The European Food Safety Authority (EFSA) has published a paper describing research and innovation needs to support the agency’s regulatory science in the coming years.

For example, advances in omics technologies could help enhance risk assessment capabilities by providing deeper insights into mechanisms of action, improving hazard identification and characterization, and enabling more precise and quantitative data. Although EFSA has made progress implementing specific omics approaches, such as whole genome sequencing (WGS), further research and innovation—including the development of standardized approaches, validation frameworks, and data analysis, as well as the execution of proof-of-concept studies—is required to facilitate the adoption of other omics technologies, such as transcriptomics and metabolomics.

Another area in which advancement could improve risk assessments is understanding the human gut microbiome. Although evidence suggests that dietary components can adversely affect the gut microbiome, little is known about the impact of food-related compounds on human microbiota and their subsequent effects on human health. EFSA underlines the need for in vitro methods to be developed to obtain data on chemical-microbiome interactions, in support of establishing microbiome-specific endpoints for risk assessments. There is also a need to develop and validate biomarkers for assessing microbiome-related adverse effects to health, as well as to gather data from targeted epidemiological cohort studies to establish links between dietary exposure to specific compounds, gut microbiome alterations, and health outcomes.

The paper also describes how new approach methodologies (NAMs), which are non-animal methods of chemical safety assessment, could be leveraged for food safety risk assessment. Currently, the integration of NAMs is hindered by regulatory data requirements that still mandate in vivo testing, as well as the need to further validate NAMs and gain confidence in their use for risk assessment through case studies. EFSA calls for the development of a workflow and toolbox for NAM integration that can demonstrate practical examples of its use and facilitate the transition toward animal-free testing.

Allergenicity risk assessment of novel proteins and food and feed is another area in which research and innovation is necessary. To ensure allergenicity risk assessments are still fit-for-purpose in light of scientific developments, EFSA underlines the need to improve predictions to aid decision-making for de novo sensitization and predictions for cross-reactivity, as well as to establish criteria on when and how in vitro (and eventually, in vivo) testing would be required to validate such predictions.

Moreover, to better ensure food chemical safety, advancements in exposure science, especially for aggregate exposure to multiple chemicals, in necessary. At present, EFSA primarily focuses on dietary exposure and legislation is mostly based on individual substances. Given that consumers are exposed to many chemicals through multiple routes, however, EFSA sees the need for three critical research and innovation actions to advance aggregate chemical exposure assessments: 1) create an EU framework for aggregate exposure assessments and develop guidance on the use of exposure models and human biomonitoring data, 2) develop a comprehensive approach for grouping chemicals, and 3) cultivate a more comprehensive understanding of the exposome and its role in human health.

Finally, there is a need to improve environmental risk assessments (ERAs), which are required for regulated products like feed additives, food flavorings, genetically modified organisms (GMOs), and plant protection products, and which evaluate the potential adverse effects of such products on the environment. As ERAs are often criticized for not aligning with ecological and practical realities, research and innovation is needed to develop and refine testing methods and approaches to assess the effects of regulated products on relevant species based on an organism’s biological and ecological traits. Work is also needed to integrate landscape factors in ERAs to provide more context-specific and ecologically relevant predictions.