Clustering & Reading Across Structurally Similar Chemicals
Exponent’s scientists have the technical expertise, expert judgment, and project management capabilities in chemistry, mammalian metabolism and toxicology, and risk assessment to develop defensible cluster analyses and subsequently read across to fill data gaps for carcinogenicity, developmental and reproductive, chronic and acute effects study data.
The biological effects of a chemical compound, for which there may be limited toxicology information, and/or data gaps, can be predicted by clustering structurally similar chemical compounds that share common physical-chemical properties and metabolic pathways. Cluster analysis provides the basis to determine whether the existing data in a cluster can be bridged to others within the cluster by reading across to the most sensitive toxicology endpoint in the existing study database. If a class of structurally similar compounds is not obvious, the use of (Quantitative) Structure Activity Relationship [(Q) SAR] modelling may be needed as the first step in identifying compounds to be included in the cluster.
Clustering of structurally similar compounds provided the foundation for eliminating unnecessary animal testing and fostering efficient safety assessments for both the US and the Organization for Economical Cooperation and Development (OECD) High Production Volume programs. EPA now accepts risk assessments based on cluster analysis for both Toxic Substances Control Act (TSCA) and New Chemical Federal Insecticide and Rodenticide Act (FIFRA) regulatory submissions. Cluster analyses also have been encouraged and accepted by the European Union for the Registration, Evaluation and Authorization of Chemicals (REACH) regulatory initiative. Because it is an efficient means to reduce the number of tests and in vitro/in vivo assays, regulators, industry and NGOs have embraced the use of cluster analysis to conserve resources – that is, the costs of conducting studies, the use of laboratory animals, and the time needed to conduct and review toxicological studies. Clustering and read-across methodologies will likely play a key role in grouping and limiting the required testing for classes of structurally similar compounds under the future TSCA Reform testing initiative.
Our cluster analysis experience includes development of successful regulatory strategies that can be based primarily on the interpretation of existing biologically relevant physical-chemical, toxicology and metabolism data. However, when necessary to do so, we can also utilize publically available models, such as the OECD (Q)SAR Tool Box and EPA Sustainable Futures P2 Framework Methods, to supplement existing toxicology data. We also have experience with Oncologic® for carcinogenic estimations and EpiSuiteTM to estimate biologically relevant physicochemical properties such as lipophilicity and water solubility.
Our applied cluster analyses accomplishments include TSCA regulatory support for EPA New Chemicals Programs Premanufacture Notices (PMNs). Under FIFRA, we have also provided clients with technical and administration support related to cluster analysis for a broad array of chemically diverse inert formulation ingredients.