Gastric cancer

Paul Lochhead and Emad M. El-Omar

Gastrointestinal Research Group, Department of Medicine and Therapeutics, Aberdeen University,
Aberdeen, UK
Background: Gastric cancer remains a major cause of mortality and morbidity
worldwide, and the total number of gastric cancer cases is predicted to rise as a
result of population growth. The pathogenesis of gastric cancer represents a
paradigm for microbially induced and inammation-driven malignancies, and
understanding this will be the best means of defeating this cancer.
Sources of data: We reviewed the relevant English language literature in
relation to gastric cancer with particular reference to the role of Helicobacter
pylori. We summarize what is known of the epidemiology, aetiology and
pathogenesis of gastric cancer. We also describe current approaches to the
detection and management of early gastric cancer and discuss the prevention
strategies.
Areas of agreement: H. pylori is the most important aetiological risk factor for
this cancer, and the pathogenesis involves the combined effects of host genetics,
bacterial virulence and environmental factors.
Areas of disagreement: Although most accept that removing Helicobacter could
prevent gastric cancer, there are still no denitive trials to prove this concept.
There is also some anxiety about the long-term effects of removing such a
prevalent chronic infection from large sections of the population.
Conclusions: Gastric cancer is now arguably one of the most understood
malignancies, and real progress is being made towards eradicating this global
killer. Much work still needs to be done to dene the optimal approach for
eradicating the causative agent, namely H. pylori infection.
Keywords: gastric cancer/helicobacter pylori/epidemiology/diet/host
genetics/screening/endoscopic therapy/cancer prevention
British Medical Bulletin 2008; 85: 87100
DOI:10.1093/bmb/ldn007
& The Author 2008. Published by Oxford University Press.
All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org
Accepted: January 22,
2008
Correspondence to:
Prof. Emad M El-Omar,
Department of Medicine
and Therapeutics,
Aberdeen University,
Institute of Medical
Sciences, Foresterhill,
Aberdeen AB25 2ZD, UK.
E-mail: e.el-omar@abdn.
ac.uk
Published Online February 10, 2008

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Introduction
The earliest descriptions of gastric cancer can be found in ancient
Egyptian medical papyri such as the Ebers papyrus dating to around
1550 BC. At the turn of the twentieth century, gastric cancer was the
leading cause of cancer-related death in the UK. The incidence of
gastric cancer has fallen dramatically in most countries over the past
70 years. Between 1975 and 2003, the incidence of gastric cancer in
males in the UK halved, and it now represents the fifth and ninth most
common cancer in males and females, respectively. Nonetheless, 8000
new cases of gastric cancer were registered in the UK in 2003, and
gastric cancer deaths in 2004 totalled over 5800.
1
Globally, gastric
cancer is the second most common cause of cancer-related death and,
as a result of population aging and growth, the predicted incidence for
2010 is 1.1 million with the majority of this health burden being borne
by economically lesser-developed countries.
2
Although two-thirds of cases of gastric cancer occur in lesser-
developed countries, the age-standardized incidence figures are in fact
comparable with those of more developed countries.
3
There are,
however, geographical variations of up to 10-fold in gastric cancer inci-
dence, even greater than those observed for oesophageal cancer. In
general, incidence rates are lowest in Western Europe and North
America and highest in Eastern Asia, South America and Eastern
Europe. The prevalence of gastric cancer in Japan is around eight times
higher than in the USA. This geographical distribution is not clear-cut
with India, for example, having a relatively low incidence when com-
pared with other Asian countries such as China, Japan and Indonesia.
3
Even within individual countries, there are both regional and racial
differences in incidence. Migrant populations from high-risk parts of
the world can represent high-risk subgroups within low-risk popu-
lations, for example, Koreans living in the USA. With subsequent gen-
erations, the risk within migrant populations falls to assume the
background risk in the host country.
4
In all populations for which data are available, the incidence of
gastric cancer in males is around twice that of females. In the UK, this
ratio is 5:3.
1
Pathological differentiation
The vast majority of malignant tumours of the stomach are adenocarci-
nomas. Historically gastric adenocarcinomas were subdivided by the
Laure n histological classification into intestinal and diffuse subtypes.
P. Lochhead and E. M. El-Omar
88 British Medical Bulletin 2008;85

