From "weight of evidence" to quantitative data integration using multicriteria decision analysis and Bayesian methods

Advancing the quantitative understanding of adverse outcome pathways: current status, methodologies, and future directions

An adverse outcome pathway (AOP) framework maps the sequence of events leading to adverse outcomes from chemical exposures, providing a mechanistic understanding often absent in traditional methods. The quantitative AOP (qAOP) advances AOP by integrating quantitative data and mathematical modeling, thereby providing a more precise comprehension of relationships between molecular initiating events, key events, and adverse outcomes. This review critically examines three primary methodologies: systems toxicology, regression modeling, and Bayesian network modeling, highlighting their strengths, limitations, and specific data requirements within toxicology. Through an analysis of current methodologies and challenges, this review emphasizes the integration of experimental and computational approaches to elucidate key event relationships and proposes strategies for overcoming limitations through standardized protocols and advanced computational tools. By outlining future research directions and the potential of qAOPs to transform chemical risk assessment, this review aims to contribute to the advancement of regulatory science and the protection of public health and the environment.

A quantitative weight-of-evidence method for confidence assessment of adverse outcome pathway networks: A case study on chemical-induced liver steatosis

The field of chemical toxicity testing is undergoing a transition to overcome the limitations of in vivo experiments. This evolution involves implementing innovative non-animal approaches to improve predictability and provide a more precise understanding of toxicity mechanisms. Adverse outcome pathway (AOP) networks are pivotal in organizing existing mechanistic knowledge related to toxicological processes. However, these AOP networks are dynamic and require regular updates to incorporate the latest data. Regulatory challenges also persist due to concerns about the reliability of the information they offer. This study introduces a generic Weight-of-Evidence (WoE) scoring method, aligned with the tailored Bradford-Hill criteria, to quantitatively assess the confidence levels in key event relationships (KERs) within AOP networks. We use the previously published AOP network on chemical-induced liver steatosis, a prevalent form of human liver injury, as a case study. Initially, the existing AOP network is optimized with the latest scientific information extracted from PubMed using the free SysRev platform for artificial intelligence (AI)-based abstract inclusion and standardized data collection. The resulting optimized AOP network, constructed using Cytoscape, visually represents confidence levels through node size (key event, KE) and edge thickness (KERs). Additionally, a Shiny application is developed to facilitate user interaction with the dataset, promoting future updates. Our analysis of 173 research papers yielded 100 unique KEs and 221 KERs among which 72 KEs and 170 KERs, respectively, have not been previously documented in the prior AOP network or AOP-wiki. Notably, modifications in de novo lipogenesis, fatty acid uptake and mitochondrial beta-oxidation, leading to lipid accumulation and liver steatosis, garnered the highest KER confidence scores. In conclusion, our study delivers a generic methodology for developing and assessing AOP networks. The quantitative WoE scoring method facilitates in determining the level of support for KERs within the optimized AOP network, offering valuable insights into its utility in both scientific research and regulatory contexts. KERs supported by robust evidence represent promising candidates for inclusion in an in vitro test battery for reliably predicting chemical-induced liver steatosis within regulatory frameworks.

Can Weight of Evidence, Quantitative Bias, and Bounding Methods Evaluate Robustness of Real-world Evidence for Regulator and Health Technology Assessment Decisions on Medical Interventions?

PURPOSE

High-quality evidence is crucial for health care intervention decision-making. These decisions frequently use nonrandomized data, which can be more vulnerable to biases than randomized trials. Accordingly, methods to quantify biases and weigh available evidence could elucidate the robustness of findings, giving regulators more confidence in making approval and reimbursement decisions.

METHODS

We conducted an integrative literature review to identify methods for determining probability of causation, evaluating weight of evidence, and conducting quantitative bias analysis as related to health care interventions. Eligible studies were published from 2012 to 2021, applicable to pharmacoepidemiology, and presented a method that met our objective.

FINDINGS

Twenty-two eligible studies were classified into 4 categories: (1) quantitative bias analysis; (2) weight of evidence methods; (3) Bayesian networks; and (4) miscellaneous. All of the methods have strengths, limitations, and situations in which they are more well suited than others. Some methods seem to lend themselves more to applications of health care evidence on medical interventions than others.

