Allergic contact dermatitis is a prototype of delayed T-cell mediated hypersensitivity as classified by Gell and Coombs (1975). Development of allergic contact dermatitis follows exposure to the relevant contact allergen in sensitised individuals. Contact allergens (haptens) are usually low molecular weight compounds with a MW below 1000 daltons. During the sensitisation process the hapten binds to skin components and the hapten-protein conjugate functions as an antigen, which by the antigen presenting cells (Langerhans cells) is presented to and recognised by T-cells. After proliferation and dissemination of specific T-cells the sensitised individual is capable of developing an allergic contact dermatitis following renewed skin contact with the hapten. Contact allergy is thus a long lasting or permanent specific immunologic hypersensitivity.

In theory, allergenicity of a chemical can be studied in vivo in animal and in man and in vitro.

2. Animal assays

The sensitising potential of chemicals can be investigated by standardised animal test methods using guinea pigs or mice (EEC, 1992). The performance of the animal assays is well documented regarding testing of different classes of chemicals. These animal assays have also been valuable in the safety testing of cosmetic ingredients. The efficiency of the test methods concerning mixtures of ingredients and products is less well documented.

2.1. Guinea pig assays

have been reported by Magnusson and Kligman (1969), Buehler (1965), Maurer et al. (1975), Maurer (1983), Klecak et al. (1977), and Klecak (1991) among others. The guinea pig assays have been reviewed by Andersen and Maibach (1985). The long tradition for using the guinea pig as animal model has given extensive experience with this laboratory animal and the performance of the various test methods is documented.

The current guideline methods are the Guinea pig maximization test (GPMT) (Magnusson and Kligman, 1969) and the Buehler test ( Buehler, 1965).

The GPMT is a highly sensitive method using Freund´s complete adjuvant as an immune enhancer. It includes both intradermal and topical induction treatment and closed challenge.

The Buehler test uses repeated closed topical applications during induction and closed challenge.

The GPMT is regarded as a more sensitive assay that may also, for certain substances, overestimate the sensitisation hazard for the compound tested. On the other hand, a negative GPMT gives a large safety margin. The Buehler test is less sensitive and may underestimate the sensitisation potential of a compound.

There are a number of pitfalls in guinea pig allergy tests (Andersen and Maibach, 1985). The technology varies from laboratory to laboratory and the choice of test concentration, choice of vehicle, practice of reading, use of control animals, and interpretation of test results may vary widely and cause considerable interlaboratory variation, which may be reduced through increased experience and standardisation of the protocols (Andersen, Vølund and Frankild, 1995). These problems are not solved in the current guidelines (EEC, 1992; OECD #406, 1992).

2.2. Mouse assays

Recently, a mouse model, the Local lymph node assay (LLNA) has been accepted by the Interagency Co-ordinating Committee on the Validation of Alternative Methods (ICCVAM) in USA as a stand alone alternative to the current guinea pig tests, and as an improvement for animal welfare (ICCVAM, 1999). ECVAM is currently examining the acceptability of the LLNA as a validated alternative test in the European Union.

The scientific basis for the test is measurement of the incorporation of 3H-methyl thymidine into lymphocytes in draining lymph nodes of mice topically exposed to the test article as a measurement of sensitisation. It does not include a challenge phase. The endpoint of interest is a stimulation index giving the ratio of thymidine incorporation in lymph nodes from dosed animals compared to the incorporation in lymph nodes from control animals. The test is positive when the stimulation index exceeds 3 (SI >ce required, and it omits the Freund´s complete adjuvant injections. Improvements of the test procedure by use of analysis of cell activation markers and flow cytometry are possible (Gerberick et al, 1999a and 1999b). If they practically can be implemented in standard LLNA protocols for routine toxicology is not determined. On the other hand, the LLNA allows for a more limited choice of test vehicles, most studies have used a mixture of acetone and olive oil. A recent study shows the variability of the results using different vehicles (Lea et al., 1999). Further, it is not possible with the LLNA to study the challenge phase, or cross reactivity patterns because the animals are sacrificed after induction treatment before the lymph nodes are harvested.

3. Human assays

Sensitisation potential has also been investigated using human volunteers, and the development of animal sensitisation tests has been partly based on comparison to human tests performed with the same chemicals. Further, human testing has the advantage that extrapolation of the test results from one species to another is avoided.

Human predictive skin sensitisation tests have been in use the last 50 years. They have been used more widely in the United States than in Europe. Contract laboratories have performed the vast majority of human sensitisation tests and the scientific literature contains a limited number of publications giving results from tests with cosmetic ingredients as preservatives and fragrance chemicals (Marzulli and Maibach, 1973 and 1980).

