Table 2.

Summary of actions needed to build AOP-based in vitro DNT screening tools for regulatory use

Creation of putative AOPs for DNT by taking existing data on basic molecular developmental neuroscience as well as DNT into account that will foster:

Targeted generation of missing molecular-, cellular-, tissue- and organism-level data using in vitro and in vivo methods to develop validated AOPs

Identification of MIEs and/or KEs in priority AOPs for which cell models/alternative organisms must be generated

Generation of chemical training and testing sets for the use in assay development and validation

Generation of data sets for large numbers of chemical that allows qualification/validation of assay use that is “fit for purpose”, including:

Comparison of results across assays with similar endpoints

Comparison of results of different assays across chemicals

Development of in silico models (e.g. QSAR, docking models)

Development of a DNT alternative methods testing battery for the use in routine screening of new and existing chemicals

Development of predictive computational models based on AOPs that assess reliability of both individual test methods and the DNT testing battery, including:

Definition of model- and endpoint-specific quantitative cut-off values for delineating adversity

Development and incorporation of qualitative and quantitative species-specific differences in signalling pathway-driven guidance of developmental processes

Generation of case studies for use of AOP-based DNT screening data in regulatory decisions, including:

Use in multiple types of regulatory decision such as read across, prioritization for further testing and replacement of in vivo testing requirements