VITO NV | Boeretang 200 | 2400 MOL | Belgium | Tel. + 32 14 33 55 11 | www.vito.be

HEALTH

AN VAN ROMPAY1, KAREN HOLLANDERS1, ELS ADRIAENS2, NATHALIE ALEPEE3, VIRGINIE LEBLANC3, HELENA KANDAROVA4, AGNIESZKA DRZEWIECKA5, PRZEMYSLAW FOCHTMAN5, KATARZYNA GRUSZKA5, ROBERT GUEST6, GARETH MAGLENNON6, JANE SCHOFIELD6, JAMIN A. WILLOUGHBY7 and SANDRA VERSTRAELEN1

1VITO nv, Applied Bio & molecular Systems, Mol, Belgium; 2Adriaens Consulting BVBA, Aalter, Belgium; 3L’Oréal Research & Innovation, Aulnay-sous-Bois, France; 4MatTek In Vitro Life Science Laboratories, Bratislava, Slovakia; 5Institute of Industrial Organic Chemistry Branch Pszczyna, Pszczyna, Poland; 6Envigo, Cambridgeshire, United Kingdom; 7Cyprotex US, LLC, Kalamazoo, MI 49008, USA

CON4EI: Funded by CEFIC LRI – AIMT6 – VITO, Eye irritation testing in vitro in practice – Database and testing strategy. Cosmetic Europe contributed in the chemical selection.

Introduction Assessment of ocular irritation is an international regulatory requirement in the safety evaluation of products. Although a number of in vitro assays exist, none are capable of fully categorizing chemicals as stand-alone assays. In CON4EI (2015-2016) the reliability of 8 in vitro test methods and computational models for 80 reference chemicals was assessed and an integrated testing strategy for eye irritation was established. One of the 8 assays was the SkinEthicTM HCE EIT (Reconstructed Human Corneal Epithelium Eye Irritation Test). Results will be published in ATLA.

Methods

Assessment of the acute eye irritation potential is part of the international regulatory requirements for testing of chemicals. The objective of the CON4EI (CONsortium for in vitro Eye Irritation testing strategy) project is to develop tiered testing strategies for eye irritation assessment for all drivers of classification. For this, a set of 80 reference chemicals (38 liquids and 42 solids) was tested with eight different alternative methods. Here, the results obtained with SkinEthicTM HCE EIT (Reconstructed Human Corneal Epithelium Eye Irritation Test) model are shown. The primary aim of this study was an evaluation of the performance of the test method to discriminate chemicals not requiring classification for serious eye damage/eye irritancy (No Category) from chemicals requiring classification and labelling (Category 1 and 2). In addition, the predictive capacity in terms of in vivo driver of classification was investigated. In a second step, it was investigated if SkinEthicTM HCE EIT can be used as part of a tiered-testing strategy for eye irritation assessment. The chemicals were tested twice, the first run was performed by VITO and the second run was performed by L’Oréal. For the SkinEthicTM HCE EIT, 100% concordance in predictions (I versus NI) between the two participating laboratories was obtained. The accuracy of the SkinEthicTM HCE was 97.5% with 100% sensitivity and 96.9% specificity. The results of this study seem promising with regard to the inclusion of this test method in an integrated testing strategy for eye irritation assessment. This research is funded by CEFIC-LRI. We acknowledge Cosmetics Europe for their contribution in chemical selection.

Keywords: CON4EI, SkinEthicTM HCE EIT, eye irritation/corrosion, ocular irritation assay, in vitro novel irritation testing; testing strategy

Abstract

SkinEthicTM HCE EIT discriminates between No Cat and Cat 1/Cat21,2. The 80 blind-coded chemicals (15 No Cat, 27 Cat 2 and 38 Cat1) were tested by the laboratories VITO and L’Oréal.

1Alépée et al., Toxicology in Vitro 2016; 31,43-53; 2Alépée et al., Toxicology in Vitro 2016; 34,55-70.

Results Concordance in prediction: in terms of the classification I versus NI, a 100% agreement in prediction between VITO and L’Oréal was obtained. Reproducibility of the viability: The high reproducibility between the runs between VITO and L’Oréal was reflected in the correlation plots. These plots illustrates the two-by-two relation between the experiments for liquids and solids separately. Dots close to the line of equality (solid line), indicate similar levels of viability for the independent experiments. The dotted lines correspond with the cut-off that distinguishes I from NI. Predictive capacity: The predictive capacity was calculated for each laboratory and for the cumulative results of the two laboratories. A sensitivity of 96.9% and specificity of 100% with an accuracy of 97.5% was obtained overall and in both laboratories separately.

Overall, four false negatives were reported, the following in vivo Cat 2 chemicals (both solids) were consistently predicted NI by the two laboratories: (2R,3R)-3-((R)-1-(Tert-butyldimethylsiloxy)ethyl)-4-oxoazetidin-2-yl acetate (in vivo Cat 2A with CO mean ≥ 1 as driver) and m-Dinitrobenzene (in vivo Cat 2B with Conj mean ≥ 2 as driver).

Boxplot showing the distribution of the viability in function of GHS Driver of classification. The dotted lines correspond with the cut-off for liquids (60%) and solids (50%) that distinguishes I from NI.

Distribution of the liquid chemicals (chem.) according to the drivers of classification (UN GHS Cat 1 and Cat 2) and according to the subgroups (UN GHS No Cat) as defined by Barroso and co-workers (2016).

Distribution of the solid chemicals (chem.) according to the drivers of classification (UN GHS Cat 1 and Cat 2) and according to the subgroups (UN GHS No Cat) as defined by Barroso and co-workers (2016).

Summary • 100 % concordance in prediction between VITO and L’Oreal was obtained. • The accuracy was 97.5% with 96.9% sensitivity and 100% specificity. • Yes, the results seem promising with regard to inclusion of SkinEthicTM HCE EIT in an

integrated testing strategy for eye irritation assessment Cefic LRI-AIMT6-VITO CON4EI project: defines applicability domains, strength and limitations of these 8 in vitro test systems: 1) BCOP (Bovine Corneal Opacity and Permeability 2) BCOP-LLBO (BCOP-Laser Light Based Opacitometer), 3) ICE (Isolated Chicken Eye), 4) EpiOcular-EIT (EpiOcular Eye Irritation Test) 5) EpiOcular-ET50 6) SkinEthicTM HCE EIT (Human Corneal Epithelial), 7) STE (Short Term Exposure) 8) SMI (Slug Mucosal Irritation) to define eye irritation

potential. It identifies tiered-testing strategies for eye irritation assessment.

Classification (Prediction)

In vitro EITL (Liquids)

In vitro EITS (Solids)

Not classified (No category)

Mean tissue viability > 60%

Mean tissue viability > 50%

Classified (Category 1 / Category 2)

Mean tissue viability ≤ 60%

Mean tissue viability ≤ 50%

SkinEthic™ HCE

EITL - Liquid:

Exposure: 30 min

Post Soak: 30 min

Rinse

EITS - Solid:

Exposure: 4 hours

Post incubation: 18 hrs

Rinse

Post Soak: 30 min

Receipt : At least overnight incubation at

37C, 5% CO2, ≥ 95% humidity

Rinsing with PBS without Ca2+ Mg2+

Topical application 37°C, 5% CO2,

≥ 95 % humidity with control and test

substances

Quantification (OD570 reading )

MTT Incubation for 3 hrs at 37C

Formazan extraction (4 hrs at RT or ON at 4C)