Available online at www.sciencedirect.

com

Regulatory Toxicology and Pharmacology 52 (2008) 323 www.elsevier.com/locate/yrtph

Dermal sensitization quantitative risk assessment (QRA) for fragrance ingredients

d,2

Research Institute for Fragrance Materials, Inc., 50 Tice Boulevard, Woodcli Lake, NJ, USA b Unilever SEAC, Colworth House, Sharnbrook, Bedford MK44 1LQ, United Kingdom Firmenich SA, Corporate Product Safety & Regulatory Aairs, Case Postale 239, 1, Route des Jeunes/de la Jonction, Geneva 8 CH-1211, Switzerland d LVMH, Fragrance Safety and Regulatory Aairs, 185 Avenue de Verdun, Saint Jean de Braye Cedex F-45804, France e Givaudan Suisse SA, 5 chemin de la parfumerie, Vernier CH 1214, Switzerland f The Procter & Gamble Company, Miami Valley Laboratories, 11810 East Miami River Road, Cincinnati, OH 45252, USA g Clariant Produkte (Deutschland) GmbH, Corporate Product Safety, Am Unisys-Park 1, 65843 Sulzbach, Germany h The Procter & Gamble Technical Centres Ltd, Whitehall Lane, Egham Surrey TW20 9NW, United Kingdom Received 16 July 2007 Available online 24 October 2007

Abstract Based on chemical, cellular, and molecular understanding of dermal sensitization, an exposure-based quantitative risk assessment (QRA) can be conducted to determine safe use levels of fragrance ingredients in dierent consumer product types. The key steps are: (1) determination of benchmarks (no expected sensitization induction level (NESIL)); (2) application of sensitization assessment factors (SAF); and (3) consumer exposure (CEL) calculation through product use. Using these parameters, an acceptable exposure level (AEL) can be calculated and compared with the CEL. The ratio of AEL to CEL must be favorable to support safe use of the potential skin sensitizer. This ratio must be calculated for the fragrance ingredient in each product type. Based on the Research Institute for Fragrance Materials, Inc. (RIFM) Expert Panels recommendation, RIFM and the International Fragrance Association (IFRA) have adopted the dermal sensitization QRA approach described in this review for fragrance ingredients identied as potential dermal sensitizers. This now forms the fragrance industrys core strategy for primary prevention of dermal sensitization to these materials in consumer products. This methodology is used to determine global fragrance industry product management practices (IFRA Standards) for fragrance ingredients that are potential dermal sensitizers. This paper describes the principles of the recommended approach, provides detailed review of all the information used in the dermal sensitization QRA approach for fragrance ingredients and presents key conclusions for its use now and renement in the future. 2008 Published by Elsevier Inc.
Keywords: Quantitative risk assessment; Dermal sensitization; Fragrance ingredients; NESIL; SAF; AEL; CEL

1. Introduction
Corresponding author. Fax: +1 201 689 8090. E-mail address: [email protected] (A.M. Api). 1 Present address: DABMEB Consultancy Ltd, Two Normans Road, Sharnbrook, Bedfordshire MK44 1PR, United Kingdom. 2 Present address: Pierre-Fabre Dermo Cosmetique, Centre de Recherche et Development, 17 Allee Camille Soula, BP 74, Vigoulet Auzil 31320, France. 0273-2300/$ - see front matter 2008 Published by Elsevier Inc. doi:10.1016/j.yrtph.2007.10.008
*

Although some substances in common use today may have the potential to cause dermal sensitization, they can be formulated into consumer products at safe levels. This is also the case for fragrance ingredients. IFRA provides the fragrance industry with risk management strategies on the use of fragrance ingredients includ-

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

ing those ingredients identied as contact allergens. Historically they achieved this through the establishment of Standards based on no-eect concentrations and translated these as maximum limits that were applied equally to all types of skin contact products with dierent limits only for non-contact products. More recently, signicant developments have been incorporated in the way dermal sensitization risk assessments are conducted for fragrance ingredients (Gerberick et al., 2001). The general toxicological principles of quantitative risk assessment can be applied here, since it is known that the induction of dermal sensitization is also a threshold based phenomenon (Kimber et al., 1999; Robinson et al., 2000). With this and based on an understanding of the chemical, cellular, and molecular principles of dermal sensitization, it is possible to conduct an exposure-based quantitative risk assessment (QRA) to determine safe use levels of fragrance ingredients in a variety of consumer product types. This paper describes the principles of the approach for fragrance ingredients in consumer products and provides detailed review of all the areas and information used. There will be other publications that demonstrate the implementation by providing practical examples for individual fragrance ingredients. 1.1. Review of dermal sensitization risk assessment methodologies for recommendation of the QRA approach for fragrance ingredients The safety assessment of chemicals that possess the ability to cause sensitization by skin contact have traditionally been done using an ad hoc comparative risk assessment technique (Robinson et al., 1989). It is only recently that the principles of exposure-based risk assessment, as an extrapolation of quantitative risk assessment methods that are widely accepted in general toxicology, have also been applied to induction of skin sensitization. Several papers (Farage et al., 2003; Felter et al., 2002, 2003; Gerberick et al., 2001; Griem et al., 2003; Robinson et al., 2000) have been published supporting the use of alternative and potentially better quantitative risk assessment approaches. For the purpose of this review, two key methods were considered in detail (Gerberick et al., 2001; Griem et al., 2003) in the evaluation of a common approach to risk assessment for fragrance ingredients that are contact allergens. Both methods are based on the same fundamental principles and have signicant common elements that were used as a starting point to dene the rened risk assessment methodology for fragrance ingredients based on the induction of dermal sensitization. The key renements that have been introduced in this paper are the establishment of known benchmarks [weight of evidence no expected sensitization induction level (NESIL)] and the determination of uncertainty factors (sensitization assessment factors). As with any risk assess-

ment, exposure is an essential element of the risk assessment process. Elements addressed here are the appropriate dose metric and how to prioritize exposure data from dierent sources. All of these renements are described in detail in this review and clear guidance is provided on their use within this dermal sensitization risk assessment approach. 1.1.1. QRA methodology for fragrance ingredients It is implicit that the conduct of a dermal sensitization QRA is necessary only for those fragrance ingredients identied as dermal sensitizers. The skin sensitization QRA approach for fragrance ingredients follows the same four fundamental steps as identied for general toxicology risk assessment. These four steps are outlined below for dermal sensitization. Hazard identication. This involves the use of experimental data to determine the skin sensitization potential of the fragrance ingredient. Typically this would involve a murine Local Lymph Node Assay (LLNA), but may also involve the use of other assays such as the guinea pig maximization test or Buehler guinea pig test. Criteria that are used to dene a dermal sensitizer and a non-sensitizer have been published in ECETOC (2003). Doseresponse assessment or hazard quantication. The doseresponse for induction of skin sensitization is typically determined in the rst instance using animal assays such as the LLNA. Conrmatory human assays such as the Human Repeat Insult Patch Test (HRIPT) may also be subsequently conducted to provide substantiation of the NOEL. Relative skin permeability and integrity are also considered in this section. Exposure assessment. Exposure to the fragrance ingredient is determined using habits and practice data for consumer product use and human parameters data. Risk characterization. The data from the previous steps are used to determine an acceptable exposure level to a fragrance ingredient against which the real life consumer exposure to that fragrance ingredient in a specic product type can be compared. The acceptability or unacceptability of real life exposures can then be determined accordingly. In developing a quantitative risk assessment method for skin sensitization of fragrance ingredients, based on the above recommended approach, some new terms have been adopted and are presented below. The new terms are No Expected Sensitizing Induction Level (NESIL) and Sensitization Assessment Factors (SAFs) that replace no observed eect level (NOEL) and uncertainty factors, respectively. These terms have been adopted to take into account unique elements of quantitative risk assessment for skin sensitization. 1.2. Hazard identication 1.2.1. Animal data Historically, there are several animal models that have been used to determine the potential for a fragrance ingre-

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

dient to induce sensitization. Guinea pig tests (adjuvant and non-adjuvant) have been used for many years to assess the inherent contact sensitization potential of chemicals. These tests can assess potency to a certain extent or antigen cross-reactivity of structurally-related chemicals. More recently the murine local lymph node assay (LLNA) has been approved by the OECD and can be used both to determine the potential of a material to induce contact sensitization and to estimate the relevant sensitizing potency of contact allergens by using the EC3 value: the concentration required to induce a threshold positive response (Basketter et al., 1999). The EC3 value has recently been demonstrated to closely correlate with the NOEL from human sensitization tests designed to conrm lack of induction (Basketter et al., 2000, 1999; Gerberick et al., 2001,a, 2004; Griem et al., 2003; Schneider and Akkan, 2004). 1.3. Doseresponse or hazard quantication 1.3.1. No Expected Sensitizing Induction Level (NESIL) The NESIL is a benchmark that is derived from animal (see above) and human data (see below) through application of weight of evidence approach to all the relevant data. The NESIL is expressed as a dose per unit area (e.g., lg/ cm2) value. In contact allergy, there is now overwhelming empirical support for using quantity per unit area rather than other dose metrics such as concentration applied to the skin (Kligman, 1966; Magnusson and Kligman, 1970; Friedmann and Moss, 1985; White et al., 1986; Rees et al., 1990; Upadhye and Maibach, 1992). An in-depth review of the published studies including those mentioned above that support the use of dose per unit area in risk assessments for induction of dermal sensitization is provided in the publication by Kimber et al. (2008). 1.3.2. Human data A human sensitization test is used to conrm the lack of sensitization at an exposure level which was identied as a NOEL in an animal model or derived as a likely NOEL from quantitative structureactivity relationships. The test most typically conducted is the human repeat insult patch test (HRIPT) (McNamee et al., 2008). Dose for dose, this test exaggerates exposure from normal use of consumer products. Such tests must meet current ethical and methodological criteria. With implementation of the QRA approach, IFRA/ RIFM are recommending the use of the RIFM standard HRIPT protocol for generation of conrmatory human data for use in QRA. Details of this standard HRIPT protocol are described by Politano and Api (2008). Diagnostic patch test data from dermatology clinics are not used in the determination of the NESIL. This is because these data are a measure of elicitation of allergic contact dermatitis, not induction of dermal sensitization. To date there are insucient data to discern any quantitative relationship between induction and elicitation. Such

information is most useful in a risk assessment approach to help determine the need for additional data, for example to indicate where current exposures to fragrance ingredients may be a source of clinically relevant positive reactions. The absence of signicant clinically relevant positive reactions following testing in dermatology clinics, will provide additional data for use in the QRA approach and may provide support for current exposures to the fragrance ingredient. 1.3.3. Weight of evidence approach for determining the NESIL for fragrance ingredients Historical data that are used to determine the sensitization potential of a material may be of variable quality and robustness. To this end, weight of evidence (WoE) guidelines (see Fig. 1) have been developed. These guidelines have been developed specically for fragrance ingredients and are intended only to be applied to fragrance ingredients. These guidelines may also address some unusual situations for which discrepancies between data generated in non-adjuvant guinea pig tests, LLNA and human data (HRIPT), human maximization test (HMT) need to be resolved. In the previous risk assessment approach for dermal sensitization, the RIFM Expert Panel (REXPAN) has been the advisory body responsible for determination of noeect levels used to establish limits of use described in the IFRA Standards. REXPAN will continue to have this responsibility, but will determine the NESIL rather than the no-eect levels for a fragrance ingredient. They have adopted the guidelines outlined below for establishing WoE NESILs for fragrance ingredients. Scientic judgment will prevail when establishing WoE NESILs for fragrance ingredients. 1.3.3.1. WoE NESILs for selected fragrance ingredients identied as potential dermal sensitizers. Animal (guinea pig and mouse), human (maximization, RIPTs and others) and diagnostic patch test data for a group of 31 fragrance ingredients were reviewed in detail. This group of fragrance ingredients was chosen to include the 26 fragrance allergens that must now be labeled on cosmetic products in Europe in line with the 7th Amendment of the EU Cosmetics Directive and an additional 5 fragrance ingredients for which an IFRA Standard based on sensitization eects exists. The guidelines detailed above were applied to all the data and a WoE NESIL was identied. These NESILs are provided in Table 1. 1.3.4. Sensitization assessment factors for fragrance ingredients In general toxicology uncertainty factors are applied to extrapolate from experimental to real life exposure scenarios. These uncertainty factors are dened from inter-species variability (Travis and White, 1988; Chappell and Mordenti, 1991) and inter-individual variability (Renwick and Lazarus, 1998; Burin and Saunders, 1999; Aldridge et al.,

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

Fig. 1. Guidelines for applying weight of evidence (WoE) approach for use of induction sensitization data on fragrance ingredients for derivation of NESILs.

