(Q)SARs – Emerging Science & Policy

Home / MAPP / Emerging Technologies / Non-test Approaches: (Q)SARs, Read-Across / (Q)SARs – Emerging Science & Policy

Non-test Approaches: (Q)SARs, Read Across

Emerging Science & Policy

Last updated: November 4, 2014

The regulatory landscape has undergone a tremendous shift in the last 10 years, and with it there has been a concerted effort to exploit the use of alternative approaches such as (Q)SARs, chemical categories, and read-across to fulfill more than just screening purposes. The shift was largely driven in the EU in part by the 7th Amendment to the Cosmetics Directive, but also notably the REACH (Registration, Evaluation, and Authorisation of Chemicals) legislation. REACH in particular encourages the use of (Q)SARs as a means of addressing endpoint information requirements, provided a number of conditions are met. Annex XI, Section 1.3, of the REACH Regulation outlines the requirements, namely that the results are derived from a valid (Q)SAR model, should be ‘adequate’ for classification and labeling and/or risk assessment purposes, the substance under consideration should fall within the applicability domain of the (Q)SAR model, and adequate and reliable documentation of the applied method should be provided.

Guiding principles for the development and application of (Q)SARs were first identified during the course of a workshop organized by the European Chemical Industry Council (CEFIC) and the International Council of Chemical Associations (ICCA) in Setubal, Portugal, in 2002 (Jaworska et al., 2003). These principles were subsequently evaluated by the OECD as part of the ad hoc group for (Q)SARs and adopted by the Joint Meeting in 2004. They are now referred to as the OECD Principles for the Validation of (Q)SARs. OECD Guidance, published in 2007, characterized each of the principles in turn. As part of the development of the REACH Technical Guidance, the European Commission published guidance on the application of (Q)SARs to characterize the REACH conditions further. Specific reporting formats were developed to help address the need for “adequate documentation.” Two reporting formats were derived, the so-named (Q)SAR Model Reporting Format (QMRF) and the (Q)SAR Prediction Reporting Format (QPRF). Both are underpinned by the OECD principles. The QMRF aims to characterize and document a (Q)SAR model in accordance with the OECD principles. The QPRF discusses the robustness and relevance of a prediction made using a given model, thus addressing the condition of “falling within the applicability domain.” To date under REACH, the greatest application for (Q)SARs has been in filling data gaps in lieu of testing for physicochemical properties, ecotoxicological, and environmental fate endpoints. (Q)SARs have also been used to help substantiate read-across through providing mechanistic information or as part of a weight of evidence approach for mammalian endpoints.

The JRC’s Institute of Health and Consumer Protection (IHCP) Unit was particularly active during the run-up to REACH in developing guidance and tools to assist both registrants and regulators in the application of (Q)SARs.  The JRC commissioned a number of open source computer-based applications, which are made freely available via their website. Tools include Toxtree (Ideaconsult, Ltd.) a platform containing many structural alert schemes: DART (Talete, Srl.), to facilitate the ranking and prioritization of chemicals; Toxmatch (Ideaconsult, Ltd.), a chemical similarity tool; as well as the JRC (Q)SAR Inventory, a repository of documented QMRFs.

The OECD has been working actively in the area of (Q)SARs since the 1990s. Since 2004, the OECD’s main focus has been on developing a (Q)SAR Application Toolbox, also known as the OECD QSAR Toolbox. A prototype was first made publicly available in March 2008. Phase 2, a 4-year project which began in late 2008, was in collaboration with ECHA. The most recent version of the Toolbox is v3.2, which was released December 2013. The Toolbox is aimed to facilitate the development, justification, and documentation of chemical categories for read-across. The workflow within the Toolbox mimics that described in both the OECD Guidance on Grouping of Chemicals, Second Edition (No. 194) (OECD, 2014a) as well as the REACH Technical Guidance on Grouping. The Toolbox contains a number of regulatory inventories, profilers to assist in grouping substances based on structural alerts, as well as experimental data from a number of sources. The ECHA REACH dissemination data (2010) was also made available to the OECD Toolbox. The data model underpinning the Toolbox utilizes the OECD Global Harmonised Templates, which facilitates export and import between IUCLID and the Toolbox. Thus, the Toolbox is uniquely positioned to allow read-across and (Q)SAR predictions to be readily imported into IUCLID for REACH purposes. A current focus of the OECD Toolbox now is in the area of Adverse Outcome Pathways (AOPs), and exploring to what extent these can assist in the grouping of substances for read-across, particularly for more complex systemic toxicity endpoints. The AOP for skin sensitization that was published by OECD in 2012 (OECD, 2012a, 2012b) has since been implemented into the Toolbox. The tutorial, “How to use the Toolbox AOP workflow for skin sensitization,” is available.

The OECD Guidance Document No. 80 (2007) was developed in parallel with that being prepared for REACH. The OECD guidance aimed to be more practical in terms of outlining the strategies for developing, justifying, and documenting chemical categories. This guidance underwent revision to incorporate experiences gained under REACH and other frameworks. The revised version published in April 2014 (OECD, 2014b) includes more practical guidance for evaluating analogue suitability as well as AOPs.

The International QSAR Foundation fulfills its mission “to develop computerized tools as alternatives to animal testing” by supporting research for the development of (Q)SAR models for regulatory testing endpoints and by hosting annual workshops. One of their recent focuses was on developing AOPs and in particular establishing Effectopedia, an open source tool that could serve as a repository for AOPs in development or completed. As part of the OECD AOP work program in 2012, a joint collaboration between the OECD, the US Environmental Protection Agency, and the European Commission Joint Research Centre was established to develop an Adverse Outcome Pathway Knowledge Base (AOP KB). This is a web-based platform which aims to bring together all knowledge on how chemicals can induce adverse effects, therefore providing a focal point for AOP development and dissemination. The first AOP KB module is the AOP Wiki, launched in September 2014, which is an interactive and virtual encyclopedia for AOP development, structured in accordance with the original OECD “Guidance document and a template for developing and assessing adverse outcome pathways” (OECD, 2013), and the more recent Handbook for AOP Developers. Another module will be to continue the development of Effectopedia.

New Perspectives

For some new perspectives on (Q)SARs and Read-across, read the following invited commentaries:

Grace Patlewicz, PhD
US Environmental Protection Agency

AltTox Editorial Board reviewer(s):
Sherry L. Ward, PhD, MBA
AltTox Contributing Editor