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

The intestinal type appears to arise from a background of chronic
atrophic gastritis, with the development of intestinal metaplasia and
transition through progressively increasing stages of dysplasia to
carcinoma. The intestinal type is the predominant histological type in
high-risk countries, more common in men, and increases significantly in
incidence with age. The diffuse type does not show marked geographic
variation, can arise in the absence of atrophic gastritis, is more common
in females, and often occurs in younger patients with a positive family
history.
46
These epidemiological differences lead to the notion that the
two histological subtypes had distinct aetiologies. There are, however,
problems with this concept. Laure n himself was unable to classify a sig-
nificant number of tumours as being either intestinal or diffuse, and
some tumours show features of both histological subtypes.
6
It has
become apparent over the recent years that gastric adenocarcinoma can
also be subdivided according to the anatomical site at which it arises.
Tumours can be said to be proximal, arising in the cardia region of the
stomach, or distal arising from non-cardia regions. As with the histologi-
cal subtypes, there are aetiological and epidemiological differences
between the two tumour sub-sites. Until recently, adenocarcinoma of
the gastric cardia represented a small proportion of gastric cancers as a
whole. The global trend of falling gastric cancer incidence therefore
reflects non-cardia cancer incidence. The incidence of carcinoma of the
gastric cardia actually appears to be rising.
3,4
Whilst this may be partly
due to more accurate reporting, the dramatic fall in the incidence of
distal cancers has not been paralleled for cardia cancers. Gastric cardia
cancer is now therefore relatively more common accounting for around
50% of cases of gastric cancer in developed countries.
5
Survival
Outside of Japan, survival rates for gastric cancer remain poor.
Estimated 5 year survival in Europe is only 2427%.
2,3
The survival
rates in North America are somewhat better perhaps as a result of
earlier detection as a result of the overall larger number of diagnostic
endoscopic procedures performed. In Japan, where population-based
endoscopic screening was introduced in the 1960s, the 5-year survival
is greater than 50%.
2
Helicobacter pylori as a risk factor for gastric cancer
The discovery of H. pylori in 1983 has proved to be pivotal in our
understanding of the aetiology of gastric cancer. H. pylori is a
Gastric cancer
British Medical Bulletin 2008;85 89

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

gram-negative bacillus capable of colonizing the gastric mucosa.
Countries with high rates of gastric cancer, such as China and Japan,
tend to have a high prevalence of H. pylori infection.
7
In 1994, the World Health Organization and the International
Agency for Research on Cancer consensus group stated that there was
sufficient evidence to classify H. pylori as a Class I (i.e. definite)
human carcinogen. The evidence as it stood at that time comprised epi-
demiological studies only, many of which were poorly controlled for
confounding risk factors for gastric cancer. Over the subsequent 10
years, numerous cohort and case-control studies were published
demonstrating an association between serological evidence of H. pylori
infection and increased risk of gastric cancer.
8,9
Much of these data
have been subjected to meta-analysis with the conclusion that H. pylori
infection carries around a 2-fold increased risk for the development of
gastric cancer.
8
This association is strongest for non-cardia cancer, but
holds for both intestinal and diffuse histological types.
10
The most compelling epidemiological evidence comes from a large
prospective series from Japan. A total of 1526 patients who had been
found to have duodenal ulcers, gastric ulcers, gastric hyperplasia or
non-ulcer dyspepsia were recruited. Endoscopy and biopsy were per-
formed at enrolment and H. pylori status established by histological,
enzymatic (urease test) or serological methods. A total of 1246 patients
had evidence of H. pylori infection, the remainder did not. The partici-
pants underwent endoscopic follow-up for the detection of early
gastric cancer (EGC) over a mean follow-up period of 7.8 years.
Gastric cancers developed in 36 (2.9%) of infected patients, but in
none of the uninfected patients.
11
Among patients with H pylori, those
with severe atrophic gastritis, corpus predominant gastritis and intesti-
nal metaplasia were found to be at highest risk of developing gastric
cancer. Interestingly, 4.7% of infected patients with non-ulcer dyspep-
sia went on to develop gastric cancer, but none of the 275 patients
with duodenal ulcer developed gastric cancer. The latter observation is
in keeping with previous studies that have shown a negative association
between history of duodenal ulcer and the risk of gastric cancer.
Since the vast majority of individuals infected with H. pylori do not
develop gastric cancer, there must be additional factors that determine
which individuals will go on to develop malignancy. One potential
factor is bacterial virulence. The most widely studied H. pylori virulence
factor is the cytotoxin-associated gene A antigen (CagA). One
meta-analysis has specifically addressed the association between infec-
tion with CagA harbouring strains of H. pylori and gastric cancer risk.
10
Among H pylori-infected populations, seropositivity for CagA increases
the risk of developing non-cardia gastric cancer by around an additional
2-fold. In the West, around 60% of H. pylori isolates possess CagA
P. Lochhead and E. M. El-Omar
90 British Medical Bulletin 2008;85