There are a number of different human sensitisation tests available. They vary with regard to the number of induction patch tests, the placing of the patches and the use of a maximisation step. However, it is not entirely clear how useful these variations are, because validation of the tests has not kept pace with development of new tests. The human sensitisation tests require great experience in design and execution of the test and a number of artefacts are possible.

Three different approaches for predictive testing in man has been in use:

1) A single induction/single challenge patch test (Schwartz-Peck Test),

2) Human Repeated Insult Patch Tests (HRIPT)

a) with interval exposure (Draize test, Shelanski-Shelanski test and Voss-Griffith test).

b) with continuous exposure (Modified Draize test).

3) Human Maximisation Test.

The performance of the different test methods depends on a number of factors including type of test substance (ingredient or finished product), chemistry and animal toxicological data available, and intended use of the product.

3.1. Schwartz-Peck Test

Schwartz and Peck (1949) and Schwartz (1951 and 1969) described the Complete and Incomplete Schwartz-Peck tests (table 1). They are based on one single induction patch of varying duration and dose followed by a single challenge patch test depending on the type of product tested. The Complete Schwartz-Peck test, further comprises a 4 week use test with the product after the challenge patch. The Schwartz-Peck test only detects potent sensitisers and is considered obsolete in comparison with other predictive human sensitisation assays.

3.2. Human Repeated Insult Patch Test (HRIPTs)

The four most used HRIPTs include: the Draize human sensitisation test (Draize et al., 1944, Draize, 1959), the Shelanski-Shelanski test (Shelanski and Shelanski, 1951; Shelanski, 1953), the Voss-Griffith test (Voss, 1958, Griffith and Buehler, 1976) and the Modified Draize human sensitisation test (Marzulli and Maibach, 1973 and 1974). These assays are very similar as shown in table 1, however, there are some important differences.

- In the original Draize test ten consecutive induction patches are applied to new skin sites on the arms or back for 24 h every other day 3 times a week. Each induction site is evaluated for erythema and edema after removal of the patch. Two weeks after the last induction, a challenge patch is applied for 24 h and subsequently read. The response after challenge is compared to the responses reported after the early induction patches.

- The Shelanski-Shelanski test is comparable to the original Draize HRIPT but employs 15 consecutive induction patches to the same site and if erythema and/or edema develops during induction the following patch should be moved to an adjacent untreated area. 2-3 weeks after the last induction a challenge patch is applied for 48 h and scored. The induction patch responses are also noted and interpreted as evidence of cumulative irritation.

- The Voss-Griffith test is also like the original Draize HRIPT with nine 24 h patch tests conducted over a 3 weeks period and challenge is performed 2 weeks later with duplicate patches applied to the induction skin site and to the opposite arm. This assay allowed testing of four materials simultaneously. Repeated challenge is recommended in case of dubious reactions.

- The Modified Draize test differs from the original Draize test by subjecting the volunteers to a continuous induction period with patch exchange 3 times a week until a total of 10 patches have been applied. The patches are reapplied to the same site, and only if moderate inflammation has developed, the next patch is moved to an adjacent skin site. Challenge is performed on naive skin two weeks later with a 72 h patch test with a non-irritating concentration of the compound.

The concentration of material chosen for induction and challenge in the HRIPT is determined by considering the following factors: previous human experience, previous sensitisation tests in guinea pigs and irritation studies in humans. It is common practice to test multiple compounds simultaneously, because it saves time and cost, but the scientific basis for multiple simultaneous inductions is not substantiated.

3.3. Human Maximisation Test

This assay was designed in 1966 as a sensitive assay for the identification of the sensitisation potential of chemicals (Kligman, 1966). The "maximisation step"nd on a slightly irritated skin site. The challenge site is scored after 72 and 96 h and the sensitisation index is noted. The maximisation test has shown to be a very sensitive test from studying more than 90 different compounds with widely varying sensitisation potential (Kligman, 1966). However, a comparison of 21 different fragrance ingredients tested with the Draize test and the maximisation test showed that 8 of the compounds tested positive in the Draize test were not detected by the maximisation test (Marzulli and Maibach, 1980). Because the human maximisation test may produce a rather dramatic effect on the skin it may be considered unacceptable today.

3.4 Concerns regarding the use of human volunteers for predictive allergenicity tests

statistical reasons up to 36 of 1000 consumers may react. The argument for reducing the number of volunteers in the human maximisation test is the amplifying step introduced by treatment with an irritant test product or sodium lauryl sulphate.

In any case, it is scientifically inadequate and unethical to perform predictive tests with a number of subjects insufficient to produce valid data.

