Is Helicobacter pylori really the cause of gastric cancer?

Since 1994, when the International Agency for Research on Cancer (IARC) designated Helicobacter pylori a human carcinogen, a considerable number of new studies have examined the association of this microorganism with risk of gastric cancer. The aim of this review was to critically evaluate current evidence of a causal relationship between the infection and gastric cancer in humans, taking these new studies into account. A MEDLINE search for the years 1982 through March 1998 yielded eight cohort studies, 40 case-control studies and eight ecologic studies. They were divided into four categories according to the weight of the evidence produced. In the category producing the strongest evidence, the cohort studies, five out of eight studies (including the three that formed the basis for the IARC decision) found a statistically significant excess risk of gastric cancer among the infected with odds ratios ranging between 2.8 and 6.0. Two cohort studies showed a non-significant tendency towards a positive association, but with point estimates indicating no more than a 50-60% excess risk among the H. pylori seropositive, and one apparently well-performed cohort study was completely negative. The two population-based case-control studies published to date found statistically non-significant odds ratios no higher than 1.3 and 1.4. Thus, the evidence of a strong causal link between H. pylori infection and gastric cancer has, if anything, become less convincing. Recent studies seem to indicate that the strength of the association may vary with type of H. pylori strain, and possibly presence of effect-modifying co-factors in the host and/or the environment. Future studies need to identify high-risk constellations of carcinogenic strains and unfavorable co-factors to make targeted prophylaxis cost-effective.