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

compared with virtually all isolates in Japan.
4
In addition to CagA,
there are other bacterial virulence factors that have been associated with
increased risk of gastric cancer. These include the vacuolating cytotoxin
VacA and some outer membrane proteins such as OipA and BabA.
Helicobacter and cancer of the gastric cardia
The association between H. pylori infection, atrophic gastritis and
cancer of the cardia region of the stomach is less clear than it is for
distal cancers. In Western countries, there appears to be a negative
association between H. pylori infection and cardia cancer, whereas
there is a trend towards positive association in Eastern Asia.
5
A recent nested casecontrol study based in Norway has helped elu-
cidate the aetiology of cardia cancer.
12
The study set out to compare
the premorbid state of the gastric mucosa between individuals with
cardia and non-cardia gastric cancer. Out of an original cohort of 101
601 individuals, 129 non-cardia and 44 cardia cancers were included
and matched with three controls each. Serum samples had been col-
lected a median of 11.9 years before the diagnosis of cancer. The
serum samples were analysed for H. pylori serology, and the ratio of
pepsinogen 1:2, which is a marker of H pylori-induced atrophic gastri-
tis. As expected, there was a strong association between H. pylori sero-
positivity and non-cardia cancer; however, there was a negative
association between H. pylori infection and cardia cancer. The predo-
minant histological subtype of cardia cancer was intestinal and was not
associated with gastric atrophy. In persons with cardia cancer plus ser-
ological evidence of atrophic gastritis, however, there was a strong
association with H. pylori infection, and the intestinal:diffuse ratio was
1:1, similar to non-cardia cancer. These findings indicate that there are
two aetiologically distinct types of cardia cancer: one associated with
H pylori-induced atrophic gastritis, similar to non-cardia cancer, and
the other associated with non-atrophic gastric mucosa and resembling
oesophageal adenocarcinoma.
12
The role of host genetics
Helicobacter pylori is responsible for three separate phenotypes in the
infected host. The first is corpus-predominant gastritis leading to
atrophic gastritis, hypochlorhydria and to the development of gastric
cancer. The second is the benign phenotype where H. pylori infection
results in a mild mixed gastritis that has minimal effect on the physi-
ology of gastric acid production. The third phenotype is the duodenal
Gastric cancer
British Medical Bulletin 2008;85 91

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

ulcer phenotype where an antrum-predominant gastritis leads to
increased gastric acid secretion and duodenal ulceration. Therein lies
the paradox that H. pylori infection can predispose to two mutually
exclusive conditions.
Individual differences in the host response to H. pylori infection,
determined by host genetic polymorphisms, might, in part, explain
why some individuals are more likely to develop the gastric cancer phe-
notype than others. Evidence for the importance of host genetic factors
initially came from the finding of an increased incidence of atrophic
gastritis and hypochlorhydria in H. pylori-infected relatives of gastric
cancer patients when compared with matched controls.
13
Interleukin-1 beta (IL-1b) is a pro-inflammatory cytokine and also a
potent inhibitor of gastric acid secretion. The IL-1 gene was therefore a
potential candidate for host genetic polymorphisms that may influence
gastric cancer risk. Individuals with pro-inflammatory IL-1 gene
cluster polymorphisms are at increased risk of developing mucosal
atrophy and hypochlorhydria in response to H. pylori infection, and
this is reflected in a 23-fold increase in the risk of non-cardia
cancer.
14,15
Following the identification of pro-inflammatory IL-1 gene
cluster polymorphisms, pro-inflammatory genotypes of tumour necro-
sis factor-a(TNF-a) and interleukin-10 (IL-10) were described as inde-
pendent risk factors for non-cardia cancer.
14
TNF-a is another
pro-inflammatory cytokine whose expression is upregulated in the
gastric mucosa in response to H. pylori infection. IL-10, on the other
hand, is an anti-inflammatory cytokine that suppresses expression of
pro-inflammatory cytokines including IL-1b, TNF-a and interferon-g.
The risk of developing gastric cancer has been studied with respect to
the inheritance of multiple pro-inflammatory polymorphisms in the
aforementioned genes. The risk increases progressively such that an
individual with three or four polymorphisms who is infected with
H. pylori has 27-fold increased risk of developing non-cardia cancer.
16
This clearly demonstrates that geneenvironment interactions are
important in the aetiology of gastric cancer, and that inflammation and
the adaptive immune response play a pivotal role in driving gastric
carcinogenesis.
Familial gastric cancer
Approximately 1015% of gastric cancers arise in individuals with a
significant family history of the condition.
17,18
An increased risk of
gastric cancer is associated with several of the recognized dominantly
inherited cancer predisposition syndromes, such as familial adenoma-
tous polyposis, hereditary non-polyposis colon cancer and
P. Lochhead and E. M. El-Omar
92 British Medical Bulletin 2008;85