IMPLICATIONS

To provide robust evidence for and improve confidence in regulatory or reimbursement decisions, we recommend applying multiple methods to triangulate associations of medical interventions, accounting for biases in different ways. This approach could lead to well-defined robustness assessments of study findings and appropriate science-driven decisions by regulators and payers for public health.

From Molecular Descriptors to Intrinsic Fish Toxicity of Chemicals: An Alternative Approach to Chemical Prioritization

The European and U.S. chemical agencies have listed approximately 800k chemicals about which knowledge of potential risks to human health and the environment is lacking. Filling these data gaps experimentally is impossible, so in silico approaches and prediction are essential. Many existing models are however limited by assumptions (e.g., linearity and continuity) and small training sets. In this study, we present a supervised direct classification model that connects molecular descriptors to toxicity. Categories can be driven by either data (using k-means clustering) or defined by regulation. This was tested via 907 experimentally defined 96 h LC50 values for acute fish toxicity. Our classification model explained ≈90% of the variance in our data for the training set and ≈80% for the test set. This strategy gave a 5-fold decrease in the frequency of incorrect categorization compared to a quantitative structure-activity relationship (QSAR) regression model. Our model was subsequently employed to predict the toxicity categories of ≈32k chemicals. A comparison between the model-based applicability domain (AD) and the training set AD was performed, suggesting that the training set-based AD is a more adequate way to avoid extrapolation when using such models. The better performance of our direct classification model compared to that of QSAR methods makes this approach a viable tool for assessing the hazards and risks of chemicals.

Probabilistic risk assessment - the keystone for the future of toxicology

Safety sciences must cope with uncertainty of models and results as well as information gaps. Acknowledging this uncer-tainty necessitates embracing probabilities and accepting the remaining risk. Every toxicological tool delivers only probable results. Traditionally, this is taken into account by using uncertainty / assessment factors and worst-case / precautionary approaches and thresholds. Probabilistic methods and Bayesian approaches seek to characterize these uncertainties and promise to support better risk assessment and, thereby, improve risk management decisions. Actual assessments of uncertainty can be more realistic than worst-case scenarios and may allow less conservative safety margins. Most importantly, as soon as we agree on uncertainty, this defines room for improvement and allows a transition from traditional to new approach methods as an engineering exercise. The objective nature of these mathematical tools allows to assign each methodology its fair place in evidence integration, whether in the context of risk assessment, sys-tematic reviews, or in the definition of an integrated testing strategy (ITS) / defined approach (DA) / integrated approach to testing and assessment (IATA). This article gives an overview of methods for probabilistic risk assessment and their application for exposure assessment, physiologically-based kinetic modelling, probability of hazard assessment (based on quantitative and read-across based structure-activity relationships, and mechanistic alerts from in vitro studies), indi-vidual susceptibility assessment, and evidence integration. Additional aspects are opportunities for uncertainty analysis of adverse outcome pathways and their relation to thresholds of toxicological concern. In conclusion, probabilistic risk assessment will be key for constructing a new toxicology paradigm - probably!

Bioinformatics analysis and quantitative weight of evidence assessment to map the potential mode of actions of bisphenol A

Bisphenol A (BPA) is a classical chemical contaminant in food, and the mode of action (MOA) of BPA remains unclear, constraining the progress of risk assessment. This study aims to assess the potential MOAs of BPA regarding reproductive/developmental toxicity, neurological toxicity, and proliferative effects on the mammary gland and the prostate potentially related to carcinogenesis by using the Comparative Toxicogenomics Database (CTD)-based bioinformatics analysis and the quantitative weight of evidence (QWOE) approach on the basis of the principles of Toxicity Testing in the 21st Century. The CTD-based bioinformatics analysis results showed that estrogen receptor 1, estrogen receptor 2, mitogen-activated protein kinase (MAPK) 1, MAPK3, BCL2 apoptosis regulator, caspase 3, BAX, androgen receptor, and AKT serine/threonine kinase 1 could be the common target genes, and the apoptotic process, cell proliferation, testosterone biosynthetic process, and estrogen biosynthetic process might be the shared phenotypes for different target organs. In addition, the KEGG pathways of the BPA-induced action might involve the estrogen signaling pathway and pathways in cancer. After the QWOE evaluation, two potential estrogen receptor-related MOAs of BPA-induced testis dysfunction and learning-memory deficit were proposed. However, the confidence and the human relevance of the two MOAs were moderate, prompting studies to improve the MOA-based risk assessment of BPA.