2003). In dermal sensitization risk assessments it is equally necessary to extrapolate from the experimental (dened and controlled exposure conditions) to real life consumer exposure (variable exposure controlled by the consumer). This is achieved by the application of a Sensitization Assessment Factor (SAF) which takes account of three parametersinter-individual variability (the same as in general toxicology), vehicle/product matrix eects, and use considerations (specic for dermal sensitization). The concept of and the parameters aecting the SAF for fragrance ingredients were originally proposed by Gerberick et al. (2001) and expanded by Felter et al. (2002). The SAFs recommended in this paper draw and build from the previous publications.

Key SAF areas to be addressed are given in the forthcoming sections. 1.3.4.1. Inter-individual variability. The SAF for inter-individual variability allows for possible variations in the sensitivity of individuals within the human population due to dierent parameters such as genetic eects, sensitive subpopulations, inherent barrier function, age, gender, and ethnicity. Genetic factors are not totally understood, but are clearly instrumental in determining individual susceptibility (Felter et al., 2002; Smith and Hotchkiss, 2001). There are several studies that address the importance of subpopulations, such as those with multiple allergies who may be more susceptible (Felter et al., 2002; Friedmann

Table 1 No expected sensitization induction level (NESIL) for fragrance ingredients derived by application of weight of evidence guidelines
Fragrance ingredient CAS No. LLNA weighted mean EC3 values (lg/cm2) [no. of studies] 2420 [4] >6250 [1]e 1475 [1]e >12,500 [1]e >12,500 [1]e 4600 [1]e 725 [1]e 2372 [6] 5250[1]e 262 [23] 1414 [11] 10,875 [1]e >6250 [1]e 2703 [6] 1200 [2] 3525 [5] 2372 [>5] 5612 [9] 4275 [1]e 498 [18] 10,075 [5] 12,650 [2] <125 [1]e <1250 Estimated 625 [1]e 5450 [1]e 962 [2] 970 [1]e 2163 [2] 1012 [2] >6250 [1]e >2500 [1]e Potency classication based on animal datab Human data NOEL HRIPT (induction) (lg/cm2) 23,622d 3543d NA 5906 59,050d 4720d 17,717d 4125 3000 591 1400 29,528d 3543 5906 2755 11,811 23,622d 5000 4000 250 10,000d 15,000d 118 24 70,866d 591 700 700 24 3898 581 NOEL HMT (induction) (lg/cm2) NA NA 3448d 6897 20,690d 5517d 20,690d NA 2759 NA NA 4138 5517 NA NA NA NA NA NA NA 5517d 13,793d NA NA NA NA NA NA NA NA NA LOELa (induction) (lg/cm2) NA NA NA 8858 NA NA NA 29,528 4724 775 3876 NA 8858 NA 6897h NA NA 5906 NA 775 NA NA 194 118 NA 1181 NA NA 236 7752 1250 WoE NESILc (lg/cm2)

a-Amylcinnamaldehyde a-Amylcinnamyl alcohol Anisyl alcohol Benzyl alcohol Benzyl benzoate Benzyl cinnamate Benzyl salicylate p-t-Butyl-a-methylhydro-cinnamic aldehyde (BMHCA) Cinnamyl alcohol Cinnamaldehyde Citral DL-Citronellol Coumarin Eugenol Farnesol Geraniol a-Hexyl-cinnamaldehyde Hydroxycitronellal 3 & 4-(4-Hydroxy-4-methylpentyl)-3-cyclohexene1-carboxaldehyde (HMPCC) Isoeugenol i D-Limonene Linalooli Methyl 2-octynoate (Methyl heptine carbonate) Methyl 2-nonynoate (Methyl octine carbonate) a-iso-Methylionone Phenylacetaldehyde Oakmoss Treemoss trans-2-Hexenal Isocyclogeraniol Cinnamyl nitrile

122-40-7 101-85-9 105-13-5 100-51-6 120-51-4 103-41-3 118-58-1 80-54-6 104-54-1 104-55-2 5392-40-5 106-22-9 91-64-5 97-53-0 4602-84-0 106-24-1 101-86-0 107-75-5 31906-04-4 97-54-1 5989-27-5 78-70-6 111-12-6 111-80-8 127-51-5 122-78-1 90028-68-5 90028-67-4 6728-26-3 68527-77-5 1885-38-7

Weak Weak Moderate Weak Weak Weak Moderate Weak Weak Moderate Moderate Weak Weak Weak Moderate Weak Weak Weak Weak Moderate Weak Weak Strong Moderate Weak Moderate Moderate Moderate Moderate Weak Weak

23,600 3500 1500 5900 59,000 4700 17,700 4100f 3000 590 1400 29,500g 3500 5900 2700 11,800 23,600 5000 4000 250 10,000 15,000 120 24 71,000 590 700j 700k 24 3900 580l

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323 7

All data in this table are available from RIFM and are listed in the RIFM database. NOEL = No observed eect level; HRIPT = human repeat insult patch test; HMT = human maximization test; LOEL = lowest observed eect level; NA = not available. a Data derived from HRIPT or HMT. b Based on animal data using classication dened in ECETOC, Technical Report No. 87, 2003. c WoE NESIL limited to three signicant gures. d MT-NOEL = Maximum tested no eect level. No sensitization was observed in human predictive studies. Doses reported reect the highest concentration tested, not necessarily the highest achievable NOEL. e EC3 value from one LLNA, not the mean. f BMHCAHRIPT LOEL data suggest that the NOEL is likely to be in the region of 29,000 lg/cm2. On this basis, the IFRA Joint Advisory Group (JAG) was asked to supply any sensitization data on nal products containing BMHCA. g DL-CitronellolIFRA Joint Advisory Group was asked to supply any sensitization data on nal products containing DL-Citronellol. h LOEL from human maximization test, not a human repeated insult patch test. i D-Limonene and linalool are not contact allergens, but some hydroperoxides formed by autoxidation are known to be dermal sensitizers. In addition, D-limonene and linalool are known human irritants. The irritancy prole of D-limonene and linalool is being further investigated by RIFM. j OakmossPending LLNA and a conrmatory HRIPT on new qualities of oakmoss, which contain signicantly lower levels of atranol and choloratranol. All data presented are on qualities of oakmoss containing typical levels of atranol and chloroatranaol. k TreemossPending LLNA and a conrmatory HRIPT on new qualities of treemoss, which contain signicantly lower levels of atranol and choloratranol. All data presented are on qualities of treemoss containing typical levels of atranol and chloroatranaol. l RIFM sponsored HRIPT with 1000 lg/cm2 cinnamyl nitrile is in progress.

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

and Moss, 1985; Moss et al., 1985). Inherent barrier function for inter-individual susceptibility is an important to consider because its function can be compromised and could lead to greater susceptibility for induction of contact allergy. Age, gender, and ethnicity may have an eect on inherent barrier function in healthy skin. Skin barrier function is very similar from infancy to adulthood (Cunico et al., 1977; Cassimos et al., 1980; West et al., 1981; Holbrook, 1982; McCormack et al., 1982; Wester and Maibach, 1982; Fairley and Rasmussen, 1983; Harpin and Rutter, 1983). Decreases in the skin barrier function can occur at either end of the age spectrum pre-term infant (Kalia et al., 1998) and geriatric under certain conditions (Leveque et al., 1984; Ghadially et al., 1995). Pre-term infants were not included in this review since they would be under medical care. While there is some indication that females are the more reactive responder population (Jordan and King, 1977; Rees et al., 1989), the weight of evidence supports that females and males react similarly to contact allergens (Robinson, 1999; Felter et al., 2002). Weight of evidence indicates individuals of dierent ethnic origins are not substantially more susceptible to induction of contact allergy (Kligman, 1966; Weigand et al., 1974). Genetic eects, sensitive subpopulations, and inherent barrier function are known to be generally more inuential than age, gender, and ethnicity (Robinson, 1999; Felter et al., 2002). 1.3.4.2. Matrix eects. The consumer can be exposed to fragrance ingredients in many dierent product forms (e.g., cream, shower gel, eau de toilette). These product formulations are of varying complexity ranging from a simple ethanol matrix to multi-phase creams. In the experimental situation, exposure to the fragrance ingredient is typically in a simple vehicle. In addition, some of the consumer product formulations may contain ingredients that are irritants or penetration enhancers. A vehicle can be a single moiety (e.g., water), mixtures (acetone/ water, ethanol/water), or a complex product formulation presented in undiluted or diluted form. The eect of complex formulation/matrix, as a vehicle, on the physical chemical parameters and bioavailability of a test material may be substantially dierent from a simple vehicle. The same is true when extrapolating from the experimental situation in which a simple vehicle is used to the real life scenario where the fragrance ingredient is typically formulated into a more complex product matrix (Felter et al., 2002). In dermal sensitization risk assessment, consideration of matrix eects encompasses extrapolation from the matrix/ vehicle used to determine the EC3/NOEL in the experimental situation to the product formulation containing the fragrance ingredient to which the consumer is exposed in real life scenarios. The larger the dierence between the experimental situation and real life exposure scenario, the greater the SAF will be.

The two areas within vehicle/matrix eects that are noteworthy are irritants and penetration enhancers. Both have the ability to promote the skin penetration of the fragrance ingredient. Irritants. Dermal irritants are known to compromise the skin barrier (Robinson et al., 2000). They are also known to serve as a promoter of dermal sensitization possibly by inuencing the magnitude of response or by inuencing other steps in the induction of allergy (Smith et al., 2000). It is apparent that some degree of direct chemical inammation or other concurrent trauma enhances the keratinocyte activity, produced by the applied chemical itself, by some other component of the chemical delivery system, or by some form of physical insult. This may account for the noted enhancing eect of primary skin irritation on the sensitization response (Cumberbatch et al., 1993; Kligman, 1966). Penetration enhancers. Some chemicals are specically known to aect the penetration of other chemicals through the stratum corneum (Scheuplein and Ross, 1970; Schaefer and Redelmeier, 1996). As such it remains important to understand the experimental matrix/vehicle as to its eect on the penetration of the fragrance ingredient since it will aect the bioavailability of the material in the experimental situation.