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

Peutz-Jeghers syndrome. Hereditary diffuse gastric cancer (HDGC) is
also inherited as a dominant trait, and in around a third of affected
kindred is caused by inactivating mutations in the CDH1 gene, which
encodes the epithelial cell adhesion protein E-cadherin.
18
HDGC is
highly penetrant and the average age at diagnosis is 38 years.
Other risk factors
Although H. pylori plays a crucial role in the pathogenesis of gastric
cancer, it cannot be considered the sole causative agent in this malig-
nancy. There are other factors that clearly contribute and perhaps the
strongest are diet and smoking.
Diet
The marked geographic variation in gastric cancer incidence suggests
that environmental influences play an important aetiological role, and
much attention has, therefore, focussed on the association between diet
and gastric cancer risk. It has been postulated that the downward trend
in gastric cancer incidence may in part be due to the advent of wide-
spread refrigeration of food, hence increased intake of fresh produce,
and less reliance on food preservation. Numerous studies have
suggested a protective effect from diets rich in fresh fruit and veg-
etables; however, when one considers case-control and prospective data
only, the evidence is less robust.
3
Data from a recent large European
prospective study (EPIC-EURGAST) failed to show an overall associ-
ation between fruit or vegetable consumption and gastric cancer risk.
19
There was, however, a statistically significant association between total
dietary vegetable content (and onion and garlic intake) and the intesti-
nal histological subtype. A non-significant negative association was
observed between cardia cancer risk and citrus fruit consumption.
Interestingly, the effect of fruit and vegetable intake appeared to be
independent of H. pylori status.
Vitamin C and other anti-oxidant nutrients have attracted a lot of
attention as the potential mediators of any dietary influence on gastric
cancer risk. Vitamin C seems a promising candidate since its levels are
reduced in the serum of H. pylori-infected individuals, and, as well as
being a free radical scavenger, it reduces the formation of potentially
carcinogenic N-nitroso compounds.
20
Furthermore, there is evidence
from case-control studies of a negative association between dietary
vitamin C and gastric cancer risk.
21
Data from the EPIC cohort show a
negative correlation between gastric cancer risk and serum vitamin C,
Gastric cancer
British Medical Bulletin 2008;85 93

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

but not dietary vitamin C, intake, and this was unaffected by H. pylori
status.
22
A review by the Cochrane Collaboration, which included a number
of high-quality randomized trials, has concluded that there is no evi-
dence that dietary supplementation with anti-oxidants, including
vitamin C, reduces gastric cancer risk.
23
Salt and nitrite are other dietary components that have been impli-
cated in gastric cancer risk. Both are frequently used in food preser-
vation. Pickled and smoked foods may also contain potential
carcinogens such as N-nitroso compounds and benzapyrene. Of note is
the fact that the Japanese diet is particularly rich in salted fish and
pickled vegetables. Dietary nitrate, of which a significant portion may
come from water depending on the source, can be converted to nitrite
by nitrate reductase-synthesizing bacteria. H. pylori infection and
hypochlorhydria facilitate the growth of such bacteria.
Dietary salt and nitroso compounds appear to exert a synergistic
effect with H. pylori in animal models of gastric carcinogenesis.
Several casecontrol studies in humans have shown a positive
association between gastric cancer risk and both salt and dietary
nitrate/nitrite intake, although data from prospective studies are
conflicting.
2426
Smoking
Smoking is an independent risk factor for gastric cancer. Prospective
studies have demonstrated a significant dose-dependent association
between tobacco smoke and gastric cancer risk.
27,28
In the EPIC
cohort, it was estimated that 17.6% of gastric cancer in this European
population was attributable to cigarette smoking.
29
Detection and treatment of EGC
As stated previously, survival rates from gastric cancer outside of Japan
remain dismal. The reason that gastric cancer caries such a poor prog-
nosis is that the presentation is invariably late in the natural history of
the disease, with local extension or metastatic disease rendering the
condition inoperable. Whilst chemotherapy, both neo-adjuvant and
palliative, has a demonstrated role in improving survival in patients
with gastric cancer,
30
resection with curative intent offers the only real
chance of long-term survival in patients with operable disease. Around
90% of cases of gastric cancer in Western countries are inoperable at
P. Lochhead and E. M. El-Omar
94 British Medical Bulletin 2008;85

