Risk characterization generally requires integration of the data and analysis of the first three elements of risk assessment to determine whether and to what extent humans will experience any of the various types of toxicity associated with exposure to a substance (Table 21.7).
Risk characterization deals with the question: what is the estimated incidence and severity of the adverse effect? It is in characterizing the risks that the major assumptions, scientific judgments, and uncertainties should be identified so that the risk estimate can be better understood. In Europe, the term risk estimation is used to characterize risk. Many uncertainties exist, and several approaches have been suggested to improve the characterization and understanding of risks. The US Environmental Protection Agency (US EPA) has considered guidelines for risk characterization. Such guidelines could include, for example, the use of sensitivity analysis on data sets employed in risk extrapolation, expressing variability in risks from a given extrapolation model, statistical levels used to project risks (e.g., median, 95th percentile, range), and ways to evaluate risks quantitatively when the qualitative weight of evidence is low.
126.96.36.199 Limitations and assumptions in risk assessment Risk assessment is a process that provides a framework for evaluating information and presenting that information in a form useful to decision makers. Risk assessment, however, is limited by:
Table 21.7 Risk characterization
Combine and integrate exposure and dose-response assessment. Estimate quantitatively some measure of risk.
Identify major assumptions, scientific judgments, and estimates of uncertainties.
- lack of data on substances and adverse health effects;
- uncertainty about the cause of disease;
- uncertainty in extrapolating human risk from animal data.
These limitations have resulted in applying a set of assumptions, or default positions. The assumptions and uncertainties that abound in the risk assessment process have generated much controversy. When there is uncertainty or a lack of data, public health officials tend to use assumptions that will not underestimate risk. Nine of the most generally agreed-upon assumptions in risk assessment have been emphasized, although many more have been identified:
1. In the absence of adequate human data, adverse effects in experimental animals are regarded as indicative of adverse effects in humans.
2. Dose-response models can be extrapolated outside the range of experimental observations to yield estimates or estimated upper bounds on low-dose risk.
3. Observed experimental results can be extrapolated from one species to the other.
4. No threshold doses (i.e., doses below which no adverse effects will occur) exist for carcinogenesis, although threshold levels may apply for other toxicological outcomes.
5. Average doses give a reasonable measure of exposure when dose rates are not constant over time.
6. In the absence of toxicokinetic data, the effective or target dose is assumed to be proportional to the administered dose.
7. The risks from multiple exposure and multiple sources of exposure to the same chemical are usually assumed to be additive.
8. Regardless of the route of exposure (dermal, oral, or inhalatory), 100% absorption across species is assumed in the absence of specific evidence to the contrary.
9. Results associated with a specific route of exposure are potentially relevant for other routes of exposure.
The risks from a substance cannot be ascertained with any degree of confidence unless dose-response relationships are quantified. In the US, the regulatory distinction between substances that cause cancer and those that do not has a major impact on the extrapolation methods used to characterize the dose-response curve in the non-observable low-dose range. All carcinogens, whether characterized as genotoxic or non-genotoxic, are considered by US regulatory agencies to pose a risk, no matter how finite, at all doses, while for non-carcinogens a threshold dose is assumed. As will be discussed in the following sections, this distinction results in a different characterization for these two classes of substances. Most European regulatory agencies, by contrast, distinguish between carcinogens characterized as genotoxic and non-genotoxic. For genotoxic carcinogens, it is assumed that there is no threshold. For non-genotoxic carcinogens, the existence of a threshold is assumed, provided the mechanism of carcinogenesis is understood. JECFA has indicated that carcinogens vary in the degree of risk they represent, and the intentional use of a food additive known to be a carcinogen should be considered only under very restricted circumstances.
For noncarcinogens, a threshold dose or level of exposure is assumed below which no effect is observed (Table 21.8). The dose-response evaluation requires estimation of the threshold dose and determination of the no-observed-adverse-effect level (NOAEL) from observations in experimental animals or exposed people. The acceptable daily intake (ADI) (also called the tolerable daily intake, or TDI, see also Chapters 16 and 17) is estimated by dividing the NOAEL by a safety or uncertainty factor. Scientific guidelines and recommendations on the development and use of ADIs have been adopted by the Joint FAO/WHO Food Standards Program (Codex Alimentarius Committee on Food Additives), the FAO Committee on Pesticide Residues, and the WHO Expert Committee on Pesticide Residues. If the maximum daily intake of a non-carcinogenic substance is estimated to be lower than the ADI, then no risk is assumed for almost all members of the general population. Critical to this estimate, however, is the magnitude of the safety or uncertainty factor, which can range from 10 to 10,000 based on the data and on the policy of different regulatory organizations. For example, for non-nutrient substances, the Center for Food Safety and Applied Nutrition at the US Food and Drug Administration (US FDA) uses safety factors of between 100 and 2000, depending on the availability and type of data for analysis. The safety factor accounts for uncertainties concerning interspecies and intraspecies variation.
Where the WHO uses tolerable daily intake instead of accepted daily intake, the US EPA uses reference dose (RfD) to avoid the value judgment implicit in the calculation of an acceptable dose. The no-observed-adverse-effect level (NOAEL) and/or the lowest-observed-adverse-effect level (LOAEL) (Lowest found concentration or amount of a substance, which causes an adverse effect) are determined for each study and type of effect. To determine the RfD, uncertainty factors are applied to the NOAEL (or LOAEL if a NOAEL has not been established).
