1. Top of page
  2. Abstract
  8. References


: The published epidemiological studies of chronic Helicobacter pylori infection and gastric cancer yield conflicting results, so there is uncertainty as to whether any material association exists and, if so, how strong it is.


: To review these studies quantitatively.


: A systematic review of sero-epidemiological studies published before 1998 of H. pylori and gastric cancer, as identified by computer-assisted literature searches of relevant journals, reference lists and discussions with authors. All relevant studies identified were included, subdivided by study design. The following was abstracted from published reports: adjusted odds ratio (or, in prospective studies, the risk ratio) and confidence interval, study design, type of controls, mean age, mean duration of follow-up, assay methods, location of study, and degree of adjustment for confounders.


: The 34 retrospective studies included in total 3300 gastric cancers, but their controls were of uncertain validity. The 10 ‘nested’ case–control comparisons in prospective studies included in total only 800 gastric cancers, and combined analysis of them yielded a risk ratio of 2.5 (95% CI: 1.9–3.4; 2P < 0.00001) for gastric cancer in people seropositive for H. pylori antibodies.


: The prospective studies suggest that gastric cancer is 2 or 3 times as common in those chronically infected by H. pylori, but to help investigate causality, further observational studies are still needed, as are large-scale randomized trials of whether antibacterial regimens reduce the eventual incidence of gastric cancer.


  1. Top of page
  2. Abstract
  8. References

When Helicobacter pylori was first identified, those that discovered it suggested that it might cause gastric cancer,1 which kills about a million people a year worldwide.2H. pylori is found in the stomachs of about one-third of the adults in developed countries (where it has been decreasing in prevalence) and in about two-thirds of the adults in developing countries.3 It is strongly correlated with lower socioeconomic status, usually acquired in early life, probably spread from person to person, and generally persists until old age unless eradicated with specific antibiotic treatment.4 A proportion of infected people develop peptic ulceration.5 There is, however, a great deal of uncertainty as to whether the infection can also predispose to gastric cancer. In 1994 the International Agency on Research in Cancer declared H. pylori a ‘carcinogen’,6 but a consensus panel of the National Institutes of Health disagreed.5 A number of additional studies have been reported recently, but expert opinions continue to differ, with claims ranging from a nine-fold association7 to no important association at all.8 To help clarify the actual evidence, this article provides a systematic review (meta-analysis) of the epidemiological studies.


  1. Top of page
  2. Abstract
  8. References

Epidemiological and clinical studies published between 1983 and 1998 that reported on H. pylori status in cases with gastric cancer and in controls without the disease were sought by Medline searches, scanning of reference lists, hand-searching of gastroenterology, epidemiology and other relevant journals, and discussions with authors of relevant reports. Combinations of key words were used in the searches (e.g. Helicobacter pylori, Campylobacter pylori, gastric adenocarcinoma, gastric cancer, stomach cancer).9[10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52]–53 (One study of 40 cases was found to be ineligible as it selected only controls who were known to be infected with H. pylori.54) Non-English articles were translated and abstracted according to a fixed protocol. The following information was obtained for each study (from published reports or from the investigators): adjusted odds ratio (or, in prospective studies, the risk ratio) and confidence interval; number of cases and controls; study design (‘nested’ case–control study or retrospective study); type of controls (prospective studies with ‘internal’ controls; healthy volunteers; hospital in-patients; patients with gastric diseases); mean age of cases (or, for prospective studies, mean age at entry and mean follow-up duration); assay methods; geographical location of study; and degree of adjustment for confounders denoted by: + age and sex only; ++ these plus smoking; +++ these plus socioeconomic status; ++++ these plus information on dietary habits. For prospective studies, combination of results involved inverse-variance-weighted log risk ratios. Heterogeneity was assessed with standard χ2-tests.


