The Potential Use of Non-Invasive Methods in the Safety Assessment of Cosmetic Products

The Report and Recommendations of ECVAM Workshop 361,2

Reprinted with minor amendments from ATLA 27, 515-537.

Appendix 1

Examples of Experience Within the Industry

Some participants in the ECVAM/EEMCO workshop presented the protocols used within their own companies. Four examples have been summarised here, but it must be emphasised that they are only intended to serve as examples, and should not be seen as guidelines.

Example 1: Evaluation Programme of Skin Mildness and Skin Safety (Irritancy Testing)

Screening tests

The formulation is first discussed with the formulator, to provide guidance on the choice of surfactants and their combination.

During or at the end of the development stage, a quick screening for skin irritation is performed, to estimate overall skin mildness. Formulations which do not meet the predefined mildness criteria are rejected.

Several in vitro skin irritation predictive tests, based on protein denaturation by the least-mild surfactant systems, have been selected for their rapidity and the possibility of running them during formulation. The Zein test (1), the collagen swelling test (2), and the pH-rise test (3) are currently used.

When specific ingredients are suspected to interfere with the in vitro tests mentioned, other tests are performed, including an ex vivo assessment of the effects of the product on the stratum corneum by corneosurfametry (4) and shortterm patch tests (15-30 minute application) on human volunteers, with subclinical evaluation of stratum corneum alterations by squamometry (5, 6). Such methods have the advantage that the products are tested directly on their real target, the stratum corneum, and they can reveal large variations between products at various mildness levels. For this reason, small panels of human volunteers (n = 6 or 8) can be used.

It should be noted that the in vitro, ex vivo and short-term in vivo tests do not necessarily demonstrate the safety of the product, but only have predictive value.

Skin mildness tests

Here, the objective is to compare the mildness of the newly developed product with that of other products already present in the intended marketplace. Several types of tests can be used. Most tests exaggerate the exposure conditions, to amplify the reactions and permit a more accurate comparison between the test products.

The most commonly used tests are as follows.

  1. The Frosch-Kligman soap chamber test (7): solutions of the products are applied to the volunteers (n = 25-30) under occlusion for 24 hours, followed by another 6 hours during 4 consecutive days. Erythema and dryness are evaluated visually by a trained assessor.
  2. The modified soap chamber test (8): solutions of the products are applied to volunteers (n = 25-30) under occlusion, for two 24-hour periods. Visual assessment of erythema and dryness is often complemented by instrumental measurements of redness (3 hours after each of the applications), of TEWL (skin barrier damage, 3 hours after the first application), and of skin capacitance (skin surface hydration/dehydration, 3-5 days after the second application [9]).
  3. The exaggerated arm wash (10), flex wash (11) and hand wash (12) tests or hand soaking (13) tests: visual assessment of erythema and dryness is usually complemented by instrumental measurement. Capacitance and TEWL measurements are the most useful. Recently, squamometry was shown to have a great potential value in such tests (14, 15): the panel of 25-30 volunteers could be reduced to 10-15 subjects.

The choice of the test to be used depends on several factors, including:

  1. the number of products to be compared: for example, in a soap chamber test, up to eight products can be compared, while in an exaggerated wash/soak procedure, usually only two products are compared ("split arm" design, however, allows four products to be tested simultaneously);
  2. the intended use of the product (for example, a wash test is more appropriate for a soap bar, and a soaking test for a foam bath product);
  3. the potential effect of mechanical action on the interaction between the product and the skin surface (for example, in the presence of slight abrasives or encapsulated actives, a wash test is more appropriate than a patch test); and
  4. whether or not subjective, self-perceived information on the effect of the product on the skin is needed (for example, a patch test does not provide such information).

Other factors can also be considered, such as the cost of the studies and the equipment, the experience of the testing laboratory, and the expected scale of differences between the products.

Safety tests

Safety tests are different from mildness tests and aim at checking that the products do not irritate skin. Some products can be "non-mild", but can bring other benefits to the skin. Although claims of mildness properties are not made, they are still "non-irritant" products.

For this purpose, two types of skin irritation tests are most commonly used.