1.3.4.3. Use considerations. Use considerations in the experimental situation are dened and controlled (e.g., site of contact, skin integrity, operator controlled, duration of exposure). On the other hand, use considerations in real life scenarios in almost all cases involve less exaggerated exposure, are more variable and are within consumers control. There are three key parameters for consideration when extrapolating from the controlled experimental situation to the real life scenario. They are site of contact, dermal integrity, and occlusion. The larger the dierence in skin site location, eect on barrier integrity, and occlusion, the greater the SAF. Regional dierences in dermal absorption can be substantial. Table 2 provides a comprehensive list of references that describe important considerations for application to dierent sites of contact. Variations in barrier integrity can be inuenced by consumer practices. Factors inuencing dermal integrity are known to have a signicant eect on dermal penetration. This might include, for example, the presence of diaper rash (Odio and Friedlander, 2000) in an infant, or dermatitis in an adult (Benfeldt et al., 1999). While less dramatic, shaving has also been shown to have an inuence (Edman, 1994). Occlusion of the skin increases the hydration of the stratum corneum, skin temperature, microbial count, pH, and dermal irritation (Zhai and Maibach, 2001) which can inuence dermal penetration. The human data used to dene NESILs are obtained under semi- or fully-occlusive experimental patch conditions. Under most circumstances

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323 Table 2 Derivation of SAFs for fragrance ingredients in dierent product types using RIFM data: rationale and the literature references Product type Interindividuala,b,c,d,e,f,
g,h,i, j,k,l,m,n,o,p,q,r,s,t, u,v,w,x,y,z,aa

Matrix SAF

Matrix SAF rationalef,z,ab,ac,ad,ae,af (experimental versus real life exposure) Matrix for the product not the same as the experimental conditions.

Use SAF

Use SAF rationale (experimental versus real life exposure)

SAF

SAF 3* 10 The area is the underarmag; the skin is easily irritatedah, highly follicularag and an area that is shaved ai. Type of occlusion is similar to that of the experimental test conditionsaj. The area is the underarmag; the skin is easily irritatedah, highly follicularag and an area that is shavedai. Type of occlusion is similar to that of the experimental test conditionsaj. The area is the underarmag; the skin is easily irritatedah, highly follicularag and an area that is shavedai. Type of occlusion is similar to that of the experimental test conditionsaj. The area is the underarmag; the skin is easily irritatedah, highly follicularag and an area that is shavedai. Type of occlusion is similar to that of the experimental test conditionsaj. The area is the underarmag; the skin is easily irritatedah, highly follicularag and an area that is shavedai. Type of occlusion is similar to that of the experimental test conditionsaj. The area is the underarmag; the skin is easily irritatedah, highly follicularag and an area that is shavedai. Type of occlusion is similar to that of the experimental test conditionsaj. The area is the underarmag; the skin is easily irritatedah, highly follicularag and an area that is shavedai. Type of occlusion is similar to that of the experimental test conditionsaj. The area is whole body including underarmag and mucous membranes ak ; the skin is easily irritatedah, highly follicularag and an area that is shavedai. Type of occlusion is similar to that of the experimental test conditionsaj. The area is the neck, wrists, antecubital fossa that may have increased permeabilityag The area is the face with increased permeabilityag, highly follicularag and possible abrasion from shavingai. The area is the face with increased permeabilityag, highly follicularag and possible abrasion from shavingai. (continued on next 300

Aerosol deodorant

10

Aerosol antiperspirant

10

3*

Matrix for the product not the same as the experimental conditions and may contain irritating active ingredients.

10

300

Stick deodorant/ antiperspirant

10

3*

Matrix for the product not the same as the experimental conditions and may contain irritating active ingredients.

10

300

Roll-on deodorant

10

3*

Matrix for the product not the same as the experimental conditions.

10

300

Roll-on antiperspirant

10

3*

Matrix for the product not the same as the experimental conditions and may contain irritating active ingredients.

10

300

Cream deodorant/ antiperspirant

10

3*

Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating active ingredients. Matrix for the product not the same as the experimental conditions and may contain irritating active ingredients.

10

300

Gel deodorant/ antiperspirant

10

3*

10

300

Deodorant cologne (body sprays)

10

3*

Matrix for the product not the same as the experimental conditions.

10

300

Hydroalcoholic products applied to unshaved skin Hydroalcoholic products applied to recently shaved skin Mens facial cream and balms

10

3*

Matrix for the product not the same as the experimental conditions. Matrix for the product not the same as the experimental conditions.

100

10

3*

10

300

10

3*

Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients.

10

300

page)

10 Table 2 (continued) Product type

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

Interindividuala,b,c,d,e,f,
g,h,i, j,k,l,m,n,o,p,q,r,s,t, u,v,w,x,y,z,aa

Matrix SAF

Matrix SAF rationalef,z,ab,ac,ad,ae,af (experimental versus real life exposure) Matrix for the product not the same as the experimental conditions, but not expected to be more irritating. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients. Matrix is very dierent from the experimental test conditions, however, it is not expected to be more irritating. Matrix for the product is not the same as the experimental conditions and may be designed to enhance penetration. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients. Matrix is very dierent from the experimental test conditions and contains highly irritating ingredients. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients. Matrix is very dierent from the experimental test conditions and may contain ingredients that are irritating. Matrix for the product not the same as the experimental conditions. Matrix for the product is very dierent from experimental conditions and may contain irritating ingredients. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients.

Use SAF

Use SAF rationale (experimental versus real life exposure)

SAF

SAF 3* 10 The area is the eye area with increased permeability and easily irritatedal. The area is the entire bodyag which may include, dry skinam, abraded skinai (e.g., underarms, legs) and semi- occlusion, due to clothing occurs. The area is mainly the hands, which may include dry skinam, there may be compromised skin due to dermatitisah, but occlusion does not occur. The area is the face with increased permeabilityag. 300

Eye products (includes: eye shadow, mascara, eyeliner, eye make-up) Body creams, lotions

10

10

3*

10

300

Hand cream

10

3*

100

Womens facial cream/ facial make-up

10

3*

100

Make-up remover

10

3*

The area is the face with increased permeabilityag.

100

Lip products

10

3*

10

Foot care products

10

3*

Shaving creams

10

3*

10

The site is highly vascular and there is exposure to mucous membranesawand possible exposure to dry or chapped lips. The area is the feet, which are less permeableag. Type of occlusion is similar to that of the experimental test conditionsaj. The area is the face with increased permeabilityag and highly follicularag and possible abrasion from shavingai.

300

30

300

Depilatory

10

10

The area is the underarm, upper part of leg and lower part of the legag. The area is the entire bodyag which may include, dry skinam, abraded skinai (e.g., underarms, legs) and possible exposure to mucous membranes ak,an, ao,ap,aq,ar,as,at,au,av . The area is the head which is highly follicularag and the scalp which is more permeableag,aw. The area is the head which is highly follicularag and the scalp which is more permeableag,aw. The area is the head which is highly follicularag and the scalp which is more permeableag,aw. The area is the head which is highly follicularag and the scalp which is more permeableag,aw. The area is mainly the hands, but may include the entire bodyag which may include, dry skinam, abraded skinai (e.g., underarms, legs), there may be compromised skin due to dermatitisah and possible exposure to mucous membranesak,an,ao,ap,aq,ar,as,at,au,av

300

Body wash/shower gels

10

3*

100

Hair styling aids (mousse, gels, leave in conditioners) Hair sprays

10

3*

100

10

3*

100

Shampoo

10

3*

100

Conditioner (rinse-o)

10

3*

100

Bar soap

10

3*

100

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323 Table 2 (continued) Product type Interindividuala,b,c,d,e,f,
g,h,i, j,k,l,m,n,o,p,q,r,s,t, u,v,w,x,y,z,aa

11

Matrix SAF

Matrix SAF rationalef,z,ab,ac,ad,ae,af (experimental versus real life exposure) Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients. Matrix for the product not the same as the experimental conditions and may be designed to enhance penetration. May contain irritating ingredients.

Use SAF

Use SAF rationale (experimental versus real life exposure)

SAF

SAF 3* 3 The area is mainly the hands, which may include dry skinam, there may be compromised skin due to dermatitisah. The area is the face with increased permeabilityag. 100

Liquid soap

10

Face washes, gels, scrubs

10

3*

100

Bath gels, foams, mousses

10

3*

Aerosol air fresheners

10

3*

Matrix for the product not the same as the experimental conditions. Matrix is dierent from the experimental test conditions and may contain irritating ingredients.

Toothpaste

10

3*

Mouthwash

10

3*

Matrix for the product not the same as the experimental conditions but, not expected to be more irritating than the experimental conditions.

Nail care

10

3*

Candle not in a jar

10

Closed air fresheners

10

Matrix for the product is not the same as the experimental conditions, is highly solvent based and expected to be more irritating than the experimental test conditions. Fragrance is not freely available for release from the matrix, unlike experimental conditions. Enclosed product; limited contact with fragrance.

The area is the entire bodyag which may include, dry skinam, abraded skinai (e.g., underarms, legs) and possible exposure to mucous membranes ak,an,ao,ap,aq,ar,as,at,au,av. Bathing involves a longer time of exposure to the product than showering. Conversely, product concentration is greater when showering than bathing. The area is the upper extremities and the face the latter of which has increased permeabilityag. The sites are the lips and mouth which are highly vascular (these areas are a mixture of keratinized and non-keratinized skin)ak,ax,ay,az,ba,bb,bc,bd,be. Data suggest the peri-oral skin (a site of concern) is highly permeablebf and is exposed to oral care products that may not be removed. For many products, especially for buccal cavity exposure, rapid dispersion, limited contact time and salivary dilution would indicate a lower SAF for use considerationsak. The sites are the lips and mouth which are highly vascular (these areas are a mixture of keratinized and non-keratinized skin)ak,ax,ay,az,ba,bb,bc,bd,be. Data suggest the peri-oral skin (a site of concern) is highly permeablebf and is exposed to oral care products that may not be removed. For many products, especially for buccal cavity exposure, rapid dispersion, limited contact time and salivary dilution would indicate a lower SAF for use considerationsak. The area is the nail, which is less permeablebg but there may be compromised skin due to dermatitisah. Brief contact with ngersbh.

100

100

100

100

100

10

Closed product, only rare accidental 10 contact may occur. (continued on next page)

12 Table 2 (continued) Product type

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

Interindividuala,b,c,d,e,f,
g,h,i, j,k,l,m,n,o,p,q,r,s,t, u,v,w,x,y,z,aa

Matrix SAF

Matrix SAF rationalef,z,ab,ac,ad,ae,af (experimental versus real life exposure) Matrix is dierent from the experimental test conditions, however, it is not expected to be more irritating and the matrix is an inert material and not skin impactful. Matrix is dierent from the experimental test conditions, however, it is not expected to be more irritating.

Use SAF

Use SAF rationale (experimental versus real life exposure)

SAF

SAF 1 10 The area is vulval mucous membraneak,an,ao,ap,aq,ar,as,at,au,av; Type of occlusion is similar to that of the experimental test conditionsaj. 100

Feminine hygiene conventional pads, liners, interlabial pads

10

Intimate wipes

10

3*

10

Tampons

10

Matrix is very dierent from the experimental test conditions, however, it is not expected to be more irritating and the matrix is an inert material and not skin impactful.

