A case-control study on gastric cancer risk in relation to nutrient composition of diet was conducted in the 2 Belgian provinces of Oost-Vlaanderen and Liège as part of a large epidemiological study on cancers of the digestive tract, also including colorectal cancer. The statistical analysis was carried out on a total of 301 men and women aged 35-74 years with histologically confirmed stomach tumors and 2,851 population controls. Dietary intake assessments were obtained by interview, using a dietary history questionnaire. Gastric cancer risk was increased for diets rich in mono- and disaccharides, according to statistical models based on energy-adjusted residuals [odds ratio (OR) = 1.88 for upper vs. lower quartiles of energy-adjusted intakes] or energy decomposition models, but showed no increase for high intake of polysaccharides. Intake of polyunsaturated fats, particularly linoleic acid, was inversely associated with gastric cancer risk. High intake of vitamin C, beta carotene and vitamins B1 (thiamine), B3 (nicotinic acid) or B6 (pyridoxine) was also associated with decreased risk, whereas increased risk was observed for high intake of vitamin A (retinol) or vitamin B2 (riboflavin). Many of these associations between gastric cancer risk and nutrient composition were similar to those found in the analysis of a parallel study on cancers of the colon and rectum.