NEXT GENERATION RISK ASSESSMENT FOR CONSUMER SAFETY OF COSMETICS: A CASE STUDY APPROACH

Transcription

1 NEXT GENERATION RISK ASSESSMENT FOR CONSUMER SAFETY OF COSMETICS: A CASE STUDY APPROACH CARL WESTMORELAND SAFETY & ENVIRONMENTAL ASSURANCE CENTRE, UNILEVER, UK

2 CAN WE USE A NEW INGREDIENT SAFELY? Can we safely use x% of ingredient y in product z?

3 MAXIMISING USE OF EXISTING INFORMATION AND NON- ANIMAL APPROACHES All available safety data in silico predictions Exposure-based waiving approaches History of safe use Read across Use of existing OECD in vitro approaches

4 ICCR PRINCIPLES OF RISK ASSESSMENT WITHOUT ANIMAL TESTING

5 ICCR NINE PRINCIPLES OF NGRA Main overriding principles: The overall goal is a human safety risk assessment The assessment is exposure led The assessment is hypothesis driven The assessment is designed to prevent harm Principles describe how a NGRA should be conducted: Following an appropriate appraisal of existing information Using a tiered and iterative approach Using robust and relevant methods and strategies Principles for documenting NGRA: Sources of uncertainty should be characterized and documented The logic of the approach should be transparently and documented

6 A CASE STUDY APPROACH IMAGINE WE HAD NO DATA...

7 EXPOSURE-LED Exposure scenario: Worst case in US µg/cm2 Used one time per day Skin surface area: cm 2 (95 percentile) Amount of product used per day: 5.18 g/day Amount of ingredient in contact with skin: 155 mg/day Laundry scenarios Systemic exposure (mg/kg bw Local dermal exposure per day) (ug/cm2) Pre-Treatment Main wash Residues on clothes Total (Main wash+ residues on clothes) Total (pre-treatment+ main wash+ residues on clothes) Skin penetration Caffeine Free Concentration PBPK model predicted free concentrations (µm) plasma heart liver brain adipose Parameters Model 4 output for Model 4 output for Pre-treatment Main wash Cmax (ug/ml) 0.1 (0.68 µm) 0.46 (3.15 µm) Tmax (h) AUC 0-inf (ug-h/ml) AUC 0-t (ug-h/ml) Cmax Liver 0.01 (0.068 µm) (0.4 µm)

8 A TIERED AND ITERATIVE APPROACH See Also: Poster P25-04 Teixeira et al High-throughput and non-depletive quantification in 3D liver microtissue in vitro assay e.g. Atlas of Molecular Initiating Events See Also: Poster P05-13 Allen et al Computational approaches for predicting molecular initiating events

9 PREDICTING SKIN SENSITISATION AOP for skin sensitisation The SARA Weight of Evidence (WoE) human potency model* is a highdimensional probability distribution describing data from the following sources: DPRA OECD TG442D (cys/lys depletion) KeratinoSens TM OECD TG442C (EC 1.5, EC 3, IC 50 ) H-Clat OECD TG442E (CD54 EC 200, CD86 EC 150, CV75) U-SENS TM OECD TG 442E (CD86) * Reynolds, J, MacKay C, Gilmour N, Miguel-Vilumbrales D and Maxwell G (Submitted for publication: Computational Toxicology) Probabilistic prediction of human skin sensitiser potency for use in next generation risk assessment

10 PREDICTION OF PROBABILITY OF SENSITISATION OCCURRING IN HRIPT FOR CASE STUDY CHEMICALS DNCB Methyl heptine carbonate Coumarin Lactic acid = No Expected Sensitization Induction Level ( See Also: Poster P13-05 Vandenbossche et al Evaluation of an integrated strategy for skin allergy risk assessment using six ingredients and two cosmetic products

