1. Fragrance in our lives – a choice worth making? Allergies in the population: The EDEN epidemiology study Thomas L. Diepgen Univ. of Heidelberg Dept. of Social Medicine, Occup. & Environmental Dermatology Germany

2. Background: The exposure to cosmetics is massive and it is inevitable that adverse effects occur. Allergic contact dermatitis from cosmetics is quite common. Fragrances and preservatives are the most culprits.

3. Outline “Sensitive” to cosmetics How to diagnose Exposure data The EDEN study Clinical consequences Fragrance allergies in the population

4. How frequent are adverse effects from fragrances? A considerable percentage of the population report adverse health effects or irritation from fragrance products UK: 23% of women and 13.8% of men experience some sort of adverse reaction to a personal care product over the course of a year. 1 U.S. population reports 2 –Scented products on others irritating 30.5% –Adverse health effects from air fresheners 19% –Irritation by scented laundry products 10.9% 1 Orton & Wilkinson Am J Clin Dermatol 2004 2 Caress & Steinemann J Environ Health 2009

5. Adverse effect to cosmetics does not mean contact sensitisation 1998 Glostrup Allergy Study (DK n=482) 1 –51.5% report skin redness, itching or rash after the use of a cosmetic product however only –2.3% sensitised to Fragrance Mix I –1.3% sensitised to balsam of Peru 1 Thyssen et al. BJD 2009

6. CompoundFormulaCAS number: Oak moss absoluteExtract IsoeugenolC 10 H 22 N 2 97-54-1 EugenolC 10 H 12 0 2 97-53-0 Cinnamic aldehydeC9H8OC9H8O104-55-2 HydroxycitronellalC 10 H 18 O 2 107-75-5 GeraniolC 10 H 18 O106-24-1 Cinnamic alcoholC 9 H 10 O104-54-1 α-amyl cinnamic aldehydeC 14 H 18 O122-40-7 Sorbitan sesquioleate (emulsifier)8007-43-0 Cross reactions: Balsam of Peru, Cassia oil, Cinnamon. Cloves, Citronella candles, Ethylene bassylate, Tiger balm Screening fragrance allergy: Fragrance Mix I Larsen WG, Arch Dermatol 113: 623-626 (1977)

7. Patch test data from Europe

8. Frequencies of positive patch test reactions to allergens of the European standard series in 10 European centres 1 AllergenTotalMF Nickel sulfate (5%)17.94.825.5 Fragrance Mix I (8%) 9.77.211.1 Cobalt chloride (1%) 5.93.7 7.2 Potassium dichromate (0.5%) 4.65.5 4.1 1 Bruynzeel et al Contact Dermatitis 53:146-9 (2005)

9. The 10 most frequent sensitisers in the standard series 1 AllergenFrequencyRange Nickel sulfate 5%17.912.6 – 24,6 Fragrance mix 8%9.75.0 – 12.6 Balsam of Peru 25% *6.02.8 – 10.9 Cobalt chloride 1%5.92.7 – 9.9 Potassium dichromate 0.5%4.61.0 – 11.0 Colophonium 20%4.02.2 – 5.2 p-Phenylenediamine base 1%3.91.3 – 7.4 Thiuram mix 1%3.21.4 – 7.0 Neomycin sulfate 20%3.01.6 – 7.7 Lanolin (wool wax alc.) 30%2.90.6 – 5.7 1 Bruynzeel et al. Contact Dermatitis 53:146-9 (2005) * Natural extract from the tree Myroxylon pereirae

10. Fragrance mix I (FM I) and II (FM II) A new fragrance mix II (contains 6 fragrance allergens) has been developed to supplement FM I A multicenter study has shown that FM II detected about 30 % more fragrance allergic patients 1 The European baseline patch test series has recently be amended to include FM II 2 1 Frosch et al. Contact Dermatitis 52: 207-15 (2005) 2 Bruze et al. Contact Dermatitis 58: 129-33 (2008)

11. Hydroxylisohexyl 3- cyclohexene carboxaldehyde (Lyral) (2.5%) Citral (1.0%) Farnesol (2.5%) Coumarin (2.5%) Citronellol (0.5%) α-hexyl cinnamaldehyde (5%) Fragrance Mix II (14% in pet.) plus Lyral 5% in the standard series

12. No. 7, PI: 5% Lyral + Neg.IR?++++++ 55

13. Germany: IVDK – DKG 2009 Allergens 2007 (n=11.099) und 2008 (n=10.745) Allergen20072008 Nickel sulfate17,3 Cobalt chloride7,47,1 Balsam of Peru6,76,6 Fragrance Mix6,5 Dichromate6,14,9 Fragrance Mix II4,64,7 Colophony3,93,8 percentages

14. Germany: IVDK – DKG 2009 Allergens 2007 (n=11.099) und 2008 (n=10.745) Allergen20072008 Propolis2,22,5 Lyral2,22,3 Thiuram-Mix2,02,3 Wollwachsalkohole2,02,3 Methyldibromoglutaronitrile MDBGN 3.92.1 MCI / MI (Kathon CG)2,1 Epoxy resin1,41,7 percentages

15. Epidemiology of fragrance allergy Most knowledge about contact dermatitis is derived from –clinical case reports, –clinical studies of in- and out-patient groups, –statistical compilations of patch test reports, and –from studies of small outbreaks of skin diseases at the work place. All these data sources have their limitations and the data must be interpreted carefully.

