Minimising data needs to support the safer design of multicomponent nanomaterials – Application of grouping

• A grouping-based strategy has been developed for design of safer materials. • Focused sub-hypotheses simplify logical evidence gathering for grouping. • Grouping sub-hypotheses address material property impacts on fate and hazard. • Enhanced properties, mixture effects are accounted for, along with transformations. • New data is generated through application of tiered testing strategies promoting the use of NAMs. There is an ongoing demand to develop options to reduce hazard testing of substances and materials on a case-by-case basis. Grouping approaches offer a way to share or re-use safety-related information between similar substances, providing insights that can inform the Safe and Sustainable by-Design (SSbD) 2 2 SSbD: Safe and Sustainable by-Design. of new materials. Here, an existing grouping hypothesis template for single-component nanomaterials (NMs) 3 3 NMs: Nanomaterials. is expanded to facilitate systematic consideration of grouping for multicomponent nanomaterials (MCNMs) 4 4 MCNMs: Multicomponent Nanomaterials. relevant to SSbD. Modifications to the template include additional information on a) the complexity of physical and chemical composition; b) the emerging properties driving the MCNM functionality; c) the potential for MCNM components to transform with different rates, leading to complex exposure scenarios; d) prioritisation and simplification of grouping decisions related to material properties (what they are), fate/toxicokinetics (where they go) and the hazard mechanisms (what they do). Existing information and data are used to formulate a matrix of sub-hypotheses that individually relate one (or more) indicators of ‘what they are’ to a single indicator of either ‘where they go’ or ‘what they do’. The resultant sub-hypotheses are easier to assess than the all-encompassing over-arching hypothesis required for regulatory application of grouping. The estimated level of impact of each indicator is used to prioritise the sub-hypothesis assessment. Accepting or rejecting each prioritised sub-hypothesis is facilitated by the application of tiered testing strategies promoting the use of relevant existing data, new approach methodologies and machine learning-based models. A case study of SiO 2 @ZnO MCNM is provided to demonstrate the template’s usefulness in an SSbD context.