Occupational exposure of dumper operators to whole-body vibration in opencast coal mines: an approach for risk assessment using a Bayesian network

Whole-body vibration (WBV) is one of the leading risk factors for development of musculoskeletal disorders (MSDs) that develop symptoms of lower back pain, pain in the neck and shoulders, digestive problems, blood pressure and diabetes among professional dumper operators. The present study specifically aimed at assessing the WBV exposure of 79 dumper operators engaged in two Indian opencast coal mines through vibration measurements followed by questionnaire survey. From the daily frequency-weighted root mean square exposure, dumper operators have experienced vibration levels higher than the Health Guidance Caution Zone (HGCZ) of Standard No. ISO 2631-1:1997. However, on the basis of daily vibration dose values, 60.8% of operators have experienced vibration levels above the HGCZ. Finally, an attempt was also made to explore the potential of a Bayesian network to predict the risk factors for WBV of dumper operators in development of MSDs to prioritize the factors for human health risk assessment.

Integrated characterization and risk management of marine sediments: The case study of the industrialized Bagnoli area (Naples, Italy)

The aim of the present work is to demonstrate the practical importance of a multidisciplinary approach and weighted criteria to synthesize and integrate different typologies of data (or lines of evidence, LOEs), including chemical levels in marine sediments, their bioavailability to specific indicator species, ecotoxicological effects measured through subcellular biomarkers and batteries of bioassays, and potential impacts of pollution on local benthic communities. The area of Bagnoli (Gulf of Naples, Southern Italy) was selected as a model case-study, as it is a coastal area chronically impacted by massive industrial contamination (trace metals and hydrocarbons), and dismissed decades ago without any subsequent remediation or habitat restoration. The results of each LOE were elaborated to provide specific hazard indices before their overall integration in a weight of evidence (WOE) evaluation. Levels of some trace metals and PAHs revealed a severe contamination in the entire study area. Bioavailability of hydrocarbons was evident particularly for high molecular weight PAHs, which also caused significant variations of cellular biomarkers, such as cytochrome P450 metabolization in fish, lysosomal membrane destabilization in mussels, genotoxic effects both in fish and molluscs. The results of a battery of bioassays indicated less marked responses compared to those obtained from chemical and biomarkers analyses, with acute toxicity still present in sediments close to the source of contamination. The analysis of benthic assemblages showed limited evidence of impact in the whole area, indicating a good functioning of local ecosystems at chronic contamination. Overall, the results of this study confirm the need of combining chemical and biological data, the quantitative characterization of various typologies of hazard and the importance of assessing an integrated environmental WOE risk, to orientate specific and scientifically-supported management options in industrialized areas.

Internationalization of read-across as a validated new approach method (NAM) for regulatory toxicology

Read-across (RAx) translates available information from well-characterized chemicals to a substance for which there is a toxicological data gap. The OECD is working on case studies to probe general applicability of RAx, and several regulations (e.g., EU-REACH) already allow this procedure to be used to waive new in vivo tests. The decision to prepare a review on the state of the art of RAx as a tool for risk assessment for regulatory purposes was taken during a workshop with international experts in Ranco, Italy in July 2018. Three major issues were identified that need optimization to allow a higher regulatory acceptance rate of the RAx procedure: (i) the definition of similarity of source and target, (ii) the translation of biological/toxicological activity of source to target in the RAx procedure, and (iii) how to deal with issues of ADME that may differ between source and target. The use of new approach methodologies (NAM) was discussed as one of the most important innovations to improve the acceptability of RAx. At present, NAM data may be used to confirm chemical and toxicological similarity. In the future, the use of NAM may be broadened to fully characterize the hazard and toxicokinetic properties of RAx compounds. Concerning available guidance, documents on Good Read-Across Practice (GRAP) and on best practices to perform and evaluate the RAx process were identified. Here, in particular, the RAx guidance, being worked out by the European Commission’s H2020 project EU-ToxRisk together with many external partners with regulatory experience, is given.

