Article Text
PDF
Abstract
Objective During the last decade, the management of gastric intestinal metaplasia (GIM) has been addressed by several distinct international evidence-based guidelines. In this review, we aimed to synthesise these guidelines and provide clinicians with a global perspective of the current recommendations for managing patients with GIM, as well as highlight evidence gaps that need to be addressed with future research.
Design We conducted a systematic review of the literature for guidelines and consensus statements published between January 2010 and February 2023 that address the diagnosis and management of GIM.
Results From 426 manuscripts identified, 16 guidelines were assessed. There was consistency across guidelines regarding the purpose of endoscopic surveillance of GIM, which is to identify prevalent neoplastic lesions and stage gastric preneoplastic conditions. The guidelines also agreed that only patients with high-risk GIM phenotypes (eg, corpus-extended GIM, OLGIM stages III/IV, incomplete GIM subtype), persistent refractory Helicobacter pylori infection or first-degree family history of gastric cancer should undergo regular-interval endoscopic surveillance. In contrast, low-risk phenotypes, which comprise most patients with GIM, do not require surveillance. Not all guidelines are aligned on histological staging systems. If surveillance is indicated, most guidelines recommend a 3-year interval, but there is some variability. All guidelines recommend H. pylori eradication as the only non-endoscopic intervention for gastric cancer prevention, while some offer additional recommendations regarding lifestyle modifications. While most guidelines allude to the importance of high-quality endoscopy for endoscopic surveillance, few detail important metrics apart from stating that a systematic gastric biopsy protocol should be followed. Notably, most guidelines comment on the role of endoscopy for gastric cancer screening and detection of gastric precancerous conditions, but with high heterogeneity, limited guidance regarding implementation, and lack of robust evidence.
Conclusion Despite heterogeneous populations and practices, international guidelines are generally aligned on the importance of GIM as a precancerous condition and the need for a risk-stratified approach to endoscopic surveillance, as well as H. pylori eradication when present. There is room for harmonisation of guidelines regarding (1) which populations merit index endoscopic screening for gastric cancer and GIM detection/staging; (2) objective metrics for high-quality endoscopy; (3) consensus on the need for histological staging and (4) non-endoscopic interventions for gastric cancer prevention apart from H. pylori eradication alone. Robust studies, ideally in the form of randomised trials, are needed to bridge the ample evidence gaps that exist.
- surveillance
- gastric carcinoma
Data availability statement
Data are available on reasonable request. All data analysed are available in proper databases depending on publisher.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Since 2010, several guidelines and expert consensus statements have been published worldwide providing evidence-based recommendations for the management of patients with gastric intestinal metaplasia (GIM), a precancerous condition associated with increased risk of gastric cancer.
Likely due to heterogeneous populations, methods and wording across different guidelines, controversy, confusion and misperceptions remain in the management of patients with GIM, which might compromise the optimal care of these potentially high-risk patients.
WHAT THIS STUDY ADDS
We conducted a systematic review and synthesised all available, published international guidelines describing the management of patients with GIM and, here, provide the first unified, global perspective for clinicians.
Irrespective of the patient’s country of origin or ethnicity, the index or surveillance endoscopy must include staging of GIM by performing endoscopic and histopathological mapping (risk stratification).
Only individuals with a high-risk GIM phenotype (ie, corpus-extended GIM, OLGIM stages III/IV, moderate-to-severe GIM, incomplete GIM subtype, persistent refractory Helicobacter pylori infection or first-degree family history of gastric cancer) merit regular surveillance, with a 3-year interval unless multiple risk factors are present.
Individuals with focal GIM changes limited to the antrum and no other risk factors for gastric cancer do not require ongoing endoscopic surveillance follow-up.
H. pylori testing (and its eradication if present) is unanimously recommended.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Clinical recommendations and policies must include risk stratification of gastric cancer among individuals with GIM.
There is an opportunity to enhance the research agenda in this field to better define: populations who warrant screening; the optimal method for endoscopic surveillance (eg, endoscopic staging, use of image enhancement, etc) and the optimal interval for surveillance; development of non-invasive diagnostic and prognostic biomarkers for GIM; assessment of interventions beyond H. pylori eradication to prevent gastric cancer and management of GIM in specific situations (eg, autoimmune gastritis without H. pylori, hereditary syndromes), among others.
