Diseases of the Esophagus (2014) 27, 645–653 DOI: 10.1111/dote.12194
Review article
Association of Helicobacter pylori infection with esophageal adenocarcinoma and squamous cell carcinoma: a meta-analysis S. Nie,1,2 T. Chen,1 X. Yang,3 P. Huai,2 M. Lu1,2 Clinical Epidemiology Unit, 3Department of Gastroenteroloy, Qilu Hospital of Shandong University, and Department of Epidemiology and Health Statistics, School of Public Health, Shandong University, Jinan, China
1
2
SUMMARY. To evaluate the relationship of Helicobacter pylori and cytotoxin-associated gene A (CagA) positive strains with esophageal neoplasm, including esophageal adenocarcinoma (EAC) and esophageal squamous cell carcinoma (ESCC), the authors conducted a meta-analysis using a predefined protocol. PubMed, Web of Science, China biology medical literature database, Wanfang, and China National Knowledge Infrastructure were searched for relevant articles from the first available year to April 8, 2013. The fixed or random effect pooled measure was selected based on heterogeneity among studies, which was evaluated using Q test and the I 2 of Higgins and Thompson. Metaregression was used to explore the sources of between-study heterogeneity. Publication bias was analyzed by Begg’s funnel plot and Egger’s regression test. The association was assessed by odds ratio (OR) with 95% confidence interval (CI). A total of 28 eligible studies were included in the meta-analysis. There was a significant inverse association between H. pylori infection (pooled OR, 0.57; 95% CI, 0.44–0.73) and EAC; CagA-positive H. pylori strains were less likely to be associated with EAC compared with CagA-negative strains (pooled OR, 0.64; 95% CI, 0.52–0.79). However, there was no statistically significant association between H. pylori/CagA-positive H. pylori strains infection and ESCC, and the pooled ORs were 1.16 (95% CI, 0.83–1.60) and 0.97 (95% CI, 0.79–1.19). But significant associations between CagA-positive H. pylori strains infection and ESCC risk were found in the stratified analysis of the study location (Asian and non-Asian), and the summary ORs were 0.74 (95% CI, 0.57–0.97) and 1.41 (95% CI, 1.02–1.94). H. pylori infection and CagA-positive strains are associated with decreased risk of EAC in the overall population. No significant association was found between H. pylori infection/CagA-positive strains and ESCC. But CagA-positive strains might have a positive association with ESCC in non-Asian population and an inverse association in Asian population. KEY WORDS: esophageal adenocarcinoma, esophageal squamous cell carcinoma, Helicobacter pylori, meta-analysis.
INTRODUCTION Helicobacter pylori infect about 50% of the world’s population.1 And many cases of peptic ulcers, gastritis, and duodenitis are confirmed to have an association with H. pylori infection.2–4 In addition, infection with cytotoxin-associated gene A (CagA)-positive Address correspondence to: Dr Ming Lu, MD, PhD, Clinical Epidemiology Unit, Qilu Hospital of Shandong University, 107 Wenhuaxi Road, Jinan 250012, China. Email: [email protected] Sources of funding: This work was supported by the National Natural Science Foundation of China (grants 81072358 and 81273151). Conflict of interest: The authors have no competing interests. Disclosure: None. © 2014 International Society for Diseases of the Esophagus
strains of H. pylori increases the risk associated with H. pylori infection alone.5 H. pylori was classified as a group I carcinogen by the International Agency for Research on Cancer and the World Health Organization in 1994.6 H. pylori have been identified as the most important risk factor for gastric cancer.7,8 Epidemiologic studies also have investigated the association between H. pylori and other gastrointestinal malignancies, including pancreatic cancer,9,10 colorectal cancer,11,12 and esophageal cancer.13–15 Esophageal cancer is a disease in epidemiologic transition. Until the 1970s, the most common type of esophageal cancer in the United States was esophageal squamous cell carcinoma (ESCC), which has smoking and alcohol consumption as risk factors; 645
646
Diseases of the Esophagus
since then, there has been a progressive increase in the incidence of esophageal adenocarcinoma (EAC), for which the most common predisposing factor is gastroesophageal reflux disease (GERD). The two main histological types of esophageal cancer (ESCC and EAC) have distinct geographic and demographic patterns. Therefore, the associations of any potential causal factors should be evaluated separately for these two cancer types. H. pylori can be detected with a number of invasive methods, such as histological examination and urease test on gastric antral biopsies, or by non-invasive methods, such as serologic tests. H. pylori infection was defined as positivity to at least one test among histology, rapid urease test, and serology for both general anti-immunoglobulin G and anti-CagA. There are different strains of H. pylori. The genome of Cag-positive strains contains the cag pathogenicity island. This island includes approximately 31 putative genes, including cagA – the gene that encodes the CagA protein.16 CagA-positive strains confer a higher risk of non-cardia gastric cancer than CagAnegative strains.5 Likewise, the association between CagA-positive and CagA-negative strains with other cancers may be different17 and should be considered in analyses. Indeed, a large number of studies have examined the association between H. pylori infection and esophageal neoplasm (EAC and ESCC).13,14,18–20 But the role of H. pylori infection in esophageal neoplasm was controversial.21,13,18,22 We conducted this meta-analysis to summarize the published literature on H. pylori and esophageal neoplasm. We analyzed and reported the results separately for ESCC and EAC. In addition, we examined the role of CagA-positive H. pylori strains in these entities.
METHODS Literature search Sources of studies included the literature databases PubMed, Web of Science, China biology medical literature database, Wanfang, and China National Knowledge Infrastructure. Studies were searched from the inception of each database through April 8, 2013 using the key words ‘Helicobacter pylori’, ‘H. pylori’, or ‘HP’, and ‘esophageal neoplasm’, ‘esophageal carcinoma’, ‘esophageal tumor’, ‘esophageal cancer’, ‘esophageal squamous cell carcinoma’, ‘esophageal squamous carcinoma’, ‘ESCC’, ‘esophageal adenocarcinoma’, ‘adenocarcinoma of the esophagus’, or ‘EAC’. The subjects of studies were defined as humans, and the languages of articles were limited to English and Chinese because the reviewers are fluent in both of these languages. We evaluated potentially associated publications by checking their
titles and abstracts, and then picked out the most relevant publications for a closer examination. Moreover, the reference lists of the selected papers were also screened for other potential articles that possibly have been missed in the initial search (references cited in the identified articles were searched manually). Study selection Eligible studies were selected according to the following explicit inclusion criteria: 1 Esophageal cancer and H. pylori could be used as an outcome in the analysis. 2 The method(s) to detect H. pylori infection must be mentioned in the article. For example, histology, rapid urease test, and serology. 3 Case–control, nested case–control, or cohort studies were considered. 4 Sufficient data for estimating an odds ratio (OR) with 95% confidence interval (CI). 5 Searching the information without language restriction. We excluded studies based on the following criteria: 1 Articles published not in English or Chinese. 2 Studies that without full text articles or could not be obtained. 3 Reviews and repeated literatures were also excluded. 4 The study was not conducted on humans. 5 Not offering the source of cases and controls and other essential information. When data from multiple reports were identified, we included only the report with the most complete relevant data. Data extraction and quality assessment Data were extracted and entered into a database. Two investigators extracted information from all eligible publications independently according to the criteria listed above. The following data were collected from each study: name of the first author, year of publication, study location, study design (source of controls), number in case (exposed) and control (unexposed) groups, and the methods to assay the H. pylori infection, respectively. Study design was stratified to population-based (PB) studies and hospital-based (HB) studies. The quality of all included studies was assessed using the Newcastle– Ottawa Scale as recommended by the Cochrane NonRandomized Studies Methods Working Group.23 Data extraction and quality assessment were performed by two independent investigators. Any disagreement was settled by discussion. Statistical analysis The association between H. pylori/CagA-positive H. pylori strains and EAC/ESCC was measured by © 2014 International Society for Diseases of the Esophagus
HP infection and esophageal cancer
647
Potentially eligible titles initially generated by the literature searches (n = 948)
592 titles rejected (titles suggested article not appropriate)
356 titles retrieved
185 abstracts rejected (duplications, review articles)
171 abstracts accepted (with extractable data eligible)
146 studies excluded as not fulfilling the inclusion criteria Studies included in qualitative synthesis (n = 28, 23 in English and 5 in Chinese) Fig. 1 Flow chart of study selection.
pooled ORs with 95% CIs. Fourfold table data were obtained from each study respectively. The heterogeneity among different studies was checked by the Q-test and I 2 statistics.24 I 2 > 50% indicated evidence of heterogeneity. The ORs were pooled through a fixed effects model,25 using the Mantel– Haenszel approach when no heterogeneity was observed among studies. Otherwise, a random effects model was adopted.26 Metaregression with restricted maximum likelihood estimation was performed to explore the potentially important sources of betweenstudy heterogeneity. Sensitivity analysis was performed to validate the stability of outcomes by sequential removal of each individual study.27 An individual study is suspected to excessively influence the point estimate if its omitted analysis lies outside the 95% CI of the combined analysis. Publication bias was diagnosed with Begg’s funnel plot and Egger’s linear regression method.28 Asymmetric or incomplete funnel shaped plots and P < 0.05 in Egger’s test indicated the presence of potential publication bias. All statistical analyses used STATA 11.0 (StataCorp LP, College Station, TX, USA).
