Risk evaluation of food chemicals

At present the approach to assessing the potential risks of exposure to a chemical in the diet involves the application of a standardised risk assessment approach consisting of three main elements - hazard identification and characterisation and exposure assessment. 

Assessment of whether a chemical has the potential to cause adverse effects in humans arises usually from direct observation of an effect in animals or humans, such as the acute poisoning episodes that have occurred when potatoes contain high levels of glycoalkaloids. Epidemiological studies have also been used to infer a possible relationship between intake of a particular type of food, or constituent of that food, and the potential to cause an adverse effect. Such observations led to the characterisation of the aflatoxins as human carcinogens. However, natural toxic substances that occur in plant foods have often been identified through observations in animals, particularly farm animals. It was observations of adverse effects in farm animals that led to the further characterisation of the phytoestrogens and the mycotoxins. In other instances, the concern arises from the chemical similarity to other known toxins. 

For chemicals in general the identification of a potential hazard normally arises from the application of in vitro tests or from short-term toxicity studies undertaken in laboratory animals (up to a period of 90 days in the case of the rat where the test material normally should not exceed 1% of the total diet). This usually enables a critical effect to be assessed. 

The characterisation of the toxicity associated with a specific chemical is invariably dependent on feeding the chemical to laboratory animals in statistically-based lifetime bioassay, guidelines for which have been agreed at the international level by the OECD (OECD, 1981) and the IPCS (IPCS, 1999), and which are undertaken strictly in compliance with the principles of good laboratory practice (GLP). The animals used in long-term bioassays are 

A typical dose-response is shown in Fig. 11.1. This assumes that a dose exists which has no effects due to the capacity of the body to reverse minor changes and maintain cellular homeostasis. The threshold dose is normally taken to be the observed experimental NOAEL, but the NOAEL could be lower than the threshold dose. The NOAEL chosen is the one that represents the most sensitive species studied since all international protocols require that the chemical is tested on at least two species (frequently the rat and the mouse). 

---