Introduction
Gastric adenocarcinoma (GC) is a preventable cancer. In 2022, there were approximately 970 000 new GC cases and nearly 760 000 related fatalities. The International Agency for Research on Cancer projects that the number of new GC cases will increase to 1.8 million by 2040, with an associated 1.3 million deaths if current trends continue. This projection underscores the critical importance of both primary and secondary prevention efforts.1 2
Helicobacter pylori eradication, healthy dietary habits and smoking cessation are mainstays for primary GC prevention. However, primary prevention alone is insufficient for successfully decreasing GC burden and mortality, and effective secondary and tertiary prevention through strategies for early diagnosis is imperative.3 4 In the majority of cases, intestinal-type GC develops through a cascade of gastric preneoplastic mucosal changes before neoplastic progression occurs. Furthermore, early gastric cancer (ie, prior to submucosal invasion) can be cured through endoscopic or surgical resection, whereas there are no curative options for advanced-stage diagnoses. The slow stepwise progression and the opportunity for curative resection of early lesions underlie the rationale and effectiveness of secondary and tertiary prevention for combatting GC burden and mortality. These key concepts can be extrapolated, in theory, to all countries that have the appropriate infrastructure, and this is reflected in international guidelines for gastric intestinal metaplasia (GIM) diagnosis and management.
The ‘Correa cascade’ considers the occurrence of GC in a chronic state of mucosal inflammation, including the presence of atrophic changes and GIM. GIM in most cases occurs on a background of atrophy, although atrophic changes may not be clinically apparent in the setting of diffuse GIM.5 Conversely, atrophy may occur in the absence of GIM. Although moderate-to-severe gastric atrophy is associated with an increased risk of GC, the endoscopic and histopathological staging and risk stratification for gastric atrophy as compared with GIM are subject to more inter-observer, and even intra-observer, variability. Therefore, we have focused on guidance for GIM management as opposed to atrophy in the absence of confirmed GIM.
The prevalence of GIM when broadly considering all adults undergoing endoscopy with biopsies ranges from 5% to 30%, and is potentially even higher among individuals with additional risk factors.6 7 Despite this high prevalence and known association with GC, no specific guidelines existed for the management of GIM until 2012 and this was reflected by significant heterogeneity in clinical practice for GIM and low adherence to endoscopic surveillance and appropriate staging protocols.8 9 To address this, scientific societies internationally independently evaluated the published literature according to predefined methodology (eg, Grading of Recommendations, Assessment, Development and Evaluations (GRADE)) to develop evidence-based guidance on the management of precancerous conditions, specifically GIM.
Currently, non-invasive biomarkers have not demonstrated sufficient, reproducible accuracy for GIM diagnosis. Endoscopy and histology thus remain the mainstay for diagnosis, surveillance and staging of GIM. We hypothesise that the availability of multiple international guidelines with varying methodologies and presentation may confuse clinicians regarding GIM management, especially endoscopic surveillance, which invariably would impact clinical care.
We therefore conducted a systematic review and qualitative synthesis of all published and indexed evidence-based guidelines or expert consensus statements on GIM management, including which individuals should receive screening for staging of the gastric mucosa for future risk determination; and which individuals should undergo GIM surveillance. This work primarily aims to provide clinicians worldwide with a comprehensive perspective, and secondarily to inform future guideline processes as well as the research agenda.
Methods
We performed a systematic review in PubMed on 20 February 2024, with the following query: (guidelines OR statements OR consensus) AND (atrophy OR atrophic gastritis OR intestinal metaplasia OR precancerous OR premalignant OR preneoplasia OR dysplasia) AND (gastric OR stomach).
We limited the search to include publications between 1 January 2010 and the search date. We also searched the complete list of references of the included articles and used experts’ (authors’) knowledge of other references to determine other guidelines that might have been missed during the electronic search. We included all guidelines and consensus statements that evaluated and discussed the diagnosis and management of GIM. Documents were considered ‘guidelines’ or ‘consensus statements’ as long as they were commissioned by an accredited international or national GI society (eg, European Society for Gastrointestinal Endoscopy, American Gastroenterological Association) and described the methodological approach for delivering clinical recommendations or suggestions. We excluded documents that did not include GIM management. Hereafter, we use the term ‘guideline’ to also encompass consensus statements and clinical practice updates, with differences in methodology acknowledged where appropriate.
From each guideline, the following information was abstracted and summarised:
Scientific society(ies) supporting the guideline.
Year of publication and literature search dates.
Scope of the guideline (eg, management of GIM only vs other, eg, H. pylori eradication).
Geographic region and population to which the guideline refers.
Methodology (GRADE vs other).
Recommendations or discussions regarding upper gastrointestinal endoscopy (UGIE) screening for eligible populations to allow for the opportunistic identification of GC and/or gastric precancerous conditions (GIM).
Recommendations or suggestions regarding surveillance once GIM is diagnosed.
Recommended/Suggested endoscopic surveillance intervals in those considered eligible for surveillance.
