The relationship between the eradication of Helicobacter pylori and the occurrence of stomach cancer: an updated meta-analysis and systemic review

Abstract

Objective

Helicobacter pylori (H. pylori) are classified as a Group 1 carcinogen by the International Agency for Research on Cancer (IARC), highlighting its well-established role in gastric carcinogenesis. While previous studies and systematic reviews suggest that H. pylori eradication may lower the incidence and mortality of gastric cancer, the evolving body of evidence necessitates continual reassessment. In light of newly available data, we conducted a comprehensive meta-analysis to evaluate the association between H. pylori eradication therapy and gastric cancer risk, aiming to strengthen the evidence base and inform clinical decision-making.

Method

We systematically searched the Cochrane Library, PubMed, Web of Science, and Embase up to December 2024, including only randomized controlled trials (RCTs) while excluding non-RCT studies. The target population comprised adults diagnosed with H. pylori infection who were either healthy or had previously undergone gastrectomy for gastric tumors. Eradication therapy served as the intervention, while placebo was the control. Eligible studies had a treatment duration exceeding seven days and a follow-up period of more than three years. The Cochrane risk-of-bias tool was used to assess methodological quality, and effect estimates were expressed as relative risk (RR) and the number needed to treat (NNT).

Outcomes

A total of 11 RCTs encompassing 104,786 individuals were analyzed. The meta-analysis revealed that H. pylori eradication significantly reduced gastric cancer risk (RR: 0.61; 95% CI: 0.47–0.79; NNT = 332). Subgroup analysis indicated that among healthy adults, the relative risk (RR) for the occurrence of gastric cancer was 0.67 (95% CI: 0.48–0.93; NNT = 476). In individuals who had undergone endoscopic mucosal resection, the reduction was even more pronounced (RR: 0.51; 95% CI: 0.36–0.71; NNT = 21). Although stomach cancer-specific mortality showed a slight decline (RR: 0.84; 95% CI: 0.69–1.01), all-cause mortality remained statistically unchanged (RR: 1.00; 95% CI: 0.89–1.13).

Conclusion

Our findings support H. pylori eradication as an effective strategy for reducing gastric cancer incidence, particularly in East Asian populations. While the effect on overall mortality remains inconclusive, the observed reduction in gastric cancer-related mortality highlights the potential clinical significance of eradication therapy as a preventive measure. Further well-designed, long-term studies are warranted to reinforce the evidence base and optimize clinical recommendations.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12876-025-03886-z.

Introduction

Gastric cancer is the fifth most common malignancy worldwide and the third leading cause of cancer-related mortality [1]. Although its incidence has declined in recent years, an increasing proportion of cases is now being diagnosed at younger ages [2]. By 2040, the global burden of gastric cancer is projected to exceed 1.77 million cases. In 2020 alone, 770,000 individuals succumbed to the disease, with 65% of fatalities occurring in men, predominantly in East Asia [3].

Helicobacter pylori (H. pylori) infection is the primary risk factor for gastric cancer, accounting for nearly 90% of distal gastric cancer cases worldwide [4]. It is estimated that 50% of the global population harbors H. pylori, with approximately 600 million cases occurring in China alone. While many infected individuals remain asymptomatic, between 1% and 3% may progress to gastric adenocarcinoma (GAC), and approximately 0.1% may develop mucosa-associated lymphoid tissue (MALT) lymphoma [5, 6].

Prior meta-analyses have established H. pylori infection as the predominant driver of gastric carcinogenesis, contributing to more than a 2.5-fold increase in gastric cancer risk [7]. The carcinogenic process may be mediated through direct dysregulation of protein expression, induction of genetic mutations, or persistent chronic inflammation, ultimately leading to neoplastic transformation. Amieva et al. demonstrated that H. pylori-induced gastric inflammation activates LGR5 stem cells within the antrum, triggering abnormal cellular proliferation [8]. Moreover, H. pylori interact with dietary factors, including iron and salt, as well as other microbial communities in the gastrointestinal tract, collectively influencing the oncogenic transformation of gastric epithelial cells.

