Global analysis of publicly available safety data for 9,801 substances registered under REACH from 2008-2014

VenomPred 2.0: A Novel In Silico Platform for an Extended and Human Interpretable Toxicological Profiling of Small Molecules

The application of artificial intelligence and machine learning (ML) methods is becoming increasingly popular in computational toxicology and drug design; it is considered as a promising solution for assessing the safety profile of compounds, particularly in lead optimization and ADMET studies, and to meet the principles of the 3Rs, which calls for the replacement, reduction, and refinement of animal testing. In this context, we herein present the development of VenomPred 2.0 (http://www.mmvsl.it/wp/venompred2/), the new and improved version of our free of charge web tool for toxicological predictions, which now represents a powerful web-based platform for multifaceted and human-interpretable in silico toxicity profiling of chemicals. VenomPred 2.0 presents an extended set of toxicity endpoints (androgenicity, skin irritation, eye irritation, and acute oral toxicity, in addition to the already available carcinogenicity, mutagenicity, hepatotoxicity, and estrogenicity) that can be evaluated through an exhaustive consensus prediction strategy based on multiple ML models. Moreover, we also implemented a new utility based on the Shapley Additive exPlanations (SHAP) method that allows human interpretable toxicological profiling of small molecules, highlighting the features that strongly contribute to the toxicological predictions in order to derive structural toxicophores.

Novel Strategy to Assess the Neurotoxicity of Organic Solvents Such as Glycol Ethers: Protocol for Combining In Vitro and In Silico Methods With Human-Controlled Exposure Experiments

BACKGROUND

Chemicals are not required to be tested systematically for their neurotoxic potency, although they may contribute to the development of several neurological diseases. The absence of systematic testing may be partially explained by the current Organisation for Economic Co-operation and Development (OECD) Test Guidelines, which rely on animal experiments that are expensive, laborious, and ethically debatable. Therefore, it is important to understand the risks to exposed workers and the general population exposed to domestic products. In this study, we propose a strategy to test the neurotoxicity of solvents using the commonly used glycol ethers as a case study.

OBJECTIVE

This study aims to provide a strategy that can be used by regulatory agencies and industries to rank solvents according to their neurotoxicity and demonstrate the use of toxicokinetic modeling to predict air concentrations of solvents that are below the no observed adverse effect concentrations (NOAECs) for human neurotoxicity determined in in vitro assays.

METHODS

The proposed strategy focuses on a complex 3D in vitro brain model (BrainSpheres) derived from human-induced pluripotent stem cells (hiPSCs). This model is accompanied by in vivo, in vitro, and in silico models for the blood-brain barrier (BBB) and in vitro models for liver metabolism. The data are integrated into a toxicokinetic model. Internal concentrations predicted using this toxicokinetic model are compared with the results from in vivo human-controlled exposure experiments for model validation. The toxicokinetic model is then used in reverse dosimetry to predict air concentrations, leading to brain concentrations lower than the NOAECs determined in the hiPSC-derived 3D brain model. These predictions will contribute to the protection of exposed workers and the general population with domestic exposures.

RESULTS

The Swiss Centre for Applied Human Toxicology funded the project, commencing in January 2021. The Human Ethics Committee approval was obtained on November 16, 2022. Zebrafish experiments and in vitro methods started in February 2021, whereas recruitment of human volunteers started in 2022 after the COVID-19 pandemic-related restrictions were lifted. We anticipate that we will be able to provide a neurotoxicity testing strategy by 2026 and predicted air concentrations for 6 commonly used propylene glycol ethers based on toxicokinetic models incorporating liver metabolism, BBB leakage parameters, and brain toxicity.

CONCLUSIONS

This study will be of great interest to regulatory agencies and chemical industries needing and seeking novel solutions to develop human chemical risk assessments. It will contribute to protecting human health from the deleterious effects of environmental chemicals.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/50300.