  1. Cumulative irritation tests: solutions of the products are applied to the backs of the volunteers (a minimum of 25 are used) under occlusion or semi-occlusion for several consecutive days (for example, the 21-day cumulative irritation test [16]). Reference products with wellknown skin tolerance in the marketplace are usually included. As the objective is to check the absence of obvious skin reactions, instrumental measurement is usually unnecessary.
  2. Home usage tests: volunteers apply the product at home under normal usage conditions for several consecutive weeks. Volunteers report on a regular basis to a competent assessor, who checks for the absence of unwanted effects. Before beginning such a test, a high level of confidence that the product is unlikely to cause irritation is required. The main advantage of such a procedure is the ability to check for the absence of subjective signs of skin irritation (for example, itching, stinging, burning), which are rarely detectable in patch tests. Large panels are recommended for home usage tests.

Example 1 summarises a full testing procedure for a product derived from another which was already known to have good skin compatibility. The information available on some components of the product or their interactions in mixtures is only rarely insufficient for tests on volunteers to be conducted, as described above. A more progressive testing procedure should then be used (17).

Example 2: Safety Testing (Irritancy) and Detection of Unexpected Events

Strategy

During new product development, all ingredients are scrutinized from a toxicological point of view. A critical review of the toxicological data provided by the suppliers of the ingredients is performed and other sources of information are checked (legal documents scientific literature, case reports, own information). If the data are of good quality and indicate an absence of toxic properties, then the formulation work can begin. The next step is to approve human volunteer testing of a number of newly suggested formulations. Tests are only considered for those formulations that merely cause reversible damage to the skin (for example, irritation). The type of testing is determined on a case-by-case basis. Slight changes in an existing formulation might not require any testing, whereas formulations with new ingredients are always tested in-house for irritancy. This confirms the safety of the product with respect to irritation (sensory irritation is not covered), and confirms that nothing unexpected has happened with the product due to interactions between its ingredients or during its manufacture.

Irritancy testing on human volunteers is initiated only when the ethical committee has given its approval, and only temporary irritation in some individuals can be expected. The new formulations are tested simultaneously with a suitable reference product, and with water as a control. The products of competitors are sometimes included. Comparative tests are considered to encourage the development toward milder products.

Procedure

The products are applied to the skin for 48 hours in large Finn chambers. Stay-on products are tested neat, and rinse-off products at a dilution of 10%. One hour after removal of the products, the areas are assessed visually for their degree of irritation, on a 4-grade scale. The following day, the test areas are also examined by using TEWL and LDF measurements. Usually, 12 volunteers are included, and up to 12 formulations are evaluated in the same study. The instrumental readings are made on one day. Statistically significant differences between formulations can usually be detected. Whether these differences are clinically relevant is another question, which needs careful consideration in the final safety assessment of the products. The visual grading of erythemacorrelates well with the LDF value. A fairly good correlation between TEWL and LDF can usually be found, although the instruments concerned measure completely different parameters. The absence of correlation provides important information on the mechanisms by which the products exert effects on the skin.

Advantages

Depending on the company involved, the instrumental assessment of the exposed skin areas can have major advantages, i.e.:

  1. two parameters relevant for irritation are measured, namely blood flow as a component of inflammation, and barrier damage as a direct effect on the stratum corneum or an indirect effect via inflammation;
  2. unexpected effects on the skin can be detected;
  3. the method is objective, it is sensitive when compared to visual assessment, the values are on a continous scale and not on a categorical scale, it is inexpensive (important for small companies and firms with products with low volumes), and it is ethical (since a limited number of volunteers need to be included in the testing procedure); and
  4. to date, after using this strategy, no product has been launched which has caused high complaint rates due to adverse skin reactions.

Example 3: Safety Assessment for Skin Care and Skin Cleansing Products

Product safety assurance is the result of a carefully planned and well executed safety assessment programme, a process by which the potential hazard of a product is determined. A product is considered to be safe, if no substantial hazard or significant risk of damage results under conditions of recommended use reasonable foreseeable misuse, or accidental exposure.

The safety assessment of an ingredient or product initially involves a review of any existing data in the scientific literature or in information provided by its manufacturer or supplier. In addition, any available in vitro data will be evaluated at this time. A suitable alternative to animal testing for ingredients and finished products is to use a battery of in vitro studies to show that they are likely to be non-irritating (18, 19). If these preliminary data indicate that the item is suitable for further testing, study requirements for human clinical research are determined. In these well-defined, carefully monitored clinical studies, specific information on how the ingredient or product affects the body is generated through a series of clinical tests.