The recommended/suggested endoscopic approach (quality parameters) to optimise the detection of gastric precancer/cancer, and staging.
The stated risk stratification parameters in patients with GIM, including histological staging systems, and respective definitions.
Recommendations/Suggestions regarding potential therapeutic interventions for GC prevention once GIM is diagnosed (eg, H. pylori testing/eradication; lifestyle modifications).
Intentionally, at least one author from each of the geographic regions encompassed was asked to co-author this manuscript. Each author was asked to critically review the summary for inconsistencies concerning their region’s published guidelines. The topics of endoscopic and histological diagnosis and staging of GIM are extensively described in prior literature and are not the focus of the article. An in-depth discussion of quality metrics in UGIE and population-based screening of GC and precancerous conditions is outside the scope of this article, apart from identifying whether these aspects are included in international guidelines and summarising the respective recommendations and guidance.
Results
Literature search
The systematic literature search yielded 426 manuscripts, of which 411 were excluded based on title and abstract screening. No additional articles were identified through manual review of the references, and one was identified through co-authors’ expert knowledge.10 A total of 16 articles met eligibility criteria and were included in this qualitative review.10–25
Seven of these 16 articles were from Europe, 5 were from the Americas (3 from the USA, 2 from Latin America) and 4 were from the Asia-Pacific region (111 from China, 2 from Japan, 1 from Taiwan). The 2012 and 2019 updates of the European Society of Gastrointestinal Endoscopy (ESGE)/European Helicobacter and Microbiota Study Group/European Society of Pathology (ESP) guidelines (MAPS I and MAPS II, respectively), the British Society of Gastroenterology (BSG) guidelines (2019), Italian (2019), Asociación Española de Gastroenterología/Sociedad Española de Endoscopia Digestiva and Sociedad Española de Anatomía Patológica (2021), Asociacion Chilena de Endoscopia Digestiva/Sociedad Chilena de Gastroenterolgía (2014), American Gastroenterology Association (AGA) guidelines (2019), American Society of Gastrointestinal Endoscopy (ASGE, 2015) and Chinese guidelines (2023) were specifically dedicated to the management of precancerous conditions, including GIM. In contrast, the Maastricht, Taipei, Kyoto, Brazilian and Japan Gastroenterological Endoscopy Society guidelines primarily focused on H. pylori management and included statements or recommendations regarding the management of GIM. The Italian and Spanish guidelines are not adaptations of MAPS I and/or II, which is why they are considered separately in this review. The AGA Clinical Practice Update published in 2021 focuses on atrophic gastritis as a precancerous condition but provides guidance on GIM management including surveillance and is therefore included. All included guidelines used either GRADE methodology or a consensus approach. Nearly all the guidelines primarily addressing the management of precancerous conditions used GRADE methodology, with the exception of the Chinese Society of Gastroenterology, Cancer Collaboration Group of Chinese Society of Gastroenterology/Chinese Medical Association guidelines, which used a consensus approach. In contrast, guidelines that included GIM management as a secondary focus more often used a consensus approach (table 1). Nevertheless, the overall clinical guidance regarding GIM management did not differ substantially based on the robustness of the methodology (ie, GRADE vs consensus).
Description of included guidelines and consensus documents according to their main purpose/scope and geographic region
Opportunistic detection of GIM during screening or diagnostic endoscopy
GIM is typically asymptomatic and can only be detected endoscopically as currently available non-invasive biomarkers have not demonstrated reliable accuracy for the diagnosis. Thus, opportunities to diagnose GIM are either via screening UGIE in asymptomatic individuals (ie, opportunistic detection) or via diagnostic UGIE in individuals with a clinical indication for endoscopy. None of the international guidelines recommend screening for GIM per se; instead, screening is for early detection of GC. Some guidelines do, though, make clear statements that all endoscopies must include an assessment for the presence of precancerous changes, namely GIM. Many guidelines align in recommending that a high-quality endoscopic examination using virtual technologies (eg, narrow band imaging (NBI), linked colour imaging (LCI), blue light imaging (BLI)) or conventional chromoendoscopy, along with endoscopic biopsies, should be performed to enhance the endoscopic detection of GIM. One exception is the Japanese guideline, which does not recommend gastric biopsies. Instead, it relies solely on endoscopic detection and staging (table 2). For those individuals diagnosed with GIM, many but not all guidelines also recommend the use of histological staging systems, primarily OLGIM, and histological subtyping of GIM given the implications for risk stratification (table 2).