Application of read-across methods as a framework for the estimation of emissions from chemical processes

The read-across method is a popular data gap filling technique with developed application for multiple purposes, including regulatory. Within the US Environmental Protection Agency's (US EPA) New Chemicals Program under Toxic Substances Control Act (TSCA), read-across has been widely used, as well as within technical guidance published by the Organization for Economic Co-operation and Development, the European Chemicals Agency, and the European Center for Ecotoxicology and Toxicology of Chemicals for filling chemical toxicity data gaps. Under the TSCA New Chemicals Review Program, US EPA is tasked with reviewing proposed new chemical applications prior to commencing commercial manufacturing within or importing into the United States. The primary goal of this review is to identify any unreasonable human health and environmental risks, arising from environmental releases/emissions during manufacturing and the resulting exposure from these environmental releases. The authors propose the application of read-across techniques for the development and use of a framework for estimating the emissions arising during the chemical manufacturing process. This methodology is to utilize available emissions data from a structurally similar analogue chemical or a group of structurally similar chemicals in a chemical family taking into consideration their physicochemical properties under specified chemical process unit operations and conditions. This framework is also designed to apply existing knowledge of read-across principles previously utilized in toxicity estimation for an analogue or category of chemicals and introduced and extended with a concurrent case study.

ELIXIR and Toxicology: a community in development

Toxicology has been an active research field for many decades, with academic, industrial and government involvement. Modern omics and computational approaches are changing the field, from merely disease-specific observational models into target-specific predictive models. Traditionally, toxicology has strong links with other fields such as biology, chemistry, pharmacology and medicine. With the rise of synthetic and new engineered materials, alongside ongoing prioritisation needs in chemical risk assessment for existing chemicals, early predictive evaluations are becoming of utmost importance to both scientific and regulatory purposes. ELIXIR is an intergovernmental organisation that brings together life science resources from across Europe. To coordinate the linkage of various life science efforts around modern predictive toxicology, the establishment of a new ELIXIR Community is seen as instrumental. In the past few years, joint efforts, building on incidental overlap, have been piloted in the context of ELIXIR. For example, the EU-ToxRisk, diXa, HeCaToS, transQST, and the nanotoxicology community have worked with the ELIXIR TeSS, Bioschemas, and Compute Platforms and activities. In 2018, a core group of interested parties wrote a proposal, outlining a sketch of what this new ELIXIR Toxicology Community would look like. A recent workshop (held September 30th to October 1st, 2020) extended this into an ELIXIR Toxicology roadmap and a shortlist of limited investment-high gain collaborations to give body to this new community. This Whitepaper outlines the results of these efforts and defines our vision of the ELIXIR Toxicology Community and how it complements other ELIXIR activities.

Data-Driven Quantitative Structure-Activity Relationship Modeling for Human Carcinogenicity by Chronic Oral Exposure

Traditional methodologies for assessing chemical toxicity are expensive and time-consuming. Computational modeling approaches have emerged as low-cost alternatives, especially those used to develop quantitative structure-activity relationship (QSAR) models. However, conventional QSAR models have limited training data, leading to low predictivity for new compounds. We developed a data-driven modeling approach for constructing carcinogenicity-related models and used these models to identify potential new human carcinogens. To this goal, we used a probe carcinogen dataset from the US Environmental Protection Agency's Integrated Risk Information System (IRIS) to identify relevant PubChem bioassays. Responses of 25 PubChem assays were significantly relevant to carcinogenicity. Eight assays inferred carcinogenicity predictivity and were selected for QSAR model training. Using 5 machine learning algorithms and 3 types of chemical fingerprints, 15 QSAR models were developed for each PubChem assay dataset. These models showed acceptable predictivity during 5-fold cross-validation (average CCR = 0.71). Using our QSAR models, we can correctly predict and rank 342 IRIS compounds' carcinogenic potentials (PPV = 0.72). The models predicted potential new carcinogens, which were validated by a literature search. This study portends an automated technique that can be applied to prioritize potential toxicants using validated QSAR models based on extensive training sets from public data resources.

