Is Helicobacter pylori infection associated with pancreatic cancer? A systematic review and meta-analysis of observational studies

Abstract

Background and Objectives:

Recent observational studies have investigated the association between Helicobacter pylori (H. pylori) infection and pancreatic cancer with conflicting data. Therefore, we conducted a systematic review and meta-analysis to assess the potential association.

Design:

This is a systematic review and meta-analysis.

Methods:

We searched three databases (PubMed, Embase, and Web of Science) from inception to 30 August 2022. The summary results as odds ratio (OR) or hazard ratio (HR) with 95% confidence interval (CI) were pooled by generic inverse variance method based on random-effects model.

Results:

A total of 20 observational studies involving 67,718 participants were included in the meta-analysis. Meta-analysis of data from 12 case–control studies and 5 nested case–control studies showed that there was no significant association between H. pylori infection and the risk of pancreatic cancer (OR = 1.20, 95% CI = 0.95–1.51, p = 0.13). Similarly, we also did not find significant association between cytotoxin-associated gene A (CagA) positive strains, CagA negative strains, vacuolating cytotoxin gene A (VacA) positive strains H. pylori infection, and the risk of pancreatic cancer. Meta-analysis of data from three cohort studies showed that H. pylori infection was not significantly associated with an increased risk of incident pancreatic cancer (HR = 1.26, 95% CI = 0.65–2.42, p = 0.50).

Conclusion:

We found insufficient evidence to support the proposed association between H. pylori infection and increased risk of pancreatic cancer. To better understand any association, future evidence from large, well-designed, high-quality prospective cohort studies that accounts for diverse ethnic populations, certain H. pylori strains, and confounding factors would be useful to settle this controversy.

Keywords

meta-analysis observational studies pancreatic cancer systematic review

Introduction

Pancreatic cancer is a major health concern worldwide, with a mortality rate that is nearly equal to the number of cases. It is the seventh leading cause of cancer death in both male and female.^1,2 During the past two and a half decades, the incidence, prevalence, and mortality of pancreatic cancer have all gone up significantly, with respective increases of 55%, 63%, and 53%.³ Surgical resection is the only method of treatment that can result in a cure; however, the 5-year survival rate postsurgery is exceedingly low, estimated to be around 10%.⁴ In view of unsatisfactory treatment effects, low survival rate of pancreatic cancer, and high mortality, the etiological prevention has become an important measure to prevent pancreatic cancer. The etiology of pancreatic cancer remains unknown, however. Environmental, lifestyle risk factors and genetic susceptibility may lead to the development of pancreatic cancer. Previous research has investigated several established risk factors of pancreatic cancer, including advanced age, male sex, smoking, obesity, diabetes, a family history of pancreatic cancer, chronic pancreatitis, and ABO blood group.^5–7 Nevertheless, established risk factors only explain about 40% of the increase in incidence and prevalence of pancreatic cancer in the United Kingdom.⁸ Therefore, further studies on its etiology are urgently needed.

Helicobacter pylori (H. pylori), a spiral, microaerophilic, gram-negative bacterium, colonizes the luminal surface of human gastric epithelium and its infection rate is at least 50% of the global population.^9,10 It is generally acknowledged that H. pylori infection is not only associated with chronic gastritis, peptic ulcers, and gastric cancer, but also may be related to many extragastrointestinal diseases.¹¹ In recent years, the link between H. pylori infection and pancreatic cancer has attracted more and more attention. Previously, several early meta-analyses^12–18 published between 2011 and 2017 have yielded conflicting and controversial results. Furthermore, the majority of their studies pooled data using data unadjusted for confounders, which would misinterpret their association of H. pylori with pancreatic cancer. In addition, a growing body of new relevant studies on the topic have been published with inconsistent conclusions in recent years,^19–23 and it is still unknown if H. pylori infection is correlated to increased risk of pancreatic cancer.

Based on these above considerations, we conducted an updated meta-analysis of observational studies to assess the association between H. pylori infection and risk of pancreatic cancer. Our aim was to determine whether, and to what extent, H. pylori may be associated with this risk. Clarification of the nature and magnitude of risk of pancreatic cancer associated with H. pylori infection might have important clinical implications for primary preventative strategies against the cancer.

Materials and methods

Protocol of registration

This study was registered on PROSPERO platform (registration no. CRD42022357054) and adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2020 statement and Meta-analysis of Observational Studies in Epidemiology (MOOSE) reporting guidelines.^24,25 No ethical approval was required as the analyzed data had been published previously.

Search strategy

Electronic searches were conducted in PubMed, Embase, and Web of Science from inception through 30 August 2022. The search strategy for PubMed was as follows: (‘helicobacter’ [MeSH Terms] OR ‘helicobacter’ [All Fields] OR ‘helicobacter pylori’ [MeSH Terms] OR ‘helicobacter pylori’ [All Fields] OR ‘helicobacter infections’ [MeSH Terms] OR H. pylori [All Fields] OR HP [All Fields] OR ‘campylobacter pylori’ [All Fields]) AND ((‘pancreas’ [MeSH Terms] OR ‘pancreas’ [All Fields]) OR (‘pancreatic’ [All Fields])) AND ((‘neoplasms’ [MeSH Terms] OR ‘neoplasms’ [All Fields] OR ‘neoplasm’ [All Fields]) OR (‘cancer’ [All Fields] OR ‘cancers’ [All Fields]) OR (‘tumour’ [All Fields] OR ‘tumor’ [All Fields]) OR (‘tumours’ [All Fields] OR ‘tumors’ [All Fields]) OR (‘carcinoma’ [MeSH Terms] OR ‘carcinoma’ [All Fields]) OR (‘adenocarcinoma’ [MeSH Terms] OR ‘adenocarcinoma’ [All Fields])). In addition, manual searches would also be performed by reviewing the reference lists of relevant studies to locate further resources. We used appropriate Medical Subject Heading terms in conjunction with text word searching. There is no language restriction. The search terms of the remaining databases (EMBASE and Web of Science) were presented in the Online Supplementary Materials Table S1.