Recommendations for opportunistic diagnosis, surveillance and therapeutic interventions in individuals with GIM according to the diverse guidelines/consensus and their scope and region
Guidelines that include opportunistic detection through endoscopic screening. As of this writing, the Japanese guideline, the Spanish guideline, the British guideline and the Maastricht guideline offer recommendations regarding true population-based screening for GC. In Japan, references to GC risk stratification and potential non-invasive approaches are named. The Spanish guidelines recommend against the routine use of UGIE for GC screening in the general Spanish population given the low GC incidence overall, and suggest screening only for those with a family history of GC. This is similar to the recommendations of the German guideline and the Maastricht guideline. The Maastricht guideline, which is focused on H. pylori diagnosis and management, recommends that endoscopy with gastric biopsies be performed in asymptomatic individuals with a family history of GC starting at age 45 years and older. Neither the Spanish nor Maastricht guidelines call out other high-risk groups (eg, early generation immigrants from countries where GC is endemic). The 2019 British guidelines state there is insufficient evidence to support endoscopic screening in the overall low-risk UK population, but do provide a weak recommendation that UGIE screening with biopsies should be considered for individuals aged ≥50 years with multiple risk factors for GC, notably rating the quality of evidence as low but achieving 100% consensus agreement. The Chilean guidelines stipulate that age >40 years is sufficient for endoscopic screening for GC and its precancerous conditions considering that Chile is a high GC incidence country; however, the Chilean guidelines list an additional criterion that patients have at least one other clinical indication for endoscopy (eg, abdominal pain) and thus, this cannot be considered true population-based screening. The Italian guideline recommends that a family history of GC and long-term PPI use warrants offering UGIE screening for precancerous conditions; clearly, though, many individuals warranting long-term PPI use may qualify for a diagnostic as opposed to a screening endoscopy. The ASGE guidelines on GIM management do not include recommendations on GC screening, but a separate document on race and ethnic considerations in UGIE more broadly mentions that endoscopic GC screening may be considered in certain high-risk populations in the USA (eg, immigrants from high-risk regions, first-degree family history) starting at age 40 years; no other guidance, such as mucosal staging protocol or recommendations for subsequent surveillance intervals, is provided.26
Guidelines that include opportunistic GIM detection through diagnostic endoscopy. The 2012 MAPS I guideline states that for H. pylori diagnosis and gastric mucosal staging, gastric samples from the antrum and corpus should be obtained during a diagnostic upper endoscopy. This recommendation was reiterated and refined in the updated 2019 MAPS II guideline. It mandated that all first-time diagnostic endoscopies include gastric antrum and corpus biopsies, without mention of whether this is necessary for patients who have been non-invasively tested for H. pylori already or who are otherwise considered low risk for GC. Similar to MAPS I/II, the Chinese guidelines recommend biopsies from the antrum and corpus to diagnose H. pylori and stage the gastric mucosa to determine the need for ongoing surveillance. The 2020 Japanese guidelines recommend endoscopic risk stratification using Kimura-Takemoto classification or virtual chromoendoscopy of GIM since advanced precancerous changes including GIM can be reliably detected endoscopically.27
Guidelines that do not address opportunistic detection of GIM. The AGA guidelines do not address endoscopic screening or opportunistic detection during a diagnostic endoscopy for a non-screening indication. Instead, the document provides recommendations for the management of patients with already confirmed GIM, which is most often identified incidentally in the USA. The AGA guidelines acknowledge risk factors for GC, such as specific races and ethnicities, early generation immigrant groups and family history of GC, where an individualised approach may be considered.
In summary, there is high heterogeneity among guidelines regarding recommendations on the use of endoscopy for GC screening or opportunistic detection of GIM. It seems reasonable to consider that the index or surveillance endoscopy must include staging of GIM, if present. Revised or updated guidelines may consider clarifying these recommendations.
Individuals with established GIM diagnosis: surveillance versus no surveillance
All guidelines are aligned regarding the need to risk stratify patients with GIM to determine their need for ongoing surveillance, as well as in mandating H. pylori eradication if present.
All guidelines agree that patients with GIM are considered to have a high-risk phenotype if GIM affects both the gastric antrum (±incisura) and corpus (corpus-extended GIM), if there is moderate-to-severe GIM staged using OLGIM (ie, OLGIM III/IV), if GIM is of the incomplete subtype (vs complete subtype), and if there is a family history of GC in a first-degree relative. These are all factors consistently highlighted in the literature as the most important predictors of progression and may be used to identify individuals at the highest risk of GC, with most guidelines distinctly prioritising histological staging.
According to most guidelines, the presence of any high-risk phenotype is sufficient to identify those patients with GIM who merit surveillance. In contrast, the Spanish guidelines recommend surveillance for extensive GIM if at least one other risk factor, for example, incomplete type or family history of GC, is present.