Advancing New Approach Methodologies (NAMs) for Tobacco Harm Reduction: Synopsis from the 2021 CORESTA SSPT-NAMs Symposium

New approach methodologies (NAMs) are emerging chemical safety assessment tools consisting of in vitro and in silico (computational) methodologies intended to reduce, refine, or replace (3R) various in vivo animal testing methods traditionally used for risk assessment. Significant progress has been made toward the adoption of NAMs for human health and environmental toxicity assessment. However, additional efforts are needed to expand their development and their use in regulatory decision making. A virtual symposium was held during the 2021 Cooperation Centre for Scientific Research Relative to Tobacco (CORESTA) Smoke Science and Product Technology (SSPT) conference (titled "Advancing New Alternative Methods for Tobacco Harm Reduction"), with the goals of introducing the concepts and potential application of NAMs in the evaluation of potentially reduced-risk (PRR) tobacco products. At the symposium, experts from regulatory agencies, research organizations, and NGOs shared insights on the status of available tools, strengths, limitations, and opportunities in the application of NAMs using case examples from safety assessments of chemicals and tobacco products. Following seven presentations providing background and application of NAMs, a discussion was held where the presenters and audience discussed the outlook for extending the NAMs toxicological applications for tobacco products. The symposium, endorsed by the CORESTA In Vitro Tox Subgroup, Biomarker Subgroup, and NextG Tox Task Force, illustrated common ground and interest in science-based engagement across the scientific community and stakeholders in support of tobacco regulatory science. Highlights of the symposium are summarized in this paper.

U.S. Federal Agency interests and key considerations for new approach methodologies for nanomaterials

Engineered nanomaterials (ENMs) come in a wide array of shapes, sizes, surface coatings, and compositions, and often possess novel or enhanced properties compared to larger sized particles of the same elemental composition. To ensure the safe commercialization of products containing ENMs, it is important to thoroughly understand their potential risks. Given that ENMs can be created in an almost infinite number of variations, it is not feasible to conduct in vivo testing on each type of ENM. Instead, new approach methodologies (NAMs) such as in vitro or in chemico test methods may be needed, given their capacity for higher throughput testing, lower cost, and ability to provide information on toxicological mechanisms. However, the different behaviors of ENMs compared to dissolved chemicals may challenge safety testing of ENMs using NAMs. In this study, member agencies within the Interagency Coordinating Committee on the Validation of Alternative Methods were queried about what types of ENMs are of agency interest and whether there is agency-specific guidance for ENM toxicity testing. To support the ability of NAMs to provide robust results in ENM testing, two key issues in the usage of NAMs, namely dosimetry and interference/bias controls, are thoroughly discussed.

Avoiding Regrettable Substitutions: Green Toxicology for Sustainable Chemistry

Green chemistry seeks to design less hazardous chemicals, but many of the efforts to replace chemicals have resulted in so-called "Regrettable Substitutions", when a chemical with an unknown or unforeseen hazard is used to replace a chemical identified as problematic. Here, we discuss the literature on regrettable substitution and focus on an oft-mentioned case, Bisphenol A, which was replaced with Bisphenol S-and the lessons that can be learned from this history. In particular, we focus on how Green Toxicology can offer a way to make better substitutions.

In vitro reconstructed 3D corneal tissue models for ocular toxicology and ophthalmic drug development

Testing of all manufactured products and their ingredients for eye irritation is a regulatory requirement. In the last two decades, the development of alternatives to the in vivo Draize eye irritation test method has substantially advanced due to the improvements in primary cell isolation, cell culture techniques, and media, which have led to improved in vitro corneal tissue models and test methods. Most in vitro models for ocular toxicology attempt to reproduce the corneal epithelial tissue which consists of 4-5 layers of non-keratinized corneal epithelial cells that form tight junctions, thereby limiting the penetration of chemicals, xenobiotics, and pharmaceuticals. Also, significant efforts have been directed toward the development of more complex three-dimensional (3D) equivalents to study wound healing, drug permeation, and bioavailability. This review focuses on in vitro reconstructed 3D corneal tissue models and their utilization in ocular toxicology as well as their application to pharmacology and ophthalmic research. Current human 3D corneal epithelial cell culture models have replaced in vivo animal eye irritation tests for many applications, and substantial validation efforts are in progress to verify and approve alternative eye irritation tests for widespread use. The validation of drug absorption models and further development of models and test methods for many ophthalmic and ocular disease applications is required.