Selection criteria

The inclusion criteria for this study were as follows: (1) case–control and cohort studies investigating the relationship between H. pylori infection and risk of pancreatic cancer; (2) H. pylori infection was determined by serological test such as enzyme-linked immunosorbent assay (ELISA), westerning blotting (WB), or another reliable methods; (3) the diagnostic method of pancreatic cancer should be stated; (4) the study presented the odds ratio (OR), hazard ratio (HR), and risk ratio (RR) with 95% confidence interval (CI), or the data presented in the study was adequate to calculate them. Exclusion criteria included conference abstracts, case series, animal studies, comments, editorials, letters, reviews and meta-analyses, as well as studies without sufficient data. In cases in which multiple studies utilized the same database or cohort, or had partially overlapping participants, the study with the most comprehensive results or the most updated study was included. Two investigators (B.-G.Z. and Y.-Z.M.) conducted an independent review of the studies by examining their titles and abstracts. The articles in full text were further obtained potentially relevant articles and reviewed for inclusion in the final analysis. Discrepancies were addressed through discussion.

Data extraction and quality assessment

B.-G.Z. and J.-S.W., two investigators, independently extracted the data and appraised the methodological quality from all included studies. Any disagreements were reconciled through consensus. From each eligible study, the following information was obtained: the surname of the first author, year of publication, study design, study country, sample size, mean age, study period, H. pylori diagnostic method, ascertainment of pancreatic cancer, OR, RR, HR with their 95% CI, relevant data of cytotoxin-associated gene A (CagA) and vacuolating cytotoxin gene A (VacA), and confounders adjustment. To appraise the methodological quality of case–control and cohort studies, the Newcastle-Ottawa Scale (NOS)²⁶ was utilized. This scale is divided into three sections: selection, comparability, and outcome of interest (for cohort studies) or exposure (for case–control studies). The quality of the studies was rated using a star system, with a maximum score of nine stars. Research was deemed to be of high quality if the total score was ⩾7, while studies received 4–6 points were qualified as moderate quality. Studies with score <4 were considered to be of low quality.²⁷

Statistical analysis

The Review Manager software 5.3 (Version 5.3, The Cochrane Collaboration, Denmark) and STATA/SE 12.0 (StataCorp, College Station, TX, USA) were utilized to conduct all statistical analyses. A generic inverse variance method of DerSimonian and Laird based on a random-effect model²⁸ was used to pool OR (for case–control studies) or HR (for cohort studies) with 95% CI. Random-effects model was used as it attempted to generalize findings beyond the included studies by assuming that the selected studies are random samples from a larger population.²⁹ Whenever feasible, we utilized adjusted ratios (i.e. those that accounted for the highest degree of adjustment for potential confounding variables) unless only unadjusted data were available. In order to assess the heterogeneity among studies, Cochran’s Q test and I² statistic were utilized. A p value of less than 0.10 on the Q test was deemed to have statistically significant heterogeneity. The degree of heterogeneity is determined by the I² values, which vary from 0–25%, 26–50%, 51–75%, and more than 75%, representing insignificant, low, moderate, and high heterogeneity, respectively.³⁰ Subgroup analyses were conducted to explore potential sources of heterogeneity and to assess the influence of various factors on the overall results. To confirm the robustness of overall results, sensitivity analyses were performed by (1) removing one study at a time, (2) eliminating studies published in non-English, and (3) excluding studies with less than 500 participants. To detect any possible publication bias, Begg’s test and Egger’s test were conducted if the overall meta-analysis comprised of at least 10 studies.^31–33 The trim-and-fill method utilized to evaluate the impact of potential missing articles or small-study effects caused by publication bias on the overall results.³⁴

Results

Study selection process

Initially, we identified 5247 records from three electronic databases. After removing 1352 duplicates, 3844 records were eliminated by examining titles or abstracts. A total of 51 records were then evaluated, with 31 articles were excluded after scrutinizing abstracts or full texts carefully (see Online Supplementary Material Table S2). Ultimately, 20 studies^{16,19–23,35–48} were included in our meta-analysis. The study selection process was presented in Figure 1.

Figure 1.

PRISMA flowchart of study selection process.

Study characteristics

Table 1 presents the main characteristics of included studies. In total, 20 studies with 67,718 participants were included in our meta-analysis. All these studies were published between 1998 and 2022. There were 12 case–control studies,^{16,19,21–23,35,38,40–42,45,46} 5 nested case–control studies,^{36,37,39,43,48} and 3 cohort studies.^20,44,47 Six studies^{36,39,42,43,47,48} were undertaken in Europe (Finland, Sweden, Poland, Germany, the United Kingdom, France, Italy, Denmark, Greece, the Netherlands, Spain, and Norway), three studies in North America (the United States),^21,37,40 nine studies in Asia (China, Japan, and India),^{19,20,22,23,38,41,44–46} and two studies in Oceania (Australia).^16,35 Regarding publication language, one study was published in Chinese,³⁸ one in Spanish,⁴² while the rest were published in English. The sample size of case–control and cohort studies ranged from 53 to 30,110 participants. The average age of the individuals was between 48 and 74 years old. The diagnosis of H. pylori infection was based on serology by ELISA (n = 16 studies), immunohistochemistry and western blotting (n = 1 case–control study), multiplex serology assay (n = 2 studies), and histopathology (n = 1 cohort study). The identification of pancreatic cancer is mainly achieved through pathologic diagnosis, International Classification of Diseases (ICD) code, imaging techniques, and clinical diagnosis. The characteristics of included studies about sex, study period, confounders’ adjustment, and corresponding data in our meta-analysis are summarized in Online Supplementary Material Table S3. As shown in Online Supplementary Material Table S4, the NOS scores of the studies that were included ranged from 5 to 9, with an average of 7.1. A total of 13 studies were appraised as high quality, while the other 7 were found to be of moderate quality.

