Reference n° ENV/CBC/MONO(2025)18

Date: 31.10.2025

Overview:

This OECD Guidance Document sets out best-practice approaches to improve generation, reporting, sharing and regulatory use of research data for chemical hazard and risk assessments. It defines reporting, reliability and relevance principles, and emphasises FAIR data, tailored reporting templates, and tools to bridge non-standard academic data with regulatory evidence needs.

The Guidance describes workflows for researchers and assessors: designing studies for regulatory utility, structured literature searching and screening, use of evaluation tools (e.g., SciRAP, CRED), systematic review methods, and case studies illustrating integration of research data into regulatory decisions and recommendations for harmonised practices.

Main points

- Regulatory Uptake of Research Data: Improving the use of non-standard research data in regulatory chemical assessments enhances scientific robustness and supports legal requirements to consider all available evidence.

• - Stakeholder Responsibilities: Researchers, funders, publishers, reviewers, repository managers, and risk assessors all share responsibility to increase the regulatory utility and uptake of research data throughout its lifecycle.
• - Principles of Data Quality: High-quality reporting, reliability (internal validity), and regulatory relevance (external validity) are essential for research data to be useful in regulatory contexts.
• - Reporting Standards and Templates: Adhering to established reporting standards (such as OECD Harmonised Templates, ARRIVE, STROBE, SciRAP, CRED) and using structured data repositories maximises data accessibility, transparency, and reuse.
• - Systematic Review and Evidence Integration: Structured approaches such as systematic reviews and systematic evidence maps (SEMs) are recommended for identifying, screening, evaluating, and integrating research data in regulatory assessments.
• - Fit-for-Purpose Evaluation Tools: Use of clear, context-appropriate evaluation tools and critical appraisal methods is necessary for transparent, consistent assessment of study reliability and relevance; qualitative tools are preferred over simple scoring systems.
• - Publication of All Results: Both positive and negative (no-effect) results should be published and made accessible to avoid bias, support model development, and prevent unnecessary repetition of studies, especially animal studies.
• - Recommendations for Harmonisation and Training: Adoption of harmonised reporting, evaluation tools, and ongoing training for all stakeholders is critical to improve the quality, consistency, and regulatory acceptance of research data.

Citations:

Please cite this publication as:
OECD (2025), OECD Guidance Document on the Generation, Reporting and

Use of Research Data for Regulatory Assessments, OECD Series on Testing and Assessment, No. 417, OECD Environment, Health and Safety, Paris, https://one.oecd.org/document/ENV/CBC/MONO(2025)18/en/pdf

Reference n° ENV/CBC/MONO(2025)19

Date: 30.10.2025

Overview:

This OECD guidance details a comprehensive, updated framework for grouping chemicals—via analogue and category approaches—to support hazard assessment, reduce animal testing, and inform regulatory decisions. It integrates traditional methods with New Approach Methodologies (NAMs), (Q)SARs, omics, AOPs, and IATA/Defined Approaches to improve read-across, trend analysis, and uncertainty characterization.

The third edition provides stepwise workflows, reporting templates, tools, and case studies for selecting analogues, forming categories, and documenting read-across justifications across diverse substance types (including UVCBs, metals, nanomaterials), emphasizing applicability domains, data quality, and iterative review.

Main points:

Chemical Grouping Approaches : Grouping chemicals enables hazard assessment by considering structurally or mechanistically similar chemicals together, using either analogue (one-to-one or few-to-few) or category (many-to-many) approaches to fill data gaps and reduce animal testing.

Read-Across and Data Gap Filling : Read-across uses data from one or more source chemicals to predict properties or hazards of a target chemical lacking data, and can be qualitative (binary) or quantitative (numerical value); it is central to both analogue and category approaches.

Category Formation and Trends : Categories are defined by common structural, physicochemical, or mechanistic features, and allow for the identification of trends (e.g., toxicity, potency) across members, which supports interpolation and extrapolation to fill data gaps.

Uncertainty Assessment : Evaluation of uncertainties is essential in grouping and read-across, considering data quality, similarity rationale (structural, physicochemical, metabolic, bioactivity, MOA), and robustness of predictions; multiple frameworks and templates exist for systematic uncertainty assessment.

Role of New Approach Methodologies (NAMs) : NAMs, including in vitro assays, omics (transcriptomics, metabolomics), high-throughput/content screening (HTS/HCS), and computational models ((Q)SARs), provide supporting evidence for similarity, mechanistic justification, and can increase confidence in grouping.

Applicability Domain and Boundaries : Clearly defining the applicability domain—structural, physicochemical, and mechanistic boundaries—determines which chemicals can be reliably included in a group or category and supports regulatory acceptance.

Reporting and Documentation : Transparent documentation is required, including the rationale, data matrices, justification for inclusion/ exclusion, uncertainty analysis, and reporting formats for analogue and category approaches, often using modular templates and data matrices.

