3,7,11-trimethyldodeca-1,6,10-trien-3-ol,mixed isomers

Toxicological Summary

Workers - Hazard via inhalation route

Systemic effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

10 mg/m³

Most sensitive endpoint:

repeated dose toxicity

DNEL related information

Overall assessment factor (AF):

18

Modified dose descriptor starting point:

NOAEC

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

DNEL related information

Local effects

Long term exposure

Hazard assessment conclusion:

no hazard identified

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

DNEL related information

Workers - Hazard via dermal route

Systemic effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

2.8 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

DNEL related information

Overall assessment factor (AF):

72

Modified dose descriptor starting point:

NOAEL

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

DNEL related information

Local effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

122.5 µg/cm²

Most sensitive endpoint:

sensitisation (skin)

DNEL related information

Overall assessment factor (AF):

10

Dose descriptor:

other: NESIL

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

Workers - Hazard for the eyes

Local effects

Hazard assessment conclusion:

low hazard (no threshold derived)

Additional information - workers

For the derivation of DNELs for systemic effects after long term exposure, the oral combined repeated dose/ reproductive toxicity screening study was chosen (BASF 96R0466/09052).

The lowest NOAEL observed in this study, i.e for general maternal toxicity (1500 ppm representing a minimum dose of approx. 100 mg/kg bw/d during the premating period) was chosen as point of departure for derivation of the inhalative and dermal long term systemic DNEL by route to route extrapolation.

On the basis of the toxicokinetic assumptions, a high rate of oral absorption in rats is considered and a 100% bioavailability has been set for the oral route. No data on dermal absorption are available for nerolidol, however, on the basis of its physicochemical characteristics, a low bioavailablility via the dermal route is to be expected and a dermal penetation rate of 50% has been chosen as a worst case for a route to route extrapolation.

Based on the availability of a sufficient toxicity dataset, the default assessment factors (acc. to ECHA GD R8) have been modified into substance specific assessment factors (AF), considering the intrinsic hazard properties of the registered substance.

The values used as point of departure for the respective systemic DNELs were based on general systemic effects of Nerolidol such as impaired body weight gains and liver effects mainly resulting from hepatic enzyme induction / metabolic activation. On the basis of the nature of these general adverse systemic effects observed after Nerolidol administration no difference in sensitivity (toxicodynamic and/or additional toxicokinetic differences) between test animals and humans is to be expected at these Nerolidol dose levels besides aspects covered by allometric scaling.This substance specific argumentation is supported by a probabilistic approach for interspecies extrapolation using the RepDose database (Escher et al. 2013), supporting that species are on average equally sensitive to equipotent doses, if doses are related to energy turnover (uptake via inhalation, drinking water or food). The remaining uncertainties observed in this study were rather low considering the included differences in study design and intraspecies variability. These findings are found to hold true for systemic as well as for local effects after inhalation. Based on this assessment, a default interspecies extrapolation distribution between animals and humans according to the respective allometric scaling factor for body doses (e.g. gavage studies) and a factor of 1 for doses in ppm or mg/m3 (e.g. uptake via food, drinking water or inhalation) is proposed.

Furthermore, due to the general systemic effects observed for Nerolidol such as impaired body weight gains and liver effects mainly resulting from hepatic enzyme induction / metabolic activation, the default assessment factors for intraspecies variability for systemic toxicity after repeated dosing have been amended to less conservative values. These findings are not considered to represent severe specific organ toxicity. The derivation of these Nerolidol specific assessment factors are guided by the following information. In an attempt to evaluate the intraspecies variability within the human population, the distribution of human data for various toxicokinetic and toxicodynamic parameters were examined (Hattis et al 1987, 1999; Hattis and Silver 1994; Renwick and Lazarus, 1998; see ECETOC TR No.86, 2003). These evaluations included data from ‘healthy adults’ of both sexes, as well as limited data from the young and elderly, mixed races and patients with various medical conditions such as cancer and hypertension. The data of Renwick and Lazarus (1998) and Hattis et al. (1999) were based exclusively on human data and similar values were obtained within each percentile. Considering that the data analysed by these authors included both sexes, a variety of disease states and ages, the use of the 95th percentile is considered sufficiently conservative to account for intraspecies variability in the general population. Thus, a default assessment factor of 5 was taken for the general population with a lower factor of 3 (i. e. closer to the 90th percentile) for the more homogenous worker population.

