A strategy for safety assessment of chemicals with data gaps for developmental and/or reproductive toxicity

- For most chemicals DART NOAELs are similar to or higher than chronic repeat dose NOAELs.
- The existing database for chemicals more frequently includes repeat dose data than DART data.
- Many of these chemicals lack structural analogs with DART data.
- Use of the DART decision tree with a database UF can provide a path for risk assessment where there is a DART data gap.
- Database UFs are suggested that can be applied to repeat dose NOAELs based on the results of the DART decision tree.

Alternative methods for full replacement of in vivo tests for systemic endpoints are not yet available. Read across methods provide a means of maximizing utilization of existing data. A limitation for the use of read across methods is that they require analogs with test data. Repeat dose data are more frequently available than are developmental and/or reproductive toxicity (DART) studies. There is historical precedent for using repeat dose data in combination with a database uncertainty factor (UF) to account for missing DART data. We propose that use of the DART decision tree (Wu et al., 2013), in combination with a database UF, provides a path forward for DART data gap filling that better utilizes all of the data. Our hypothesis was that chemical structures identified by the DART tree as being related to structures with known DART toxicity would potentially have lower DART NOAELs compared to their respective repeat dose NOAELs than structures that lacked this association. Our analysis supports this hypothesis and as a result also supports that the DART decision tree can be used as part of weight of evidence in the selection of an appropriate DART database UF factor.