A general procedure is outlined below, to illustrate how the clinical safety testing of skin care products (stay-on category) and skin cleansing products (rinse-off category) is performed on human volunteers.

Skirt care products

A distinction must be made as to whether Minor alterations were made to an existing formulation, or whether new ingredients have been introduced.

Mirror alterations
Examination of the formulation is followed by an in-use conditions test, in which the results are quantified by corneometry (skin hydration) and profilometry (skin roughness).Twenty-five volunteers are included in this test, and the product is applied twice daily on the volar forearm as the test site of choice. Bioengineering measurements are conducted after one week of application. The application of the product is continued for an additional week. In the case of significant erythema or alterations of the skin surface, the volunteer must immediately discontinue applying the product. New ingredients/new formulations
A single application is made in a closed 24-hour patch test on the backs of volunteers. A negative (very mild skin care formulation) and a positive control (0.1% solution of sodium dodecylsulphate) are included, to permit comparisons between tests. If the test sample shows negative results, a repeated closed patch test is performed for further characterization of skin compatibility, in which the product is applied to the backs of human volunteers on four consecutive days, for 21 hours per day. The test is visually scored by a trained investigator. If erythema or any skin reaction occurs, the re-application of the sample is stopped immediately. Bioengineering techniques, such as LDF imaging or calorimetry, are occasionally applied to quantify the erythema. The measurement of biophysical parameters provides highly objective data on blood flow or skin redness. However, to assess the occurrence of an erythema, these latter techniques are not necessarily of higher sensitivity than the subjective visual inspection.

Skin cleansing products

Minor alterations
After examination of the formula, an in-use conditions test, such as the forearm wash test, is performed. The forearm wash test is a method for estimating the relative irritation of personal cleansers. The protocol is based on consumer washing habits, and is more useful than many other methods for evaluating personal cleanser mildness. A standard forearm wash test involves 12-16 volunteers. Two skin sites are treated with cleansing formulae and one site is treated with water only. One skin site is an untreated control. Washing occurs twice a day for 6 days. Quantification of the results is performed by using corneometry, TEWL measurement, desquamation techniques (D-squame) and visual assessment.

New ingredients/new formulations
A single application, 24-hour closed patch test is conducted on the backs of the volunteers. If negative or low positive results are found, a repeated application, closed patch test is performed on four consecutive days with a daily 6-hour application. Visual scoring is used to detect erythema. TEWL measurements and LDF imaging are occasionally conducted, to measure the damage to the skin barrier and to quantify the erythema, respectively. The results are compared with those obtained with products already on the market.

These clinical tests focus on the possible primary or cumulative irritation potentials of an ingredient or a formulation. In parallel with these investigations, the sensitising potential of an ingredient or a product is evaluated in appropriate in vitro studies (20, 21) and in repeated insult patch tests performed on human volunteers. In addition, for face care cosmetics, testing for phototoxicity and photoallergy must be taken into consideration.

In summary, the main parameter in safety testing on humans is the visual inspection of the skin site involved. Bioengineering methods offer an opportunity for the quantification and documentation of skin reactions. To ensure reliable conclusions in the future, a regimen of an appropriate combination of visual assessment and instrumental investigation will probably be built up as the standard approach to safety testing on human volunteers.

Example 4: The Determination of the Horny Layer Reservoir of Human Skin as a Parameter for Percutaneous Absorption

Studies on percutaneous absorption are rarely undertaken to establish a potential hazard for the skin itself. However, the major safety concern is that substances, which under in-use conditions provoke few or no apparent cutaneous side-effects, might pose systemic risks.

There is no direct approach to measuring exposure to cosmetic ingredients in humans in vivo. Due to intrinsic slow penetration kinetics and the low penetration rates of such materials, serum levels can only be assessed in humans in vivo in exceptional cases, and then only with radiolabelled compounds. However, in almost every case, the exposure to the radiolabel would exceed the limits posed by ethical rules. The measurement of radioactivity in excrete can only be interpreted on the basis of the knowledge of the excretion kinetics after intraveneous injection of the radiolabelled substance, an approach which is unlikely to gain ethics committee approval. However, a sufficient number of medicines have been investigated in vivo in humans, and the results provided can serve as reference data.

