvantage of nonselectivity.145 PI3K is ubiquitously present in many mammalian
cells,145 thus the use of wortmannin and LY294002 may be
toxic to the cells that do not display any alteration of the
pathway. The lack of stability of wortmannin and the lack of
solubility of LY294002 are additional things that hampered
further clinical studies of these agents.145
Another possibility of therapy is the use of inhibitors of
the downstream targets of the PI3K pathway. PIK3CA mutations lead to an increase in the phosphorylation of AKT
with its concomitant activation.148-152 The repression of the
activation of AKT represents an attractive possibility of
inhibiting PI3K pathway, although no small AKT inhibitors
have been established yet.143
Further, the inhibition of PI3K–AKT target genes may
represent a fighting chance. mTOR is an example of this
possibility. Some inhibitors of mTOR are under preclinical
and clinical investigations.145 It is the case of rapamycin
and its derivatives (Rad001, CCI-779, and AP23573),153
that were shown to work as potent radiosensitizers of endothelial cells in vitro and led to improved tumor-growth
delay of glioma xenographs in vivo.145 Together, these
results imply that inhibition of the downstream targets may
be a valuable approach to indirectly block the oncogenic
ability of active PI3K–AKT pathway. However, it should be
taken into account that the downstream effectors of PI3K
that are activated on deregulation of the pathway may
depend on the cell type or tumor model, and in the case of
GC, it will be of limited use.
In the future, the development, if possible, of inhibitors
that target specifically the mutant forms of oncogenes that
are activated in MSI GC would be of particular interest.
Concluding remarks and future challenges
In this review, we focused on several genes and downstream
pathways shown to play a role in gastric carcinogenesis,
aiming at analyzing putative targets for therapy and biological modifiers in gastric carcinoma.
From the experience of our group and the data on record,
we would highlight the following key issues:
283
● Identification and characterization of germline CDH1
mutations play a pivotal role in decision making on
prophylactic gastrectomies in the setting of HDGC.
● The characterization of the second-hit mechanisms of
CDH1 inactivation in HDGC tumors will be relevant for
the therapeutic management of HDGC patients.
● EGFR becomes (abnormally) activated in cells with Ecadherin loss of function. Thus, it is tempting to test the
use of EGFR pharmacological inhibitors in patients with
tumors harboring CDH1 mutations, namely whenever
such mutations are localized in the extracellular domain.
● CDH1 mutations lead to an increased resistance to apoptosis mediated by Taxol. This experimental result questions the effectiveness of Taxol in the treatment of advanced gastric patients with tumors harboring CDH1
mutations and calls for the need for further research on
the subject.
● Activation of RTKs such as EGFR and ERBB2 by mutations or amplification is rare in GC. However, they are
good molecular targets for anti-RTKs therapies with
ectodomain-binding antibodies or small-molecule TK inhibitors.
● In GC, MSI is a molecular marker of good prognosis.
● A subset of MSI GC carries mutations in KRAS or
PIK3CA genes. These molecules or their downstream
effectors are promising therapeutic targets.
Acknowledgments
This work is supported in part by grants from The
Portuguese Foundation for Science and Technology (FCT)
(POCTI/SAU-OBS/58111/2004 and POCI/SAU-OBS/
57275/2004) and Sixth Framework Programme from EUFP6 (LSHC-CT-2005-018754).
References
1. Parkin DM, Bray F, Ferlay J, et al: Global cancer statistics, 2002. CA
Cancer J Clin 55:74-108, 2005
2. Ferlay J, Bray F, Pisani P, et al (eds): GLOBOCAN 2002. Cancer
Incidence, Mortality and Prevalence Worldwide. IARC Cancer Base
No. 5, Version 2.0. Lyon, France, IARC Press, 2004
3. Carneiro F: Classification of gastric carcinomas. Curr Diagn Pathol
4:51-59, 1997
4. Laurén P: The two histological main types of gastric carcinoma:
diffuse and so-called intestinal-type carcinoma. An attempt at a
histoclinical classification. Acta Pathol Microbiol Scand 64:31-49,
1965
5. Henson DE, Dittus C, Younes M, et al: Differential trends in the
intestinal and diffuse types of gastric carcinoma in the United States,
1973-2000: increase in the signet ring cell type. Arch Pathol Lab Med
128:765-770, 2004
6. Carneiro F, Seixas M, Sobrinho–Simões M: New elements for an
updated classification of the carcinomas of the stomach. Pathol Res
Pract 191:571-584, 1995
7. Ekstrom AM, Serafini M, Nyren O, et al: Dietary antioxidant intake
and the risk of cardia cancer and noncardia cancer of the intestinal
284
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
Seminars in Diagnostic Pathology, Vol 25, No 4, November 2008
and diffuse types: a population-based case-control study in Sweden.