Gastric intestinal metaplasia (GIM), a precursor to gastric cancer, has been extensively studied. Research by Sugano et al. highlighted that incomplete GIM poses a higher carcinogenic risk than complete GIM. Chronic gastritis induced by H. pylori significantly increases the prevalence of incomplete GIM, further escalating cancer risk [9, 10]. A study by Li et al. demonstrated that even short-term H. pylori eradication can effectively reduce the incidence and mortality of gastric cancer [11]. Accumulating evidence supports the notion that early eradication therapy can markedly lower gastric cancer risk and mortality. Meta-analyses by Ford et al. and Lee et al. further corroborated that H. pylori eradication significantly reduces gastric cancer incidence [12, 13].

Given that four years have elapsed since the last major meta-analysis, newly published trials with longer follow-up durations and greater clinical significance may provide additional insights. This study aims to re-evaluate the relationship between H. pylori eradication therapy and gastric cancer incidence through a comprehensive meta-analysis, integrating the most recent evidence to refine our understanding and inform clinical strategies.

Methods

This study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [14]. The research protocol was registered in PROSPERO (registration number: CRD42024617886), although a separate protocol document was not prepared.

Literature search

We systematically searched Web of Science, Embase, the Cochrane Library, and PubMed for studies published up to December 10, 2024, using controlled vocabulary (e.g., MeSH), keywords, and free-text terms (see Appendix 1). To ensure comprehensiveness, gray literature sources such as ClinicalTrials.gov and ICTRP were screened. Major gastroenterology conference proceedings from 2001 to 2024 were also reviewed. Where only abstracts were available or studies were ongoing, we contacted authors to obtain full texts or updated data. Additionally, reference lists of included studies were checked to identify further relevant trials.

Criteria for inclusion and exclusion

Eligibility was based on the PICOS framework [15]. Studies were included if they (1) involved adults with H. pylori infection and no prior stomach cancer, including those post-curative treatment for early gastric tumors; (2) provided eradication therapy ≥ 7 days; (3) had a control group receiving placebo or symptomatic care; (4) reported gastric cancer incidence (primary outcome), and mortality (secondary outcomes); (5) were randomized controlled trials; and (6) were published in English [16].

Studies were excluded if they: (1) included participants with stomach cancer at baseline; (2) used active eradication in controls; (3) lacked relevant outcome data; (4) were non-RCTs or not in English; or (5) focused solely on endoscopic outcomes without long-term clinical follow-up. All exclusions were applied at the study level, as individual patient data were not available.

Study selection

Search results were imported into EndNote, and duplicates were removed. Two independent researchers screened titles and abstracts using the predefined criteria, followed by full-text review to confirm eligibility. Discrepancies were resolved through discussion or, if needed, consultation with a senior investigator. Reference lists of selected studies were also reviewed for additional eligible RCTs.

Data collection

Two researchers independently extracted data using Microsoft Excel, including study characteristics (author, year, title), participant demographics (sample size, sex, age), intervention details, methodology, and outcomes. A third investigator verified all entries.

Participants post-randomization who were ineligible (e.g., baseline cancer), did not receive the intervention, or were lost to follow-up were excluded. As all trials had ≥ 3 years follow-up, a complete-case analysis was conducted. The inter-rater reliability (k = 0.82) indicated high agreement [17].

Evaluation of evidence certainty and bias risk

Risk of bias was assessed independently by two reviewers using the Cochrane Risk of Bias 2 (ROB2) tool [18], covering five domains: randomization, deviations from intended interventions, missing outcome data, outcome measurement, and selective reporting. Disagreements were resolved by discussion or third-party consultation.

Certainty of evidence was evaluated using the GRADE approach [19–21], with RCTs as the starting point. Outcomes were downgraded for risk of bias, inconsistency, indirectness, imprecision, or publication bias, ensuring transparent quality assessment.

Statistical analysis

For each outcome, relative risk (RR) and 95% confidence intervals (CI) were calculated, with a p-value of < 0.05 considered statistically significant.

