SEURAT-1: Why predictive safety science is important to regulatory acceptance of alternative methods

Transcription

1 1 SEURAT-1: Why predictive safety science is important to regulatory acceptance of alternative methods Ian A Cotgreave Swedish Toxicology Sciences Research Center On behalf of all SEURAT-1 partners 10 th EPAA conference November 19,

2 2 Predictive safety science: Animals, cells, computers: All three or just one or two? The predicament: Is this safe??

3 3 Emerging trends in safety Assessment: A paradigm shift? Pathology Pathway 3

4 4 The Concept of the shift: From molecule to man Via in vitro/in silico MOA 3R wins!! Pathological observations MOA MOA (A+B=C+D)

5 5 So where are the trends in the literature and legislation? Proving concepts to build confidence and gain acceptance Seurat-1 begins

6 Major European research initiative addressing the global long-term strategic target SEURAT - Safety Evaluation Ultimately Replacing Animal Testing. Jointly funded by the European Commission and Cosmetics Europe for 5 years starting from Jan Aim: Provide a blueprint for future implementation of mechanismbased, integrated toxicity testing strategies into modern safety assessment approaches. SEURAT-1 at WC9 Launch of the 4th Annual Report: Tuesday at at the SEURAT-1 corner hosted on the JRC ECVAM booth Meet SEURAT-1, get information about SEURAT-1 presence at WC9 and gather contacts

7 7 Who? Call «FP7-health-2010 Alternative Testing» Projects 7 SEP COACH (secretariat) Working groups Cross-cluster Cooperation Gold Compounds; Data Analysis; Mode of Action; Biokinetics; Stem Cells; Safety Assessment Towards the replacement of in vivo repeated dose systemic toxicity testing

8 8 How? Cosmetics Europe and EC jointly fund multidisciplinary projects to advance the scientific knowledge and technology building blocks - EUR50 M over 5 years Initial focus on a small number of adverse outcome pathways (AOPs) relevant to human safety Proof of Concept studies with well-characterised toxicants that have led to dosedependent adverse effects in humans Once proof-of concept assessments are possible some initial case studies with chemicals of cosmetic relevance will be explored

9 9 What? Harness existing knowledge for chemicals of cosmetic relevance (chemistry and repeated dose toxicity data) Develop a mechanistic understanding of the perturbations in biological processes that lead to adverse effects in humans Develop a toolbox that will enable evaluation of the dose response in these changes and link changes seen to adverse Propose a risk assessment approach based on ensuring consumer use of a chemical in cosmetic products would not lead to adverse effect effects and/or perturbations in the key biological processes that lead to adverse effects

10 10 Why? To evaluate the safety of cosmetic ingredients for repeated exposure in humans without using animals. To gain acceptance for these safety assessments by appropriate regulatory authorities. Best achieved simultaneously!

11 11 The SEURAT Vision and Strategy

12 12 SEURAT-1 Roadmap We are here!

13 13 Examples of available tissues to SEURAT hes hips Adult Keratinocytes 3D skin Hepatocyte-like cells hips-hep Human primary hepatocytes Neuronal cells Cardiac cells 13 MPC

14 14 SEURAT-1 safety assessment Case Studies - Making sense of knowledge and technology 1 st level - Knowledge 1. Challenging the predictive power and robustness of an AOP construct from bile salt export pump inhibition to cholestatic injury 2 nd level - Methodology 1. Investigation of the fibrotic response induced by methotrexate and acetaminophen in the HeMiBio bioreactor 2. Evaluation of valproic acid induced steatosis in HepaRG cells 3. Use of biomarkers to substantiate the read across prediction 4. Developing chemotypes for mitochondrial toxicity 5. Screening of perturbed toxicity pathways by transcriptomics fingerprinting of data poor substances 6. Mode-of-action-based classification model for repeated dose liver toxicity 3 rd level Safety Assessment 1. Read-across using SEURAT-1 evidence 2. Ab initio case study

15 15 Level 1: Development of AOP Descriptions of key events based on in vivo data Three liver-related AOP descriptions Exposure Molecular Event Organelle Events Cellular Events Tissue Events Organ Events Individual Response Populatio n Response Further AOP s related to neurotoxicity Knowledge describe selected AOPs to a sufficient extent so that they can be used as blueprints for system design.

16 16 AOP Protein Alkylation Liver Fibrosis Vinken (2013), Toxicol. 312,

17 17 Level 2: Development of test systems 6 case studies on systems level, prediction goals: Liver fibrosis (HeMiBio) Liver steatosis (NOTOX) Liver, kidney and cardiotoxicity (organ specific) (DETECTIVE) Liver and heart toxicity (non-organ specific -> general AOPs) Mitochondrial toxicity (non-organ specific -> general AOPs) (COSMOS) Liver / non-liver toxicity (JRC) Methodology demonstrate integrated systems for associating a chemical with an AOP category and predicting effect levels.

18 Level 3: development of case studies with regulatory understanding Knowledge describe selected AOPs to a sufficient extent so that they can be used as blueprints for system design.? Systems demonstrate integrated systems for associating a chemical with an AOP category and predicting effect levels. Application / Regulatory Implementation use the information derived from predictive systems to support safety assessment processes and decisions.?