RESULTS Characteristics of studies The literature search identified 948 potentially relevant articles. Based on our inclusion criteria, © 2014 International Society for Diseases of the Esophagus
a total of 28 studies13–15,17–22,29–47 were included in qualitative synthesis (Fig. 1). The total population of the included studies was 10 285 persons, including 3269 cases and 7016 controls. Of the 28 studies, 23 are English and five are Chinese papers. Detailed characteristics of the aggregated data for the 28 studies are summarized in Table 1. In total, nine studies14,17,18,33,36,38,40,41,46 concerned EAC, 15 concerned ESCC,13,15,20,22,29–32,34,35,39,42–45 and four19,21,37,47 concerned both. There were 13 studies in Asian populations and 15 studies in non-Asian subjects.
QUANTITATIVE SYNTHESIS AND TEST OF HETEROGENEITY To evaluate the association between H. pylori infection and EAC risk, 1145 cases and 3453 controls were obtained from 13 studies, of which 372 cases and 1380 controls were H. pylori positive, and 773 cases and 2073 controls were H. pylori negative. As shown in Figure 2, the overall OR was 0.57 (95% CI 0.44–0.73). Significant study heterogeneity(I 2 = 53.3%) was found so a random effects model was used to pool the ORs. For the assessment of the association of CagApositive H. pylori strains with EAC, 562 cases and 1439 controls were obtained from seven studies, of which 225 cases and 712 controls were positive, and 337 cases and 727 controls were negative. As shown in Figure 3, the overall OR was 0.64 (95% CI
Lu H Venerito M Khoshbaten M Mao K Cook MB Wu IC Fruh M Zheng Y Kamangar F Siman JH Iijima K Anderson LA Derakhshan MH Xu H Wu DC de Martel C Ye W El-Omar EM Wang KX Wu AH Grimley CE Vicari JJ Chow WH Qing Y Talley NJ Hu HM Weston AP Whiteman DC
2012 2011 2011 2011 2010 2009 2008 2008 2007 2007 2007 2007 2007 2006 2005 2005 2004 2003 2003 2002 1999 1998 1998 1998 1991 2009 2000 2010
China Germany Iran China Finland Taiwan United States China China Sweden Japan Ireland Iran China Taiwan United States Sweden United States China United States England United States United States China United States Taiwan United States Australia
Study location PCR S,H,U S PCR S S S U S S S,H,U S S U S S S S S S S S S U S S H S
HP Dx — — — — — — 36/100 — — 7/12 — 100/208 9/19 — — 19/51 18/97 35/108 — 49/80 16/40 7/21 38/129 — — — 3/20 35/260
HP+
0/7 12/38 — — — — —
9/18 42/97 — — 18/80
— —
— — — — — — 29/100 — — 6/12 — 109/208
CagA+
Case
— — — — — — 30/101 — — 32/47 — 150/253 — — — 44/71 293/499 — — 106/356 — 11/26 46/86 — — — — —
CagA+
Control
— — — — — — 43/101 — — 35/47 — 157/253 28/38 — — 74/150 198/499 84/210 — 230/356 21/46 26/57 86/223 — — — 96/127 302/1316
HP+
EAC
17/25 53/75 58/100 122/160 57/71 112/317 — 67/96 254/335 24/37 60/73 — — 23/35 28/127 — 32/85 31/53 33/63 — — — — 39/43 20/41 66/180 — 54/208
HP+ — 42/75 28/100 — 32/46 91/99 — — 178/254 24/37 — — — — — — 63/85 — — — — — — 82 — — — —
CagA+
Case
40/75 36/100 — 41/60 142/153 — — 552/727 82/129 — — — — — — 293/499 — — — — — — 131 — — — —
CagA+
Control
9/25 53/75 83/100 39/80 68/87 160/395 — 88/157 727/992 87/129 56/73 — — 308/353 74/171 — 198/499 84/210 145/310 — — — — 43/170 96/252 102/194 — 302/1316
HP+
ESCC
HB HB HB HB PB PB HB HB PB PB HB PB HB HB PB PB PB PB NR PB HB HB PB HB HB HB HB PB
Study type
No Yes Yes No No Yes Yes No No Yes Yes Yes Yes No No Yes No Yes Yes Yes No No Yes No No Yes No Yes
Matched
5 8 8 6 7 8 8 6 7 9 8 9 8 6 7 9 9 9 7 9 6 6 9 6 6 8 6 9
Scores
—, no data available; EAC, esophageal adenocarcinoma; ESCC, esophageal squamous cell carcinoma; H, histology; HB, hospital based; HP Dx, HP detection method; NR, not reported; PB, population based; PCR, polymerase chain reaction; S, serology; U, rapid urease test.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Publication year
Characteristics of included studies on H. pylori and esophageal neoplasm
First author
Table 1
648 Diseases of the Esophagus
© 2014 International Society for Diseases of the Esophagus
HP infection and esophageal cancer
649
Fig. 2 Random effects meta-analysis of the association between H. pylori (HP) and esophageal adenocarcinoma (EAC). CI, confidence interval; OR, odds ratio.
0.52–0.79). There was no significant study heterogeneity (I 2 = 0.0%) so a fixed effects model was used to pool the ORs. To evaluate the possible association of H. pylori infection with ESCC risk, 2124 cases and 5588 controls obtained from 19 studies were included, of which 1150 cases and 2722 controls were H. pylori positive, and 974 cases and 2866 controls were H. pylori negative. As shown in Figure 4, the overall OR was 1.16 (95% CI 0.83–1.60). There was statistical heterogeneity (I 2 = 85.7%) among studies so a random effects model was used to pool the ORs. For the assessment of the association of CagApositive H. pylori strains with ESCC, 696 cases and 1743 controls were obtained from seven studies, of which 458 cases and 1186 controls were positive, and 238 cases and 557controls were negative. As shown in Figure 5, the overall OR was 0.97 (95% CI 0.79– 1.19). There was no statistical study heterogeneity (I 2 = 49.7%) so a fixed effects model was used to pool the ORs. Exploration of the sources of heterogeneity As seen above, strong heterogeneity among studies on H. pylori and ESCC was demonstrated. To further explore the potential sources of heterogeneity and test © 2014 International Society for Diseases of the Esophagus
the effects of study characteristics on the overall estimates, exploratory univariate metaregression was performed with study location (Asian and nonAsian), source of controls (PB or HB), and publication year. However, none of the variables was identified as potentially important source of betweenstudy heterogeneity. Sensitivity analysis and publication bias The influence of each study on the pooled ORs was examined by repeating the meta-analysis while sequentially omitting individual studies. Both sensitivity analysis for EAC and ESCC indicated that no single study influenced the pooled ORs qualitatively, suggesting that the results of our meta-analysis are stable. We plotted Begg’s funnel plot to examine small study effects. We also used Begg’s and Egger’s weighted regression method to calculate P for bias. The shape of the funnel plots did not reveal any evidence of obvious asymmetry. And the P-values of Egger’s test were 0.216, 0.170, 0.280, and 0.496 for H. pylori versus EAC, CagA-H. pylori versus EAC, H. pylori versus ESCC, and CagA-H. pylori versus ESCC, respectively, implying no existence of publication bias.
650
Diseases of the Esophagus
Fig. 3 Fixed effects meta-analysis of the association between cytotoxin-associated gene A (CagA)-positive H. pylori strains and esophageal adenocarcinoma (EAC). CagA-HP, CagA-positive H. pylori strains; CI, confidence interval; OR, odds ratio.
