Abbreviations
AGREE:
Appraisal of Guidelines for Research and Evaluation
AUC:
area under the curve
CE:
chromoendoscopy
CI:
confidence interval
COX:
cyclo-oxygenase
EGC:
early gastric cancer
EHMSG:
European Helicobacter and Microbiota Study Group
ESD:
endoscopic submucosal dissection
ESGE:
European Society of Gastrointestinal Endoscopy
ESP:
European Society of Pathology
GI:
gastrointestinal
GRADE:
Grading of Recommendations Assessment, Development, and Evaluation
HD-WLE:
high definition white-light endoscopy
HGD:
high grade dysplasia
HR:
hazard ratio
IM:
intestinal metaplasia
LGD:
low grade dysplasia
MAPS:
Management of precancerous conditions and lesions in stomach
NBI:
narrow-band imaging
NSAID:
nonsteroidal anti-inflammatory drug
OLGA:
Operative Link on Gastritis Assessment
OLGIM:
Operative Link on Gastritis Assessment based on Intestinal Metaplasia
OR:
odds ratio
RCT:
randomized controlled trial
RR:
relative risk
SIR:
standardized incidence ratio
SPED:
Sociedade Portuguesa de Endoscopia Digestiva
This Guideline is an official statement of the European Society of Gastrointestinal
Endoscopy (ESGE), the European Helicobacter and Microbiota Study Group (EHMSG), the
European Society of Pathology (ESP), and the Sociedade Portuguesa de Endoscopia Digestiva
(SPED). Based on new evidence, it makes recommendations on the diagnostic assessment
and management of individuals with atrophic gastritis, intestinal metaplasia and dysplasia
of the stomach, updating the 2012 MAPS guideline.
1 Introduction
Gastric cancer is still a major world problem, ranking fifth for incidence and third
for cancer-related mortality worldwide in the latest published global cancer statistics
[1 ]. Even though early recognition and treatment is possible, most cases are diagnosed
at a late stage and thus most patients with a diagnosis of gastric cancer die of the
disease [1 ]. Screening and surveillance of people at risk may decrease gastric cancer mortality
by allowing early detection and treatment, often by endoscopy instead of more invasive
surgery, and have therefore been recommended [2 ]
[3 ].
In 2012, the European Society of Gastrointestinal Endoscopy (ESGE), the Sociedade
Portuguesa de Endoscopia Digestiva (SPED), the European Helicobacter and Microbiota
Study Group (EHMSG), and the European Society of Pathology (ESP) produced the first
international guideline on the management of precancerous conditions and lesions in
the stomach (MAPS) [4 ]
[5 ]. Its recommendations were then presented in various countries, and were adapted
and translated in some. Moreover, the MAPS Guideline was incorporated into ESGE guidelines
on quality parameters for upper gastrointestinal (GI) endoscopy [6 ].
This document aims to update the first MAPS guideline (referred to here as MAPS I)
and to summarize current evidence on the management of patients with precancerous
conditions and lesions, focusing on the evidence published after 2010.
Scope
Management (diagnostic assessment, treatment, and surveillance) of individuals with
atrophic gastritis, intestinal metaplasia, and dysplasia of the stomach.
2 Methods
These recommendations were developed according to the Appraisal of Guidelines for
Research and Evaluation (AGREE) process for the development of clinical practice guidelines
[7 ]. In October 2016, on behalf of ESGE, EHMSG, ESP, and SPED, the coordinators of the
previous 2012 Guideline (MAPS I) assembled a panel of European gastroenterologists
and pathologists in order to produce an updated guideline, MAPS II.
Working groups were set up to cover the following topics: (1) Definitions and prevalence;
(2) Endoscopic diagnosis; (3) Biopsies and histology; (4) Noninvasive assessment;
(5) Follow-up; (6) Helicobacter pylori treatment; (7) Other therapies; (8) Management; and (9) Cost-effectiveness. (See
online-only Supplementary material .)
The evidence-based Delphi process was applied to develop consensus statements. First,
key questions were agreed and statements were proposed by the MAPS II coordinators
(P.P.N. and M.D.R.), considering previous MAPS I statements and potential changes
to previous recommendations. Each working group considered their statements, and changed
these according to evidence if necessary. A literature search was done using PubMed
(until March 2018) with a focus on articles published after the MAPS I literature
search (November 2010). Each working group rated the quality level of the available
evidence and the strength of recommendations using the Grading of Recommendations
Assessment, Development, and Evaluation (GRADE) process [8 ]
[9 ]. The MAPS II coordinators evaluated and grouped each statement and evidence in a
single document with all the necessary bibliography. This document was then sent to
every participant and statements were voted upon online. At this stage, changes were
made if necessary and statements with less than 75 % agreement were excluded. A final
version with the consensus recommendations ([Table 1 ]) was sent to and approved by every author. Finally, the manuscript was reviewed
by two members of the ESGE Governing Board and sent for further comments to the National
Societies and Individual Members. Suggestions were considered, and after agreement
on a final version the manuscript was submitted for publication.
Table 1
Management of epithelial precancerous conditions and lesions in the stomach (MAPS)
Guidelines: summary of all MAPS I and MAPS II recommendations. Changes from MAPS I
(new or modified recommendations) are shown in bold.
MAPS I
MAPS II (in bold if modified)
Definitions and prevention aims
1 Patients with chronic atrophic gastritis or intestinal metaplasia should be considered
to be at higher risk for gastric adenocarcinoma
1 Patients with chronic atrophic gastritis or intestinal metaplasia are at risk for
gastric adenocarcinoma (high quality evidence)
2 Histologically confirmed intestinal metaplasia is the most reliable marker of atrophy
in gastric mucosa (high quality evidence)
3 Patients with advanced stages of gastritis, that is atrophy and/or intestinal metaplasia
affecting both antral and corpus mucosa, should be identified as they are considered
to be at higher risk for gastric adenocarcinoma (moderate quality evidence, strong
recommendation)
2 High grade dysplasia and invasive carcinoma should be regarded as the outcomes to
be prevented when patients with chronic atrophic gastritis or intestinal metaplasia
are managed
4 High grade dysplasia and invasive carcinoma should be regarded as the outcomes to
be prevented when patients with chronic atrophic gastritis or intestinal metaplasia
are managed (moderate quality evidence, strong recommendation)
3 Patients with endoscopically visible high grade dysplasia or carcinoma should undergo
staging and adequate management
5 Patients with an endoscopically visible lesion harboring low or high grade dysplasia
or carcinoma should undergo staging and treatment (high quality evidence, strong recommendation)
Diagnosis and staging
4 Conventional white light endoscopy cannot accurately differentiate and diagnose
preneoplastic gastric conditions
5 Magnification chromoendoscopy and narrow band imaging (NBI), with or without magnification,
improve the diagnosis of gastric preneoplastic conditions/lesions
6 High definition endoscopy with chromoendoscopy (CE) is better than high definition
white-light endoscopy alone for the diagnosis of gastric precancerous conditions and
early neoplastic lesions (high quality evidence)
6 Within this context, diagnostic upper gastrointestinal endoscopy should include
gastric biopsies sampling
7 Whenever available and after proper training, virtual CE, with or without magnification,
should be used for the diagnosis of gastric precancerous conditions, by guiding biopsy
for staging atrophic and metaplastic changes and by helping to target neoplastic lesions
(moderate quality evidence, strong recommendation)
7 Atrophic gastritis and intestinal metaplasia are often unevenly distributed throughout
the stomach. For adequate staging and grading of gastric precancerous conditions,
at least four non-targeted biopsies of two topographic sites (at the lesser and greater
curvature, from both the antrum and the corpus) should be taken and clearly labelled
in separate vials; additional target biopsies of lesions should be taken
8 For adequate staging of gastric precancerous conditions, a first-time diagnostic upper
gastrointestinal endoscopy should include gastric biopsies both for Helicobacter pylori infection diagnosis and for identification of advanced stages of atrophic gastritis
(moderate quality evidence, strong recommendation)
9 Biopsies of at least two topographic sites (from both the antrum and the corpus, at
the lesser and greater curvature of each) should be taken and clearly labelled in
two separate vials. Additional biopsies of visible neoplastic suspicious lesions should
be taken (moderate quality evidence, strong recommendation)
8 Systems for histopathological staging (e. g. operative link for gastritis assessment
[OLGA] and operative link for gastric intestinal metaplasia [OLGIM] assessment) may
be useful for categorization of risk of progression to gastric cancer
10 Systems for histopathological staging (e. g. Operative Link on Gastritis Assessment
[OLGA] and Operative Link on Gastric Intestinal Metaplasia [OLGIM] assessment) can
be used to identify patients with advanced stages of gastritis. If these systems are
used to stratify patients, additional biopsy of the incisura should be considered
(moderate quality evidence, weak recommendation)
9 Serum pepsinogen levels can predict extensive atrophic gastritis
10 In patients with low pepsinogen test levels, H. pylori serology may be useful for further detection of high risk individuals
11 Low pepsinogen I serum levels or/and low pepsinogen I/II ratio identify patients with
advanced stages of atrophic gastritis and endoscopy is recommended for these patients,
particularly if H. pylori serology is negative (moderate quality evidence, strong recommendation )
11 Family history of gastric cancer should be taken into account in the follow-up
of precancerous conditions
12 Even though diverse studies assessed age, gender, and H. pylori virulence factors as well as host genetic variations, no clinical recommendations
can be made for targeted management based on these factors with regard to diagnosis
and surveillance
12 Even though diverse studies assessed age, gender, and H. pylori virulence factors, as well as host genetic variations, no clinical recommendations
regarding diagnosis and surveillance can be made for targeted management based on
these factors (low quality evidence, weak recommendation)
Surveillance
13 Patients with low grade dysplasia in the absence of an endoscopically defined lesion
should receive follow-up within 1 year after diagnosis. In the presence of an endoscopically
defined lesion, endoscopic resection should be considered, to obtain a more accurate
histological diagnosis
13 In patients with dysplasia in the absence of an endoscopically defined lesion immediate
high quality endoscopic reassessment with CE (virtual or dye-based) is recommended.
