Barcelona April, 2008 Overview of the QSAR Application Toolbox Gilman Veith International QSAR Foundation Duluth, Minnesota

QSAR Foundation Goals  Identify major scientific gaps in QSAR capabilities for modeling regulatory endpoints and develop an agenda to bridge them  Develop high quality databases for QSAR modeling (repeated dose, metabolism, nucleophile reactivity profiles)  Provide QSAR training for regulators, industrial users and students

QSAR at a Glance  Chemistry is based on the premise that similar chemicals will have similar chemical behaviours, including toxic effects  QSAR is the science of chemical similarity and grouping chemicals by mechanisms  QSAR methods use a few measurements in each group to estimate untested chemicals

QSAR Purpose  It is not possible to test and assess all chemicals for all known hazards  Only a small fraction of chemicals are likely to be found to pose significant hazards in any given test guideline  QSAR is needed to identify chemicals with minimal hazards and focus assessments on the chemicals posing the greatest hazards

Illustrating the Small Percentage of Chemicals with Relevant ER Relative Binding Affinity (RBA) Among the 39,436 TSCA Chemicals RBA

QSAR Methods  QSAR fills Data Gaps by first grouping chemicals and then using Existing Data within a group to estimate Missing Values  When the chemical group is identified by a common mechanism, QSAR models accurately describe the trends

Aquatic Toxicity for NonPolar Industrial Chemicals have Consistent Trends over 4-5 Orders of Magnitude

Log P Log Molar Concentration Many Mechanisms Give Similar Trends LC hr Water Solubility

Grouping Chemicals is Known as the Category Approach The reliability and transparency of QSAR are based on grouping common mechanisms Chemical mechanisms are easily encoded into computers for practical use by assessors The OECD Toolbox was created to simplify grouping chemicals and filling data gaps

What do we mean by Chemical Categories? A group of chemicals that have some features that are common –Structurally similar e.g. common substructure –Property e.g. similar physicochemical, topological, geometrical, or surface properties –Behaviour e.g. (eco)toxicological response underpinned by a common Mechanism of Action –Functionality e.g. preservatives, flavourings, detergents, fragrances

Substances whose physicochemical, toxicological and ecotoxicological properties are likely to be similar or follow a regular pattern as a result of structural similarity may be considered as a group, or “category” of substances. Application of the group concept requires that physicochemical properties, human health effects and environmental effects or environmental fate may be predicted from data for a reference substance within the group by interpolation to other substances in the group (read-across approach). This avoids the need to test every substance for every endpoint. Annex IX of REACH

Rationales for Chemical Categories CHEMICAL CATEGORY Similar chemical structureSimilar physicochemical properties (or predictable pattern) Similar environmental fate properties (or predictable pattern) Similar (eco)toxicological effects (or predictable pattern) Key words: Structural similarity (structural analogues) Predictable pattern (trends in behaviour)

Organization for Economic Co-operation and Development QSAR Application Toolbox -filling data gaps using available information- Training Workshop Barcelona

QSAR Application Toolbox Organization for Economic Co-operation and Development First “organized” discussions – ‘Red Lobsters’, Duluth filling data gaps using available information- Historical Notes Organized actions of EU and OECD – coming with REACH The role of the “revolutionary” notions – category, analogues OECD and EU Guidance documents on ‘Category’, ‘QSAR’ Need for translation documents into a working machinery

QSAR Application Toolbox Organization for Economic Co-operation and Development Improve accessibility of (Q)SAR methods and databases Facilitate selection of chemical analogues and categories Integrate metabolism/mechanisms with categories/(Q)SAR Assist in the estimation of missing values for chemicals -ENV/JM(2006)47 -filling data gaps using available information- General Objectives

Special thanks to: DG Environment European Chemicals Bureau Danish Ministry of the Environment US EPA Environment Canada NITE Japan CEFIC MultiCase (USA) SRC (USA) A collaborative effort of all member countries and stakeholders

Typical queries included in the (Q)SAR Application Toolbox Is the chemical included in regulatory inventories or existing chemical categories? Has the chemical already been assessed by other agencies/organisations? Would you like to search for available data on assessment endpoints for each chemical?

Explore a chemical list for possible analogues using predefined, mechanistic, empiric and custom built categorization schemes? Group chemicals based on common chemical/toxic mechanism and/or metabolism? Design a data matrix of a chemical category? Typical Queries included in the (Q)SAR Application Toolbox

Fill data gaps in a chemical category using: –read-across, –trend analysis or –QSAR models Report the results: –Work history –Export the data matrix –IUCLID 5 harmonized templates Typical Queries included in the (Q)SAR Application Toolbox

System Workflow

Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

User Alternatives for Chemical ID: A. Single target chemical Name CAS# SMILES/InChi Draw Chemical Structure Select from User List/Inventory B. Group of chemicals User List Inventory Specialized Databases Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

User Alternatives for Chemical ID: A. Single target chemical Name CAS# SMILES/InChi Draw Chemical Structure Select from User List/Inventory B. Group of chemicals User List/Inventory Specialized Databases Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage  Toolbox Inventories: US EPA TSCA Canadian DSL OECD HPVCs, US EPA HPVCs EU EINECS Japanese MITI DANISH EPA

General characterization by the following grouping schemes: Substance information Predefined Mechanistic Empirical Custom Metabolism Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

General characterization by the following grouping schemes: Substance information: CAS Name Structural formula OECD Global portal Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

General characterization by the following grouping schemes: Substance information Predefined: US EPA categorization OECD categorization Database affiliation Inventory affiliation Substance type: polymers, mixtures, discrete, hydrolyzing Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

General characterization by the following grouping schemes: Substance information Predefined Mechanistic: Acute Toxicity MOA Protein binding (OASIS) DNA binding (OASIS) Electron reach fragments (Superfragments) BioBite Cramer Classification Tree (ToxTree) Veerhar/Hermens reactivity rules (ToxTree) Lipinski rules (MultiCase) Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

General characterization by the following grouping schemes: Substance information Predefined Mechanistic Empirical: Chemical elements Groups of elements Natural functional groups AIM (EPA/SRC) Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

Chemical input ProfilingCategory Definition Filling data gap Report Finding Data for SIDS and Other Endpoints Selecting Data Base(s): Toolbox databases  Publicly available  Proprietary databases Toolbox Links to External Databases (DSSTOX) Selecting type of extracting data: Measured Data Estimated Data Both Endpoints Logical sequence of components usage

Measured data summary of the Current Toolbox 1.Biodegradation DB – 745 chemicals 2.Genotox DB chemicals 3.ISSCAN Genotox – 873 chemicals 4.Skin sensitization DB chemicals 5.Estrogen RBA chemicals 6.Bioaccumulation DB – 700 chemicals 7.ECOTOX database – 5071 chemicals 8.ECETOC database – 777 chemicals

Estimated Data Summary of the Current Toolbox 1. Danish EPA DB - data for chemicals

Forming and Pruning Categories: Predefined Mechanistic Empirical Custom Metabolism Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

Forming and Pruning Categories: Predefined OECD categorization US EPA categorization Inventory affiliation Database affiliation Substance type Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage

Data gaps filling approaches Read-across Trend analysis QSAR models Chemical input ProfilingCategory Definition Filling data gap Report Endpoints Logical sequence of components usage

Report the results: QMRF/QPRF IUCLID 5 Harmonized Templates SIDS Dossiers (Data matrix) History of the Toolbox Application User-Defined Reports Documentation: Description of the system Description of Category Editor Chemical input ProfilingCategory Definition Filling data gap ReportEndpoints Logical sequence of components usage