Quantifying Uncertainty in Ecotoxicological Risk Assessment: MUST, a Modular Uncertainty Scoring Tool

Whether conducting a risk, hazard, or alternatives assessment, one invariably struggles with the task of reconciling multiple available values of toxicological thresholds into a single outcome. When combining multiple pieces of evidence from many different sources, it is important to consider the role of data uncertainty. Uncertainty is inherent to all scientific data. However, in toxicological assessments, controversies and uncertainties are typically understated; they lack methodological transparency; or they poorly integrate qualitative and quantitative sources of information. Similarly, in model development, data curation is rarely performed with sufficient rigor, particularly when applying big data statistics. To overcome the hurdles of a decision process that must reconcile divergent data, we developed an uncertainty scoring tool that can be trained to reproduce specific decision-making paradigms and ensure consistency in the practitioner's judgment across complex scenarios. While designed to aid with ecotoxicological assessments and predictive model development, the tool's applicability extends to any decision-making process that calls for synthesis of incongruent data. Here, we highlight the development process, as well as demonstrate the method's utility in several prototypical ecotoxicological case studies.

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.

Use of LUCS (Light-Up Cell System) as an alternative live cell method to predict human acute oral toxicity

LUCS (Light-Up Cell System) is a new live cell test that allows assessment of a cell's homeostasis and its alteration by a toxic agent. To evaluate the effectiveness of LUCS as an alternative test method for acute oral toxicity, we compared EC₅₀s determined in HepG2 cells treated with 53 chemicals selected from the ACuteTox EU database with corresponding human blood LC₅₀s derived from human acute poisoning cases. Linear regression analysis showed that LUCS results predict human data to 69 %. Rodent oral LD₅₀s and LUCS EC₅₀s were then correlated to human LC₅₀s using shared data sets. Linear regression analyses comparing LUCS and animal data clearly showed that LUCS always predicts human toxicity better than animal data do. These successful prediction values prompted us to simplify the LUCS test, adapting it to regulatory and high throughput applications, resulting in a new protocol with consistent dose-response profiles and EC₅₀s. This study demonstrates that the LUCS test method could be relevant for assessing human acute oral toxicity with a simplified protocol adapted to commercially available fluorescence readers. We suggest that this new alternative method can be used for acute systemic toxicity testing in combination with other tests under European REACH and other regulations, wherever pertinent alternative methods are still lacking.

Predicting toxicity of chemicals: software beats animal testing

We created earlier a large machine-readable database of 10,000 chemicals and 800,000 associated studies by natural language processing of the public parts of Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) registrations until December 2014. This database was used to assess the reproducibility of the six most frequently used Organisation for Economic Co-operation and Development (OECD) guideline tests. These tests consume 55% of all animals in safety testing in Europe, i.e. about 600,000 animals. With 350-750 chemicals with multiple results per test, reproducibility (balanced accuracy) was 81% and 69% of toxic substances were found again in a repeat experiment (sensitivity 69%). Inspired by the increasingly used read-across approach, we created a new type of QSAR, which is based on similarity of chemicals and not on chemical descriptors. A landscape of the chemical universe using 10 million structures was calculated, when based on Tanimoto indices similar chemicals are close and dissimilar chemicals far from each other. This allows placing any chemical of interest into the map and evaluating the information available for surrounding chemicals. In a data fusion approach, in which 74 different properties were taken into consideration, machine learning (random forest) allowed a fivefold cross-validation for 190,000 (non-) hazard labels of chemicals for which nine hazards were predicted. The balanced accuracy of this approach was 87% with a sensitivity of 89%. Each prediction comes with a certainty measure based on the homogeneity of data and distance of neighbours. Ongoing developments and future opportunities are discussed.