20ak

Baby diapers

10

Matrix is very dierent from the experimental test conditions, however, it is not expected to be more irritating and the matrix is an inert material and not skin impactful.

10

Baby wipes

10

3*

Matrix is dierent from the experimental test conditions, however, it is not expected to be more irritating.

10

Baby shampoo

10

3*

Matrix for the product is very dierent from experimental conditions and may contain irritating ingredients. Matrix for the product is very dierent from experimental conditions and may contain irritating ingredients. Matrix for the product is designed to enhance penetration.

Baby wash, bath

10

3*

Baby cream

10

3*

10

Baby oil

10

3*

Matrix for the product is designed to enhance penetration.

10

The area is vulval mucous membraneak,an,ao,ap,aq,ar,as,at,au,av and outer labia, which are highly follicularag. Type of occlusion, due to under clothing, is similar to that of the experimental test conditionsaj. The area is vaginal mucous membraneak,an,ao,ap,aq,ar,as,at,au,av includes non-keratinized mucous membrane-increased permeabilityaz,bf,bi. The nature of occlusion is dierent, but eect is expected to be similar to that of the experimental test conditionsaj. The area is the babys buttocks, groin, lower stomach and upper thighs and the skin integrity may be compromised (diaper rash)bj and involve mucous membrane exposureak,an,ao,ap,aq,ar,as,at,au,av. There is occlusion through diaper usebj. The area is primarily the babys buttocks, groin, lower stomach and upper thighs and the skin integrity may be compromised (diaper rash)bj and involve mucous membrane exposureak,an,ao,ap,aq,ar,as,at,au,av. There may be occlusion through diaper usebj. The area is the head (scalp more permeable)ag or possibly whole bodyag and mucous membrane exposure (body wash)ak,an,ao,ap,aq,ar,as,at,au,av. The area is possibly whole bodyag and the skin integrity may be compromised (diaper rash)bj and mucous membrane exposure (body wash)ak,an,ao,ap,aq,ar,as,at,au,av. The area is possibly whole bodyag or head (scalp more permeable)ag and the skin integrity may be compromised (diaper rash)bj and mucous membrane exposure (body wash)ak,an,ao,ap,aq,ar,as,at,au,av. There may be occlusion through diaper usebj The area is possibly whole bodyag or head (scalp more permeable)ag and the skin integrity may be compromised (diaper rash)bj and mucous membrane exposure (body wash)ak,an,ao,ap,aq,ar,as,at,au,av. There may be occlusion through diaper usebj.

300

200

100

300

100

100

300

300

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323 Table 2 (continued) Product type Interindividuala,b,c,d,e,f,
g,h,i, j,k,l,m,n,o,p,q,r,s,t, u,v,w,x,y,z,aa

13

Matrix SAF

Matrix SAF rationalef,z,ab,ac,ad,ae,af (experimental versus real life exposure) Matrix is dierent from the experimental test conditions, however, it is not expected to be more irritating and the matrix is an inert material and not skin impactful. Matrix for the product not the same as the experimental conditions.

Use SAF

Use SAF rationale (experimental versus real life exposure)

SAF

SAF 1 10 The area is possibly whole bodyag and the skin integrity may be compromised (diaper rash)bi and mucous membrane exposureak,an,ao,ap,aq,ar,as,at,au,av. There may be occlusion through diaper usebj. The area is the lower extremities which may include, dry skinam, abraded skinai (e.g., shaved legs) and semi- occlusion. The area is the exposed skin (25% of their average total body surface area bk ) which may include, hands, head, forearms, legs, dry skinam, abraded skinai (e.g., legs). Hands and lower armsag. May involve skin sites with dermatitisah. 100

Baby powder

10

Tights with moisturizers

10

3*

10

300

Insect Repellents (intended to be applied to the skin)

10

3*

Matrix for the product not the same as the experimental conditions. May contain irritating ingredients.

10

300

Handwash laundry detergent

10

3*

Laundry pre-treatment

10

3*

Hand dishwashing detergent

10

3*

Hard surface cleaner

10

3*

Matrix for the product is very dierent from experimental conditions and may contain irritating ingredients. Matrix for the product is very dierent from experimental conditions and may contain irritating ingredients. Matrix for the product is very dierent from experimental conditions and may contain irritating ingredients. Matrix for the product is dierent from experimental conditions and may contain solvents and other irritating ingredients.

100

Hands and lower armsag. May involve skin sites with dermatitisah.

100

Hands and lower armsag. May involve skin sites with dermatitisah.

100

Hands and lower armsag. May involve skin sites with dermatitisah.

100

Note: Products that contain sunscreens are not addressed separately but are included in the major product types (e.g., lip creams with sunscreen are included in lip product category). a Travis and White (1988). aa Weigand et al. (1974). ab Robinson et al. (2000). ac Smith et al. (2000). ad Cumberbatch et al. (1993). ae Scheuplein and Ross (1970). af Schaefer and Redelmeier (1996). ag Feldmann and Maibach (1967). ah Benfeldt et al. (1999). ai Edman (1994). aj Bucks et al. (1989). ak Farage et al. (2003). al Nuutinen et al. (2003). am Matts and Rawlings (2005). an Britz and Maibach (1979). ao Britz and Maibach (1979a). ap Elsner and Maibach (1990). aq Elsner et al. (1990). ar Elsner et al. (1990a). as Elsner et al. (1990b). at Elsner et al. (1990c). au Elsner et al. (1991). av Farage and Maibach (2004). aw Zhai et al. (2004). ax Kobayashi and Tagami (2004). ay de Vries et al. (1991). az Harris and Robinson (1992). (continued on next page)

14

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323 Table 2 (continued)

Chappell and Mordenti (1991). Lesch et al. (1989). bb Squier and Hall (1985). bc Squier (1986). bd Squier (1991). be Sayani and Chien (1996). bf Kobayashi and Tagami (2004a). bg American Beauty Association (2002). bh Selim (2005). bi Thompson et al. (2001). bj Odio and Friedlander (2000). bk EPA (2001b). c Renwick and Lazarus (1998). d Burin and Saunders (1999). e Aldridge et al. (2003). f Felter et al. (2002). g Robinson (1999). h Smith and Hotchkiss (2001). i Dupuis (1979). j Friedmann and Moss (1985). k Moss et al. (1985). l Cassimos et al. (1980). m Cunico et al. (1977). n West et al. (1981). o Holbrook (1982). p McCormack et al. (1982). q Wester and Maibach (1982). r Fairley and Rasmussen (1983). s Harpin and Rutter (1983). t Kalia et al. (1998). u Leveque et al. (1984). v Ghadially et al. (1995). w Jordan and King (1977). x Rees et al. (1989). y Young et al. (1988). z Kligman (1966). * For practical purposes the number 3 is the practical representation of 3.16 (half log of 10)].
ba

consumers are exposed to products under less than full occlusive conditions (examples of exceptions are diapers and axillary products). For those products where occlusion in the consumer exposure scenario is greater than that of the experimental situation, the SAF is increased. For example if the NESIL is derived from patch test data generated on the arm or back and the product is meant to be used in the axillae where the skin is easily irritated, highly follicular, occluded and may be abraded by shaving, this would increase the SAF to reect the large differences between the experimental situation and real life scenarios here. 1.3.4.4. Dening SAF numbers. The question that is probably most apparent at this point is which number to assign each component of the SAF. For inter-individual variability, a value of 10 is assigned. This is based on well established principles of general toxicology and is meant to reect not only the average consumer but also more susceptible sub-populations. For matrix eects and use considerations the number that is assigned to each area is dependent upon how dierent the experimental situation is versus the real life sce-

nario. For example, with vehicle eects if the vehicle in which the experimental data (used to dene the WoE NESIL) is generated is the same as that to which the consumer is exposed in the nished product then a SAF of 1 would be assigned. In general, the more impactful the difference between the experimental vehicle and the consumer product, the bigger the SAF up to a maximum of 10. It is also important to take into account the eect of the product matrix on the skin since a product matrix can be radically dierent in chemical composition from the experimental vehicle but be expected to have no eect on the skin, e.g., talcum powder versus an alcohol-based experimental vehicle. Although any value between 1 and 10 may be assigned for the SAFs relating to matrix/product eects and use considerations, it is considered pragmatic to limit the values used to 1, 3.16 (half log of 10), and 10. (For purposes practical in this paper the value of 3.16 is represented simply as the number 3.) A value of 1 denes an experimental condition that is identical or essentially identical to the real life scenario. A value of 10 denes an experimental condition that is unrelated or nearly unrelated to the real life scenario. A value of 3 is used to dene dierences between the

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

15

experimental conditions and the real life scenarios that are greater than 1 (none or minimal dierences), but less than 10 (maximal dierences). These values chosen are consistent with the approach used by EPA for general risk assessment (Dourson et al., 1996). This lends appropriate conservatism and simplicity to the approach. The overall SAF is a combination of the three key parameters dened above and is calculated by multiplying the inter-individual variability by vehicle/matrix eects and by use considerations. In theory, SAFs could range from 10 (inter-individual = 10, vehicle/matrix = 1, use considerations = 1) to 1000 (inter-individual = 10, vehicle/ matrix = 10, use considerations = 10). In reality, for fragrance ingredients it is unlikely that the SAF would exceed 300. However, exceptions could include where there is mucosal contact where higher SAFs for use considerations are assigned (Farage et al., 2003). The SAFs for dermal sensitization risk assessments for fragrance ingredients are specic for this toxicity endpoint and cannot be compared to the values dened for uncertainty factors in general toxicology. Fig. 2 illustrates the approach to assign SAFs. 1.3.4.5. Rationale for fragrance ingredients SAFs in dierent product types based on RIFM data. When considering the SAFs for fragrance ingredients, the SAF of inter-individual variability was given a value of 10. Since the parameters used to determine inter-individual variability in general toxicology are equally applicable to the identication to SAFs for the induction of skin sensitization, there is no scientic basis to change from the value of 10 used in general toxicology. For vehicle/matrix eects based on RIFM data, the SAFs for fragrance ingredients are based on the use of a vehicle containing ethanol. Key factors in determining this SAF are: an evaluation of the skin eects of ethanol (drying and barrier function decrease) in the experimental situation versus the consumer product matrix. the presence and level of formulation ingredients that are known to be irritants in the consumer products. formulation dierences other than the presence of ingredients that are skin irritants that would impact the integrity of the skin barrier. For use considerations based on RIFM data, the SAFs for fragrance ingredients are based on the use of conrmatory human data which were generated using the RIFM standard HRIPT protocol in which the fragrance ingredient is applied to the back or the upper arm and conducted under full occlusion for 24 h per patch application. Key factors in determining this SAF are primarily site of contact and personal practices that impact barrier function. Table 2 details the numbers assigned to each of the components of the SAF for fragrance ingredients in different types of products. The table also includes the

10 Inter-Individual Variability 1 3 Vehicle/Product Matrix Effects 1 3 Use Considerations Overall SAF 10 30 100 300 1000 10 10

Note: for practical purposes the number 3 is the practical representation of 3.16 (half log of 10)]

Fig. 2. Sensitization assessment factor (SAF). (SAF is calculated by multiplying sub-factors for inter-individual variability, vehicle/matrix eects, and use considerations.)