Special Considerations for Complex Substances and Nanomaterials : Grouping and read-across principles apply to substances of unknown or variable composition (UVCBs), metals, inorganics, and nanomaterials, but require additional attention to compositional, physicochemical, and transformation characteristics due to their complexity and variability.

Summary

Introduction

This document provides comprehensive guidance on the grouping of chemicals for hazard assessment, offering methodologies to increase efficiency, reduce animal testing, and ensure scientific robustness in regulatory and scientific contexts.

Key Insights and Themes

• Grouping Approaches enable the assessment of chemicals as analogues or categories, allowing data from tested chemicals to predict properties of untested ones, thus reducing the need for extensive testing.

• Analogue Approach uses empirical data from one or more structurally or mechanistically similar chemicals to predict properties for a specific target chemi- cal, emphasizing the importance of shared mode or mechanism of action.

• Category Approach organizes chemicals into groups with similar or regularly patterned properties, supporting hazard assessment through trend analysis and read-across within the group.

• Read-Across and Data Gap Filling are central techniques, where information from one chemical or group is used to fill data gaps for others, and can be applied qualitatively or quantitatively.

• Uncertainty Analysis is integral, requiring systematic identification, characteri- zation, and documentation of uncertainties in both data and similarity rationales to ensure robust predictions.

• New Approach Methodologies (NAMs), such as in vitro assays, omics tech- nologies, high-throughput screening, and computational models, are increasingly used to substantiate similarity and support grouping hypotheses.

• Bioactivity Similarity leverages biological response data (e.g., from omics or HTS/HCS) as evidence for grouping, with confidence strengthened by mechanistic links to endpoints or adverse outcome pathways.

• Adverse Outcome Pathways (AOPs) provide mechanistic frameworks linking molecular events to adverse effects, supporting grouping and read-across by clari- fying the biological plausibility of groupings.

• Integrated Approaches to Testing and Assessment (IATA) and Defined Ap- proaches (DA) combine multiple evidence sources, including grouping, to guide hazard and risk assessment in a structured manner.

• Applicability Domains and Boundaries must be clearly defined for both ana- logues and categories, specifying structural, physicochemical, and mechanistic cri- teria for group membership and reliable predictions.

• Subcategories and Breakpoints may arise within categories when trends do not apply uniformly, requiring endpoint-specific justifications and potentially lead- ing to subcategorization for regulatory clarity.

• Regulatory Context and Evolution drive the development of grouping guid- ance, with frameworks like EU REACH and ECHA’s Read-Across Assessment Frame- work (RAAF) shaping scientific and documentation standards.

• Reporting Formats for analogue and category approaches are standardized to ensure transparency, reproducibility, and comprehensive justification, including data matrices and explicit uncertainty assessments.

• Computational Tools such as the OECD QSAR Toolbox, GenRA, and others support analogue identification, trend analysis, and category development by pro- viding systematic and reproducible methods.

• Special Considerations are addressed for complex substances (UVCBs), met- als, inorganic compounds, and nanomaterials, with tailored grouping and read- across strategies reflecting their unique characteristics and data challenges.

• Weight of Evidence (WoE) Approaches are recommended to integrate multi- ple lines of evidence, address data gaps, and support regulatory decision-making with transparent confidence assessments.

• Continuous Evolution of the guidance is expected, reflecting advances in sci- ence, technology, and regulatory experience, with periodic updates to incorporate new data sources, methodologies, and case studies.

• International Collaboration underpins the development and harmonization of grouping approaches, with contributions from global regulatory agencies, scientif- ic experts, and industry stakeholders.

Conclusion

Grouping of chemicals, supported by robust methodologies, uncertainty analysis, and evolving scientific tools, enables more efficient, ethical, and scientifically sound hazard assessment for regulatory and research purposes.

Citation: not available

Reference n° ENV/CBC/MONO(2025)16

Date: 14.10.2025

Overview:

The OECD monograph presents Integrated Approaches for Testing and Assessment (IATA) using combined bioinformatics tools to extrapolate chemical toxicity across species. It details the complementary use of SeqAPASS and G2P-SCAN to evaluate protein and pathway conservation, supporting hazard identification, prioritization, and regulatory decision-making while reducing animal testing.

The document describes workflows, case studies (PPARα, ESR1, GABRA1), tool inputs/outputs, uncertainties, and regulatory applications (ERA, endocrine disruption, endangered species). It provides method guidance, software instructions, and evidence synthesis strategies for implementing NAMs in cross-species chemical safety assessments.

Main points:

Bioinformatics Integration: Combining the SeqAPASS and G2P-SCAN computational tools enables robust cross-species extrapolation of chemical toxicity by assessing both protein target and biological pathway conservation, supporting more informed chemical safety assessments while reducing animal testing.

Key Tools: SeqAPASS predicts species susceptibility to chemicals by evaluating protein sequence and structural conservation across thousands of species, while G2P-SCAN analyzes the conservation of biological pathways using human gene inputs and model organisms.

Regulatory Relevance: These approaches align with global regulatory trends favoring New Approach Methodologies (NAMs), facilitating mechanistic, transparent, and ethical chemical risk assessments that minimize reliance on animal testing.