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

presentation in contrast to Japan where EGCs comprise 50% of gastric
cancers detected.
31
Early gastric cancer is defined as gastric carcinoma confined to the
mucosa or sub-mucosa irrespective of the presence or absence of lymph
node metastases.
31
In Japan, 5-year survival rates following resection
of EGC exceed 90%.
32
The detection of EGC has, therefore, become
the focus of endoscopic screening programmes in Japan, and the
impressive overall survival rates for gastric cancer attest to the success
of this approach. Population-based endoscopic screening, such as that
employed in Japan, is not in practice in most other countries, and it
would, therefore, be immensely helpful to be able to identify individ-
uals at risk of EGC so that screening endoscopy could be offered on a
targeted basis.
Could symptoms help predict those individuals who are likely to
have EGC? Referral guidelines for urgent endoscopy, at least in the
UK, often include the so-called alarm symptoms such as weight loss,
anaemia, dysphagia and anorexia. Symptom analysis in patients aged
,45 years diagnosed with gastric cancer has shown that fewer than a
half described alarm symptoms.
31
As one might imagine, alarm symp-
toms tend to indicate advanced disease and, therefore, urgent endo-
scopy frequently has little impact on the final outcome. Unfortunately,
no symptom complex has been shown to predict EGC, and most
patients with EGC are in fact asymptomatic or have non-specific dys-
peptic symptoms.
31
Since H. pylori infection is common (around half of the Worlds
population is infected with the organism), the magnitude of screening
all infected individuals is also impractical. Serological analysis of pepsi-
nogen I and II and gastrin levels has been employed in an attempt to
non-invasively detect the presence of the atrophic gastritis. Several
large studies from Japan have produced promising results using serolo-
gical screening, although data demonstrating sufficient sensitivity and
specificity in non-high risk populations are currently lacking.
31,33
Endoscopic identification of early gastric cancer
As experience grew in the endoscopic detection of EGC in Japan, it
was felt necessary to devise a new macroscopic classification for gastric
cancers.
34
This system was subsequently adopted internationally. A
lesion that is felt endoscopically to represent an EGC is denoted as 0
rather than T1 as would be used in pathological staging. Lesions are
then sub-classified as being protruding (.2.5 mm), flat or excavated.
Flat lesions are further subdivided according to whether they are com-
pletely flat or slightly elevated or depressed. In the West, most EGCs
Gastric cancer
British Medical Bulletin 2008;85 95

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

are identified because of the presence of an obvious mass or ulcer seen
at endoscopy; flat lesions are likely to be easily overlooked.
31
Indeed,
there are data from the UK demonstrating that a significant proportion
of patients presenting with gastric or oesophageal cancer have had an
endoscopy within the preceding 3 years at which time the diagnosis
had been missed.
35
Endoscopist error may be partly explained by the
use of acid-suppressing medications that may cause re-epithelialization
and masking of an underlying malignancy. Another explanation is that
our approach to diagnostic endoscopy in a patient with dyspeptic
symptoms is to exclude peptic ulcer disease and not to examine the
whole of the gastric mucosa in detail for subtle alterations in appear-
ance.
31
This is where our endoscopic practice differs from that of the
Japanese.
The vast majority of diagnostic upper gastrointestinal (GI) procedures
performed in the UK use standard white-light endoscopy (WLE).
Standard WLE has its limitations when carrying out surveillance endo-
scopy for dysplastic pre-malignant lesions. This is clear from our
experience of screening in Barretts oesophagus where repeated WLE
with systematic biopsy is time-consuming both for the endoscopist and
pathologist, and results in poor diagnostic yield. The Japanese have led
the way in the adoption of technologies whereby the detection of EGC
can be enhanced. There are numerous established and emerging diag-
nostic modalities including high-resolution and magnification endo-
scopy, chromoendoscopy, point spectroscopy, fluorescence imaging
and confocal endoscopy. These techniques are reviewed in detail else-
where.
36
Chromoendoscopy is being increasingly employed in endo-
scopy units in the West. The technique involves the topical application
of stains to aid the endoscopic visualization of lesions. The nature and
mechanism of action of the stains varies. Methylene blue, for example,
stains intestinal-type epithelium, but not normal gastric mucosa. Indigo
carmine accentuates mucosal patterns helping identify distorted surface
topography. Techniques such as point spectroscopy and confocal endo-
scopy allow the mucosa to be assessed in vivo for spectroscopic or
microscopic evidence of dysplasia or malignancy. Although these tech-
niques do not replace conventional biopsy for histology, they allow
biopsies to be targeted and thus increasing diagnostic yield.
Endoscopic therapy for early gastric cancer
As a result of the increasing numbers of EGCs that have been detected
in Japan over the past 15 years, techniques have evolved that allow
EGCs to be treated endoscopically, thus reducing the morbidity and
mortality associated with traditional gastric surgery. The completeness
P. Lochhead and E. M. El-Omar
96 British Medical Bulletin 2008;85