A number of investigations have previously been performed on rats in vivo, and it has been found that rat skin is 2-fold to 5-fold more permeable than human skin. Furthermore, considerable experience has been gathered with human skin in vitro, and a satisfactory agreement with the in vivo situation has been established (22-24). However, it remains difficult to mimic in-use conditions in vitro for cosmetic preparations, and repeated applications are not possible with the currently available in vitro systems.

Therefore, a non-invasive in vivo method for the assessment of percutaneous absorption in humans would be welcome. Such a method would be considered suitable when a relationship between the horny layer reservoir (which builds up rapidly after topical application of any kind of material) and subsequent penetration rates could be assessed (25). In fact, such a direct relationship appears to have been well-established by the "stripping method" (26-28). There is a quantitative relationship between the material found in the horny layer of rats after 30 minutes (when the product is not yet in the dermis), as recovered by repetitive stripping with adhesive tape, and the total amount that passes into and through rat skin in vivo over 4 days. Furthermore, the same correlation has been shown in humans in vivo for some compounds concentrations and vehicles. In fact, the limitations of this technique can be examined by using reference substances, namely, a very hydrophylic and a very lipophylic compound. Thus, the respective results can be expressed in total penetrating quantities per exposure as µg/cm2 and mg/kg body weight and combined with data on the no adverse effect level (NOAEL), determined in toxicological studies. A safety factor for a cosmetic ingredient could be generated by using this information.

