Ames, pesticides, and cancer revisited

The case for continuing use of existing levels of pesticides in agriculture, espoused by Bruce Ames, is refuted. Ames' contentions that naturally occurring carcinogens are far more widespread than man-made ones, that pesticides prevent cancer by providing fruits and vegetables at lower costs to the poor, and that animal data on high risks with high doses cannot predict low risks from low doses in humans do not address key issues: 1) fruits and vegetables contain mixtures of carcinogens and anti-carcinogens, and selection effects from human exposures to these mixtures go back more than' a million years; 2) exposures from bioconcentrations of biopersistent organochlorines in the food chain create particular risks for meat-eaters, who have higher cancer risks than vegetarians; 3) even low doses from ingestion of produce containing pesticide residues can cause tissue injury, which could itself promote cancer; 4) epidemiologic data show rises in cancer incidences in older people in many countries, major differences in cancer risks between countries, and converging trends in risks for populations migrating to certain countries; 5) studies of pesticide-exposed workers consistently show increased rates of cancers and birth defects and cancers in their offspring; 6) epidemiologic studies based on large databases tend to underestimate risks from environmental causes because of exposure misclassification; 7) exposures to many organochlorines may have pervasive effects on endocrine function; 8) crop yields can be increased with less use of pesticides. Studies demonstrating the latter need replication, and should be supported as part of a coherent government agenda to develop alternative farming methods.