rationale for selection of the specic number and lists the literature cited references. These SAFs are specic for fragrance ingredients. SAFs for other types of ingredients may vary from these based on the considerations discussed above. 1.3.4.6. Choice of consumer product types. The application of the QRA for fragrance ingredients required the identication of a range of product types. The list of product types is given in Table 2, column 1 and is based on those products listed in the SCCNFP Notes of Guidance (SCCNFP, 2003), on products surveyed by CTFA and Colipa, on products specied in the IFRA Standards and the experience of the authors. This list is not intended to be all inclusive. 1.4. Exposure 1.4.1. Dose metric As indicated above, the dose metric recommended for use in dermal sensitization risk assessments for fragrance ingredients is dose/area (lg/cm2). Support for this position is based on an understanding of the immunological principles of induction of dermal sensitization and from clinical and pre-clinical data. Based upon the understanding of the immunological mechanism involved, it is logical to assume that for an immune response to be initiated, a certain number of Langerhans Cells (LC) are required to be activated and to migrate out of the skin to the nearest lymph node in order to initiate the cascade of events to exceed a threshold of induction for skin sensitization. This would suggest that for the induction of contact allergy, the application of an amount of allergen expressed as percent weight volume is not as important as understanding both the dose applied and the surface area over which the allergen is applied. This is diagrammatically expressed in Fig. 3. Published data that support the use of this dose metric for the induction of skin sensitization is both robust and convincing in humans and animals. There are a number

16

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

Fig. 3. Importance of dose/unit area for the induction of skin sensitization.

of literature references to support this position (Kligman, 1966; Magnusson and Kligman, 1970; Friedmann and Moss, 1985; White et al., 1986; Rees et al., 1990; Upadhye and Maibach, 1992). One of the most important pieces of work in this area was conducted by Kligman in 1966 in which he investigated sensitizing areas of exposure, stimulation of more than one lymph node using various sites, use of a number of smaller patches versus one larger patch and dierent exposure conditions. Work conducted by Friedmann and his colleagues in the 1980s (Friedmann et al., 1983) clearly demonstrated that the total dose of allergen per area of skin (e.g., lg/cm2) is the critical exposure determinant for the induction of contact sensitization. This was conrmed by White et al. in 1986. Moreover, the work of Rees et al. (1990) identied that for very small areas, under 0.1 cm2, the doseresponse is signicantly diminished. This suggested there is a minimum area of contact required to induce contact allergy. The animal data are consistent with the human clinical data. Magnusson and Kligman (1970) guinea pig data showed that the concentration of allergen per unit area was most important. A comprehensive in-depth review of the published studies including those mentioned above that support the use of dose per unit area in risk assessments for induction of dermal sensitization is provided in the publication by Kimber et al. (2008). The eectiveness with which a material can cause dermal sensitization depends on a number of factors. Of key importance is the skin penetration of the material, i.e., the topical dose versus the dose delivered to the viable epidermis in the skin. In addition to skin penetration, other factors, such as evaporation, metabolism (either inactivation of activation), sequestration in the stratum corneum, binding to protein or cells in the epidermis, and uptake and presentation by antigen-presenting cells, determine if and how strong an immune response is triggered. Typically there is very little information available about the bioavailability of the material in either the experimental situation or real life exposure scenario. The application of the SAF account for this area of uncertainty. Consequently, for QRA, topical doses, expressed as dose/unit area, can be used in the denition of NESIL and CEL.

1.4.2. Consumer exposure level (CEL) Consumer exposure level (CEL) is an essential element of QRA. As such a prerequisite for risk characterization is to understand how consumers will be exposed to fragrance ingredients from use of the consumer products. The CEL (expressed as dose/unit area/day) is a measure of exposure under intended and foreseeable conditions of use (but not abuse) and takes account of the frequency of use, habits, and practices (e.g., how consumers use the product), duration of use and amount of product used per application/use. It should be noted that the CEL dened within this paper addresses consumer products that are bought for personal use. Occupational/professional exposure is not addressed in this paper because comprehensive exposure data are not available. If the frequency of product use may be more than once a day, material accumulation on the same skin site should be considered, (depending upon the physical chemical properties of the material). For frequency of use less than once per day, the conservative default of once per day was used with the exception of nail care products. When it is known that products are used in a regimen, such cumulative exposure should be taken into account. Although it is desirable to use aggregate exposure, there are insucient data to allow this to occur at this time. This is identied as an area of renement for a QRA approach. It is important to have reliable habits and practices and accurate human parameters data. Skin penetration is not specically addressed in measuring consumer exposure since the dose metric is unit weight applied per unit area to the outer surface of the skin. As such, using a conservative approach, the topical dose is taken to be the delivered dose. Dierences in skin penetration due to dierent product matrices are accounted for in the nal risk assessment by use of the matrix SAF as previously discussed. Using these criteria, the data sources given in Table 3 were used in the calculation of CEL. A hierarchy was established for how to use the data based on robustness and scope. When measured data for the same product type were available from more than one source, the most conservative value (i.e., the highest value) was used unless there was a sound scientic rationale to use data from

Table 3 Calculation of consumer exposure levels (CEL) from available habits and practices and human parameters data for dierent product types (exposures used in the QRA method are bold and italicized)
Product Type Surface area, cm2 Surface area reference Retention factora EC or SCCNFPa CTFAb Cano and Rich (2001); Tozer et al. (2004); Cano (2006)c 95th Percentile (mg/cm2/day) Colipad HERAe (mg/cm2/ day) mg/cm2/ day Dec. 2005 (mg/cm2/day) FMAf (mg/cm2/ day) RIFMg (mg/cm2/ day)

mg/ application

applications/ day

mg/cm2/ day

90th Percentile mg/ day mg/cm2/ day

90th Percentile mg/ day

Deo/AP-type not specied Deo/AP spray Deo/AP non-spray Deo/AP all over body Solid AP Shaving cream/ depilatoryg,h Lip products Eye productsi Body cream/lotion Mens facial cream Toothpaste

100 100 100 100 100 305 4.8 24 12,895 775 216.8

Bremmer et al. (2003), per axillae Bremmer et al. (2003), per axillae Bremmer et al. (2003), per axillae Bremmer et al. (2003), per axillae Bremmer et al. (2003), per axillae Bremmer et al. (2003) (1/4 area head, male) Ferrario et al. (2000) Bremmer et al. (2003) EPA (1997) (area body head and 1/2 trunk, female)j Bremmer et al. (2003) (1/4 area head + 1/2 area hands, male) Collins and Dawes (1987); Ferrario et al. (2000) (buccal + lips) Collins and Dawes (1987); Ferrario et al. (2000) (buccal + lips) Bremmer et al. (2003) (1/4 area head + 1/2 area hands, male) Bremmer et al. (2003), perfume spray EPA (1997) (1/2 area head, female) EPAc (1/2 area head, female) EPA (1997) (1/2 area head, female) EPA (1997) (1/2 area head, female) EPA (1997) (1/2 area head, female) Bremmer et al. (2003) & EPA (1997) (1/2 area hands +1/2 head) EPA (1997) (area hands + 1/2 head) EPA (1997) (area hands +1/2 head) EPA (1997) (1/2 area head, female) RIVMb EPA (1997) (area hands) EPA (1997) (area hands) EPA (1997) (area hands) EPA (1997) (1/2 area head, female) EPA (1997) (body area, female) EPA (1997) (body area, female)

1 1 1 1 1 0.01 1 1 1 1 0.1
k

500

2.50 6100* 1500 6500* 1700 8.50 30.5* 7.5 32.5* 9.1**

2000 10 10 8000 800 1400

1 4 2 0.5 2 2

0.07 8.33 0.83 0.31 2.06 1.29 2700 1.25 1.0k 55 52 14,400 11.46 2.17 1.12 56 7800 11.67 2.5 0.60

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

Mouthwash

216.8

0.01k

10,000

1.38

1.38

1.0k

Hydroalcoholic products for shaved skinl Hydroalcoholic products for unshaved skin Womens facial cream Womens facial liquid make-up Hair spraystype not specied Hair spraysaerosoll Hair sprayspump sprayl Hair styling aids

775

2.21

100

1770

17.70

2.21

555 555 555 555 555 1010

1 1 0.1 0.1 0.1 0.1

800

2.88

3500 1760

6.31 3.17

1500

2.70 1.08

2700

0.97 7730 12220 1.39 2.20

0.45

5000

0.99

0.15 (mousse) 0.5 (gel) 23630 28200 0.17 0.20 0.3 10500 0.07

Shampoo Conditioners, rinseo Make-up remover Nail care Bar soaps Liquid soap Hand cream Face washes, gels, scrubs Body wash gels, foams, mousses Bath foams, gels, moussesh

1430 1430 555 11 840 840 840 555 16,900 16,900

0.01 0.01 0.1 0.1 0.01 0.01 1 0.01 0.01 0.01

8000 14000 2500 250 800

1 1 2 0.43 6

0.056 0.098 0.90 0.97 0.057

0.05 0.2 4.2 8300 25500 0.15 0.015 0.009

800 5000 17000

2 2 1

0.03 0.006 0.010

17

(continued on next page)

18

Table 3 (continued) Product Type Surface area, cm2 Surface area reference Retention factora EC or SCCNFPa CTFAb Cano and Rich (2001); Tozer et al. (2004); Cano (2006)c 95th Percentile (mg/cm2/day) Colipad HERAe (mg/cm2/ day) mg/cm2/ day Dec. 2005 (mg/cm2/day) FMAf (mg/cm2/ day) RIFMg (mg/cm2/ day)

mg/ application

applications/ day

mg/cm2/ day

90th Percentile mg/ day mg/cm2/ day

90th Percentile mg/ day

Feminine hygienetampons Feminine Hygienepads Feminine Hygieneliners Baby diapers Baby wipes Intimate wipes Aerosol air freshener

2.9 0.14 0.14 0.0006 4.0 4.4 0.025

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

3425

Insect repellent (intended to be applied to the skin) Handwash laundry Laundry tablets & Powder Hand dishwashing Fabric clothing Tights with moisturizers Hard surface cleaner Candles

4225

EPA (1997) (1/2 area head + upper extremities, female) EPA (2001b) (25% body area, femalehead, hands, forearms, legs)

3.02

0.1 Insignicant 0.01 Insignicant 6570 EPA (1997) (lower extremities, female) 1 0.12 0.00033 0.00005

Note: Products that contain sunscreen are not addressed separately but are included in the major product type (e.g., lip creams with sunscreen are included in lip product category). Hair sprayexposure for the pump spray is recommended for all hair sprays since this gure was the most conservative (e.g., highest) value.
a b c d e f g

EC (1996) or SCCNFP (2003) Guidelines. Loretz et al. (2005, 2006); CTFA (2005,a,b). Cano and Rich (2001); Tozer et al. (2004); Cano (2006). Colipa (2005). AISE/HERA (2002). Api et al. (2007).

RIFM (2005), AM Api, Internal memo December 12, 2005, on dermal exposure to pressurized aerosol air fresheners; RIFM (2006), Memo to AM Api from RIFM Member Company, May 2006 on exposure to feminine hygiene products, diapers, intimate wipes and baby wipes; RIFM (2007), Memo to AM Api from RIFM Member Company, January 2007 on exposure to tights with moisturizers, RIFM (2007), Memo to AM Api from RIFM Member Company, January 2007 on exposure to insect repellents intended to be applied on the skin, Cowan-Ellsberry et al. (2008), exposure on axilla surface area.
h Shaving cream/depilatory cream productsthe amount used was derived from the EC (1996) Technical Guidance Document in support of Commission Directive 93/67/EEC on risk assessment for new notied substances and Commission Regulation (EC) No. 1488/94 on risk assessment for existing substances. This reference did not distinguish between shaving the face or shaving the leg. As such, the dose/unit area for shaving the face was calculated and the same value was applied to shaving or depilating the legs. In the absence of more robust data, this was assumed to be a reasonable and conservative approach. i j

For frequency of use less than once per day, the default of once per day was used with the exception of nail care products.