All guidelines are also aligned in their recommendations that patients with a low-risk GIM phenotype, which is substantially more common than the high-risk phenotype, do not require ongoing endoscopic surveillance. Most guidelines agree that the presence of antrum-limited GIM that is non-severe (OLGIM I-II) may not warrant surveillance, assuming adequate staging metrics during the index gastroscopy and the absence of high-risk qualifiers, including persistent H. pylori infection, incomplete GIM subtype and family history of GC in a first-degree relative. However, factors such as tobacco smoking and dietary factors, which are associated with GC, are not fully accounted for as risk stratification metrics. In patients with limited, mild GIM but with additional risk factors for gastric cancer (eg, family history or high-risk race/ethnicity), MAPS II, BSG, China, Chilean and the AGA guidelines recommend endoscopic surveillance, whereas this is not mentioned in the Japanese, Taipei and Brazilian guidelines. However, the Japanese guideline recommends surveillance regardless of the extent of GIM because of the high GC incidence in the population and the relative availability of endoscopy.
Most guidelines do not provide guidance on how to properly risk stratify patients when information is missing. For example, if patients have insufficient biopsies (eg, only antrum), complete histological staging may not be possible. However, the AGA guidelines, MAPS I/II and the Spanish guidelines do provide guidance on short-interval endoscopy (<12 months) for risk stratification purposes if the quality of the initial examination is questionable.
In summary, advanced stages of GIM—such as GIM affecting both the gastric antrum (±incisura) and corpus (corpus-extended GIM), moderate-to-severe GIM staged using OLGIM (ie, OLGIM III/IV) or GIM of the incomplete subtype—are considered the phenotypes of interest for endoscopic surveillance to detect early neoplasia. In contrast, low-risk GIM phenotypes generally do not warrant surveillance in the absence of additional risk factors for advanced gastric neoplasia.
Approach to endoscopic surveillance
When surveillance is recommended, all guidelines recommend endoscopy as the only acceptable modality given the insufficient performance of currently available non-invasive biomarkers and biomarker panels. Few guidelines detail upper endoscopy quality metrics—for example, mucosal cleansing protocol, visualisation technique (high-definition white light±virtual or conventional chromoendoscopy), photodocumentation, or withdrawal time—within the guideline document, except for the BSG guidelines. Most guidelines do provide suggestions for the biopsy protocol, although with heterogeneity. Most guidelines suggest that biopsies from the antrum/incisura and corpus be obtained separately following a systematic protocol5 to allow for histological staging, while targeted biopsies should also be obtained separately for lesions suspicious for advanced pathology. However, MAPS II states that the benefit of performing biopsies in patients under surveillance is not established, and biopsies should only be performed if any irregular area/suspicious lesion is identified. Figure 1 summarises these approaches. Some guidelines have accompanying articles (eg, AGA, Spanish society) that detail quality metrics in upper endoscopy with specific sections on best practices for the detection and surveillance of preneoplastic conditions.28 29
Synthesis of most commonly accepted biopsies protocols according to endoscopic findings. In the presence of gastric intestinal metaplasia (GIM), targeted biopsies should be performed. In the absence of GIM, random biopsies from the antrum and corpus are acceptable, with a biopsy from the incisura angularis being optimal but optional.5 HD-WLE, high-definiton wight light endoscopy; IEE, image- enhanced endoscopy; NBI, narrow band imaging.
In summary, practice guidelines vary with respect to details and approach to the endoscopic surveillance exam.
Surveillance intervals
While there is generally agreement regarding who warrants surveillance, there is notable heterogeneity regarding recommended surveillance intervals. Surveillance every 3 years for patients with any high-risk GIM phenotype is the most commonly recommended (7 out of 16 guidelines) (figure 2). If there is a family history of GC in a first-degree relative, MAPS II, the BSG and the Chinese guidelines suggest considering a shorter surveillance interval. In contrast, the Chilean, Brazilian, Japanese and Taiwanese guidelines also recommend considering an interval shorter than 3 years, but they do not provide additional criteria for determining who qualifies for the shortened interval.
Algorithm synthesising the major recommendations for managing individuals with gastric intestinal metaplasia (GIM) according to country of origin when applicable. Both the European Society of Gastrointestinal Endoscopy (ESGE) guidelines and three US guidance documents (two American Gastroenterology Association, one American Society of Gastrointestinal Endoscopy) are included. *Patients with gastric cancer (GC) should still undergo endoscopic and histological staging of the surrounding mucosa to facilitate additional risk stratification. EGD, esophagogastroduodenoscopy; Hp, Helicobacter pylori; y, years; IEE, image-enhanced endoscopy.
