Although the terms are often used interchangeably, hazard and risk have different meanings in toxicology.

Hazard refers to the intrinsic ability of a substance to cause adverse biological effects, independent of exposure conditions.

Risk describes the likelihood that those harmful effects will actually occur under specific exposure conditions. Risk depends on both the hazardous properties of a substance and the level of human exposure.

For example, a chemical may have hazardous properties, but if exposure is negligible or effectively controlled, the actual risk to human health may be very low.

For this reason, regulatory toxicology focuses on risk characterization, which considers both hazard and exposure when determining whether a substance can be used safely.

A toxicological risk assessment is required whenever the safety of a chemical substance, material, or product must be demonstrated for human exposure.

TRAs are routinely prepared to support:

• Regulatory submissions
• Product safety evaluations
• Market authorization applications
• Regulatory compliance
• Product development
• Material selection
• Manufacturing changes
• Investigation of potential safety concerns

They are widely used across industries including:

• Medical devices
• Pharmaceuticals
• Chemicals (including REACH registrations)
• Cosmetics and personal care products
• Food and feed
• Food-contact materials
• Consumer products

A toxicological risk assessment may also be necessary when introducing a new substance, modifying a formulation, changing manufacturing processes, or when existing safety data are insufficient to demonstrate safe use.

A toxicological risk assessment is relevant for any organization that develops, manufactures, imports, or markets products that may result in human exposure to chemical substances.

Typical users include:

• Medical device manufacturers
• Pharmaceutical companies
• Chemical manufacturers
• Cosmetic and personal care companies
• Food and feed producers
• Packaging manufacturers
• Consumer product manufacturers
• Regulatory affairs professionals
• Product safety and toxicology teams

The scope of the assessment depends on several factors, including the substance involved, the intended use, expected exposure, target population, and applicable regulatory requirements.

A well-prepared toxicological risk assessment helps organizations demonstrate product safety, support regulatory submissions, and reduce compliance risks.

The toxicity of a chemical substance is influenced by multiple interacting factors that determine the likelihood and severity of adverse effects. The toxicological profile of a substance depends not only on its intrinsic properties, but also on exposure conditions and biological factors.

Key factors influencing chemical toxicity include:

• Dose, duration, timing and frequency of exposure
• Route of exposure (e.g. oral, dermal, inhalation)
• Chemical and physical properties (e.g. molecular structure, particle size, surface characteristics, solubility, reactivity, stability, and metabolism)
• Bioavailability and systemic distribution
• Species differences and human variability

Understanding these factors is essential for interpreting toxicological data, evaluating hazard and exposure, and ensuring accurate toxicological risk characterization in regulatory assessments.

A toxicological risk assessment (TRA) relies on high-quality scientific data to evaluate whether human exposure to a substance is safe under its intended conditions of use. The amount and type of information required depend on the substance, exposure scenario, regulatory framework, and available evidence.

A comprehensive toxicological risk assessment typically includes:

• Toxicological data, including in vivo, in vitro, and in silico information
• Dose–response data and identification of relevant points of departure (e.g., NOAEL, LOAEL, or Benchmark Dose (BMD))
• Physicochemical properties that may influence absorption, distribution, metabolism, and elimination (ADME)
• Exposure assessment, including the route, frequency, duration, and estimated level of human exposure
• Existing regulatory evaluations and authoritative scientific opinions
• Peer-reviewed scientific literature
• Human data, including epidemiological studies or clinical evidence where available.

The strength of a toxicological risk assessment depends on the quality, relevance, and completeness of the available evidence. When data are limited, scientifically accepted approaches such as read-across, weight-of-evidence evaluations, and computational toxicology can often be used to support a robust safety assessment.

Reliable toxicological data are available from a wide range of scientific and regulatory sources. Identifying the most relevant and highest-quality evidence is a critical part of every toxicological risk assessment.