Int J Cancer 87:133-140, 2000
Palli D, Galli M, Caporaso NE, et al: Family history and risk of
stomach cancer in Italy. Cancer Epidemiol Biomarkers Prev 3:15-18,
1994
Carneiro F, Oliveira C, Suriano G, et al: Molecular pathology of
familial gastric cancer, with an emphasis on hereditary diffuse gastric
cancer (HDGC). J Clin Pathol 61:25-30, 2008
Guilford P, Hopkins J, Harraway J, et al: E-cadherin germline mutations in familial gastric cancer. Nature 392:402-405, 1998
Caldas C, Carneiro F, Lynch HT, et al: Familial gastric cancer:
overview and guidelines for management. J Med Genet 36:873-880,
1999
Chun YS, Lindor NM, Smyrk TC, et al: Germline E-cadherin gene
mutations: is prophylactic total gastrectomy indicated? Cancer 92:
181-187, 2001
Huntsman DG, Carneiro F, Lewis FR, et al: Early gastric cancer in
young, asymptomatic carriers of germ-line E-cadherin mutations.
N Engl J Med 344:1904-1909, 2001
Carneiro F, Huntsman DG, Smyrk TC, et al: Model of the early
development of diffuse gastric cancer in E-cadherin mutation carriers
and its implications for patient screening. J Pathol 203:681-687, 2004
Charlton A, Blair V, Shaw D, et al: Hereditary diffuse gastric cancer:
predominance of multiple foci of signet ring cell carcinoma in distal
stomach and transitional zone. Gut 53:814-820, 2004
Blair V, Martin I, Shaw D, et al: Hereditary diffuse gastric cancer:
diagnosis and management. Clin Gastroenterol Hepatol 4:262-275,
2006
Rogers WM, Dobo E, Norton JA, et al: Risk-reducing total gastrectomy for germline mutations in E-cadherin (CDH1): pathologic findings with clinical implications. Am J Surg Pathol 32:799-809, 2008
Lynch HT, Kaurah P, Wirtzfeld D, et al: Hereditary diffuse gastric
cancer: diagnosis, genetic counseling, and prophylactic total gastrectomy. Cancer 112:2655-2663, 2008
Shore EM, Nelson WJ: Biosynthesis of the cell adhesion molecule
uvomorulin (E-cadherin) in Madin-Darby canine kidney epithelial
cells. J Biol Chem 266:19672-19680, 1991
Mareel M, Leroy A: Clinical, cellular, and molecular aspects of
cancer invasion. Physiol Rev 83:337-376, 2003
Berx G, Becker KF, Höfler H, et al: Mutations of the human Ecadherin (CDH1) gene. Hum Mutat 12:226-237, 1998
Oliveira C, Seruca R, Carneiro F: Genetics, pathology and clinics of
familial gastric cancer. Int J Surg Pathol 14:21-33, 2006
Grady WM, Willis J, Guilford PJ, et al: Methylation of the CDH1
promoter as the second genetic hit in hereditary diffuse gastric
cancer. Nat Genet 26:16-17, 2000
Oliveira C, de Bruin J, Nabais S, et al: Intragenic deletion of CDH1
as the inactivating mechanism of the wild-type allele in an HDGC
tumour. Oncogene 23:2236-2240, 2004
Knudson AG Jr: Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 68:820-823, 1971
Kaurah P, MacMillan A, Boyd N, et al: Founder and recurrent CDH1
mutations in families with hereditary diffuse gastric cancer. J Am
Med Assoc 297:2360-2372, 2007
Lynch HT, Grady W, Suriano G, et al: Gastric cancer: new genetic
developments. J Surg Oncol 90:114-133, 2005
Fitzgerald RC, Caldas C: Clinical implications of E-cadherin associated hereditary diffuse gastric cancer. Gut 53:775-778, 2004
