Evaluation of in vitro assays for the assessment of the skin sensitisation hazard of functional polysiloxanes and silanes
A published research programme investigating the skin sensitisation potential of a group of functional polysiloxanes and silanes in vivo and in vitro is an instructive example of what can happen in ‘real life’ when testing a group of chemical substances that seemingly belong to the same class of chemicals for skin sensitisation. It demonstrates the importance of having the flexibility to adjust the testing strategies to address the skin sensitisation endpoint to the chemical nature of the test substance.
In August, the latest of a series of three articles concerning the assessment of the skin sensitisation hazard of a diverse group of polyfunctional polysiloxanes and silanes were published in Regulatory Toxicology and Pharmacology. The research programme underlying the articles was funded by the Silicones Europe (Centre Européen des Silicones) and spanned over more than 10 years with the first publication in 2012, the second in 2017 and the last one now in 2018.
The programme evaluated the suitability of recently validated in vitro skin sensitisation assays DPRA, KeratinoSens™ and h-CLAT as well as the human 3D tissue-based SENS-IS for the skin sensitisation hazard of the investigated silicon materials by assessing the in vitro findings against in vivo outcomes.
One could have expected, some heterogeneity of the results in vitro, for a subgroup of the substances, but the difficulties started already with interpreting the in vivo data: while the polymeric siloxanes were consistently negative in the GPMT, they revealed skin sensitisation potential in the LLNA. Only a thorough weight-of-evidence analysis covering PC properties, structure activity analyses and existing human information allowed drawing the conclusion that the polymeric siloxanes do not represent a skin sensitisation hazard to humans. In contrast, the low molecular weight siloxanes and silanes were consistently positive for skin sensitisation in the GPMT and the LLNA.
In vitro, the validated assays DPRA, KeratinoSens™ and h-CLAT provided a heterogeneous picture. Not all substances considered could be tested, largely for solubility reasons. In vivo outcomes were inconsistently predicted particularly for the non-polymeric siloxanes and silanes. Even when using recently proposed ‘2 out of 3’ approaches, 3 out of 5 skin sensitisers would have been missed. The SENS-IS provided, however, some promising results. First, it allowed testing of all substances and it matched all in vivo outcomes.
The programme on functional polysiloxanes and silanes is an instructive example of what can happen in ‘real life’ when investigating the skin sensitisation potential of group of chemical substances that seemingly belong to the same class of chemicals. Solubility issues in assay media or issues related to the multiconstituent nature of the commercial grade testing materials prevented the testing of some of the substances in those assays that are validated alternatives to in vivo testing. It should be noted that the testing was already initiated at time point when some of the protocols were not yet finally validated with regard to their applicability domains. But how to assess the skin sensitisation potential of those substance which are outside the applicability domain and in vivo testing is not an option for regulatory reasons? There is still quite a bit to do. The good news is that with the SENS-IS there was an assay that delivered promising results. There are also a number of further new assays and approaches in the pipeline which may support the skin sensitisation assessment of those substances which cannot be tested or provided unreliable results in currently validated test systems for technical reason. It’ll be important that also from a regulatory point of view a certain degree of flexibility in the choice of the final testing strategy will be permitted.