Annex D Background information on sensitization tests for skin sensitization
Sensitization in man occurs after single or multiple epicutaneous exposures and is initiated and elicited by components of the immune system. First, the hapten (chemical) must penetrate the skin. It then reacts with skin proteins to form immunogenic complexes. Langerhans cells at the epidermal/dermal border present the antigen to specific lymphocytes which are then activated to initiate the immune responses. A small percentage of these lymphocytes are long-lived memory cells, and these serve as the primary activators during the challenge phase. Thus, subsequent re-exposures can result in adverse reactions that are mediated by lymphokines released by the activated lymphocytes and other inflammatory cells that are attracted to the area of the lesion.
In 1895, Jadassohn employed the patch test to disclose contact allergy to mercury in a clinical patient. This innovative approach provided the scientific basis for subsequent tests aimed at diagnosis and prediction of contact allergy in man and animals. The development of prospective/predictive tests for evaluating the sensitizing potential of chemicals followed the pioneering work of Landsteiner and Chase[11], who firmly substantiated the use of the guinea pig for studying skin sensitization.
In 1969, Magnusson and Kligman[12] explored many of the variables of guinea pig testing and presented a procedure, the GPMT, based on intradermal injections (with and without Freund's complete adjuvant), followed by topical application of the test material to the same area. The original procedure requires pretreatment of the test site if the test material is a non-irritant. By definition, it reputedly detects weak skin sensitizers, because “weak” included a zero incidence of positive reactors. It is a sensitive test and has been extensively used. The use of Freund's complete adjuvant increases the sensitivity of the test method and it can in some cases overestimate the sensitizing potential of the compound in question.
In 1965, Buehler[7] advocated the use of the closed patch test to provide occlusion as a method to optimize exposure and to mimic the procedures used in humans [human repeat insult patch test (HRIPT)]. It was suggested that the occlusive patch procedure was sensitive and can accurately predict moderate to severe skin sensitizers, thus avoiding exposure of human subjects to the prospect of adverse reaction during HRIPTs. The data presented demonstrated the superiority of occlusion over intradermal injections and open-type topical application. Stimulation of the immune system by adjuvants was not used. This method is established as a technique that is sufficiently sensitive to detect most weak skin sensitizers and has been shown to be sufficiently flexible to be used in the risk assessment process. However, the closed-patch test (Buehler test) is less sensitive compared with the GPMT. See Reference [9].
These two tests, the closed-patch test in the United States and the GPMT in Europe, have been the most frequently used for safety assessment. The result from guinea pig sensitization assays depends on many animal-related and technical factors explaining the interlaboratory variation in test results, for example, animal strain, sex, age, ambient test conditions, test site on the animal, method of hair removal (clipping/shaving or chemical depilation), type of patch design, quantity of test material, quality of occlusion, exposure time and reading of the tissue response. Numerous other tests have been employed and investigated and all of these have their proponents. There are currently several procedures that have been recognized as acceptable for regulatory purposes, provided the procedure is properly documented and validated by the investigator. In all cases, the procedures should be performed in accordance with the original references. A list of other tests is given below.
Reference [8] gives an update on skin sensitization testing.
- a) Freund's complete adjuvant test.
- b) Split adjuvant test.
- c) Open epicutaneous test.
- d) Mauer optimization test.
- e) Footpad test in guinea pig.
- f) Cumulative contact enhancement test.
- g) Scratched skin (adjuvant and patch) test.
- h) Mouse ear swelling test.
The LLNA is now preferred to the GPMT and the Buehler occluded patch test for hazard identification of chemicals. LLNA has been accepted in 2010 by the OECD as a stand-alone alternative to the current guinea pig tests, and as an improvement for animal welfare. See Reference [33]. The LLNA has been validated for the determination of sensitizing activity of chemicals. See References [39] and [40].
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 sample as a measurement of sensitization. It does not include a challenge phase. The end point of interest is a stimulation index giving the ratio of thymidine incorporation in lymph nodes from dosed animals compared with the incorporation in lymph nodes from control animals. The test is positive when the stimulation index exceeds 3 (Is > 3). An intra- and inter-laboratory evaluation of the LLNA has demonstrated a reproducible dose-response relationship within and between laboratories. See References [16], [17], [21], [25], [26], [28] and [32]. However, difficulties in differentiating between irritating and allergenic substances with the LLNA have been reported. See References [18], [25] and [28]. Thus, the LLNA can give false-positive results with irritants and can overestimate the allergenicity of substances with both irritating and allergenic properties. See Reference [16]. However, the LLNA has advantages compared with the guinea pig assays because of shorter test duration, a more objective end point, less test substance 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. See References [23] and [24]. Whether they can be practically implemented in standard LLNA protocols for routine toxicology is not determined. On the other hand, the LLNA allows 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. See Reference [27]. 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.
The popliteal lymph node assay (PLNA) using subcutaneous administration in the footpad (see References [19], [22] and [31]) is an alternative to the lymph node assay. In the latter assay, in addition to direct measurement of lymph node activation, reporter antigens may be used for further clarification of the immunomodulation caused by the chemical under investigation. See Reference [15].
The risk assessment process should not rely on a single model or approach but should be thoughtfully conducted to provide maximum assurance of safety to the consumer. Generally, this entails both animal or non-animal methods and human experimental models. There should be flexibility in the choice of models and approaches, as long as the rationale is documented and/or validated.
Negative tests in guinea pigs, when they are properly conducted, can generally be definitive if the test concentration has a sufficient safety factor over use conditions. However, one should avoid classifying test materials solely on the basis of incidence and/or severity, without due consideration of eventual product usage.
The risk, i.e. incidence and severity, of the allergic reaction to the product is determined mainly with the following four factors: the sensitizing potency of the chemical allergen, its amount in the product, bioavailability and the exposure conditions. The relative sensitizing potencies of chemicals can be defined in terms of the minimum induction concentration required to induce a given level of sensitization: the lower this concentration the more potent the sensitizer. See References [6] and [30]. It was shown that the significant incidence of allergic contact dermatitis was found in users when the residue level of the allergen in the product exceeded its minimum induction concentration obtained by GPMT. See Reference [14].
On the other hand, predictive testing of mixtures and products is much less validated and can be performed following testing of product ingredients. Accordingly, test design and result interpretation are subject to uncertainty, but several examples document this possibility. In animal experiments with acetone extracts from a sweater that had caused contact dermatitis in humans, allergens (phosgene chlorophenylhydrazones) were demonstrated. See Reference [11]. In another case, where animal experiments with acetone/chloroform extracts from rubber boots had caused contact dermatitis in man, mercaptobenzothiazole and dibenzothiazyldisulfide were eventually found to be the causative allergens. See Reference [10]. The importance of using an appropriate organic solvent was clearly demonstrated. The extracts made with organic solvent induced skin sensitization in the guinea pigs, while the saline extracts failed to do so.
Reference [4] adopts the sample preparation procedure with organic solvent followed by evaporation of the solvent to obtain the residue, and the risk assessment procedure by comparing the percent residue yield from the material with the minimum percent dilution of the residue (mixture) that still induced skin sensitization in animals.
In vitro methods for skin sensitization testing are not yet available for routine use, but in view of new regulations in Europe which ban the use of animal tests for cosmetics, it seems likely that novel strategies will become available for the identification of skin sensitizers. See Annex C.