Suriano G, Oliveira C, Ferreira P, et al: Identification of CDH1
germline missense mutations associated with functional inactivation
of the E-cadherin protein in young gastric cancer probands. Hum Mol
Genet 12:575-582, 2003
Suriano G, Seixas S, Rocha J, et al: A model to infer the pathogenic
significance of CDH1 germline missense variants. J Mol Med 84:
1023-1031, 2006
Becker KF, Höfler H: Frequent somatic allelic inactivation of the
E-cadherin gene in gastric carcinomas. J Natl Cancer Inst 87:10821084, 1995
32. Machado JC, Oliveira C, Carvalho R, et al: E-cadherin gene (CDH1)
promoter methylation as the second hit in sporadic diffuse gastric
carcinoma. Oncogene 20:1525-1528, 2001
33. Corso G, Roviello F, Paredes J, et al: Characterization of the P373L
E-cadherin germline missense mutation and implication for clinical
management. Eur J Surg Oncol 33:1061-1067, 2007
34. Chang YF, Imam JS, Wilkinson MF: The nonsense-mediated decay
RNA surveillance pathway. Annu Rev Biochem 76:51-74, 2007
35. Holbrook JA, Neu–Yilik G, Hentze MW, et al: Nonsense-mediated
decay approaches the clinic. Nat Genet 36:801-808, 2004
36. Wilkinson MF: A new function for nonsense-mediated mRNAdecay
factors. Trends Genet 21:143-148, 2005
37. Karam R, Carvalho J, Bruno I, et al: The NMD mRNA surveillance
pathway downregulates aberrant E-cadherin transcripts in gastric
cancer cells and in CDH1 mutation carriers. Oncogene 2008 Apr 21
[Epub ahead of print]
38. Sasaki CY, Lin H, Morin PJ, et al: Truncation of the extracellular
region abrogates cell contact but retains the growth-suppressive activity of E-cadherin. Cancer Res 60:7057-7065, 2000
39. Yap AS, Kovacs EM: Direct cadherin-activated cell signaling: a view
from the plasma membrane. J Cell Biol 160:11-16, 2003
40. Crepaldi T, Pollack AL, Prat M, et al: Targeting of the SF/HGF
receptor to the basolateral domain of polarized epithelial cells. J Cell
Biol 125:313-320, 1994
41. Hoschuetzky H, Aberle H, Kemler R: Beta-catenin mediates the
interaction of the cadherin– catenin complex with epidermal growth
factor receptor. J Cell Biol 127:1375-1380, 1994
42. Qian X, Karpova T, Sheppard AM, et al: E-cadherin-mediated adhesion inhibits ligand-dependent activation of diverse receptor tyrosine
kinases. EMBO J 23:1739-1748, 2004
43. Mateus AR, Seruca R, Machado JC, et al: EGFR regulates RhoAGTP dependent cell motility in E-cadherin mutant cells. Human Mol
Genet 16:1639-1647, 2007
44. Yarden Y, Sliwkowski MX: Untangling the ErbB signaling network.
Nat Rev Mol Cell Biol 2:127-137, 2001
45. den Hartigh JC, van Bergen en Henegouwen PM, Verkleij AJ, et al:
The EGF receptor is an actin-binding protein. J Cell Biol 119:349355, 1992
46. Hall A: Rho GTPases and the actin cytoskeleton. Science 279:509514, 1998
47. Nobes CD, Hall A: Rho GTPases control polarity, protrusion, and
adhesion during cell movement. J Cell Biol 144:1235-1244, 1999
48. Comoglio PM, Boccaccio C, Trusolino L: Interactions between
growth factor receptors and adhesion molecules: breaking the rules.