  1. Top of page
  2. Abstract
  8. References

Thirty-four retrospective epidemiological studies, involving in total about 3300 cases, have reported on the possible association between H. pylori and gastric cancer. Of these, the nine (including 571 cases) in which gastric biopsy was used to assess H. pylori status are not reviewed because the detection method is insensitive.34[35][36][37][38][39][40][41]–42 This leaves 25 retrospective case-control studies of H. pylori seropositivity (with 2700 cases). The results of each are given in Figure 1, but in general the controls (particularly those who had gastric disease) were of uncertain validity, and no formal combination of the published findings is appropriate. In addition, there were 10 ‘nested’ case–control comparisons within prospective studies (with about 800 cases, Figure 2) of H. pylori seropositivity and gastric cancer. These prospective studies avoid many of the problems of control selection, but it is still difficult to adjust properly for potential confounding factors. All made adjustments for age and sex, and several also attempted adjustment for smoking, social class or diet. The weighted mean age at baseline of the participants in the 10 studies was 56 years, with a weighted mean follow-up of 8 years (range 2–14 years). There was no evidence of significant heterogeneity among the prospective studies (χ29 = 14.5, > 0.05), and a combined analysis was done that yielded a risk ratio of 2.5 (95% CI: 1.9–3.4; 2P < 0.00001). Subdivision of the published prospective evidence by possibly relevant characteristics (e.g. study location, anatomical site of gastric cancer, duration of follow-up) is not possible because the data are too sparse, or reported in insufficient detail, or both.


Figure 1. . Odds ratios in retrospective sero-epidemiological studies of H. pylori and gastric cancer. Areas of black squares are proportional to numbers of cases. Degree of adjustment: + age and sex only; ++ these plus smoking; +++ these plus socioeconomic status; ++++ these plus information on dietary habits.

Download figure to PowerPoint


Figure 2. . Risk ratios in prospective sero-epidemiological studies of H. pylori and gastric cancer.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  8. References

The present meta-analysis differs from a recent review of H. pylori and gastric cancer55 in two main ways. First, whereas the present review involves a total of 44 different studies including more than 4000 cases, the previous report identified only 19 studies with 2500 cases, leaving it prone to substantial ‘publication bias’.56 Second, the previous review statistically combined the results of all 19 identified studies regardless of the study design. This is inappropriate because retrospective studies of H. pylori and gastric cancer suffer from important methodological limitations due to the use of inappropriate controls and limited adjustment for possible confounding factors. Figure 1 illustrates just such distortions: the use of healthy volunteers can over-represent controls from higher social classes, thereby tending to overestimate the strength of the association of H. pylori with gastric cancer, whereas the use of controls with conditions that are themselves associated with H. pylori infection (e.g. peptic ulceration or chronic gastritis) can lead to underestimation. Moreover, studies that failed to adjust appropriately for age, sex, cigarette smoking and indicators of socioeconomic status, either because such characteristics were not recorded or because they were not measured accurately, were prone to inflated estimates.

The prospective studies, by contrast, use internal controls ‘nested’ within cohorts from which blood samples were taken some years before the onset of clinical disease ( Figure 2). These studies should be more informative than retrospective reports because they avoid selection bias, assess infection before the onset of clinical disease, and assess H. pylori status using serum IgG antibodies (raised concentrations of which are usually reliable indicators of chronic gastric infection57 and, in the absence of specific treatment, generally persist without much fluctuation58). Figure 2 suggests a less important role for H. pylori than claimed in some previous commentaries.6 Moreover, there is still the possibility of some ‘publication bias’56 (whereby some of the studies with less extreme associations may not yet have been published), and of some residual confounding. Conversely, the infection’s role may have been underestimated by studies from developing countries that tested for H. pylori antigens from Western populations (because the prevalence of antigens can vary substantially between regions), or by studies that failed to exclude cancers reported in the first few years of follow-up (because disease itself may render some cases sero-negative).59

If even a two-fold relative risk could be confirmed (or, if stronger associations could be reliably demonstrated with specific H. pylori subtypes, such as cytotoxin-positive strains60), the effect would be sufficient to be of some practical relevance, for H. pylori would then be responsible for about one-quarter of the one million gastric cancer deaths each year. Large randomized trials with prolonged follow-up could ascertain whether some of these deaths might be avoidable by antibiotic treatment in middle age, but, even collectively, the presently recruiting trials of H. pylori eradication are too small (and, at present, too brief) to detect even a 25% reduction in the incidence of gastric cancer.61 In the UK, for example, a trial of H. pylori eradication at age 60 might have to randomize about 100 000 individuals (four times as many as are now being randomized in total worldwide) with at least one decade of follow-up to achieve statistically reliable results.62 Even larger numbers might be needed if there is a delay of some years before benefits emerge, if antibiotic treatments become more widely used for ‘non-ulcer dyspepsia’ or if gastric cancer rates fall more steeply than anticipated.