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

of resection must not be compromised by the decision to utilize thera-
peutic endoscopy, and patients must, therefore, be carefully selected
with adherence to treatment guidelines.
32
Lymph node metastases are a
contraindication to endoscopic resection since formal surgical lymph
node dissection is required. The presence of lymph node metastases
correlates with the depth of tumour invasion, and endoscopic ultraso-
nography (EUS) in combination with endoscopy is helpful in staging
tumour depth and detecting perigastric nodal metastases.
32
The two endoscopic techniques most widely used in the treatment of
EGC are endoscopic mucosal resection (EMR) and endoscopic submu-
cosal dissection (ESD). Both therapies are tissue preserving, allowing
the specimen to be retrieved and sent for histology, this is in contrast
to ablative techniques. There are a variety of different EMR techniques;
however, these all tend to involve the use of a snare to remove the
affected mucosa, often following injection of saline into the submucosa
to elevate the lesion.
32
EMR is generally limited to lesions ,2 cm in
diameter. Larger areas of mucosa can be removed by ESD where more
extensive submucosal injection is carried out followed by dissection of
the submucosa from the muscularis propria using specialized endo-
scopic cutting devices. ESD is claimed to have higher en bloc resection
rates (over 90%) and lower local recurrence rates than EMR.
37
Prevention of gastric cancer by H. pylori eradication
Screening for EGC is not feasible in most populations, and therefore,
the prevention represents the most promising approach to reducing
gastric cancer mortality on a global scale. Although the exact mechan-
isms by which H. pylori promotes gastric carcinogenesis remain
unclear, it is the single most important aetiological agent in gastric
cancer and therefore the target of preventative strategies. H. pylori is
believed to be acquired in infancy or childhood. There is therefore a
long latency period before the development of malignancy during
which H. pylori eradication may be effective in preventing gastric car-
cinogenesis.
38
Despite what appears to be a logical assumptionthat
H. pylori eradication in infected individuals will prevent the develop-
ment of gastric cancerthere is minimal supporting clinical evidence
from controlled trials.
39,40
This is partly due to the difficulty faced in
recruiting infected individuals who may be randomized to placebo, and
the consequent ethical issues this generates. In addition, the follow-up
period required for such studies is often prohibitively lengthy.
39
The
largest randomized controlled trial, conducted in a high-risk region of
China, was published by Wong et al..
41
In this study, 1630 healthy
H. Pylori carriers were randomized to receive either eradication
Gastric cancer
British Medical Bulletin 2008;85 97

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

therapy or placebo. There was no overall difference in the incidence of
gastric cancer between the two groups over a follow-up period of 7.5
years. Within the subgroup of patients who had no endoscopic evi-
dence of atrophy or intestinal metaplasia at recruitment, there was,
however, a significant reduction in the risk of gastric cancer in the
active treatment group. This suggests that there is likely to be a point
of no return in gastric carcinogenesis, beyond which H. pylori eradica-
tion is ineffective in preventing the progression to carcinoma. The evi-
dence that eradicating H. pylori leads to regression of premalignant
mucosal changes is conflicting, although it appears that, at least in
some individuals, both atrophic gastritis and, to a lesser extent, intesti-
nal metaplasia can regress following eradication of H. pylori, but this
remains highly controvercial.
39
It is likely that not all gastric cancers
will be prevented by H. pylori eradication since some individuals will
already have developed premalignant mucosal changes that may no
longer be dependent on H. pylori for driving the progression to
carcinoma.
Summary
Our understanding of the mechanisms by which gastric cancer arises is
more advanced than for the majority of other solid tumours. H. pylori
is the single most important environmental risk factor for the develop-
ment of non-cardia gastric cancer as well as for at least a proportion of
cardia cancers. Prevention of gastric cancer will only be achievable
through prevention of acquisition of H pylori, or eradication of the
infection before an as yet undetermined critical committal stage in
the gastric carcinogenesis. Further studies are required to determine the
efficacy of H. pylori eradication as a preventative strategy in gastric
cancer. The evidence implicating dietary factors, and the potential for
dietary modification to reduce gastric cancer risk, is conflicting and
merits further investigation.
Screening for EGC in high-risk populations reduces gastric cancer
mortality and facilitates endoscopic therapy, thus reducing the morbid-
ity associated with the condition. It may eventually be possible to
target screening to those individuals most at risk of developing gastric
cancer within non-high-risk populations. Risk estimation is likely to
incorporate factors such as age, family history and H. pylori status
combined with serological testing for gastric atrophy, and possibly
molecular genetic analysis for predisposing polymorphisms and bac-
terial virulence factors.
P. Lochhead and E. M. El-Omar
98 British Medical Bulletin 2008;85