11 PROBABILITY OF CONSUMER BECOMING SENSITISED Face cream Shampoo Probability of sensitising one individual See Also: Poster P13-05 Vandenbossche et al Evaluation of an integrated strategy for skin allergy risk assessment using six ingredients and two cosmetic products

12 TOXCAST: COMBINING IN VITRO ACTIVITY AND DOSIMETRY Rotroff, et al. Tox.Sci 2010 Slide from Dr Rusty Thomas, EPA, with thanks

13 UNILEVER/US EPA : JOINT CRADA CASE STUDY CHEMICALS 1. Caffeine 2. Curcumin 3. Bisdemethoxycurcumin 4. Tetrahydrocurcumin 5. 6-Gingerol 6. Coumarin 7. Hydroquinone 8. Doxorubicin

14 CELL STRESS PANEL Stress pathways Mitochondrial Toxicity Oxidative Stress DNA damage Inflammation ER Stress Metal Stress Osmotic Stress Heat Shock Hypoxia Cell Health 14 chemicals, including Low-risk compounds: Phenoxyethanol Niacinamide Caffeine Known high-risk compounds: Doxorubicin Diclofenac Troglitazone Platform Technology: High content imaging Cell line: HepG2 Timepoints: 1, 6 & 24 hours Calculate free concentration Use in vitro exposure models: Groothuis et al (2015) Toxicology, 332, 30-40

15 CELL STRESS PANEL Coumarin 6 hours 24 hours Doxorubicin 6 hours 24 hours

16 HIGH THROUGHPUT TRANSCRIPTOMICS NOTEL* is the derived concentration of a compound that does not elicit a meaningful change in gene expression (i.e. the threshold of the concentration that elicits minimal mechanistic activity). Thomas et al (2013). Toxicol Sci, 134, Recommended approaches in the application of toxicogenomics to derive points of departure for chemical risk assessment Farmahin et al (2017) Arch Tox 91, *NOTEL = No observed transcriptional effect level

17 Accumulation HIGH THROUGHPUT TRANSCRIPTOMICS Fold change >1.5 Genes in pathway >3 Fishers exact test >0.1 (3 independent experiments) HepG2 BMD minimum 295 BMD medium 450 BMD maximum 962 POD - NOTEL value MCF7 BMD minimum 117 BMD medium 167 BMD maximum 290 Conc Phenoxyethanol (µm) SEAC HepaRG BMD minimum 237 BMD medium 344 BMD maximum 913 See also poster: P12-35 Nepelska et al Unilever Information: Internal Use 17

18 Log (10) Consumer Exposure (µm) NGRA EXPOSURE-DRIVEN CASE STUDIES Log (10) POD (µm)

19 Log (10) Consumer Exposure (µm) NGRA EXPOSURE-DRIVEN CASE STUDIES Log (10) POD (µm)

20 ICCR PRINCIPLES OF RISK ASSESSMENT AND WHAT WE RE LEARNING FROM CASE STUDIES Importance of understanding consumer exposure including the relevance of metabolism Non-standard, bespoke data generation driven by the risk assessment question Ensuring quality, robustness of non-standard (non-tg, non-glp?) work. In silico modelling approaches and bespoke in vitro solutions Importance of defining points-of-departure and understanding adverse vs. adaptive responses Understanding uncertainty in risk assessments to allow informed decision-making

21 ACKNOWLEDGEMENTS AMMS Prof Peng Jiabin Guo Haitao Yuan US EPA Josh Harrell Richard Judson Imran Shah Rusty Thomas Cyprotex Paul Walker SEAC Unilever The SEAC NGRA Team with special thanks to the following for their help with this talk: Maria Baltazar Sarah Cooper Matt Dent Nicky Gilmour Steve Gutsell Julia Head Cameron MacKay Sophie Malcomber Gavin Maxwell Alistair Middleton Victor Oreffo Gurjit Phgura Paul Russell Hequn Li Jin Li Paul Carmichael Joe Reynolds Andy White Unilever Information: Internal Use 21