Eye productsthis is based on the CTFA measured data for all types of eye shadows from a specically designed exposure study for eye products. The SCCNFP (2003) exposure data on mascara product types were not used for the eye product category because there is little if any skin contact from this product type.
k l

Body cream/lotionthe surface area comprises the total body surface area for a female minus the area of the head and half the trunk. This is based on habits and practices data for adults that indicate that body lotion is not applied to the head or the back.

These are product dilution factors. Dierent dilution factors are used for mouthwashes and toothpastes. The dilution factor used for mouthwashes is 1% or 0.01 and that used for toothpastes is 10% or 0.1. These values are dierent from the values used in the SCCNFP (2003) Guidelines, but considered to be more relevant since it takes into account the amount remaining in the oral cavity and perioral area rather than that ingested. It also takes into account salivation and distribution across the oral cavity surface (Muhlemann and Rudolf, 1975; Zero et al., 1988; Issa and Toumba, 2004). The dierence in the dilution factors used for mouthwashes and toothpastes is based on the fact that while very dierent volumes of each product are applied (i.e., 30 g/day of mouthwash versus 2.7 g of toothpaste), it is reasonable to expect that similar amounts of product would be in contact with the mouth (buccal cavity and lips) at any one time since the same surface area is involved. The exposure to oral care products (toothpastes and mouthwashes) is impacted by salivation, product dilution and distribution across the oral surfaces and the focus for sensitization reactions is the perioral area. As such, in order to benchmark against the exposure approach used here, a worst case exposure scenario was evaluated using the principles of HERA. In HERA, it was assumed that a 0.01 cm lm thickness was left on the skin (Vermeire et al., 1993) from a 10% aqueous product solution. This would result in a worst case exposures of 1 mg/cm2, assuming 100% retention of the fragrance ingredient from the product solution. This is consistent with the value identied by the primary exposure approach.
* ** ***

These data should not be used due to logistical diculties with determination of the actual amount of product delivered on skin (Colipa, 2005). This exposure value is used in the QRA for fragrance ingredients for all types of deodorants and antiperspirants. This exposure value is used in the QRA for fragrance ingredients for all types of hair sprays.

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

19

another source, e.g., (1) Cano & Rich hydroalcoholic data were used over the CTFA hydroalcoholic data because the former reported distributions of amount, frequency, and surface area in the same study while CTFA reported a distribution only of amounts in their study, (2) the Colipa (2005) exposure study data were used over the CTFA data published in Loretz et al. (2005) on the basis the Colipa study participants used their own products rather than products supplied by the study investigator as in the CTFA study, (3) Cowan-Ellsberry et al. (2008) deodorant/antiperspirant data were used over CTFA and Colipa data because Cowan-Ellsberry et al. (2008) used measured 90th percentile exposure (amount) and surface area data and integrated it into a per diem exposure). All of these sources of exposure data are based on information of varying detail and completeness. This means that the robustness of the exposure data can also be dierent. For these reasons when evaluating a distribution of exposure data, the same percentile data point cannot be selected for each set of exposure data. For example, the 90th percentile was chosen from the Colipa exposure study to dene the most appropriate exposure level given the conservatisms in the model (Colipa, 2005). On the other hand, whilst the study conducted by Cano and Rich (2001); Tozer et al. (2004); and Cano (2006) measured distribution of amount, frequency of use and surface area it did not include the same conservatisms as the Colipa study. On this basis it was more appropriate to choose a higher percentile from this study and therefore the 95th percentile was chosen. Several authoritative sources of human parameters data (i.e., skin site surface areas) (Collins and Dawes, 1987; EPA, 1997; Ferrario et al., 2000; Bremmer et al., 2003; Cowan-Ellsberry et al., 2008) were used and a hierarchal approach applied. Preference was given conservatively to the smaller surface area (i.e., 50th percentile in combination with the measured CTFA and Colipa exposure data). The exceptions to this are studies in which exposure and surface area data are integrated (i.e., Cano and Rich, 2001; Tozer et al., 2004; Cano, 2006; CowanEllsberry et al., 2008). Within these human parameters data sources, the individual references used to support the consumer exposure to dierent product types are detailed in Table 3. 1.5. Risk characterization There are two key elements involved in risk characterization in the recommended approach. These are the Acceptable Exposure Level (AEL) and the comparison of that AEL to the CEL. The practical application of risk characterization to the identication of product categories is detailed below. 1.5.1. Acceptable exposure level (AEL) The AEL is determined by dividing the WoE NESIL by the product type SAF.

Table 4 Risk characterization: calculation of AEL for a hypothetical fragrance ingredient (X) in a deodorant product and hydroalcoholic product for unshaved skin Fragrance ingredient X WoE NESIL SAF AEL = WoE NESIL/SAF AEL Deodorant 500 lg/cm2 300 =500/300 1.7 lg/cm2 Hydroalcoholic product for unshaved skin 500 lg/cm2 100 =500/100 5.0 lg/cm2

AEL

WoE NESIL SAF

The AEL is expressed in terms of dose/unit area/day. The denition of this AEL allows identication of exposures to fragrance ingredients that are acceptable (below the AEL) or unacceptable (above the AEL). This is demonstrated below in Table 4 for a hypothetical fragrance ingredient (X) in a deodorant product and a hydroalcoholic product for unshaved skin. 1.5.2. AEL/CEL ratio To establish the acceptability of consumer exposure to a fragrance ingredient in a given product, the ratio of the AEL to the CEL is determined by dividing the AEL by the CEL (AEL/CEL). The percent concentration of the fragrance ingredient in a product type is acceptable if the AEL exceeds the CEL. The converse, where the CEL exceeds the AEL, would require re-evaluation of the risk management and may lead to a decrease in the concentration of fragrance ingredient in that product type. This is demonstrated below in Table 5 for the same hypothetical fragrance ingredient (X), which is being used at 0.1% in a deodorant product and in a hydroalcoholic product for unshaved skin. For the purposes of these practical examples, for an acceptable risk assessment, the AEL has to be greater than or equal to the CEL (i.e., AEL P CEL). 1.5.3. Product categories A practical application of the recommended risk assessment approach for fragrance ingredients is to form product categories for the implementation of IFRA Standards. The process to dene product categories and the use of this approach to establish IFRA Standards is described in separate publications (Api and Vey, submitted for publication). 2. Conclusions QRA represents a very important step forward in skin sensitization risk assessment. Implementation by the fragrance industry of the QRA approach for fragrance ingredients described in this review has now begun. Principles of general toxicology risk assessment can be applied to induction of skin sensitization since this

20

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

Table 5 Risk characterization: determination of acceptability for 0.1% of fragrance ingredient X in a deodorant product and in a hydroalcoholic product for unshaved skin Fragrance ingredient X WoE NESIL SAF AEL Product exposurea Concentration of fragrance X in the product CEL Risk assessment
a

Deodorant 500 lg/cm2 300 1.7 lg/cm2 9.1 mg/cm2/day 0.1% =0.1% * 9.1 mg/cm2 * 1000 lg/mg =9.1 lg/cm2 Unacceptable (AEL < CEL)

Hydroalcoholic product for unshaved skin 500 lg/cm2 100 5 lg/cm2 2.2 mg/cm2/day 0.1% =0.1% * 2.2 mg/cm2 * 1000 lg/mg =2.2 lg/cm2 Acceptable (AEL P CEL)

Product exposure selected for this example is the data from Cowan-Ellsberry et al. (2008) for antiperspirants; and the 95 percentile data from Tozer et al. (2004) study for hydroalcoholic products for unshaved skin.

is also a threshold phenomenon. However, these general principles require tailoring to take into account unique elements of dermal sensitization as a toxicity endpoint. Following identication of a fragrance ingredient as a potential dermal sensitizer, a weight of evidence approach is used to determine its NESIL. This introduces a better approach to allergen potency evaluation for use in risk assessment. SAFs within the dermal sensitization QRA approach are based on published peerreviewed scientic data and have been predened for certain product types. As with all risk assessment, exposure is a critical element and in this approach the CEL is calculated using the best available habits and practices and human parameters data. The NESIL, CEL, and AEL are expressed in quantity of allergen per unit area in keeping with empirical evidence. The dermal sensitization QRA approach can be used to estimate safe exposure levels for fragrance ingredients. In this way, it can be used as a basis for risk management. For fragrance ingredients QRA could be used both prospectively and retrospectively. Prospective use of QRA in this context would address identifying acceptable levels in products for which IFRA Standards do not exist. Retrospective use of QRA could help to determine the acceptability or unacceptability of current IFRA Standards. With the implementation of the QRA approach, IFRA/ RIFM are recommending the use of the RIFM standard HRIPT protocol for generation of conrmatory human data for use in QRA. Details of this standard HIRPT protocol are available from Politano and Api (2008). Diagnostic patch test data from dermatology clinics are not used in the determination of the NESIL. This is because these data are a measure of elicitation of allergic contact dermatitis, not induction of dermal sensitization. To date there are insucient data to discern any quantitative relationship between induction and elicitation. Clinical results from the dermatology community and company post-market surveillance data should be used to conrm the eectiveness of QRA-based risk management procedures. There may be renements to this dermal sensitization QRA approach for fragrance ingredients in the future as new information becomes available. Some key areas for potential renement are (1) improved exposure data (i.e.,

habits and practices, human parameter data) to further rene CEL and extend it to include occupational/professional exposure to consumer products; (2) the inuence of LLNA EC3 values on the WoE NESIL determinations, may be re-evaluated as more experience is gained with its use as a indicator of human allergenic potency; and (3) SAFs, where additional data (e.g., the inuence of evaporation, of retention factors) may lead to renement.

Conict of Interest Anne Marie Api is an employee of the Research Institute for Fragrance Materials, an independent research institute supported by the manufacturers of fragrances and consumer products containing fragrances.

Funding Source This research was supported by the Research Institute for Fragrance Materials, an independent research institute that is funded by the manufacturers of fragrances and consumer products containing fragrances.

Acknowledgments The authors thank the fragrance, cosmetic, and consumer products industries, their member companies and their respective trade associations (e.g., the European Flavor and Fragrance Association (EFFA); the International Fragrance Association (IFRA); the Research Institute for Fragrance Materials, Inc. (RIFM); the Cosmetic Toiletries and Fragrance Association (CTFA); the European Cosmetic, Toiletry and Perfumery Association (COLIPA); the International Association for Soaps, Detergents and Maintenance Products (A.I.S.E.); the Soap and Detergent Association (SDA)) for all their support, input and review of the work undertaken by this expert group and its nal outcome, the technical dossier on dermal sensitization QRA for fragrance ingredients.