Table 1.

Main characteristics of included studies in the meta-analysis.

Author	Country, Language	Study location	Study design	Sample size	Mean age/age range (years)	H. pylori detection method	Ascertainment of pancreatic cancer	NOS score
Raderer et al.³⁵	Austria, English	Oceania	Case–control	154	58/56	Serum IgG by ELISA	Pathologic diagnosis	5
Stolzenberg-Solomon et al.³⁶	Finland, English	Europe	Nested case–control	347	58/58	Serum IgG and CagA by ELISA	Pathologic diagnosis	7
de Martel et al.³⁷	The United States, English	North America	Nested case–control	366	50/50	Serum IgG and CagA by ELISA	Tumor registry reports	8
Dou and Wang³⁸	China, Chinese	Asia	Case–control	221	53/53	Serum IgG by ELISA	Clinical and pathologic diagnosis	6
Lindkvist et al.³⁹	Sweden, Spanish	Europe	Nested case–control	350	48/48	Serum IgG by ELISA	ICD-7 code	8
Risch et al.⁴⁰	The United States, English	North America	Case–control	1063	67/68	Serum IgG and CagA by ELISA	Clinical and pathologic diagnosis	8
Shimoyama et al.⁴¹	Japan, English	Asia	Case–control	53	67/62	Serum IgG by ELISA	Medical records	5
Gawin et al.⁴²	Poland, Polish	Europe	Case–control	316	63/58	Serum IgG by ELISA and CagA by WB	Pathologic diagnosis and imaging techniques	7
Yu et al.⁴³	Finland, English	Europe	Nested case–control	706	57/57	Multiplex serology assay	Medical records	7
Hsu et al.⁴⁴	China (Taiwan), English	Asia	Cohort study	30,110	51.0	Histopathology	ICD-9 code	8
Risch et al.⁴⁵	China, English	Asia	Case–control	1555	65/65	Serum IgG and CagA by ELISA	Pathologic diagnosis, medical records	8
Ai et al.⁴⁶	China, English	Asia	Case–control	116	57/55	Serum IgG and IgM by ELISA, CagA by WB	Pathological diagnosis or clinical diagnosis	6
Schulte et al.¹⁶	Australia, English	Oceania	Case–control	1206	67/67	Serum IgG and CagA by ELISA	Pathological diagnosis or clinical diagnosis	8
Chen et al.⁴⁷	Germany, English	Europe	Cohort study	9504	50–75	Serum IgG and CagA by ELISA	ICD-10 code	9
Huang et al.⁴⁸	10 European countries,a English	Europe	Nested case–control	896	58/58	Serum IgG and CagA by ELISA	ICD-10 code	8
Geng et al.¹⁹	China, English	Asia	Case–control	138	56/54	Immunohistochemistry and WB	Pathological diagnosis and clinical diagnosis	6
Hirabayashi et al.²⁰	Japan, English	Asia	Cohort study	20,116	57	Serum IgG by ELISA	ICD code	9
Permuth et al.²¹	The United States, English	North America	Case–control	262	68/59	Multiplex serology assay	Pathological diagnosis	8
Laya et al.²²	India, English	Asia	Case–control	155	56/53	Serum IgG and CagA by ELISA	Imaging techniques or histopathology report	6
Osaki et al.²³	Japan, English	Asia	Case–control	84	69/74	Serum IgG by ELISA	Imaging techniques or pathological diagnosis	5

CagA, cytotoxin-associated gene A; ELISA, enzyme-linked immunosorbent assay; H. pylori, Helicobacter pylori; ICD, International Classification of Diseases; IgG, immunoglobulin G; IgM, immunoglobulin M; NOS, Newcastle-Ottawa Scale; WB, western blotting.

Included Denmark, France, Germany, Greece, Italy, the Netherlands, Spain, Norway, Sweden, and the United Kingdom.

H. pylori infection and risk of pancreatic cancer from case–control studies

Overall meta-analysis

A total of 17 studies^{16,19,21–23,35–48} (12 case–control studies and 5 nested case–control studies) with 7988 participants investigated the relationship between H. pylori infection and the risk of pancreatic cancer. The pooled meta-analysis revealed no significant association between them (OR = 1.20, 95% CI = 0.95–1.51, p = 0.13). The analysis revealed a moderate degree of heterogeneity (I² = 70%, p < 0.00,001) (Figure 2).

Figure 2.

Forest plot of overall meta-analysis of association between H. pylori infection and risk of pancreatic cancer from case–control studies.

To investigate the potential sources of statistical heterogeneity and factors influencing the overall results, we conducted subgroup analyses according to study design, study location, publication year, H. pylori detection method, and adjustment of confounders. No significant association was observed when the subgroup analyses were conducted based on study design, study location, publication year, and H. pylori detection method. In subgroup analysis stratified by whether to adjust the confounding factors, no significant association were observed based on data from adjusted confounders (adjusted OR = 0.99, 95% CI = 0.79–1.26, p = 0.96). Data from unadjusted confounders, however, showed that H. pylori infection was significantly linked to an increased risk of pancreatic cancer (unadjusted OR = 1.83, 95% CI = 1.33–2.52, p = 0.0,002). The results of subgroup analyses were presented in Table 2 and Supplementary Material Figures S1–S5.

