Brussels, Belgium – 1 October 2026 – The ONTOX project concluded its fifth and final Annual Meeting, held on 30 September and 1 October 2026 at Vrije Universiteit Brussel. The meeting brought together the ONTOX community to reflect on more than five years of collaborative research and to discuss how the knowledge, methods, computational tools and scientific networks established during the project can continue beyond its formal conclusion on 31 October 2026.
Since 2021, ONTOX has brought together 19 partners from Europe and the United States with the ambition of advancing New Approach Methodologies (NAMs) for repeated-dose toxicity assessment and supporting the transition towards more human-relevant, animal-free approaches to chemical safety assessment.
The project has addressed six toxicological adversities affecting the liver, kidney and developing brain, combining biological, toxicological, chemical and kinetic knowledge with in vitro methods, computational modelling, ontologies and artificial intelligence. The final Annual Meeting brought these different strands together, reviewing both the progress achieved and the challenges that remain for further development, validation, implementation and regulatory uptake.
From biological understanding to risk assessment
A central theme of the final meeting was how the different scientific components developed within ONTOX can contribute to the transition from hazard identification towards next-generation risk assessment (NGRA).
The first day followed this progression deliberately. Sessions addressed the biological, toxicological, chemical, kinetic and data-integration domains of the ONTOX framework, followed by organ-specific NAMs for hepatotoxicity, nephrotoxicity and developmental neurotoxicity. A dedicated session, “From hazard to risk assessment,” then focused on how these different evidence streams can begin to support risk-assessment applications. The programme included an overview of the ONTOX strategy towards risk assessment and implementation, presentation of ONTOX case studies, discussion of the emerging ONTOX Probabilistic Risk Assessment concept (OPRA), and a demonstration of ToxIndex, an AI-supported tool for toxicological hazard and risk prediction.
The case-study approach has been particularly important in examining how experimental NAM data, computational predictions, kinetic information, mechanistic knowledge and existing evidence can be brought together around concrete chemical-safety questions. OPRA, still at an early stage of development, explores how such evidence streams might in future be combined within a more structured probabilistic framework, with explicit consideration of uncertainty and variability.
The ONTOX consortium has also developed and/or applied a broader ecosystem of computational resources, including ToxIndex and ToxTransformer, agentic AI platforms supporting regulatory toxicology; SysRev, an AI-assisted platform that accelerates systematic literature review; DockTox, an online molecular-docking platform; VEGA QSAR, part of the VEGAHUB suite, providing computational models for the prediction of chemical properties and toxicological endpoints; ECCS-DT, a computational workflow for classifying small molecules according to their predominant clearance mechanism; QSAR Physchem, a suite of models for physicochemical property prediction; LLEMY, an LLM-based system for exploring molecular maps; and NeuroTox-KPGT, an automated AI tool that transforms raw ChEMBL bioactivity data into optimized deep-learning models for predicting molecular initiating events. Together with physiological maps, adverse outcome pathway networks, kinetic models, experimental NAM data and project case studies, these resources provide a scientific foundation for continued testing, refinement and application beyond the project.
“ONTOX started with the ambition to connect different types of scientific evidence into a practical framework for human-relevant chemical safety assessment without animal testing. Five years later, we have developed methods, case studies and computational tools that demonstrate what such integration may look like in practice. Our final meeting is therefore not simply the conclusion of a project, but also a handover. The next challenge is to continue testing, refining and evaluating these approaches and to support their wider scientific and regulatory use,” said Prof. Mathieu Vinken, ONTOX Coordinator, Vrije Universiteit Brussel.
Moving from research towards implementation
The final meeting also placed strong emphasis on innovation, impact, sustainability and regulatory uptake. The programme included discussions on innovative services for chemical safety, sustainability through collaboration, the ASPIS Cluster and ASPIS Academy, NAM policy and the translation of research into regulatory uptake. These discussions reflected an important message emerging from ONTOX: scientific innovation alone is not sufficient to transform chemical safety assessment. New approaches require clearly defined contexts of use, transparent documentation, robust performance evaluation and continued dialogue among researchers, regulators, industry and other end users. The ONTOX Hub is being developed to help maintain access to project outputs and expertise, including NAM and NGRA resources, computational tools, knowledge and services generated across the consortium.
The second day continued this focus on sustainability, with dedicated sessions on data management and quality control, dissemination, exploitation, communication, coordination and feedback from the Scientific Advisory Board.
“ONTOX has made important progress in connecting mechanistic science with risk assessment and in exploring how computational and AI-supported tools can help integrate increasingly complex evidence. At the same time, continued validation, transparency and regulatory dialogue will be essential before many of these approaches can achieve wider use. The young scientists trained through ONTOX are a particularly important part of this legacy, because they will continue to test, improve and implement these methods,” said Dr. Paul Whaley, Lancaster University, UK.
The end of the project – but not of its scientific impact
ONTOX will formally conclude on 31 October 2026, but many of the scientific questions, tools and collaborations established during the project will continue. The project’s legacy lies not in a single method or computational platform, but in the combination of mechanistic understanding, experimental NAMs, computational approaches, kinetic information, case studies, risk-assessment concepts, training and interdisciplinary collaboration.
By bringing these elements together, ONTOX has contributed to the scientific foundations needed for a longer-term transition towards more transparent, mechanistically informed, human-relevant and animal-free chemical safety assessment in Europe and internationally.
***************************
About ONTOX:
Ontology-driven and artificial intelligence-based repeated dose toxicity testing of chemicals for next-generation risk assessment (ONTOX) is a project funded by the European Commission’s Horizon 2020 (H2020) research and innovation program under Grant Agreement No. 963845. The vision of ONTOX is to provide a functional and sustainable solution for advancing human risk assessment of chemicals without the use of animals in line with the principles of 21st-century toxicity testing and next-generation risk assessment. ONTOX is part of the ASPIS cluster, the collaboration of three EU-funded H2020 projects (ONTOX, PrecisionTox and RISK-HUNT3R) that represents Europe’s €60M effort towards the sustainable, animal-free and reliable chemical risk assessment of tomorrow.
- Website: https://ontox-project.eu
- LinkedIn: @ONTOX
- X: @ONTOX_EUproject
- Facebook: @ONTOX