Whether or not the relatively weak association described in Figure 2 reflects causality, H. pylori cannot plausibly account for the sharp variations in gastric cancer mortality between different populations (e.g. 20-fold higher in certain parts of China than in the USA63) or between past and present (e.g. the five-fold decrease in Scotland between 1950 and 199064). Also, gastric cancer is twice as common in men as in women and may occur less frequently in people with duodenal ulceration, whereas H. pylori occurs roughly equally in both sexes and is strongly associated with peptic ulceration.65 These observations imply that important cause(s) of gastric cancer still await discovery.

In conclusion, published prospective sero-epidemiological studies indicate that gastric cancer is 2 or 3 times as common in those chronically infected by H. pylori, but it is not known how well publication bias and confounding have been controlled.


  1. Top of page
  2. Abstract
  8. References

Rory Collins, Richard Doll, Hellen Gelband, Richard Peto and Martin Vessey commented helpfully; Helgi Sigvaldason provided additional details from a study published as an abstract; Paul Appleby plotted the figures; Young-Ha Go and Francesca Signorelli translated articles. The study was supported by a Frohlich award.


  1. Top of page
  2. Abstract
  8. References
  • 1
    Warren JR & Marshall BJ. Unidentified curved bacilli on gastric epithelium in active chronic gastritis. Lancet 1983; 1: 12731273.
  • 2
    Murray CJL & Lopez AD. Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study. Lancet 1997; 349: 1498 504.
  • 3
    Pounder RE & Ng D. The prevalence of Helicobacter pylori infection in different countries. Aliment Pharmacol Ther 1995; 9(Suppl. 2): 33 9.
  • 4
    Goodwin CS, Mendall MM, Northfield TC. Helicobacter pylori infection. Lancet 1997; 349: 265 9.
  • 5
    NIH Consensus Development Panel. Helicobacter pylori in peptic ulcer disease. J Am Med Assoc 1994; 272: 65 9.
  • 6
    IARC Working Group. Monographs on the evaluation of the carcinogenic risks to humans: schistosomes, liver flukes and Helicobacter pylori. IARC Publications, Geneva, 1994, 177 241.
  • 7
    Webb PM & Forman D. Helicobacter pylori as a risk factor for cancer. Balliere’s Clin Gastroenterol 1996; 9: 563 82.
  • 8
    Crespi M & Citarda F. Helicobacter pylori and gastric cancer: an overrated risk? Scand J Gastroenterol 1996; 31: 1041 6.
  • 9
    Archimandritis A, Bitsakas J, Tjivras M, et al. Non-cardia gastric adenocarcinoma and Helicobacter pylori infection. Ital J Gastroenterol 1993; 25: 368 71.
  • 10
    Asaka M, Kimura T, Kato M, et al. Possible role of Helicobacter pylori infection in early gastric cancer development. Cancer 1994; 73: 2691 4.
  • 11
    Asaka M, Kato M, Kudo M, et al. Relationship between Helicobacter pylori infection, atrophic gastritis and gastric carcinoma in a Japanese population. Eur J Gastroenterol Hepatol 1995; 7(Suppl. 1): S7 10.
  • 12
    Kikuchi S, Wada O, Nakajima T, et al. Serum anti-Helicobacter pylori antibody and gastric carcinoma among young adults. Cancer 1995; 75: 2789 93.
  • 13
    Li D, Gao X, Guan H. Relationship between Helicobacter pylori infection and diffuse and intestinal types of stomach cancer. Chung Hua Liu 1995; 16: 299 302.
  • 14
