Risk Assessment

1983

when was risk assessment defined as specific activity by National

Research council

banned carcinogens from food supply

in 1958, what did the Delaney clause do

hazard x exposure

risk =

cancer

the inital focus of risk assessment was on what

no, now it includes focus across lifestages and for ecosystems

does the focus of risk assessment continue to be on cancer?

extrapolations

risk assessment and management require ______ that go far beyond observation of actual effects

• each problem must be placed into environmental and/or public health context

• must engage all “stakeholders” at all stages

According to the National Research Council (NRC, 1994), what makes risk assessment and management effective?

• protect human and ecological health (toxic substances)

• balance risks and benefits (drugs, pesticides)

• set target levels of risk (food contaminants and water pollutants)

• set priortieis for program activities (regulatory agencies, manufacturers, environmental/consumer organizations)

• estimate residual risks and extent of risk reduction after steps are taken to reduce risks

what are the main objectives of risk assessment

risk assessment

Systematic scientific evaluation of potential adverse health effects resulting from human exposures to hazardous agents or situations

• requires integration of qualitative (overall evidence) and quantitative scientific (exposure data) data

• carries uncertainty and variability

risk

probability of an adverse outcome based on exposure and potency of the hazardous agent

hazard

Potential source of harm

• Inherent toxic properties

• Under typical exposure scenarios

exposure

dose and route subjected to organism or environment

risk characterization

combination of qualitative and quantitative data

• Risk estimate

• Consideration of uncertainties, mode of action, type of affect across species or context

risk management

Process by which policy decisions are chosen to control hazards

• Considers scientific evidence

• Statutory, engineering, economic, social, and political factors

risk communication

Process of making risk assessment and management information comprehensible to the public

• Perception vs. intent

• ‘Outrage factor’

Decision making is governed by diverse statutes, some of which rely on risk alone while others balance risk vs. benefit, leading to different regulatory outcomes.

Why does risk-based decision making vary across environmental and public health issues?

“risk alone” decisions are based solely on the magnitude of health “risk vs. benefit” decisions weigh health risks against economic, medical, or societal benefits.

What is the difference between “risk alone” and “risk vs. benefit” decision frameworks?

effective decision making requires collaboration across disciplines, including toxicology, epidemiology, medicine, economics, and policy, because risk decisions are not purely scientific.

Why is interdisciplinary work important in risk management decision making?

they provide direct evidence that exposure has occurred, linking environmental exposure to biological effects

Why are biomarkers of exposure valuable in decision making?

detemining whether an agent is capable of causing toxic effects, focusing on the presence and type of toxicity rather than the magnitude of risk.

What is the goal of hazard identification in risk assessment?

predicts potential toxicity based on chemical structure similarities, allowing hazard identification when experimental data are limited.

How are structure–activity relationships (SAR) used in hazard identification?

use cells or tissues to identify toxic mechanisms and screen for potential hazards, but lack whole-organism context.

What role do in vitro studies play in hazard identification?

assess toxicity in whole organisms, revealing systemic effects, target organs, and dose–response relationships.

Why are in vivo studies important for hazard identification?

provides direct evidence of toxicity in exposed populations, though results may be limited by confounding and exposure uncertainty.

How does human epidemiology contribute to hazard identification?

they can help guide early decisions—whether to continue chemical development, submit a Pre-Manufacturing Notice (PMN), or require additional testing—saving time, money, and animals before expensive long-term rodent studies ($2–4 million, 3–5 years)

Why are SAR predictions used before full rodent carcinogenicity studies?

• Structure

• Solubility

• Stability

• pH sensitivity

• Electrophilicity

• Volatility

• Chemical reactivity

what are the key factors of SAR

Key molecular structures of chemicals guide regulators on when to assess hazard potential.

What provides regulators with most information for assessing hazard potential in SAR?

aromatic amine

8 of 1st 14 regulated occupational carcinogens belonged to what class

PMNs

the EPA uses SARs to respond to

N-nitroso or aromatic amines, amino azo dyes, and phenanthrene nuclei.