16. Contact sensitisation to fragrances in the general population Systematic review (19 studies) Sample size 82 – 2545 Participation rate 21.3 – 83.1 % True test, Epiquick (8); standard patch test (11) Reading day 3 but in 6 studies only day 2 Average weighted prevalence in adults (13 studies): –Fragrance Mix I 3.7% –Balsam of Peru 1.6% No study on Fragrance Mix II 1 Thyssen et al. BJD 2009

17. Contact sensitisation to fragrances in the general population 1 Thyssen et al. BJD 2009 AuthorYearNSampleF*M* Dotterud20071236GP (N)1.71.8 Svedmann2007484Stent pat (S)4.95.0 White20072545GP (Thai)2.72.1 Bryld2003627Twins (DK)1.3 Nielsen2001469GP (DK)3.21.0 Greig2000219Volunteers (Aus)4.53.5 Nielsen1992567GP (DK)1.01.1 Seidenari1990593Male cadets-0.5 * Percentage of positive reaction to FM I

18. 1146 schoolchildren in DK 12-16 yrs of age Positive reactions to FM I – Girls1.6 % – Boys2.1 % Contact sensitisation to fragrances in adolescents Mortz et al. Acta Derm Venereol 2002

19. Fragrance allergy – a challenge to interpretation Problems: 1.Irritant versus allergic patch test reaction. 2.Contact allergy versus allergic contact dermatitis 3.Did we patch tested the responsible allergen?

20. Do we test with the right FM I ? Patch test of FM I (True Test and Trolab) in 5006 eczema patients* The clinical relevance was recorded as present or past 495 (9,9%) positive reactions –187 positive according to both patch test systems –277 positive only to Trolab – 31 positive only to True Test Significant difference between the 2 tests in frequency of positive reactions (p<0.0001). *personal communication Andersen 2010

21. Do we test with the right FM I ? If a patient had a positive reaction to one of the FM I tests, the individual constituents of FM I (Trolab®) were tested: –+ reactions show only few reactions to the individual constituents of the FM I (True Test 54%, Trolab 29%) –Only ++ or +++ reactions show also in a high percentage positive reactions to the individual constituents of the FM I (True Test: 83.3 %; 100%; Trolab: 74.6%; 96.7%) *personal communication Andersen 2010

22. No. 3, PI: 8% Fragr. Mix + Neg.IR?++++++ 19

23. Individuals (%) reported a rash from scented products Dermatitis patients –with positive patch test to FM I62.8% –with negative patch test to FM I40,8% General population31.8% 1 Johansen et al. Am J Contact Derm 1998

24. 7th amendment of the European Union (EU) Cosmetics Directive & the Detergents Regulations of the EU (2005) cosmetic products and detergents must be labelled for 26 individual named fragrances, if their concentrations in the product exceed –leave-on products 0.001% (> 10 ppm) –rinse-off products 0.01% (> 100 ppm) –household products 0.01% (> 100 ppm) Screening by using FM I and II

25. Exposure data Exposure pattern to fragrance of the U.K. consumer was assessed (300 products) 6 most frequently labelled fragrances: –linalool (63%), –limonene (63%), –citronellol (48%), FM II –geraniol (42%), (FM I) –butyl phenyl methyl propional (Lilial) (42%) –hexyl cinnamal (42%) FM II oxidized limonene and oxidized linalool should be added to the test series 11% contained none of the listed fragrances but were labelled as containing 'parfum' or 'aroma' 1 Buckley BJD 2007

26. Exposure pattern to fragrance of the U.K. consumer was assessed (300 products) FM I: –geraniol 42% –eugenol 27% –hydroxycitronellal 17% –isoeugenol 9% –cinnamic alcohol 8% –amyl cinnamal 7% –cinnamal 6% –oak moss absolute 4% FM II: –geraniol 42% –hexyl cinnamal 42% –coumarin 30% –Lyral 29% –citral 25% –farnesol 8% 1 Buckley BJD 2007