DISCUSSION The results showed an inverse statistically significant relationship of H. pylori infection and CagA-positive
H. pylori strains with EAC. In terms of ESCC, no significant association was found between H. pylori/CagA-positive H. pylori strains and ESCC. In the stratified analysis of study location, an
Fig. 4 Random effects meta-analysis of the association between H. pylori (HP) and esophageal squamous cell carcinoma (ESCC). CI, confidence interval; OR, odds ratio. © 2014 International Society for Diseases of the Esophagus
HP infection and esophageal cancer
651
Fig. 5 Random effects meta-analysis of the association between cytotoxin-associated gene A (CagA)-positive H. pylori strains and esophageal squamous cell carcinoma (ESCC). CagA-HP, CagA-positive H. pylori strains; CI, confidence interval; OR, odds ratio.
inverseassociation was found between CagA-positive H. pylori strains and ESCC in Asian populations; but a positive association was found in non-Asian populations. The reason for this phenomenon may be the difference in the carcinogenesis pathways between ESCC and EAC. As known to all, possible risk factors for ESCC include cigarette smoking, alcohol consumption, hot-temperature food, low intake of vegetables, salty food, pickled vegetables, nutrient deficiency, chronic mucosal irritation, and a family history of cancer,48 while EAC is closely related to Barrett’s esophagus.49,50 Barrett’s esophagus plays a pivotal role in the development of EAC. And the occurrence of Barrett-type ulcer is due to frequent gastroesophageal reflux, which leads to chronic inflammation and ulceration of the squamous epithelial lining, followed by re-epithelialization by gastric-type or intestinaltype epithelium in the low-PH microenvironment. Researchers have reported that H. pylori infection can protect against the development of GERD.51,52 The probable mechanisms of the protective effect may be that H. pylori infection can lead to gastric atrophy, lower intragastric acidity, and increased intragastric ammonia production. A study also revealed that eradication of the infection might be a risk factor for GERD, especially in Asian populations.53 Our study also found that H. pylori is a strong protective factor against EAC. The possible mechanisms are as follows. First, it may be related to gastroesophageal reflux. Mager54 considers that the © 2014 International Society for Diseases of the Esophagus
radical treatment of the H. pylori infection may bring about the increasing of the ghrelin in the serum,55 resulting in the occurrence of obesity.56,57 The function of the lower esophageal sphincter will be slack with the increasing of weight, leading to the increasing of GERD. And GERD is a risk factor for BE and EAC.58 Second, it may be that the decreased iron state reduces the morbidity of EAC. It is not the atrophic gastritis and gastratrophy reducing the incidence of EAC in Ye et al’s view.37 For example, people with pernicious anemia also have gastric mucosal atrophy, but the incidence of EAC is not declining. Third, Jones et al. have reported that H. pylori could induce apoptosis in Barrett’s-derived esophageal adenocarcinoma cells through the Fas-caspase cascade.59 This meta-analysis shows an inverse relationship of CagA-positive H. pylori strains with ESCC in Asian populations. But in non-Asian populations, the relationship was positive. There are several possible explanations for the inverse association. Hu et al. suggest that cyclooxygenase-2–1195 polymorphism plays a role in modifying the inverse association between H. pylori infection and risk of ESCC.45 There is also a hypothesis that postulates that the apparent protection associated with H. pylori infection is mediated via gastric atrophy and a reduced load of esophageal acid.51 Besides, a recent study provides indirect evidence of the inverse association between H. pylori infection and ESCC risk, which is possibly due to the H. pylori-induced apoptosis in ESCC cells.32 H. pylori does not live in the esophagus,
652
Diseases of the Esophagus
but H. pylori-induced gastric atrophy and subsequently reduced gastric acidity may allow overgrowth of bacteria, which produce nitrosamines and thereby increase esophageal cancer risk.37 The reason for the different association in Asian and non-Asian population may be the different diet and lifestyle. Although there were other meta-analyses examining the association between H. pylori and esophageal neoplasm.60–62 Our meta-analysis differs from these previous evaluations in several respects. One difference is the choice of the published articles. The former meta-analyses contain only English papers, but in this meta-analysis Chinese papers also have been absorbed. Although some articles are the same with the formers, new studies were added. Another difference is the result. In the stratified analysis of study location, an inverse association was found between CagA-positive H. pylori strains and ESCC in Asian populations; but a positive association was found in non-Asian populations. Except for these differences, all meta-analyses found that H. pylori may be a protective factor for EAC. Same with the former meta-analysis, the association of H. pylori with EAC was largely homogeneous across studies, but there was substantial and statistically significant heterogeneity in results with respect to ESCC. Strengths of this meta-analysis included a large number of cases/controls and several methods to examine study heterogeneity. However, the potential limitations of this meta-analysis should be considered. First, this meta-analysis included only English- and Chinese-language articles, which might influence the pooled estimated value. Second, the papers searched in our study were limited to those openly published. It is possible that some related published or unpublished studies that might have met the inclusion criteria were missed; hence, some inevitable publication biases might exist. Third, because of the inability to obtain raw data, we could perform only a study-level but not a patient-level meta-analysis, which would have enabled us to adjust for multiple factors. In conclusion, H. pylori and CagA-positive H. pylori strains infection might be a preventive factor for EAC. No association was found between H. pylori/H. pylori CagA-positive and ESCC. But in the stratified analysis of study location, an inverse association was found between CagA-positive H. pylori strains and ESCC in Asian populations, and a positive association was found in non-Asian populations. Because the potential biases and confounders in the meta-analysis could not be ruled out completely, these results need to be confirmed by further studies. References 1 Sachs G, Scott D R. Helicobacter pylori: eradication or preservation. F1000 Med Rep 2012; 4: 7.
2 Arkkila P E, Kokkola A, Seppala K, Sipponen P. Size of the peptic ulcer in Helicobacter pylori-positive patients: association with the clinical and histological characteristics. Scand J Gastroenterol 2007; 42: 695–701. 3 Li X B, Ge Z Z, Chen X Y, Liu W Z. Duodenal gastric metaplasia and Helicobacter pylori infection in patients with diffuse nodular duodenitis. Braz J Med Biol Res 2007; 40: 897–902. 4 Mhaskar R S, Ricardo I, Azliyati A et al. Assessment of risk factors of Helicobacter pylori infection and peptic ulcer disease. J Glob Infect Dis 2013; 5: 60–7. 5 Huang J Q, Zheng G F, Sumanac K, Irvine E J, Hunt R H. Meta-analysis of the relationship between cagA seropositivity and gastric cancer. Gastroenterology 2003; 125: 1636–44. 6 International Agency for Research on Cancer. Schistosomes, liver flukes and Helicobacter pylori. IARC Working Group on the evaluation of carcinogenic risks to humans. Lyon, 7–14 June 1994. IARC Monogr Eval Carcinog Risks Hum 1994; 61: 1–241. 7 Wang C, Yuan Y, Hunt R H. The association between Helicobacter pylori infection and early gastric cancer: a metaanalysis. Am J Gastroenterol 2007; 102: 1789–98. 8 Bourke B. Will treatment of Helicobacter pylori infection in childhood alter the risk of developing gastric cancer? Can J Gastroenterol 2005; 19: 409–11. 9 Stolzenberg-Solomon R Z, Blaser M J, Limburg P J et al. Helicobacter pylori seropositivity as a risk factor for pancreatic cancer. J Natl Cancer Inst 2001; 93: 937–41. 10 Raderer M, Wrba F, Kornek G et al. Association between Helicobacter pylori infection and pancreatic cancer. Oncology 1998; 55: 16–9. 11 Limburg P J, Stolzenberg-Solomon R Z, Colbert L H et al. Helicobacter pylori seropositivity and colorectal cancer risk: a prospective study of male smokers. Cancer Epidemiol Biomarkers Prev 2002; 11 (10 Pt 1): 1095–9. 12 Shmuely H, Passaro D, Figer A et al. Relationship between Helicobacter pylori CagA status and colorectal cancer. Am J Gastroenterol 2001; 96: 3406–10. 13 Khoshbaten M, Zadimani A, Bonyadi M R, Mohammadzadeh M, Gachkar L, Pourhoseingholi M A. Helicobacter pylori infection reduces the risk of esophageal squamous cell carcinoma: a case-control study in Iran. Asian Pac J Cancer Prev 2011; 12: 149–51. 