Overall, it needs to be pointed out, that several factors of conservatism have been applied to the systemic DNEL derivation made. Next to the use of the 90-95th percentile concerning intraspecies variability, the multiplicatory principle of AF provide a sufficient degree of conservatism.

 

For the worker, the following DNELs were derived:

The NOAEL of the oral study was set at 100 mg/kg bw/d Nerolidol covering males and females. For derivation of the inhalative DNEL, the oral NOAEL was converted into a corrected inhalative NOAEC of 176 mg/m3 according to the procedure, recommended in the current guidance document (R8, ECHA 2008).

 

NOAECinhal corrected= 100*(1/0.38)*(100/100)*(6.7/10) = 176 mg/m3

 

The following assessment factors (AF) were applied:

·       allometric scaling = 1 (not applicable according to R8 ECHA 2008)

·       remaining differences = 1 (On the basis of the general adverse systemic effects observed, i.e. body weight changes and liver effects mainly resulting from hepatic enzyme induction / metabolic activation at high doses of Nerolidol, no difference in sensitivity (toxicodynamic and/or additional toxicokinetic differences) between test animals and humans is to be expected besides aspects already covered in the route to route extrapolation performed above).

·       intraspecies = 3 (based on the main substance related adverse effects observed after repeated oral dosing and general considerations)

·       exposure duration = 6 (subacute to chronic);

·       quality of whole database = 1 (based on validity of studies performed).

 

AF = 1 x 1 x 3 x 6 x 1 = 18.Consequently, the inhalative long-term systemic DNEL was set at 10 mg/m3 for the worker.

 

For derivation of the long-term systemic dermal DNEL, the oral NOAEL was converted into a corrected dermal NOAEL of 200 mg/kg bw/d according to the procedure, recommended in the current guidance document (R8, ECHA 2008).

 

NOAEL corrected dermal = 100*(100/50) = 200 mg/kg bw/d

 

The following assessment factors (AF) were applied:

·       allometric scaling = 4 (according to R8 ECHA 2008)

·       remaining differences = 1 (On the basis of the general adverse systemic effects observed, i.e. body weight changes and liver effects mainly resulting from hepatic enzyme induction / metabolic activation at high doses of Nerolidol, no difference in sensitivity (toxicodynamic and/or additional toxicokinetic differences) between test animals and humans is to be expected besides aspects already covered by allometric scaling).

·       intraspecies = 3 (based on the main substance related adverse effects observed after repeated oral dosing and general considerations)

·     exposure duration = 6 (subchronic to chronic);

·     quality of whole database = 1 (based on validity of studies performed).

 

AF = 4 x 1 x 3 x 6 x 1 = 72. Consequently, the dermal long-term systemic DNEL derived was 2.8 mg/kg bw/d for the worker.

No DNELs were derived for local effects after short term or after long term inhalative exposure and no DNELs were also derived for systemic effects after short term dermal or inhalative exposure, as the substance exhibits no hazardous potential in terms of these endpoints. Hence at these points there is no need for classification and therefore, there is consequently no need for deriving additional DNELs for the worker.

No hazard concerning dermal irritation exist. To assess the DNEL for local effects after short term and long term dermal exposure, data for skin sensitization were considered.

In the chosen key study, i.e. a murine LLNA according to OECD TG 429 and GLP, (BASF SE; 58V0466/09A141), an EC3 for 3H-thymidine incorporation was determined to be 4.9%. According to the ECHA guidance document R8, this EC3 is converted as follows:

EC3 [μg/cm2] = EC3 [%]*250 [μg/cm2/% ] = 4.9 * 250 = 1225 µg/cm2

Furthermore, human data for Nerolidol need to be taken into account, such as a human maximization test conducted on 25 healthy volunteers, showing no dermal effects at a concentration of 4 % (2760 µg/cm2) tetrahydrolinalool (Kligman 1973). Different patch tests in patients revealed sporadic postive skin reactions, however, these data do not provide a sound basis to derive a skin sensitization induction threshold level relevant for the DNEL derivation. 

