Optimised strategies for the identification of skin sensitising chemicals
JRC scientists have been developing and optimising animal-free testing
strategies for the identification of chemicals with the potential to cause skin
allergy, as a contribution towards international efforts aimed at fulfilling
regulatory information requirements on chemical safety while at the same time
reducing or avoiding the use of vertebrate animals.
Traditionally, this information has been obtained by performing an animal test,
such as the Local Lymph Node Assay (LLNA) in mice. However, for scientific and
animal welfare reasons, regulatory bodies are increasingly requiring that
information on skin sensitisation potential is provided instead by mechanistic
information generated by using alternative (non-animal) test methods.
At present, EU legislation does not provide detailed and explicit guidance on
how the different pieces of mechanistic information, generated by non-animal
methods, should be combined in order to predict the potential for chemically
induced skin sensitisation. Therefore, in line with the EURL ECVAM skin
sensitisation
strategy<http://publications.jrc.ec.europa.eu/repository/handle/JRC79446>, JRC
scientists have been exploring the development of these integrated prediction
models.
A detailed analysis<http://dx.doi.org/10.1016/j.tiv.2016.07.014> of a high
quality skin sensitisation dataset by JRC scientists (Asturiol et al, 2016) has
shown that sensitising and non-sensitising chemicals can be identified with a
high level of overall predictive accuracy (approximately 93%) by using decision
trees based on easily computable properties of chemicals, such as their
reactivity towards proteins. This is therefore an efficient and cost-effective
strategy for hazard assessment. In a separate study, JRC scientists contributed
to an investigation led by researchers at the University of Wagenigen
(Leontaridou et al, 2016). This
study<http://www.atla.org.uk/evaluation-of-non-animal-methods-for-assessing-skin-sensitisation-hazard-a-bayesian-value-of-information-analysis/>
provides a novel illustration of how a decision theory approach can be applied
to address cost-benefit questions related to the marketing and use of chemical
products.
Read more in:
Asturiol et al (2016), "Consensus of classification trees for skin
sensitisation hazard prediction<http://dx.doi.org/10.1016/j.tiv.2016.07.014>",
Toxicology in Vitro 36, 197–209,
doi:dx.doi.org/10.1016/j.tiv.2016.07.014<http://dx.doi.org/10.1016/j.tiv.2016.07.014>
Leontaridou et al (2016), "Evaluation of Non-animal Methods for Assessing Skin
Sensitisation Hazard: A Bayesian Value-of-Information
Analysis<http://www.atla.org.uk/evaluation-of-non-animal-methods-for-assessing-skin-sensitisation-hazard-a-bayesian-value-of-information-analysis/>",
ATLA 44, 255–269.
