How Well Do Defined Approaches Predict the Human Skin Sensitisation Potential of Previously Untested Chemicals?

SOT 2022 -- The in vitro prediction of human skin sensitisation took a significant step forward with the publication of the OECD's Health Effects Guideline 497 on Defined Approaches for Skin Sensitisation in 2021. Prior to that, in vitro data was used as a substitute to in vivo methods and validation efforts notoriously presented comparison figures on sensitivity and specificity between the new in vitro method and the existing in vivo method. Publication of the skin sensitisation AOP solidified the relationship between molecular perturbations, pathway biology and the resultant adverse human outcome(s) when exposed to chemical substances. Since then, test methods for three of the four key events have been developed and in 2015/2016, Test Guidelines for each were published by OECD. Combining multiple data sets to provide an overall conclusion is a complex process so in recognition of this, additional guidance to assist data interpretation was also published along with case studies demonstrating applications. Since then, the Defined Approach (DA) concept has gained further traction and only five years later, OECD Guideline 497 was published, containing three DAs from the sub-set of DAs previously documented. Each of these combines information sources (in chemico, in vitro and/or in silico) to predict skin sensitisation hazard and in some cases GHS potency sub-classification. The true goal is to predict the human outcome and thus, in order to challenge the DAs described in OECD 497, we have tested 22 chemicals [13 human sensitisers (3 were Cat 1A) and 9 non-sensitisers] that had existing animal and/or human data but did not have in chemico/in vitro data. The novel data were then assessed using the 3 DAs in OECD 497 and an additional DA designed by Lhasa Limited.