Meta-analyses were conducted using the Stata Meta-Analysis package. Heterogeneity was assessed using Cochran’s Q-test and quantified using the I² statistic, with I² ≥ 50% and P < 0.1 indicating substantial heterogeneity. For outcomes with significant heterogeneity, additional exploratory analyses were performed, including Labbe plots, Galbraith plots, and influence analysis (metaninf), to visually examine potential sources of heterogeneity. A random-effects model was applied for all meta-analyses, as it accounts for between-study variability and provides more conservative and reliable results.

To further investigate the sources of heterogeneity, meta-regression analyses were conducted, evaluating potential effect modifiers such as baseline characteristics, follow-up duration, geographic region, mean age, and H. pylori eradication rates. A p-value < 0.05 in meta-regression indicated a statistically significant contribution of a covariate to heterogeneity. Additionally, subgroup analyses were performed to assess variations in treatment effects across different study populations and methodological characteristics.

For each outcome, the number needed to treat (NNT) and its 95% CI were computed using the formula: NNT = 1 / (ACR × (1-RR)), where the Absolute Control Risk (ACR) represents the incidence of the outcome in the control group. NNT indicates the number of individuals who need to receive the intervention rather than control treatment to prevent one additional adverse event (e.g., gastric cancer case). Publication bias was assessed using funnel plots and Egger’s regression test, with p < 0.05 indicating significant publication bias.

Result

Our literature search initially identified 2464 studies. After removing 918 duplicates, 1486 studies were excluded during abstract screening. Of the remaining studies, five were unavailable in full-text form. Additional exclusions based on intervention, design, and outcomes led to removal of 44 studies. Ultimately, 11 RCTs were included (Fig. 1). No further eligible studies were identified from gray literature or author contacts. Table 1 summarizes included study characteristics.

Fig. 1.

Literature screening flow chart

Table 1.