If no lesion is detected in this high quality endoscopy, biopsies for staging of gastritis
(if not previously done) and endoscopic surveillance within 6 months (if high grade
dysplasia) to 12 months (if low grade dysplasia) are recommended (low quality evidence,
strong recommendation)
14 For patients with high grade dysplasia in the absence of endoscopically defined
lesions, immediate endoscopic reassessment with extensive biopsy sampling and surveillance
at 6-month to 1-year intervals is indicated
15 For those patients with mild to moderate atrophy/intestinal metaplasia restricted
to the antrum there is no evidence to recommend surveillance
14 For patients with mild to moderate atrophy restricted to the antrum there is no
evidence to recommend surveillance (moderate quality evidence, strong recommendation)
15 Patients with IM at a single location have a higher risk of gastric cancer. However,
this increased risk does not justify surveillance in most cases, particularly if a
high quality endoscopy with biopsies has excluded advanced stages of atrophic gastritis
(moderate quality evidence, strong recommendation)
16 In patients with IM at a single location but with a family history of gastric cancer,
or with incomplete IM, or with persistent H. pylori gastritis, endoscopic surveillance with chromoendoscopy and guided biopsies in 3
years’ time may be considered (low quality evidence, weak recommendation)
16 Endoscopic surveillance should be offered to patients with extensive atrophy and/or
intestinal metaplasia (i. e., atrophy and/or intestinal metaplasia in the antrum and
corpus)
17 Patients with extensive atrophy and/or intestinal metaplasia should receive follow-up
every 3 years after diagnosis
17 Patients with advanced stages of atrophic gastritis (severe atrophic changes or intestinal
metaplasia in both antrum and corpus, OLGA/OLGIM III/IV) should be followed up with
a high quality endoscopy every 3 years (low quality evidence, strong recommendation)
18 Patients with advanced stages of atrophic gastritis and with a family history of gastric
cancer may benefit from a more intensive follow-up (e. g. every 1 – 2 years after
diagnosis) (low quality evidence, weak recommendation)
19 Patients with autoimmune gastritis may benefit from endoscopic follow-up every 3 – 5
years (low quality evidence, weak recommendation)
Therapy
18 H. pylori eradication heals nonatrophic chronic gastritis and it may lead to partial regression
of atrophic gastritis
20
H. pylori eradication heals nonatrophic chronic gastritis, may lead to regression of atrophic
gastritis, and reduces the risk of gastric cancer in patients with nonatrophic and
atrophic gastritis, and, therefore, it is recommended in patients with these conditions
(high quality evidence, strong recommendation)
19 In patients with intestinal metaplasia, H. pylori eradication does not appear to reverse intestinal metaplasia but it may slow progression
to neoplasia, and therefore it is recommended
21 In patients with established IM, H. pylori eradication does not appear to significantly reduce the risk of gastric cancer, at
least in the short term, but reduces inflammation and atrophy and, therefore, it should
be considered (low quality evidence, weak recommendation)
20 H. pylori eradication is recommended for patients with previous neoplasia after endoscopic
or surgical therapy
22 H. pylori eradication is recommended for patients with gastric neoplasia after endoscopic therapy
(high quality evidence, strong recommendation)
21 Currently, the use of cyclo-oxgenase-2 (COX-2) inhibitors cannot be supported as
an approach to decrease the risk of progression of gastric precancerous lesions
23 Even though cyclo-oxygenase (COX)-1 or COX-2 inhibitors may slow progression of gastric
precancerous conditions, they cannot be recommended specifically for this purpose
(low quality evidence, weak recommendation)
22 The use of dietary supplementation with antioxidants (ascorbic acid and betacarotene)
cannot be supported as a therapy to reduce the prevalence of atrophy or intestinal
metaplasia
24 Low dose daily aspirin may be considered for prevention of various cancers, including
gastric cancer, in selected patients (moderate quality evidence, weak recommendation)
Cost-effectiveness
23 After endoscopic resection of early gastric cancer, H. pylori eradication is cost-effective
25 In intermediate to high risk regions, identification and surveillance of patients
with precancerous gastric conditions is cost-effective (moderate quality evidence)
24 Currently available evidence does not allow an accurate estimation of the cost-effectiveness
of surveillance for premalignant gastric conditions worldwide
3 Definitions and prevention aims
3 Definitions and prevention aims
3.1 Gastric carcinogenesis
1 Patients with chronic atrophic gastritis or intestinal metaplasia are at risk for
gastric adenocarcinoma. High quality evidence (100 % agree [94 % strongly or moderately agree]).
2 Histologically confirmed intestinal metaplasia is the most reliable marker of atrophy
in gastric mucosa. High quality evidence (100 % agree [100 % strongly or moderately agree]).
3 Patients with advanced stages of gastritis, that is, atrophy and/or intestinal metaplasia
affecting both antral and corpus mucosa, should be identified as they are considered
to be at higher risk for gastric adenocarcinoma.
Moderate quality evidence, strong recommendation (94 % agree [94 % strongly or moderately
agree]).
4 High grade dysplasia and invasive carcinoma should be regarded as the outcomes to
be prevented when patients with chronic atrophic gastritis or intestinal metaplasia
are managed.
Moderate quality evidence, strong recommendation (100 % agree [100 % strongly or moderately
agree]).
Intestinal-type gastric adenocarcinoma represents the final outcome of the inflammation–atrophy–metaplasia–dysplasia–carcinoma
sequence, known as the Correa cascade [10 ]
[11 ]
[12 ]
[13 ]
[14 ].
Chronic atrophic gastritis and intestinal metaplasia (IM) are considered to be precancerous conditions because they independently confer a risk for development of gastric cancer and constitute
the background in which dysplasia and adenocarcinoma may occur [11 ]
[15 ]
[16 ]
[17 ]. Diverse efforts have been made to stage or classify individuals according to the
severity and/or extent of these changes. Advanced stages of atrophic gastritis should
be defined as significant (moderate to marked) atrophy or as IM (as the best and more
reliable marker of atrophy) affecting both antral and corpus mucosa. In MAPS I, the Operative Link on Gastritis Assessment (OLGA), and Operative Link on
Gastritis Assessment based on Intestinal Metaplasia (OLGIM) systems were proposed
for staging of atrophy and IM, respectively. A large body of evidence, consolidated
in a recent meta-analysis, is now available ascertaining OLGA/OLGIM reliability, with
minor differences between the two systems regarding predictive value for gastric cancer
risk [18 ]. A recent study pointed out that the likelihood for progression to gastric cancer
of high versus low OLGIM stages is two times that of high versus low OLGA stages [19 ]. As the diagnosis of atrophic gastritis needs grading of the severity of gland loss
and this shows poor inter- and intraobserver agreement, we recommend that OLGIM should
be preferred whenever the aim is staging of mucosal changes [19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]. OLGIM can be widely applied with higher accuracy and cost-effectiveness, and also
has lower technical requirements regarding orientation of biopsy samples [23 ]. OLGIM III and IV stages may thus identify patients at a higher risk for gastric
cancer [18 ]
[19 ].
Gastric dysplasia represents the penultimate stage of the gastric carcinogenesis sequence.
It is defined as histologically unequivocal neoplastic epithelium without evidence
of tissue invasion, and is thus a direct neoplastic precancerous lesion
[25 ]. The World Health Organization (WHO) has reiterated the classification of dysplasia/intraepithelial
neoplasia [26 ]:
Intraepithelial neoplasia/dysplasia comprises unequivocally epithelial and neoplastic proliferations characterized by
variable cellular and architectural atypia, but without convincing evidence of invasion.
Low grade intraepithelial neoplasia/dysplasia shows minimal architectural disarray and only mild to moderate cytological atypia.
High grade intraepithelial neoplasia/dysplasia comprises neoplastic cells that are usually cuboidal, rather than columnar, with
a high nucleus-to-cytoplasm ratio, prominent amphophilic nucleoli, more pronounced
architectural disarray, and numerous mitoses, which can be atypical. Importantly,
the nuclei frequently extend into the luminal aspect of the cell, and nuclear polarity
is usually lost. Most patients harboring lesions classified as high grade dysplasia (HGD) are at high risk for either synchronous invasive carcinoma or its rapid development.
Intramucosal invasive neoplasia/intramucosal carcinoma defines carcinomas that invade the lamina propria and are distinguished from intraepithelial
neoplasia/dysplasia not only by desmoplastic changes that can be minimal or absent,
but also by distinct structural anomalies, such as marked glandular crowding, excessive
branching, budding, and fused or cribriform glands. The diagnosis of intramucosal
carcinoma means that there is an increased risk of lymphatic invasion and lymph node
metastasis, although with certain features this risk is absent or minimal.
Guidelines for endoscopic treatment of early gastric cancer (EGC) are beyond the scope
of this manuscript but can be found in published ESGE guidelines [2 ]
[3 ].
3.2 Gastric precancerous and early cancer lesions
5 Patients with an endoscopically visible lesion harboring low or high grade dysplasia
or carcinoma should undergo staging and treatment. High quality evidence, strong recommendation (94 % agree [94 % strongly or moderately
agree]).
In the MAPS I Guideline, we recommended that “Patients with endoscopically visible
high grade dysplasia or carcinoma should undergo staging and adequate management.”
