Using epidemiology to regulate food additives: saccharin case-control studies.

The increasing use of nonnutritive sweeteners and the widely publicized 1969 ban on cyclamate led to additional investigations in rodents of the carcinogenic potential of saccharin. Preliminary results of a long-term feeding study indicated formation of bladder tumors in rodents, and collective experimental evidence has demonstrated that high doses of the synthetic sweetener saccharin can cause bladder cancer in rodents. Based on the results of that and other rodent studies indicating an increased risk of bladder cancer associated with saccharin, the Commissioner of the Food and Drug Administration announced the agency's intention to propose a ban on saccharin. This intention was made known in April 1977 under the Delaney Clause of the Food, Drug, and Cosmetic Act. The clause essentially states that no additive shall be deemed safe if it is found to induce cancer in man or animals, or if it is found, after tests appropriate for the evaluation of the safety of food additives, to induce cancer in man or animals. Also in 1977, a group of epidemiologists began to assess the available epidemiologic information to determine the potential human risk. This report describes the assessment of several human epidemiologic studies available then and the results of more recent epidemiologic studies.

Use of artificial sweeteners by cancer patients.

Current daily use of artificial sweeteners (AS) and diet drinks was evaluated for 1,862 patients hospitalized for cancer and for 10,874 "control" patients hospitalized for other conditions believed not to be associated with use of these substances. The data were derived from an ongoing survey in seven countries. For cancer of most sites, the age-standardized proportion of users of AS was somewhat less than that for controls. A greater proportion of users among cancer patients than among controls was noted only for cancer of the stomach among women. Little information on urinary tract cancer was available; there were no users of AS among 13 patients with cancer of the bladder, 5 with cancer of the renal pelvis, and 2 with cancer of the ureter. There were 455 cancer patients known to have been interviewed during their initial hospitalization for the disease. Based on these cases, an age-sex-country-standardized estimate of cancer incidence for users of AS, relative to nonusers, was 1.0. Only a very small proportion of patients reported daily use of diet drinks, and the proportion of users did not differ substantially between cancer patients and controls. The present data provide virtually no support for an overall positive association of AS with cancer.

Food additives (natural and synthetic).

Food additives can be divided into the following categories: intentional, unintentional, contaminants, and those resulting from food processing procedures. Representative food additives from each category are discussed, with special attention being paid to the status of those suspected or proven to be toxic to humans. In addition, certain chemical components of food and methods for testing food additives are considered. Areas requiring additional testing include saccharin, cooking procedures, especially charcoal broiling, and hydrozines in mushrooms. The more recent developments in test procedures, including in vitro test methods, the transplacental exposure route, the use of maximal tolerated dose, and the initiation--promotion sequence, are evaluated.

Conditioned and latent inhibition in taste-aversion learning: clarifying the role of learned safety.

Experiments 1-3 investigated the applicability of the classical conditioning concept of conditioned inhibition to taste-aversion learning. Rats made ill after drinking saccharin and subsequently administered a "safe" exposure to saline (or casein hydrolysate) evidenced an enhanced preference for the safe fluid (relative to either a third, slightly aversive, solution or to water) when compared to controls in which saccharin was not previously poisoned. Such active condition inhibition was significantly reduced in Experiment 4 when two safe exposures to saline preceded saccharin-illness pairings. These results indicate that conditioned inhibition can be established in a taste-aversion procedure and that a latent inhibition manipulation reduces the ability of a taste to become a signal for safety. Implications of these findings for the learned safety theory of taste-aversion learning and the relevance to bait-shyness of principles established within the classical conditioning paradigm are considered.