Table of basic information of included literature

Study	Location	Last point of follow-up	(Number of cancers) Number of treatment groups	(Number of cancers) Number of control groups	Technique employed to verify the existence of Helicobacter pylori	Regimen employed for the eradication of Helicobacter pylori	Method of ascertainment of stomach cancer cases	Characteristics of participants	Health status during the baseline period
Correa, P. 2000 Correa, P. 2001	Two communities in Narino Province, Colombia	6 years	(3)321	(2)309	Histological analysis of gastric biopsies procured during upper gastrointestinal endoscopy.	Bismuth subsalicylate 262 mg, amoxicillin 500 mg, and metronidazole 375 mg to be administered three times daily for a duration of two weeks.	Histological analysis of gastric biopsies collected during upper gastrointestinal endoscopy at the age of 6 years.	The average age is 51.1 years, with a range spanning from 29 to 69 years, and 46.1% of the population is male.	healthy individuals
Fukase, K. 2008	Fifty-one hospitals in Japan	3 years	(9)255	(24)250	Histological analysis and rapid urease testing were conducted on gastric biopsies collected during upper gastrointestinal endoscopy.	Lansoprazole 30 mg, amoxicillin 750 mg, and clarithromycin 200 mg to be administered twice daily for a duration of one week.	Histological analysis of gastric biopsies collected during upper gastrointestinal endoscopy conducted at a minimum of three years.	The median age is 68.5 years, with a range spanning from 20 to 79 years, and a male representation of 76.4%.	Patients diagnosed with early-stage gastric carcinoma who are receiving endoscopic mucosal resection
Wong, B. C. Y. 2012	Twelve villages in Linqu County, Shandong Province, China.	5 years	(6)553	(3)466	Carbon-urea breath testing.	Omeprazole 20 mg, amoxicillin 1 g, and clarithromycin 500 mg administered twice daily for a duration of one week.	Histological analysis of gastric biopsies collected during upper gastrointestinal endoscopy at the five-year mark.	The average age is 53.0 years, with a range spanning from 35 to 64 years, and 46.4% of the population is male.	healthy individuals
Zhou, L.Y. 2014	11 villages located in Yantai County, Shandong Province, China.	10 years	(2)185	(7)193	Histological analysis and rapid urease testing conducted on gastric biopsies acquired during upper gastrointestinal endoscopy.	Omeprazole 20 mg, amoxicillin 1 g, and clarithromycin 500 mg to be administered twice daily for one week.	Histological analysis of gastric biopsies collected during upper gastrointestinal endoscopy at the 2, 5, 8, and 10-year marks.	The average age is 52.0 years, with a range of 35 to 75 years, and 47.8% of the population is male.	healthy individuals
Kim, Y. I. 2016	One hospital in Goyang. South Korea.	9.4 years	(6)87	(7)82	Both the rapid urea test and histology prior to distal gastric surgery	Rabeprazole 10 mg, amoxicillin 1 g, and clarithromycin 500 mg twice daily for a duration of 1 week.	Engage in periodic gastroscopy to ascertain the possibility of cancer recurrence.	Media age 57.0 years (range 18–70 years), 69.2% male	Patients diagnosed with early-stage gastric carcinoma who are receiving endoscopic mucosal resection
Choi, J. M. 2018	One hospital in Seoul, South Korea.	6 years	(18)437	(36)440	Histological analysis and swift urease testing were conducted on gastric biopsies acquired during upper gastrointestinal endoscopy.	Omeprazole 20 mg, amoxicillin 1 g, and clarithromycin 500 mg were administered b.i.d. for a period of one week.	Histological analysis of gastric specimens obtained during upper gastrointestinal endoscopy.	The average age is 60.4 years, with a range spanning from 20 to 75 years, and a male representation of 67.7%.	Patients diagnosed with early-stage gastric carcinoma who are receiving endoscopic mucosal resection
Choi, I. J. 2018	One hospital in Goyang, South Korea.	5.9 years	(14)194	(27)202	Histological analysis and rapid urease testing conducted on gastric biopsies acquired during upper gastrointestinal endoscopy.	Rabeprazole 10 mg, amoxicillin 1 g, and clarithromycin 500 mg twice daily for a duration of one week.	Histological analysis of gastric biopsies collected during upper gastrointestinal endoscopy at the 3-month and 3-year marks.	The average age is 59.8 years, with a range from 18 to 75 years, and 75.3% of the participants are male.	Patients diagnosed with early-stage stomach cancer who are receiving endoscopic mucosal resection
Li, W. Q. 2019	Thirteen villages in Linqu County, Shandong Province, China.	22.3 years	(41)1130	(78)1128	Serological testing.	Omeprazole 20 mg and amoxicillin 1 g administered twice daily for a duration of two weeks.	Histological analysis of gastric biopsies acquired during upper gastrointestinal endoscopy, or derived from clinical, laboratory, or pathological data.	The average age is 46.8 years (range 35–64 years), with 50.0% males.	healthy individuals
Choi, I. J. 2020	One hospital in Goyang, South Korea.	9.2 years	(10)832	(23)844	Histological analysis and rapid urease testing conducted on gastric biopsies acquired during upper gastrointestinal endoscopy	Lansoprazole 30 mg, amoxicillin 1 g, and clarithromycin 500 mg administered twice daily for a duration of one week.	Utilization of the Korean National Cancer Incidence Database is essential to verify all instances of stomach cancer identified through endoscopic surveillance throughout the duration of the trial.	The average age is 48.8 years, with a range spanning from 40 to 65 years, and the male population constitutes 49.5% of the total.	healthy individuals
Yan, L. 2022	Seven settlements located in Changle County, Fujian Province, China.	26.5 years	(21)817	(35)813	Histological analysis and rapid urease testing conducted on gastric biopsies acquired during upper gastrointestinal endoscopy.	Omeprazole at a dosage of 20 mg, coamoxiclav at 750 mg, and metronidazole at 400 mg, twice daily for a duration of 2 weeks.	Histological analysis of gastric biopsies collected during upper gastrointestinal endoscopy at 7.5 years, or, in cases diagnosed prior to this age, an evaluation of clinical records and pathology specimens conducted by three blinded clinicians.	The average age is 42.2 years, with a range spanning from 35 to 65 years, and a male representation of 54.0%.	healthy individuals
Pan, K. F. 2024	980 villages located in Linqu County, Shandong Province, China.	11.8 years	(354)47,601	(399)47,647	Carbon-urea breath testing.	Consisting of 20 mg of omeprazole, 750 mg of tetracycline, 400 mg of metronidazole, and 300 mg of bismuth citrate for a duration of 10 days.	The trial’s Database Management System was connected to the internet-based Cancer Registry in Linqu. Staff conducted visits to each township hospital every three months to collect data on cancer incidence and cancer-specific mortality, while village visits occurred every six months to verify new cancer events.	Mean age 42.5 years (range 25–54 years), 46.2% male.	healthy individuals