    Estevens J, Fidalgo P, Tendeiro T, et al. Anti-Helicobacter pylori antibodies: prevalence and gastric adenocarcinoma in Portugal: report of a case-control study. Eur J Cancer Prev 1993; 2: 377 80.
  • 15
    Miglio F, Miglioli M, Holton J, et al. Prevalence of Helicobacter pylori in patients with gastric carcinoma. Irish J Med Sci 1992; 161(Suppl. 10): GCE 37GCE 37(Abstract).
  • 16
    Kang HC & Chung IS. Helicobacter pylori infection and gastric adenocarcinoma in Korea: prevalence and distribution of H. pylori in resected specimen of gastric cancer. J Cath Med Coll 1992; 45: 849 63.
  • 17
    Menegatti M, Vaira D, Miglioli M, et al. Helicobacter pylori in patients with gastric and nongastric cancer. Am J Gastroenterol 1995; 90: 1278 81.
  • 18
    Kurtenkov O, Klaamas K, Cavacci A, Wadstrom T. IgG immune response to a recombinant fragment of cagA in patients with gastric cancer: relation to the ABO(h) blood group phenotype and the stage of cancer. Gut 1997; 41(Suppl. 1): A48A48(Abstract).
  • 19
    Lin JT, Wang JT, Wu MS, Huang SC, Wang TH. Serum levels of pepsinogen I and gastrin in gastric carcinoma: the influence of Helicobacter pylori infection and tumor characteristics. Hepatogastroenterology 1993; 40: 600 3.
  • 20
    Hansson LE, Engstrand L, Nyren O, et al. Helicobacter pylori infection: independent risk indicator of gastric adenocarcinoma. Gastroenterology 1993; 105: 1098 103.
  • 21
    Talley NJ, Zinsmeister AR, Weaver A, et al. Gastric adenocarcinoma and Helicobacter pylori infection. J Natl Cancer Inst 1991; 83: 1734 9.
  • 22
    Rudi J, Muller M, Von Herbay A, et al. Lack of association of Helicobacter pylori seroprevalence and gastric cancer in a population with low gastric cancer incidence. Scand J Gastroenterol 1995; 30: 958 63.
  • 23
    Rudi J, Kolb C, Maiwald M, et al. Serum antibodies against Helicobacter pylori proteins VacA and CagA are associated with increased risk for gastric adenocarcinoma. Dig Dis Sci 1997; 42: 1652 9.
  • 24
    Barreto-Zuniga R, Maruyama M, Kato Y, et al. Significance of Helicobacter pylori infection as a risk factor in gastric cancer: serological and histological studies. J Gastroenterol 1997; 32: 289 94.
  • 25
    Sipponen P, Kosunen TU, Valle J, Rihela M, Seppala K. Helicobacter pylori infection and chronic gastritis in gastric cancer. J Clin Pathol 1992; 45: 319 23.
  • 26
    Blaser MJ, Kobayashi K, Cover TL, Cao P, Feurer ID, Perez-Perez GI. Helicobacter pylori infection in Japanese patients with adenocarcinoma of the stomach. Int J Cancer 1993; 55: 799 802.
  • 27
    Fukuda H, Saito D, Hayashi S, et al. Helicobacter pylori infection, serum pepsinogen level and gastric cancer: a case-control study in Japan. Jpn J Cancer Res 1995; 86: 64 71.
  • 28
    Chira C, Rovinaru I, Pop F, Raducan L, Copaci I, Vadan R. Gastric cancer and Helicobacter pylori infection. Gut 1996; 39(Suppl. 3): A68A68(Abstract).
  • 29
    Kuipers EJ, Gracia-Casanova M, Pena AS, et al. Helicobacter pylori serology in patients with gastric carcinoma. Scand J Gastroenterol 1993; 28: 433 7.
  • 30
    Kato S, Onda M, Matsukura N, et al. Genetic polymorphisms of the cancer related gene and Helicobacter pylori infection in Japanese gastric cancer patients. Cancer 1996; 77: 1654 61.
  • 31
    Ng TM, Fock KM, Khim MM, et al. Gastric cancer and H. pylori infection in Singapore: lack of association with cagA status. Gut 1997; 41(Suppl. 1): A47A47(Abstract).