Table 2.

Subgroup analyses of case–control studies.

Subgroup	Number of studies	OR (95% CI)	p _association	I² (%)	p _{heterogeneity}
Overall studies	17	1.20 (0.95–1.51)	0.13	70	<0.00,001
Study design
Case–control	12	1.27 (0.91–1.76)	0.16	77	<0.00,001
Nested case–control	5	1.08 (0.83–1.39)	0.58	29	0.23
Study location
Europe	5	1.14 (0.88–1.46)	0.32	21	0.28
North America	3	0.98 (0.62–1.54)	0.93	51	0.13
Asia	7	1.36 (0.74–2.49)	0.32	83	<0.00,001
Oceania	2	1.37 (0.67–2.81)	0.39	76	0.04
Publication year
<2015	10	1.20 (0.86–1.67)	0.28	76	<0.0,001
⩾2015	7	1.19 (0.84–1.69)	0.32	60	0.02
H. pylori detection method
Serology by ELISA	14	1.22 (0.95–1.57)	0.13	72	<0.0,001
Multiple detection method	3	1.08 (0.50–2.35)	0.84	73	0.02
Adjustment for confounders
Adjusted	10	0.99 (0.79–1.26)	0.96	66	0.002
Unadjusted (crude)	7	1.83 (1.33–2.52)	0.0,002	15	0.31

CI, confidence interval; ELISA, enzyme-linked immunosorbent assay; OR, odds ratio.

We conducted sensitivity analyses to assess the robustness of overall results and explore potential sources of heterogeneity. We found that the magnitude and direction of the summary estimates remained unchanged when eliminating studies published in non-English and studies with <500 participants. Upon performing sensitivity analyses by omitting single study at a time and combining the remaining studies, the overall results did not significantly vary except when excluding the study conducted by Risch et al.⁴⁵ Notably, when excluding the study conducted by Risch et al.,⁴⁵ the I² was reduced to 45%, indicating that the study could be the origin of statistical heterogeneity. These results of sensitivity analyses were summarized in Online Supplementary Material Table S5.

Regarding publication bias, Begg’s funnel plot displayed slightly asymmetrical distribution (Online Supplementary Material Figure S6), yet Begg’s test revealed no evidence of substantial publication bias (p_Begg = 0.650). Conversely, Egger’s test indicated the presence of publication bias (p_Egger = 0.022). To evaluate the impact of publication bias on the outcome of the meta-analysis, the trim-and-fill method was utilized. The result of the trim-and-fill method (filled OR = 0.99, 95% CI = 0.79–1.26, p = 0.958) did not significantly modify the overall results, suggesting that the outcome of our meta-analysis was robust and reliable.

CagA positive H. pylori infection and risk of pancreatic cancer

The result of the meta-analysis of 12 studies^{16,21,22,36–38,40,42,43,45,46,48} on the association between CagA positive H. pylori infection and risk of pancreatic cancer showed no significant association (OR = 1.03, 95% CI = 0.83–1.29, p = 0.78). Moderate heterogeneity was observed (I² = 55%, p = 0.01) (Figure 3).

Figure 3.

Forest plot of association between CagA positive H. pylori infection and risk of pancreatic cancer from case–control studies.

CagA negative H. pylori infection and risk of pancreatic cancer

The result of the meta-analysis of six studies^{16,36,37,40,45,48} indicated no significant association between CagA negative H. pylori infection and risk of pancreatic cancer (OR = 1.22, 95% CI = 1.00–1.49, p = 0.05). No heterogeneity was observed (I² = 0%, p = 0.52) (Online Supplementary Material Figure S7).

VacA positive H. pylori infection and risk of pancreatic cancer

The meta-analysis of the data from three studies^21,38,43 demonstrated a lack of significant association between VacA positive H. pylori infection and risk of pancreatic cancer (OR = 1.22, 95% CI = 0.66–2.25, p = 0.52). The heterogeneity of the analysis was moderate (I² = 62%, p = 0.07) (Online Supplementary Material Figure S8).

H. pylori infection and risk of incident pancreatic cancer from cohort studies

The meta-analysis of three cohort studies^20,44,47 with 59,730 participants, investigating the association between H. pylori infection and the risk of incident pancreatic cancer, revealed no significant increase in the risk (HR = 1.26, 95% CI = 0.65–2.42, p = 0.50). Heterogeneity was assessed to be moderate (I² = 69%, p = 0.04) (Figure 4). These findings were based on the adjusted HR (adjusted for various confounders). The results of sensitivity analyses remained largely unchanged when each study was removed in turn (Online Supplementary Material Table S6). In addition, one study⁴⁷ identified the effect of CagA on pancreatic cancer. The results showed that there was no association between CagA positive or CagA negative H. pylori infection and risk of incident pancreatic cancer (CagA positive: HR = 1.08, 95% CI = 0.52–2.24; CagA negative: HR = 1.61, 95% CI = 0.82–3.18).

Figure 4.

Forest plot of association between H. pylori infection and risk of incident pancreatic cancer from cohort studies.