14 Fruh M, Zhou W, Zhai R et al. Polymorphisms of inflammatory and metalloproteinase genes, Helicobacter pylori infection and the risk of oesophageal adenocarcinoma. Br J Cancer 2008; 98: 689–92. 15 Kamangar F, Qiao Y L, Blaser M J et al. Helicobacter pylori and oesophageal and gastric cancers in a prospective study in China. Br J Cancer 2007; 96: 172–6. 16 Hatakeyama M. Oncogenic mechanisms of the Helicobacter pylori CagA protein. Nat Rev Cancer 2004; 4: 688–94. 17 Chow W H, Blaser M J, Blot W J et al. An inverse relation between cagA+ strains of Helicobacter pylori infection and risk of esophageal and gastric cardia adenocarcinoma. Cancer Res 1998; 58: 588–90. 18 Anderson L A, Murphy S J, Johnston B T et al. Relationship between Helicobacter pylori infection and gastric atrophy and the stages of the oesophageal inflammation, metaplasia, adenocarcinoma sequence: results from the FINBAR casecontrol study. Gut 2008; 57: 734–9. 19 Siman J H, Engstrand L, Berglund G, Forsgren A, Floren C H. Helicobacter pylori and CagA seropositivity and its association with gastric and oesophageal carcinoma. Scand J Gastroenterol 2007; 42: 933–40. 20 Iijima K, Koike T, Abe Y et al. Extensive gastric atrophy: an increased risk factor for superficial esophageal squamous cell carcinoma in Japan. Am J Gastroenterol 2007; 102: 1603– 9. 21 El-Omar E M, Rabkin C S, Gammon M D et al. Increased risk of noncardia gastric cancer associated with proinflammatory cytokine gene polymorphisms. Gastroenterology 2003; 124: 1193–201. 22 Wu D C, Wu I C, Lee J M et al. Helicobacter pylori infection: a protective factor for esophageal squamous cell carcinoma in a Taiwanese population. Am J Gastroenterol 2005; 100: 588– 93. © 2014 International Society for Diseases of the Esophagus
HP infection and esophageal cancer 23 Wells G A, Shea B, O’Connell D et al. The Newcastle–Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses. 2000. [Cited 3 Apr 2013.] Available from URL: http://www.ohri.ca/programs/clinical_epidemiology/ oxford.asp. 24 Higgins J P, Thompson S G. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 1539–58. 25 Higgins J P, Thompson S G, Deeks J J, Altman D G. Measuring inconsistency in meta-analyses. BMJ 2003; 327: 557–60. 26 DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 177–88. 27 Tobias A. Assessing the influence of a single study in the metaanalysis estimate. Stata Tech Bull 1999; 8: 15–7. 28 Egger M, Davey S G, Schneider M, Minder C. Bias in metaanalysis detected by a simple, graphical test. BMJ 1997; 315: 629–34. 29 Lu H Effect of HPV and HP on the incidence of esophageal cancer and its association with p53 and p63 (Master Thesis). Shijiazhuang, Hebei, China: Hebei Medical University, 2012. (in Chinese). 30 Mao K, Ding X, Cui J, Et A. An analysis of Helicobacter pylori infection for esophageal cancer in high-risk and low-risk areas in southern and northern China. Chin J Diffic and Compl Cas 2011; 10: 676–8. (in Chinese). 31 Cook M B, Dawsey S M, Diaw L et al. Serum pepsinogens and Helicobacter pylori in relation to the risk of esophageal squamous cell carcinoma in the alpha-tocopherol, betacarotene cancer prevention study. Cancer Epidemiol Biomarkers Prev 2010; 19: 1966–75. 32 Wu I C, Wu D C, Yu F J et al. Association between Helicobacter pylori seropositivity and digestive tract cancers. World J Gastroenterol 2009; 15: 5465–71. 33 Derakhshan M H, Malekzadeh R, Watabe H et al. Combination of gastric atrophy, reflux symptoms and histological subtype indicates two distinct aetiologies of gastric cardia cancer. Gut 2008; 57: 298–305. 34 Zheng Y, Wang S, Yang C. A research of the relationship between the infection of Helicobacter pylori and esophageal squamons cell carcinoma. Int J Dig Dis 2008; 28: 433–4. (in Chinese). 35 Xu H, Wu B. The relationship between infection of HP and disease of upper digestive tract. Mod Med J China 2006; 8: 7–8, (in Chinese). 36 de Martel C, Llosa A E, Farr S M et al. Helicobacter pylori infection and the risk of development of esophageal adenocarcinoma. J Infect Dis 2005; 191: 761–7. 37 Ye W, Held M, Lagergren J et al. Helicobacter pylori infection and gastric atrophy: risk of adenocarcinoma and squamouscell carcinoma of the esophagus and adenocarcinoma of the gastric cardia. J Natl Cancer Inst 2004; 96: 388–96. 38 Wu A H, Crabtree J E, Bernstein L et al. Role of Helicobacter pylori CagA+ strains and risk of adenocarcinoma of the stomach and esophagus. Int J Cancer 2003; 103: 815–21. 39 Wang K X, Wang X F, Peng J L, Cui Y B, Wang J, Li C P. Detection of serum anti-Helicobacter pylori immunoglobulin G in patients with different digestive malignant tumors. World J Gastroenterol 2003; 9: 2501–4. 40 Grimley C E, Holder R L, Loft D E, Morris A, Nwokolo C U. Helicobacter pylori-associated antibodies in patients with duodenal ulcer, gastric and oesophageal adenocarcinoma. Eur J Gastroen Hepat 1999; 11: 503–9. 41 Vicari J J, Peek R M, Falk G W et al. The seroprevalence of cagA-positive Helicobacter pylori strains in the spectrum of gastroesophageal reflux disease. Gastroenterology 1998; 115: 50–7. 42 Qing Y, Chen Y, Xu J, Huang S, Xu Q. Correlation between Helicobacter Pylori infection and esophageal cancer. Pract J Cancer 1998; 13: 35–6. (in Chinese).
© 2014 International Society for Diseases of the Esophagus
653
43 Talley N J, Zinsmeister A R, Weaver A et al. Gastric adenocarcinoma and Helicobacter pylori infection. J Natl Cancer Inst 1991; 83: 1734–9. 44 Venerito M, Kohrs S, Wex T et al. Helicobacter pylori infection and fundic gastric atrophy are not associated with esophageal squamous cell carcinoma: a case-control study. Eur J Gastroenterol Hepatol 2011; 23: 859–64. 45 Hu H M, Kuo C H, Lee C H et al. Polymorphism in COX-2 modifies the inverse association between Helicobacter pylori seropositivity and esophageal squamous cell carcinoma risk in Taiwan: a case control study. BMC Gastroenterol 2009; 9: 37. 46 Weston A P, Badr A S, Topalovski M, Cherian R, Dixon A, Hassanein R S. Prospective evaluation of the prevalence of gastric Helicobacter pylori infection in patients with GERD, Barrett’s esophagus, Barrett’s dysplasia, and Barrett’s adenocarcinoma. Am J Gastroenterol 2000; 95: 387–94. 47 Whiteman D C, Parmar P, Fahey P et al. Association of Helicobacter pylori infection with reduced risk for esophageal cancer is independent of environmental and genetic modifiers. Gastroenterology 2010; 139: 73–83. e11–2. 48 Palladino-Davis A G, Mendez B M, Fisichella P M, Davis C S. Dietary habits and esophageal cancer. Dis Esophagus 2013; doi:10.1111/dote.12097 (Epub ahead of print). 49 Iftikhar S Y, James P D, Steele R J, Hardcastle J D, Atkinson M. Length of Barrett’s oesophagus: an important factor in the development of dysplasia and adenocarcinoma. Gut 1992; 33: 1155–8. 50 Williamson W A, Ellis F J, Gibb S P et al. Barrett’s esophagus. Prevalence and incidence of adenocarcinoma. Arch Intern Med 1991; 151: 2212–6. 51 Richter J E, Falk G W, Vaezi M F. Helicobacter pylori and gastroesophageal reflux disease: the bug may not be all bad. Am J Gastroenterol 1998; 93: 1800–2. 52 Varanasi R V, Fantry G T, Wilson K T. Decreased prevalence of Helicobacter pylori infection in gastroesophageal reflux disease. Helicobacter 1998; 3: 188–94. 53 Xie T, Cui X, Zheng H, Chen D, He L, Jiang B. Meta-analysis: eradication of Helicobacter pylori infection is associated with the development of endoscopic gastroesophageal reflux disease. Eur J Gastroenterol Hepatol 2013; 25: 1195–205. 54 Mager D L. Bacteria and cancer: cause, coincidence or cure? A review. J Transl Med 2006; 4: 14. 55 Nwokolo C U, Freshwater D A, O’Hare P, Randeva H S. Plasma ghrelin following cure of Helicobacter pylori. Gut 2003; 52: 637–40. 56 Nakazato M, Murakami N, Date Y et al. A role for ghrelin in the central regulation of feeding. Nature 2001; 409: 194–8. 57 Wren A M, Seal L J, Cohen M A et al. Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 2001; 86: 5992. 58 Spechler S J. Screening and surveillance for complications related to gastroesophageal reflux disease. Am J Med 2001; 111 (8, Suppl. 1): 130–6. 59 Jones A D, Bacon K D, Jobe B A, Sheppard B C, Deveney C W, Rutten M J. Helicobacter pylori induces apoptosis in Barrett’sderived esophageal adenocarcinoma cells. J Gastrointest Surg 2003; 7: 68–76. 60 Islami F, Kamangar F. Helicobacter pylori and esophageal cancer risk: a meta-analysis. Cancer Prev Res (Phila) 2008; 1: 329–38. 61 Zhuo X, Zhang Y, Wang Y, Zhuo W, Zhu Y, Zhang X. Helicobacter pylori infection and oesophageal cancer risk: association studies via evidence-based meta-analyses. Clin Oncol (R Coll Radiol) 2008; 20: 757–62. 62 Rokkas T, Pistiolas D, Sechopoulos P, Robotis I, Margantinis G. Relationship between Helicobacter pylori infection and esophageal neoplasia: a meta-analysis. Clin Gastroenterol Hepatol 2007; 5: 1413–7.