Common sources of toxicological information include:

• Peer-reviewed scientific publications
• Regulatory assessment reports and evaluation documents
• GLP and non-GLP toxicology study reports
• Public toxicology databases
• Industry submission dossiers
• International guidance documents
• Human epidemiological studies where available

Frequently consulted databases include:

• European Chemicals Agency (ECHA)
• U.S. Environmental Protection Agency (EPA)
• National Toxicology Program (NTP)
• PubChem
• Other specialised toxicology and chemical safety databases

Not all data have equal scientific value. Each study should be critically evaluated for its reliability, methodological quality, relevance, and applicability to the intended exposure scenario. Regulatory acceptance depends on both the quality of the evidence and the scientific justification for its use.

Limited toxicological data do not automatically prevent a scientifically sound risk assessment. Instead, toxicologists apply structured approaches to make the best possible use of the available evidence while identifying and addressing data gaps.

Common approaches include:

• Read-across using structurally or toxicologically similar substances
• In silico methods, including QSAR models and computational toxicology
• New Approach Methodologies (NAMs) such as advanced in vitro methods
• Weight-of-evidence assessments integrating multiple independent data sources
• Targeted testing, where additional information is scientifically justified and necessary

The objective is to develop a scientifically robust and regulatorily defensible assessment while avoiding unnecessary testing whenever suitable alternative data are available.

Yes. In many cases, a toxicological risk assessment can be completed without conducting new animal studies.

Modern risk assessments increasingly rely on existing scientific evidence and alternative methods, including:

• Published toxicological studies
• Regulatory assessments
• Read-across from similar substances
• Computational (in silico) models
• In vitro testing
• New Approach Methodologies (NAMs)
• Weight-of-evidence evaluations

Whether additional animal testing is necessary depends on factors such as:

• The quality and completeness of the available data
• The intended use of the substance
• Expected human exposure
• Applicable regulatory requirements
• Remaining scientific uncertainties

When sufficient evidence already exists to demonstrate safety, further animal testing is often unnecessary. This approach supports the internationally recognised 3Rs principles—Replacement, Reduction, and Refinement—which aim to minimise animal use while maintaining high scientific standards.

Common approaches include:

• Read-across using structurally or toxicologically similar substances
• In silico methods, including QSAR models and computational toxicology
• New Approach Methodologies (NAMs) such as advanced in vitro methods
• Weight-of-evidence assessments integrating multiple independent data sources
• Targeted testing, where additional information is scientifically justified and necessary

The objective is to develop a scientifically robust and regulatorily defensible assessment while avoiding unnecessary testing whenever suitable alternative data are available.

New Approach Methodologies (NAMs) are modern scientific methods that improve toxicological assessments by generating human-relevant safety information while reducing reliance on traditional animal testing.

NAMs include a wide range of innovative approaches, such as:

• Advanced in vitro cell and tissue models
• Computational toxicology and in silico modelling
• Quantitative Structure–Activity Relationship (QSAR) models
• Omics technologies
• High-throughput screening methods
• Integrated testing strategies
• Defined approaches that combine multiple sources of evidence

These methods help toxicologists:

• Identify potential hazards
• Investigate mechanisms of toxicity
• Fill data gaps
• Prioritise substances for further evaluation
• Support regulatory decision-making
• Improve the efficiency of safety assessments

Although NAMs are becoming increasingly important in regulatory toxicology, their use requires careful scientific interpretation and must be appropriate for the specific regulatory context. When combined with other available evidence, NAMs can strengthen toxicological risk assessments while supporting more efficient, ethical, and human-relevant safety evaluations.

In silico methods use computational models, algorithms, and existing scientific data to predict the toxicological properties of chemical substances. They are an important component of modern toxicological risk assessment and help generate scientifically robust safety information without relying solely on experimental testing.

These methods support risk assessments by:

• Predicting potential toxicological endpoints based on chemical structure
• Identifying data gaps and prioritising substances for further evaluation
• Supporting read-across assessments
• Complementing in vitro and in vivo studies
• Contributing to weight-of-evidence evaluations
• Reducing unnecessary animal testing where scientifically justified

One of the most widely used in silico approaches is the Quantitative Structure–Activity Relationship (QSAR) model, which predicts biological activity by analysing the relationship between a substance's chemical structure and known toxicological data.