Curr Opin Cell Biol 15:565-571, 2003
49. Ruoslahti E, Reed JC: Anchorage dependence, integrins, and apoptosis. Cell 77:477-478, 1994
50. Frisch SM, Ruoslahti E: Integrins and anoikis. Curr Opin Cell Biol
9:701-706, 1997
51. Ihara A, Koizumi H, Hashizume R, et al: Expression of epithelial
cadherin and a- and b-catenins in nontumoral livers and hepatocellular carcinomas. Hepatology 23:1441-1447, 1996
52. Hirata K, Ajiki T, Okazaki T, et al: Frequent occurrence of abnormal
E-cadherin/-catenin protein expression in advanced gallbladder cancers and its association with decreased apoptosis. Oncology 71:102110, 2006
53. Ferreira P, Oliveira MJ, Beraldi E, et al: Loss of functional Ecadherin renders cells more resistant to the apoptotic agent taxol in
vitro. Exp Cell Res 310:99-104, 2005
54. Ullrich A, Schlessinger J: Signal transduction by receptors with
tyrosine kinase activity. Cell 61:203-212, 1990
55. Klapper LN, Kirchbaum MH, Sela M, et al: Biochemical and clinical
implications of the ErbB/HER signaling network of growth factor
receptors. Adv Cancer Res 77:25-79, 2000
56. Hynes NE, Lane HA: ERBB receptors and cancer: the complexity of
targeted inhibitors. Nat Rev Cancer 5:341-354, 2005
57. Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the
epidermal growth factor receptor underlying responsiveness of non-
Carneiro et al
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
Gastric Cancer
small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139,
2004
Paez JG, Janne PA, Lee JC, et al: EGFR mutations in lung cancer:
correlation with clinical response to gefitinib therapy. Science 304:
1497-1500, 2004
Moutinho C, Mateus AR, Milanezi F, et al: Epidermal growth factor
receptor structural alterations in gastric cancer. BMC Cancer 8:10,
2008
Akiyama T, Sudo C, Ogawara H, et al: The product of the human
c-erbB-2 gene: a 185-kilodalton glycoprotein with tyrosine kinase
activity. Science 232:1644-1646, 1986
Popescu NC, King CR, Kraus MH: Localization of the human erbB-2
gene on normal and rearranged chromosomes 17 to bands q12-21.32.
Genomics 4:362-366, 1989
Slamon DJ, Godolphin W, Jones LA, et al: Studies of the HER-2/neu
proto-oncogene in human breast and ovarian cancer. Science 244:
707-712, 1989
Dowsett M, Bartlett J, Ellis IO, et al: Correlation between immunohistochemistry (HercepTest) and fluorescence in situ hybridization
(FISH) for HER-2 in 426 breast carcinomas from 37 centres. J Pathol
199:418-423, 2003
McKenzie SJ, DeSombre KA, Bast BS, et al: Serum levels of HER-2
neu (C-erbB-2) correlate with overexpression of p185neu in human
ovarian cancer. Cancer 71:3942-3946, 1993
Hirashima N, Takahashi W, Yoshii S, et al: Protein overexpression
and gene amplification of c-erb B-2 in pulmonary carcinomas: a
comparative immunohistochemical and fluorescence in situ hybridization study. Mod Pathol 14:556-562, 2001
Cohen JA, Weiner DB, More KF, et al: Expression pattern of the neu
(NGL) gene-encoded growth factor receptor protein (p185neu) in
normal and transformed epithelial tissues of the digestive tract. Oncogene 4:81-88, 1989
David L, Seruca R, Nesland JM, et al: c-erbB-2 expression in primary
gastric carcinomas and their metastases. Mod Pathol 5:384-390, 1992
Nakajima M, Sawada H, Yamada Y, et al: The prognostic significance of amplification and overexpression of c-met and c-erb B-2 in
human gastric carcinomas. Cancer 85:1894-1902, 1999
Takehana T, Kunitomo K, Kono K, et al: Status of c-erbB-2 in gastric
adenocarcinoma: a comparative study of immunohistochemistry, fluorescence in situ hybridization and enzyme-linked immuno-sorbent
assay. Int J Cancer 98:833-837, 2002
Varis A, Zaika A, Puolakkainen P, et al: Coamplified and overexpressed genes at ERBB2 locus in gastric cancer. Int J Cancer 109:
548-553, 2004
Park DI, Yun JW, Park JH, et al: HER-2/neu amplification is an
independent prognostic factor in gastric cancer. Dig Dis Sci 51:13711379, 2006
Kim MA, Jung EJ, Lee HS, et al: Evaluation of HER-2 gene status in
gastric carcinoma using immunohistochemistry, fluorescence in situ
hybridization, and real-time quantitative polymerase chain reaction.