Review article
Association of Helicobacter pylori infection with esophageal adenocarcinoma and squamous cell carcinoma: a meta-analysis S. Nie,1,2 T. Chen,1 X. Yang,3 P. Huai,2 M. Lu1,2 Clinical Epidemiology Unit, 3Department of Gastroenteroloy, Qilu Hospital of Shandong University, and Department of Epidemiology and Health Statistics, School of Public Health, Shandong University, Jinan, China
1
2
SUMMARY. To evaluate the relationship of Helicobacter pylori and cytotoxin-associated gene A (CagA) positive strains with esophageal neoplasm, including esophageal adenocarcinoma (EAC) and esophageal squamous cell carcinoma (ESCC), the authors conducted a meta-analysis using a predefined protocol. PubMed, Web of Science, China biology medical literature database, Wanfang, and China National Knowledge Infrastructure were searched for relevant articles from the first available year to April 8, 2013. The fixed or random effect pooled measure was selected based on heterogeneity among studies, which was evaluated using Q test and the I 2 of Higgins and Thompson. Metaregression was used to explore the sources of between-study heterogeneity. Publication bias was analyzed by Begg’s funnel plot and Egger’s regression test. The association was assessed by odds ratio (OR) with 95% confidence interval (CI). A total of 28 eligible studies were included in the meta-analysis. There was a significant inverse association between H. pylori infection (pooled OR, 0.57; 95% CI, 0.44–0.73) and EAC; CagA-positive H. pylori strains were less likely to be associated with EAC compared with CagA-negative strains (pooled OR, 0.64; 95% CI, 0.52–0.79). However, there was no statistically significant association between H. pylori/CagA-positive H. pylori strains infection and ESCC, and the pooled ORs were 1.16 (95% CI, 0.83–1.60) and 0.97 (95% CI, 0.79–1.19). But significant associations between CagA-positive H. pylori strains infection and ESCC risk were found in the stratified analysis of the study location (Asian and non-Asian), and the summary ORs were 0.74 (95% CI, 0.57–0.97) and 1.41 (95% CI, 1.02–1.94). H. pylori infection and CagA-positive strains are associated with decreased risk of EAC in the overall population. No significant association was found between H. pylori infection/CagA-positive strains and ESCC. But CagA-positive strains might have a positive association with ESCC in non-Asian population and an inverse association in Asian population. KEY WORDS: esophageal adenocarcinoma, esophageal squamous cell carcinoma, Helicobacter pylori, meta-analysis.
INTRODUCTION Helicobacter pylori infect about 50% of the world’s population.1 And many cases of peptic ulcers, gastritis, and duodenitis are confirmed to have an association with H. pylori infection.2–4 In addition, infection with cytotoxin-associated gene A (CagA)-positive Address correspondence to: Dr Ming Lu, MD, PhD, Clinical Epidemiology Unit, Qilu Hospital of Shandong University, 107 Wenhuaxi Road, Jinan 250012, China. Email: [email protected] Sources of funding: This work was supported by the National Natural Science Foundation of China (grants 81072358 and 81273151). Conflict of interest: The authors have no competing interests. Disclosure: None. © 2014 International Society for Diseases of the Esophagus
strains of H. pylori increases the risk associated with H. pylori infection alone.5 H. pylori was classified as a group I carcinogen by the International Agency for Research on Cancer and the World Health Organization in 1994.6 H. pylori have been identified as the most important risk factor for gastric cancer.7,8 Epidemiologic studies also have investigated the association between H. pylori and other gastrointestinal malignancies, including pancreatic cancer,9,10 colorectal cancer,11,12 and esophageal cancer.13–15 Esophageal cancer is a disease in epidemiologic transition. Until the 1970s, the most common type of esophageal cancer in the United States was esophageal squamous cell carcinoma (ESCC), which has smoking and alcohol consumption as risk factors; 645
646
Diseases of the Esophagus
since then, there has been a progressive increase in the incidence of esophageal adenocarcinoma (EAC), for which the most common predisposing factor is gastroesophageal reflux disease (GERD). The two main histological types of esophageal cancer (ESCC and EAC) have distinct geographic and demographic patterns. Therefore, the associations of any potential causal factors should be evaluated separately for these two cancer types. H. pylori can be detected with a number of invasive methods, such as histological examination and urease test on gastric antral biopsies, or by non-invasive methods, such as serologic tests. H. pylori infection was defined as positivity to at least one test among histology, rapid urease test, and serology for both general anti-immunoglobulin G and anti-CagA. There are different strains of H. pylori. The genome of Cag-positive strains contains the cag pathogenicity island. This island includes approximately 31 putative genes, including cagA – the gene that encodes the CagA protein.16 CagA-positive strains confer a higher risk of non-cardia gastric cancer than CagAnegative strains.5 Likewise, the association between CagA-positive and CagA-negative strains with other cancers may be different17 and should be considered in analyses. Indeed, a large number of studies have examined the association between H. pylori infection and esophageal neoplasm (EAC and ESCC).13,14,18–20 But the role of H. pylori infection in esophageal neoplasm was controversial.21,13,18,22 We conducted this meta-analysis to summarize the published literature on H. pylori and esophageal neoplasm. We analyzed and reported the results separately for ESCC and EAC. In addition, we examined the role of CagA-positive H. pylori strains in these entities.
METHODS Literature search Sources of studies included the literature databases PubMed, Web of Science, China biology medical literature database, Wanfang, and China National Knowledge Infrastructure. Studies were searched from the inception of each database through April 8, 2013 using the key words ‘Helicobacter pylori’, ‘H. pylori’, or ‘HP’, and ‘esophageal neoplasm’, ‘esophageal carcinoma’, ‘esophageal tumor’, ‘esophageal cancer’, ‘esophageal squamous cell carcinoma’, ‘esophageal squamous carcinoma’, ‘ESCC’, ‘esophageal adenocarcinoma’, ‘adenocarcinoma of the esophagus’, or ‘EAC’. The subjects of studies were defined as humans, and the languages of articles were limited to English and Chinese because the reviewers are fluent in both of these languages. We evaluated potentially associated publications by checking their
titles and abstracts, and then picked out the most relevant publications for a closer examination. Moreover, the reference lists of the selected papers were also screened for other potential articles that possibly have been missed in the initial search (references cited in the identified articles were searched manually). Study selection Eligible studies were selected according to the following explicit inclusion criteria: 1 Esophageal cancer and H. pylori could be used as an outcome in the analysis. 2 The method(s) to detect H. pylori infection must be mentioned in the article. For example, histology, rapid urease test, and serology. 3 Case–control, nested case–control, or cohort studies were considered. 4 Sufficient data for estimating an odds ratio (OR) with 95% confidence interval (CI). 5 Searching the information without language restriction. We excluded studies based on the following criteria: 1 Articles published not in English or Chinese. 2 Studies that without full text articles or could not be obtained. 3 Reviews and repeated literatures were also excluded. 4 The study was not conducted on humans. 5 Not offering the source of cases and controls and other essential information. When data from multiple reports were identified, we included only the report with the most complete relevant data. Data extraction and quality assessment Data were extracted and entered into a database. Two investigators extracted information from all eligible publications independently according to the criteria listed above. The following data were collected from each study: name of the first author, year of publication, study location, study design (source of controls), number in case (exposed) and control (unexposed) groups, and the methods to assay the H. pylori infection, respectively. Study design was stratified to population-based (PB) studies and hospital-based (HB) studies. The quality of all included studies was assessed using the Newcastle– Ottawa Scale as recommended by the Cochrane NonRandomized Studies Methods Working Group.23 Data extraction and quality assessment were performed by two independent investigators. Any disagreement was settled by discussion. Statistical analysis The association between H. pylori/CagA-positive H. pylori strains and EAC/ESCC was measured by © 2014 International Society for Diseases of the Esophagus
HP infection and esophageal cancer
647
Potentially eligible titles initially generated by the literature searches (n = 948)
592 titles rejected (titles suggested article not appropriate)
356 titles retrieved
185 abstracts rejected (duplications, review articles)
171 abstracts accepted (with extractable data eligible)
146 studies excluded as not fulfilling the inclusion criteria Studies included in qualitative synthesis (n = 28, 23 in English and 5 in Chinese) Fig. 1 Flow chart of study selection.
pooled ORs with 95% CIs. Fourfold table data were obtained from each study respectively. The heterogeneity among different studies was checked by the Q-test and I 2 statistics.24 I 2 > 50% indicated evidence of heterogeneity. The ORs were pooled through a fixed effects model,25 using the Mantel– Haenszel approach when no heterogeneity was observed among studies. Otherwise, a random effects model was adopted.26 Metaregression with restricted maximum likelihood estimation was performed to explore the potentially important sources of betweenstudy heterogeneity. Sensitivity analysis was performed to validate the stability of outcomes by sequential removal of each individual study.27 An individual study is suspected to excessively influence the point estimate if its omitted analysis lies outside the 95% CI of the combined analysis. Publication bias was diagnosed with Begg’s funnel plot and Egger’s linear regression method.28 Asymmetric or incomplete funnel shaped plots and P < 0.05 in Egger’s test indicated the presence of potential publication bias. All statistical analyses used STATA 11.0 (StataCorp LP, College Station, TX, USA).
