REACH, risk assessment and the need for a change in mind set Kees van Leeuwen

1. REACH, risk assessment and the need for a change in mind set Kees van Leeuwen TNO Quality of Life The Netherlands http://www.tno.nl/ k.vanleeuwen@tno.nl

2. CONTENTS • Key elements of REACH • Risk assessment will change • REACH data requirements and hazard assessment options • Why a change in mind set is needed • Concluding remarks

3. I. Key elements of REACH See also: • http://ec.europa.eu/echa/ • McKim-I and II • Van Leeuwen, C.J., B.G. Hansen and J.H.M. de Bruijn. 2007.Management of industrial chemicals in the European Union (REACH). 2007. In: Risk Assessment of Chemicals. An Introduction (2nd edition). Van Leeuwen, C.J. and T.G. Vermeire, eds. Springer Publishers, Dordrecht, The Netherlands, pp 511-551.

4. The REACH legislation • Final version published in OJ L136, volume 50, on 29 May 2007 (Corrigendum to Regulation 1907/2006) • 141 articles • In 15 Titles • With 17 annexes • ≈ 850 pages (including annexes) OJ = Official Journal of the European Union Although it is European legislation, it covers imports and so it will have global implications

5. A new REACH language • CMR = Carcinogenic, Mutagenic, Reprotoxic • CSA/CSR= Chemical Safety Assessment / Report • DU = Downstream User • ERA/HRA = Environmental/Human Risk Assessment • M/I = Manufacturer/Importer • PBT = Persistent Bioaccumulating & Toxic • PPROD = Product and Process Oriented Research and Development • QSAR = Quantitative Structure Activity Relationship • RIP = REACH Implementation Project • RMM = Risk Management Measure • SIEF = Substance Information Exchange Forum • vPvB = very Persistent & very Bioaccumulating

6. Main features of REACH • Single Coherent System for new (non phase-in) and existing (phase-in) substances • Responsibility for all manufacturers and importers • Registration of substances produced/imported above 1 tonne/year • Evaluation by the Agency and Member States • Authorization for substances of very high concern • Restrictions - the safety net • Agency to manage the system

7. Registration under REACH • Aim: to ensure that industry adequately manages the risk arising from its substances (starting at 1 tonne/y) • Method: • Manufacturer/importer obtains adequate data • Provides a registration dossier which includes a chemical safety report for substances above 10 tonnes/y • Submits to authorities (enforcement, transparency) • Increased info requirements according to tonnage (testing proposal) • Reduced requirements for polymers and intermediates

8. Registration timelines

9. Authorization under REACH • Deals with CMR (category 1 or 2), PBT, vPvB, and substances of “equivalent concern” • Focus on PBT or vPvB properties, wide dispersive use and high volumes • Prioritized by the Agency with input from the Member States • Considerations: • The risks and adequate control of these risks • Social and economic benefits/implications of a refusal to authorize • The analysis of alternatives submitted • Available information on risks of any alternative substances or technologies • Commission decision based on Agency opinion

10. Core tools under REACH • The Chemical Safety Assessment (CSA) is the tool used to determine • The Chemical Safety Report (CSR) is the tool used to record/document • The Safety Data Sheet (SDS) is the tool used to communicate

11. Guidance on Annex XV for C&L Guidance on evaluation Guidance mainly for authorities ≈ 650 pages Guidance on Identification of SVHC Guidance on SEA -Restrictions Guidance on Annex XIV inclusion Guidance on Annex XV for restrictions ready not ready

12. Guidance on registration Guidance on pre-registration Guidance on SEA-Authorisation Guidance on data sharing Guidance for authorisation application Guidance mainly for industry ≈ 800 pages Guidance for Downstream users Guidance on intermediates Guidance for articles Guidance on polymers Guidance on C&L notification Guidance on PPRODR ready notready

13. Guidance on substance identification Guidance on CPL Guidance on REACH methods ≈ 4400 pages Guidance on IUCLID Guidance on information requirements Guidance on priority setting ready not ready

14. Comprehensive guidance and tools The total package of REACH (legislation and guidance) comes close to 7000 pages and: • Some guidance documents still not available • Not all guidance is comprehensive and concise • Many methodologies still not harmonized So: Implementation will be a long process A process of “learning by doing” Quality and acceptance take time

15. Conclusions I. Lessons from the past • Science is like life in general: a continuous forgetting and rediscovering old thruths (Brody, 1945) • Legislation is important but implementation is key (De Bruijn, Hansen and van Leeuwen, 2000) • Legislation is only as strong as its implementation and enforcement (US comments on REACH, 2006)

16. TGD EUSES Data evaluation II. Risk assessment will change Exposure assessment Effects assessment Data set Single species toxicity data Emission rates Extrapolation Distribution PNEC PEC Risk characterization PEC/PNEC

17. Hazard information Exposure information REACH: safe use of chemicals Safe use?

18. Exposure scenarios under REACH are:an integral approach to control risks Formal definition: the set of conditions, including operational conditions and risk management measures, that describe how the substance is manufactured or used during its life-cycle and how the manufacturer or importer controls, or recommends downstream users to control, exposures of humans and the environment. These exposure scenarios may cover one specific process or use or several processes or uses as appropriate.