19 19 Context for the level 3 case studies Threshold of toxicological concern Read across Mechanism-based Ab initio Low High Systemic exposure Similarity to current chemical space High NOW: Low freedom to operate Low AMBITION: High freedom to operate Underpinned by international scientific co-operation and regulatory acceptance

20 20 1) SEURAT-1 read-across Case Study Read-across case study to illustrate how new approach data can be used to improve the quality of read-across arguments; i.e. to increase the confidence in the case or to extend the scope of read-across or to expand categories This is an achievable target within SEURAT-1 & is important to demonstrate the usefulness of the project Conceptual framework enables rational integration of evidence, notably existing animal studies & read-across predictions

21 26 How to assess the impact of SEURAT-1 evidence for readacross? Expert judgement of the case before & after the additional evidence is added to give a qualitative assessment of the robustness of the toxicity prediction. The toxicity of the target substance could be predicated by read-across then validated against classical toxicity test data. Target standard for read-across is to fill a registration data element in REACH, i.e. the prediction is (more or less) equivalent to the omitted standard animal study

22 22 Compound Groups for the Scenarios Direct acting toxicants, similar mechanism of action (e.g. perfluorinated acids) Metabolism-driven toxicity (e.g. α,β-unsaturated alcohols) Toxicants with no obvious reactive or specific mode-of-action (e.g. saturated aliphatic alcohols) Toxicants with overt toxicity and a presumed mode-of-action (e.g. short-chain carboxylic acids)

23 23 2) Ab initio Assessment Goal (ultimately): Predict toxicity based on a forward modelling exercise Categorize chemicals with in silico tools, check the categorization in vitro, predict toxicity Requirement: Full representation of mechanisms of toxicity Achievable? For repeated dose systemic toxicity?

24 24 Ab initio Assessment Problem: Single case study -> absence of steatosis. Regulator: I m also interested in 50 other apical endpoints, thank you very much, not useful All the test systems together (steatosis, fibrosis, cholestasis) -> backbone for assessment strategy addressing liver toxicity? Agreement on a few test chemicals among the case studies! (meta-analysis: Comparing dose-response in different test systems). Candidates: Valproic acid, methotrexate, doxorubicin + 1 cosmetic ingredient (to be defined by Cosmetics Europe) Real safety assessment application case?

25 25 Design of ab initio case study Two anchoring exposures chosen for in vitro experimentation, one toxic and o non toxic based on known human PK/exposure data. Testing in a battery of best-practice systems with wide coverage of tissues and A Concentration ranges tested around these initial achoring exposures. Appraisal of the in vitro PK/PD responses of the various systems versus in vivo toxicity data. Re-performance of safety assessment based on in vitro data only. 25

26 26 Areas of concern Ability to determine a priori the human relevant mode of action of primary concern, and thus focus on part 2 of the Seurat safety framework. Ability to distinguish outcomes (likelihood to be within a region of safety) for an exposure scenario. Ability to address the uncertainty of adverse risk to consumers under different consumer populations & exposure scenarios. Must ultimately cover chemical spaces represented in cosmetic ingredients.

27 27 Areas of uncertainties Exposure Application Systemic exposure levels. Extrapolation from in vitro systems to in vivo Use of in vitro assays to move beyond hazard characterisation Population effects. Benchmark use and where we are by using gold compounds which have known in vivo adverse effects at a population level. Combinatorial AOP thinking in multiple mechanisms of action in risk assessment. Mechanisms of action involving homeostatic responses: Damage versus reparation = remodelling.

28 28 Q R S In vitro to in vivo extrapolation How do we make the link between in vitro hazard data and a risk to consumers in vivo? In vitro adaptive/adverse threshold concentration (um) measured as free media concentration Free plasma concentration (um) corresponding to consumer use, from PBPK modelling Exposure due to consumer use mg/kg/day

29 Judson et al Chem Res Tox 29 In vitro to in vivo extrapolation Pharmacodynamics Adverse Effect MOA Key Events Pharmacokinetics Dose-to-Concentration Scaling Function (C ss /DR) Probability Distribution Toxicity Pathway BPADL HTS Assays Probability Distribution for Dose that Activates Biological Pathway PK Model Populations Biological Pathway Activating Concentration (BPAC) Probability Distribution Intrinsic Clearance Plasma Protein Binding

30 30 Combinatorial AOP thinking: High Throughput Risk Assessment (HTRA) Estimated Exposure Low Upper no effect dose HTRA Report Card For Chemical: ABC Exposure / Dose Critical Effect No detect No detect No detect High Pathway / Target / Model CAR/PXR Pathway ER / AR / Endocrine Targets ReproTox Signature DevTox Signature Vascular Disruption Signature Thyroid Cancer Signature 30 Enable further refinement of BPAD use of repeat dose rather than acute assays + estimate actual exposure in vitro rather than nominal concentration

31 SEURAT-1 Conceptual Framework : a generic IATA to combine evidence Hypothesis generation regarding mode of action Pieces of evidence and initial considerations Purpose of the assessment Exposure context Expert knowledge and judgement based on existing evidence / data Type of adversity General adversities Organ specific adversities Definition of relevant dose range Toxicokinetics Assessment of ADME properties Determination of point of departure Toxicodynamics Many biological targets (based on chemical structure, e.g. alkylating agents) Specific targets present in many cells / tissues / organs (e.g. AhR-pathway) Toxicodynamics Target organ: full assessment based on Adverse Outcome Pathway (AOP) 1 Non-target organ: limited assessment Evaluation Overall Assessment (including uncertainties and knowledge gaps) Result Improve assessment if necessary Use of prediction for pre-defined purpose (with consideration of acceptable uncertainty) 1) The steps in the AOP (molecular initiating event, key events) will be assessed using a selection of tools including in silico predictions and in vitro tests.

32 32 So.. perhaps the most improtant factor for success within the SEURAT-1 project is that novel predictive safety assessment science has interacted with regulatory thinking and n from day-1!

33 33 Thanks to: All colleagues within the SEURAT-1 cluster for the last 4 years! Derek Knight, Cathy Mahony, Andy White and Scott Boyer for slide material All colleagues at SWETOX for 3R discussions EPAA for the invitation! 33