  • 32
    Lin JT, Wang JT, Wang TH, Wu MS, Lee TK, Chen CJ. Helicobacter pylori infection in a randomly selected population, healthy volunteers, and patients with gastric ulcer and gastric adenocarcinoma. Scand J Gastroenterol 1993; 28: 1067 72.
  • 33
    Igarashi H, Takahashi S, Ishiyama N, et al. Is Helicobacter pylori a causal agent in gastric carcinoma? Irish J Med Sci 1992; 161(Suppl. 10): GCE 36GCE 36(Abstract).
  • 34
    Fiocca R, Luinetti O, Villani L, et al. High incidence of Helicobacter pylori colonization in early gastric cancer and the possible relationship to carcinogenesis. Eur J Gastroenterol Hepatol 1993; 5(Suppl. 2): S2 8.
  • 35
    Loffeld Rjfl, Willems I, Flendrig JA, Arends JW. Helicobacter pylori and gastric carcinoma. Histopathology 1990; 17: 537 41.
  • 36
    Kokkola A, Valle J, Haapainen R, Sipponen P, Kivilaakso E, Puolakkainen P. Helicobacter pylori infection in young patients with gastric carcinoma. Scand J Gastroenterol 1996; 31: 643 7.
  • 37
    Parsonnet J, Vandersteen D, Goates J, Sibley RK, Pritikin J, Chang Y. Helicobacter pylori infection in intestinal- and diffuse-type adenocarcinomas. J Natl Cancer Inst 1991; 83: 640 3.
  • 38
    Boixeda D, Gisbdert JP, Arocena C, Baleriola I, Martin de Argila C, Garcia Plaza A. Prevalencia de la infeccion por Helicobacter pylori en el adenocarcinoma gastrico y en la gastritis cronica. Rev Esp Enf Digest 1996; 88: 403 8.
  • 39
    Craanen ME, Blok P, Dekker W, Tytgat GN. Helicobacter pylori and early gastric cancer. Gut 1994; 35: 1372 4.
  • 40
    Shimoyama T, Fukuda S, Tanaka M, Mikami T, Saito Y, Munakata A. High prevalence of the CagA-positive Helicobacter pylori strains in Japanese asymptomatic patients and gastric cancer patients. Scand J Gastroenterol 1997; 32: 465 8.
  • 41
    Tokumaru K, Kojima T, Satoh K, et al. CagA gene of Helicobacter pylori is a not a carcinogenic marker in Japanese patients. Gut 1997; 41(Suppl. 1): A47A47(Abstract).
  • 42
    Kim HY, Cho BD, Chang WK, et al. Helicobacter pylori infection and the risk of gastric cancer among the Korean population. J Gastroenterol Hepatol 1997; 12: 100 3.
  • 43
    Webb PM, Yu MC, Forman D, et al. An apparent lack of association between Helicobacter pylori infection and risk of gastric cancer in China. Int J Cancer 1996; 67: 603 7.
  • 44
    Aromaa A, Kosunen TU, Knekt P, et al. Circulating anti-Helicobacter pylori immunoglobulin A antibodies and low serum pepsinogen I level are associated with increased risk of gastric cancer. Am J Epidemiol 1996; 144: 142 9.
  • 45
    Lin JT, Wang LY, Wang JT, Wang TH, Yang CS, Chen CJ. A nested case-control study on the association between Helicobacter pylori infection and gastric cancer in a cohort of 9775 men in Taiwan. Anticancer Res 1995; 15: 603 6.
  • 46
    Forman D, Newell DG, Fullerton F, et al. Association between infection with Helicobacter pylori and risk of gastric cancer: evidence from a prospective investigation. Br Med J 1991; 302: 1302 5.
  • 47
    Wald NJ, Law MR, Morris JK, Bagnall AM. Helicobacter pylori and mortality from ischaemic heart disease: a negative result from a large prospective study. Br Med J 1997; 315: 1199 201.
  • 48
    Parsonnet J, Friedman GD, Vandersteen DP, et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991; 325: 1127 31.