3.5. Ethical considerations

The basic principles for testing in humans are provided by the following documents:

- World Medical Association Declaration of Helsinki in its current revisions (1964-1975-1983-1989-1996)

- Recommendations N° R(90)3, of the Committee of Ministeries/Council of Europe, adopted on 4 ^th February 1990

- Draft Directive on Good Clinical Practices for Trials on Medicinal Products in the European Community

- National regulations regarding human studies.

The performance of human sensitisation tests raises ethical considerations, in particular concerning the risk for the volunteers, especially the risk that a patch test sensitisation elicits a clinical disease in the subject.

In the literature there is no answer on the consequences of such testing on human volunteers. A request of information about the risk involved was sent to COLIPA in December 1998. The answer dated 22 March 1999 gave the following information provided by member companies:

- Dermatological testing to confirm skin compatibility is common practice; data on 470000 human volunteers covering 2000 products did not reveal any positive results identified as due to sensitisation;

- Reported data covering HRIPT tests carried out during the last 10 years and related to 2044 different products tested on a total of 136765 persons showed 123 cases of probable/confirmed sensitisations.

In conclusion a risk for human volunteers cannot be excluded. There is still a lack of information on the severity and frequency of adverse effects.

4. Alternative tests

There are currently no validated alternative predictive contact allergenicity tests available.

5. Opinion of the SCCNFP

The predictive animal assays described in this document used to evaluate the cutaneous sensitising potential of cosmetic ingredients and chemicals are reliable. The performance of the tests when testing mixtures of ingredients and products is less well documented.

questionable whether predictive testing on humans contributes to human safety in comparison with animal testing. At the present, no alternative method for predicting sensitisation has been validated.

6. References

Andersen KE, Maibach HI (1985) Guinea pig sensitisation assays. An overview. In: Andersen KE, Maibach HI., eds. Contact allergy. Predictive tests in guinea pigs. Basel: Karger, 263-90.

Andersen KE, Vølund Aa, Frankild S (1995) The guinea pig maximization test - with a multiple dose design. Acta Derm Venereol 75:463-469.

Basketter DA, Lea LJ, Cooper K, Stocks J, Dickens A, Pate I, Dearman RJ and Kimber I (1999) Threshold for classification as a skin sensitizer in the Local Lymph Node Assay: a statistical evaluation. Food Chem Toxicol 37:1167-74.

Basketter DA, Roberts DW, Cronin M, Scholes EW (1992) The value of the local lymph node assay in quantitative structure-activity investigations. Contact Dermatitis 27:137-42.

Basketter DA, Scholes EW, Kimber I. (1994) The performance of the local lymph node assay with chemicals identified as contact allergens in the human maximization test. Food Chem Toxicol 32:543-7.

Buehler EV (1965) Delayed contact hypersensitivity in the guinea pig. Arch Dermatol 91:171-7.

Dearman RJ, Basketter DA, Kimber I. (1999) Local lymph node assay: use in hazard and risk assessment. J Appl Toxicol 19: 299-306.

Draize JH (1959) Dermal toxicity. Appraisal of the safety of chemicals in foods, drugs and cosmetics. Association of food and drug officials of the United States, Texas State Department of Health. Texas: Austin.

Draize JH, Woodard G, Calvery HD (1944) Methods for the study of irritation and toxicology of substances applied topically to the skin and mucous membrane. J Pharmacol Exp Ther 83: 377-90.

European Economic Community (29 December 1992) Annex to Commission Directive 92/69/EEC of 31 July 1992 adapting to technical progress for the 17 ^th time Council Directive 67/548/EEC on the approximation of the laws, regulations and administrative provisions relating to the classification, packaging and labeling of dangerous substances. Off J Eur Commun L383 A:131

Gell PDH, Coombs RRA, Lachman R (1975) Clinical aspects of immunology, 3 ^rd. Edn. Blackwell, London.

Gerberick GF, Cruse LW and Ryan CA (1999a) Local Lymph Node Assay: differentiating allergic and irritant responses using flow cytometry Methods 19:48-55

Gerberick GF, Cruse LW, Miller CM and Ridder GM (1999 b ) Selective modulation of B-cell activation markers CD86 and I-Ak on murine draining lymph node cells following allergen or irritant treatment Toxic Appl Pharmcol 159:42-151.

ICCVAM (Interagency coordinating committee on the validation of alternative methods). Peer review panel evaluation of the local lymph node assay (LLNA). (1999) NIH Publication No. 99-4494.

Ikarashi Y, Tsukamoto Y, Tsuchiya T, Nakamura A (1993) Influence of irritants on lymph node cell proliferation and the detection of contact sensitivity to metal salts in the murine local lymph node assay. Contact Dermatitis 29:128-32.