Leveraging heterogeneous data from GHS toxicity annotations, molecular and protein target descriptors and Tox21 assay readouts to predict and rationalise acute toxicity

Despite the increasing knowledge in both the chemical and biological domains the assimilation and exploration of heterogeneous datasets, encoding information about the chemical, bioactivity and phenotypic properties of compounds, remains a challenge due to requirement for overlap between chemicals assayed across the spaces. Here, we have constructed a novel dataset, larger than we have used in prior work, comprising 579 acute oral toxic compounds and 1427 non-toxic compounds derived from regulatory GHS information, along with their corresponding molecular and protein target descriptors and qHTS in vitro assay readouts from the Tox21 project. We found no clear association between the results of a FAFDrugs4 toxicophore screen and the acute oral toxicity classifications for our compound set; and a screen using a subset of the ToxAlerts toxicophores was also of limited utility, with only slight enrichment toward the toxic set (odds ratio of 1.48). We then investigated to what degree toxic and non-toxic compounds could be separated in each of the spaces, to compare their potential contribution to further analyses. Using an LDA projection, we found the largest degree of separation using chemical descriptors (Cohen’s d of 1.95) and the lowest degree of separation between toxicity classes using qHTS descriptors (Cohen’s d of 0.67). To compare the predictivity of the feature spaces for the toxicity endpoint, we next trained Random Forest (RF) acute oral toxicity classifiers on either molecular, protein target and qHTS descriptors. RFs trained on molecular and protein target descriptors were most predictive, with ROC AUC values of 0.80–0.92 and 0.70–0.85, respectively, across three test sets. RFs trained on both chemical and protein target descriptors combined exhibited similar predictive performance to the single-domain models (ROC AUC of 0.80–0.91). Model interpretability was improved by the inclusion of protein target descriptors, which allow the identification of specific targets (e.g. Retinal dehydrogenase) with literature links to toxic modes of action (e.g. oxidative stress). The dataset compiled in this study has been made available for future application.

Advancing Computational Toxicology in the Big Data Era by Artificial Intelligence: Data-Driven and Mechanism-Driven Modeling for Chemical Toxicity

In 2016, the Frank R. Lautenberg Chemical Safety for the 21st Century Act became the first US legislation to advance chemical safety evaluations by utilizing novel testing approaches that reduce the testing of vertebrate animals. Central to this mission is the advancement of computational toxicology and artificial intelligence approaches to implementing innovative testing methods. In the current big data era, the terms volume (amount of data), velocity (growth of data), and variety (the diversity of sources) have been used to characterize the currently available chemical, in vitro, and in vivo data for toxicity modeling purposes. Furthermore, as suggested by various scientists, the variability (internal consistency or lack thereof) of publicly available data pools, such as PubChem, also presents significant computational challenges. The development of novel artificial intelligence approaches based on public massive toxicity data is urgently needed to generate new predictive models for chemical toxicity evaluations and make the developed models applicable as alternatives for evaluating untested compounds. In this procedure, traditional approaches (e.g., QSAR) purely based on chemical structures have been replaced by newly designed data-driven and mechanism-driven modeling. The resulting models realize the concept of adverse outcome pathway (AOP), which can not only directly evaluate toxicity potentials of new compounds, but also illustrate relevant toxicity mechanisms. The recent advancement of computational toxicology in the big data era has paved the road to future toxicity testing, which will significantly impact on the public health.

How in silico and QSAR approaches can increase confidence in environmental hazard and risk assessment

In silico methods are typically underrated in the current risk assessment paradigm, as evidenced by the recent document from the European Chemicals Agency (ECHA) on animal alternatives, in which quantitative structure-activity relationships (QSARs) were practically used only as a last resort. Their primary use is still to provide supporting evidence for read-across strategies or to add credence to experimental results of unknown or limited validity (old studies, studies without good laboratory practices [GLPs], limited information reported, etc.) in hazard assessment, but under the pressure of increasing burdens of testing, industry and regulators alike are at last warming to them. Nevertheless, their true potential for data-gap filling and for resolving sticking points in risk assessment methodology and beyond has yet to be recognized. We postulate that it is possible to go beyond the level of simply increasing confidence to the point of using in silico approaches to accurately predict results that cannot be resolved analytically. For example, under certain conditions it is possible to obtain meaningful results by in silico extrapolation for tests that would be technically impossible to conduct in the laboratory or at least extremely challenging to obtain reliable results. The following and other concepts are explored in this article: the mechanism of action (MechoA) of the substance should be determined, as an aid verifying that the QSAR model is applicable to the substance under review; accurate QSARs should be built with high-quality data that were not only curated but also validated with expert judgment; although a rule of thumb for acute to chronic ratios appears applicable for nonpolar narcotics, it seems unlikely that a "one-value-fits-all" answer exists for other MechoAs; a holistic approach to QSARs can be employed (via reverse engineering) to help validate or invalidate an experimental endpoint value on the basis of multiple experimental studies. Integr Environ Assess Manag 2019;15:40-50. © 2018 SETAC.