The AGA, on the other hand, recommends that a surveillance interval of up to 5 years may also be acceptable, but does not provide additional information on who might qualify for this extended interval. Notably, the AGA Clinical Practice Update advocates for a 3-year interval in individuals with advanced atrophic gastritis with or without GIM.
In summary, international guidelines vary with respect to evidence-based guidance regarding appropriate endoscopic surveillance intervals among individuals with GIM enrolled in surveillance. Surveillance is generally suggested every 3 years, but the South American and Asian guidelines favour a lower threshold for more intensive surveillance in patients with additional risk factors, while the US guidelines allow for potentially longer intervals.
Non-endoscopic therapeutic interventions
Other than surveillance, there is no definitive endoscopic management for GIM in the absence of visible neoplastic lesions. The mainstay non-endoscopic intervention for patients with GIM is testing for H. pylori infection, either through histology or other non-invasive non-serological methods. Confirming eradication after treatment in those who test positive is a strong recommendation in all guidelines. GIM can progress even once H. pylori is successfully eradicated, and this fact is acknowledged in most guidelines.
There is more heterogeneity regarding non-endoscopic non-H. pylori interventions in patients with GIM for GC risk/mortality reduction. MAPS II and the Chinese guidelines suggest offering low-dose aspirin to patients at higher risk of GC who also have a cardiovascular indication. In contrast, the German guidelines advise against the use of aspirin if the sole indication is GC prevention. The Chinese recommendations also refer to the supplementation of certain vitamins/minerals, including selenium and folate, and garlic. The Maastricht guidelines state that medical and special dietary chemoprevention cannot be recommended in patients with severe GIM (or severe atrophy) post-H. pylori eradication in the absence of sufficient evidence for benefit. None of the other guidelines offered guidance for any other non-invasive interventions for individuals with GIM.
Specific situations
Management of individuals with GIM in special situations, for example, postpartial gastrectomy and hereditary syndromes (eg, Lynch syndrome, familial adenomatous polyposis) is not addressed in general. Many guidelines offer limited guidance for patients with autoimmune gastritis diagnosed with GIM and for those diagnosed with gastric neoplasia (dysplasia/GC). A comprehensive discussion of these topics is beyond the scope of this document.
Discussion
Reducing the global burden of GC is a shared goal worldwide. Most GCs have precursor conditions and lesions that can be detected endoscopically and histologically, which facilitates opportunities for prevention and curative resection of early GC. Population-based screening for GC is not recommended outside of a few East Asian countries.30 That said, several international societies including from Europe and the Americas acknowledge the potential impact of targeted screening for high-risk groups, although with varying definitions of what qualifies as ‘high-risk’, which most certainly reflects the limited literature. Current studies are conducted to determine the role of endoscopy and other measures for screening in Europe (eg, Towards Gastric Cancer Screening Implementation in the European Union (TOGAS EU4H–2022–PJ–01)). Endoscopic screening for early GC detection and prevention (eg, resection of dysplasia) must be balanced with the fact that endoscopy in most countries is a limited resource, is costly, invasive, inconvenient for patients and poses small but measurable risk.
GIM is a defined precancerous condition that is amenable to surveillance for early GC detection. However, most patients with GIM are at low risk of progression and the risk/burden of endoscopy likely outweighs the theoretical benefit of detecting early neoplasia in these patients. Accordingly, clinicians must be equipped with clear evidence-based guidance to ensure that resources are appropriately allocated to high-risk patients with GIM while minimising the burden and potentially unnecessary anxiety among low-risk patients with GIM. Furthermore, there is a push to maximise the sustainability of endoscopy for cancer prevention by optimising the populations for which it is recommended, and ensuring each exam generates the highest-quality information to minimise unnecessary repeat procedures.31 32 The presence of GIM, along with the determination of a high-risk versus low-risk phenotype, is the most consistent marker for GC risk, and thus appropriate endoscopic surveillance and adjunct therapeutic interventions (eg, H. pylori eradication, smoking cessation) allow an invaluable opportunity for efficient secondary prevention of GC. For these reasons, we aimed to generate a document that summarises and compares the currently published guidelines and consensus documents on GIM management.