19. Initial ES • Short title • Operational conditions • Risk management measures If risk not demonstrated to be controlled Building Exposure Scenarios Decisions by M/I • Refine hazard assessment • Refine exposure assessment • Modify RMM or OC • Advise against use Final ES • Short title • Operational conditions • Risk management measures

20. REACH and the use of test animals • Testing on vertebrate animals shall be undertaken only as a last resort (art. 25) • Information may be generated by other means than tests, in particular through in vitro methods, (Q)SARs and read-across (art. 13)  Legislative text + guidance should limit use of animals and prevent box-ticking

21. Conclusions II • RMMs are the start of a RA. The focus is on exposure • It requires multidisciplinary and integrative thinking & expertise right from the start • Dialogue up and down the supply chain between actors in the supply chain is key to success! • It requires paradigm shifts: • effects-based → exposure-driven • risk assessment → risk management • It requires detailed information on use and exposure of substances (in products) which is generally not available to the authorities (Haigh and Bailly, 1992!) • Expertise in and outside industry is scarce (aging population) • Animal welfare considerations and resources (time, cost and laboratory capacities) require other approaches to hazard assessment

22. III. REACH data requirements and hazard assessment options Annex VII (≥ 1 tonne per year) • Physicochemical properties • Human health: in vitro irritation, sensitization, mutagenicity, acute toxicity (one route) • Environmental: acute aquatic toxicity (daphnia, algae), biodegradation Annex VIII (≥ 10 tonnes per year) • Human health: including in vivo irritation, and 28-day repeat dose studies • Environmental: acute toxicity fish, fate studies (hydrolysis, adsorption / desorption) Annex IX ( ≥ 100 tonnes per year) • Long term, repeat dose, chronic toxicity, fate etc Annex X (≥ 1000 tonnes per year) • Further long term, repeat dose, chronic toxicity, fate etc

23. 1-10 t Annex VII 30 k€ 20,000 REACH data requirements: a phased approach Estimated costs per dossier Estimated number of substances 10-100 t Annex VIII 100-1000 t Annex IX 70-330 k€ 4,600 > 1000 t AnnexX 400-875 k€ 2,900 400k-2 M€ 2,600 2008…..2010……………..2013……………………………………………….2018

24. Estimated test animal need (van der Jagt et al., 2004)

25. REACH ambition: reduction of animal use June 2013 Registration Annex IX (2.900 substances) Animal use June 2018 Registration Annex VII & VIII (24.600 substances) 2010 2008 2030 2020 Dec 2010 Registration Annex X (2.600 substances) + CMR (cat 1+2) > 1 tonne/y (? substances) + very toxic (R50/53) > 100 tonnes/y (? substances) Dec 2008 Pre-registration

26. Full replacement of animal testing (SCCNFP, 2004)

27. Hazard Characterization Risk Characterization U.S. NRC Toxicity Testing in the 21st Century (2007) Dose Response Assessment Chemical Characterization Mode of Action Population Based Studies Compounds Dose Response Analysis for Perturbations of Toxicity Pathways Affected Pathway Assess Biological Perturbation Calibrating in vitro and human Dosimetry Exposure Guideline Measures of dose in vitro Metabolite(s) Human Exposure Data Exposure Assessment

28. U.S. NRC Toxicity Testing in the 21st Century Dose Response Analysis for Perturbations of Toxicity Pathways Toxicity Pathway: A cellular response pathway that, when sufficiently perturbed, is expected to result in adverse health effects

29. U.S. NRC (2007) New Paradigm: Exposure Tissue dose Biological interaction Pertubation Normal Biologic Function Biological inputs Early Cellular Changes Adaptive Stress Responses Cell injury Morbidity and mortality Activation of Toxicity Pathways

30. Options for Future Toxicity Testing Strategies (Modified after the US NRC 2007)

31. National Academy of Sciences Report (2007) Toxicity Testing in the Twenty-first Century: A Vision and a Strategy Science: Feb 15, 2008

32. Complete replacement with in vitro or (Q)SAR methodologies is not possible for the most relevant endpoints (SCCNFP, 2004) The USNRC described a vision and a strategy but the expectations to replace animal tests seem to be running ahead of scientific reality Industry concern: new technologies can only be used to replace traditional testing approaches unless relevance for safety assessment has been demonstrated Dose / exposureinformation is the critical component in hazard and risk assessment independent of the technology used So, what do we do in the next 20 years? Focus on effects only or also on exposure? And what can we learn from current regulatory programs? Options for Future Toxicity Testing Strategies