27. Frequencies of reactions to 26 fragrances in a German multicentre study 1 tree moss 2.4% Lyral2.3 % oak moss 2.0 % hydroxycitronellal 1.3 % isoeugenol 1.1 % cinnamic aldehyde 1.0 % farnesol 0.9 % cinnamic alcohol 0.6 % citral 0.6 % citronellol 0.5 % geraniol 0.4 % eugenol 0.4 % coumarin 0.4 % lilial 0.3 % amyl-cinnamic alcohol 0.3 % benzyl cinnamate 0.3 % benzyl alcohol 0.3 % linalool 0.2 % methylheptin carbonate 0.2 % amyl-cinnamic aldehyde 0.1 % hexyl-cinnamic aldehyde 0.1 % limonene 0.1 % benzyl salicylate 0.1 % gamma-methylionon 0.1 % benzyl benzoate 0.0 % anisyl alcohol 0.0 % 1 Schnuch et al. Contact Dermatitis 57: 1-10 2007

28. Fragrance allergy – oxidation forms sensitizers R-Limonene is a common fragrance terpene, widely used in cosmetics and in domestic and industrial products R-limonene is a very weak sensitizer but forms allergenic oxidation products upon contact with air (= prehapten). Oxidized (ox.) limonene is a frequent cause of contact allergy in clinical testing 1 Oxygen-centred radicals are important in hapten- protein complex formation of hydroperoxides 1 Christensen et al. Contact Dermatitis 59: 344-52 (2008)

29. Fragrance allergy – ROAT Whenever a certain cosmetic product is suspected of causing an allergic reaction a repeated open application test (ROAT) should be performed (twice daily for 7 days) A ROAT is used to mimic the normal exposure Fragrance ingredients are special as significant amounts of allergen may evaporate from the skin -> Lyral: the ROAT threshold in dose per area per application is lower than the patch test threshold 1 1 Fischer et al. Br J Dermatol 2009

30. Safety assessment FRAGRANCE ALLERGY Epidemiology Study Elicitation Threshold Oxidation Research Quantitative Risk Assessment Supplying Patch Test Materials Other Clinical Diagnostic Tests Other Other Clinical Diagnostic Tests

31. An international study assessing the prevalence of contact allergy to fragrances in the general population: The European Dermato- epidemiology Network (EDEN) Fragrance Study in collaboration with ESCD supported by RIFM

34. Study has two major parts: –Assess the prevalence & severity of contact dermatitis, avoidance of products in the general population through a standardized interview. –Assess the rate of sensitization to fragrances through standard patch test procedures Patch testing (47 substances) –2 Fragrance mixes + 14 materials that comprise it –True Test panel 1, 2, 3 (29 substances) Geographical Regions –Italy, Sweden, The Netherlands, Germany (East & West), Portugal STUDY PROTOCOL

36. 36 Pilot Study (n= 589): Fragrance substances ++ / ++++ Ir? - Fragrance Mix I6712347 Fragrance Mix II 5211355 Fragrance Mix I 5502352 with sorbitan sesq. Single fragrances* 14199135041 * positive reactions in 0.65 %

37. 37 Fragrance sensitisation, avoidance and skin problems ++ / ++++Ir?- Total sensitisation rate131598319 Avoidance of products N (%) Yes11 (84.6) 6 (40.0) 3 (33.3) 4 (50.0) 125 (39.2) No2 (15.4) 9 (60.0) 6 (66.7) 4 (50.0) 194 (60.8) Skin rash (1 yr. Prevalence) Yes 5826129 No 8772190

38. Patch test reactions and fragrance allergy 589 subjects 28 (4.7%) (17 f; 11 m) 18 (3.0%) no clear reaction: 7 f; 3 m no avoidance: 3 f; 3 m no skin rash: 3 f; 3 m 12 (2.0%) 7 (1.2%)

39. Fragrance allergy – a challenge to interpretation Most people in modern society are exposed daily to fragrance ingredients from one or more sources. Fragrance ingredients are also one of the most frequent causes of contact allergic reactions. Patch testing is mandatory – however do we patch test the right substances?

40. The most frequent sensitizers in cosmetics Fragrances Methylchlorisothiazolinone and Methylisothiazolinone Methlydibromoglutaronitrile Formaldehyde Formaldehyde donors Quartenium 15 Imidazolidinylurea Diazolidinylurea Bronopol DMDM hydantoin Parabens Iodopropynyl butlycarbamate Formaldehyde resin P-Phenylendiamine and related hair dyes Cocamidopropyl betain UV filters Lanolin and derivates Glycerylthioglycolate Propylene glycol Antioxidants

41. Fragrance allergy – Clinical consequencies The diagnosis is made by patch testing with a mixture of fragrance ingredients: FM I and FM II With investigator-loaded Finn Chamber the major problem is false-positive reactions Weak positive reactions to FM should be evaluated carefully for clinical relevance. Whenever there is a positive reaction to FM, additional patch testing with the ingredients is recommended. In addition, testing of 'whole' products and ROAT can be used to confirm the relevance of reactions in cases of doubt.

42. Thank you

44. No. 2, PI: After shave ++ Neg.IR?++++++ 37

45. No. 8: 1% Shampoo (soap effect) IR Neg.IR?++++++ 10