188.8.131.52 Characterization of cancer risks From a scientific standpoint, substantial progress has been and is being made in understanding the mechanisms of toxicity, including carcinogenesis, and the causal relationships on which safety assessments are based. It is recognized to an increasing extent that "carcinogen" is difficult to define and that distinctions can be made among carcinogens based on the differing underlying mechanisms. Some substances initiate cancer directly and others are only involved secondarily in carcinogenesis. Thus for some carcinogens, as for non-carcinogens, there may be levels of exposure for which the possibility of harm to humans can be ruled out with reasonable certainty, a threshold dose determined, and for which instead a safety-factor or uncertainty-factor evaluation may be appropriate. A scheme for determining how chemical carcinogens could be identified is presented in Figure 21.2.
In the US, however, under Section 409 of the Food, Drug and Cosmetic Act, the Delaney Clause prohibits the use of food additives found to induce cancer in animals or humans.
Table 21.8 Characterization of risks
For food additives, the no-observed-adverse-effect level (NOAEL) is divided by a safety or uncertainty factor to estimate an acceptable daily intake (ADI).
For systemic toxicants, US EPA developed the reference dose (RfD) approach, where the NOAEL is divided by an uncertainty factor and a modifying factor. Generally, the RfD is an estimate of a daily exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of harmful effects during a lifetime.
Risk is estimated from the cumulative dose and/or the dose-response curve extrapolation.
Mathematical models are used to extrapolate to low-dose response.
A range of risks might be produced using different models and assumptions about dose-response curves, relative sensitivities of humans and animals, and for different estimates of human doses.
reaction with DNA?
yes genotoxic no non - genotoxic
1 threshold unlikely
2 dose-response may be affected; cell proliferation (usually toxicity-related at high doses)
reaction with cell receptor?
1 threshold questionable
2 usually effective at low doses no proliferative
1 threshold likely
2 usually related to toxicity and regeneration
Figure 21.2 Proposed scheme for classification. Source: Cohen and Ellwein, 1990. With permission.
In contrast to the general safety standard for non-carcinogens, which recognizes the impossibility of assessing the complete absence of risk, the Delaney Clause has been interpreted as taking a zero risk approach to substances implicated as carcinogens. It should be stressed that this clause was enacted during a period when relatively few carcinogens had been identified and even fewer were believed to be present or associated with food. The result in the US was the assumption that there was no threshold dose for carcinogens and that oncogenic risk was a function of the cumulative lifetime dose (Table 21.8).
The non-nutritive sweetener saccharin has been shown to induce bladder cancer in rats. It is not metabolically activated when ingested by humans or animals, does not react with DNA, and is not mutagenic in short-term tests. Therefore, it is considered non-genotoxic. The lowest dose for an effect in rats is 2.5% sodium saccharin in the diet, while there is no effect with acid saccharin at 7.5% in the diet. The NOAEL in rats is 1.0% in the diet; there is no effect in the animals if the urine is acidified. At higher doses, increased cell proliferation of the adult rat bladder epithelium is observed. Urinary silicate precipitates and/or microcrystals are critical phenomena in the development of the lesions in rats. There is no evidence of any interaction of saccharin with cell receptors; relatively high doses are required for the effect in the bladder. Thus, as suggested in Figure 21.2, the proliferative response is probably related to toxicity and cellular regeneration, and a threshold dose is likely for this effect.
184.108.40.206 Characterization of risk using mathematical modeling As most of the information on whether a substance is capable of inducing cancer is obtained from animal studies at high doses, statisticians developed mathematical models to extrapolate from these high-dose level studies to determine the risk at the low doses to which humans would be potentially exposed. This modeling process is used for quantitative risk assessment of chemical carcinogens and involves eight steps (Table 21.9). It has been termed an empirical risk assessment model or default carcinogen risk assessment methodology. Starting with carcinogenicity in the rodent bioassay, the procedures and calculations are outlined to reach the exposure analysis and risk-benefit analysis needed to determine exposure levels and cancer risks that society can tolerate.
Table 21.9 Empirical risk assessment model or default carcinogenic risk assessment methodology
1. Positive response in rodent bioassay
2. Appropriate dose measure; typically mg/kg body weight per day
3. Dose-response function selected for risk to rodents; typically linearized multi-stage model
4. Estimate of the variability of the dose-response function; typically 95% confidence interval
5. Linearized upper 95% bound on risk in the 1 in 10-6 region selected to determine quantitative value for risk assessment for rodents
6. Interspecies extrapolation to estimate risk for humans in dose-region of interest
7. Extrapolation from one exposure route to the other
8. Exposure analysis and risk benefit analysis to determine exposure levels and risks society can tolerate
The modeling and extrapolation processes employed in quantitative risk assessment are considered by many to be the most important sources of uncertainty in the risk assessment process. A quantitative estimate of the risk from a substance at a particular low-dose level is highly dependent on the mathematical form of the presumed dose-response relationship. Differences between models of at least three to five orders of magnitude are not uncommon. One difficulty with low-dose extrapolation is that some methods fit the data from animal experiments reasonably well, and it is impossible to distinguish their validity on the basis of a good fit. From a mathematical point of view, distinguishing between the models on the basis of their fit with experimental data would require an extremely large experiment which, from a practical point of view, is probably impossible. The different approaches used by the various regulatory agencies for assessing risk are, for example, reflected in the acceptable exposure levels set for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (Table 21.10). Not all agencies assume that no threshold exists for carcinogenesis. Although TCDD has proved to be extremely toxic to some rodents, its carcinogenic potential to humans has been the subject of considerable scientific controversy. TCDD has also been shown to induce a wide spectrum of adverse effects, not only carcinogenicity, in experimental animals. Use of the NOAEL and a general safety standard as well as a cancer dose-modeling approach yields a 2,000-fold range of allowable exposure levels by various regulatory agencies (Table 21.10).
Was this article helpful?
Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.