Epidemiological research in stomach cancer: progress over the last ten years

Abstract

In this paper the progress of epidemiological research in stomach cancer during 1980-1990 is reviewed in respect to regional variation, etiology, and formation of carcinogens. The evaluation of 4 cohort and 16 case-control studies revealed a consistently inverse relationship of stomach cancer risk with raw vegetables, fruit, and wholemeal bread consumption and with vitamin C and carotene intake. Milk, cooked vegetables and vitamins A and E were not consistently found to be related to stomach cancer risk. Positive associations of increasing consumption with stomach cancer risk were occasionally found for processed or particularly prepared meat and fish, and for nitrite. Dietary nitrate intake did not appear to be related to stomach cancer risk in these studies. This latter observation is also supported by metabolic studies in high- and low-risk areas for stomach cancer. Consistently among studies, increased risk for stomach cancer was also found for later availability of refrigeration facilities in the household, non-centralized water supply (especially well water), and high salt intake. Prospective studies agreed in an increased risk for stomach cancer for cigarette smoking, but not for alcohol drinking, whereas case-control studies showed divergent results on these factors. Recent metabolic studies in high- and low-risk areas for stomach cancer or in groups with precursor lesions, with the N-nitrosoproline test as a marker for endogenous nitrosation, revealed inconsistent results. Higher nitrite concentration and increased pH in stomach juice were found to be associated with precursor conditions for stomach cancer. It is still not clear whether intake of preformed carcinogens or endogenous formation in the stomach with or without the inclusion of nitrite is the most important source of tumor-initiating or -promoting substances. Preservation or preparation of meat and fish may play an important role in this process, and vitamin C may be an inhibiting substance.