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.

Machine Learning of Toxicological Big Data Enables Read-Across Structure Activity Relationships (RASAR) Outperforming Animal Test Reproducibility

Earlier we created a chemical hazard database via natural language processing of dossiers submitted to the European Chemical Agency with approximately 10 000 chemicals. We identified repeat OECD guideline tests to establish reproducibility of acute oral and dermal toxicity, eye and skin irritation, mutagenicity and skin sensitization. Based on 350-700+ chemicals each, the probability that an OECD guideline animal test would output the same result in a repeat test was 78%-96% (sensitivity 50%-87%). An expanded database with more than 866 000 chemical properties/hazards was used as training data and to model health hazards and chemical properties. The constructed models automate and extend the read-across method of chemical classification. The novel models called RASARs (read-across structure activity relationship) use binary fingerprints and Jaccard distance to define chemical similarity. A large chemical similarity adjacency matrix is constructed from this similarity metric and is used to derive feature vectors for supervised learning. We show results on 9 health hazards from 2 kinds of RASARs-"Simple" and "Data Fusion". The "Simple" RASAR seeks to duplicate the traditional read-across method, predicting hazard from chemical analogs with known hazard data. The "Data Fusion" RASAR extends this concept by creating large feature vectors from all available property data rather than only the modeled hazard. Simple RASAR models tested in cross-validation achieve 70%-80% balanced accuracies with constraints on tested compounds. Cross validation of data fusion RASARs show balanced accuracies in the 80%-95% range across 9 health hazards with no constraints on tested compounds.

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.

3S - Systematic, systemic, and systems biology and toxicology

A biological system is more than the sum of its parts - it accomplishes many functions via synergy. Deconstructing the system down to the molecular mechanism level necessitates the complement of reconstructing functions on all levels, i.e., in our conceptualization of biology and its perturbations, our experimental models and computer modelling. Toxicology contains the somewhat arbitrary subclass "systemic toxicities"; however, there is no relevant toxic insult or general disease that is not systemic. At least inflammation and repair are involved that require coordinated signaling mechanisms across the organism. However, the more body components involved, the greater the challenge to reca-pitulate such toxicities using non-animal models. Here, the shortcomings of current systemic testing and the development of alternative approaches are summarized. We argue that we need a systematic approach to integrating existing knowledge as exemplified by systematic reviews and other evidence-based approaches. Such knowledge can guide us in modelling these systems using bioengineering and virtual computer models, i.e., via systems biology or systems toxicology approaches. Experimental multi-organ-on-chip and microphysiological systems (MPS) provide a more physiological view of the organism, facilitating more comprehensive coverage of systemic toxicities, i.e., the perturbation on organism level, without using substitute organisms (animals). The next challenge is to establish disease models, i.e., micropathophysiological systems (MPPS), to expand their utility to encompass biomedicine. Combining computational and experimental systems approaches and the chal-lenges of validating them are discussed. The suggested 3S approach promises to leverage 21st century technology and systematic thinking to achieve a paradigm change in studying systemic effects.