What structural alerts trigger further cancer testing?

A few chemical classes, including valproic acid, retinoic acid, phthalate esters, and glycol esters.

What structural alerts trigger further noncancer testing?

through Toxicity Equivalence Factors (TEFs) for closely related compounds; for example, TCDD (dioxin) uses Ah receptor mechanism and estimated toxicity = sum of product concentrations × TEF values.

How is SAR used to assess complex mixtures?

improve drug design efficiency but are more effective for designing drugs than predicting toxicity.

How do combinatorial chemistry and 3-D molecular modeling relate to SAR?

Ames, which assesses reverse mutations in bacteria to detect potential mutagenic chemicals.

What is the primary in vitro assay used to detect mutagenicity, and what does it assess?

Skin painting in mice (local carcinogenicity) and altered liver foci in rats (early liver cancer markers); primarily used for mechanistic evaluation.

what are examples of in vitro studies

Validation of new assays is difficult; results are mainly mechanistic. Nonmutagenicity results can support nonlinear cancer risk assessment paradigms for regulatory decisions.

What are the limitations and regulatory uses of in vitro and short-term studies?

Chemicals that cause tumors in animals may also cause tumors in humans; all known human carcinogens produce positive results in at least one animal model, but not all chemicals positive in animals are carcinogenic to humans.

What is the basic premise of animal bioassays in hazard identification?

follow a similar rationale: testing for adverse effects in sensitive life stages, with the premise that effects in animals may indicate human hazard.

How do developmental and reproductive assays relate to standard animal bioassays?

Two species, both sexes, ~50 animals per dose group; key choices include strain selection and dose levels (90%, 50%, 10–25% of maximally tolerated dose).

What is the standard design for cancer bioassays?

Increased tumor incidence at specific organs, induction of rare tumors, earlier induction of common tumors, and increases in total tumor numbers.

What are indicators of positive results in animal cancer bioassays?

If only a few high doses are tested or systemic toxicity occurs, it is hard to predict effects at low, human-relevant exposures.

Why is low-dose extrapolation difficult in animal bioassays?

• Rat vs. mouse: ~70% concordance

• Humans vs. rodents: likely <70%

• Male vs. female: ~65%

• Sensitivity varies; e.g., rats are ~100,000× more sensitive to aflatoxin B1 than mice.

What is known about concordance and sensitivity in animal bioassays?

epidemiological data

what is the most convincing line of evidence for human risk

it’s well conducted and there’s positive association between exposure and disease

why is the most convincing line of evidence for human risk epidemiological data

• Robustness of exposure estimates

• Many health effects have long latencies

• Multiple exposures

• Trade off between detailed information on a few and limited

information on a large number of individuals

what are some limitations of epidemiological data specifically for human risk

cross-sectional, cohort, and case-control

what are the major types of epidemiological data

cross-sectional

study type: surveys groups, identifies risk factors (exposures associated with disease, not useful for cause and effect

cohort

study type: evaluates individuals based on exposure, monitored for development of disease

case-control

study type: selected on basis of disease status, exposure history analysis

• Strength of association

• Consistency of observations

• Specificity

• Temporal relationship – exposure precedes disease

• Biological plausibility

how do we judge epidemiological findings

molecular epidemiology

integrates molecular biology into traditional epidemiology

• Biomarkers of exposure, susceptibility

• Genetic factors

• Link molecular events in the causal disease pathway

threshold

dose-response assessment: noncancer endpoints and nongenotoxic cancer endpoints; assumes a safe exposure exists below the threshold.

nonthreshold

dose-response assessment: genotoxic cancer endpoints; assumes any exposure carries some risk.

Human exposure data are limited; extrapolation of high doses in rodents to low doses in humans; choose most robust data set that shows effects at the lowest level

Why are animal bioassays predominantly used for dose-response assessment?

