From the start of the project, we have been cementing our three main pillars: setting up our science and technology, making sure that our solutions can work in a regulatory context, and recruiting and engaging the people that share our vision and make it a reality.
Ultimately, we want to be able to answer the question whether a certain compound is safe, by using a safety testing pipeline that works for humans and that does not rely on animal testing. To this end, we are developing our NGRA workflow (ASPA) that is modular in nature, meaning that it is flexible and future-proof. Synergy on these activities has been achieved by common working groups within the joint ASPIS cluster.
Where the first project period saw the setting up and testing of the various workflow modules in exploratory studies, the second period focused more on their integration. To this end over 10 case studies has been delineated. Data generation and interpretation for these cases is greatly advancing and is guided by the developing ASPA in a tiered fashion.
To test if a chemical will enter the human body upon exposure, we implemented in vitro tests for uptake via the lung or gut. Moreover, our biokinetic models can predict concentrations in our tests while accounting for metabolism in the ADME pillar of ASPA, as well as for biological variability.
To predict toxicological hazards that might occur when chemicals are taken up by the body, we established computational pipelines based on chemical structures of compounds to predict to which targets they may bind for various tissue types. We also generated a panel of pluripotent stem cells with stress response reporters for high-throughput hazard screening in different target organ lineages, like hepatocytes, renal proximal tubular cells and cardiomyocytes. High-throughput data for over 100 compounds include phenotypic, MIE (molecular initiating event) and KE (key event) assay data.
For transcriptomics, which can show changes in all cellular processes at the same time, we set up comprehensive toxicogenomics prediction platforms for kidney and liver cells, including stem cell-based models for these cell types, as well as for mature peripheral neurons. We also set up state-of-the-art combined transcriptome and metabolome studies to determine cell fate. Kidney and liver organoids were further developed, used for transcriptome mapping, and combined in a two-organ chip. Challenging compounds were assembled and tested to show how to avoid false negatives.
Using network mapping, we delineated gene and protein networks and compounds associated with kidney toxicity, as well as putative new kidney biomarkers. We can connect transcriptomic and morphological data with clinical information and we constructed a database for all project compounds based on existing human exposome knowledge. We can quantify AOPs, all the way from MIE to late KEs, and developed a framework to evaluate such qAOPs. Meanwhile we started to implement our uncertainty framework for application in NGRA.
RH3R put in place a data and knowledge infrastructure with a harmonised data template and compound database, always respecting FAIR data storage criteria. For our NAMs we keep track of readiness levels for relevant applications, required for their success and commercial prospects. We also refined the ASPA workflow via interaction with stakeholders (regulators, industry, academia, NGOs) at multiple international events. A major innovation that has started to bring ASPA to the end-users is the NAMASTOX online dashboard to transparently operate on the workflow whilst integrating and documenting the safety assessment procedure.
RH3R established a strong internet presence. Key project publications and events can be found at the project website, in newsletters, and social media.
Lastly, RH3R has continuously provided highly rated training on NGRA to the new generation of scientists on risk assessment, including the practical context of linking lab research and regulatory reality.