However, several studies have shown that low grade dysplasia (LGD) also has a real
potential for malignancy and, even more importantly, visible lesions with LGD on biopsy
may in fact already be malignant lesions. Moreover, some biopsies may be negative
for dysplasia in the face of a true neoplastic lesion [27 ]. In a Western endoscopic submucosal dissection (ESD) series, there was a histological
upstaging after resection for 33 % of the lesions [28 ]. Similarly, an Eastern study that analyzed 1850 lesions, focusing on the discrepancy
between endoscopy biopsies and endoscopic resection specimens, concluded that the
overall discrepancy rate was 32 % [27 ]. A meta-analysis that specifically investigated the upstaging of gastric LGD after
endoscopic resection found that this happens in 25 % of lesions, with 7 % being upstaged
to malignant [29 ]. Taking all this evidence together, we can conclude that endoscopic biopsies are
insufficient for correct diagnosis of visible gastric lesions and that an endoscopically
visible lesion with any neoplastic change should be considered for treatment.
4 Diagnosis and staging
4.1 Endoscopy
6 High definition endoscopy with chromoendoscopy (CE) is better than high definition
white-light endoscopy alone for the diagnosis of gastric precancerous conditions and
early neoplastic lesions. High quality evidence (94 % agree [94 % strongly or moderately agree.
7 Whenever available and after proper training, virtual CE, with or without magnification,
should be used for the diagnosis of gastric precancerous conditions, by guiding biopsy
for staging atrophic and metaplastic changes and by helping to target neoplastic lesions.
Moderate quality evidence, strong recommendation (94 % agree [94 % strongly or moderately
agree]).
Classical studies of conventional white-light endoscopy (WLE) showed that the correlation
between histological and endoscopic findings for the diagnosis of gastric precancerous
conditions was poor [30 ]
[31 ]
[32 ]
[33 ]
[34 ]. However, recent studies with high definition WLE (HD-WLE) presented promising results.
For preneoplastic conditions, a cross-sectional study showed that HD-WLE had a global
accuracy of 88 % for the diagnosis of IM with a sensitivity of 75 % and specificity
of 94 % [35 ]. In a real-time multicenter prospective study, the global accuracy of HD-WLE was
83 %, with a specificity of 98 % for IM but with only 53 % sensitivity [36 ]. These results were confirmed in another multicenter prospective study, that showed
a 98 % specificity for IM but again with a low sensitivity of 59 % [37 ]. For the diagnosis of neoplastic lesions these two studies showed low sensitivities
of 74 % and 29 %, respectively, although the specificities were higher than 95 % [36 ]
[37 ]. HD-WLE with magnification may improve these results; however, the data are too
scarce to provide definitive conclusions [38 ]
[39 ]
[40 ]. So, even though these results for HD-WLE are satisfactory for IM and for early
neoplastic lesions they are far from perfect, particularly regarding the sensitivity
in the diagnosis of these lesions.
Conventional CE with application of dyes (indigo carmine, methylene blue, acetic acid,
or hematoxylin) has consistently been associated with the detection of gastric preneoplastic
or neoplastic conditions or lesions with high accuracy [41 ]
[42 ]
[43 ]
[44 ]
[45 ]. In a recently published meta-analysis including 10 studies, 699 patients, and 902
lesions, the pooled sensitivity, specificity, and area under the curve (AUC) of dye-CE
were 0.90 (95 % confidence interval [CI] 0.87 – 0.92), 0.82 (95 %CI 0.79 – 0.86),
and 0.95, respectively, these results being significantly better than WLE alone (risk
difference of 0.36 for neoplasia and 0.17 for premalignant conditions) [46 ]. However, dye-CE is cumbersome and significantly lengthens endoscopic procedures.
This favors virtual CE which is available at the touch of a button.
Several studies focused on the role of virtual CE in the diagnosis of gastric precancerous
conditions. A systematic review showed that most studies addressed narrow-band imaging
(NBI) (mainly with magnification). The pooled sensitivity and specificity for the
diagnosis of IM were 86 % and 77 %, and for dysplasia/early cancer these values were
90 % and 83 %, respectively [47 ]. However, the authors concluded that few studies addressed interobserver reliability
and that there was no validated classification. Other investigators evaluated all
the NBI patterns previously described, and created and validated a simplified NBI
(without magnification) classification using only reproducible NBI features ([Fig. 1 ]) [48 ]. The global accuracy for the diagnosis of IM was 84 % and for dysplasia it was 95 %.
However, these results clearly depend on training and are better with experienced
endoscopists [48 ]
[49 ]. External validation of this classification in a prospective multicenter study involving
five international Western centers (some using near-focus and second-generation NBI)
showed a sensitivity and specificity of 87 % and 97 % for the diagnosis of IM and
92 % and 99 % for the diagnosis of dysplasia [36 ]. The diagnostic accuracy rate was of 94 % (11 % higher than HD-WLE), with the greatest
advantage of NBI over or after WLE being sensitivity for detecting IM (87 % vs. 53 %,
P < 0.001) and sensitivity for neoplasia (92 % vs. 74 %) [36 ]. Altogether these results supported the ESGE Technology Review on advanced imaging
which suggested this classification as the one to be used in this context [50 ].
Fig. 1 Simplified narrow-band imaging (NBI) classification for the endoscopic diagnosis
of gastric precancerous conditions and lesions. a Antrum (upper panel) and body (lower panel): normal gastric mucosa. A regular and
circular/oval mucosal pattern with regular thin/peripheral (body) or thick/central
(antrum) vessels is highly predictive of a normal mucosa. b Intestinal metaplasia (IM). Regular vessels with ridge/tubular or tubulovillous glands,
particularly with a light blue crest, are highly suggestive of IM. In general, these
areas of mucosa alternate with areas of normal but atrophic mucosa. c Dysplasia/carcinoma. Irregular vessels and glands (upper panel, high grade dysplasia
lesion), or absent glands with complete architectural loss of the mucosal and vascular
pattern (lower panel, intramucosal adenocarcinoma) predict neoplastic changes of the
mucosa.
Other studies comparing HD-WLE to NBI consistently showed better results with NBI
for detecting both IM and EGC. A large multicenter prospective randomized study reached
the same conclusions this time in an Eastern population. Again, even though specificities
for IM and cancer were the same, the sensitivities for IM (92 % vs. 59 %) and particularly
for cancer (100 % vs. 29 %) were much higher with second-generation NBI when compared
to HD-WLE [37 ]. In an Indian randomized prospective crossover study the conclusions were very similar
regarding IM, with the frequency of IM detection by NBI being significantly higher
than by WLE (P = 0.001) [51 ].
We can conclude that NBI is better than HD-WLE for the detection and diagnosis of
IM, but is it better than standard nontargeted biopsy sampling? In a comparative study
including 119 patients, the overall sensitivity, specificity, and accuracy of WLE
nontargeted biopsies taken according to the Sydney-Houston protocol were compared
with NBI-guided biopsies. For predicting atrophy, the WLE nontargeted biopsies vs.
NBI results were 86 % vs. 62 %, 100 % vs. 97 %, and 93 % vs. 80 %, respectively; for
IM they were 80 % vs. 72 %, 100 % vs. 93 %, and 90 % vs. 82 %. These results were
slightly better for nontargeted protocol biopsies. This was only significant with
respect to the detection of atrophy (P = 0.03) and not for IM or dysplasia [52 ]. A prospective blinded trial detected higher proportions of patients with IM by
NBI-guided (65 %) or WLE nontargeted mapping (76 %) versus HD-WLE-guided biopsies
(29 %; P < 0.005 for both comparisons). In this study the best results would have been obtained
by combining NBI with mapping (detection of 100 % of patients with IM and 95 % of
areas with IM), with NBI identifying different patients and sites with IM that would
not have been detected by mapping alone [53 ]. We can firstly conclude that random biopsies may detect some cases that are not
detected by NBI alone. However, most of these cases will have mild/moderate atrophy
(that has no validated NBI pattern) or mild/focal IM that does not require surveillance.
In fact, it has been shown that, with an appropriately experienced operator, second-generation
NBI may detect almost every case of extensive atrophy/IM without the need of biopsies
[54 ]. Secondly, in expert hands, NBI-guided biopsies may increase the diagnostic yield
of mapping biopsies. When the two modalities are combined, they will detect almost
all cases of gastric precancerous conditions. Finally, it seems clear that atrophic
changes and IM are unevenly distributed throughout the stomach. In this context, an
Endoscopic Grading of Gastric Intestinal Metaplasia (EGGIM) system has been proposed,
and in theory it may allow better staging of gastritis than histology alone since
it takes into account complete assessment of the gastric mucosa [36 ]. However, future studies are needed to validate the EGGIM classification before
routine clinical application.
Regarding EGC diagnosis, other investigators showed that magnifying NBI was better
than HD-WLE with an accuracy of 90 % (vs. 65 %, P < 0.001) [55 ]. A recent meta-analysis also concluded that magnifying NBI is better than WLE alone
with an AUC of 0.96 for diagnosing EGC [56 ]. Two studies (one Eastern, the other Western) compared NBI without magnification
to HD-WLE and suggested increased diagnostic accuracy of NBI over HD-WLE alone [36 ]
[37 ]. In one study, 7 EGC lesions were detected by NBI, but only 2 by HD-WLE [37 ]. In another study, 5 EGC lesions were misdiagnosed by WLE alone, with 2 of them
being only seen with NBI [36 ]. Even though both studies were underpowered for this purpose, they support the estimate
that use of only HD-WLE may miss almost 10 % of neoplastic lesions. This estimate
is in accordance with the 10 % miss rate for gastric cancer observed during upper
GI endoscopy [57 ]
[58 ]. Thus, NBI outperforms HD-WLE for diagnosis and characterization and it may increase
detection of EGC lesions.
There is less evidence to support other methods of virtual CE such as i-Scan digital
contrast and flexible spectral imaging color enhancement (FICE). There are currently
insufficient data to recommend routine clinical use of these techniques, even though
in theory and after proper training they could have similar applications [47 ]
[59 ]. A few prospective studies suggested that blue laser imaging may achieve similar
results to NBI [60 ]
[61 ]
[62 ]. Other emerging technologies, such as confocal endomicroscopy, endocytoscopy, Raman
spectroscopy, and polarimetry, may have a future role but at this stage cannot be
recommended for routine clinical use [59 ].