Risk of bias assessment

Risk of bias was assessed using ROB2 across five domains in 11 RCTs. Overall, four studies had low risk, four raised some concerns, and three had high risk of bias. Eight trials showed low bias risk for randomization, with three causing some concerns. For deviations from interventions, eight had low risk, one raised concerns, and two had high risk. Regarding missing outcome data, nine trials had low risk, one had concerns, and one had high risk. Outcome measurement bias was low in nine studies, with two causing some concerns. All trials had low bias in reporting results (Appendix 3).

GRADE evidence quality assessment

Although all included studies were RCTs, evidence certainty for gastric cancer incidence was moderate due to allocation concealment, blinding issues, incomplete reporting, moderate heterogeneity, and indirectness from studies mostly conducted in high-incidence areas. Publication bias was low.

For secondary outcomes (all-cause and gastric cancer-specific mortality), certainty was also moderate due to bias risks, indirectness, and imprecision from wide confidence intervals. Minimal publication bias and high consistency were noted, but further high-quality studies are needed. GRADE details are in Appendix 2.

Primary outcome: gastric Cancer incidence

Eleven studies were included for gastric cancer incidence. Moderate heterogeneity (I² = 53.5%, p = 0.018) was observed visually (Labbe, Galbraith plots). Using a random-effects model, H. pylori eradication significantly reduced gastric cancer incidence (RR = 0.61, 95% CI:0.47–0.79, p < 0.001), indicating a 39% lower risk versus controls.

Sensitivity analysis indicated no undue influence from individual studies (Appendix 4). Meta-regression found no significant sources of heterogeneity from factors such as baseline cancer prevalence, follow-up duration, region, age, or eradication rates (Appendix 5).

Subgroup analysis showed risk reduction was significant in healthy individuals (RR = 0.67, 95% CI:0.48–0.93, NNT = 476) and more pronounced among those post-endoscopic mucosal resection (RR = 0.51, 95% CI:0.36–0.71, NNT = 21). Figure 2 illustrates subgroup details.

Fig. 2.

Subgroup analysis of the incidence rate of stomach cancer. 0: healthy individuals; 1: Individuals with early stomach cancer undergoing endoscopic mucosal resection

Funnel plot suggested slight asymmetry, but Egger’s test showed no significant publication bias (p = 0.08; Fig. 3).

Fig. 3.

Funnel plot

Secondary outcomes

Six studies evaluated all-cause mortality. No significant heterogeneity (I² = 0%, p = 0.763) was found. Pooled analysis (fixed-effects) indicated no significant impact of eradication therapy (RR = 1.00, 95% CI:0.89–1.13, p = 0.956; Fig. 4).

Fig. 4.

Forest plot of the fixed-effect model for all-cause mortality

For gastric cancer-specific mortality (six studies), heterogeneity was negligible (I² = 0%, p = 0.638). Pooled analysis showed a nominal, non-significant reduction (RR = 0.84, 95% CI:0.69–1.01, p = 0.059; Fig. 5). Results suggest a potential protective effect, though additional studies are needed.

Fig. 5.

Forest plot of the fixed-effect model for stomach cancer mortality rate

Funnel plots for mortality outcomes showed no significant asymmetry; Egger’s tests for all-cause (p = 0.11) and cancer-specific mortality (p = 0.43) supported absence of publication bias (Fig. 5).

Summary of findings

This meta-analysis provides moderate-quality evidence supporting H. pylori eradication to reduce gastric cancer incidence, particularly after endoscopic resection. Effects on gastric cancer-specific mortality were suggestive but not statistically significant, and no effect was found for all-cause mortality. These findings highlight the potential preventive role of eradication therapy, but additional high-quality RCTs with longer follow-up are warranted.