We conclude that international guidelines are aligned in two main recommendations regarding (1) the need for risk-stratified surveillance (as opposed to universal surveillance) and (2) mandatory H. pylori eradication in patients with GIM. Otherwise, there was heterogeneity regarding recommendations for opportunistic detection of GIM during UGIE, the execution of risk stratification (ie, use of histological staging systems such as OLGIM), the approach to the endoscopic examination with respect to defined quality metrics and gastric sampling protocols, surveillance intervals and adjunctive modalities for non-invasive interventions for GC risk reduction in patients with GIM (eg, chemoprevention agents apart from H. pylori eradication; diet modifications and supplementation). This heterogeneity reflects the current gaps in evidence and lack of robust, high-quality data, as well as the heterogeneity of the populations served by each of the guidelines. Indeed, it is expected that guidelines serving high-GC incidence populations (eg, Latin America, Asia-Pacific) may favour a stricter approach, for example, shorter endoscopic intervals, while those serving low-GC incidence populations may favour a less aggressive approach. However, this may not be appropriate, particularly since there are no compelling data to support changing surveillance practices based solely on the country of practice or patient ethnicity, independent of other risk parameters.33
Certainly, the prevalence of gastric precancerous conditions and GC is higher in certain populations based on country of origin and ethnicity. Additionally, the risk of developing GC overall (and mortality) is higher in early generation immigrants from high-GC regions who have moved to low-GC regions.34 35 However, among individuals already diagnosed with precancerous conditions, the risk of progression to GC is independent of the country of origin and ethnicity,33 meaning that other factors (eg, GIM subtype, persistent H. pylori, family history) should be used to determine the need and intervals for surveillance. Thus, the variations in recommendations, especially regarding the surveillance interval, across international guidelines may relate to perceived risk of progression but lacks strong evidence. For example, the Chilean guidelines recommend a 1-year surveillance interval in patients with OLGIM III/IV, whereas the Spanish guidelines recommend surveillance only if there is at least one additional risk factor for progression (eg, family history of GC), and the AGA suggests that up to a 5-year interval may be appropriate.
Other nuances when considering recommendations for GIM management in the context of each guideline and the population served relate to the respective healthcare infrastructure and resource availability, age of the population as well as cultural, lifestyle and dietary practices. The training of the clinical providers including the endoscopists and pathologists and their familiarity with GIM detection and staging is also of relevance. For example, broadly speaking, endoscopists in the USA have lower familiarity with GIM detection and endoscopic staging versus endoscopists from high-GC incidence countries, particularly Japan and South Korea where routine population-based screening occurs. In the USA, histological staging with OLGIM and GIM subtyping are not routine practice, particularly compared with many European and Latin American countries. The primary benefit of endoscopic surveillance is the opportunity for early detection of neoplasia and curative resection, which has substantial implications for GC mortality reduction. However, the success of endoscopic surveillance hinges on the accurate detection of precancerous and cancerous lesions. Considering that the endoscopic miss rate for early GC is approximately 10% based on a meta-analysis, and may be higher in less experienced hands, interventions are needed to improve and standardise the endoscopic surveillance exam. To this end, few guidelines specify the quality metrics that must be achieved to qualify as a high-quality upper endoscopic exam. However, the importance of such metrics is becoming increasingly recognised. In 2024, the AGA published its first consensus report providing ‘best practice advice’ statements for how to perform a high-quality upper GI endoscopy and included metrics specific for the GIM surveillance exam, which complements the AGA guideline on GIM management published in 2020.28 Similarly, the Spanish GI society also published a supplemental document on quality metrics in upper endoscopy to accompany the guideline, which lacked such metrics. The ESGE addresses appropriate biopsy sampling according to MAPS guidelines as a quality criterion together with others to define the quality of an upper endoscopy.29 32
Accurate endoscopic and histological staging of GIM is critical, as this is a primary determinant for the branchpoint of whether or not to recommend endoscopic surveillance. While there was consistency regarding which factors (eg, OLGIM III/IV, incomplete GIM subtype, anatomic extension) constitute a high-risk phenotype, there was heterogeneity regarding which metrics are routinely recommended. For example, in the USA, there is no guideline recommendation for pathologists to perform OLGIM staging or to report the histological subtype when GIM is present, and therefore this information is rarely available for clinical decision-making. In addition, there was heterogeneity regarding the ‘weight’ of each risk stratification parameter. Family history and OLGIM III/IV generally hold the most weight and in some guidelines warrant a shorter than 3-year surveillance interval; however, other guidelines, such as the Spanish guidelines, recommended surveillance in patients with extensive GIM only if additional risk factors are present (eg, incomplete GIM, family history of GIM), or if there is a prior or current history of gastric neoplasia. This heterogeneity stems both from the lack of robust evidence, as no randomised clinical trials exist, and from the diverse classifications that determine the presence of this phenotype.36–38 More robust and precise risk stratification markers (eg, tissue-based markers) could improve risk stratification and, ideally, consistency across guidelines.