Hum Pathol 38:1386-1393, 2007
Kimura M, Tsuda H, Morita D, et al: A proposal for diagnostically
meaningful criteria to classify increased epidermal growth factor
receptor and c-erbB-2 gene copy numbers in gastric carcinoma, based
on correlation of fluorescence in situ hybridization and immunohistochemical measurements. Virchows Arch 445:255-262, 2004
Vulfovich M, Rocha–Lima C: Novel advances in pancreatic cancer
treatment. Expert Rev Anticancer Ther 8:993-1002, 2008
Slamon DJ, Leyland–Jones B, Shak S, et al: Use of chemotherapy
plus a monoclonal antibody against HER2 for metastatic breast
cancer that overexpresses HER2. N Engl J Med 344:783-792, 2001
Romond EH, Perez EA, Bryant J, et al: Trastuzumab plus adjuvant
chemotherapy for operable HER2-positive breast cancer. N Engl
J Med 353:1673-1684, 2005
Piccart–Gebhart MJ, Procter M, Leyland–Jones B, et al: Trastuzumab
after adjuvant chemotherapy in HER2-positive breast cancer. N Engl
J Med 353:1659-1672, 2005
285
78. Tanner M, Hollmen M, Junttila TT, et al: Amplification of HER-2 in
gastric carcinoma: association with Topoisomerase IIalpha gene amplification, intestinal type, poor prognosis and sensitivity to trastuzumab. Ann Oncol 16:273-278, 2005
79. Rebischung C, Barnoud R, Stefani L, et al: The effectiveness of
trastuzumab (Herceptin) combined with chemotherapy for gastric
carcinoma with overexpression of the c-erbB-2 protein. Gastric Cancer 8:249-252, 2005
80. Peltomaki P: Deficient DNA mismatch repair: a common etiologic
factor for colon cancer. Hum Mol Genet 10:735-740, 2001
81. Liu B, Nicolaides NC, Markowitz S, et al: Mismatch repair gene
defects in sporadic colorectal cancers with microsatellite instability.
Nat Genet 9:48-55, 1995
82. Peltomaki P, Lothe RA, Aaltonen LA, et al: Microsatellite instability
is associated with tumors that characterize the hereditary non-polyposis colorectal carcinoma syndrome. Cancer Res 53:5853-5855, 1993
83. Seruca R, Santos NR, David L, et al: Sporadic gastric carcinomas
with microsatellite instability display a particular clinicopathologic
profile. Int J Cancer 64:32-36, 1995
84. dos Santos NR, Seruca R, Constância M, et al: Microsatellite instability at multiple loci in gastric carcinoma: clinicopathologic implications and prognosis. Gastroenterology 110:38-44, 1996
85. Chung YJ, Park SW, Song JM, et al: Evidence of genetic progression
in human gastric carcinomas with microsatellite instability. Oncogene 15:1719-1726, 1997
86. Hayden JD, Cawkwell L, Quirke P, et al: Prognostic significance of
microsatellite instability in patients with gastric carcinoma. Eur J
Cancer 33:2342-2346, 1997
87. Oliveira C, Seruca R, Seixas M, et al: The clinicopathological features of gastric carcinomas with microsatellite instability may be
mediated by mutations of different “target genes”: a study of the
TGFbeta RII, IGFII R, and BAX genes. Am J Pathol 153:1211-1219,
1998
88. Fleisher AS, Esteller M, Wang S, et al: Hypermethylation of the
hMLH1 gene promoter in human gastric cancers with microsatellite
instability. Cancer Res 59:1090-1095, 1999
89. Halling KC, Harper J, Moskaluk CA, et al: Origin of microsatellite
instability in gastric cancer. Am J Pathol 155:205-211, 1999
90. Kang GH, Shim YH, Ro JY, et al: Correlation of methylation of the
hMLH1 promoter with lack of expression of hMLH1 in sporadic
gastric carcinomas with replication error. Lab Invest 79:903-909,
1999
91. Leung SY, Yuen ST, Chung LP, et al: hMLH1 promoter hypermethylation and lack of hMLH1 expression in sporadic gastric carcinomas
