Mortality rates for gastric cancer have been declining in recent decades, most prominently in Western countries. Worldwide, however, gastric cancer is the second most commonly occurring malignancy, second only to lung cancer. Acknowledged risk factors for gastric cancer include Helicobacter pylori gastric infection; the World Health Organization has even classified this bacterium as a class I carcinogen. But how can infectious agents such as bacteria cause cancer?

As Richard Peek and Martin Blaser discuss in their Review on page 28, a complex interaction between host and bacterial factors seems to be involved, and less than 3% of individuals who carry H. pylori develop gastric neoplasia. A report recently published in Science shows that a protein produced by a particular strain of H. pylori can activate a putative oncogenic signalling cascade in gastric cells (see highlight on page 6). Further studies are required to identify the strains of H. pylori that are most carcinogenic and the host genotypes that confer the greatest cancer risk.

Prevention of gastric cancer by eradication of H. pylori infection has been considered in several countries. This might not be feasible, however, as almost half of the world's population is infected. Furthermore, the rise in the incidence of oesophageal adenocarcinoma has been attributed to a declining prevalence of H. pylori infection in Western countries, and gastric infection with H. pylori is believed to protect the oesophagus from gastro-oespohageal reflux disease. Therefore, elimination of the purported cause of one type of cancer might only lead to an increase in the incidence of other diseases. A better approach might be to improve our understanding of the pro-oncogenic and anti-oncogenic mechanisms of this pathogen and to develop therapeutics that target those pathways — or target specific bacterial strains in high-risk individuals.