Discussion

This updated meta-analysis aimed to synthesize all available evidence regarding the relationship between H. pylori infection and the risk of pancreatic cancer. Meta-analysis of data from 17 case–control studies revealed no substantial association between H. pylori infection and the risk of pancreatic cancer (OR = 1.20, 95% CI = 0.95–1.51, p = 0.13). The subgroup analyses revealed that the results were consistent with overall results regardless of study design, study location, year of publication, and H. pylori detection method. Similar result was observed from the meta-analysis of three cohort studies (HR = 1.26, 95% CI = 0.65–2.42, p = 0.50). Given that CagA and VacA are the primary virulence factor of H. pylori,⁴⁹ we further analyze the effect of H. pylori infection with CagA positive strains, CagA negative strains, and VacA positive strains on pancreatic cancer. Our findings, however, failed to demonstrate any considerable correlation between CagA positive and negative strains, VacA positive strains H. pylori infection, and the risk of pancreatic cancer.

Our meta-analysis is, to the best of our knowledge, the largest, most updated, and most comprehensive assessment regarding the relationship between H. pylori infection and risk of pancreatic cancer. Previously, Trikudanathan et al.¹² conducted a meta-analysis of six case–control studies and found evidence to support a positive correlation between H. pylori infection and risk of pancreatic cancer. Subsequently, two meta-analyses conducted by Xiao et al.¹³ (n = 9 studies) and Guo et al.¹⁷ (n = 8 studies) yielded similar results. Compared with previous three positive results of meta-analyses, this study encompassed the majority of their studies, excluding a few of lower quality. In 2014, Wang et al.¹⁴ conducted a meta-analysis of nine observational studies (four nested case–control and five case–control studies) which concluded that H. pylori and CagA positive infection are linked to a decreased risk of pancreatic cancer in eastern populations, yet there was no significant correlation in western countries. The meta-analysis conducted by the authors included nine observational studies,^{35–37,39–43,45} which have also been incorporated into this study. In 2015, Schulte et al.¹⁶ performed a meta-analysis of 10 studies, and the results indicated that there was no significant overall association between H. pylori infection and pancreatic cancer risk (OR = 1.13, 95% CI = 0.86–1.50). CagA seropositivity, however, was linked to a significantly reduced risk of pancreatic cancer (OR = 0.78, 95% CI = 0.67–0.91), while CagA negative H. pylori seropositivity was associated with a significantly increased pancreatic cancer risk (OR = 1.30, 95% CI = 1.02–1.65). This study included all of the studies included by Schulte et al.¹⁶ except for one low-quality conference abstract.⁵⁰ Given that different study designs may have an impact on the association between H. pylori and pancreatic cancer, two meta-analyses only explored the relationship between them in cohort and nested case–control studies. In 2015, Chen et al.¹⁵ and Liu et al.,¹⁸ who included five nested case–control studies, found that the presence of H. pylori did not have a statistically significant effect on the likelihood of developing pancreatic cancer (OR = 0.99, 95% CI = 0.65–1.50; p = 0.96). A positive association, however, was observed between pancreatic cancer and CagA – nonvirulent strains of H. pylori (OR = 1.47, 95% CI = 1.11–1.96; p = 0.008), whereafter Liu et al.¹⁸ included seven nested case–control and cohort studies, and the findings were similar. All these studies included in these two meta-analyses were included in this study. Recently, in the course of our meta-analysis, Xu et al.⁵¹ conducted a meta-analysis involving 17 observational studies (4 cohort and 13 case–control studies), showing that H. pylori infection is associated with an increased risk of pancreatic cancer in general. Nevertheless, no correlation was found between CagA/VacA-positive H. pylori infection and pancreatic cancer risk. Compared with Xu et al.’s⁵¹ meta-analysis, our updated meta-analysis encompassed the majority of the studies with the exception of the cohort study from Chen et al.⁵² The purpose of this cohort study was to investigate the relationship between H. pylori and pancreatic cancer mortality, rather than H. pylori and pancreatic cancer risk. Unfortunately, the authors misextracted their data and pooled them.

Compared with all previous meta-analyses,^12–18,51 our updated meta-analysis furthers the research findings of past work in a variety of ways. First, our updated meta-analysis is the most comprehensive to date, including 20 studies with 67,718 participants, providing the latest and most sufficient epidemiologic evidence on the topic. Second, we performed a more thorough statistical analysis, for instance, performing separate analyses for case–control and cohort studies, as well as performing multiple subgroup and sensitivity analyses. The meta-analysis results of included case–control and cohort studies are in agreement, and the results of sensitivity analyses were basically stable. As a consequence of this methodological effort, the current meta-analysis has yielded reliable result that there is no substantial link between H. pylori infection and the likelihood of developing pancreatic cancer. Third, previous meta-analyses mainly analyzed the relationship between CagA positive/CagA negative strains of H. pylori infection and pancreatic cancer risk, but rarely analyzed VacA positive strains at the same time. This study, however, explored the relationship between CagA positive, CagA negative, VacA positive strains of H. pylori infection, and pancreatic cancer risk at the same time, which makes us understand the relationship more comprehensively.

