Computer Models and QSARs

The Consortium and its members have funded the development of computer models to characterise chemical properties and decrease animal testing. In 2009, members funded a successful international review of a method to predict endocrine activity of chemicals; in 2010 and 2012, members funded a series of workshops to create computer models to predict cancer-causing activity. Computer models developed with the support of Consortium members, including the endocrine model, have been incorporated into a publicly available predictive tool managed by the OECD called the OECD QSAR Toolbox, which is now widely used to fulfil data requirements for regulatory programmes around the globe. As part of the International Council on Animal Protection in OECD Programmes (ICAPO), the Consortium partnered with computational toxicology experts to train US Environmental Protection Agency regulators in the use of the OECD QSAR Toolbox.

Consortium member PETA US funded the 2nd McKim Workshop on Reducing Data Redundancy in Cancer Assessment, which was hosted by the International QSAR Foundation and held in Baltimore, Maryland, in May 2012. Participants reviewed recent developments in QSAR screening methods for grouping chemicals based on their potential to cause cancer through epigenetic pathways and evaluated the combined use of structural domains and in vitro data to reduce the use of the rodent carcinogen assay. Of particular note, Romualdo Benigni of the Istituto Superiore di Sanità in Rome, Italy, described his PETA US-funded research demonstrating that a tiered approach–including the Ames bacterial test, newly identified structural alerts, and an in vitro cell transformation assay–correctly detects 90 to 95 per cent of carcinogens tested.

“The science PETA is sponsoring is critical to the elimination of animal use. PETA puts its money where its mouth is and has steadfastly supported better use of science to reduce reliance on animal testing.”
-Dr Gilman Veith
Chair of the Board
International QSAR Foundation