RESULTS Characteristics of studies The literature search identified 948 potentially relevant articles. Based on our inclusion criteria, © 2014 International Society for Diseases of the Esophagus
a total of 28 studies13–15,17–22,29–47 were included in qualitative synthesis (Fig. 1). The total population of the included studies was 10 285 persons, including 3269 cases and 7016 controls. Of the 28 studies, 23 are English and five are Chinese papers. Detailed characteristics of the aggregated data for the 28 studies are summarized in Table 1. In total, nine studies14,17,18,33,36,38,40,41,46 concerned EAC, 15 concerned ESCC,13,15,20,22,29–32,34,35,39,42–45 and four19,21,37,47 concerned both. There were 13 studies in Asian populations and 15 studies in non-Asian subjects.
QUANTITATIVE SYNTHESIS AND TEST OF HETEROGENEITY To evaluate the association between H. pylori infection and EAC risk, 1145 cases and 3453 controls were obtained from 13 studies, of which 372 cases and 1380 controls were H. pylori positive, and 773 cases and 2073 controls were H. pylori negative. As shown in Figure 2, the overall OR was 0.57 (95% CI 0.44–0.73). Significant study heterogeneity(I 2 = 53.3%) was found so a random effects model was used to pool the ORs. For the assessment of the association of CagApositive H. pylori strains with EAC, 562 cases and 1439 controls were obtained from seven studies, of which 225 cases and 712 controls were positive, and 337 cases and 727 controls were negative. As shown in Figure 3, the overall OR was 0.64 (95% CI
Lu H Venerito M Khoshbaten M Mao K Cook MB Wu IC Fruh M Zheng Y Kamangar F Siman JH Iijima K Anderson LA Derakhshan MH Xu H Wu DC de Martel C Ye W El-Omar EM Wang KX Wu AH Grimley CE Vicari JJ Chow WH Qing Y Talley NJ Hu HM Weston AP Whiteman DC
2012 2011 2011 2011 2010 2009 2008 2008 2007 2007 2007 2007 2007 2006 2005 2005 2004 2003 2003 2002 1999 1998 1998 1998 1991 2009 2000 2010
China Germany Iran China Finland Taiwan United States China China Sweden Japan Ireland Iran China Taiwan United States Sweden United States China United States England United States United States China United States Taiwan United States Australia
Study location PCR S,H,U S PCR S S S U S S S,H,U S S U S S S S S S S S S U S S H S
HP Dx — — — — — — 36/100 — — 7/12 — 100/208 9/19 — — 19/51 18/97 35/108 — 49/80 16/40 7/21 38/129 — — — 3/20 35/260
HP+
0/7 12/38 — — — — —
9/18 42/97 — — 18/80
— —
— — — — — — 29/100 — — 6/12 — 109/208
CagA+
Case
— — — — — — 30/101 — — 32/47 — 150/253 — — — 44/71 293/499 — — 106/356 — 11/26 46/86 — — — — —
CagA+
Control
— — — — — — 43/101 — — 35/47 — 157/253 28/38 — — 74/150 198/499 84/210 — 230/356 21/46 26/57 86/223 — — — 96/127 302/1316
HP+
EAC
17/25 53/75 58/100 122/160 57/71 112/317 — 67/96 254/335 24/37 60/73 — — 23/35 28/127 — 32/85 31/53 33/63 — — — — 39/43 20/41 66/180 — 54/208
HP+ — 42/75 28/100 — 32/46 91/99 — — 178/254 24/37 — — — — — — 63/85 — — — — — — 82 — — — —
CagA+
Case
40/75 36/100 — 41/60 142/153 — — 552/727 82/129 — — — — — — 293/499 — — — — — — 131 — — — —
CagA+
Control
9/25 53/75 83/100 39/80 68/87 160/395 — 88/157 727/992 87/129 56/73 — — 308/353 74/171 — 198/499 84/210 145/310 — — — — 43/170 96/252 102/194 — 302/1316
HP+
ESCC
HB HB HB HB PB PB HB HB PB PB HB PB HB HB PB PB PB PB NR PB HB HB PB HB HB HB HB PB
Study type
No Yes Yes No No Yes Yes No No Yes Yes Yes Yes No No Yes No Yes Yes Yes No No Yes No No Yes No Yes
Matched
5 8 8 6 7 8 8 6 7 9 8 9 8 6 7 9 9 9 7 9 6 6 9 6 6 8 6 9
Scores
—, no data available; EAC, esophageal adenocarcinoma; ESCC, esophageal squamous cell carcinoma; H, histology; HB, hospital based; HP Dx, HP detection method; NR, not reported; PB, population based; PCR, polymerase chain reaction; S, serology; U, rapid urease test.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Publication year
Characteristics of included studies on H. pylori and esophageal neoplasm
First author
Table 1
648 Diseases of the Esophagus
© 2014 International Society for Diseases of the Esophagus
HP infection and esophageal cancer
649
Fig. 2 Random effects meta-analysis of the association between H. pylori (HP) and esophageal adenocarcinoma (EAC). CI, confidence interval; OR, odds ratio.
0.52–0.79). There was no significant study heterogeneity (I 2 = 0.0%) so a fixed effects model was used to pool the ORs. To evaluate the possible association of H. pylori infection with ESCC risk, 2124 cases and 5588 controls obtained from 19 studies were included, of which 1150 cases and 2722 controls were H. pylori positive, and 974 cases and 2866 controls were H. pylori negative. As shown in Figure 4, the overall OR was 1.16 (95% CI 0.83–1.60). There was statistical heterogeneity (I 2 = 85.7%) among studies so a random effects model was used to pool the ORs. For the assessment of the association of CagApositive H. pylori strains with ESCC, 696 cases and 1743 controls were obtained from seven studies, of which 458 cases and 1186 controls were positive, and 238 cases and 557controls were negative. As shown in Figure 5, the overall OR was 0.97 (95% CI 0.79– 1.19). There was no statistical study heterogeneity (I 2 = 49.7%) so a fixed effects model was used to pool the ORs. Exploration of the sources of heterogeneity As seen above, strong heterogeneity among studies on H. pylori and ESCC was demonstrated. To further explore the potential sources of heterogeneity and test © 2014 International Society for Diseases of the Esophagus
the effects of study characteristics on the overall estimates, exploratory univariate metaregression was performed with study location (Asian and nonAsian), source of controls (PB or HB), and publication year. However, none of the variables was identified as potentially important source of betweenstudy heterogeneity. Sensitivity analysis and publication bias The influence of each study on the pooled ORs was examined by repeating the meta-analysis while sequentially omitting individual studies. Both sensitivity analysis for EAC and ESCC indicated that no single study influenced the pooled ORs qualitatively, suggesting that the results of our meta-analysis are stable. We plotted Begg’s funnel plot to examine small study effects. We also used Begg’s and Egger’s weighted regression method to calculate P for bias. The shape of the funnel plots did not reveal any evidence of obvious asymmetry. And the P-values of Egger’s test were 0.216, 0.170, 0.280, and 0.496 for H. pylori versus EAC, CagA-H. pylori versus EAC, H. pylori versus ESCC, and CagA-H. pylori versus ESCC, respectively, implying no existence of publication bias.
650
Diseases of the Esophagus
Fig. 3 Fixed effects meta-analysis of the association between cytotoxin-associated gene A (CagA)-positive H. pylori strains and esophageal adenocarcinoma (EAC). CagA-HP, CagA-positive H. pylori strains; CI, confidence interval; OR, odds ratio.