References

  1. Gotte, E. (1966). Synthetische tenside in medizinisch-kosmetischen badern. Aesthetische Medizin 10: 313-319.
  2. Blake-Haskins, J.C., Scala, D., Rhein, L.D. & Robbins, C.R. (1986). Predicting surfactant irritation from the swelling response of a collagen film. Journal of the Society of Cosmetic Chemists 37: 199-210.
  3. Tavss, E.A., Eigen, E. & Kligman, A.M. (1988). Anionic detergent-induced skin irritation and anionic detergent-induced pH rise of bovine serum albumin. Journal of the Society of Cosmetic Chemists 39: 267-272.
  4. Piérard, G.E., Goffin, V. & Piérard-Franchimont, C. (1994). Squamometry and corneosurfametry in rating interactions of cleansing products with stratum corneum. Journal of the Society of Cosmetic Chemists 45: 269-277.
  5. Piérard, G.E., Piérard-Franchimont, C., Saint-Léger, D. & Kligman, A.M. (1992). Squamometry: the assessment of xerosis by colonmetry of D-squame adhesive discs. Journal of the Society of Cosmetic Chemists 40: 29.
  6. Paye, M., Gomes, G., Zerweg, C., Piérard, G.E. & Grove, G.G. (1999). A hand immersion test in laboratory-controlled usage conditions: a need for sensitive and controlled assessment methods. Contact Dermatitis 40: 133-138.
  7. Frosch, P.J. & Kligman, A.M. (1979). The soap chamber test: a new method for assessing the irritancy of soaps. Journal of the American Academy of Dermatology 1: 35-41.
  8. Simion, F.A., Rhein, L.D., Grove, G.L., Wojtkowski, J.M., Cagan, R.H. & Scala, D.D. (1991). Sequential order of skin responses to surfactants during a soap chamber test. Contact Dermatitis 25: 242-249.
  9. Paye, M., Van de Gaer, D. & Morrison, B.M. Jr (1995). Corneometry measurements to evaluate skin dryness in the modified soap chamber test. Skin Research and Technology 1: 123-127.
  10. Sharko, P.T., Murahata, R.l., Leyden, J.J. & Grove, G.L. (1992). Arm wash evaluation with instrumental evaluation: a sensitive technique for differentiating the irritation potential of personal washing products. Journal of the Dermatological and Clinical Eualuation Society 43: 187-192.
  11. Strube, D.D., Koontz, S.W., Murahata, R.l. & Theiler R.F. (1989). The flex wash test: a method for evaluating the mildness of personal washing products. Journal of the Society of Cosmetic Chemists 40: 297-306.
  12. Charbonnier, V., Morrison, B.M., Paye, M. & Maibach, H.l. (1999). Open application assay in investigation of subclinical irritant dermatitis induced by sodium lauryl sulfate (SLS) in man: advantage of squamometry. Skin Research and Technology, in press.
  13. Clarys, P., Manou, I. & Barel, A.O. (1997). Influence of temperature on irritation in the hand/forearm immersion test. Contact Dermatitis 36: 240-245.
  14. Paye, M. & Cartiaux, Y. (1999). Squamometry: a tool to move from exaggerated to more and more realistic application conditions for comparing human skin compatibility of surfactant-based products. International Journal of Cosmetic Science, in press.
  15. Morrison, B.M., Paye, M., Charbonnier, V. & Maibach, H.I. (1998). Demonstrating invisible (subclinical) sodium lauryl sulfate irritation with squamometry. American Academy of Dermatology Meeting, Orlando, FL, USA, 27 February-4 March, PO98.
  16. Lanman, B.M., Elvers, E.B. & Howard, C.J. (1968). The role of human patch testing in a product development program. Joint Conference on Cosmetic Science. The Toilet Goods Association, pp. 21-23. CTFA: Washington, DC, USA: CTFA.
  17. Walker, A.P., Basketter, D.A., Baverel, M., Diembeck, W., Matthies, W., Mougin, D., Paye, M., Röthlisberger, R. & Dupuis, J. (1996). Test guidelines for assessment of skin compatibility of cosmetic finished products in man. Food and Chemical Toxicology 34: 651-660.
  18. Pape, W.J. & Hoppe U. (1990). Standardization of an in vitro red blood cell test for evaluating the acute cytotoxic potential of tensides. Arzneimittelforschung 40: 498-502.
  19. Pape, W.J. & Hoppe U. (1991). In vitro methods for the assessment of primary local effects of topically applied preparations. Skin Pharmacology 4: 205-212.
  20. Degwert, J., Steckel, F. & Hoppe U. (1995). Antigen presenting and primary in vitro sensitizing capacity of CD1a+ dendritic cells generated from human blood. Aduances in Experimental Medicine and Biology 378: 187-189.
  21. Reutter, K., Jager, D., Degwert, J. & Hoppe, U. (1997). In vitro model for contact sensitization. ll. Induction of IL-1 mRNA in human blood-derived dendritic cells by contact sensitizers. Toxicology in Vitro 11: 619-626.
  22. Pershing, L.K., Lambert, I., Shah, V.P. & Lam, S.Y. (1992). Variability and correlation of chromameter and tape stripping methods with the visual skin blanching assay in the quantitative assessment of topical 0.05% betamethasone dipropionate bioavailability in humans. International Journal of Pharmaceutics 86: 201-210.
  23. Pershing, L.K., Parry, G.E., Bunge, A., Krueger G.G. & Shah, V.P. (1993). Assessment of topical corticosteroid treatment in vivo and in vitro. In Topical Drug Bioauailability: Bioequioalence and Penetration (ed. V.P. Shah & H.l. Maibach), pp. 351-391. New York, USA: Plenum Press.
  24. Pershing, L.K., Lambert, L., Wright, E.D., Shah V.P. & Williams, R.L (1994). Topical 0.05% betamethasone dipropionate: pharmacokinetic and pharmacodynamic dose response studies in humans. Archives of Dermatology 130: 740-747.
  25. Rougier, A., Dupuis, D., Lotte, C. Roguet, R. & Schaefer, H. (1983). In vivo correlation between stratum corneum reservoir function and percutaneous absorption. Journal of Investigative Dermatology 81: 275-278.
  26. Rougier, A., Dupuis, D., Lotte, C. & Maibach, H.l. (1989). Stripping method for measuring percutaneous absorption in vivo. In Percutaneous Absorption: Mechanism, Methodology, Drug Delivery (ed. R.L. Bronaugh & H.l. Maibach), pp. 415 435: Basel, Switzerland: Marcel Dekker.
  27. Rougier, A. & Lotte, C. (1993). Predictive approaches. I. The stripping technique. In Topical Drug Biocuailability, Bioequioalence and Penetration (ed. V.P. Shaw & H.l. Maibach), pp. 163-182. New York, USA: Plenum Press.
  28. Coderch, L., Oliva, M., Pons, M., de la Maza, A. Manich, A.M. & Parra, J.L. (1996). Percutaneous penetration using the tape stripping technique. International Journal of Pharmaceutics 139: 197-203.

Return to Main Article