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323

21

References
Aldridge, J.E., Gibbons, J.A., Flaherty, M.M., Kreider, M.L., Romano, J.A., Levin, E.D., 2003. Heterogeneity of toxicant response: source of human variability. Toxicological Sciences 76 (1), 320. AISE/HERA, 2002. AISE (Association Internationale de la Savonnerie, de la Detergence et des Produits dEntretien): table of habits and practices for consumer products in Western Europe. Developed with the HERA project in 2002. AISE. American Beauty Association, 2002. Analysis of the subungual penetration of 100% dibutyl phthalate (DBP) in human ngernails. Unpublished Report to CIR. Api, A.M., Bredbenner, A., McGowen, M., Niemiera, D., Parker, L., Renskers, K., Selim, S., Sgaramella, R., Signorelli, R., Tedrow, S., Troy, W., 2007. Skin contact transfer of three fragrance residues from candles to human hands. Regulatory Toxicology and Pharmacology 48, 279283. Api, A.M., Vey, M., 2008. Implementation of the dermal sensitization quantitative risk assessment (QRA) for fragrance ingredients. Regulatory Toxicology and Pharmacology 52, 5361. Basketter, D.A., Blaikie, L., Dearman, R.J., Kimber, I., Ryan, C.A., Gerberick, G.F., Harvey, P., Evans, P., White, I.R., Rycroft, R.J.G., 2000. Use of the local lymph node assay for the estimation of relative contact allergenic potency. Contact Dermatitis 42 (6), 344348. Basketter, D.A., Lea, L.J., Dickens, A., Briggs, D., Pate, I., Dearman, R.J., Kimber, I., 1999. A comparison of statistical approaches to the derivation of EC3 values from local lymph node assay dose responses. Journal of Applied Toxicology 39 (4), 621627. Benfeldt, E., Serup, J., Menne, T., 1999. Eect of barrier perturbation on cutaneous salicylic acid penetration in human skin: in vivo pharmacokinetics using microdialysis non-invasive quantication of barrier function. British Journal of Dermatology 140, 739748. Bremmer H.J., PrudHomme de Lodder, L.C.H., van Engelen, J.G.M., 2003. Cosmetics Fact Sheet to Assess The Risks For the Consumer. RIVM Report 6128100. Britz, M.B., Maibach, H.I., 1979. Human cutaneous vulvar reactivity to irritants. Contact Dermatitis 5, 375377. Britz, M.B., Maibach, H.I., 1979a. Human labia majora skin: transepidermal water loss in vivo. Acta Dermato-Venereologica 59, 2325. Bucks, D.A.W., Maibach, H.I., Guy, R.H., 1989. Occlusion does not uniformly enhance penetration in vivo. In: Percutaneous Absorption: MechanismsMethodologyDrug Delivery, Chapter 5. Marcel Dekker, pp. 7789. Burin, G.J., Saunders, B.R., 1999. Addressing human variability in risk assessment. The robustness of the intraspecies uncertainty factor. Regulatory, Toxicology, Pharmacology 30, 209216. Cano, M.-F., Rich, K., 2001. Presentation at the 2001 Colipa Scientic Forum. Cano, M.-F., 2006. Personal communication on exposure data for hydroalcoholic products for shaved skin. Cassimos, C., Kanakoudi-Tsakalidis, F., Spyroglou, K., Ladianos, M., Tzaphi, R., 1980. Skin sensitization to 2,4-dinitrochlorobenzene (DNCB) in the rst months of life. Journal of Clinical Laboratory Immunology 3 (2), 111113, 48632. Chappell, W.R., Mordenti, J., 1991. Extrapolation of toxicological and pharmacological data from animals to humans. Advances in Drug Research 20, 1116. Collins, L.M.C., Dawes, C., 1987. The surface area of the adult human mouth and thickness of the salivary lm covering the teeth and oral mucosa. Journal of Dental Research 66 (8), 13001302. Colipa, 2005. Updated daily consumer exposure to cosmetic products. Unpublished submission to SCCP, December 2005. Cowan-Ellsberry, C., McNamee, P.M., Leazer, T., 2008. Axilla surface area for males and females: measured distribution. Regulatory Toxicology and Pharmacology 52, 4652. CTFA, 2005. Summary distributions of facial cleanser use data. Unpublished report, August 5, 2005.

CTFA, 2005a. Summary distributions of powder eye shadow use data. Unpublished report, August 5, 2005. CTFA, 2005b. Summary distributions of rinse-o hair conditioner use data. Unpublished report, August 5, 2005. Cumberbatch, M., Scott, R.C., Basketter, D.A., Scholes, E.W., Hilton, J., Dearman, R.J., Kimber, I., 1993. Inuence of sodium lauryl sulfate on 2, 4-dinitrochlorobenzene-induced lymph node activation. Toxicology 77, 181191. Cunico, R.L., Maibach, H.I., Khan, H., Bloom, E., 1977. Skin barrier properties in the newborn. Biologia Neonatorum 32, 177182. de Vries, M.E., Boddle, H.E., Verhoef, J.C., Junginger, H.E., 1991. Developments in buccal drug delivery. Critical Reviews in Therapy, Drug Carrier Systems 8, 271303. Dourson, M.L., Felter, S.P., Robinson, D., 1996. Evolution of sciencebased uncertainty factors in noncancer risk assessment. Regulatory, Toxicology and Pharmacology 24, 108120. Dupuis, G., 1979. Studies on poison ivy. In vitro lymphocyte transformation by urushiol-protein conjugates. British Journal of Dermatology 101, 617624. EC, 1996. Technical guidance document in support of Commission Directive 93/67/EEC on risk assessment for new notied substances and Commission Regulation (EC) No. 1488/94 on risk assessment for existing substances. European Commission, Luxembourg. ECETOC, 2003. Contact Sensitisation: Classication According to Potency. Technical Report No. 87. Edman, B., 1994. The inuence of shaving method on perfume allergy. Contact Dermatitis 31 (5), 291292. Elsner, P., Maibach, H.I., 1990. The eect of prolonged drying on transepidermal water loss, capacitance and pH of human vulvar and forearm skin. Acta Dermato-Venereologica 70, 105109. Elsner, P., Wilhelm, D., Maibach, H.I., 1990. Frictional properties of human forearm and vulvar skin: inuence of age and correlation with transepidermal water loss and capacitance. Dermatologica 181, 8891. Elsner, P., Wilhelm, D., Maibach, H.I., 1990a. Irritant eect of a model surfactant on the human vulvar and forearm skin. Age-related dierences. Journal of Reproductive Medicine 35, 10351039. Elsner, P., Wilhelm, D., Maibach, H.I., 1990b. Physiological skin surface water loss dynamics of human vulvar and forearm skin. Acta Dermato-Venereologica 70, 141144. Elsner, P., Wilhelm, D., Maibach, H.I., 1990c. Sodium lauryl sulphateinduced irritant contact dermatitis in vulvar and forearm skin of premenopausal and postmenopausal women. Journal of the American Academy of Dermatology 23, 648652. Elsner, P., Wilhelm, D., Maibach, H.I., 1991. Eect of low-concentration sodium lauryl sulphate on human vulvar and forearm skin. Agerelated dierences. Journal of Reproductive Medicine 36, 7781. EPA, 1997. Exposure Factors Handbook. Doc EPA/600/P-95/002Fa. Oce of Research and Development, U.S. Environmental Protection Agency, Washington, DC. EPA (U.S. Environmental Protection Agency), Whalan, J., Weiss, S., 2001b. Human Health Risk Assessment for KBR 3023 All Family Insect Repellent Spray & Cream. DP Barcode: D279005, PC Code: 070705. November 16, 2001 Memorandum to K. Sweeney. Oce of Prevention, Pesticides and Toxic Substances, U.S. EPA, Washington, DC. EU Cosmetics Directive 76/768/EECSeventh Amendment (Council Directive 2003/15/EC). Fairley, J.A., Rasmussen, J.E., 1983. Comparison of stratum corneum thickness in children and adults. Journal of the American Academy of Dermatology 8 (5), 652654. Farage, M., Maibach, H.I., 2004. The vulvar epithelium diers from the skin: implications for cutaneous testing to address topical vulvar exposures. Contact Dermatitis 51 (4), 201209. Farage, M.A., Bjerke, D.L., Mahony, C., Blackburn, K.L., Gerberick, G.F., 2003. Quantitative risk assessment for the induction of allergic contact dermatitis: uncertainty factors for mucosal exposures. Contact Dermatitis 49 (3), 140147.

22

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323 Kobayashi, H., Tagami, H., 2004a. Distinct locational dierences observable in biophysical functions of the facial skin: with special emphasis on the poor functional properties of the stratum corneum of the perioral region. International Journal of Cosmetic Science 26, 91101. Lesch, C.A., Squier, C.A., Cruchley, A., Williams, D.M., Speight, P., 1989. The permeability of human oral mucosa and skin to water. Journal of Dental Research 68, 13451349. Leveque, J.L., Corcu, P., de Rigal, J., Agache, P., 1984. In vivo studies of the evolution of physical properties of the human skin with age. International Journal of Dermatology 23 (5), 322329. Loretz, L., Api, A.M., Barraj, L.M., Burdick, J., Dressler, W.E., Gettings, S.D., Hsu, H.H., Pan, Y.H.L., Re, T.A., Renskers, K.J., Rothenstein, A., Scraord, C.G., Sewall, C., 2005. Exposure data for cosmetic products: lipstick, body lotion, and face cream. Food and Chemical Toxicology 43, 279291. Loretz, L., Api, A.M., Barraj, L.M., Burdick, J., Davis, D.A., Dresser, W., Gilberti, E., Jarrett, G., Mann, S., Pan, Y.H.L., Re, T.A., Renskers, K.J., Scraord, C., Vater, S., 2006. Exposure data for personal care products: hair spray, spray perfume, liquid foundation, shampoo, body wash, and solid antiperspirant. Food and Chemical Toxicology 44, 20082018. Magnusson, B., Kligman, A.M., 1970. Allergic Contact Dermatitis in the Guinea Pig. C.C. Thomas, Springeld, IL, pp. 5056. Matts, P., Rawlings, A., 2005. The dry skin cycle: introducing a new model for the induction and propagation of xerosis in normal skin. Journal of American Academy Dermatology 52 (3 Suppl. 1), 94. McCormack, J.J., Boisits, E.K., Fisher, L.B., 1982. An in vitro comparison of the permeability of adult versus neonatal skin. in: neonatal skin: structure and function. In: Maibach, H.I., Boisits, E.K. (Eds.), Neonatal Skin: Structure and Function. Marcel Dekker Inc., New York, pp. 149164. McNamee, P.M., Api, A.M., Basketter, D.A., Gerberick, G.F., Gilpin, D.A., Hall, B.M., Jowsey, I., Robinson, M.K., 2008. A review of critical factors in the conduct and interpretation of the human repeat insult patch test. Regulatory Toxicology and Pharmacology 52, 2434. Moss, C., Friedmann, P.S., Shuster, S., Simpson, J.M., 1985. Susceptibility and amplication of sensitivity in contact dermatitis. Clinical and Experimental Immunology 61, 232241. Muhlemann, H.R., Rudolf, E.R., 1975. Fluoride retention after rinsing with sodium uoride and amine uoride. Helvetica Odontologica Acta 19 (1), 8184, 48762. Nuutinen, J., Harvima, I., Lahtinen, M.-R., Lahtinen, T., 2003. Water loss through the lip, nail, eyelid skin, scalp skin and axillary skin measured with a closed-chamber evaporation principle. British Journal of Dermatology 148 (4), 839841. Odio, M., Friedlander, S.F., 2000. Diaper dermatitis and advances in diaper technology. Current Opinions in Pediatrics 12 (4), 342346. Politano, V.T., Api, A.M., 2008. The research institute for fragrance materials human repeated insult patch test protocol. Regulatory Toxicology and Pharmacology 52, 3538. Rees, J.L., Friedmann, P.S., Matthews, J.N., 1989. Sex dierences in susceptibility to development of contact hypersensitivity to dinitrochlorobenzene (DNCB). British Journal of Dermatology 120, 371374. Rees, J.L., Friedmann, P.S., Matthews, J.N.S., 1990. The inuence of area of application on sensitization by dinitrochlorobenzene. British Journal of Dermatology 122, 2931. Renwick, A.G., Lazarus, N.R., 1998. Human variability and non-cancer risk assessment. An analysis of the default uncertainty factor. Regulatory, Toxicology and Pharmacology 27, 320. Research Institute for Fragrance Materials Inc. (RIFM), 2005. A.M. Api internal memo, December 12, 2005 on dermal exposure to pressurized aerosol air fresheners. Research Institute for Fragrance Materials Inc. (RIFM), 2006. Memo to A.M. Api from RIFM Member Company, May 2006 on exposure to feminine hygiene products, diapers, intimate wipes and baby wipes. Research Institute for Fragrance Materials Inc. (RIFM), 2007. Memo to A.M. Api from RIFM Member Company, January 2007 on exposure to tights with moisturizers.