DISCUSSION The results showed an inverse statistically significant relationship of H. pylori infection and CagA-positive
H. pylori strains with EAC. In terms of ESCC, no significant association was found between H. pylori/CagA-positive H. pylori strains and ESCC. In the stratified analysis of study location, an
Fig. 4 Random effects meta-analysis of the association between H. pylori (HP) and esophageal squamous cell carcinoma (ESCC). CI, confidence interval; OR, odds ratio. © 2014 International Society for Diseases of the Esophagus
HP infection and esophageal cancer
651
Fig. 5 Random effects meta-analysis of the association between cytotoxin-associated gene A (CagA)-positive H. pylori strains and esophageal squamous cell carcinoma (ESCC). CagA-HP, CagA-positive H. pylori strains; CI, confidence interval; OR, odds ratio.
inverseassociation was found between CagA-positive H. pylori strains and ESCC in Asian populations; but a positive association was found in non-Asian populations. The reason for this phenomenon may be the difference in the carcinogenesis pathways between ESCC and EAC. As known to all, possible risk factors for ESCC include cigarette smoking, alcohol consumption, hot-temperature food, low intake of vegetables, salty food, pickled vegetables, nutrient deficiency, chronic mucosal irritation, and a family history of cancer,48 while EAC is closely related to Barrett’s esophagus.49,50 Barrett’s esophagus plays a pivotal role in the development of EAC. And the occurrence of Barrett-type ulcer is due to frequent gastroesophageal reflux, which leads to chronic inflammation and ulceration of the squamous epithelial lining, followed by re-epithelialization by gastric-type or intestinaltype epithelium in the low-PH microenvironment. Researchers have reported that H. pylori infection can protect against the development of GERD.51,52 The probable mechanisms of the protective effect may be that H. pylori infection can lead to gastric atrophy, lower intragastric acidity, and increased intragastric ammonia production. A study also revealed that eradication of the infection might be a risk factor for GERD, especially in Asian populations.53 Our study also found that H. pylori is a strong protective factor against EAC. The possible mechanisms are as follows. First, it may be related to gastroesophageal reflux. Mager54 considers that the © 2014 International Society for Diseases of the Esophagus
radical treatment of the H. pylori infection may bring about the increasing of the ghrelin in the serum,55 resulting in the occurrence of obesity.56,57 The function of the lower esophageal sphincter will be slack with the increasing of weight, leading to the increasing of GERD. And GERD is a risk factor for BE and EAC.58 Second, it may be that the decreased iron state reduces the morbidity of EAC. It is not the atrophic gastritis and gastratrophy reducing the incidence of EAC in Ye et al’s view.37 For example, people with pernicious anemia also have gastric mucosal atrophy, but the incidence of EAC is not declining. Third, Jones et al. have reported that H. pylori could induce apoptosis in Barrett’s-derived esophageal adenocarcinoma cells through the Fas-caspase cascade.59 This meta-analysis shows an inverse relationship of CagA-positive H. pylori strains with ESCC in Asian populations. But in non-Asian populations, the relationship was positive. There are several possible explanations for the inverse association. Hu et al. suggest that cyclooxygenase-2–1195 polymorphism plays a role in modifying the inverse association between H. pylori infection and risk of ESCC.45 There is also a hypothesis that postulates that the apparent protection associated with H. pylori infection is mediated via gastric atrophy and a reduced load of esophageal acid.51 Besides, a recent study provides indirect evidence of the inverse association between H. pylori infection and ESCC risk, which is possibly due to the H. pylori-induced apoptosis in ESCC cells.32 H. pylori does not live in the esophagus,
652
Diseases of the Esophagus
but H. pylori-induced gastric atrophy and subsequently reduced gastric acidity may allow overgrowth of bacteria, which produce nitrosamines and thereby increase esophageal cancer risk.37 The reason for the different association in Asian and non-Asian population may be the different diet and lifestyle. Although there were other meta-analyses examining the association between H. pylori and esophageal neoplasm.60–62 Our meta-analysis differs from these previous evaluations in several respects. One difference is the choice of the published articles. The former meta-analyses contain only English papers, but in this meta-analysis Chinese papers also have been absorbed. Although some articles are the same with the formers, new studies were added. Another difference is the result. In the stratified analysis of study location, an inverse association was found between CagA-positive H. pylori strains and ESCC in Asian populations; but a positive association was found in non-Asian populations. Except for these differences, all meta-analyses found that H. pylori may be a protective factor for EAC. Same with the former meta-analysis, the association of H. pylori with EAC was largely homogeneous across studies, but there was substantial and statistically significant heterogeneity in results with respect to ESCC. Strengths of this meta-analysis included a large number of cases/controls and several methods to examine study heterogeneity. However, the potential limitations of this meta-analysis should be considered. First, this meta-analysis included only English- and Chinese-language articles, which might influence the pooled estimated value. Second, the papers searched in our study were limited to those openly published. It is possible that some related published or unpublished studies that might have met the inclusion criteria were missed; hence, some inevitable publication biases might exist. Third, because of the inability to obtain raw data, we could perform only a study-level but not a patient-level meta-analysis, which would have enabled us to adjust for multiple factors. In conclusion, H. pylori and CagA-positive H. pylori strains infection might be a preventive factor for EAC. No association was found between H. pylori/H. pylori CagA-positive and ESCC. But in the stratified analysis of study location, an inverse association was found between CagA-positive H. pylori strains and ESCC in Asian populations, and a positive association was found in non-Asian populations. Because the potential biases and confounders in the meta-analysis could not be ruled out completely, these results need to be confirmed by further studies. References 1 Sachs G, Scott D R. Helicobacter pylori: eradication or preservation. F1000 Med Rep 2012; 4: 7.
2 Arkkila P E, Kokkola A, Seppala K, Sipponen P. Size of the peptic ulcer in Helicobacter pylori-positive patients: association with the clinical and histological characteristics. Scand J Gastroenterol 2007; 42: 695–701. 3 Li X B, Ge Z Z, Chen X Y, Liu W Z. Duodenal gastric metaplasia and Helicobacter pylori infection in patients with diffuse nodular duodenitis. Braz J Med Biol Res 2007; 40: 897–902. 4 Mhaskar R S, Ricardo I, Azliyati A et al. Assessment of risk factors of Helicobacter pylori infection and peptic ulcer disease. J Glob Infect Dis 2013; 5: 60–7. 5 Huang J Q, Zheng G F, Sumanac K, Irvine E J, Hunt R H. Meta-analysis of the relationship between cagA seropositivity and gastric cancer. Gastroenterology 2003; 125: 1636–44. 6 International Agency for Research on Cancer. Schistosomes, liver flukes and Helicobacter pylori. IARC Working Group on the evaluation of carcinogenic risks to humans. Lyon, 7–14 June 1994. IARC Monogr Eval Carcinog Risks Hum 1994; 61: 1–241. 7 Wang C, Yuan Y, Hunt R H. The association between Helicobacter pylori infection and early gastric cancer: a metaanalysis. Am J Gastroenterol 2007; 102: 1789–98. 8 Bourke B. Will treatment of Helicobacter pylori infection in childhood alter the risk of developing gastric cancer? Can J Gastroenterol 2005; 19: 409–11. 9 Stolzenberg-Solomon R Z, Blaser M J, Limburg P J et al. Helicobacter pylori seropositivity as a risk factor for pancreatic cancer. J Natl Cancer Inst 2001; 93: 937–41. 10 Raderer M, Wrba F, Kornek G et al. Association between Helicobacter pylori infection and pancreatic cancer. Oncology 1998; 55: 16–9. 11 Limburg P J, Stolzenberg-Solomon R Z, Colbert L H et al. Helicobacter pylori seropositivity and colorectal cancer risk: a prospective study of male smokers. Cancer Epidemiol Biomarkers Prev 2002; 11 (10 Pt 1): 1095–9. 12 Shmuely H, Passaro D, Figer A et al. Relationship between Helicobacter pylori CagA status and colorectal cancer. Am J Gastroenterol 2001; 96: 3406–10. 13 Khoshbaten M, Zadimani A, Bonyadi M R, Mohammadzadeh M, Gachkar L, Pourhoseingholi M A. Helicobacter pylori infection reduces the risk of esophageal squamous cell carcinoma: a case-control study in Iran. Asian Pac J Cancer Prev 2011; 12: 149–51. 