4.2 Biopsy sampling
8 For adequate staging of gastric precancerous conditions, a first-time diagnostic
upper gastrointestinal endoscopy should include gastric biopsies both for H. pylori infection diagnosis and for identification of advanced stages of atrophic gastritis
Moderate quality evidence, strong recommendation (88 % agree [77 % strongly or moderately
agree]).
9 Biopsies of at least two topographic sites (from both the antrum and the corpus,
at the lesser and greater curvature of each) should be taken and clearly labelled
in two separate vials. Additional biopsies of visible neoplastic suspicious lesions
should be taken. Moderate quality evidence, strong recommendation (94 % agree [82 % strongly or moderately
agree]).
10 Systems for histopathological staging (e. g. OLGA and OLGIM assessment) can be used
to identify patients with advanced stages of atrophic gastritis. If these systems
are used to stratify patients, additional biopsy of the incisura should be considered.
Moderate quality evidence, weak recommendation (88 % agree [58 % strongly or moderately
agree]).
Considering that most endoscopists are not yet familiar with advanced imaging patterns,
at present we cannot recommend exclusively endoscopic staging of gastritis without
biopsies. However, current evidence suggests that CE-targeted biopsies plus mapping
biopsies are the best way of detecting most cases of advanced gastritis. For these
reasons we recommend that, when available, CE should be used for targeted biopsies.
When CE is not available (or the endoscopist doubts the advanced imaging diagnosis),
the number of biopsies needed for correct staging is debated. More biopsies will allow
better staging. However, in clinical practice more biopsies mean more time and higher
procedure costs. In the MAPS I Guideline, we recommended at least two biopsies from
the antrum and two from the corpus, and the lack of obligatory biopsy of the incisura
was a matter of some controversy. In fact, the incisura may be the anatomical location
with the highest incidence and severity of IM [63 ]
[64 ]
[65 ]. This is used to support an additional biopsy of the incisura. The updated Sydney
system is the most widely accepted protocol for the classification and grading of
gastritis. It recommends at least five biopsies: two from the antrum (from the greater
and lesser curvature, 3 cm from the pylorus); one from the incisura; and two from
the body (from the lesser curvature, 4 cm proximal to the incisura, and from the greater
curvature, middle). This differed from the initial Sydney protocol that recommended
only two biopsies from the corpus and two from the antrum [20 ]. However, the need to sample the incisura was based mostly on the notion that atrophic/metaplastic
changes appear first in the incisura even though there were no data suggesting a clinical
benefit. In this regard a large study, published after the MAPS I Guideline, evaluated
400 738 biopsy sets and found that compliance with the original Sydney system (two
antrum, two corpus) had the highest yield for the diagnosis of H. pylori infection and IM when compared with all other biopsy strategies [66 ]. More biopsies or inclusion of an incisura biopsy yielded minimal additional diagnostic
information with more costs. Some studies, published after MAPS I, specifically addressed
the benefit of incisura biopsy sampling. The inclusion of incisura biopsy increased
the proportion of patients classified with high risk stages (OLGA III/IV or OLGIM
III/IV) in three studies (two European studies including nonselected populations [65 ]
[67 ] and one Korean study in high risk patients [64 ]). All found that the incisura biopsy increased the proportion of patients with high
risk stages. Considering the two European studies from nonselected populations together,
without the incisura biopsy, there was a downgrading from high risk OLGA stages to
low risk OLGA in 14/1048 patients (absolute difference 1.33 %) and from high risk
OLGIM to low risk OLGIM in 13/1048 patients (absolute difference 1.24 %). This translates
into a number needed to treat of 75 – 80, meaning that one in 75 – 80 patients will
not be correctly included in a high risk group if incisura biopsy is not performed.
Another European study evaluated classification systems in a high risk population
(first-degree relatives of early onset gastric cancer patients) using OLGA and OLGIM
staging systems that were modified by exclusion of the incisura biopsy, and demonstrated
an overall 15 % and 30 % downgrade of staging in comparison with the original OLGA/OLGIM
systems. In high risk stages, the downgrade of staging was less pronounced (5 %) for
both modified staging systems in comparison with the original OLGA system [68 ]. Another study comparing different biopsy protocols reported that biopsy of the
incisura did not provide additional benefit as the prevalence of IM in the incisura
was similar to that in other biopsy sites, although the impact of incisura biopsies
in high risk phenotypes was not assessed [66 ].
In summary, this small additional yield from an incisura biopsy needs to be balanced
against costs and workload. We therefore recommend a minimum of two biopsies from
the antrum and two biopsies from the corpus, noting that adding an incisura biopsy
can be considered in order to maximize the detection of patients with precancerous
conditions, especially in cases where CE is not available to target biopsies. Moreover,
this additional biopsy will allow more precise evaluation of OLGA and OLGIM stages,
that have been proven to correlate with risk for cancer progression [69 ]
[70 ]
[71 ].
Regarding the number of vials, even though separate vials may not be required among
expert pathologists, as antral and corpus mucosa can be easily distinguished in the
absence of severe atrophic changes, use of a single vial cannot be recommended in
all cases. Future studies should evaluate specific scenarios when antrum, incisura,
and corpus samples can be sent in the same vial.
4.3 Noninvasive assessment
11 Low pepsinogen I serum levels or/and a low pepsinogen I/II ratio identify patients
with advanced stages of atrophic gastritis, and endoscopy is recommended for these
patients, particularly if H. pylori serology is negative.
Moderate quality evidence, strong recommendation (88 % agree [76 % strongly or moderately
agree]).
As stated in the MAPS I Guideline, a low pepsinogen I serum level, a low pepsinogen
I/II ratio, or both, are good indicators of atrophic changes in the gastric mucosa.
A 2004 meta-analysis suggested that pepsinogen I ≤ 50 ng/mL and pepsinogen I/II ratio
≤ 3 were the best cutoff values for dysplasia diagnosis [72 ]. Several articles published after MAPS I confirm levels of pepsinogens to be good
indicators of extensive atrophic gastritis and of gastric cancer [73 ]
[74 ]. A 2015 meta-analysis on pepsinogen tests in gastric cancer and atrophic gastritis
suggested a good correlation between decreased pepsinogen serum levels and atrophy
[75 ]. In this meta-analysis, the summary sensitivity and summary specificity for gastric
cancer diagnosis were 0.69 (95 %CI 0.60 – 0.76) and 0.73 (95 %CI 0.62 – 0.82), respectively.
Corresponding values for atrophic gastritis diagnosis were 0.69 (95 %CI 0.55 – 0.80)
and 0.88 (95 %CI 0.77 – 0.94), respectively. The AUC for gastric cancer diagnosis
was 0.76 (95 %CI 0.72 – 0.80) and for atrophic gastritis it was 0.85 (95 %CI 0.82 – 0.88).
A Fagan plot indicated that the use of pepsinogen serum levels could moderately improve
the gastric cancer and atrophy detection rate, confirming a moderate efficiency of
pepsinogen serum levels for gastric cancer and atrophic gastritis diagnosis. In a
subgroup analysis the authors concluded that combining low pepsinogen I level with
the pepsinogen I/II ratio is the best way of detecting gastric cancer (AUC 0.78) and
atrophic gastritis (AUC 0.87). However, different cutoff values were used, although
most studies used pepsinogen I < 70 ng/mL and pepsinogen I/II ratio < 3 as the best
cutoff values. In fact, these are widely accepted cutoff values for gastric cancer
screening in Japan [76 ]. The authors concluded that pepsinogen serum levels have a potentially significant
role in the identification of populations at high risk for gastric cancer and could
be used for mass screening. However, they note that there was great heterogeneity
between studies. Moreover, different methods are used for quantifying levels of pepsinogens
and in this meta-analysis enzyme-linked immunoassay (ELISA) was slightly superior
to the other methods, with this difference possibly inducing heterogeneity [75 ]. In fact, different methods may be used for pepsinogen quantification and results
may differ between tests [77 ]. Therefore, cutoff values validated for a particular assay should be used, and cannot
be generalized to all assays.
Other serum molecule levels were studied as markers of gastric atrophy. A 2017 systematic
review and meta-analysis focused on the combination of pepsinogen I/II, gastrin-17,
and anti-Helicobacter antibodies for diagnosing atrophic gastritis [78 ]. However, the design of this meta-analysis does not allow assessment of the individual
performance of each marker for detecting atrophy. Moreover, previously published evidence
demonstrated little yield from adding gastrin-17 to pepsinogen assessment for detecting
atrophy [79 ]. On the other hand, adding H. pylori serology to pepsinogen level evaluation may help to detect patients at higher risk
of gastric cancer [80 ]
[81 ]. In a 2014 cohort of 4655 patients followed up for 16 years, there was a progressive
increase in cancer risk, going from those with no gastritis to those with chronic
H. pylori -positive gastritis without extensive atrophy (H. pylori -positive, normal pepsinogen levels; hazard ratio [HR] 8.9, 95 %CI 2.7 – 54.7), to
those with extensive chronic atrophic gastritis (defined by pepsinogen I < 70 ng/mL
and pepsinogen I/II ratio < 3) with H. pylori -positive serology (HR 17.7, 95 %CI 5.4 – 108), and finally to those with atrophic
gastritis with H. pylori -negative serology, suggestive of extensive IM (HR 69.7, 95 %CI 14 – 503) [81 ].
Other methods for noninvasive assessment of gastric mucosal atrophy, including evaluation
of decreased serum ghrelin [82 ]
[83 ]
[84 ], trefoil factors [85 ], a panel of microRNAs [86 ], and volatile organic compounds in exhaled air [87 ], have been suggested, with good results. However, the available evidence for these
tests is not sufficient and further studies are required before they can be recommended
for clinical application.
