Threshold dose, safety assessment

Although dose-response assessments for deterministic and stochastic effects are discussed separately in this Report, it should be appreciated that many of the concepts discussed in Section 3.2.1.2 for substances that cause deterministic effects apply to substances that cause stochastic effects as well. The processes of hazard identification, including identification of the critical response, and development of data on dose-response based on studies in humans or animals are common to both types of substances. Based on the dose-response data, a NOAEL or a LOAEL can be established based on the limited ability of any study to detect statistically significant increases in responses in exposed populations compared with controls, even though the dose-response relationship is assumed not to have a threshold. Because of the assumed form of the dose-response relationship, however, NOAEL or LOAEL is not normally used as a point of departure to establish safe levels of exposure to substances causing stochastic effects. This is in contrast to the common practice for substances causing deterministic effects of establishing safe levels of exposure, such as RfDs, based on NOAEL or LOAEL (or the benchmark dose) and the use of safety and uncertainty factors. 

In many respects, the foundations and framework of the proposed risk-based hazardous waste classification system and the recommended approaches to implementation are intended to be neutral in regard to the degree of conservatism in protecting public health. With respect to calculations of risk or dose in the numerator of the risk index, important examples include (1) the recommendation that best estimates (MLEs) of probability coefficients for stochastic responses should be used for all substances that cause stochastic responses in classifying waste, rather than upper bounds (UCLs) as normally used in risk assessments for chemicals that induce stochastic effects, and (2) the recommended approach to estimating threshold doses of substances that induce deterministic effects in humans based on lower confidence limits of benchmark doses obtained from studies in humans or animals. Similarly, NCRP believes that the allowable (negligible or acceptable) risks or doses in the denominator of the risk index should be consistent with values used in health protection of the public in other routine exposure situations. NCRP does not believe that the allowable risks or doses assumed for purposes of waste classification should include margins of safety that are not applied in other situations. 

The effects of genotoxic compounds are considered non-threshold. Thus, risk assessment for a given exposure is usually performed by a linear or sub-linear extrapolation from the high dose effects observed in animals to the lower human exposure. Since the outcome of the extrapolation depends on the model applied and extrapolation over different orders of magnitude is error prone, the European Food and Safety Authority (EFSA 2005) recommended to avoid this extrapolation and proposed the MOE approach. This approach uses the benchmark dose, or the T25 calculated from a carcinogenicity study and compares this with human exposure. A MOE of 10,000 and more is considered to be of minor concern. The advantage is that neither a debatable extrapolation from high to low doses needs to be performed nor are hypothetical cancer cases calculated. For details of the different approaches see, SCHER, SCCP, SCENIHR (2008). 

For nongenotoxic chemicals, risk assessment is based on the concept of threshold doses, below which no adverse effect results from exposure. From human or experimental animal data, one tries to establish the no observable adverse effect level (NOAEL) and the lowest observed adverse effect level (LOAEL). In order to establish safe levels of exposure to potentially toxic agents, the NOAEL is divided by a safety factor (often named uncertainty factor). When the risk assessment is based on data from experimental animals, a default safety factor of 100 is usually applied. The safety factor constitutes a factor of 10 for potential differences in susceptibility between animals and man, and another factor of 10 for interindividual differences among humans. The factors are combinations of differences in toxicokinetics and toxicodynamics, both in animals and man. If true factors are known, the size of the safety factor may be changed accordingly. When risk assessment is based on human data, a safety factor of 10 is applied in most cases, for instance, for food additives. However, for natural toxins in food, smaller factors are usually applied. This is a risk management decision, often based on information on the absence of adverse health effects at intake levels close to the estimated LOAELs. 

In classifying waste, deterministic responses generally should be of concern only for hazardous chemicals (see Section 3.2.2.1). Therefore, the only important issue for risk assessment is the most appropriate approach to estimating thresholds for induction of responses in humans. The primary concern here is that consistent approaches should be used for all substances that induce deterministic effects. NCRP s recommendation that nominal thresholds in humans should be estimated using the benchmark dose method and a safety factor of 10 or 100, depending on whether the data were obtained in a study in humans or animals (see Section 6.1.2.1), is intended to provide consistency in estimating thresholds for all substances that cause deterministic effects. 

A knowledge of the relation between the dose of a chemical and the response it produces enables its safety to be assessed and predictions to be made. By knowing the relationship and the threshold it is possible to estimate a safe dose of a chemical (see Chapter 12). 

