Concept for a generic information strategy for PBT and vPvB assessement of UVCB substances

Within the context of the European REACH (Registration, Evaluation, Authorization, and restriction of Chemicals) Regulation, the SVHC Roadmap aims to identify the highest priority potential substances of very high concern (SVHC) before 2020. With regard to environmental concerns, substances will be assessed for persistency, bioaccumulation, and toxicity (PBT) properties. One group of substances which are more difficult to assess for PBT properties are the UVCBs: substances of unknown or variable composition, complex reaction products or biological materials.
This report aims to set a first step in the process of developing a generic information strategy for the PBT assessment of UVCBs. Two examples of UVCB substances are used in this report to illustrate specific parts of the information strategy: petroleum and coal stream (PetCo) substances and chlorinated paraffines.
The proposed strategy is based on current guidance documents, experience from case studies, and current state of science. The concept follows two lines of evidence: a constituents approach and a substance approach. The first line is based on the identification of the constituents of the UVCB, followed by grouping (blocks) of constituents with similar properties (e.g. physico-chemical, fate, mode of action), and the selection of representative structures for each block. The representative structures are individually assessed for PBT properties using in silico predictions and experimental work. In general, the characterisation of the exact molecular structure of constituents in UVCB substances is challenging or extremely time consuming due to the complexity of the mixtures, even with the latest analytical techniques available. However, for most UVCBs generic structures of the constituents can be generated. This generic structures can be used for blocking constituents with similar properties and to generate representative structures for each block. It is therefore suggested that the generation of generic structures can be used as a first tier assessment of UVCBs. Yet, if information on specific constituents with potential PBT properties are available these should be assessed as well. Predictions of the physico-chemical properties of the representative structures together with (Q)SAR PBT models are then used for the PBT assessment. This information can further guide the experimental work on representative structures and therefore reduce animal testing. Together with the whole substance testing it provides weight-of-evidence on the PBT properties of the UVCB substance.
The second line assesses the UVCB substance as a whole, which can be followed by fractionation of the substance in fractions before experimental testing. With the later concept the main challenges are in the testing methodologies as many UVCB substances are hydrophobic and therefore alternative testing methods are needed.
In this concept there are a number of challenges which are in the field of i) testing methodologies, ii) the applicability domain of the in silico models, and iii) each UVCB will have its own specific challenges and therefore a case-by case approach is needed. The current momentum on the development of the strategies for UVCB assessment needs multi-stakeholder involvement from regulators, industry and academia.