In conclusion, pepsinogen serum levels are currently the best evaluated noninvasive
test for detecting patients with advanced atrophic gastritis. Low pepsinogen I serum
levels, particularly when associated with H. pylori -negative serological status, may identify patients at higher risk of gastric cancer
to whom endoscopy should be offered.
4.4 Additional risk factors
12 Even though diverse studies assessed age, gender, and H. pylori virulence factors, as well as host genetic variations, no clinical recommendations
regarding diagnosis and surveillance can be made for targeted management based on
these factors. Low quality evidence, weak recommendation (100 % agree [88 % strongly or moderately
agree]).
Assuming the gene-environment interaction for gastric cancer, multiple risk factors
have been linked to the multistep progression from chronic nonatrophic gastritis to
atrophic gastritis, IM, dysplasia, and finally cancer [10 ].
H. pylori plays a pivotal role in this progression and was classified as a type 1 carcinogen
in 1994 by the WHO [88 ]. It is believed that the combination of a virulent organism in a genetically susceptible
host is associated with more severe chronic inflammation and more rapid progression
to gastric cancer, at least for the Lauren intestinal type [89 ]
[90 ]
[91 ].
Different strains of H. pylori vary in their carcinogenic potential, with those containing virulence factors, such
as the cytotoxin-associated antigen (cagA) protein and the vacuolating toxin A (vacA),
inducing a higher degree of inflammation and increasing the risk for gastric cancer
[92 ]
[93 ]
[94 ]
[95 ]
[96 ]
[97 ]. Nevertheless, there are no studies addressing the clinical usefulness of genotyping
H. pylori strains with regard to the management and surveillance of gastric precancerous conditions/lesions.
An immense number of studies have addressed the implications of genes and genetic
host variations for gastric carcinogenesis. The best characterized are those that
play a role in the inflammatory response to H. pylori infection and inflammation of the gastric mucosa, leading to mucosal atrophy and
progression to cancer. These include host genetic interleukin polymorphisms of IL-1B,
IL1-receptor antagonist (IL-1RN), IL8, IL10, and TNF-α [98 ]
[99 ]
[100 ]
[101 ]
[102 ]
[103 ]
[104 ]. However, the heterogeneity of the results makes it difficult to translate them
into recommendations for daily clinical practice.
5 Surveillance
5.1 Dysplasia
13 In patients with dysplasia in the absence of an endoscopically defined lesion immediate
high quality endoscopic reassessment with CE (virtual or dye-based) is recommended.
If no lesion is detected in this high quality endoscopy, biopsies for staging of gastritis
(if not previously done) and endoscopic surveillance within 6 months (if high grade
dysplasia) to 12 months (if low grade dysplasia) are recommended.
Low quality evidence, strong recommendation (88 % agree [88 % strongly or moderately
agree]).
Most routine gastroscopies are performed with standard definition WLE. As we have
seen, CE (virtual or dye-based) increases accuracy for detection of dysplasia. A prospective
study that included 20 patients with a diagnosis of HGD or carcinoma, without visible
endoscopic lesions in the index endoscopy, showed that immediate endoscopic reassessment
with high definition endoscopes and virtual CE allowed the identification of visible
lesions and adequate treatment in 18 patients [105 ]. Conventional CE also improves the detection of precancerous conditions and lesions
[106 ]. A systematic review and meta-analysis reported that approximately 10 % of the patients
with a gastric cancer diagnosis had undergone a recent endoscopy in which the gastric
cancer was not diagnosed (because of both missed endoscopic lesions or nonmalignant
pathology diagnosis). A recent study also showed that 8.6 % of the patients with EGCs
had a simultaneous lesion that was not detected in the diagnostic endoscopy [57 ]
[107 ]. The rate of missed lesions tended to be higher in primary care and screening settings
than in secondary and tertiary care. Another study showed that a longer endoscopy
time (> 7 minutes) was associated with a higher likelihood of detecting neoplastic
lesions (odds ratio [OR] 3.42, 95 %CI 1.25 – 10.38) [108 ]. Moreover, a finding of dysplasia in nontargeted biopsies significantly increases
the risk of gastric cancer, which may be as high as 6 % per year [109 ]. A recent Swedish study, that to the best of our knowledge is the largest follow-up
study to date among patients with gastric precancerous conditions, suggested a lower
risk of gastric cancer for patients with dysplasia. However, they excluded the first
2 years of follow-up and concluded that this might be the reason for the lower risk
of gastric cancer since many lesions might have been there already [110 ].
“Indefinite for dysplasia/neoplasia” should not be viewed initially as an innocuous
diagnosis although in the majority of patients the prognosis is favorable. Indeed,
a study found that 26.8 % of resected lesions that had been characterized as indefinite
for dysplasia/neoplasia in preresection biopsies were in fact neoplastic (5.0 % adenomas
and 21.8 % EGCs) [111 ]. Another study found that reassessment of indefinite for dysplasia biopsies by three
expert gastrointestinal pathologists changed the diagnosis to dysplasia in 11 /46
patients (10 LGD and 1 HGD) [112 ].
All of this suggests that patients with diagnoses from nontargeted biopsies of indefinite
for dysplasia, of dysplasia, or of carcinoma benefit from a careful endoscopic reassessment
in centers with experience in the diagnosis and endoscopic treatment of EGC. We recommend
that pathology slides should be reviewed by an expert GI pathologist and recommend
immediate (as soon as possible) high quality endoscopic reassessment with CE. If a
lesion is seen and the endoscopic assessment suggests dysplasia, we recommend resection
without need of further biopsies. If endoscopic reassessment with CE does not reveal
a visible lesion and repeat nontargeted biopsies do not show dysplasia/neoplasia,
then staging the severity and extent of preneoplastic conditions in such cases can
help to define the surveillance program. A retrospective study of patients with indefinite
for dysplasia lesions at enrollment and OLGA staging, with a median follow-up of 31
months, did not detect dysplasia in any patient with OLGA 0 /I/II, while 6 cases of
LGD/HGD were detected in 25 patients with OLGA III/IV during follow-up [113 ].
With the above considerations in mind, patients with a diagnosis of indefinite for
dysplasia/neoplasia or of dysplasia/intramucosal carcinoma in random biopsies (i. e.,
no clear lesion identified at endoscopy) should be promptly referred to an expert
endoscopy center and have an endoscopic reassessment with high definition endoscopes
and CE (dye or virtual). If no lesion is identified in this high quality endoscopy,
endoscopic revaluation is recommended at a 6-month (if previous HGD) to 12-month (if
previous LGD) interval, with further adjustment according to the severity and extent
of precancerous conditions ([Fig. 2 ]).
Fig. 2 Proposed management for patients with atrophic gastritis, gastric intestinal metaplasia,
or gastric epithelial dysplasia. OLGA, Operative Link on Gastritis Assessment; OLGIM,
Operative Link on Gastritis Assessment based on Intestinal Metaplasia.
1 Advanced stages of atrophic gastritis warranting surveillance should be defined as
significant (moderate to marked) atrophy or intestinal metaplasia (IM) affecting both
antral and corpus mucosa or as OLGA/OLGIM stages III/IV. Mild atrophy without IM,
even when affecting antrum and corpus, should not be considered to be an advanced
stage of gastritis.
2 First-degree family history of gastric cancer is an important risk factor for gastric
cancer and even though the evidence is scarce these patients may benefit from a more
intensive follow-up. These recommendations do not apply to hereditary/familial diffuse
gastric cancer.
3 When reported, incomplete IM may identify patients with a higher risk of gastric cancer.
However, additional studies are required before subtyping can be routinely recommended.
4 After diagnosis of dysplasia, revision of pathology slides by an expert gastrointestinal
pathologist should be considered, particularly when no lesion is seen after a high
quality endoscopy. If expert revision does not confirm the diagnosis of dysplasia
then the patient may be released from intensive follow-up.
5.2 Atrophic gastritis/intestinal metaplasia
14 For patients with mild to moderate atrophy restricted to the antrum there is no evidence
to recommend surveillance.
Moderate quality evidence, strong recommendation (100 % agree [100 % strongly or moderately
agree]).
15 Patients with IM at a single location have a higher risk of gastric cancer. However,
this increased risk does not justify surveillance in most cases, particularly if a
high quality endoscopy with biopsies has excluded advanced stages of atrophic gastritis.
Moderate quality evidence, strong recommendation (100 % agree [82 % strongly or moderately
agree]).
16 In patients with IM at a single location but with a family history of gastric cancer,
or with incomplete IM, or with persistent H. pylori gastritis, endoscopic surveillance with CE and guided biopsies in 3 years’ time may
be considered.
Low quality evidence, weak recommendation (82 % agree [76 % strongly or moderately
agree]).
17 Patients with advanced stages of atrophic gastritis (severe atrophic changes or IM
in both antrum and corpus, OLGA/OLGIM III/IV) should be followed up with a high quality
endoscopy every 3 years.
Low quality evidence, strong recommendation (100 % agree [94 % strongly or moderately
agree]).
18 Patients with advanced stages of atrophic gastritis and with a family history of
gastric cancer may benefit from a more intensive follow-up (e. g. every 1 – 2 years
after diagnosis).
Low quality evidence, weak recommendation (82 % agree [65 % strongly or moderately
agree]).
Gastric precancerous conditions are frequent in the general population (although with
wide geographical variability according to H. pylori infection prevalence). The annual incidence of gastric cancer has been reported to
be 0.1 % – 0.25 % in patients with chronic atrophic gastritis and 0.25 % in patients
with IM, and may be as high as 1.36 % person-year for any gastric neoplasia (including
dysplasia and neuroendocrine tumors) [109 ]
[114 ]. Cumulative incidences of gastric cancer of 2.4 % at 10 years in patients with IM
were reported, and a Swedish study reported a cumulative incidence at 20 years of
approximately 2 % in patients with atrophic gastritis and of 2.5 % in patients with
IM [110 ]. A Japanese study found higher cumulative incidences of gastric cancer at 5 years,
reaching 1.9 % – 10 % in patients with extensive endoscopic atrophy and 5.3 % – 9.8 %
in patients with IM [115 ].