An integrated pathway based on in vitro data for the human hazard assessment of nanomaterials

In line with the 3R concept, nanotoxicology is shifting from a phenomenological to a mechanistic approach based on in vitro and in silico methods, with a consequent reduction in animal testing. Risk Assessment (RA) and Life Cycle Assessment (LCA) methodologies, which traditionally rely on in vivo toxicity studies, will not be able to keep up with the pace of development of new nanomaterials unless they adapt to use this new type of data. While tools and models are already available and show a great potential for future use in RA and LCA, currently none is able alone to quantitatively assess human hazards (i.e. calculate chronic NOAEL or ED₅₀ values). By highlighting which models and approaches can be used in a quantitative way with the available knowledge and data, we propose an integrated pathway for the use of in vitro data in RA and LCA. Starting with the characterization of nanoparticles' properties, the pathway then investigates how to select relevant in vitro human data, and how to bridge in vitro dose-response relationships to in vivo effects. If verified, this approach would allow RA and LCA to stir up the development of nanotoxicology by giving indications about the data and quality requirements needed in risk methodologies.

Managing evidence in food safety and nutrition

Evidence (‘data’) is at the heart of EFSA's 2020 Strategy and is addressed in three of its operational objectives: (1) adopt an open data approach, (2) improve data interoperability to facilitate data exchange, and (3) migrate towards structured scientific data. As the generation and availability of data have increased exponentially in the last decade, potentially providing a much larger evidence base for risk assessments, it is envisaged that the acquisition and management of evidence to support future food safety risk assessments will be a dominant feature of EFSA's future strategy. During the breakout session on ‘Managing evidence’ of EFSA's third Scientific Conference ‘Science, Food, Society’, current challenges and future developments were discussed in evidence management applied to food safety risk assessment, accounting for the increased volume of evidence available as well as the increased IT capabilities to access and analyse it. This paper reports on presentations given and discussions held during the session, which were centred around the following three main topics: (1) (big) data availability and (big) data connection, (2) problem formulation and (3) evidence integration.

Big-data and machine learning to revamp computational toxicology and its use in risk assessment

The creation of large toxicological databases and advances in machine-learning techniques have empowered computational approaches in toxicology. Work with these large databases based on regulatory data has allowed reproducibility assessment of animal models, which highlight weaknesses in traditional in vivo methods. This should lower the bars for the introduction of new approaches and represents a benchmark that is achievable for any alternative method validated against these methods. Quantitative Structure Activity Relationships (QSAR) models for skin sensitization, eye irritation, and other human health hazards based on these big databases, however, also have made apparent some of the challenges facing computational modeling, including validation challenges, model interpretation issues, and model selection issues. A first implementation of machine learning-based predictions termed REACHacross achieved unprecedented sensitivities of >80% with specificities >70% in predicting the six most common acute and topical hazards covering about two thirds of the chemical universe. While this is awaiting formal validation, it demonstrates the new quality introduced by big data and modern data-mining technologies. The rapid increase in the diversity and number of computational models, as well as the data they are based on, create challenges and opportunities for the use of computational methods.

Computational approaches to chemical hazard assessment

Computational prediction of toxicity has reached new heights as a result of decades of growth in the magnitude and diversity of biological data. Public packages for statistics and machine learning make model creation faster. New theory in machine learning and cheminformatics enables integration of chemical structure, toxicogenomics, simulated and physical data in the prediction of chemical health hazards, and other toxicological information. Our earlier publications have characterized a toxicological dataset of unprecedented scale resulting from the European REACH legislation (Registration Evaluation Authorisation and Restriction of Chemicals). These publications dove into potential use cases for regulatory data and some models for exploiting this data. This article analyzes the options for the identification and categorization of chemicals, moves on to the derivation of descriptive features for chemicals, discusses different kinds of targets modeled in computational toxicology, and ends with a high-level perspective of the algorithms used to create computational toxicology models.