Workflow Process: The integrated workflow involves identifying a chemical’s molecular target, mapping its associated biological pathways, prioritizing key pathway proteins, and using both tools to predict conservation and susceptibility across species, which informs regulatory decision-making.

Expert Judgment Requirement: Critical steps such as target identification, isoform selection, and pathway prioritization rely on expert judgment, often requiring literature review, evaluation of empirical evidence, and knowledge of protein structure-function relationships.

Limitations: The approach is limited by the availability and quality of protein and gene sequence data, incomplete pathway annotation in non-model species, and is most applicable when the chemical’s molecular target is known and well characterized.

Regulatory Applications: Results are used for hazard identification, prioritization, and as additional lines of evidence in weight-of-evidence evaluations for regulatory frameworks such as the Endangered Species Act, endocrine disruptor screening, and pesticide registration.

Case Study Outcomes: Application to chemicals like 2-ethylhexanoic acid, diethylstilbestrol, and topiramate demonstrated that consensus between SeqAPASS and G2P-SCAN enhances confidence in predicting pathway conservation and chemical susceptibility across mammalian and other vertebrate species.

Summary

Introduction

This document presents integrated bioinformatics approaches for extrapolating chemical toxicity knowledge across species to inform chemical safety assessments, aiming to reduce reliance on animal testing and improve regulatory decision-making.

Key Insights and Themes

• Global regulatory shift increasingly favors New Approach Methodologies (NAMs), such as computational and cell-based tools, to assess chemical safety in an ethical, efficient, and scientifically robust manner.

• Cross-species extrapolation is essential for chemical risk assessment, enabling predictions about chemical effects in untested or protected species based on data from surrogate organisms.

• Traditional methods like safety factors and species sensitivity distributions have limitations due to assumptions about interspecies relatedness and data availability.

• Bioinformatics tools—notably SeqAPASS (Sequence Alignment to Predict Across Species Susceptibility) and G2P-SCAN (Genes-to-Pathways Species Conser- vation Analysis)—enable systematic evaluation of protein and pathway conservation across species.

• SeqAPASS assesses protein sequence and structural conservation to predict potential chemical-protein interactions across thousands of species, supporting chemical susceptibility predictions.

• G2P-SCAN analyzes conservation of biological pathways using human gene inputs, mapping orthologs and functional families across key model species to infer pathway conservation.

• Combined application of SeqAPASS and G2P-SCAN strengthens the weight of evidence by integrating molecular (protein) and pathway-level data, enhancing confidence in species extrapolation.

• Integrated approach supports hazard assessment, prioritization in early screening, and intelligent test design by identifying the most biologically relevant species for further testing.

• Regulatory context is evolving to accept mechanistic, cell-based, and computational data, requiring transparent, scientifically robust methods for cross-species extrapolation.

• Applicability domain for these approaches requires a known chemical-biomolecule interaction in at least one species and relies on high-quality sequence and annotation data.

• Workflow involves identifying chemical targets, mapping pathways, prioritizing key proteins, and combining evidence from both tools to generate lists of susceptible species.

• Expert judgment is critical at multiple decision points, including target selection, isoform choice, and interpretation of orthology and pathway conservation results.

• Uncertainties arise from incomplete knowledge of chemical-protein interactions, taxonomic coverage, and lack of empirical evidence for some pathways or species, but can be reduced by integrating multiple data sources.

• Case studies demonstrate the approach using chemicals such as 2-ethylhexanoic acid (PPARα target), diethylstilbestrol and butylparaben (ESR1 target), and topiramate (GABRA1 target), showing practical application and limitations.

• Results from combined approaches provide lists of species likely to be susceptible based on pathway and protein conservation, supporting regulatory decisions for hazard identification, prioritization, and species protection.

• Links to Adverse Outcome Pathways (AOPs) allow mapping of molecular and functional data to regulatory-relevant outcomes, helping to define the biologically plausible taxonomic domain of applicability (tDOA) for AOPs.

• Regulatory applications include support for the Endangered Species Act, endocrine disruptor screening, and pesticide registration, by identifying at-risk species and guiding intelligent testing strategies.

• Open-source and accessibility of tools like SeqAPASS and G2P-SCAN, along with transparent workflows, facilitate broader adoption and continuous improvement as more sequence data become available.

Conclusion

Integrating computational bioinformatics tools like SeqAPASS and G2P-SCAN enables more transparent, scientifically robust, and ethically responsible cross-species extrapolation for chemical safety assessment, supporting regulatory decisions and reducing reliance on animal testing.

Citation:

OECD (2025), Case Studies for the Integrated Approaches for Testing and Assessment in the Application of Combined Bioinformatics Approaches for Cross Species Extrapolation of Toxicity Knowledge to inform Chemical Safety. Tenth Review Cycle (2024), OECD Series on Testing and Assessment, No 415, OECD Environment, Health and Safety, Paris, https://one.oecd.org/official-document/ENV/CBC/ MONO(2025)16/en