Knowledge of the relationship between dose and response (effect), and the threshold for this, is crucial in defining the risk of exposure to a chemical. Safety evaluation is a legal requirement for drugs, food additives, and contaminants in food, and a risk assessment has to be carried out in order to set the limits of exposure. The relationship between the dose and the response (effect) can be established and plotted as a graph. This is called a dose-response curve (see Figure 29 and box), which often shows that there is a dose(s) of the chemical that has no effect and another, higher dose(s) which has the maximum effect. It is a visual representation of the Paracelsus principle that, at some dose, all chemicals are toxic. The corollary to this is that there is a dose(s) at which there is no effect. 

The dose-response relationship therefore allows the toxicologist to establish a threshold, the dose at which there is no adverse effect, which is vital for the proper assessment of risL Information from the dose-response relationship is used to determine the therapeutic index and the margin of safety for drugs, which indicate how safe the drug is. The greater the value the greater the difference between the dose at which there are adverse effects and the therapeutic dose. This is used as part of the risk assessment process. 

For chemicals such as food additives, food contaminants, and industrial chemicals the threshold, that is the dose at which toxic effects become apparent, is determined from the dose-response graph and used in the risk assessment process. The threshold value is used, together with safety factors, to determine the acceptable daily intake (ADI) of a food additive, or the tolerable daily intake (TDI) of a food contaminant, or the threshold limit value (TLV in the USA, or maximum exposure limit (MEL) in the UK), for an industrial chemical (see box for calculation). For a drug, information about the dose in animals below which there are no adverse effects will be necessary before human volunteers can be exposed in clinical trials. More extensive safety evaluation is carried out for drugs than for... 

If, in the safety evaluation of a new food additive, tumours are found in adult animals given reasonable doses, especially if this occurs in more than one species of animal and if the underl5dng mechanism is relevant to humans, this would be incorporated into the risk assessment. The chemical would almost certainly not be licensed as a food additive. If the adverse effect caused by a food additive is of a different type and is believed to show a threshold, and provided a NOAEL can be determined from the available data, an ADI can be set, as described above. If a NOAEL cannot be determined, then a larger safety factor will be used. The risks have to be considered in relation to the benefits. 

Two examples of alternative approaches to cancer risk assessment would be estimations based on threshold-response (EPA, 2005a) and benchmark dose modeling (EPA, 1995, 2000). As a practical matter, if the proposed basis of safety relies on a threshold or mode-of-action characterization to dismiss or mitigate animal tumor data, PDA would reconunend that the safety narrative clearly discuss the scientific rationale and present all relevant data for consideration. In the absence of adequate evidence to the contrary, PDA presumes that certain assumptions are appropriately protective of safety, namely that (i) the induction of tumors in animals is relevant to human... 

The existing methods available for scientifically defensible risk characterization are not yet ideal since each step has an associated uncertainty resulting from data limitation and incomplete knowledge on exact mechanism of action of the toxic chemical on the human body. For noncancer end points, safety factors or uncertainty factors are applied since these effects are assumed to have a threshold below which no adverse effect is expected to be observed. US EPA has used the concept of a reference concentration (RfC) to estimate acceptable daily human exposure from HAPs. The RfC was adapted for inhalation studies based on a reference dose (RfD) method previously used for oral exposure assessment. The derivation of the RfC differs from that for the RfD in the use of dosimetric adjustment to extrapolate the exposure concentration for animals to a human equivalent concentration. Both are estimates, with uncertainty spaiming perhaps an order of magnitude, of a daily exposure to the human population, including sensitive subgroups, which would be without appreciable risk of deleterious effects over a lifetime. 

Now I hope you see why I believe that it is criminal to sound an alarm when a pesticide residue is detected on produce at 100 to 1000 times below the tolerance. The tolerance is related to a safety threshold, not a toxic one. This misuse of data totally ignores the entire risk assessment process that is appropriately conservative and protective of human health. What is even more ironic is that the very chemicals that the public is made to fear in food at these below-tolerance levels are often used by the same individuals in household and garden pesticide products at much higher doses. These are generally proved to be safe for most individuals, making worrying about food concentrations millions of times lower a totally worthless experience that detracts us from more serious concerns. 

Preclinical studies must be performed before an ingredient can be considered for clinical studies in humans in order to determine the potential toxicity of the ingredient and its metabolites and their effects in the matrix. Preclinical studies allow researchers to expose cell cultures and experimental animals to doses of ingredients not normally encountered in human consumption. Results of such assessments are used to determine the threshold of toxicity for the given ingredient (i.e., the margin of safety). 

---