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

References
1 http://info.cancerresearchuk.org/cancerstats/types/stomach.
2 Parkin DM, Bray F, Ferlay J, Pisani P (2005) Global Cancer Statistics, 2002. CA Cancer J
Clin, 55, 74108.
3 Forman D, Burley VJ (2006) Gastric cancer: global pattern of the disease and an overview of
environmental risk factors. Best Pract Res Clin Gastroenterol, 20, 633649.
4 Crew KD, Neugut AI (2006) Epidemiology of gastric cancer. World J Gastroenterol, 21,
354362.
5 McColl KE (2006) Cancer of the gastric cardia. Best Pract Res Clin Gastroenterol, 20,
687696.
6 Woolf N (1998) The Gastrointestinal System. Pathology, Basic and Systemic. London: WB
Saunders, 513518.
7 Prinz C, Schwendy S, Voland P (2006) H. pylori and gastric cancer: shifting the global
burden. World J Gastroenterol, 12, 54585464.
8 Eslick GD (2006) Helicobacter pylori infection causes gastric cancer? A review of the epide-
miological, meta-analytic, and experimental evidence. World J Gastroenterol, 12,
29912999.
9 Lochhead P, El-Omar EM (2007) Helicobacter pylori infection and gastric cancer. Best Pract
Res Clin Gastroenterol, 21, 281297.
10 Huang JQ, Sridhar S, Chen Y, Hunt RH (1998) Meta-analysis of the relationship between
Helicobacter pylori seropositivity and gastric cancer. Gastroenterology, 114, 11697.
11 Uemura N, Okamoto S, Yamamoto S et al. (2001) Helicobacter pylori infection and the
development of gastric cancer. N Engl J Med, 345, 784789.
12 Hansen S, Vollset SE, Derakhshan MH et al. (2007) Two distinct aetiologies of cardia
cancer; evidence from premorbid serological markers of gastric atrophy and H. pylori status.
Gut (Epub ahead of print).
13 El-Omar EM, Oien K, Murray LS et al. (2000) Increased prevalence of precancerous changes
in relatives of gastric cancer patients: critical role of H. pylori. Gastroenterology, 118,
2230.
14 El-Omar EM (2006) Role of host genes in sporadic gastric cancer. Best Pract Res Clin
Gastroenterol, 20, 675686.
15 Smith MG, Hold GL, Tahara E, El-Omar EM (2006) Cellular and molecular aspects of
gastric cancer. World J Gastroenterol, 12, 29792990.
16 El-Omar EM, Rabkin CS, Gammon MD et al. (2003) Increased risk of noncardia gastric
cancer associated with proinflammatory cytokine gene polymorphisms. Gastroenterology,
124, 11931201.
17 Barber M, Fitzgerald RC, Caldas C (2006) Familial gastric canceraetiology and pathogen-
esis. Best Pract Res Clin Gastroenterol, 20, 721734.
18 Firth VF, Hurst JA, Hall JG (2005) Oxford Desk Reference: Clinical genetics. New York:
Oxford University Press, 450451.
19 Gonzalez CA, Pera G, Agudo A et al. (2006) Fruit and vegetable intake and the risk of
stomach and oesophagus adenocarcinoma in the European Prospective Investigation into
Cancer and Nutrition (EPIC-EURGAST). Int J Cancer, 118, 25592566.
20 Woodward M, Tunstall-Pedoe H, McColl K (2001) Helicobacter pylori infection reduces sys-
temic availability of dietary vitamin C. Eur J Gastroenterol Hepatol, 13, 233237.
21 Mayne ST, Risch HA, Dubrow R et al. (2001) Nutrient intake and risk of subtypes of eso-
phageal and gastric cancer. Cancer Epidemiol Biomarkers Prev, 10, 10551062.
22 Jenab M, Riboli E, Ferrari P et al. (2006) Plasma and dietary vitamin C levels and risk of
gastric cancer in the European Prospective Investigation into Cancer and Nutrition
(EPIC-EURGAST). Carcinogenesis, 27, 22502257.
23 Bjelakovic G, Nikolova D, Simonetti RG, Gluud C (2004) Antioxidant supplements for pre-
venting gastrointestinal cancers. Cochrane Database of Systematic Reviews Issue 4.
24 Tokui N, Yoshimura T, Fujino Y et al. (2005) Dietary habits and stomach cancer risk in the
JACC Study. J Epidemiol, 15 (Suppl 2), 98108.
Gastric cancer
British Medical Bulletin 2008;85 99