with high-frequency microsatellite instability. Cancer Res 59:159164, 1999
92. Yamamoto H, Perez–Piteira J, Yoshida T, et al: Gastric cancers of the
microsatellite mutator phenotype display characteristic genetic and
clinical features. Gastroenterology 116:1348-1357, 1999
93. Pinto M, Oliveira C, Machado JC, et al: MSI-L gastric carcinomas
share the hMLH1 methylation status of MSI-H carcinomas but not
their clinicopathological profile. Lab Invest 80:1915-1923, 2000
94. Aarnio M, Salovaara R, Aaltonen LA, et al: Features of gastric cancer
in hereditary non-polyposis colorectal cancer syndrome. Int J Cancer
74:551-555, 1997
95. Pinto M, Wu Y, Mensink RG, et al: Somatic mutations in mismatch
repair genes in sporadic gastric carcinomas are not a cause but a
consequence of the mutator phenotype. Cancer Genet Cytogenet
18:110-114, 2008
96. Suzuki H, Itoh F, Toyota M, et al: Distinct methylation pattern and
microsatellite instability in sporadic gastric cancer. Int J Cancer
83:309-313, 1999
97. Toyota M, Ahuja N, Suzuki H, et al: Aberrant methylation in gastric
cancer associated with CpG island methylator phenotype. Cancer Res
59:5438-5442, 1999
98. Latham KE: X chromosome imprinting and inactivation in the early
mammalian embryo. Trends Genet 12:134-138, 1996
286
Seminars in Diagnostic Pathology, Vol 25, No 4, November 2008
99. Issa JP, Ottaviano YL, Celano P, et al: Methylation of the oestrogen
receptor CpG island links ageing and neoplasia in human colon. Nat
Genet 7:536-540, 1994
100. Esteller M, Corn PG, Baylin SB, et al: A gene hypermethylation
profile of human cancer. Cancer Res 61:3225-3229, 2001
101. Nguyen C, Liang G, Nguyen TT, et al: Susceptibility of nonpromoter
CpG islands to de novo methylation in normal and neoplastic cells.
J Natl Cancer Inst 93:1465-1472, 2001
102. Kikuchi T, Itoh F, Toyota M, et al: Aberrant methylation and histone
deacetylation of cyclooxygenase 2 in gastric cancer. Int J Cancer
97:272-277, 2002
103. Kang GH, Shim YH, Jung HY, et al: CpG island methylation in
premalignant stages of gastric carcinoma. Cancer Res 61:2847-2851,
2001
104. Fleisher AS, Esteller M, Tamura G, et al: Hypermethylation of the
hMLH1 gene promoter is associated with microsatellite instability in
early human gastric neoplasia. Oncogene 20:329-335, 2001
105. Carvalho B, Pinto M, Cirnes L, et al: Concurrent hypermethylation of
gene promoters is associated with a MSI-H phenotype and diploidy in
gastric carcinomas. Eur J Cancer 39:1222-1227, 2003
106. Song SH, Jong HS, Choi HH, et al: Transcriptional silencing of
cyclooxygenase-2 by hyper-methylation of the 50 CpG island in
human gastric carcinoma cells. Cancer Res 61:4628-4635, 2001
107. Nakagawa H, Nuovo GJ, Zervos EE, et al: Age-related hypermethylation of the 50 region of HMLH1 in normal colonic mucosa is
associated with microsatellite-unstable colorectal cancer development. Cancer Res 61:6991-6995, 2001
108. Tamura G, Yin J, Wang S, et al: E-Cadherin gene promoter hypermethylation in primary human gastric carcinomas. J Natl Cancer Inst
92:569-573, 2000
109. Lee TL, Leung WK, Chan MW, et al: Detection of gene promoter
hypermethylation in the tumor and serum of patients with gastric
carcinoma. Clin Cancer Res 8:1761-1766, 2002
110. Esteller M, Hamilton SR, Burger PC, et al: Inactivation of the DNA
repair gene O6-methylguanine-DNA methyltransferase by promoter
hypermethylation is a common event in primary human neoplasia.