Our meta-analysis did not demonstrate a significant association between H. pylori infection and pancreatic cancer. Nevertheless, some studies have proposed the potential connection between them, yet the exact mechanism is still unknown. Proposed explanations have been put forth to account for this potential correlation. First, it is possible that inflammatory mechanisms may be a contributing factor in the development of pancreatic cancer as a result of H. pylori infection. Chronic inflammation is a characteristic feature of multiple cancers.⁵³ Farrow et al.⁵⁴ proposed that pancreatic cancer may be caused by a long-term inflammatory process which involves creating a stroma. Nilsson et al.⁵⁵ discovered the presence of Helicobacter 16S ribosomal DNA in pancreas tumor and surrounding tissue. It is believed that reactive oxygen species, proinflammatory cytokines, and other inflammatory mediators, which are all caused by H. pylori, can lead to chronic tissue inflammation. These factors may result in an increase of genomic DNA damage, cell proliferation, and the inactivation of tumor suppressor genes, which further facilitating malignant transformation of pancreatic cells.⁵⁵ Second, it has been suggested that presence of H. pylori in the gastric antrum is linked to an increase in gastric acid secretion. This causes the unstrained release of duodenal secretin, which is then followed by an increased basal pancreatic bicarbonate output and DNA synthesis, further leading to pancreatic ductal hyperplasia.^56,57 Third, it was also assumed that the colonization of H. pylori in the gastric antrum leads to decreased gastric acid production and hypergastrinemia. The resulting low acidity promotes the excessive proliferation of bacteria and secretion of N-nitroso compounds, which could be activated in the ductal epithelium and transported to the pancreas through the bloodstream.^57,58 In addition, a recent study⁵⁹ has shown a link between mucin 4 (MUC4) and H. pylori infection, proposing that MUC4 may be a cytokine involved in the pathogenesis of pancreatic cancer. All in all, to gain a better understanding of the relationship between H. pylori and pancreatic cancer, it is essential to conduct comprehensive studies that consider the influence of environmental and genetic susceptibility factors.

Strengths and limitations

This meta-analysis has several strengths. First, as previously discussed, this study is the most comprehensive to date, consisting of 20 studies, which analyze H. pylori infection and risk of pancreatic cancer. The considerable sample size enabled us to evaluate the association between H. pylori infection and pancreatic cancer in a quantitatively precise manner. Second, we registered our meta-analysis in advance on PROSPERO, and used stringent search strategy and stringent criteria for inclusion, as well as adhered to the PRISMA and MOOSE reporting statement, making our results more reliable and transparent. Third, ORs and HRs were pooled separately according to different study designs, and the results of our meta-analysis mainly based on adjusted data, which makes our results more convincing and reliable. Finally, the studies included in our meta-analysis originated from four continents, and most of them featured high quality, thus rendering the results of the analysis more robust and generalizable.

Nevertheless, there are several limitations should be considered. First, this study is based on pooled data from observational studies, which are susceptible to inherent confounders in the individual included study. Although the majority of included studies have adjusted for confounders (age, sex, and smoking), a few studies controlled for other known confounders of pancreatic cancer, such as heavy alcohol intake, blood group, diabetes, obesity, chronic pancreatitis, and a family history of pancreatic cancer.^5,7 Furthermore, several studies failed to take into account adjusted confounders. The results show that, based on data from unadjusted confounders, H. pylori infection increases risk of pancreatic cancer. That said, these confounders may influence the magnitude and nature of the observed association. Thus, to validate the relationship, further studies must be conducted that take into account potential confounding variables. Second, our meta-analysis has moderate heterogeneity, which could be attributed to a variety of sources such as diversity in the population characteristics, inconsistency of adjustment for confounding factors, and H. pylori detection methods. Heterogeneity may limit the reliability of the conclusions of this study. Third, for meta-analysis from 17 case–control studies, the results of all our sensitive analyses were stable, except that the case–control study⁴⁵ published by Risch and colleagues in 2014 was excluded from the pooled analysis of the remaining studies for sensitivity analysis, and the results were reversed. This study was conducted in Shanghai (China), including the largest sample of pancreatic cancer cases (761 cases) compared with other case–control studies, which showed that individuals who tested seropositive for CagA had a reduced risk of developing pancreatic cancer when compared with those who tested seronegative for both H. pylori and CagA (OR = 0.68, 95% CI = 0.54–0.84). Up to 95% of the people in this place are colonized by H. pylori strains carrying the virulent protein CagA, while only 65% of the people in the United States are colonized.⁴⁵ Evidence from prior study has indicated that CagA positive H. pylori strains may be associated with a decreased risk of pancreatic cancer.⁴⁰ The effect that may help explain differences in effect estimates can be seen in populations with different serological positive rates for CagA. In fact, it is hard to disentangle the possible reversal causation. There is a lack of relevant information on socioeconomic status and dietary intakes, which may contribute to disease susceptibility and worse health outcomes. Individuals with a poor socioeconomic status are more likely to be infected with H. pylori.¹¹ Different dietary intakes may have different pancreatic cancer risk. Several studies show that folate, vitamin D, and methionine protect against development of pancreatic cancer.⁶ High-fat and high-cholesterol diet, however, are associated with an increased pancreatic cancer.⁵ Thus, future high-quality studies with large samples take fully account of socioeconomic status, dietary intakes, and CagA positive/negative strains are necessary to validate their relationship. Fourth, most of studies used serological tests (serum IgG antibodies) to detect H. pylori, positive serological tests did not distinguish between current or previous infection. Furthermore, the IgG antibodies persist for several months or even years after H. pylori eradication.¹¹ It is estimated that more than half of the world’s population will be infected with H. pylori at some stage in their lives, leading to a high rate of H. pylori seropositivity. This widespread seropositivity could potentially distort the association findings. Fifth, almost all included studies did not identify the specific type of pancreatic cancer with the exception of the study conducted by Raderer et al.³⁵ The fact, however, is that more than 90% of pancreatic cancer is pancreatic ductal adenocarcinoma;⁶⁰ we therefore believe that the pathological type has little impact on the outcome. Finally, only three cohort studies evaluated the relationship between H. pylori infection and risk of incident pancreatic cancer. It is well known that cohort studies are superior to case–control studies to confirm the etiological factor. Hence, more high-quality prospective cohort studies are required to verify the causal relationship.

Conclusion

In summary, we found insufficient evidence to support the proposed association between H. pylori infection and increased risk of pancreatic cancer. In order to better understand any association, future evidence from large, well-designed, high-quality prospective cohort studies that accounts for diverse ethnic populations, certain H. pylori strains, and confounding factors would be useful to settle this controversy.

