The test strategy consists of the following steps:
• Select an adverse outcome pathway (AOP) for grouping and testing chemicals;
• Align the AOP with in-silico methods, i.e. (quantitative) structure activity relationship and molecular docking, and in-vitro assays;
• Use available literature data, in-silico predictions and/or exposure results for priority setting;
• Test priority chemicals in vitro, model the results to derive chemical potencies and test the appropriateness of the dose addition assumption;
• Develop relevant IVIVE models;
• Compare in-vitro results with in-vivo experiments for verification;
• Perform MRA using the newly generated data.
The test strategy is elaborated for three outcomes: fatty changes in lever, feminisation and cranio-facial malformation. In-silico predictions and Threshold of Toxicological Concern values are available for 1630 substances (10 chemical classes). A probabilistic model for estimating co-exposure of chemicals to prioritise mixtures was programmed and applied for 10 EU countries. These activities have resulted in a list of priority chemicals for further testing.
Relevant in-vitro assays aligning the AOPs have been established and validated. Priority chemicals were tested alone for chemical potencies or as mixtures displaying similar or dissimilar mode of action (MoA). So far, dose-addition seems to be the most common mechanism. In-vivo studies confirmed the in-vitro findings for feminisation and cranio-facial malformation. Results of the in vivo study for liver steatosis were inconclusive and need more investigation. For IVIVE, 9 chemical specific and 1 generic PB-TK models as part of IVIVE were developed.
The web-based EuroMix model and data platform contains relevant models and data for efficient MRA:
• Existing and new developed models for hazard (BMD and PB-TK) and exposure assessment (combined and aggregated);
• Data obtained from in-silico predictions and in-vitro assays;
• Concentration and consumption data.
The EuroMix model and data platform was validated and fits into the European modern ICT infrastructure strategy (e.g. interagency i-cloud).
Case studies using the EuroMix model and data platform were performed. For aggregated exposure, three case studies for multiple pesticides, bisphenols, and pyrethroids were performed. A case study addressing MRA of pesticides, additives and contaminants was performed for 8 EU countries. Examples implementing use of in-silico and in-vitro data in MRA were also performed.
A human biomonitoring study focusing on substances relevant for aggregated exposure was finalised. Biological samples, food consumption data and use of cosmetics were obtained and analysis of samples and data interpretation has been performed. The biomonitoring results were compared with predicted exposure based on residue data and consumption patterns.
Access to the tools was facilitated by physical training and webinars. A practical guidance on how to use the EuroMix tests and models, in line with international developments, was written. Dissemination and harmonization of the approach was achieved by involving key-experts, EFSA, WHO and US-EPA, four harmonization workshops and a WHO expert consultation on how the EuroMix tools can be used in Europe and in a global setting.