Donohoe T, Garnett K, Lansink AO, Afonso A, Noteborn H. Emerging risks identification on food and feed - EFSA

The European Food Safety Authority's has established procedures for the identification of emerging risk in food and feed. The main objectives are to: (i) to carry out activities aiming at identifying, assessing and disseminating information on emerging issues and ensure coordination with relevant networks and international organisations; (ii) promote the identification of data sources and data collection and /or data generation in prioritised emerging issues; and the (iii) evaluate of the collected information and identify of emerging risks. The objective(s) of the Standing Working Group on Emerging Risks (SWG-ER) is to collaborate with EFSA on the emerging risks identification (ERI) procedure and provide strategic direction for EFSA work building on past and ongoing projects related to EFSA ERI procedure. The SWG-ER considered the ERI methodologies in place and results obtained by EFSA. It was concluded that a systematic approach to the identification of emerging issues based on experts' networks is the major strength of the procedure but at present, it is mainly focused on single issues, over short to medium time horizons, no consistent weighting or ranking is applied and clear governance of emerging risks with follow-up actions is missing. The analysis highlighted weaknesses with respect to data collection, analysis and integration. No methodology is in place to estimate the value of the procedure outputs in terms of avoided risk and there is urgent need for a communication strategy that addresses the lack of data and knowledge uncertainty and addresses risk perception issues. Recommendations were given in three areas: (i) Further develop a food system-based approach including the integration of social sciences to improve understanding of interactions and dynamics between actors and drivers and the development of horizon scanning protocols; (ii) Improve data processing pipelines to prepare big data analytics, implement a data validation system and develop data sharing agreements to explore mutual benefits; and (iii) Revise the EFSA procedure for emerging risk identification to increase transparency and improve communication.

Linking People to Coastal Habitats: A meta-analysis of final ecosystem goods and services on the coast

Coastal ecosystem goods and services (EGS) have steadily gained traction in the scientific literature over the last few decades, providing a wealth of information about underlying coastal habitat dependencies. This meta-analysis summarizes relationships between coastal habitats and final ecosystem goods and services (FEGS) users. Through a "weight of evidence" approach synthesizing information from published literature, we assessed habitat classes most relevant to coastal users. Approximately 2,800 coastal EGS journal articles were identified by online search engines, of which 16% addressed linkages between specific coastal habitats and FEGS users, and were retained for subsequent analysis. Recreational (83%) and industrial (35%) users were most cited in literature, with experiential-users/hikers and commercial fishermen most prominent in each category, respectively. Recreational users were linked to the widest diversity of coastal habitat subclasses (i.e., 22 of 26). Whereas, mangroves and emergent wetlands were most relevant for property owners. We urge EGS studies to continue surveying local users and identifying habitat dependencies, as these steps are important precursors for developing appropriate coastal FEGS metrics and facilitating local valuation. In addition, understanding how habitats contribute to human well-being may assist communities in prioritizing restoration and evaluating development scenarios in the context of future ecosystem service delivery.

Guidance on the use of the weight of evidence approach in scientific assessments

EFSA requested the Scientific Committee to develop a guidance document on the use of the weight of evidence approach in scientific assessments for use in all areas under EFSA's remit. The guidance document addresses the use of weight of evidence approaches in scientific assessments using both qualitative and quantitative approaches. Several case studies covering the various areas under EFSA's remit are annexed to the guidance document to illustrate the applicability of the proposed approach. Weight of evidence assessment is defined in this guidance as a process in which evidence is integrated to determine the relative support for possible answers to a question. This document considers the weight of evidence assessment as comprising three basic steps: (1) assembling the evidence into lines of evidence of similar type, (2) weighing the evidence, (3) integrating the evidence. The present document identifies reliability, relevance and consistency as three basic considerations for weighing evidence.