Supplemental Material

sj-docx-1-taj-10.1177_20406223231155119 – Supplemental material for Is Helicobacter pylori infection associated with pancreatic cancer? A systematic review and meta-analysis of observational studies

Supplemental material, sj-docx-1-taj-10.1177_20406223231155119 for Is Helicobacter pylori infection associated with pancreatic cancer? A systematic review and meta-analysis of observational studies by Ben-Gang Zhou, Yu-Zhou Mei, Jing-Shu Wang, Jian-Lei Xia, Xin Jiang, Sheng-Yong Ju and Yan-Bing Ding in Therapeutic Advances in Chronic Disease

Footnotes

Acknowledgements

None.

Declarations

ORCID iD

Ben-Gang Zhou

Supplemental material

Supplemental material for this article is available online.

References

Sung

Ferlay

Siegel

, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021; 71: 209–249.

Chen

Liu

, et al. Global, regional and national burden of pancreatic cancer, 1990 to 2017: results from the Global Burden of Disease Study 2017. Pancreatology 2020; 20: 462–469.

Lippi

Mattiuzzi

The global burden of pancreatic cancer. Arch Med Sci 2020; 16: 820–824.

Conroy

Hammel

Hebbar

, et al. FOLFIRINOX or gemcitabine as adjuvant therapy for pancreatic cancer. N Engl J Med 2018; 379: 2395–2406.

Hidalgo

Pancreatic cancer. N Engl J Med 2010; 362: 1605–1617.

Raimondi

Maisonneuve

Lowenfels

AB.

Epidemiology of pancreatic cancer: an overview. Nat Rev Gastroenterol Hepatol 2009; 6: 699–708.

Maisonneuve

Lowenfels

AB.

Risk factors for pancreatic cancer: a summary review of meta-analytical studies. Int J Epidemiol 2015; 44: 186–198.

Parkin

Boyd

Walker

LC.

The fraction of cancer attributable to lifestyle and environmental factors in the UK in2010. Br J Cancer 2011; 105: S77–S81.

Crowe

SE.

Helicobacter pylori infection. N Engl J Med 2019; 380: 1158–1165.

10.

Hunt

Xiao

Megraud

, et al. World Gastroenterology Organisation global guideline: Helicobacter pylori in developing countries. J Clin Gastroenterol 2011; 45: 383–388.

11.

Venkatanarayanan

De Deyn

MLZQ

, et al. A meta-analysis of the association between Helicobacter pylori (H. pylori) infection and hyperemesis gravidarum. Helicobacter 2018; 23: e12455.

12.

Trikudanathan

Philip

Dasanu

, et al. Association between Helicobacter pylori infection and pancreatic cancer: a cumulative meta-analysis. JOP 2011; 12: 26–31.

13.

Xiao

Wang

Gao

Association between Helicobacter pylori infection and pancreatic cancer development: a meta-analysis. PLoS ONE 2013; 8: e75559.

14.

Wang

Zhang

Wang

YJ.

Helicobacter pylori and pancreatic cancer risk: a meta-analysis based on 2,049 cases and 2,861 controls. Asian Pac J Cancer Prev 2014; 15: 4449–4454.

15.

Chen

Wang

Chen

, et al. Cytotoxin-associated gene A-negative strains of Helicobacter pylori as a potential risk factor of pancreatic cancer: a meta-analysis based on nested case-control studies. Pancreas 2015; 44: 1340–1344.

16.

Schulte

Pandeya

Fawcett

, et al. Association between Helicobacter pylori and pancreatic cancer risk: a meta-analysis. Cancer Causes Control 2015; 26: 1027–1035.

17.

Guo

Liu

Helicobacter pylori infection and pancreatic cancer risk: a meta-analysis. J Cancer Res Ther 2016; 12: C229–C232.

18.

Liu

Chen

Wang

, et al. Helicobacter pylori infection, atrophic gastritis, and pancreatic cancer risk: a meta-analysis of prospective epidemiologic studies. Medicine 2017; 96: e7811.

19.

Geng

Nie

, et al. Expression of Helicobacter pylori antibody in patients with pancreatic cancer. Acta Medica Mediterranea 2019; 35: 51–54.

20.

Hirabayashi

Inoue

Sawada

, et al. Helicobacter pylori infection, atrophic gastritis, and risk of pancreatic cancer: a population-based cohort study in a large Japanese population: the JPHC study. Sci Rep 2019; 9: 6099.

21.

Permuth

Rahman

Chen

D-T

, et al. A case control study of the seroprevalence of Helicobacter pylori proteins and their association with pancreatic cancer risk. J Pancreat Cancer 2021; 7: 57–64.

22.

Laya

Anandhi

Gurushankari

, et al. Association between Helicobacter pylori and periampullary and pancreatic cancer: a case-control study. J Gastrointest Cancer 2022; 53: 902–907.

23.

Osaki

Lin

Sasahira

, et al. Prevalence estimates of Helicobacter species infection in pancreatic and biliary tract cancers. Helicobacter 2022; 27: e12866.

24.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021; 372: n71.

25.

Stroup

Berlin

Morton

, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting – Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000; 283: 2008–2012.

26.

Stang

Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010; 25: 603–605.

27.

Khan

Yuan

Iqbal

, et al. No association linking short-term proton pump inhibitor use to dementia: systematic review and meta-analysis of observational studies. Am J Gastroenterol 2020; 115: 671–678.

28.

DerSimonian

Laird

Meta-analysis in clinical trials revisited. Contemp Clin Trials 2015; 45: 139–145.

29.