Surveillance of patients with precancerous conditions allows the detection of lesions
at early stages (with a significant proportion being amenable to endoscopic resection)
and was recommended in the MAPS I Guideline in patients with extensive atrophy or
IM (in both corpus and antrum). The extent of preneoplastic changes was identified
as a risk factor for progression, as well as family history of gastric cancer and
type III incomplete IM.
Extent and presence of IM Some recent studies confirmed the presence and extent of IM as risk factors for gastric
cancer. An Italian prospective study found a significantly increased risk of gastric
neoplasia in patients with OLGA and OLGIM stages III/IV at baseline, while extensive
atrophy (antrum and corpus) was also associated with a trend to higher risk of progression
although this was not statistically significant on multivariable analysis (HR 7.2,
95 %CI 0.7 – 6.84) [114 ]. Extensive IM was also found to be associated with a higher risk of progression
in a US study [116 ]. A Japanese study also found that IM in the corpus (isolated or antrum and corpus)
and extensive endoscopic atrophy at baseline were independent predictors of gastric
cancer at follow-up [115 ]. A case-control study found that OLGIM II-IV (but not OLGA II-IV) and corpus-predominant
gastritis were significantly more frequent in gastric cancer patients than in controls
[117 ]. Another case-control study found that OLGA III/IV, OLGIM III/IV, and endoscopically
classified moderate-to-severe atrophy were significantly more frequent in gastric
cancer patients [118 ]. Another study using endoscopic grading of atrophy found that gastric cancer risk
was increased in patients with extensive atrophy (5.33 % in patients with atrophy
present in the entire stomach vs. 0 % and 0.25 % in patients with atrophy limited
to the gastric antrum and atrophy in the incisura or lower corpus, respectively) [119 ]. A Korean study reported that OLGA III/IV and OLGIM I – IV were independent risk
factors for gastric cancer, especially the intestinal type, showing that even nonextensive
IM may significantly increase the risk of gastric cancer [120 ]. The adjusted odds ratios for the different stages were: OLGA III 2.09, OLGA IV
2.04; OLGIM I 2.38, OLGIM II 2.97, OLGIM III 7.89, OLGIM IV 13.20 (all statistically
significant). OLGA IV, histological IM, and a higher classification of endoscopic
atrophy were also identified as independent risk factors in a prospective Korean study
with follow-up > 3 years [121 ]. These studies suggest that the presence of IM (as a surrogate of advanced gastritis)
may be of equal or more importance than the extent of atrophy without IM, since the
risk of gastric cancer was higher with OLGIM I/II than with OLGA III/IV. This accords
with other previous studies that evaluated the risk of gastric cancer in patients
with only atrophy or IM (independently of extent), and which showed that the risk
of gastric cancer is higher in IM patients (not considering extent) than in patients
with atrophy [109 ]. In agreement, a recent study in Sweden that analyzed more than 400 000 patients
concluded that IM (independently of extent) significantly increases the risk of gastric
cancer. Interestingly, it showed that a second endoscopic surveillance with biopsies
can have significant prognostic value, since downgrading of gastritis (to no IM detected)
is associated with less risk of progression to cancer (and then these patients may
not benefit from follow-up) [110 ].
Nevertheless, the prevalence of focal IM in the population may be as high as 25 %
and it seems unreasonable to follow up all of these patients [122 ]. Moreover, even though the present authors recognize that focal IM may increase
the risk of gastric cancer compared to no IM or even to only atrophy, this risk appears
too small to justify surveillance [19 ]. On the other hand, extensive IM significantly increases the risk of gastric cancer
compared to focal IM, and in this scenario, surveillance is recommended.
Other factors may influence the risk for cancer:
Incomplete IM A Spanish prospective multicenter study with a mean follow-up of 12 years found that
incomplete IM was associated with a significantly higher risk of gastric cancer when
compared with complete IM (HR 2.57, 95 %CI 1.06 – 6.26) [123 ]. A systematic review from the same authors also reported that in 10 follow-up studies,
incomplete type III IM was associated with significantly higher risk of gastric cancer
in 6 studies, with a 6 – 11-fold higher risk [124 ]. A recent study with a follow-up of 16 years also showed that incomplete-type IM
was associated with a higher risk of progression to cancer than the complete type
(OR 11.3, 95 %CI 1.4 – 91.4) [19 ]. These findings suggest that incomplete IM is associated with a risk of progression
similar to that attributed to extensive atrophy or family history of gastric cancer.
For these reasons, when reported, this information can have prognostic value and can
aid in the selection of patients for surveillance. However, incomplete IM is not always
found in the gastrectomy specimens of gastric cancer patients [125 ]
[126 ]
[127 ]. Additional studies are required before subtyping can be routinely recommended.
Family history Although most gastric cancers are sporadic, some kind of familial aggregation occurs
in 10 % of cases [128 ]. Having a first-degree relative with gastric cancer is a consistent risk factor
for gastric cancer, with an odds ratio varying from 2 to 10 in relation to geographic
region and ethnicity [129 ]. Importantly, adjustment for environmental factors does not alter this risk. Having
a second-degree relative with gastric cancer also confers a higher risk of development
of the disease, but to a lesser extent [130 ]. It is believed that this familial clustering of gastric cancer is due to an inherited
genetic susceptibility, shared environmental or lifestyle factors, shared susceptibility
to H. pylori , sharing the same cytotoxic H. pylori strain, or a combination of these factors. Accordingly, a meta-analysis showed that
first-degree relatives of gastric cancer patients have an increased prevalence of
H. pylori infection (OR 1.93), gastric atrophy (OR 2.2) and IM (OR 1.98) [131 ]. Also, first-degree relatives of early-onset gastric cancer patients have increased
prevalences of high stage gastritis (OLGA stage III/IV) and dysplasia that seem to
be associated with high virulence H. pylori strains and pro-inflammatory host genotypes [68 ]
[132 ].
Thus, these data show that first-degree relatives of gastric cancer patients have
an increased prevalence of H. pylori infection and precancerous conditions/lesions, as well as an increased risk for gastric
cancer.
Regarding progression of precancerous conditions, a US study found an increased risk
for progression in patients with IM and a family history of gastric cancer (P = 0.002) [116 ]. In an Italian cohort, family history was also associated with a higher risk for
progression in patients with gastric atrophy although this was not statistically significant
[114 ]. Although there is only scarce evidence that precancerous conditions in relatives
of a gastric cancer patient progress more rapidly through the carcinogenic cascade
to cancer than similar conditions in matched controls in a general population, it
seems reasonable to recommend a more intensive follow-up in patients with extensive
atrophy/IM and a first-degree family history of gastric cancer.
In sum, there is a significantly higher risk of progression to cancer in patients
with dysplasia, extensive atrophy/IM, and/or OLGA/OLGIM stage III/IV, and we recommend
endoscopic surveillance of these patients, ideally by a high quality endoscopy. However,
the risk of gastric cancer is also increased, even though with a lower magnitude,
in patients with less advanced stages of preneoplastic change, such as those with
focal IM (OLGIM I/II), particularly if there is also incomplete IM and/or a family
history of gastric cancer. There are also some practical problems related to the adequate
staging of precancerous conditions in routine clinical practice, since an important
proportion of gastroscopies are performed with standard definition WLE (nontargeted
biopsies) and the adherence to biopsy protocols is variable. We thus recommend (based
on expert opinion) that patients with only nonguided biopsies at the antrum showing
IM should be reassessed after 3 years with biopsies from antrum and corpus in separate
vials, ideally with HD-NBI to allow targeted biopsies and restaging, if this was not
previously done. If this high quality endoscopy excludes extensive IM then these patients
may be released from endoscopic surveillance. Exceptions may be cases of family history
of gastric cancer, incomplete IM in biopsies, and persistent H. pylori .
The benefit of doing nontargeted biopsies in patients under surveillance and already
with correct staging of gastritis has not been established. For this reason, for patients
with an indication for surveillance we recommend a high quality endoscopy with CE
and biopsies of only the irregular/suspicious for dysplasia areas with no further
biopsies being needed ([Fig. 2 ]).
Regarding the schedule for surveillance, the MAPS I Guideline recommended, based on
expert opinion, surveillance every 3 years in patients at higher risk for gastric
cancer. A recent prospective cohort study supports this recommendation, showing a
neoplasia risk of 36.5 per 1000 person-years in OLGA III patients (95 %CI 13.7 – 97.4)
and 63.1 per 1000 person-years in OLGA IV patients (95 %CI 20.3 – 195.6) [133 ]. The authors suggested that the best follow-up surveillance interval would be 2
instead of 3 years. However, the cost-effectiveness of a 2-year interval for every
patient may not be ideal and the evidence is not strong enough to change the recommended
3-year surveillance interval. Nevertheless, the present authors recognize that patients
with extensive IM, and also with at least one of persistent H. pylori infection, incomplete IM, or, particularly, a first-degree family history of gastric
cancer, may benefit from a tighter endoscopic surveillance schedule (e. g. every 1 – 2
years). These recommendations do not apply to hereditary/familial diffuse gastric
cancer, for which there are specific guidelines [134 ].
5.3 Autoimmune gastritis
19 Patients with autoimmune gastritis may benefit from endoscopic follow-up every 3 – 5
years. Low quality evidence, weak recommendation (82 % agree [76 % strongly or moderately
agree]).
Autoimmune gastritis is a chronic progressive inflammatory condition that results
in the replacement of the parietal cell mass by atrophic and metaplastic mucosa, leading
to a corpus-predominant atrophic gastritis, reduced or absent acid production, and
loss of intrinsic factor which may progress to a severe form of vitamin B12-deficiency
anemia known as pernicious anemia. Both gastric carcinoma and neuroendocrine tumors
are the most dreaded long-standing complications of pernicious anemia.