Feldmann, R.J., Maibach, H.I., 1967. Regional variation in percutaneous penetration of 14C Cortisol in man. Journal of Investigative Dermatology 48, 181183. Felter, S.P., Robinson, M.K., Basketter, D.A., Gerberick, G.F., 2002. A review of the scientic basis for uncertainty factors for use in quantitative risk assessment for the induction of allergic contact dermatitis. Contact Dermatitis 47 (5), 257266. Felter, S.P., Ryan, C.A., Basketter, D.A., Gerberick, G.F., 2003. Application of the risk assessment paradigm to the induction of allergic contact dermatitis. Regulatory Toxicology and Pharmacology 37, 110. Ferrario, V.F., Sforza, C., Schmitz, J.H., Ciusa, V., Colombo, A., 2000. Normal growth and development of the lips: a 3-dimensional study from 6 years to adulthood using a geometric model. Journal of Anatomy 196, 415423. Friedmann, P.S., Moss, C., Shuster, S., Simpson, J.M., 1983. Quantitative relationships between sensitizing dose of DNCB and reactivity in normal subjects. Clinical and Experimental Immunology 53, 709715. Friedmann, P.S., Moss, C., 1985. Quantication of contact hypersensitivity in man. In: Maibach, H.I., Lowe, N.J. (Eds.), . In: Models in Dermatology, vol. 2l. Karger, Basel, pp. 75281. Gerberick, G.F., Robinson, M.K., Felter, S.P., White, I.R., Basketter, D.A., 2001. Understanding fragrance allergy using an exposure-based risk assessment approach. Contact Dermatitis 45 (6), 333340. Gerberick, G.F., Robinson, M.K., Ryan, C.A., Dearman, R.J., Kimber, I., Basketter, D.A., Wright, Z., Marks, J.G., 2001a. Contact allergenic potency: correlation of human and local lymph node assay data. American Journal of Contact Dermatitis 12 (3), 156161. Gerberick, G.F., Ryan, C.A., Kern, P.S., Dearman, R.J., Kimber, I., Patlewicz, G.Y., Basketter, D.A., 2004. A chemical dataset for evaluation of alternative approaches to skin-sensitization testing. Contact Dermatitis 50 (5), 274288. Ghadially, R., Brown, B.E., Sequeira-Martin, S.M., Feingold, K.R., Elias, P.M., 1995. The aged epidermal permeability barrier. Journal of Clinical Investigation 95, 22812290. Griem, P., Goebel, C., Scheer, H., 2003. Proposal for a risk assessment methodology for skin sensitization potency data. Regulatory Toxicology and Pharmacology 38, 269290. Harpin, V.A., Rutter, N., 1983. Barrier properties of the newborn infants skin. Journal of Pediatrics 102, 419425. Harris, D., Robinson, J.R., 1992. Drug delivery via the mucous membranes of the oral cavity. Journal of Pharmaceutical Sciences 81 (1), 110. Holbrook, K.A., 1982. A histological comparison of infant and adult skin. In: Maibach, H.I., Boisits, E.K. (Eds.), Neonatal Skin, Structure and Function. Marcel Dekker Inc., New York. Issa, A.L., Toumba, K.J., 2004. Oral uoride retention in saliva following toothbrushing with child and adult dentifrices with and without water rinsing. Caries Research 38 (1), 1519. Jordan Jr., W.P., King, S.E., 1977. Delayed hypersensitivity in females. The development of allergic contact dermatitis in females during the comparison of two predictive patch tests. Contact Dermatitis 3, 1926. Kalia, Y.N., Nonato, L.B., Lund, C.H., Guy, R.H., 1998. Development of skin barrier function in premature infants. Journal of Investigative Dermatology 111 (2), 320326. Kimber, I., Dearman, R.J., Basketter, D.A., Ryan, C.A., Gerberick, G.F., McNamee, P.M., Lalko, J., Api, A.M., 2008. Dose metrics in the acquisition of skin sensitization: thresholds and importance of dose per unit area. Regulatory Toxicology and Pharmacology 52, 3945. Kimber, I., Gerberick, G.F., Basketter, D.A., 1999. Thresholds in contact sensitization: theoretical and practical considerations. Food and Chemical Toxicology 37 (5), 553560. Kligman, A.M., 1966. The identication of contact allergens by human assay. II. Factors inuencing the induction and measurement of allergic contact dermatitis. Journal of Investigative Dermatology 47, 375392. Kobayashi, H., Tagami, H., 2004. Functional properties of the surface of the vermilion border of the lips are distinct from those of the facial skin. British Journal of Dermatology 150 (3), 563567.

A.M. Api et al. / Regulatory Toxicology and Pharmacology 52 (2008) 323 Research Institute for Fragrance Materials Inc. (RIFM), 2007. Memo to A.M. Api from RIFM Member Company, January 2007 on exposure to insect repellents intended to be applied on the skin. Robinson, M.K., 1999. Population dierences skin structure and physiology and the susceptibility to irritant and allergic contact dermatitis: implications for skin safety testing and risk assessment. Contact Dermatitis 41 (2), 6579. Robinson, M.K., Stotts, J., Danneman, P.J., Nusair, T.L., 1989. A risk assessment process for allergic contact sensitization. Food and Chemical Toxicology 27 (7), 479489. Robinson, M.K., Gerberick, F.G., Ryan, C.A., McNamee, P.M., White, I.R., Basketter, D.A., 2000. The importance of exposure estimation in the assessment of skin sensitization risk. Contact Dermatitis 42 (5), 251259. Sayani, A.P., Chien, Y.W., 1996. Systemic delivery of peptides and proteins across absorptive mucosae. Critical Reviews in Therapeutic Drug Carrier Systems 13, 85184. SCCNFP, 2003. The SCCNFPs Notes of Guidance For the Testing of Cosmetic Ingredients and Their Safety Evaluation, 5th Revision. Adopted by the SCCNFP during the 25th plenary meeting of 20 October 2003. SCCNFP/0690/03 Final. Schaefer, H., Redelmeier, T.E., 1996. Skin barrier. Principles of percutaneous absorption. In: Skin Barrier: Principles of Percutaneous Absorption, Chapter 5, pp. 153212. Scheuplein, R., Ross, L., 1970. Eects of surfactants and solvents on the permeability of epidermis. Journal of Society of Cosmetic Chemists 21, 853873. Schneider, K., Akkan, Z., 2004. Quantitative relationship between the local lymph node assay and human skin sensitization assays. Regulatory Toxicology and Pharmacology, 245255. Smith, C.K., Hotchkiss, S.A.M., 2001. Allergic Contact Dermatitis: Chemical and Metabolic Mechanisms. Taylor & Francis, London. Smith, H.R., Holloway, D., Armstrong, D.K.B., Basketter, D.A., McFadden, J.P., 2000. Irritant thresholds in subjects with colophony allergy. Contact Dermatitis 42, 62. Squier, C.A., Hall, B.K., 1985. The permeability of skin and oral mucosa to water and horseradish peroxidase as related to the thickness of the permeability barrier. Journal of Investigative Dermatology 84, 176179. Squier, C.A., 1986. Penetration of nicotine and nitrosonornicotine across porcine oral mucosa. Journal of Applied Toxicology 6 (2), 123128, 3567. Squier, C.A., 1991. The permeability of oral mucosa. Critical Reviews in Oral Biology and Medicine 2, 1332. Technical Guidance Document, 2003. Technical Guidance Document in support of Commission Directive 93/67/EEC on risk assessment for

23

new notied substances and Commission Regulation (EC) No. 1488/ 94 on risk assessment for existing substances. Thompson, I.O., van der Bijl, P., van Wyk, C.W., van Eyk, A.D., 2001. A comparative light-microscopic, electron-microscopic and chemical study of human vaginal and buccal epithelium. Archives of Oral Biology 46, 10911098. Tozer, S.A., OKeee, L., Cowan-Ellsberry, C.E., Rich, K., 2004. Use of probabilistic analysis in the renement of exposure data for hydroalcoholic perfume products. Toxicology 202 (12), 123124. Travis, C.C., White, R.K., 1988. Interspecic scaling of toxicity data. Risk Analysis 8 (1), 116125. Upadhye, M.R., Maibach, H.I., 1992. Inuence of area of application of allergens on sensitization in contact dermatitis. Contact Dermatitis 27, 281286. Vermeire, T.G., van der Poel, P., van de Laar, R.T.H., Roelfzema, H., 1993. Estimation of consumer exposure to chemicals: application of simple models. The Science of The Total Environment 136 (12), 155 176. Weigand, D.A., Haygood, C., Gaylor, J.R., 1974. Cell layers and density of Negro and Caucasian stratum corneum. Journal of Investigative Dermatology 62, 563568. West, D.P., Worobec, S., Solomon, L.M., 1981. Pharmacology and toxicology of infant skin. Journal of Investigative Dermatology 76, 147150, 1909. Wester, R.C., Maibach, H.I., 1982. Comparative percutaneous absorption. in: neonatal skin: structure and function. In: Maibach, H.I., Boisits, E.K. (Eds.), Neonatal Skin: Structure and Function. Marcel Dekker Inc., New York, pp. 137147. White, S.I., Friedmann, P.S., Moss, C., Simpson, J.M., 1986. The eect of altering area of application and dose per area on sensitization by DNCB. British Journal of Dermatology 155, 663668. Young, E., van Weelden, H., van Osch, L., 1988. Age and sex distribution of the incidence of contact sensitivity to standard allergens. Contact Dermatitis 19, 307308. Zero, D.T., Fu, J., Espeland, M.A., Featherstone, J.D., 1988. Comparison of uoride concentrations in unstimulated whole saliva following the use of a uoride dentifrice and a uoride rinse. Journal of Dental Research 67 (10), 12571262. Zhai, H., Fautz, R., Fuchs, A., Bhuandarkar, S., Maibach, H.I., 2004. Human scalp irritation compared to that of the arm and back. Contact Dermatitis 51 (4), 196200. Zhai, H., Maibach, H.I., 2001. Eects of skin occlusion on percutaneous absorption: an overview. Skin Pharmacology and Applied Skin Physiology 14 (1), 110.