Integrating causation in investigative ecological weight of evidence assessments

Weight of evidence (WOE) frameworks integrate environmental assessment data to reach conclusions regarding relative certainty of adverse environmental effects due to stressors, possible causation, and key uncertainties. Such studies can be investigative (i.e., determining whether adverse impact is occurring to identify a need for management) or retrospective (i.e., determining the cause of a detected impact such that management efforts focus on the correct stressor). Such WOE assessments do not themselves definitively establish causation; they provide the basis for subsequent follow-up studies to further investigate causation. We propose a modified investigative WOE framework that includes an additional weighting step, which we term "direction weighting." This additional step allows for the examination of alternative hypotheses and provides improved certainty regarding possible causation. To our knowledge, this approach has not been previously applied in investigative ecological WOE assessments. We provide a generic example of 2 conflicting hypotheses related to a mine discharging treated effluent to a freshwater lake: chemical toxicity versus nutrient enrichment. Integr Environ Assess Manag 2017;13:702-713. © 2016 SETAC.

A multi-attribute methodology for the prioritisation of oil contaminated sites in the Niger Delta

The Ogoniland region of the Niger Delta contains a vast number of sites contaminated with petroleum hydrocarbons that originated from Nigeria's active oil sector. The United Nations Environment Programme (UNEP) reported on this widespread contamination in 2011, however, wide-scale action to clean-up these sites has yet to be initiated. A challenge for decision makers responsible for the clean-up of these sites has been the prioritisation of sites to enable appropriate allocation of scarce resources. In this study, a risk-based multi-criteria decision analysis framework was used to prioritise high-risk sites contaminated with petroleum hydrocarbons in the Ogoniland region of Nigeria. The prioritisation method used a set of risk-based attributes that took into account chemical and ecological impacts, as well as socio-economic impacts, providing a holistic assessment of the risk. Data for the analysis was taken from the UNEP Environmental Assessment of Ogoniland, where over 110 communities were assessed for oil-contamination. Results from our prioritisation show that the highest-ranking sites were not necessarily the sites with the highest observed level of hydrocarbon contamination. This differentiation was due to our use of proximity as a surrogate measure for likelihood of exposure. Composite measures of risk provide a more robust assessment, and can enrich discussions about risk management and the allocation of resources for the clean-up of affected sites.

Systems Biology to Support Nanomaterial Grouping

The assessment of potential health risks of engineered nanomaterials (ENMs) is a challenging task due to the high number and great variety of already existing and newly emerging ENMs. Reliable grouping or categorization of ENMs with respect to hazards could help to facilitate prioritization and decision making for regulatory purposes. The development of grouping criteria, however, requires a broad and comprehensive data basis. A promising platform addressing this challenge is the systems biology approach. The different areas of systems biology, most prominently transcriptomics, proteomics and metabolomics, each of which provide a wealth of data that can be used to reveal novel biomarkers and biological pathways involved in the mode-of-action of ENMs. Combining such data with classical toxicological data would enable a more comprehensive understanding and hence might lead to more powerful and reliable prediction models. Physico-chemical data provide crucial information on the ENMs and need to be integrated, too. Overall statistical analysis should reveal robust grouping and categorization criteria and may ultimately help to identify meaningful biomarkers and biological pathways that sufficiently characterize the corresponding ENM subgroups. This chapter aims to give an overview on the different systems biology technologies and their current applications in the field of nanotoxicology, as well as to identify the existing challenges.

A Tractable Method for Measuring Nanomaterial Risk Using Bayesian Networks

While control banding has been identified as a suitable framework for the evaluation and the determination of potential human health risks associated with exposure to nanomaterials (NMs), the approach currently lacks any implementation that enjoys widespread support. Large inconsistencies in characterisation data, toxicological measurements and exposure scenarios make it difficult to map and compare the risk associated with NMs based on physicochemical data, concentration and exposure route. Here we demonstrate the use of Bayesian networks as a reliable tool for NM risk estimation. This tool is tractable, accessible and scalable. Most importantly, it captures a broad span of data types, from complete, high quality data sets through to data sets with missing data and/or values with a relatively high spread of probability distribution. The tool is able to learn iteratively in order to further refine forecasts as the quality of data available improves. We demonstrate how this risk measurement approach works on NMs with varying degrees of risk potential, namely, carbon nanotubes, silver and titanium dioxide. The results afford even non-experts an accurate picture of the occupational risk probabilities associated with these NMs and, in doing so, demonstrated how NM risk can be evaluated into a tractable, quantitative risk comparator.

