Data from a series of case-control studies, conducted in Italy and Switzerland between 1991 and 2001, have been analyzed to evaluate the role of n-3 polyunsaturated fatty acid (PUFA) intake in the etiology of cancer of oral cavity and pharynx (736 cases, 1772 controls), esophagus (395 cases, 1066 controls), large bowel (1394 colon, 886 rectum, 4765 controls), breast (2900 cases, 3122 controls) and ovary (1031 cases, 2411 controls). Controls were patients admitted to hospital for acute, non-neoplastic conditions, unrelated to modifications in diet. The multivariate odds ratios (OR) for the highest quintile of n-3 PUFAs compared to the lowest one were 0.5 for oral and pharyngeal cancer, 0.5 for oesophageal cancer, 0.7 for colon cancer, 0.8 for rectal and breast cancer and 0.6 for ovarian cancer; the estimates and the trends in risk were significant for all cancer sites, excluding rectal and breast cancer. The estimates for an increase in n-3 PUFAs of 1 g/week were 0.70 for oral and pharyngeal cancer, 0.71 for oesophageal, 0.88 for colon, 0.91 for rectal, 0.90 for breast and 0.85 for ovarian cancer. All the estimates were statistically significant, excluding that for rectal cancer, and consistent across strata of age and gender. These results suggest that n-3 PUFAs decrease the risk of several cancers.