Cancer Res 59:793-797, 1999
111. Park TJ, Han SU, Cho YK, et al: Methylation of O(6)-methylguanine-DNA methyltransferase gene is associated significantly with
K-ras mutation, lymph node invasion, tumor staging, and disease free
survival in patients with gastric carcinoma. Cancer 92:2760-2768,
2001
112. Mitra G, Pauly GT, Kumar R, et al: Molecular analysis of O6substituted guanine-induced mutagenesis of ras oncogenes. Proc Natl
Acad Sci USA 86:8650-8654, 1989
113. Beghelli S, de Manzoni G, Barbi S, et al: Microsatellite instability in
gastric cancer is associated with better prognosis in only stage II
cancers. Surgery 139:347-356, 2006
114. Duval A, Reperant M, Compoint A, et al: Target gene mutation
profile differs between gastrointestinal and endometrial tumors with
mismatch repair deficiency. Cancer Res 62:1609-1612, 2002
115. Davalos V, Dopeso H, Velho S, et al: High EPHB2 mutation rate in
gastric but not endometrial tumors with microsatellite instability.
Oncogene 26:308-311, 2007
116. Amendt C, Schirmacher P, Weber H, et al: Expression of a dominant
negative type II TGF-beta receptor in mouse skin results in an
increase in carcinoma incidence and an acceleration of carcinoma
development. Oncogene 17:25-34, 1998
117. Attisano L, Wrana JL, Lopez–Casillas F, et al: TGF-beta receptors
and actions. Biochim Biophys Acta 1222:71-80, 1994
118. Derynck R: TGF-beta-receptor-mediated signaling. Trends Biochem
Sci 19:548-553, 1994
119. Wrana JL, Attisano L, Carcamo J, et al: TGF beta signals through a
heteromeric protein kinase receptor complex. Cell 71:1003-1014,
1992
120. Heldin CH, Miyazono K, ten Dijke P: TGF-beta signaling from cell
membrane to nucleus through SMAD proteins. Nature 390:465-471,
1997
121. Carcamo J, Zentella A, Massague J: Disruption of transforming
growth factor beta signaling by a mutation that prevents transphosphorylation within the receptor complex. Mol Cell Biol 15:15731581, 1995
122. Markowitz S, Wang J, Myeroff L, et al: Inactivation of the type II
TGF receptor in colon cancer cells with microsatellite instability.
Science 268:1336-1338, 1995
123. Souza RF, Appel R, Yin J, et al: Microsatellite instability in the
insulin-like growth factor II receptor gene in gastrointestinal tumours.
Nature Genet 14:255-257, 1996
124. Dahms NM, Lobel P, Kornfeld S: Mannose 6-phosphate receptors
and lysosomal enzyme targeting. J Biol Chem 264:12115-12118,
1989
125. Dennis PA, Rifkin DB: Cellular activation of latent transforming
growth factor beta requires binding to the cation-independent mannose 6-phosphate/insulin-like growth factor type II receptor. Proc
Natl Acad Sci USA 88:580-584, 1991
126. Kornfeld S: Structure and function of the mannose 6-phosphate/
insulin-like growth factor type II receptors. Annu Rev Biochem
61:307-330, 1992
127. Falchetti M, Saieva C, Lupi R, et al: Gastric cancer with high-level
microsatellite instability: target gene mutations, clinicopathologic
features, and long-term survival. Hum Pathol 39:925-932, 2008
128. Minniti CP, Kohn EC, Grubb JH, et al: The insulin-like growth factor
II (IGF-II)/mannose 6-phosphate receptor mediates IGF-II-induced
motility in human rhabdomyosarcoma cells. J Biol Chem 267:90009004, 1992
129. Davies H, Bignell GR, Cox C, et al: Mutations of the BRAF gene in
human cancer. Nature 417:949-954, 2002
130. Rajagopalan H, Bardelli A, Lengauer C, et al: Tumorigenesis: RAF/
RAS oncogenes and mismatch-repair status. Nature 418:934, 2002
131. Yuen ST, Davies H, Chan TL, et al: Similarity of the phenotypic