Discussion

This meta-analysis provides updated evidence on the long-term effects of H. pylori eradication therapy in reducing stomach cancer incidence and stomach cancer-specific mortality. Our findings demonstrate distinct differences in outcomes across various patient subgroups, underscoring the heterogeneity in treatment effects. Eradication therapy significantly lowers the risk of developing stomach cancer in both asymptomatic individuals and those who have undergone endoscopic resection for early-stage malignancies, underscoring its potential preventive role across diverse patient populations. Our results are consistent with the findings from a recent meta-analysis [22], which also reported a significant reduction in gastric cancer incidence with H. pylori eradication therapy in both healthy individuals and patients with gastric neoplasia undergoing EMR. However, while both studies demonstrate the protective effects of eradication therapy, the RR reported in our study (0.67 for healthy individuals and 0.51 for post-resection patients) are slightly higher than those reported in the recent meta-analysis (0.64 for healthy individuals and 0.52 for patients with gastric neoplasia undergoing EMR).

Notably, the magnitude of risk reduction was more pronounced in patients with a history of endoscopic mucosal resection (RR: 0.51; 95% CI: 0.36–0.71; NNT = 21) compared to healthy individuals (RR: 0.67; 95% CI: 0.48–0.93; NNT = 476). This suggests that individuals with a higher baseline risk may derive greater benefit from eradication therapy, potentially due to a higher likelihood of harboring premalignant lesions. In the recent meta-analysis, the NNT for healthy individuals was reported as 72, which is considerably lower than the 476 observed in our study [22]. This discrepancy may be attributed to differences in study populations, geographic regions, and treatment protocols.

Additionally, our results confirm that H. pylori eradication is associated with a reduction in stomach cancer-related mortality, although this effect varied across subgroups. While a modest reduction was observed in overall stomach cancer-specific mortality (RR: 0.83; 95% CI: 0.69–1.01), this effect was more evident in post-resection patients, suggesting that early intervention following tumor resection may confer additional survival benefits. However, no significant impact was observed on overall mortality, which may be attributable to variations in baseline risk, comorbidities, and non-gastric causes of mortality among different study populations. This disparity highlights the need for further research to clarify whether H. pylori eradication exerts broader survival benefits beyond gastric cancer prevention.

By incorporating newly available RCTs and updated long-term follow-up data, our study provides a refined and more comprehensive assessment of the clinical benefits of H. pylori eradication therapy, reinforcing its protective role yet highlighting the potential limitations of its impact beyond gastric cancer.

With regard to the number needed to treat (NNT), our study reported an overall NNT of 332, whereas the NNT for healthy individuals was 476. These values are considerably higher than the NNT of 72 for healthy individuals reported by Ford et al. [12]. However, among post-resection patients, the NNT was markedly lower at 21, reflecting the greater absolute benefit in this high-risk population. Several factors may explain these discrepancies. First, our meta-analysis incorporated a newly published large-scale RCT [23] with 95,248 participants, which significantly influenced the pooled estimates. Second, we included updated long-term follow-up data from an existing trial [24] along with data from an additional new trial [25]. These more recent datasets suggest that the protective effect of H. pylori eradication therapy in healthy individuals may not be as pronounced as previously estimated, potentially due to differences in participant demographics, regional prevalence of H. pylori, or adherence to treatment protocols.

This study has several limitations that should be acknowledged. First, although the study by Saito et al. was not included in our meta-analysis due to methodological constraints, it provided valuable insights into the potential benefits of H. pylori eradication in reducing precancerous lesions [24]. Their findings suggest that eradication therapy may contribute to histological improvements, particularly in patients with gastric intestinal metaplasia (GIM). However, due to the lack of long-term follow-up data on gastric cancer incidence and survival outcomes, direct comparability with the studies included in our analysis was not feasible. Given that the presence of premalignant lesions is a key determinant of gastric cancer risk, future research should aim to bridge this gap by integrating histological findings with long-term clinical outcomes to better inform prevention strategies.

