Community Research and Development Information Service - CORDIS

Abstract

This review describes the state of (Q)SARs/expert systems for skin sensitisation from the early work on structural alerts to the evolvement of QSAR models through to the development of a mechanistically based QSARs. The review considers the merits of the different approaches with particular focus on their applicability for potential regulatory use. There is a strong mechanistic understanding with respect to skin sensitisation which has facilitated the development of a wide range of models. The majority of the existing models fall into one of two main categories - either they are local in nature, typically specific to a chemical class or else they are global in form, derived empirically using statistical methods. Several of the published global QSARs have been recently characterised and evaluated in accordance with the OECD principles. There are also several expert systems capable of predicting skin sensitisation, these are briefly described. Recently, a new perspective regarding the development of mechanistic skin sensitisation QSARs so-called Quantitative Mechanistic Modelling (QMM) was proposed, where reactivity and hydrophobicity, are used as the key parameters in mathematically modelling skin sensitisation. This approach appears to be the most promising means of deriving robust and mechanistically interpretable models that are of value in a regulatory context. Hydrophobicity can be conveniently modelled using log P, the octanol-water partition coefficient. This is readily determined from chemical structure. No such surrogate for reactivity exists, rate constants are best derived from experimental studies. Initiatives are in progress to generate reactivity data for reactions relevant to skin sensitisation but more resources are required to realise a comprehensive set of reactivity data. This is a fundamental and necessary requirement for the future assessment of skin sensitisation. A framework of how information from using RAI approaches can be used in the evalu

Additional information

Authors: PATLEWICZ G, European Commission, Joint Research Centre, Institute for Health and Consumer Protection, Ispra (IT);APTULA AO, European Commission, Joint Research Centre, Institute for Health and Consumer Protection, Ispra (IT);ROBERTS DW, European Commission, Joint Research Centre, Institute for Health and Consumer Protection, Ispra (IT)
Bibliographic Reference: EUR 22872 EN (2007), various paging. Free of charge
Availability: Katalogue Number: LB-NA-22872-EN-C The PDF version can be downloaded from: http://bookshop.europa.eu
Record Number: 200819821 / Last updated on: 2008-05-05
Category: PUBLICATION
Original language: en
Available languages: en