The reliability of in silico predictions depends on factors such as model validation, applicability domain, data quality, and expert interpretation. For regulatory submissions, computational predictions are typically considered alongside experimental data and other lines of evidence rather than as standalone proof of safety.

As computational toxicology continues to evolve, in silico methods are becoming increasingly valuable for improving the efficiency, consistency, and scientific basis of toxicological risk assessments.

Toxicological read-across is a scientifically accepted approach used to predict the toxicological properties of one substance (the target substance) by using data from one or more structurally or biologically similar substances (source substances).

Read-across is particularly valuable when experimental data for the target substance are limited or unavailable, allowing existing evidence to be used instead of generating new studies.

A successful read-across assessment requires a scientifically justified demonstration that the substances share relevant similarities, including:

• Chemical structure
• Physicochemical properties
• Toxicokinetic behaviour (ADME)
• Metabolic pathways
• Mechanism of action
• Expected exposure scenario

A robust read-across justification explains why data from the source substance can be reliably applied to the target substance and discusses any uncertainties associated with the prediction.

When appropriately justified, read-across can:

• Address critical data gaps
• Support regulatory submissions
• Reduce the need for additional animal testing
• Improve the efficiency of chemical safety assessments
• Contribute to weight-of-evidence evaluations

Read-across is widely used within regulatory frameworks such as REACH and other international chemical safety programmes.

Health-based exposure limits (HBELs) are scientifically derived values that define levels of human exposure that are considered unlikely to cause adverse health effects under specified conditions of use.

These values are essential tools in toxicological risk assessment because they allow estimated human exposure to be compared with an acceptable exposure level.

Common health-based exposure limits include:

• Permitted Daily Exposure (PDE): A Permitted Daily Exposure (PDE) is primarily used in pharmaceutical risk assessments. It defines the maximum acceptable daily exposure to a substance throughout a lifetime without appreciable health risk and is commonly applied in contamination control and cross-contamination assessments.
• Tolerable Daily Intake (TDI): A Tolerable Daily Intake (TDI) represents the amount of a substance that can be consumed daily over an entire lifetime without causing significant health effects. TDIs are widely used in food safety and environmental toxicology.
• Derived No-Effect Level (DNEL): A Derived No-Effect Level (DNEL) is established under chemical safety regulations such as REACH. It represents the exposure level below which adverse health effects are not expected for workers or the general population.
• Tolerable Intake (TI): A Tolerable Intake (TI) is commonly used in medical device toxicological risk assessments. It defines an acceptable level of exposure to substances released from device materials during patient use.

Although these terms are used in different regulatory sectors, they are all derived from toxicological data and incorporate uncertainty factors to ensure that exposure limits remain protective of human health.

Health-based exposure limits are established through a structured scientific evaluation of toxicological evidence. The process begins with a review of available data, including animal studies, human data, mechanistic information, published literature, and regulatory assessments.

A critical step is the identification of the critical effect, defined as the most relevant adverse health effect observed at the lowest dose level suitable for risk assessment. The critical effect is selected based on factors such as severity, biological relevance, dose-response relationship, and quality of available data. It is used to determine the Point of Departure (POD), which may include a No-Observed-Adverse-Effect Level (NOAEL), Lowest-Observed-Adverse-Effect Level (LOAEL), or Benchmark Dose (BMD).

Uncertainty factors are then applied to account for differences between laboratory animals and humans, variability within the human population, exposure duration, data limitations, and remaining uncertainties. The resulting health-based exposure limit is interpreted according to the intended exposure scenario, including route of exposure, frequency and duration, target population, product use, and regulatory requirements.

Health-based exposure limits provide scientifically justified benchmarks for comparing estimated human exposure with acceptable exposure levels during toxicological risk assessment.

Common examples include:

• Permitted Daily Exposure (PDE): Used mainly in pharmaceutical risk assessments to define the maximum acceptable daily exposure to a substance over a lifetime without appreciable health risk.

• Tolerable Daily Intake (TDI): The amount of a substance consumed daily over a lifetime without significant health effects, commonly used in food and environmental safety.

• Derived No-Effect Level (DNEL): An exposure level below which adverse effects are not expected, used in chemical safety assessments such as REACH.