The Threshold of Toxicological Concern for prenatal developmental toxicity in rats and rabbits

The Threshold Toxicological Concern (TTC) is based on the concept that in absence of experimental data reasonable assurance of safety can be given if exposure is sufficiently low. Using the REACH database the low 5th percentile of the NO(A)EL distribution, for prenatal developmental toxicity (OECD guideline 414) was determined. For rats, (434 NO(A)ELs values) for maternal toxicity, this value was 10 mg/kg-bw/day. For developmental toxicity (469 NO(A)ELs): 13 mg/kg-bw/day. For rabbits, (100 NO(A)ELs), the value for maternal toxicity was 4 mg/kg-bw/day, for developmental toxicity, (112 NO(A)EL values): 10 mg/kg-bw/day. The maternal organism may thus be slightly more sensitive than the fetus. Combining REACH- (industrial chemicals) and published BASF-data (mostly agrochemicals), 537 unique compounds with NO(A)EL values for developmental toxicity in rats and 150 in rabbits were evaluated. The low 5th percentile NO(A)EL for developmental toxicity in rats was 10 mg/kg-bw/day and 9.5 mg/kg-bw/day for rabbits. Using an assessment factor of 100, a TTC value for developmental toxicity of 100 μg/kg-bw/day for rats and 95 μg/kg-bw/day for rabbits is calculated. These values could serve as guidance whether or not to perform an animal experiment, if exposure is sufficiently low. In emergency situations this value may be useful for a first tier risk assessment.

Green Toxicology: a strategy for sustainable chemical and material development

Green Toxicology refers to the application of predictive toxicology in the sustainable development and production of new less harmful materials and chemicals, subsequently reducing waste and exposure. Built upon the foundation of "Green Chemistry" and "Green Engineering", "Green Toxicology" aims to shape future manufacturing processes and safe synthesis of chemicals in terms of environmental and human health impacts. Being an integral part of Green Chemistry, the principles of Green Toxicology amplify the role of health-related aspects for the benefit of consumers and the environment, in addition to being economical for manufacturing companies. Due to the costly development and preparation of new materials and chemicals for market entry, it is no longer practical to ignore the safety and environmental status of new products during product development stages. However, this is only possible if toxicologists and chemists work together early on in the development of materials and chemicals to utilize safe design strategies and innovative in vitro and in silico tools. This paper discusses some of the most relevant aspects, advances and limitations of the emergence of Green Toxicology from the perspective of different industry and research groups. The integration of new testing methods and strategies in product development, testing and regulation stages are presented with examples of the application of in silico, omics and in vitro methods. Other tools for Green Toxicology, including the reduction of animal testing, alternative test methods, and read-across approaches are also discussed.

Evaluating the Impact of the U.S. National Toxicology Program: A Case Study on Hexavalent Chromium

BACKGROUND

Evaluating the impact of federally funded research with a broad, methodical, and objective approach is important to ensure that public funds advance the mission of federal agencies.

OBJECTIVES

We aimed to develop a methodical approach that would yield a broad assessment of National Toxicology Program's (NTP's) effectiveness across multiple sectors and demonstrate the utility of the approach through a case study.

METHODS

A conceptual model was developed with defined activities, outputs (products), and outcomes (proximal, intermediate, distal) and applied retrospectively to NTP's research on hexavalent chromium (CrVI). Proximal outcomes were measured by counting views of and requests for NTP's products by external stakeholders. Intermediate outcomes were measured by bibliometric analysis. Distal outcomes were assessed through Web and LexisNexis searches for documents related to legislation or regulation changes.

RESULTS

The approach identified awareness of NTP's work on CrVI by external stakeholders (proximal outcome) and citations of NTP's research in scientific publications, reports, congressional testimonies, and legal and policy documents (intermediate outcome). NTP's research was key to the nation's first-ever drinking water standard for CrVI adopted by California in 2014 (distal outcome). By applying this approach to a case study, the utility and limitations of the approach were identified, including challenges to evaluating the outcomes of a research program.

CONCLUSIONS

This study identified a broad and objective approach for assessing NTP's effectiveness, including methodological needs for more thorough and efficient impact assessments in the future. Citation: Xie Y, Holmgren S, Andrews DMK, Wolfe MS. 2017. Evaluating the impact of the U.S. National Toxicology Program: a case study on hexavalent chromium. Environ Health Perspect 125:181-188; http://dx.doi.org/10.1289/EHP21.

Alternative approaches for identifying acute systemic toxicity: Moving from research to regulatory testing

Acute systemic toxicity testing provides the basis for hazard labeling and risk management of chemicals. A number of international efforts have been directed at identifying non-animal alternatives for in vivo acute systemic toxicity tests. A September 2015 workshop, Alternative Approaches for Identifying Acute Systemic Toxicity: Moving from Research to Regulatory Testing, reviewed the state-of-the-science of non-animal alternatives for this testing and explored ways to facilitate implementation of alternatives. Workshop attendees included representatives from international regulatory agencies, academia, nongovernmental organizations, and industry. Resources identified as necessary for meaningful progress in implementing alternatives included compiling and making available high-quality reference data, training on use and interpretation of in vitro and in silico approaches, and global harmonization of testing requirements. Attendees particularly noted the need to characterize variability in reference data to evaluate new approaches. They also noted the importance of understanding the mechanisms of acute toxicity, which could be facilitated by the development of adverse outcome pathways. Workshop breakout groups explored different approaches to reducing or replacing animal use for acute toxicity testing, with each group crafting a roadmap and strategy to accomplish near-term progress. The workshop steering committee has organized efforts to implement the recommendations of the workshop participants.

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.

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.

Analysis of public oral toxicity data from REACH registrations 2008-2014

The European Chemicals Agency, ECHA, made available a total of 13,832 oral toxicity studies for 8,568 substances up to December 2014. 75% of studies were from the retired OECD Test Guideline 401 (11% TG 420, 11% TG 423 and 1.5% TG 425). Concordance across guidelines, evaluated by comparing LD50 values ≥ 2000 or < 2000 mg/kg body weight from chemicals tested multiple times between different guidelines, was at least 75% and for their own repetition more than 90%. In 2009, Bulgheroni et al. created a simple model for predicting acute oral toxicity using no observed adverse effect levels (NOAEL) from 28-day repeated dose toxicity studies in rats. This was reproduced here for 1,625 substances. In 2014, Taylor et al. suggested no added value of the 90-day repeated dose oral toxicity test given the availability of a low 28-day study with some constraints. We confirm that the 28-day NOAEL is predictive (albeit imperfectly) of 90-day NOAELs, however, the suggested constraints did not affect predictivity. 1,059 substances with acute oral toxicity data (268 positives, 791 negatives, all Klimisch score 1) were used for modeling: The Chemical Development Kit was used to generate 27 molecular descriptors and a similarity-informed multilayer perceptron showing 71% sensitivity and 72% specificity. Additionally, the k-nearest neighbors (KNN) algorithm indicated that similarity-based approaches alone may be poor predictors of acute oral toxicity, but can be used to inform the multilayer perceptron model, where this was the feature with highest information value.

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.

Analysis of Draize eye irritation testing and its prediction by mining publicly available 2008-2014 REACH data

Public data from ECHA online dossiers on 9,801 substances encompassing 326,749 experimental key studies and additional information on classification and labeling were made computable. Eye irritation hazard, for which the rabbit Draize eye test still represents the reference method, was analyzed. Dossiers contained 9,782 Draize eye studies on 3,420 unique substances, indicating frequent retesting of substances. This allowed assessment of the test's reproducibility test based on all substances tested more than once. There was a 10% chance of a non-irritant evaluation given after a prior severe-irritant result as given by UN GHS classification criteria. The most reproducible outcomes were the results negative (94% reproducible) and severe eye irritant (73% reproducible). To evaluate whether other GHS categorizations predict eye irritation we built a dataset of 5,629 substances (1,931 'irritant' and 3,698 'non-irritant'). The two best decision trees with up to three other GHS classifications resulted in balanced accuracies of 68% and 73%, i.e., in the rank order of the Draize rabbit eye test itself, but both use inhalation toxicity data ("May cause respiratory irritation"), which is not typically available. Next, a dataset of 929 substances with at least one Draize study was mapped to PubChem to compute chemical similarity using 2D conformational fingerprints and Tanimoto similarity. Using a minimum similarity of 0.7 and simple classification by the closest chemical neighbor resulted in balanced accuracy from 73% over 737 substances to 100% at a threshold of 0.975 over 41 substances. This represents a strong support of read-across and (Q)SAR approaches in this area.