  • 49
    Hansen S, Vollset SE, Melby K, Jellum E. Strong association of previous Helicobacter pylori infection to distal gastric adenocarcinoma. Am J Gastroenterol 1994; 89: A1358A1358(Abstract).
  • 50
    Nomura A, Stemmermann GN, Chyou PH, Kato I, Perez-Perez GI, Blaser MJ. Helicobacter pylori infection and gastric carcinoma among Japanese Americans in Hawaii. N Engl J Med 1991; 325: 1132 6.
  • 51
    Siman JH, Forsgren A, Floren CH. Association between Helicobacter pylori and gastric carcinoma in the city of Malmo, Sweden—a prospective study . Gut 1997; 41(Suppl. 1): A45A45(Abstract).
  • 52
    Watanabe Y, Kurata JH, Mizuno S, et al. Helicobacter pylori infection and gastric cancer: a nested case-control study in a rural area of Japan. Dig Dis Sci 1998; 42: 1383 7.
  • 53
    Tulinius H, Ogmundsdottir H, Kristinsson KG, Tryggvadottir L, Jonasson I, Sigfusson N. A retrospective case-control study of H. pylori infection and subsequent gastric cancer in Iceland. IACR Annual Meeting, November 1997, abstract.
  • 54
    Mitchell HM, Hazell SL, Li YY, Hu PJ. Serological response to specific Helicobacter pylori antigens: antibody against CagA antigen is not predictive of gastric cancer in a developing country. Am J Gastroenterol 1996; 91: 1785 8.
  • 55
    Huang JQ, Sridhar S, Chen Y, Hunt R. Meta-analysis of the relationship between Helicobacter pylori seropositivity and gastric cancer. Gastroenterology 1998; 114: 1169 79.
  • 56
    Easterbrook P, Berlin J, Gopalan R, Matthews D. Publication bias in clinical research. Lancet 1991; 337: 867 72.
  • 57
    Feldman RA & Evans SJW. Accuracy of diagnostic methods used for epidemiological studies of Helicobacter pylori. Aliment Pharmacol Ther 1995; 9(Suppl. 2): 21 31.
  • 58
    Rosenstock SJ, Andersen LP, Bonnevie O, Jorgensen T. Sero-conversion and sero-reversion in IgG antibodies to H. pylori: an 11-year follow-up of 2,523 randomly selected Danes . Gut 1996; 39(Suppl. 3): A3A3(Abstract).
  • 59
    Munoz N. Is Helicobacter pylori a cause of gastric cancer? An appraisal of the seroepidemiological evidence. Cancer Epidemiol, Biomarkers Prevention 1994; 3: 445 51.
  • 60
    Blaser MJ, Perez-Perez GI, Kleanthous H. Infection with Helicobacter pylori strains possessing cagA is associated with an increased risk of developing adenocarcinoma of the stomach. Cancer Res 1995; 55: 2111 15.
  • 61
    Forman D. Lessons from ongoing intervention studies. In: Hunt R, Tytgat G, eds. Helicobacter pylori: Basic Mechanisms to Clinical Cure. Amsterdam: Kluwer Academic Publishers, 1998.
  • 62
    Danesh J, Forman D, Collins R, Peto R. Helicobacter pylori screening and gastric cancer. Lancet 1996; 348: 758 9.
  • 63
    Peto R. General reflections on principles and purposes: indirect and direct purposes of the present monograph. In: Junshi C, Campbell TC, Junyao L, Peto R, eds. Diet, Life-Style and Mortality in China: a Study of the Characteristics of 65 Chinese Counties. Oxford: Oxford University Press, 1990: 72–8.
  • 64
    Swerdlow AJ, Dos Santos Silva I, Reid A, Qiao Z, Brewster DH, Arrundale J. Trends in cancer incidence and mortality in Scotland. Br J Cancer 1998; 77(Suppl. 3): 1 16.
  • 65
    Howson CP, Hiyama T, Wynder EL. The decline in gastric cancer: epidemiology of an unplanned triumph. Epidemiol Rev 1986; 8: 1 27.