Soh

AYS

Loke

, et al. Systematic review with meta-analysis: the association between post-traumatic stress disorder and irritable bowel syndrome. J Gastroenterol Hepatol 2019; 34: 68–73.

30.

Higgins

Thompson

Deeks

, et al. Measuring inconsistency in meta-analyses. BMJ 2003; 327: 557–560.

31.

Sterne

Sutton

Ioannidis

, et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ 2011; 343: d4002.

32.

Begg

Mazumdar

Operating characteristics of a rank correlation test for publication bias. Biometrics 1994; 50: 1088–1101.

33.

Egger

Davey Smith

Schneider

, et al. Bias in meta analysis detected by a simple, graphical test. BMJ 1997; 315: 629–634.

34.

Duval

Tweedie

Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 2000; 56: 455–463.

35.

Raderer

Wrba

Kornek

, et al. Association between Helicobacter pylori infection and pancreatic cancer. Oncology 1998; 55: 16–19.

36.

Stolzenberg-Solomon

Blaser

Limburg

, et al. Helicobacter pylori seropositivity as a risk factor for pancreatic cancer. J Natl Cancer Inst 2001; 93: 937–941.

37.

de Martel

Llosa

Friedman

, et al. Helicobacter pylori infection and development of pancreatic cancer. Cancer Epidemiol Biomarkers Prev 2008; 17: 1188–1194.

38.

Dou

Wang

JB.

Detection of H. pylori infection in patients with pancreatic cancer and clinical significance. J Jilin Univ Med Ed 2008; 34: 317–319.

39.

Lindkvist

Johansen

Borgström

, et al. A prospective study of Helicobacter pylori in relation to the risk for pancreatic cancer. BMC Cancer 2008; 8: 321.

40.

Risch

, et al. ABO blood group, Helicobacter pylori seropositivity, and risk of pancreatic cancer: a case-control study. J Natl Cancer Inst 2010; 102: 502–505.

41.

Shimoyama

Takahashi

Abe

, et al. Serological analysis of Helicobacter hepaticus infection in patients with biliary and pancreatic diseases. J Gastroenterol Hepatol 2010; 25: S86–S89.

42.

Gawin

Wex

Ławniczak

, et al. Helicobacter pylori infection in pancreatic cancer. Pol Merkur Lekarski 2012; 32: 103–107.

43.

Murphy

Michel

, et al. Seropositivity to Helicobacter pylori and risk of pancreatic cancer. Cancer Epidemiol Biomarkers Prev 2013; 22: 2416–2419.

44.

Hsu

Lin

, et al. The relationship between Helicobacter pylori and cancer risk. Eur J Intern Med 2014; 25: 235–240.

45.

Risch

Kidd

, et al. Helicobacter pylori seropositivities and risk of pancreatic carcinoma. Cancer Epidemiol Biomarkers Prev 2014; 23: 172–178.

46.

Hua

Liu

, et al. Preliminary study of pancreatic cancer associated with Helicobacter pylori infection. Cell Biochem Biophys 2015; 71: 397–400.

47.

Chen

Schöttker

Castro

, et al. Association of Helicobacter pylori infection and chronic atrophic gastritis with risk of colonic, pancreatic and gastric cancer: a ten-year follow-up of the ESTHER cohort study. Oncotarget 2016; 7: 17182–17193.

48.

Huang

Zagai

Hallmans

, et al. Helicobacter pylori infection, chronic corpus atrophic gastritis and pancreatic cancer risk in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort: a nested case-control study. Int J Cancer 2017; 140: 1727–1735.

49.

Lima de Souza Gonçalves

Cordeiro Santos

Silva Luz

, et al. From Helicobacter pylori infection to gastric cancer: current evidence on the immune response. World J Clin Oncol 2022; 13: 186–199.

50.

Wadström

Fryzek

Demirjian

, et al. Antibodies to Helicobacter bilis in patients with pancreatic carcinoma. Helicobacter 2004; 9: 538–539.

51.

Zhou

Yin

, et al. The relationship between Helicobacter pylori and pancreatic cancer: a meta-analysis. Transl Cancer Res 2022; 11: 2810–2822.

52.

Chen

Segers

Blaser

MJ.

Association between Helicobacter pylori and mortality in the NHANES III study. Gut 2013; 62: 1262–1269.

53.

Marx

Cancer research: inflammation and cancer: the link grows stronger. Science 2004; 306: 966–968.

54.

Farrow

Sugiyama

Chen

, et al. Inflammatory mechanisms contributing to pancreatic cancer development. Ann Surg 2004; 239: 763–769; discussion 769.

55.

Nilsson

Stenram

Ihse

, et al. Helicobacter species ribosomal DNA in the pancreas, stomach and duodenum of pancreatic cancer patients. World J Gastroenterol 2006; 12: 3038–3043.

56.

Risch

HA.

Etiology of pancreatic cancer, with a hypothesis concerning the role of N-nitroso compounds and excess gastric acidity. J Natl Cancer Inst 2003; 95: 948–960.

57.

Panthangi

Cyril Kurupp

Raju

, et al. Association between Helicobacter pylori infection and the risk of pancreatic cancer: a systematic review based on observational studies. Cureus 2022; 14: e28543.

58.

Houben

Stockbrügger

RW.

Bacteria in the aetio-pathogenesis of gastric cancer: a review. Scand J Gastroenterol Suppl 1995; 212: 13–18.

59.

Gong

Chen

, et al. MUC4 is a novel mediator in H. pylori infection-related pancreatic cancer. Oncol Lett 2021; 21: 123.

60.

Jiang

Fagman

, et al. A comprehensive review of pancreatic cancer and its therapeutic challenges. Aging 2022; 14: 7635–7649.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB