Multi-tissue proteogenomic analysis for mechanistic toxicology studies in non-model species.

New approach methodologies (NAM), including omics and in vitro approaches, are contributing to the implementation of 3R (reduction, refinement and replacement) strategies in regulatory science and risk assessment. In this study, we present an integrative transcriptomics and proteomics analysis workflow for the validation and revision of complex fish genomes and demonstrate how proteogenomics expression matrices can be used to support multi-level omics data integration in non-model species in vivo and in vitro. Using Atlantic salmon as an example, we constructed proteogenomic databases from publicly available transcriptomic data and in-house generated RNA-Seq and LC-MS/MS data. Our analysis identified ∼80,000 peptides, providing direct evidence of translation for over 40,000 RefSeq structures. The data also highlighted 183 co-located peptide groups that supported a single transcript each, and in each case, either corrected a previous annotation, supported Ensembl annotations not present in RefSeq, or identified novel previously unannotated genes. Proteogenomics data-derived expression matrices revealed distinct profiles for the different tissue types analyzed. Focusing on proteins involved in defense against xenobiotics, we detected distinct expression patterns across different salmon tissues and observed homology in the expression of chemical defense proteins between in vivo and in vitro liver systems. Our study demonstrates the potential of proteogenomic analyses in extending our understanding of complex fish genomes and provides an advanced bioinformatic toolkit to support the further development of NAMs and their application in regulatory science and (eco)toxicological studies of non-model species.

Establishing allometric relationships between microsomal protein and cytochrome P450 content with body weight in vertebrate species.

Data from in vitro studies are routinely used to estimate in vivo hepatic clearance of chemicals and this information is needed to parameterise physiologically based kinetic models. Such clearance data can be obtained from laboratory experiments using liver microsomes, hepatocytes, precision-cut liver slices or recombinant enzymes. Irrespective of the selected test system, scaling factors are required to convert the in vitro measured intrinsic clearance to a whole liver intrinsic clearance. Scaling factors such as the hepatic microsomal protein per gram of liver and/or the amount of cytochrome P450 per hepatocyte provide a means to calculate the whole liver intrinsic clearance. Here, a database from the peer-reviewed literature has been developed and provides quantitative metrics on microsomal protein (MP) and cytochrome P450 contents in vertebrate orders namely amphibians, mammals, birds, fish and reptiles. This database allows to address allometric relationships between body weight and MP content, and body weight and cytochrome P450 content. A total of 85 and 74 vertebrate species were included to assess the relationships between log10 body weight versus log10 MP, and between log10 body weight and log10 cytochrome P450 content, respectively. The resulting slopes range from 0.76 to 1.45 in a range of vertebrate species. Such data-driven allometric relationships can be used to estimate the MP content necessary for in vitro to in vivo extrapolation of in vitro clearance data. Future work includes applications of these relationships for different vertebrate taxa using quantitative in vitro to in vivo extrapolation models coupled to physiologically based kinetic models using chemicals of relevance as case studies including pesticides, contaminants and feed additives.

Quasi-SMILES: Self-consistent models for toxicity of organic chemicals to tadpoles.

Simplified molecular input-line entry systems (SMILES) are the representation of the molecular structure that can be used to establish quantitative structure-property/activity relationships (QSPRs/QSARs) for various endpoints expressed as mathematical functions of the molecular architecture. Quasi-SMILES is extending the traditional SMILES by means of additional symbols that reflect experimental conditions. Using the quasi-SMILES models of toxicity to tadpoles gives the possibility to build up models by taking into account the time of exposure. Toxic effects of experimental situations expressed via 188 quasi-SMILES (the negative logarithm of molar concentrations which lead to lethal 50% tadpoles effected during 12 h, 24 h, 48 h, 72 h, and 96 h) were modelled with good results (the average determination coefficient for the validation sets is about 0.97). In this way, we developed new models for this amphibian endpoint, which is poorly studied.

Human Variability in Carboxylesterases and carboxylesterase-related Uncertainty Factors for Chemical Risk Assessment.

Carboxylesterases (CES) are an important class of enzymes involved in the hydrolysis of a range of chemicals and show large inter-individual variability in vitro. An extensive literature search was performed to identify in vivo probe substrates for CES1 and CES2 together with their protein content and enzymatic activity. Human pharmacokinetic (PK) data on Cmax, clearance, and AUC were extracted from 89 publications and Bayesian meta-analysis was performed using a hierarchical model to derive CES-related variability distributions and related uncertainty factors (UF). The CES-related variability indicated that 97.5% of healthy adults are covered by the kinetic default UF (3.16), except for clopidogrel and dabigatran etexilate. Clopidogrel is metabolised for a small amount by the polymorphic CYP2C19, which can have an impact on the overall pharmacokinetics, while the variability seen for dabigatran etexilate might be due to differences in the absorption, since this can be influenced by food intake. The overall CES-related variability was moderate to high in vivo (

Human variability in polymorphic CYP2D6 metabolism: Implications for the risk assessment of chemicals in food and emerging designer drugs.

The major human cytochrome P450 CYP2D6 isoform enzyme plays important roles in the liver and in the brain with regards to xenobiotic metabolism. Xenobiotics as CYP2D6 substrates include a whole range of pharmaceuticals, pesticides and plant alkaloids to cite but a few. In addition, a number of endogenous compounds have been shown to be substrates of CYP2D6 including trace amines in the brain such as tyramine and 5-methoxytryptamine as well as anandamide and progesterone. Because of the polymorphic nature of CYP2D6, considerable inter-phenotypic and inter-ethnic differences in the pharmaco/toxicokinetics (PK/TK) and metabolism of CYP2D6 substrates exist with potential consequences on the pharmacology and toxicity of chemicals. Here, large extensive literature searches have been performed to collect PK data from published human studies for a wide range of pharmaceutical probe substrates and investigate human variability in CYP2D6 metabolism. The computed kinetic parameters resulted in the largest open source database, quantifying inter-phenotypic differences for the kinetics of CYP2D6 probe substrates in Caucasian and Asian populations, to date. The database is available in supplementary material (CYPD6 DB) and EFSA knowledge junction (DOI to added). Subsequently, meta-analyses using a hierarchical Bayesian model for markers of chronic oral exposure (oral clearance, area under the plasma concentration time curve) and acute oral exposure (maximum plasma concentration (Cmax) provided estimates of inter-phenotypic differences and CYP2D6-related uncertainty factors (UFs) for chemical risk assessment in Caucasian and Asian populations classified as ultra-rapid (UM), extensive (EMs), intermediate (IMs) and poor metabolisers (PMs). The model allowed the integration of inter-individual (i.e. inter-phenotypic and inter-ethnic), inter-compound and inter-study variability together with uncertainty in each PK parameter. Key findings include 1. Higher frequencies of PMs in Caucasian populations compared to Asian populations (>8% vs 1-2%) for which EM and IM were the most frequent phenotype. 2. Large inter-phenotypic differences in PK parameters for Caucasian EMs (coefficients of variation (CV) > 50%) compared with Caucasian PMs and Asian EMs and IMs (i.e CV < 40%). 3. Inter-phenotypic PK differences between EMs and PMs in Caucasian populations increase with the quantitative contribution of CYP2D6 for the metabolism (fm) for a range of substrates (fmCYP2D6 range: 20-95% of dose) (range: 1-54) to a much larger extent than those for Asian populations (range: 1-4). 4. Exponential meta-regressions between FmCYP2D6 in EMs and inter-phenotypic differences were also shown to differ between Caucasian and Asian populations as well as CYP2D6-related UFs. Finally, implications of these results for the risk assessment of food chemicals and emerging designer drugs of public health concern, as CYP2D6 substrates, are highlighted and include the integration of in vitro metabolism data and CYP2D6-variability distributions for the development of quantitative in vitro in vivo extrapolation models.

EFSA's OpenFoodTox: An open source toxicological database on chemicals in food and feed and its future developments.

Since its creation in 2002, the European Food Safety Authority (EFSA) has produced risk assessments for over 5000 substances in >2000 Scientific Opinions, Statements and Conclusions through the work of its Scientific Panels, Units and Scientific Committee. OpenFoodTox is an open source toxicological database, available both for download and data visualisation which provides data for all substances evaluated by EFSA including substance characterisation, links to EFSA's outputs, applicable legislations regulations, and a summary of hazard identification and hazard characterisation data for human health, animal health and ecological assessments. The database has been structured using OECD harmonised templates for reporting chemical test summaries (OHTs) to facilitate data sharing with stakeholders with an interest in chemical risk assessment, such as sister agencies, international scientific advisory bodies, and others. This manuscript provides a description of OpenFoodTox including data model, content and tools to download and search the database. Examples of applications of OpenFoodTox in chemical risk assessment are discussed including new quantitative structure-activity relationship (QSAR) models, integration into tools (OECD QSAR Toolbox and AMBIT-2.0), assessment of environmental footprints and testing of threshold of toxicological concern (TTC) values for food related compounds. Finally, future developments for OpenFoodTox 2.0 include the integration of new properties, such as physico-chemical properties, exposure data, toxicokinetic information; and the future integration within in silico modelling platforms such as QSAR models and physiologically-based kinetic models. Such structured in vivo, in vitro and in silico hazard data provide different lines of evidence which can be assembled, weighed and integrated using harmonised Weight of Evidence approaches to support the use of New Approach Methodologies (NAMs) in chemical risk assessment and the reduction of animal testing.

Metabolism and pharmacokinetics of pharmaceuticals in cats (Felix sylvestris catus) and implications for the risk assessment of feed additives and contaminants.

In animal health risk assessment, hazard characterisation of feed additives has been often using the default uncertainty factor (UF) of 100 to translate a no-observed-adverse-effect level in test species (rat, mouse, dog, rabbit) to a 'safe' level of chronic exposure in farm and companion animal species. Historically, both 10-fold factors have been further divided to include chemical-specific data in both dimensions when available. For cats (Felis Sylvestris catus), an extra default UF of 5 is applied due to the species' deficiency in particularly glucuronidation and glycine conjugation. This paper aims to assess the scientific basis and validity of the UF for inter-species differences in kinetics (4.0) and the extra UF applied for cats through a comparison of kinetic parameters between rats and cats for 30 substrates of phase I and phase II metabolism. When the parent compound undergoes glucuronidation the default factor of 4.0 is exceeded, with exceptions for zidovudine and S-carprofen. Compounds that were mainly renally excreted did not exceed the 4.0-fold default. Mixed results were obtained for chemicals which are metabolised by CYP3A in rats. When chemicals were administered intravenously the 4.0-fold default was not exceeded with the exception of clomipramine, lidocaine and alfentanil. The differences seen after oral administration might be due to differences in first-pass metabolism and bioavailability. Further work is needed to further characterise phase I, phase II enzymes and transporters in cats to support the development of databases and in silico models to support hazard characterisation of chemicals particularly for feed additives.

Human variability in isoform-specific UDP-glucuronosyltransferases: markers of acute and chronic exposure, polymorphisms and uncertainty factors.

UDP-glucuronosyltransferases (UGTs) are involved in phase II conjugation reactions of xenobiotics and differences in their isoform activities result in interindividual kinetic differences of UGT probe substrates. Here, extensive literature searches were performed to identify probe substrates (14) for various UGT isoforms (UGT1A1, UGT1A3, UGT1A4, UGT1A6, UGT1A9, UGT2B7 and UGT2B15) and frequencies of human polymorphisms. Chemical-specific pharmacokinetic data were collected in a database to quantify interindividual differences in markers of acute (Cmax) and chronic (area under the curve, clearance) exposure. Using this database, UGT-related uncertainty factors were derived and compared to the default factor (i.e. 3.16) allowing for interindividual differences in kinetics. Overall, results show that pharmacokinetic data are predominantly available for Caucasian populations and scarce for other populations of different geographical ancestry. Furthermore, the relationships between UGT polymorphisms and pharmacokinetic parameters are rarely addressed in the included studies. The data show that UGT-related uncertainty factors were mostly below the default toxicokinetic uncertainty factor of 3.16, with the exception of five probe substrates (1-OH-midazolam, ezetimibe, raltegravir, SN38 and trifluoperazine), with three of these substrates being metabolised by the polymorphic isoform 1A1. Data gaps and future work to integrate UGT-related variability distributions with in vitro data to develop quantitative in vitro-in vivo extrapolations in chemical risk assessment are discussed.

Human variability in influx and efflux transporters in relation to uncertainty factors for chemical risk assessment.

Transporters are divided into the ABC and SLC super-families, mediating the cellular efflux and influx of various xenobiotic and endogenous substrates. Here, an extensive literature search was performed to identify in vivo probe substrates for P-gp, BCRP and OAT1/3. For other transporters (e.g. OCT, OATP), no in vivo probe substrates could be identified from the available literature. Human kinetic data (Cmax, clearance, AUC) were extracted from 142 publications and Bayesian meta-analyses were performed using a hierarchical model to derive variability distributions and related uncertainty factors (UFs). For P-gp, human variability indicated that the kinetic default UF (3.16) would cover over 97.5% of healthy individuals, when considering the median value, while the upper confidence interval is exceeded. For BCRP and OAT1/3 human variability indicated that the default kinetic UF would not be exceeded while considering the upper confidence interval. Although limited kinetic data on transporter polymorphisms were available, inter-phenotypic variability for probe substrates was reported, which may indicate that the current default kinetic UF may be insufficient to cover such polymorphisms. Overall, it is recommended to investigate human genetic polymorphisms across geographical ancestry since they provide more robust surrogate measures of genetic differences compared to geographical ancestry alone. This analysis is based on pharmaceutical probe substrates which are often eliminated relatively fast from the human body. The transport of environmental contaminants and food-relevant chemicals should be investigated to broaden the chemical space of this analysis and assess the likelihood of potential interactions with transporters at environmental concentrations.

Bayesian meta-analysis of inter-phenotypic differences in human serum paraoxonase-1 activity for chemical risk assessment.

Human variability in paraoxonase-1 (PON1) activities is driven by genetic polymorphisms that affect the internal dose of active oxons of organophosphorus (OP) insecticides. Here, an extensive literature search has been performed to collect human genotypic frequencies (i.e. L55M, Q192R, and C-108T) in subgroups from a range of geographical ancestry and PON1 activities in three probe substrates (paraoxon, diazoxon and phenyl acetate). Bayesian meta-analyses were performed to estimate variability distributions for PON1 activities and PON1-related uncertainty factors (UFs), while integrating quantifiable sources of inter-study, inter-phenotypic and inter-individual differences. Inter-phenotypic differences were quantified using the population with high PON1 activity as the reference group. Results from the meta-analyses provided PON1 variability distributions and these can be implemented in generic physiologically based kinetic models to develop quantitative in vitro in vivo extrapolation models. PON1-related UFs in the Caucasian population were above the default toxicokinetic UF of 3.16 for two specific genotypes namely -108CC using diazoxon as probe substrate and, -108CT, -108TT, 55MM and 192QQ using paraoxon as probe substrate. However, integration of PON1 genotypic frequencies and activity distributions showed that all UFs were within the default toxicokinetic UF. Quantitative inter-individual differences in PON1 activity are important for chemical risk assessment particularly with regards to the potential sensitivity to organophosphates' toxicity.

An open source physiologically based kinetic model for the chicken (Gallus gallus domesticus): Calibration and validation for the prediction residues in tissues and eggs.

Xenobiotics from anthropogenic and natural origin enter animal feed and human food as regulated compounds, environmental contaminants or as part of components of the diet. After dietary exposure, a chemical is absorbed and distributed systematically to a range of organs and tissues, metabolised, and excreted. Physiologically based kinetic (PBK) models have been developed to estimate internal concentrations from external doses. In this study, a generic multi-compartment PBK model was developed for chicken. The PBK model was implemented for seven compounds (with log Kow range -1.37-6.2) to quantitatively link external dose and internal dose for risk assessment of chemicals. Global sensitivity analysis was performed for a hydrophilic and a lipophilic compound to identify the most sensitive parameters in the PBK model. Model predictions were compared to measured data according to dataset-specific exposure scenarios. Globally, 71% of the model predictions were within a 3-fold change of the measured data for chicken and only 7% of the PBK predictions were outside a 10-fold change. While most model input parameters still rely on in vivo experiments, in vitro data were also used as model input to predict internal concentration of the coccidiostat monensin. Future developments of generic PBK models in chicken and other species of relevance to animal health risk assessment are discussed.

Generic physiologically based kinetic modelling for farm animals: Part II. Predicting tissue concentrations of chemicals in swine, cattle, and sheep.

The development of three generic multi-compartment physiologically based kinetic (PBK) models is described for farm animal species, i.e. cattle, sheep, and swine. The PBK models allow one to quantitatively link external dose and internal dose for risk assessment of chemicals relevant to food and feed safety. Model performance is illustrated by predicting tissue concentrations of melamine and oxytetracycline and validated through comparison with measured data. Overall, model predictions were reliable with 71% of predictions within a 3-fold of the measured data for all three species and only 6% of predictions were outside a 10-fold of the measured data. Predictions within a 3-fold change were best for cattle, followed by sheep, and swine (82%, 76%, and 63%). Global sensitivity analysis was performed to identify the most sensitive parameters in the PBK model. The sensitivity analysis showed that body weight and cardiac output were the most sensitive parameters. Since interspecies differences in metabolism impact on the fate of a wide range of chemicals, a key step forward is the introduction of species-specific information on transporters and metabolism including expression and activities.

Generic physiologically based kinetic modelling for farm animals: Part I. Data collection of physiological parameters in swine, cattle and sheep.

Physiologically based kinetic (PBK) models for farm animals are of growing interest in food and feed safety with key applications for regulated compounds including quantification of tissue concentrations, kinetic parameters and the setting of safe exposure levels on an internal dose basis. The development and application of these models requires data for physiological, anatomical and chemical specific parameters. Here, we present the results of a structured data collection of anatomical and physiological parameters in three key farm animal species (swine, cattle and sheep). We performed an extensive literature search and meta-analyses to quantify intra-species variability and associated uncertainty of the parameters. Parameters were collected for organ weights and blood flows in all available breeds from 110 scientific publications, of which 29, 48 and 33 for cattle, sheep, and swine, respectively. Organ weights were available in literature for all three species. Blood flow parameter values were available for all organs in sheep but were scarcer in swine and cattle. Furthermore, the parameter values showed a large intra-species variation. Overall, the parameter values and associated variability provide reference values which can be used as input for generic PBK models in these species.

Physiologically based kinetic models for farm animals: Critical review of published models and future perspectives for their use in chemical risk assessment.

Physiologically based kinetic (PBK) models in the 10 most common species of farm animals were identified through an extensive literature search. This resulted in 39 PBK models, mostly for pharmaceuticals. The models were critically assessed using the WHO criteria for model evaluation, i.e. 1) purpose, 2) structure and mathematical representation, 3) computer implementation, 4) parameterisation, 5) performance, and 6) documentation. Overall, most models were calibrated and validated with published data (92% and 67% respectively) but only a fraction of model codes were published along with the manuscript (28%) and local sensitivity analysis was performed without considering global sensitivity analysis. Hence, the reliability of these PBK models is hard to assess and their potential for use in chemical risk assessment is limited. In a risk assessment context, future PBK models for farm animals should include a more generic and flexible model structure, use input parameters independent on calibration and include assessment tools to assess model performance. Development and application of PBK models for farm animal species would furthermore benefit from the setup of structured databases providing data on physiological and chemical-specific parameters as well as enzyme expression and activities to support the development of species-specific QIVIVE models.

Next generation physiologically based kinetic (NG-PBK) models in support of regulatory decision making.

The fields of toxicology and chemical risk assessment seek to reduce, and eventually replace, the use of animals for the prediction of toxicity in humans. In this context, physiologically based kinetic (PBK) modelling based on in vitro and in silico kinetic data has the potential to a play significant role in reducing animal testing, by providing a methodology capable of incorporating in vitro human data to facilitate the development of in vitro to in vivo extrapolation of hazard information. In the present article, we discuss the challenges in: 1) applying PBK modelling to support regulatory decision making under the toxicology and risk-assessment paradigm shift towards animal replacement; 2) constructing PBK models without in vivo animal kinetic data, while relying solely on in vitro or in silico methods for model parameterization; and 3) assessing the validity and credibility of PBK models built largely using non-animal data. The strengths, uncertainties, and limitations of PBK models developed using in vitro or in silico data are discussed in an effort to establish a higher degree of confidence in the application of such models in a regulatory context. The article summarises the outcome of an expert workshop hosted by the European Commission Joint Research Centre (EC-JRC) - European Union Reference Laboratory for Alternatives to Animal Testing (EURL ECVAM), on "Physiologically-Based Kinetic modelling in risk assessment - reaching a whole new level in regulatory decision-making" held in Ispra, Italy, in November 2016, along with results from an international survey conducted in 2017 and recently reported activities occurring within the PBK modelling field. The discussions presented herein highlight the potential applications of next generation (NG)-PBK modelling, based on new data streams.

Pre- and Postharvest Strategies to Minimize Mycotoxin Contamination in the Rice Food Chain.

Rice is part of many people's diet around the world, being the main energy source in some regions. Although fewer reports exist on the occurrence of mycotoxins in rice compared to other cereals, fungal contamination and the associated production of toxic metabolites, even at lower occurrence levels compared to other crops, are of concern because of the high consumption of rice in many countries. Due to the diversity of fungi that may contaminate the rice food chain, the co-occurrence of mycotoxins is frequent. Specific strategies to overcome these problems may be applied at the preharvest part of the crop chain, while assuring good practices at harvest and postharvest stages, since different fungi may find suitable conditions to grow at the various stages of the production chain. Therefore, the aim of this review is to present the state-of-the-art knowledge on such strategies in an integrated way, from the field to the final products, to reduce mycotoxin contamination in rice.

Predicting acute contact toxicity of pesticides in honeybees (Apis mellifera) through a k-nearest neighbor model.

Ecological risk assessment of plant protection products (PPPs) requires an understanding of both the toxicity and the extent of exposure to assess risks for a range of taxa of ecological importance including target and non-target species. Non-target species such as honey bees (Apis mellifera), solitary bees and bumble bees are of utmost importance because of their vital ecological services as pollinators of wild plants and crops. To improve risk assessment of PPPs in bee species, computational models predicting the acute and chronic toxicity of a range of PPPs and contaminants can play a major role in providing structural and physico-chemical properties for the prioritisation of compounds of concern and future risk assessments. Over the last three decades, scientific advisory bodies and the research community have developed toxicological databases and quantitative structure-activity relationship (QSAR) models that are proving invaluable to predict toxicity using historical data and reduce animal testing. This paper describes the development and validation of a k-Nearest Neighbor (k-NN) model using in-house software for the prediction of acute contact toxicity of pesticides on honey bees. Acute contact toxicity data were collected from different sources for 256 pesticides, which were divided into training and test sets. The k-NN models were validated with good prediction, with an accuracy of 70% for all compounds and of 65% for highly toxic compounds, suggesting that they might reliably predict the toxicity of structurally diverse pesticides and could be used to screen and prioritise new pesticides.

Extending standard testing period in honeybees to predict lifespan impacts of pesticides and heavy metals using dynamic energy budget modelling.

Concern over reported honeybee (Apis mellifera spp.) losses has highlighted chemical exposure as a risk. Current laboratory oral toxicity tests in A. mellifera spp. use short-term, maximum 96 hour, exposures which may not necessarily account for chronic and cumulative toxicity. Here, we use extended 240 hour (10 day) exposures to examine seven agrochemicals and trace environmental pollutant toxicities for adult honeybees. Data were used to parameterise a dynamic energy budget model (DEBtox) to further examine potential survival effects up to 30 day and 90 day summer and winter worker lifespans. Honeybees were most sensitive to insecticides (clothianidin > dimethoate ≫ tau-fluvalinate), then trace metals/metalloids (cadmium, arsenic), followed by the fungicide propiconazole and herbicide 2,4-dichlorophenoxyacetic acid (2,4-D). LC50s calculated from DEBtox parameters indicated a 27 fold change comparing exposure from 48 to 720 hours (summer worker lifespan) for cadmium, as the most time-dependent chemical as driven by slow toxicokinetics. Clothianidin and dimethoate exhibited more rapid toxicokinetics with 48 to 720 hour LC50s changes of <4 fold. As effects from long-term exposure may exceed those measured in short-term tests, future regulatory tests should extend to 96 hours as standard, with extension to 240 hour exposures further improving realism.

Applicability and feasibility of systematic review for performing evidence-based risk assessment in food and feed safety.

Food and feed safety risk assessment uses multi-parameter models to evaluate the likelihood of adverse events associated with exposure to hazards in human health, plant health, animal health, animal welfare, and the environment. Systematic review and meta-analysis are established methods for answering questions in health care, and can be implemented to minimize biases in food and feed safety risk assessment. However, no methodological frameworks exist for refining risk assessment multi-parameter models into questions suitable for systematic review, and use of meta-analysis to estimate all parameters required by a risk model may not be always feasible. This paper describes novel approaches for determining question suitability and for prioritizing questions for systematic review in this area. Risk assessment questions that aim to estimate a parameter are likely to be suitable for systematic review. Such questions can be structured by their "key elements" [e.g., for intervention questions, the population(s), intervention(s), comparator(s), and outcome(s)]. Prioritization of questions to be addressed by systematic review relies on the likely impact and related uncertainty of individual parameters in the risk model. This approach to planning and prioritizing systematic review seems to have useful implications for producing evidence-based food and feed safety risk assessment.

Risk assessment of coccidostatics during feed cross-contamination: animal and human health aspects.

Coccidiosis, an intestinal plasmodium infection, is a major infectious disease in poultry and rabbits. Eleven different coccidiostats are licensed in the EU for the prevention of coccidiosis in these animal species. According to their chemical nature and main biological activity, these compounds can be grouped as ionophoric (monensin, lasalocid sodium, salinomycin, narasin, maduramicin and semduramicin) or non-ionophoric (robenidine, decoquinate, nicarbazin, diclazuril, and halofuginone) substances. Coccidiostats are used as feed additives, mixed upon request into the compounded feed. During the technical process of commercial feed production, cross-contamination of feed batches can result in the exposure of non-target animals and induce adverse health effects in these animals due to a specific sensitivity of mammalian species as compared to poultry. Residue formation in edible tissues of non-target species may result in unexpected human exposure through the consumption of animal products. This review presents recent risk assessments performed by the Scientific Panel on Contaminants in the Food Chain (CONTAM) of the European Food Safety Authority (EFSA). The health risk to non-target species that would result from the consumption of cross-contaminated feed with coccidostats at levels of 2, 5 or 10% was found to be negligible for most animal species with the exception of salinomycin and monensin in horses because of the particular sensitivity for which toxicity may occur when cross-contamination exceeds 2% and 5% respectively. Kinetic data and tissue analyses showed that residues of coccidiostats may occur in the liver and eggs in some cases. However, the level of residues of each coccidiostat in edible animal tissues remained sufficiently low that the aggregate exposure of consumers would not exceed the established acceptable daily intake (ADI) of each coccidiostat. It could be concluded that technical cross-contamination of animal feeds would not be expected to adversely affect the health of consumers.