Read-across is a data gap filling technique within chemical category and analogue approaches used to fulfil information requirements whilst minimising animal testing in regulatory programmes worldwide. The workflow starts by assessing the data gaps of a substance of interest (target substance), then identifying (source) analogues and evaluating their suitability for the endpoint of concern and for the decision context of interest. Typically source analogues are identified on the basis of structure similarity but evaluating their suitability relies upon assessing their similarity with respect to other parameters: similarity in metabolism, mechanism of action, bioavailability and reactivity as well as the quality of the underlying data. This assessment relies upon expert judgement which can lead to inconsistent predictions and offers little insight into the general performance of read-across. We sought to evaluate the baseline performance of read-across for a large set of chemicals in a systematic, objective and reproducible manner and to provide quantitative measures of uncertainty for the predictions derived. The approach developed, generalised read-across (GenRA), relies on chemical descriptor information and/or in vitro bioactivity data (from ToxCast high throughput screening data) to derive read-across predictions of toxicity effects observed in in vivo repeat-dose toxicity studies. The approach was also implemented into a web-based tool structured around the typical category workflow, enabling users to make objective and reproducible read-across predictions yet providing some degree of flexibility in evaluating the analogues identified. More recent work has investigated physicochemical similarity as a surrogate for bioavailability and its impact in improving the read-across performance from the baseline GenRA. Here we present examples where GenRA has been applied to identify source analogues, evaluate their validity, and make read-across predictions for specific chemicals of interest. We draw examples from Superfund chemicals and highlight the insights gained from being able to make reproducible predictions with quantitative measures of uncertainty. This abstract does not necessarily represent U.S. EPA policy.

Generalised read-across GenRA, research, implementation and practical application

255th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.