A critical review of frameworks used for evaluating reliability and relevance of (eco)toxicity data: Perspectives for an integrated eco-human decision-making framework

Considerable efforts have been invested so far to evaluate and rank the quality and relevance of (eco)toxicity data for their use in regulatory risk assessment to assess chemical hazards. Many frameworks have been developed to improve robustness and transparency in the evaluation of reliability and relevance of individual tests, but these frameworks typically focus on either environmental risk assessment (ERA) or human health risk assessment (HHRA), and there is little cross talk between them. There is a need to develop a common approach that would support a more consistent, transparent and robust evaluation and weighting of the evidence across ERA and HHRA. This paper explores the applicability of existing Data Quality Assessment (DQA) frameworks for integrating environmental toxicity hazard data into human health assessments and vice versa. We performed a comparative analysis of the strengths and weaknesses of eleven frameworks for evaluating reliability and/or relevance of toxicity and ecotoxicity hazard data. We found that a frequent shortcoming is the lack of a clear separation between reliability and relevance criteria. A further gaps and needs analysis revealed that none of the reviewed frameworks satisfy the needs of a common eco-human DQA system. Based on our analysis, some key characteristics, perspectives and recommendations are identified and discussed for building a common DQA system as part of a future integrated eco-human decision-making framework. This work lays the basis for developing a common DQA system to support the further development and promotion of Integrated Risk Assessment.

Adverse Outcome Pathways for Regulatory Applications: Examination of Four Case Studies With Different Degrees of Completeness and Scientific Confidence

Adverse outcome pathways (AOPs) offer a pathway-based toxicological framework to support hazard assessment and regulatory decision-making. However, little has been discussed about the scientific confidence needed, or how complete a pathway should be, before use in a specific regulatory application. Here we review four case studies to explore the degree of scientific confidence and extent of completeness (in terms of causal events) that is required for an AOP to be useful for a specific purpose in a regulatory application: (i) Membrane disruption (Narcosis) leading to respiratory failure (low confidence), (ii) Hepatocellular proliferation leading to cancer (partial pathway, moderate confidence), (iii) Covalent binding to proteins leading to skin sensitization (high confidence), and (iv) Aromatase inhibition leading to reproductive dysfunction in fish (high confidence). Partially complete AOPs with unknown molecular initiating events, such as 'Hepatocellular proliferation leading to cancer', were found to be valuable. We demonstrate that scientific confidence in these pathways can be increased though the use of unconventional information (eg, computational identification of potential initiators). AOPs at all levels of confidence can contribute to specific uses. A significant statistical or quantitative relationship between events and/or the adverse outcome relationships is a common characteristic of AOPs, both incomplete and complete, that have specific regulatory uses. For AOPs to be useful in a regulatory context they must be at least as useful as the tools that regulators currently possess, or the techniques currently employed by regulators.

Toward Good Read-Across Practice (GRAP) guidance

Grouping of substances and utilizing read-across of data within those groups represents an important data gap filling technique for chemical safety assessments. Categories/analogue groups are typically developed based on structural similarity and, increasingly often, also on mechanistic (biological) similarity. While read-across can play a key role in complying with legislation such as the European REACH regulation, the lack of consensus regarding the extent and type of evidence necessary to support it often hampers its successful application and acceptance by regulatory authorities. Despite a potentially broad user community, expertise is still concentrated across a handful of organizations and individuals. In order to facilitate the effective use of read-across, this document presents the state of the art, summarizes insights learned from reviewing ECHA published decisions regarding the relative successes/pitfalls surrounding read-across under REACH, and compiles the relevant activities and guidance documents. Special emphasis is given to the available existing tools and approaches, an analysis of ECHA's published final decisions associated with all levels of compliance checks and testing proposals, the consideration and expression of uncertainty, the use of biological support data, and the impact of the ECHA Read-Across Assessment Framework (RAAF) published in 2015.