Most of the evidence on the risk of gastric cancer associated with pernicious anemia
comes from case-control [135 ]
[136 ] and cohort studies [114 ]
[137 ]
[138 ]
[139 ]
[140 ]
[141 ]
[142 ]
[143 ]
[144 ]. One study based on the Surveillance, Epidemiology, and End Results (SEER) database
compared 1 138 390 pernicious anemia cases to 100 000 matched individuals [135 ]. Individuals with pernicious anemia were at increased risk for noncardia gastric
adenocarcinoma (OR 2.18, 95 %CI 1.94 – 2.45) and gastric carcinoid tumors (OR 11.43,
95 %CI 8.90 – 14.69). However, the diagnosis of autoimmune gastritis in this study
was rather flawed as it was solely based on low levels of vitamin B12 [145 ]. Therefore, many of the patients supposedly with autoimmune gastritis probably had
other causes of low serum vitamin B12, and the risk of cancer for genuine autoimmune
gastritis patients was likely underestimated. A Swedish study followed 21 265 patients
with pernicious anemia for an average of 7.1 years [138 ]. These patients had a significant excess risk for gastric cancer distal to the cardia
(standardized incidence ratio [SIR] 2.4, 95 %CI 2.1 – 2.7). The excess risks increased
with increasing follow-up duration. Among distal gastric cancers, the most conspicuous
excess risk was for carcinoid tumors (SIR 26.4, 95 %CI 14.8 – 43.5). The abovementioned
criticism with respect to the diagnosis of autoimmune gastritis also pertained to
this study.
A recent meta-analysis with 27 studies and a total of 22 417 patients showed that
the calculated pooled gastric cancer incidence rate was 0.27 % per person-year and
the overall gastric cancer relative risk in pernicious anemia was 6.8 (95 %CI 2.6 – 18.1)
[146 ]. The drawback of this meta-analysis is again that many patients included in these
studies may have had low vitamin B12 serum levels because of conditions other than
autoimmune gastritis.
Therefore, there is some evidence suggesting that autoimmune gastritis is a precancerous
condition that may justify endoscopic monitoring. Nevertheless, there is no recommended
follow-up interval to date.
Since the largest excess risk of gastric cancer incidence among patients with pernicious
anemia has been found during the first year of follow-up [141 ]
[143 ], there is evidence to recommend endoscopic screening to all patients at the time
of the diagnosis.
Several cohort studies prospectively evaluated the risk of gastric cancer in patients
with pernicious anemia, with varying follow-up intervals from 3 to 7 years [114 ]
[139 ]
[140 ]
[142 ]
[147 ]
[148 ]
[149 ]. One study [147 ] performed follow-up gastroscopies 3 years after primary screening examination of
56 patients and identified on follow-up 2 patients with gastric adenocarcinoma, no
patient with HGD, and 49 patients with IM. Another study [142 ] followed up a group of 27 patients for 6 to 7 years after initial investigation.
None of the patients had developed gastric cancer since the initial endoscopy and
the distribution of dysplasia was virtually unchanged. The only randomized controlled
trial (RCT) to determine the most effective time interval for the first follow-up
endoscopy after diagnosis of corpus-predominant atrophic gastritis randomly assigned
24 patients to a 24- or 48-month follow-up interval [148 ]. No gastric cancer was found in either group, but a patient from the 48-month group
developed a neuroendocrine tumor. The authors concluded that the first follow-up need
not be earlier than 4 years after diagnosis, with this interval being satisfactory
for detection of potential neoplastic lesions. Considering the heterogeneity of the
described cohorts and the absence of larger RCTs with longer follow-up, we recommend
follow-up endoscopy at 3- to 5-year intervals in patients with autoimmune gastritis.
6 Therapy
6.1 Helicobacter pylori eradication
20 H. pylori eradication heals nonatrophic chronic gastritis, may lead to regression of atrophic
gastritis, and reduces the risk of gastric cancer in patients with nonatrophic and
atrophic gastritis, and, therefore, it is recommended in patients with these conditions.
High quality evidence, strong recommendation (87 % agree [87 % strongly or moderately
agree]).
21 In patients with established IM, H. pylori eradication does not appear to significantly reduce the risk of gastric cancer, at
least in the short term, but reduces inflammation and atrophy and, therefore, it should
be considered.
Low quality evidence, weak recommendation (87 % agree [75 % strongly or moderately
agree]).
22 H. pylori eradication is recommended for patients with gastric neoplasia after endoscopic therapy.
High quality evidence, strong recommendation (100 % agree [100 % strongly or moderately
agree]).
Since publication of the MAPS I Guideline, three meta-analyses have been performed
regarding the effect of H. pylori eradication on chronic gastritis and risk of gastric cancer [150 ]
[151 ]
[152 ]. The first meta-analysis included only prospective trials and RCTs on H. pylori eradication with a focus on histology (before and after treatment) and not on the
risk of gastric cancer [151 ]. The authors concluded that IM in the antrum and atrophic gastritis in both the
antrum and corpus regressed after eradication of H. pylori , although this effect was not seen for IM in the corpus. This meta-analysis was statistically
more powerful than previous ones on this subject and strongly suggests that H. pylori eradication halts progression of precancerous conditions even after IM has appeared.
In fact, when studies with a longer follow-up (> 5 years) and with larger groups are
analyzed, they do show a statistical improvement for IM both at the antrum and at
the corpus after H. pylori eradication therapy has been received [19 ]
[153 ]
[154 ]
[155 ]. One study even showed no statistically significant difference with regard to IM
in comparison to an H. pylori -negative group, for the corpus 3 years after H. pylori eradication and for the antrum 5 years after H. pylori eradication [153 ]. Despite the possibility of sampling error, it appears logical that for a lesion
that occurs after decades of infection, reversion also only occurs after a very long
period and the risk of gastric cancer may also only decrease in the long but not the
short term.
Both meta-analyses that focused on the risk of gastric cancer after H. pylori eradication concluded that H. pylori eradication significantly decreases the risk of gastric cancer in patients with chronic
atrophic or nonatrophic gastritis (pooled relative risk [RR] 0.64, 95 %CI 0.48 – 0.85)
but not in patients with IM or dysplasia (RR 0.88, 95 %CI 0.59 – 1.31) [150 ]
[152 ]. However, only a few of the studies included in these meta-analyses had a long follow-up
period (more than 10 years).
In conclusion, there is strong evidence suggesting that H. pylori eradication is highly beneficial in patients with chronic nonatrophic and atrophic
gastritis, both histologically and in reducing gastric cancer risk. At later stages
of gastritis (established IM) weaker evidence suggests that H. pylori eradication has beneficial histological effects, with no conclusive effect, however,
on gastric cancer risk reduction. Nevertheless, no study suggested that H. pylori eradication has negative effects on patients with IM and, so, considering the positive
histological effects of H. pylori eradication, it is the opinion of the present authors that H. pylori eradication should also be offered to patients with IM. It is also important to note
that H. pylori infection is now considered an infectious disease and eradication is recommended
in most cases, regardless of the presence of precancerous conditions [156 ].
There is also controversy about H. pylori eradication therapy after endoscopic removal of gastric superficial neoplasia. After
publication of the MAPS I Guideline, a multicenter retrospective study including six
Japanese centers and 268 patients contradicted its recommendation in favor of such
treatment [157 ]. In that study, even though metachronous gastric cancer developed in 14.3 % versus
8.5 % of the patients of the H. pylori -persistent vs. the H. pylori -eradicated group, the baseline severity of mucosal atrophy and a follow-up of more
than 5 years were the only independent risk factors for metachronous neoplasia [157 ]. A prospective RCT that included 901 patients failed to show that H. pylori eradication reduced the risk of metachronous lesions (2.2 % treated vs. 3.7 % nontreated,
P = 0.15) [158 ]. This trial contradicted another previously published RCT on this subject that showed
a significant reduction of gastric cancer in H. pylori -eradicated patients [159 ]. Moreover, in another retrospective study that included 2089 patients who underwent
endoscopic resection of a superficial lesion, the incidence of metachronous gastric
cancer was 10.9 cases per 1000 person-years in the H. pylori -negative group, 14.7 cases per 1000 person-years in the H. pylori -eradicated group, and 29.7 cases per 1000 person-years in the group without H. pylori eradication (HR 1.9 when compared to H. pylori -eradicated group, P = 0.02) [160 ]. Two meta-analyses on this subject included the same 10 studies (8 nonrandomized,
2 randomized), but with 5914 and 5881 patients because of different inclusion criteria.
They reached the identical conclusion that H. pylori eradication reduces the risk of metachronous lesions with a risk ratio of 0.467 (95 %CI
0.362 – 0.602; P < 0.001) [161 ]
[162 ]. Finally, a 2018 double-blind, placebo-controlled RCT that included 396 patients
in the intention-to-treat analysis conclusively showed that H. pylori eradication in this group of patients reduced the risk of metachronous lesions to
almost half (7 % vs. 13 %; HR 0.5, 95 %CI 0.26 – 0.94) [163 ].
In summary, H. pylori eradication has the largest impact on gastric cancer risk in patients with nonatrophic
gastritis and early stages of atrophy. Nevertheless, a small benefit of eradication
is still seen at later stages of gastritis and even after resection of a lesion. For
this reason, we recommend that H. pylori eradication should always be considered.
6.2 Other therapies
23 Even though cyclo-oxygenase (COX)-1 or COX-2 inhibitors may slow progression of gastric
precancerous conditions, they cannot be recommended specifically for this purpose.
Low quality evidence, weak recommendation (100 % agree [94 % strongly or moderately
agree]).
24 Low dose daily aspirin may be considered for prevention of various cancers, including
gastric cancer, in selected patients.