Fortunately, most people with GIM who originate from low GC-risk countries fall into an overall low-risk phenotype and may not warrant or benefit from endoscopic surveillance. Accordingly, most guidelines are aligned in recommending against surveillance in patients with low-risk GIM phenotype. The two exceptions are the Chilean and Japanese guidelines, which recommend that all individuals with GIM should be surveilled once diagnosed. This universal recommendation reflects the high GC incidence and mortality in these countries and acknowledges that mild antrum-limited GIM carries some, although very low, level of elevated GC risk. However, the sustainability of such an approach has been questioned, given that endoscopy is a limited and costly resource. Among patients with GIM, the condition is extensive in 7%–25% based on studies from low GC incidence countries and in 16%–26% based on studies from high-incidence areas.5 39 40 The true prevalence of incomplete GIM versus complete GIM (vs mixed) is not well established, and studies reporting these figures vary significantly. However, incomplete GIM is more commonly observed in patients with extensive GIM and those with prior or current H. pylori infection. Notably, all of the guidelines that mention histological subtyping of GIM consider the presence of incomplete GIM as sufficient for categorising someone as high-risk and warranting endoscopic surveillance. However, the merits of this classification are based solely on observational studies (and meta-analyses of these studies), which are subject to bias. Despite this, studies consistently demonstrate that incomplete GIM is associated with several-fold higher risk of GC compared with patients with only complete-type GIM.
Unmet needs
Randomised trials on GC screening, gastric precancer surveillance and intervention strategies are lacking, and this stands in stark contrast to colorectal cancer screening and postpolypectomy colonoscopy surveillance, as well as oesophageal adenocarcinoma screening and Barrett’s oesophagus surveillance. Generating robust evidence is a critical unmet need, particularly as it relates to endoscopic surveillance intervals and developing, validating and positioning risk stratification metrics with high discrimination values. Specific research priority areas include non-invasive and tissue-based biomarkers for diagnosis, staging and prognosis of GIM; improved uptake and consistency of adherence to guidelines, the development of quality metrics linked to patient-specific outcomes (eg, gastric neoplasia detection rate), leveraging non-invasive therapeutic interventions for halting or reversing GIM progression (eg, diet modifications; chemoprevention) apart from H. pylori eradication, leveraging newer technologies such as artificial intelligence and image-enhanced endoscopy to improve the detection of gastric preneoplasia/neoplasia and specific guidance regarding GIM management in special situations such as hereditary syndromes, postsurgical stomach, after endoscopic resection of neoplasia, and in autoimmune gastritis.
This is the first systematic review of all position statements regarding GIM. At the time of this writing, new guidelines are being updated or developed in the USA, Germany and by ESGE (MAPS III), but are not yet publicly available. Reassuringly, GIM guidelines have demonstrated increased uptake in clinical practice during the last decade. For example, retrospective studies from the USA,41 reported that H. pylori diagnosis, multiple biopsies and the recommendation for surveillance significantly increased among US doctors following the 2019 AGA guideline. Also, in Europe, the ESGE considered biopsy sampling according to the so-called MAPS protocol as a minor but relevant quality parameter.32
In conclusion, irrespective of the country of origin when performing an UGIE, the best care must include staging of GIM by performing histopathological mapping. In most guidelines, only individuals at a higher risk (corpus extended GIM, OLGIM stages III/IV, incomplete GIM subtype, persistent H. pylori infection or first-degree family history of gastric cancer) merit regular surveillance at a 3-year interval if no concurrent risk factors are present. Importantly, the most common group, those individuals with focal GIM changes limited to the antrum and no other risk factors for GC, do not require follow-up in most cases, but there are exceptions. H. pylori testing (and its eradication if present) is unanimously recommended and should be synergistically suggested for primary prevention of GC.
Supplemental material
Data availability statement
Data are available on reasonable request. All data analysed are available in proper databases depending on publisher.
Ethics statements
Patient consent for publication
Ethics approval
This manuscript did not involve animal experiences or patients' data or interventions.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
X @mdinisribeiro, @ShailjaShahMD, @diogolibanio, @MoreiraR_Leti, @emadelomar
MD-R and SS contributed equally.
Correction notice This article has been corrected since it publishes Online First. The first sentence of the results section in the abstract has been corrected.
Contributors MD-R, SS and EME-O developed the protocol, conducted the review and revised the final draft of the manuscript. All the authors collected data, provided input for the protocol and revised the draft of the manuscript, approving the final version. MD-R, SS and EME-O are the guarantors.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. JB is supported by the UK Medical Research Council in the context of the Clinical Academic Research Partnership (MRC CARP) scheme (Grant ref.: MR/W029960/1).
Competing interests No conflicts of interest are declared by MD-R, SS, HE-S, MB, NU, HT, LGC, DL, EL, AR, J-YF, LM, JB, EJK, EME-O. PM is a member of advisory board/lecturer of Aboca, Alfasigma, Allergosan, Bayer, Biocodex, Menarini advisory boards/lectures.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.