5 TH INTERNATIONAL WORKSHOP ON GENOTOXICITY TESTING

1. 5TH INTERNATIONAL WORKSHOP ON GENOTOXICITY TESTING August 17-19, 2009 Topic 3: In Vitro Test Approaches with Better Predictivity Stefan Pfuhler (Procter & Gamble, USA) ICEM Florence Aug 20-25, 2009

2. Workshop participants: Jan van Benthem (RIVM, The Netherlands), Rapporteur Gladys Ouédraogo (L’Oreal, France) RaffaellaCorvi (ECVAM, Itraly) Rodger Curren (IIVS, USA) Kerry Dearfield (USDA FSIS, USA) Azeddine Elhajouji (Novartis, Switzerland) Mick Fellows (AstraZeneca, UK), Co-chair Paul Fowler (Covance, UK) Roland Frötschl (BfarM, Germany) Ludovic Le Hegarat (Inserm, France) Toshio Kasamatsu (KAO, Japan) Hajime Kojima (JaCVAM, Japan) Stefan Pfuhler (Procter & Gamble, USA), Chair Andrew Scott (Unilever, UK) Gunter Speit (Univ Ulm, Germany) ICEM Florence Aug 20-25, 2009

3. Background: • High rate of misleading positive results in our current battery of in vitro tests. As high as 80% when mammalian cell assays are combined (i.e. Chrom Abs or MLA) • 7th Amendment to EU Cosmetics Directive (3-test battery): Marketing and testing ban on ingredients tested in vivo came into force March 2009 • -> Valuable compounds may be unnecessarily discarded • Chemicals: REACH (3-test battery) • -> misleading positive results will lead to unnecessary in vivo follow-up tests ICEM Florence Aug 20-25, 2009

4. Presentations: • Which cell lines should we be using • Paul Fowler presented results from the Colipa “false positives” project which compares V79, CHO, CHL, TK6, HepG2, L5178Y, HuLy. • Mick Fellows presented data generated in conjunction with the micronucleus OECD guideline finalization • Azeddine Elhajouji presented data generated at Novartis for comparison of several cell lines • Ludowig Le Hegarat presented data generated with HepaRG cells • 2) Promising new approaches • Rodger Curren presented micronucleus data generated in human 3-dimensional skin models (Colipa 3D skin project plus additional data from IIVS, PG, MatTek) • Gladys Ouédraogo presented Comet assay data in 3-dimensional skin models (Colipa 3D skin project) • Hajime Kojima presented 3D Comet assay data generated by JaCVAM ICEM Florence Aug 20-25, 2009

5. Questions that were addressed: • Choice of cell line: • Are there restrictions for the choice of the cells used for testing? • Should there be recommendations for the choice of a cell line? Can that be done now or are more data needed? • New approaches: • How do we rate the status of the genetox assays performed with 3D human skin equivalents? • What is the applicability domain of that assay? • What data would the team like to see before considering this a valid assay? ICEM Florence Aug 20-25, 2009

6. Resorcinol (3hr + 21hr, +S9) ICEM Florence Aug 20-25, 2009 Data presented by Paul Fowler

7. 3D skin Models in Genotoxicity testing, results from phase 2: Evaluation of Coded Chemicals Three chemicals were sent coded from Covance UK to P&G, IIVS, L’Oreal. Each chemical was tested in at least 2 independent studies in each lab. Results sent to ECVAM for decoding and analysis ICEM Florence Aug 20-25, 2009 Data presented by Rodger Curren

8. Data presented by Rodger Curren P&G IIVS L’Oreal 48h 72h Open symbols are % binucleated cells; Closed symbols are % micronucleated cells ICEM Florence Aug 20-25, 2009

9. IWGT consensus statements • Data were presented indicating that p53 compromised rodent cell lines over-estimate genotoxic potential in the micronucleus test. Therefore, IWGTsuggests using p53 competent cells in in vitro MN- or CA-tests. • It has been demonstrated that cell line stability and source can affect the outcome of genotoxicity assays. Therefore IWGTrecommends to adhere to good cell practice, characterize all new cells, check regularly for drift and work from low passage stocks. A common genotoxicity cell bank with fully characterized stocks of all cells would be useful. • Genotoxicity testing in 3D human reconstructed skin (MN test and Comet assay ) is a promising new in vitro test for dermally applied chemicals ICEM Florence Aug 20-25, 2009

10. Statement 1: Use of p53 competent cells: • The data presented at the IWGT from the OECD MNvit test showed that all cell types correctly identify clastogens and aneugens. • However, from the data presented on the “misleading compounds” using the MNvit test there were clear differences in the response of the various cell types. • The group agreed that the above would also be applicable to the CA based on the compatibility of both assays. • HepaRG is a promising model as the cells appear to have better phase I and II metabolizing potential than other cell lines. However, further evaluation is required to confirm the value of this model for genotoxicity testing ICEM Florence Aug 20-25, 2009

11. Statement 3 • Use of 3D human reconstructed skin in MN and Comet: • The advantage of the model is that it resembles the properties of human skin (barrier function, metabolism) and the route of exposure is relevant for dermally applied chemicals (e.g. cosmetics) • The data presented show that the 3D skin Micronucleus assay is further advanced and inter and intra-lab reproducibility has been demonstrated. • IWGT agreed that the Comet assay should be further evaluated. The Comet assay is seen as a valuable addition as it is not dependent on cell proliferation and covers a wider spectrum of DNA damage. ICEM Florence Aug 20-25, 2009

12. Statement 3 • Further remarks: • The metabolic capacity needs to be further evaluated (ongoing) • It would be valuable to capture the kinetics of penetration and toxicity in order to establish the ideal sampling time(s) for the Comet assay • Recommendations on the use of appropriate vehicles should be established • It was agreed that 3D genotoxicity models, once validated, will be useful (at least) to follow up on positive results from standard in vitro assays for dermally applied chemicals. • The applicability domain will be established once the validation is completed ICEM Florence Aug 20-25, 2009