Moderate quality evidence, weak recommendation (94 % agree [47 % strongly or moderately
agree]).
COX inhibitors Earlier meta-analyses (2003, 2010) had suggested a lower risk of gastric cancer in
users of COX-inhibitors [164 ]
[165 ]. Since the publication of MAPS I, few studies have specifically addressed this issue
but all added evidence supporting this position. In 2013, a prospective nonrandomized
study reported on the role of selective COX-2 inhibitor treatment in patients with
precancerous gastric conditions: after 1 year of treatment with celecoxib following
H. pylori eradication, IM regression was more frequent in the treatment versus the control
group (44.3 % vs. 14.3 %, total 140 patients) [166 ]. Other studies suggest that inhibition of COX may slow progression of gastric precancerous
conditions and in theory may decrease the risk of gastric cancer development. A double-blind
RCT, including 1024 participants who received H. pylori eradication treatment or placebo followed by celecoxib or placebo (i. e., four different
groups studied), showed that regression of gastric precancerous conditions significantly
increased both in the eradication group (59 % vs. 41 % placebo) and in the celecoxib
group (53 % vs. 41 % placebo) with an OR of 1.72 (95 %CI 1.07 – 2.76) for celecoxib
and 2.19 (95 %CI 1.32 – 3.64) for H. pylori eradication [167 ]. However, in this study no statistically significant benefit was observed for celecoxib
after H. pylori eradication. Moreover, 9 cancers developed in this study but it was underpowered
to show a therapeutic benefit of any strategy.
Nonsteroidal anti-inflammatory drugs (NSAIDs) Several meta-analyses investigating the role of aspirin and other NSAIDs on the risk
of gastric cancer have accumulated, all of them showing a favorable effect [168 ]
[169 ]
[170 ]
[171 ]
[172 ]
[173 ]. In the most recent meta-analysis, which included 24 studies, both nonaspirin NSAIDs
(RR 0.86, 95 %CI 0.80 – 0.94) and aspirin (RR 0.70, 95 %CI 0.62 – 0.80) significantly
reduced noncardia gastric cancer risk [168 ]. However, the vast majority of original studies did not include patients with precancerous
gastric conditions. Moreover, the protective effect of aspirin/NSAIDs was more marked
for noncardia gastric cancer and in H. pylori -positive individuals. In fact, when only H. pylori -negative patients were considered, the effect of aspirin was nonsignificant (RR 0.81,
95 %CI 0.52 – 1.26). In sum, evidence suggests that NSAIDs may slow down the progression
of gastric precancerous conditions. However, this effect is small and eventually nonsignificant
after H. pylori eradication or in more advanced lesions. Considering that NSAIDs have a potential
for serious adverse events it is the opinion of the present authors that they cannot
be recommended specifically for this purpose. The exception may be low dose aspirin
since it has a better safety profile and its beneficial effects are more generalized,
reducing also cardiovascular death risk and the risk of development of other cancers,
and therefore it could be considered in selected patients.
Rebamipide and moluodan The effects of two other drugs on pathological findings has been investigated; the
drugs were administered for 6 – 12 months following H. pylori eradication therapy if required. Rebamipide, a free-radical scavenger, reduced inflammation,
IM, and LGD in one RCT [174 ], and reduced chronic inflammation but not IM in another RCT [175 ]; the RCTs included a total of 280 patients. Moluodan, a preparation of Chinese medicine
herbs, was associated with a decrease in the dysplasia score, with dysplasia disappearance
reported in 24.6 % of patients in an RCT (196 patients) [176 ]. Future studies should confirm these results before any recommendation can be made
regarding these therapies.
Antioxidant vitamins With respect to antioxidant vitamin supplementation, no new studies evaluating their
effects on gastric precancerous conditions were identified. In the general population,
intake of some vitamins may decrease the risk of gastric cancer (RR 0.77, 95 %CI 0.71 – 0.83)
according to a meta-analysis of 47 studies including 1 221 392 participants [177 ]. A significant risk reduction of approximately one third was found for vitamins
A, C, and E at daily doses of 1.5 mg, 100 mg, and 10 mg, respectively. The risk reduction
was noted only in studies where low dose dietary vitamins were used, not when vitamins
were administered at high dose or as a drug supplement. Some authors have suggested
that an efficient intervention would aim at nutrient repletion (i. e., with physiological
as opposed to pharmacological doses) in high risk populations with poor nutrition
[178 ]. Regarding potential harm, a meta-analysis (53 RCTs, 241 883 participants) found
that supplementation with vitamins A and E in doses higher than the recommended daily
allowances (as used in some of the studies included in the abovementioned meta-analysis
[177 ]) were associated with increased mortality [179 ].
In line with this, two meta-analyses found a twofold higher risk of gastric cancer
in individuals consuming low versus high amounts of allium vegetables [180 ], and in those consuming a “Western/unhealthy” diet, rich in starchy foods, meat,
and fats, versus a “prudent/healthy” diet rich in fruits and vegetables [181 ].
7 Cost-effectiveness of surveillance and screening
7 Cost-effectiveness of surveillance and screening
25 In intermediate to high risk regions, identification and surveillance of patients
with precancerous gastric conditions is cost-effective.
Moderate quality evidence) (100 % agree [94 % strongly or moderately agree]).
Surveillance of precancerous conditions In 2012 when the MAPS I Guideline was issued, three studies had been published regarding
the cost-effectiveness of surveillance of precancerous conditions. They had provided
conflicting results, mainly because of different estimates for progression to dysplasia
or cancer. Since then only three further studies have been published, all in intermediate
risk countries.
Areia et al. used a Markov model for Portugal, comparing the cost-utility of three
different endoscopic surveillance strategies, every 3, 5, or 10 years, for patients
with extensive precancerous conditions aged 50 – 75 years [182 ]. It showed that endoscopic surveillance of patients with extensive precancerous
conditions every 3 years was cost-effective compared to no surveillance and was better
than the 5- and 10-year strategies.
Zhou et al. also applied a Markov model in Singapore and compared the cost-utility
of several endoscopic surveillance or screening strategies, every 1 or 2 years, for
patients aged 50 – 69 [183 ]. It showed that endoscopic surveillance of patients with precancerous conditions
every 2 years was the most cost-effective strategy while screening strategies were
extendedly dominated (i. e., had a lower incremental cost-effectiveness or were cost-ineffective).
A third cost-utility model from Wu et al., also for Singapore, compared endoscopic
surveillance every 1 year versus endoscopic screening every 2 years versus nothing,
for patients aged 50 – 69, and concluded that annual endoscopic surveillance was cost-effective
for patients with precancerous conditions [184 ].
All three models demonstrated that endoscopic surveillance of patients with precancerous
conditions in countries with an intermediate risk for gastric cancer was cost-effective,
as suggested by the MAPS I Guideline. The recommended 3-year interval was specifically
modelled only in the study from Portugal and proved to be better than longer 5- or
10-year intervals, while the studies from Singapore showed that a 1- or 2-year interval
was most suitable but did not model the 3-year option.
In conclusion, endoscopic surveillance every 3 years of patients with precancerous
conditions in countries with an intermediate risk for gastric cancer is cost-effective,
but further economic studies would be welcome to further define the optimal interval
for endoscopy.
Endoscopic screening for gastric cancer. Regarding endoscopic screening for gastric cancer in the general population, at present
it is applied only in high risk populations, such as those of Japan and Korea, and
several economic studies have already been published proving its cost-effectiveness
[185 ].
Since 2012, two further studies have been published for the Korean population, again
concluding that endoscopic screening for this high risk population is cost-effective.
Chang et al., using a cost-utility model from a societal perspective for the population
aged 50 – 80 years, concluded that endoscopic screening annually for men and biennially
for women was the optimal cost-effective option among 12 different strategies [186 ]. Another cost-effectiveness model, also from a societal perspective in patients
older than 40 years, demonstrated that annual endoscopic screening was better than
X-ray screening or no screening [187 ].
Regarding the cost-effectiveness of endoscopic screening for gastric cancer in intermediate
or low incidence countries, three models have been published since preparation of
the MAPS I Guideline. Yeh et al. investigated a one-time gastric cancer screening
strategy for an American population at age 50, with a cost-utility analysis from a
societal perspective, comparing serum pepsinogen test screening followed by endoscopy
if results were positive versus endoscopic screening; they concluded that for this
low risk population neither option was cost-effective [188 ]. Others also used a Markov cost-utility model for the US population, and analyzed
the option of adding a one-time screening upper endoscopy at the time of screening
colonoscopy in 50-year-old patients. They concluded that this option was not cost-effective
for this low risk population despite the reduced endoscopy costs [189 ]. Finally, Areia et al. modelled the option of adding a screening upper endoscopy
at the time of screening colonoscopy, in Portugal, an intermediate to high risk country
for gastric cancer [190 ]. Using a Markov cost-utility analysis they compared three screening strategies:
stand-alone upper endoscopy, endoscopy combined with a colorectal cancer screening
colonoscopy after a positive fecal occult blood test, or pepsinogen serology screening.
The conclusion was that endoscopic gastric cancer screening was cost-effective if
combined with a screening colonoscopy in Portugal. Furthermore, this strategy might
prove to be cost-effective in other European countries with a gastric cancer risk
≥ 10 per 100 000 inhabitants, depending on further specific economic studies for each
local setting.
8 Research agenda
During the updating of this Guideline, gaps in evidence remained noticeable. Further
research is needed, including:
Large randomized trials with different surveillance schedules, that would elucidate
the natural history of each stage of gastric changes (and identify further variables
for predicting progression), and also clarify the benefit of each surveillance interval;
Adequately designed trials to assess the effect of interventions such as H. pylori eradication and others regarding the premalignant stomach;
Studies to address methodologies and target populations for screening these lesions
in Western countries.
Disclaimer
The legal disclaimer for ESGE guidelines [191 ] applies to the current Guideline.