Fish consumption and cancer risk

Abstract
Background: Although several studies have investigated the relation between fish consumption and the risk of cardiovascular diseases, less attention has been paid to the relation between fish consumption and cancer risk. Objective: The relation between frequency of consumption of fish and risk of selected neoplasms was analyzed by using data from an integrated series of case-control studies conducted in northern Italy between 1983 and 1996. Design: The overall data set included the following incident, histologically confirmed neoplasms: oral cavity and pharynx (n = 181), esophagus (n = 316), stomach (n = 745), colon (n = 828), rectum (n = 498), liver (n = 428), gallbladder (n = 60), pancreas (n = 362), larynx (n = 242), breast (n = 3412), endometrium (n = 750), ovary (n = 971), prostate (n = 127), bladder (n = 431), kidney (n = 190), thyroid (n = 208), Hodgkin disease (n = 80), non-Hodgkin lymphomas (n = 200), and multiple myelomas (n = 120). Control subjects were 7990 patients admitted for acute, nonneoplastic conditions unrelated to long-term modifications of diet. Odds ratios (ORs) were computed for subsequent levels of fish consumption compared with no or occasional consumption (Results: There was a consistent pattern of protection against the risk of digestive tract cancers with fish consumption: oral cavity and pharynx, OR = 0.5 for the highest compared with the lowest level of consumption; esophagus, OR = 0.6; stomach, OR = 0.7; colon, OR = 0.6; rectum, OR = 0.5; and pancreas, OR = 0.7. There were inverse trends in risk of larynx (OR = 0.7), endometrial (OR = 0.8), and ovarian (OR = 0.7) cancers and multiple myeloma (OR = 0.5). No pattern of cancer risk in relation to fish consumption was observed for cancers of the liver, gallbladder, breast, bladder, kidney, or thyroid or for lymphomas. Conclusion: This study suggests that the consumption of even relatively small amounts of fish is a favorable indicator of the risk of several cancers, especially of the digestive tract.