• NOAEL: No observed adverse effect level

• LOAEL: Lowest observed adverse effect level

• POD: Point of departure, estimated near the lower end of observed effects; exposures below POD require extrapolation

What key terms are used in threshold dose-response assessment?

Reference doses (RfD)

threshold approach: Daily oral exposure that is likely to be without appreciable risk

Acceptable daily intake (ADI)

threshold approach: very similar to RfD (used by EPA), some

differences in safety factors from agency to agency

NOAEL/UF x MF

RfD =

• Interspecies: 10× (animal to human)

• Intraspecies: 10× (human variability)

• Additional: 10× for children under Food Quality Protection Act

What are typical uncertainty factors in dose-response assessment?

Short-term to chronic, low animal numbers, or only LOAEL available

what are some examples of other UFs

modifying factors

• Used to adjust UFs

• Data on mechanism, PK, relevance of animal studies to human

Example, if the same metabolic pathway is known to occur in humans and rodents, UF may be reduced to 3

Using chemical-specific, data-driven adjustments for interspecies (2.5–4×) and intraspecies (3.16×) pharmacokinetic and toxicokinetic factors.

How can uncertainty in dose-response assessment be reduced?

• NOAEL must be an experimental point

• Ignores dose response curve, using only NOAEL

what are some limits of the NOEFL approach

• Dose-response curve statistically modeled (95% confidence limit)

• BMD set - 10%, typical

• RfD (high) = BMD10/UF x MF

• RfD (low) = BMDL10/UF x MF

what are some limits of NOEFL’s alternative approach, BMD (benchmark dose)

Uses dose response curve, expresses variability, consistent benchmark response across studies

what are some advantages of BMD

Margin of exposure (MOE)

• No uncertainty factors

• Not used to establish safe daily dose

<100

an MOE of what triggers further testing

NOAEL/Probable human exposure level

MOE =

• Need for extrapolation to low risk levels (10⁻⁴ to 10⁻⁶)

• EPA guidelines: Define POD/LOAEL → extrapolate to low environmentally relevant exposures → linear or nonlinear model depending on data.

How are nonthreshold dose-response assessments conducted for genotoxic cancer endpoints?

1. Probability distribution: Models individual tolerance levels using cumulative dose-response data.

2. Mechanistic: Models toxic effect as the random occurrence of one or more biological events; mathematical equation consistent with biological mechanisms.

What are the two main extrapolation methods in dose-response assessment?

• Source

• Type

• Magnitude

• Duration

primary objectives of exposure assessment are to determine what

• Who is exposed?

• Route of exposure

• Quantification of each pathway

• Plausible estimate of the upper 90-95%

what are the key variables of exposure assessment

it’s a conservative point estimate representing a high-end exposure value, meaning there is 95% confidence that actual exposure is below this level, used to protect sensitive populations in risk assessment.

In exposure assessment, what does using the “upper 95% confidence interval” mean?

• Summary of risk assessment components

• Outline key findings

• Informs risk manager in public health decisions

what are the goals of risk characterization

testing approaches that primarily use invertebrates or cell-based systems, allowing a higher percentage of chemicals to be tested with lower regulatory and ethical concerns.

What are New Approach Methodologies (NAMs) in toxicology?

• Zebrafish embryos

• Nematodes

• Inducible pluripotent stem cells (iPSCs)

What are some examples of New Approach Methodologies?

“More likely than not” (≥50% probability); in special cases, “as likely as not.”

What standard of proof is used in civil trials to establish causation from exposure?

Courts consider whether the adverse outcome did happen or could have happened, ensuring the science is credible and consistent with known mechanisms.

How is scientific plausibility evaluated in civil trials?

Experts provide reports, depositions, and testimony, translating technical findings for a lay audience (jury) while distinguishing between technical and lay explanations.

What is the role of expert testimony in civil trials?

• Greater reliance on biological data

• Genetic influences

• Lessened use of 10x factors

• Harmonization

what does the future of risk assessment include