Kimber I, Hilton J, Dearman RJ, Gerberick GF, Ryan CA, Basketter DA, Scholes EW, Ladics GS, Loveless SE, House RV et al. (1995) An international evaluation of the murine local lymph node assay and comparison of modified procedures. Toxicology 103:63-73.

Klecak G (1991) Identification of contact allergens: predictive tests in animals. In: Marzulli FN, Maibach HI., eds. Dermatotoxicology. New York: Hemisphere Publishing Corporation 363-413.

Klecak G, Geleick H, Frey JR (1977) Screening of fragrance materials for allergenicity in the guinea pig. I. Comparison of four testing methods. J Soc Cosmet Chem 28:28-53.

Kligman AM (1966) The identification of contact allergens by human assay. III. The Maximisation Test: a procedure for screening and rating contact sensitisers. J Invest Dermatol 47:393-409.

Kligman AM, Epstein W (1975) Updating the maximisation test for identifying contact allergens. Contact Dermatitis 1:231-9.

Lea LJ, Warbrick EV, Dearman RJ, Kimber I and Basketter DA (1999) The impact of vehicle on assessment of relative skin sensitization potency of 1,4-dihydroquinone in the Local Lymph Node Assay. Am. J. Cont. Derm. 10/4:213-218.

Loveless SE, Ladics GS, Gerberick GF, Ryan CA, Basketter DA, Scholes EW, House RV, Hilton J, Dearman RJ, Kimber I (1996) Further evaluation of the local lymph node assay in the final phase of an international collaborative trial. Toxicology 108:141-52.

Magnusson B, Kligman AM (1969) The identification of contact allergens by animal assay. The guinea pig maximization test. J Invest Dermatol 52:268-76.

Marzulli FN, Maibach HI (1973) Antimicrobials: Experimental contact sensitisation in man. J Soc Cosmet Chem 24:399-421.

Marzulli FN, Maibach HI (1980) Contact allergy: predictive testing of fragrance ingredients in humans by Draize and Maximisation methods. J Environ Pathol Toxicol 3:235-45.

Maurer T. Contact and photocontact allergens (1983) A manual of predictive test methods. New York: Marcel Dekker.

Maurer T, Thomann P, Weirich EG et al. (1975) The optimization test in the guinea-pig. A method for the predictive evaluation of the contact allergenicity of chemicals. Agents Actions 5:174-9.

Montelius J, Wahlkvist H, Boman A, Fernström P, Gråberg L, Wahlberg JE. (1994) Experience with the murine local lymph node assay: inability to discriminate between allergens and irritants. Acta Derm Venereol 74:22-7.

OECD Guideline 406 on Skin Sensitisation (Adopted on 17-07-1992)

Patrick E, Maibach HI (1995) Predictive assays : animal and man, and in vitro and in vivo. In Textbook of Contact Dermatitis, 2 ^nd edn., Rycroft RJG, Menne T, Frosch P (eds), Springer Verlag, Berlin, 705-747.

Schwartz L (1951) The skin testing of new chemicals. J Soc Cosmet Chem 2:321-4.

Schwartz L (1969) Twenty-two years' experience in the performance of 200,000 prophetic patch tests. South Med J 53:478-84.

Schwartz L, Peck SM (1944) The patch test in contact dermatitis. Public Health Rep 59: 546-57.

Shelanski HA (1951) Experience with and considerations of the human patch test method. J Soc Cosmet Chem 2:324-31.

Shelanski HA, Shelanski WV (1953) A new technique of human patch tests. Proc Sci Sekt Toilet Goods Assoc 19:46-9.

Warbrick EV, Dearman RJ, Lea LJ, Basketter DA and Kimber I. (1999) Local Lymph Node Assay responses to paraphenylenediamine: intra- and inter-laboratory evaluations J Appl Toxicol 19:255-260.

Table 1

Predictive human skin sensitisation assays (Patrick and Maibach, 1995).

Test	No.Subjects	Induction site	No. of exposures	Duration of exposure(h)	FrequencyOf exposure	Rest (days)	Challenge
Schwartz -Peck	200	Upper arm	1	24-72-96	-----	10 - 14	48 h patch test+4 weeks use test in the Complete Schwartz -Peck
Draize	200	Upper back or arm;	10	24	3 / week	10 - 14	24 h patch test
Shelanski-Shelanski	200	Upper arm	15	24	3 / week	14 - 21	48 h patch test
Voss-Griffith	200	Upper arm	9	24	3 / week	14	24 h patch test
Modified Draize	200	Lower or upper back	10	48	Continuous	14	72 h patch test
Human Maximization test	25	SLS pretreated forearm or calf	5	48	2-3 / week	14	2% SLS for 1/2 h followed by 48 h patch with test material