Supporting read-across using biological data

Read-across, i.e. filling toxicological data gaps by relating to similar chemicals, for which test data are available, is usually done based on chemical similarity. Besides structure and physico-chemical properties, however, biological similarity based on biological data adds extra strength to this process. In the context of developing Good Read-Across Practice guidance, a number of case studies were evaluated to demonstrate the use of biological data to enrich read-across. In the simplest case, chemically similar substances also show similar test results in relevant in vitro assays. This is a well-established method for the read-across of e.g. genotoxicity assays. Larger datasets of biological and toxicological properties of hundreds and thousands of substances become increasingly available enabling big data approaches in read-across studies. Several case studies using various big data sources are described in this paper. An example is given for the US EPA's ToxCast dataset allowing read-across for high quality uterotrophic assays for estrogenic endocrine disruption. Similarly, an example for REACH registration data enhancing read-across for acute toxicity studies is given. A different approach is taken using omics data to establish biological similarity: Examples are given for stem cell models in vitro and short-term repeated dose studies in rats in vivo to support read-across and category formation. These preliminary biological data-driven read-across studies highlight the road to the new generation of read-across approaches that can be applied in chemical safety assessment.

Analysis of publically available skin sensitization data from REACH registrations 2008-2014

The public data on skin sensitization from REACH registrations already included 19,111 studies on skin sensitization in December 2014, making it the largest repository of such data so far (1,470 substances with mouse LLNA, 2,787 with GPMT, 762 with both in vivo and in vitro and 139 with only in vitro data). 21% were classified as sensitizers. The extracted skin sensitization data was analyzed to identify relationships in skin sensitization guidelines, visualize structural relationships of sensitizers, and build models to predict sensitization.

A chemical with molecular weight > 500 Da is generally considered non-sensitizing owing to low bioavailability, but 49 sensitizing chemicals with a molecular weight > 500 Da were found.

A chemical similarity map was produced using PubChem’s 2D Tanimoto similarity metric and Gephi force layout visualization. Nine clusters of chemicals were identified by Blondel’s module recognition algorithm revealing wide module-dependent variation.

Approximately 31% of mapped chemicals are Michael’s acceptors but alone this does not imply skin sensitization. A simple sensitization model using molecular weight and five ToxTree structural alerts showed a balanced accuracy of 65.8% (specificity 80.4%, sensitivity 51.4%), demonstrating that structural alerts have information value.

A simple variant of k-nearest neighbors outperformed the ToxTree approach even at 75% similarity threshold (82% balanced accuracy at 0.95 threshold). At higher thresholds, the balanced accuracy increased. Lower similarity thresholds decrease sensitivity faster than specificity.

This analysis scopes the landscape of chemical skin sensitization, demonstrating the value of large public datasets for health hazard prediction.

How decision analysis can further nanoinformatics

The increase in nanomaterial research has resulted in increased nanomaterial data. The next challenge is to meaningfully integrate and interpret these data for better and more efficient decisions. Due to the complex nature of nanomaterials, rapid changes in technology, and disunified testing and data publishing strategies, information regarding material properties is often illusive, uncertain, and/or of varying quality, which limits the ability of researchers and regulatory agencies to process and use the data. The vision of nanoinformatics is to address this problem by identifying the information necessary to support specific decisions (a top-down approach) and collecting and visualizing these relevant data (a bottom-up approach). Current nanoinformatics efforts, however, have yet to efficiently focus data acquisition efforts on the research most relevant for bridging specific nanomaterial data gaps. Collecting unnecessary data and visualizing irrelevant information are expensive activities that overwhelm decision makers. We propose that the decision analytic techniques of multicriteria decision analysis (MCDA), value of information (VOI), weight of evidence (WOE), and portfolio decision analysis (PDA) can bridge the gap from current data collection and visualization efforts to present information relevant to specific decision needs. Decision analytic and Bayesian models could be a natural extension of mechanistic and statistical models for nanoinformatics practitioners to master in solving complex nanotechnology challenges.