Second, only one trial in our analysis was conducted outside East Asia [26], suggesting that the applicability of H. pylori eradication therapy to populations in other regions remains uncertain. Most included trials were conducted in high-incidence regions such as rural areas in Fujian and Guangdong, China, as well as South Korea and Japan [27–29]. This geographic concentration may limit the generalizability of our findings to lower-incidence regions, where the benefits of eradication therapy might differ. Regional variations in gastric cancer risk, dietary patterns, and H. pylori virulence factors could all contribute to the observed heterogeneity.

Third, the extended follow-up durations of the included studies led to a high rate of patient attrition, which may affect the reliability of the findings. Significant discrepancies were observed in patient populations across different studies, particularly in terms of the number of participants included and those who successfully completed follow-up. These variations contributed to the heterogeneity of the results. Additionally, differences in the implementation of eradication therapy across studies, including variations in treatment duration, eradication regimens, and effectiveness, may have influenced the overall estimates of risk reduction, potentially introducing bias. Further standardization of eradication protocols across clinical trials is warranted to minimize these confounding effects.

When analyzing stomach cancer incidence, moderate heterogeneity was detected among the included studies. However, meta-regression analysis did not identify any specific factors responsible for this variability. Although we used a random-effects model to minimize the impact of heterogeneity and conducted subgroup analyses, residual variability may still influence the robustness of our findings. The lack of a significant association between H. pylori eradication and all-cause mortality further suggests that other competing risks may obscure the long-term benefits of eradication therapy. Additionally, our decision to include only English-language studies may have excluded relevant research published in other languages, potentially introducing selection bias. Taken together, these differences in study design, participant demographics, and treatment protocols could account for the varying outcomes observed across trials and highlight the need for continued investigation into context-specific factors influencing H. pylori eradication efficacy.

From a clinical perspective, our findings provide valuable guidance on the role of H. pylori eradication in stomach cancer prevention, particularly in East Asia, where nearly half of the world’s new and fatal stomach cancer cases occur annually. Given the differential benefits observed across subgroups, clinicians should consider individual patient risk factors, such as previous endoscopic resection history and the presence of precancerous lesions, when determining the appropriateness of eradication therapy. Previous research has shown that compared to standard care and periodic gastroscopy, H. pylori eradication therapy is the most cost-effective strategy for stomach cancer prevention [30]. However, the relatively high NNT reported in this study suggests that the economic feasibility of large-scale eradication programs may be less promising than initially expected. Nevertheless, eradication therapy remains a highly effective measure for reducing stomach cancer incidence and mortality, making it a crucial component of risk mitigation strategies in high-incidence populations.

Beyond its role in stomach cancer prevention, H. pylori eradication therapy has also been shown to be cost-effective in preventing other gastrointestinal diseases, such as peptic ulcers and functional dyspepsia. This broader clinical utility may further increase physician adherence to eradication programs, thereby indirectly contributing to stomach cancer prevention [31]. In Japan, the inclusion of H. pylori eradication therapy in national health insurance coverage since 2013 has had a demonstrable impact on reducing stomach cancer incidence. This example underscores the potential benefits of integrating eradication therapy into public health policies in other high-incidence countries [32].

The evidence from our analysis strongly supports the conclusion that H. pylori eradication therapy significantly reduces the incidence and mortality of stomach cancer among at-risk populations. In clinical practice, especially in high-incidence regions and among high-risk individuals, systematic screening for H. pylori and prompt eradication therapy should be prioritized. Clinicians can use the NNT values reported in this study to evaluate the balance between treatment costs and benefits, enabling more precise incorporation of eradication therapy into patient-specific prevention strategies. Further well-designed, long-term RCTs with diverse populations and standardized eradication protocols are needed to refine global recommendations for H. pylori eradication in stomach cancer prevention.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Zhouhan Wu conceived the study, designed the research, and wrote the initial draft. Yonghui Xu, Meiwen Tang, Zhoutong Wu contributed to data collection and analysis. Yi Tang revised the manuscript critically for important intellectual content. All authors read and approved the final manuscript.

Funding

National Natural Science Foundation of China (82360959) will fund the publish, The study and manuscript were done independently from funding sources.

Data availability

The data and materials used in this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors have given their consent for publication.

Competing interests

The authors declare no competing interests.