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m

25 Shikata K, Kiyohara Y, Kubo M et al. (2006) A prospective study of dietary salt intake and
gastric cancer incidence in a defined Japanese population: the Hisayama study. Int J Cancer,
119, 196201.
26 van Loon AJ, Botterweck AA, Goldbohm RA, Brants HA, van Klaveren JD, van den Brandt
PA (1998) Intake of nitrate and nitrite and the risk of gastric cancer: a prospective cohort
study. Br J Cancer, 78, 129135.
27 Nishino Y, Inoue M, Tsuji I et al. (2006) Tobacco smoking and gastric cancer risk: an evalu-
ation based on a systematic review of epidemiologic evidence among the Japanese population.
Jpn J Clin Oncol, 36, 800807.
28 Koizumi Y, Tsubono Y, Nakaya N et al. (2004) Cigarette smoking and the risk of gastric
cancer: a pooled analysis of two prospective studies in Japan. Int J Cancer, 112, 10491055.
29 Gonzalez CA, Pera G, Agudo A et al. (2003) Smoking and the risk of gastric cancer in the
European Prospective Investigation Into Cancer and Nutrition (EPIC). Int J Cancer, 107,
629634.
30 Wagner AD, Schneider PM, Fleig WE (2006) The role of chemotherapy in patients with
established gastric cancer. Best Pract Res Clin Gastroenterol, 20, 789799.
31 Axon A (2006) Symptoms and diagnosis of gastric cancer at early curable stage. Best Pract
Res Clin Gastroenterol, 20, 69770.
32 Yamamoto Kita H (2005) Endoscopic therapy of early gastric cancer. Best Pract Res Clin
Gastroenterol, 19, 909926.
33 Yoshida S, Kozu T, Gotoda T, Saito D (2006) Detection and treatment of early cancer in
high-risk populations. Best Pract Res Clin Gastroenterol, 20, 745765.
34 The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach and
colon: November 30 to December 1, 2002. Gastrointest Endosc, 58, (Suppl), 34.
35 Yalamarthi S, Witherspoon P, McCole D, Auld CD (2004) Missed diagnoses in patients with
upper gastrointestinal cancers. Endoscopy, 36, 874879.
36 Wong Kee Song LM, Wilson BC (2005) Endoscopic detection of early upper GI cancers. Best
Pract Res Clin Gastroenterol, 19, 833856.
37 Fujishiro M (2006) Endoscopic submucosal dissection for stomach neoplasms. World J
Gastroenterol, 12, 51085112.
38 Correa P (2004) Is gastric cancer preventable? Gut, 53, 12171219.
39 Trautmann K, Stolte M, Miehlke S (2006) Eradication of H. pylori for the prevention of
gastric cancer. World J Gastroenterol, 12, 51015107.
40 Malfertheiner P, Fry LC, Monkemuller K (2006) Can gastric cancer be prevented by
Helicobacter pylori eradication? Best Pract Res Clin Gastroenterol, 20, 709719.
41 Wong BC, Lam SK, Wong WM et al. (2004) Helicobacter pylori eradication to prevent
gastric cancer in a high-risk region of China: a randomized controlled trial. JAMA, 291,
187194.
P. Lochhead and E. M. El-Omar
100 British Medical Bulletin 2008;85

b
y

g
u
e
s
t

o
n

A
p
r
i
l

2
7
,

2
0
1
4
h
t
t
p
:
/
/
b
m
b
.
o
x
f
o
r
d
j
o
u
r
n
a
l
s
.
o
r
g
/
D
o
w
n
l
o
a
d
e
d

f
r
o
m