1 Background

1.1 Emphasis on consumer safety

According to the Council Directive "a cosmetic product put on the market within the Community must not cause damage to human health when applied under normal or reasonably foreseeable conditions of use" (76/768/EEC). In order to achieve this goal of product safety, toxicological data on cosmetic ingredients are needed as outlined in the SCCNFP Notes of Guidance for Testing of Cosmetic Ingredients for their Safety Evaluation 2nd Rev. (XXIV/1878/97). Among the data mentioned, also «human data» are cited. However, the document does not specify these in detail. Regarding skin irritation, the SCCNFP considers that at present human testing of cosmetic ingredients or mixtures of ingredients should not be preferred to animal testing.

1.2 Animal tests for assessment of safety to be replaced by alternative methods

In the past, most of the toxicological data mentioned above have been generated by testing on animals. However, according to Council Directive 76/768/EEC, the marketing of cosmetic products containing ingredients or combinations of ingredients tested on animals after 30 June 2000 in order to meet the requirements of this Directive shall be prohibited. The Commission's general policy regarding research on animals supports the development of alternative methods to reduce or replace animal testing when possible.

1.3 Testing of cosmetic ingredients in humans

In this context, the scientific and ethical considerations for testing cosmetic ingredients or mixtures of ingredients in human subjects need to be defined more clearly. The skin irritancy reaction in humans is not an absolute measure and must be related to appropriate controls defining the range of response.

The SCCNFP stresses three points:

1. Since tests in animals or validated alternative methods may be limited regarding their predictive value for exposure of a human population, confirmatory safety tests in humans may be necessary scientifically and ethically, provided that the toxicological profile of an ingredient or a mixture of ingredients based on animal or alternative methods is available and that a high degree of safety is to be expected.

2. Confirmatory tests of ingredients or mixtures of ingredients in humans must be limited to situations where no irreversible damaging effects are to be expected for the volunteers and where the study goal is reasonably achievable with a study population of limited size.

3. The recruitment of human volunteers should be in line with the "World Medical Doctors Association Declaration of Helsinki" and "the Good Clinical Practice for trials on Medicinal Products in the European Community."

2 Procedure of irritancy assessment

The following text outlines the steps of an assessment of the irritancy of an ingredient or mixture of ingredients. While this text focuses on irritancy, it is understood that other aspects of toxicity have to be considered in parallel before performing tests in humans.

2.1 Initial considerations

Available chemical and physico-chemical data and structure-activity relationships making use of computer programs and databases for the prediction of skin irritation potential should be used.

2.2 Evaluation of irritation

Ingredients or mixtures of ingredients should be tested on animals and humans only at non-corrosive concentrations. This decision may be based on pH and acid/alkaline reserve measurements and on in vitro tests for skin corrosivity. At the present, in vitro methods for the assessment of irritancy have not yet been validated.

2.3 Confirmation by human volunteer testing

On the basis of a low irritation potential as proven by animal or future validated in vitro methods, the skin tolerability of an ingredient or a mixture of ingredient can be confirmed by testing in human volunteers. A number of test protocols are available such as open and closed patch tests, single and repeated-exposure tests. They should be chosen on the basis of the relevant use pattern of the ingredient or mixture of ingredients (1).

- In the open test, the tested ingredient or mixture of ingredients is applied on the skin without occlusion for time periods between 15 min and 24 h.
- In closed patch tests, diluted or undiluted products are applied under occlusive chambers over 24 or 48 hours.
- Cumulative or repetitive closed patch tests involve applications on the same test site between 1 and 7 times per week over a period of 1 to 5 weeks. These repetitive tests allow the assessment of cumulative irritation that is missed by single application tests.
- Controlled use or repeated open application tests (ROAT) imply the repeated application of an ingredient or a mixture of ingredients closely modelled to the use-situation.

While these tests historically have been assessed by clinical methods, non-invasive bioengineering technology such as measurement of transepidermal water loss or of blood flow may provide higher sensitivity and objectivity of these tests and thereby reduce the exposure and risk to volunteers.

However, neither the above confirmatory tests nor the use of bioengineering methods have been validated according to modern scientific criteria. The SCCNFP recommends the Commission to support further research in this area.

2.4 Consumer market surveillance

The evaluation of irritation of an ingredient or mixture of ingredients is not finished with the introduction of respective cosmetic products on the market, but it should continue by making use of data generated by consumer market surveillance and other sources.

3 Ethical considerations

Confirmatory skin tolerance tests of cosmetic ingredients in humans are subject to ethical concerns. In order to take account of these concerns, to minimise the risk to volunteers and to safeguard their rights, test protocols should be submitted to an acknowledged ethical committee and be in compliance with the followings :

- World Medical Doctors Association Declaration of Helsinki in its current revision (2).
Human testing is to be conducted and monitored under the direction of relevantly trained personnel to ensure the health and well being of volunteer subjects involved in the testing. The health and welfare of the subject has first priority and must be highly protected. Importantly, the human testing that is conducted for chemicals and consumer products is associated with minimal risk as it is conducted:

i) to supplement non-clinical information,

ii) to confirm that exposure will not cause significant harm, and/or,

iii) in a controlled fashion that minimises subject risk (4).

- National regulations regarding human studies

- Good Clinical Practice for Trials on Medicinal Products in the European Community (3).

The investigator(s) in skin tolerability tests of cosmetic ingredients should fulfil the qualifications as mentioned in the CPMP Working Party on Efficacy of Medicinal Products Note for Guidance on Good Clinical Practice (3).

4 References

(1) Patrick E, Maibach HI: Predictive assays: Animal and man, and in vitro and in vivo. In: Rycroft R.J.G., Menne T., Frosch P.J., (1995) Textbook of Contact Dermatitis. Springer Heidelberg New York.

(2) World Medical Doctors Association Declaration of Helsinki, (1997) JAMA 227: 925-926

(3) CPMP Working Party on Efficacy of Medicinal Products Note for Guidance: Good Clinical Practice for Trials on Medicinal Products in the European Community (1990) CB-55-89-706-EN-C.

(4) Organisation for Economic Co-operation and Development: Development of OECD Test Guidelines for Use in Tests with Human Volunteers. 27th Joint Meeting of the Chemicals Group and Management Committee, 11th-13th February 1998 (ENV/MC/CHEM/RD (98))