14 Fruh M, Zhou W, Zhai R et al. Polymorphisms of inflammatory and metalloproteinase genes, Helicobacter pylori infection and the risk of oesophageal adenocarcinoma. Br J Cancer 2008; 98: 689–92. 15 Kamangar F, Qiao Y L, Blaser M J et al. Helicobacter pylori and oesophageal and gastric cancers in a prospective study in China. Br J Cancer 2007; 96: 172–6. 16 Hatakeyama M. Oncogenic mechanisms of the Helicobacter pylori CagA protein. Nat Rev Cancer 2004; 4: 688–94. 17 Chow W H, Blaser M J, Blot W J et al. An inverse relation between cagA+ strains of Helicobacter pylori infection and risk of esophageal and gastric cardia adenocarcinoma. Cancer Res 1998; 58: 588–90. 18 Anderson L A, Murphy S J, Johnston B T et al. Relationship between Helicobacter pylori infection and gastric atrophy and the stages of the oesophageal inflammation, metaplasia, adenocarcinoma sequence: results from the FINBAR casecontrol study. Gut 2008; 57: 734–9. 19 Siman J H, Engstrand L, Berglund G, Forsgren A, Floren C H. Helicobacter pylori and CagA seropositivity and its association with gastric and oesophageal carcinoma. Scand J Gastroenterol 2007; 42: 933–40. 20 Iijima K, Koike T, Abe Y et al. Extensive gastric atrophy: an increased risk factor for superficial esophageal squamous cell carcinoma in Japan. Am J Gastroenterol 2007; 102: 1603– 9. 21 El-Omar E M, Rabkin C S, Gammon M D et al. Increased risk of noncardia gastric cancer associated with proinflammatory cytokine gene polymorphisms. Gastroenterology 2003; 124: 1193–201. 22 Wu D C, Wu I C, Lee J M et al. Helicobacter pylori infection: a protective factor for esophageal squamous cell carcinoma in a Taiwanese population. Am J Gastroenterol 2005; 100: 588– 93. © 2014 International Society for Diseases of the Esophagus
HP infection and esophageal cancer 23 Wells G A, Shea B, O’Connell D et al. The Newcastle–Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses. 2000. [Cited 3 Apr 2013.] Available from URL: http://www.ohri.ca/programs/clinical_epidemiology/ oxford.asp. 24 Higgins J P, Thompson S G. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 1539–58. 25 Higgins J P, Thompson S G, Deeks J J, Altman D G. Measuring inconsistency in meta-analyses. BMJ 2003; 327: 557–60. 26 DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 177–88. 27 Tobias A. Assessing the influence of a single study in the metaanalysis estimate. Stata Tech Bull 1999; 8: 15–7. 28 Egger M, Davey S G, Schneider M, Minder C. Bias in metaanalysis detected by a simple, graphical test. BMJ 1997; 315: 629–34. 29 Lu H Effect of HPV and HP on the incidence of esophageal cancer and its association with p53 and p63 (Master Thesis). Shijiazhuang, Hebei, China: Hebei Medical University, 2012. (in Chinese). 30 Mao K, Ding X, Cui J, Et A. An analysis of Helicobacter pylori infection for esophageal cancer in high-risk and low-risk areas in southern and northern China. Chin J Diffic and Compl Cas 2011; 10: 676–8. (in Chinese). 31 Cook M B, Dawsey S M, Diaw L et al. Serum pepsinogens and Helicobacter pylori in relation to the risk of esophageal squamous cell carcinoma in the alpha-tocopherol, betacarotene cancer prevention study. Cancer Epidemiol Biomarkers Prev 2010; 19: 1966–75. 32 Wu I C, Wu D C, Yu F J et al. Association between Helicobacter pylori seropositivity and digestive tract cancers. World J Gastroenterol 2009; 15: 5465–71. 33 Derakhshan M H, Malekzadeh R, Watabe H et al. Combination of gastric atrophy, reflux symptoms and histological subtype indicates two distinct aetiologies of gastric cardia cancer. Gut 2008; 57: 298–305. 34 Zheng Y, Wang S, Yang C. A research of the relationship between the infection of Helicobacter pylori and esophageal squamons cell carcinoma. Int J Dig Dis 2008; 28: 433–4. (in Chinese). 35 Xu H, Wu B. The relationship between infection of HP and disease of upper digestive tract. Mod Med J China 2006; 8: 7–8, (in Chinese). 36 de Martel C, Llosa A E, Farr S M et al. Helicobacter pylori infection and the risk of development of esophageal adenocarcinoma. J Infect Dis 2005; 191: 761–7. 37 Ye W, Held M, Lagergren J et al. Helicobacter pylori infection and gastric atrophy: risk of adenocarcinoma and squamouscell carcinoma of the esophagus and adenocarcinoma of the gastric cardia. J Natl Cancer Inst 2004; 96: 388–96. 38 Wu A H, Crabtree J E, Bernstein L et al. Role of Helicobacter pylori CagA+ strains and risk of adenocarcinoma of the stomach and esophagus. Int J Cancer 2003; 103: 815–21. 39 Wang K X, Wang X F, Peng J L, Cui Y B, Wang J, Li C P. Detection of serum anti-Helicobacter pylori immunoglobulin G in patients with different digestive malignant tumors. World J Gastroenterol 2003; 9: 2501–4. 40 Grimley C E, Holder R L, Loft D E, Morris A, Nwokolo C U. Helicobacter pylori-associated antibodies in patients with duodenal ulcer, gastric and oesophageal adenocarcinoma. Eur J Gastroen Hepat 1999; 11: 503–9. 41 Vicari J J, Peek R M, Falk G W et al. The seroprevalence of cagA-positive Helicobacter pylori strains in the spectrum of gastroesophageal reflux disease. Gastroenterology 1998; 115: 50–7. 42 Qing Y, Chen Y, Xu J, Huang S, Xu Q. Correlation between Helicobacter Pylori infection and esophageal cancer. Pract J Cancer 1998; 13: 35–6. (in Chinese).
© 2014 International Society for Diseases of the Esophagus
653
43 Talley N J, Zinsmeister A R, Weaver A et al. Gastric adenocarcinoma and Helicobacter pylori infection. J Natl Cancer Inst 1991; 83: 1734–9. 44 Venerito M, Kohrs S, Wex T et al. Helicobacter pylori infection and fundic gastric atrophy are not associated with esophageal squamous cell carcinoma: a case-control study. Eur J Gastroenterol Hepatol 2011; 23: 859–64. 45 Hu H M, Kuo C H, Lee C H et al. Polymorphism in COX-2 modifies the inverse association between Helicobacter pylori seropositivity and esophageal squamous cell carcinoma risk in Taiwan: a case control study. BMC Gastroenterol 2009; 9: 37. 46 Weston A P, Badr A S, Topalovski M, Cherian R, Dixon A, Hassanein R S. Prospective evaluation of the prevalence of gastric Helicobacter pylori infection in patients with GERD, Barrett’s esophagus, Barrett’s dysplasia, and Barrett’s adenocarcinoma. Am J Gastroenterol 2000; 95: 387–94. 47 Whiteman D C, Parmar P, Fahey P et al. Association of Helicobacter pylori infection with reduced risk for esophageal cancer is independent of environmental and genetic modifiers. Gastroenterology 2010; 139: 73–83. e11–2. 48 Palladino-Davis A G, Mendez B M, Fisichella P M, Davis C S. Dietary habits and esophageal cancer. Dis Esophagus 2013; doi:10.1111/dote.12097 (Epub ahead of print). 49 Iftikhar S Y, James P D, Steele R J, Hardcastle J D, Atkinson M. Length of Barrett’s oesophagus: an important factor in the development of dysplasia and adenocarcinoma. Gut 1992; 33: 1155–8. 50 Williamson W A, Ellis F J, Gibb S P et al. Barrett’s esophagus. Prevalence and incidence of adenocarcinoma. Arch Intern Med 1991; 151: 2212–6. 51 Richter J E, Falk G W, Vaezi M F. Helicobacter pylori and gastroesophageal reflux disease: the bug may not be all bad. Am J Gastroenterol 1998; 93: 1800–2. 52 Varanasi R V, Fantry G T, Wilson K T. Decreased prevalence of Helicobacter pylori infection in gastroesophageal reflux disease. Helicobacter 1998; 3: 188–94. 53 Xie T, Cui X, Zheng H, Chen D, He L, Jiang B. Meta-analysis: eradication of Helicobacter pylori infection is associated with the development of endoscopic gastroesophageal reflux disease. Eur J Gastroenterol Hepatol 2013; 25: 1195–205. 54 Mager D L. Bacteria and cancer: cause, coincidence or cure? A review. J Transl Med 2006; 4: 14. 55 Nwokolo C U, Freshwater D A, O’Hare P, Randeva H S. Plasma ghrelin following cure of Helicobacter pylori. Gut 2003; 52: 637–40. 56 Nakazato M, Murakami N, Date Y et al. A role for ghrelin in the central regulation of feeding. Nature 2001; 409: 194–8. 57 Wren A M, Seal L J, Cohen M A et al. Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 2001; 86: 5992. 58 Spechler S J. Screening and surveillance for complications related to gastroesophageal reflux disease. Am J Med 2001; 111 (8, Suppl. 1): 130–6. 59 Jones A D, Bacon K D, Jobe B A, Sheppard B C, Deveney C W, Rutten M J. Helicobacter pylori induces apoptosis in Barrett’sderived esophageal adenocarcinoma cells. J Gastrointest Surg 2003; 7: 68–76. 60 Islami F, Kamangar F. Helicobacter pylori and esophageal cancer risk: a meta-analysis. Cancer Prev Res (Phila) 2008; 1: 329–38. 61 Zhuo X, Zhang Y, Wang Y, Zhuo W, Zhu Y, Zhang X. Helicobacter pylori infection and oesophageal cancer risk: association studies via evidence-based meta-analyses. Clin Oncol (R Coll Radiol) 2008; 20: 757–62. 62 Rokkas T, Pistiolas D, Sechopoulos P, Robotis I, Margantinis G. Relationship between Helicobacter pylori infection and esophageal neoplasia: a meta-analysis. Clin Gastroenterol Hepatol 2007; 5: 1413–7.