• Tolerable Intake (TI): Used in medical device assessments to define acceptable exposure to substances released from device materials.

Although terminology differs between sectors, all health-based exposure limits are derived from toxicological evidence and incorporate uncertainty factors to protect human health.

Uncertainty is an inherent part of every toxicological risk assessment (TRA) because scientific data are rarely complete. A transparent evaluation of uncertainty helps ensure that risk assessment conclusions remain scientifically robust, protective of human health, and suitable for regulatory decision-making.

Sources of uncertainty may include:

• Differences between laboratory animals and humans
• Variability within the human population
• Limited or incomplete toxicological data
• Uncertainty in exposure estimates
• Differences in study quality or reliability
• Limited understanding of a substance's mode of action
• Variability in real-world use conditions

To address these uncertainties, toxicologists apply internationally accepted scientific approaches, including:

• Assessment (uncertainty) factors to account for biological variability and data limitations
• Weight-of-evidence evaluations that integrate multiple independent data sources
• Conservative assumptions where information is limited
• Expert scientific judgement supported by regulatory guidance
• Sensitivity analyses, where appropriate, to evaluate the impact of key assumptions

Rather than eliminating uncertainty, these approaches ensure that it is clearly identified, scientifically justified, and appropriately incorporated into the final risk characterization. A transparent uncertainty assessment increases confidence in the conclusions and supports regulatory compliance.

The Threshold of Toxicological Concern (TTC) is a science-based risk assessment approach used to evaluate chemicals for which little or no substance-specific toxicological data are available.

The TTC concept is based on extensive toxicological databases showing that chemicals with similar structures and toxicological characteristics can often be grouped into categories with comparable levels of concern. For each category, a conservative exposure threshold is established below which adverse health effects are considered unlikely.

The TTC approach is particularly useful when:

• Substance-specific toxicity data are limited
• Human exposure is expected to be very low
• A rapid preliminary risk assessment is required
• Prioritising substances for further investigation
• Supporting product development or regulatory decision-making

The applicability of the TTC approach depends on several factors, including:

• Chemical structure
• Expected exposure level
• Intended use
• Toxicological alerts
• Applicable regulatory framework

Although TTC does not replace a full toxicological risk assessment where comprehensive data are available, it is a valuable screening tool that helps prioritise resources while maintaining a high level of health protection.

Extractables and Leachables (E&L) are chemical substances that can migrate from materials into pharmaceutical products, medical devices, packaging systems, manufacturing equipment, or other products that come into contact with humans.

Evaluating extractables and leachables is an essential part of product safety and regulatory toxicology because these substances may result in patient or consumer exposure.

What are extractables?

Extractables are compounds that can be released from a material under controlled laboratory conditions using aggressive extraction techniques. They represent the potential chemical profile that could migrate from a material.

Extractables studies are typically performed during product development to identify substances that may require further toxicological evaluation.

What are leachables?

Leachables are compounds that actually migrate from a material into the finished product or use environment under normal storage or clinical use conditions.

Leachables therefore represent the chemicals to which patients or consumers may be exposed during real-world use.

Why are extractables and leachables important?

An Extractables and Leachables (E&L) assessment helps determine whether detected compounds could present a risk to human health.

A comprehensive E&L risk assessment typically includes:

• Identification of detected compounds
• Estimation of patient or consumer exposure
• Hazard identification and toxicological evaluation
• Comparison with appropriate health-based exposure limits
• Risk characterization
• Identification of compounds requiring additional assessment

For pharmaceutical products and medical devices, toxicologists often use an Analytical Evaluation Threshold (AET) to determine which detected compounds require further toxicological review.

An E&L assessment integrates analytical chemistry, exposure assessment, and toxicological risk assessment to demonstrate product safety and support regulatory submissions. It plays a critical role in ensuring that materials, packaging components, manufacturing systems, and medical devices remain safe throughout their intended lifecycle.

ISO 10993-17:2023 describes a systematic approach for establishing allowable exposure limits for substances released from medical devices, including hazard identification, dose-response assessment, exposure estimation, and risk characterization.