patterns associated with BRAF and KRAS mutations in colorectal
neoplasia. Cancer Res 62:6451-6455, 2002
132. Brennetot C, Duval A, Hamelin R, et al: Frequent Ki-ras mutations in
gastric tumors of the MSI phenotype. Gastroenterology 125:1282,
2003
133. Oliveira C, Pinto M, Duval A, et al: BRAF mutations characterize
colon but not gastric cancer with mismatch repair deficiency. Oncogene 22:9192-9196, 2003
134. Lee JW, Soung YH, Kim SY, et al: PIK3CA gene is frequently
mutated in breast carcinomas and hepatocellular carcinomas. Oncogene 24:1477-1480, 2005
135. Velho S, Oliveira C, Ferreira A, et al: The prevalence of PIK3CA
mutations in gastric and colon cancer. Eur J Cancer 41:1649-1654,
2005
136. Li V, Wong C, Chan T, et al: Mutations of PIK3CA in gastric
adenocarcinoma. BMC Cancer 5:29, 2005
137. Samuels Y, Wang Z, Bardelli A, et al: High frequency of mutations
of the PIK3CA gene in human cancers. Science 304:554, 2004
138. Katso R, Okkenhaug K, Ahmadi K, et al: Cellular function of
phosphoinositide 3-kinases: Implications for development, immunity,
homeostasis, and cancer. Annu Rev Cell Dev Biol 17:615-675, 2001
139. Wymann MP, Pirola L: Structure and function of phosphoinositide
3-kinases. Biochim Biophys Acta 1436:127-150, 1998
140. Byun DS, Cho K, Ryu BK, et al: Frequent monoallelic deletion of
PTEN and its reciprocal association with PIK3CA amplification in
gastric carcinoma. Int J Cancer 104:318-327, 2003
141. Vivanco I, Sawyers CL: The phosphatidylinositol 3-Kinase AKT
pathway in human cancer. Nat Rev Cancer 2:489-501, 2002
142. Samuels Y, Velculescu VE: Oncogenic mutations of PIK3CA in
human cancers. Cell Cycle 3:1221-1224, 2004
143. Osaki M, Kase S, Adachi K, et al: Inhibition of the PI3K-Akt
signaling pathway enhances the sensitivity of Fas-mediated apoptosis
in human gastric carcinoma cell line. MKN-45. J Cancer Res Clin
Oncol 130:8-14, 2004
Carneiro et al
Gastric Cancer
144. Osaki M, Oshimura M, Ito H: PI3K-Akt pathway: its functions and
alterations in human cancer. Apoptosis 9:667-676, 2004
145. Kim DW, Huamani J, Fu A, et al: Molecular strategies targeting the
host component of cancer to enhance tumor response to radiation
therapy. Int J Radiat Oncol Biol Phys 64:38-46, 2006
146. Ng SSW, Tsao M-S, Chow S, et al: Inhibition of phosphatidylinositide 3-kinase enhances gemcitabine-induced apoptosis in human pancreatic cancer cells. Cancer Res 60:5451-5455, 2000
147. Sarkaria JN, Tibbetts RS, Busby EC, et al: Inhibition of phosphoinositide 3-kinase related kinases by the radiosensitizing agent wortmannin. Cancer Res 58:4375-82, 1998
148. Samuels Y, Diaz LAJ, Schmidt-Kittler O, et al: Mutant PIK3CA
promotes cell growth and invasion of human cancer cells. Cancer
Cell 7:561-573, 2005
287
149. Ikenoue T, Kanai F, Hikiba Y, et al: Functional analysis of PIK3CA
gene mutations in human colorectal cancer. Cancer Res 65:45624567, 2005
150. Isakoff SJ, Engelman JA, Irie HY, et al: Breast cancer-associated
PIK3CA mutations are oncogenic in mammary epithelial cells. Cancer Res 65:10992-11000, 2005
151. Zhao JJ, Liu Z, Wang L, et al: The oncogenic properties of mutant
p110{alpha} and p110{beta} phosphatidylinositol 3-kinases in human mammary epithelial cells. Proc Natl Acad Sci USA 102:1844318448, 2005
152. Kang S, Bader AG, Zhao L, et al: Mutated PI 3-kinases: cancer
targets on a silver platter. Cell Cycle 4:578-581, 2005
153. Chan S: Targeting the mammalian target of rapamycin (mTOR): a
new approach to treating cancer. Br J Cancer 91:1420-1424, 2004