Preliminary safety assessment without specific toxicology data

Members of the food industry may hesitate to develop a particular flavoring agent for which they know or suspect there is very little available toxicology information. However, a preliminary safety assessment may still be feasible based on the concept of analogous substances. Although this approach will not take the place of confirmatory animal testing, and is unlikely to be an acceptable method that will gain approvals from the FDA, it can be employed to predict the expected safety of food substances based on their structural similarity to other food substances with established safety profiles. How this concept can be applied is illustrated in the following examples.

Pyrazines constitute a broad family of chemical substances that result from heat-associated non-enzymatic browning of food (generally referred to as the Maillard reaction), which tends to impart earthy, nutty, roasted flavors to cooked foods. Terpenes constitute a wide-ranging family of flavoring agents that are naturally abundant in botanical species as diverse as fruits, flowers, nuts, and spices to the trees, seeds, stems and roots that yield them. There are countless different naturally-occurring pyrazines and terpenes, many of which can also be manufactured and used to flavor food. Developing a pyrazine substance that has no specific toxicology data may not be a roadblock to further development, though, because the base structural unit of these substances is the pyrazine molecule, which is a weak base. It is known that the gastrointestinal (GI) absorption of weak bases is optimal at pH’s that are typical of the intestine (i.e., pH 5–7), so pyrazines are expected to be easily absorbed across the GI epithelium, a phenomenon that also occurs with terpenes due to their lipid solubility. Thus, absorbed pyrazines and terpenes have ready accessibility to the blood where they are transported to the liver and other tissues for further metabolism.

Based on their chemical structure, both substance types are susceptible to biotransformation to substances that will either be excreted in the urine as oxidative metabolites, or further biotransformed by conjugation processes prior to urinary excretion. Therefore, if the expected daily intake of a pyrazine or terpene under development does not exceed the expected safe daily intake of analogous pyrazines or terpenes, it is unlikely that the lack of a complete toxicology data package for a particular pyrazine or terpene would inhibit further confirmatory safety testing that would be expected to support a finding of it being a safe food ingredient. Thus, even though the full-spectrum of desired toxicity of a specific food substance may not be known, its safety may be predicted simply by knowing the chemical structure, assuming that it has a structure similar to a food substance (or substances) known to have acceptable safe daily intake.

In summary, the concept of safety assessment by analogous substances may constitute a valid avenue of preliminary safety investigation for new food substances. However, it is obligatory to understand that the degrees of uncertainty brought by such predictions effectively limit their applicability to one time uses, and even then, are confined to substances that are expected to have intakes below a relatively small exposure such as one milligram per day or less.