33. Mortality • Systemic Toxicity • Disease • Cancer Reproductive fitness • Viable Offspring • Fertility Current regulatory toxicology testing paradigm To generate in vivo animal data for all possible outcomes to determine which of all possible effects are relevant Developmental impairment • Terato • Prenatal Deficits BATTERY of Animal Testing (Jones and Bradbury, USEPA, 2005)

34. Current regulatory testing strategies

35. Cl Cl Cl OH OH Cl C C Cl Cl C C2Cl3 C2Cl3 C2Cl3 Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl OH Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl Future testing paradigm CHEMICAL INVENTORIES Existing data Read-across Categories QSARs TTC in vitro Molecular Interactions Biochemical Responses Cellular Responses Prioritization/Screening Tissue/Organ Function EFFICIENT, FOCUSSED ANIMAL TESTING Adverse Outcomes (Jones and Bradbury, USEPA, 2005)

36. A paradigm shift is needed In the context of regulatory programs, the challenge is to move in a scientifically credible and transparent manner from a paradigm that requires extensive hazard testing to one in which a hypothesis- and risk-driven approach can be used to identify the most relevant in vivo information (Bradbury, Feytel and Van Leeuwen, 2004)

37. Intelligent Testing Strategies (ITS) The most efficient way to carry out hazard and risk assessments of large numbers of chemicals, while reducing costs to industry and minimising animal testing, is to obtain the necessary information by means of intelligent testing strategies (ITS). Intelligent testing strategies are integrated approaches comprising of multiple elements aimed at speeding up the risk assessment process while reducing costs and animal tests (Bradbury, Feytel and Van Leeuwen, 2004)

38. INTELLIGENT TESTING STRATEGIES (Q)SAR In vitro Grouping Exposure

39. Experiences from the US HPV Challenge Program

40. Experiences from the US HPV Challenge Program • Until now 2200 chemicals have been evaluated • For the human health data 88% of the missing data was estimated using read-across. • For the environmental data 83% of the missing data was estimated using either QSAR or read-across.

41. Expectations in Europe: REACH saving potential of ITS(Van der Jagt et al., 2004; EUR report 21405) • Testing costs: € 800-1130 million • Number of animals: 1.3-1.9 million The most likely scenario for REACH according to the JRC: 2.6 million vertebrate animals and € 1.5 billion for testing(http://ecb.jrc.it/ )

42. Towards a 7-R strategy implementing ITS • Risks Focus on risks (include exposure) • RepetitiveA tiered approach should be applied, going from simple, to refined or comprehensive, if necessary, to quickly assess chemicals of low concern and to prevent animal testing. • Relatives The focus should be on families or categories of chemicals (a group-wise approach) using read-across, QSARs and exposure categories: move away from the chemical-by-chemical approach. • Restrictionof testing (waiving of testing) where possible and carry out in-vivo testing where needed in order to prevent damage to human health and/or the environment.The strategy should also encompass the current 3-R strategy of: • Replacement (substitution) • Refinement (reduce suffering and distress) • Reduction

43. OECD TOOLBOX Tracks and Workflow

44. Problems and Challenges “A major focus for the future of computational toxicology will be integration and analysis of large data sets. The current state of toxicity databases is something of a mess. There are a number of databases, each with differing content, architecture, and searchability, that makes the task of integration extremely difficult.” (Jacobs and Marnett, 2007) New toxicity data models provide the means, and the data entry tools provide the mechanism, for migrating previously inaccessible data and new data into a standardized, relational format (Richard, Yang and Judson, 2008)

45. ConclusionsIII • ITS / innovative toxicology is on the political agenda, partnerships with industry have been set up and has a great animal-saving and cost-saving potential • A paradigm shift is needed from extensive animal testing to efficient, focussed animal testing applying the 7-R approach • Expectations to replace animal tests seem to be running ahead of scientific reality (CSTEE & SCCNFP, 2004 and Greim et al. 2006) • A long-term commitment for further scientific work is needed for the next two decades • Timeline driven regulatory programs such as REACH cannot wait. Tools and approaches are urgently needed in the next few years. The OECD Toolbox is a major step to fulfil the needs for category approaches • The availability and accessibility of high quality in vivo data can be a limiting factor both for the development of in silico and in vitro methodologies • Industry involvement and leadership is key and so is the change in mind set for all stakeholders

46. IV. Why a change in mind set is needed(three scenarios for implementation) • The past • The present (REACH as planned) • The future (REACH as needed)

47. Public Availability of Data on HPVCs(Allanou, Hansen and van Der Bilt, 1999) • 14 %:base set data • 65%: less than base set • 21%: no data 86%

48. Implementation of EU measures Tour de EU Agreed risk reduction strategy Agreed RAR Draft RAR Priority list 1-? y SETAC BRIGHTON CONFERENCE (2000) JdB, BH, KvL