The human data available for Nerolidol, i.e. HMT, indicate the absences of a skin sensitization potential at a concentration above the EC3 (µg/cm2) determined in the murine LLNA determined (2760 µg/cm2 versus 1225 µg/cm2, respectively). Since no evident species specific differences in potencies in terms of a higher human susceptibility are indicated between the murine test and human subjects, the use of an additional interspecies assessment factor is not plausible. Although the human data are considered reliable, the point of departure for the local long term dermal DNEL is based on the murine LLNA EC3 as a conservative approach, due to limitations in study design of the available HMT, such as the number of participants.

It is recognized that a general DNEL must take into account that the threshold for skin sensitization varies between individuals. This may be due to differences in parameters such as genetic effects, sensitive subpopulations, inherent barrier function, age, gender, and ethnicity (Api et al., 2008). Whereas the latter three are recognized to have some effect on the sensitization threshold, it is generally recognized that genetic differences, the inherent barrier function and especially sensitive subpopulations play a major role (Api et al., 2008). The barrier function of the skin may be compromised which in turn may lead to a greater susceptibility of the individual. At the same time the barrier function is thought to be very similar from infancy to adulthood. The influence of the genetic setting is not well understood, however, may be plausible in the light of the immunological effect under consideration. The term sensitive subpopulations refers mostly to individuals who have previously been sensitized to other substances which may increase the susceptibility to further sensitizers (Api et al., 2006, Api et al., 2008). Overall, an assessment factor of 10 for intraspecies differences is applied to adequately address the combined influence of these effects.

The relevant parameters, i. e. induction of skin sensitization, are considered to depend on threshold concentrations and not on exposure duration. Therefore no assessment factor concerning exposure duration is deemed necessary for the derivation of the long term local dermal DNEL. The concept of threshold concentrations for the induction of these effects is generally well accepted (see e.g. Api et al. 2006; or Api et al. 2008). Furthermore, the ECHA guidance document does not indicate at all, that an assessment factor for exposure duration needs to be taken into account for the derivation of a DNEL for skin sensitization.

Therefore, a DNEL for skin sensitization was set at 122.5 µg/cm2. Using this DNEL allows for a quantitative risk characterization for skin sensitization.

Furthermore, Nerolidol is to be classified as eye irritant (category 2) according to 1272/2008/EEC. Since no quantitative data addressing the hazard of eye irritation are available, a respective no effect concentration cannot be derived and included in the derivation of the short term/long term local dermal DNEL. However, a qualitative risk characterisation including the implementation of suitable risk management measures is performed in the CSR.

 

·   Api AM, Basketter DA, Cadby PA, Cano M-F, Graham E, Gerberick F, Griem P, McNamee P, Ryan CA, Safford B (2006). Dermal Sensitization Quantitative Risk Assessment (QRA) for fragrance ingredients. Technical dossier. March 15, 2006 (revised May 2006).

·   Api AM, Basketter, DA, Cadby PA, Cano M-F, Ellis G, Gerberick GF, Griem P, McNamee PM, Ryan CA, Safford R (2008). Dermal sensitization quantitative risk assessment (QRA) for fragrance ingredients. Reg Toxicol Pharmacol 52: 3-23.

·   ECETOC (2003). Contact Sensitization: classification according to potency. Technical Report No. 87, April 2003.

·   Escher S, Batke M, Hoffmann-Doerr S, Messinger H, Mangelsdorf I (2013). Interspecies extrapolation based on the RepDose database—A probabilistic approach. Toxicology Letters 218: 159– 165

General Population - Hazard via inhalation route

Systemic effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

2.9 mg/m³

Most sensitive endpoint:

repeated dose toxicity

DNEL related information

Overall assessment factor (AF):

30

Modified dose descriptor starting point:

NOAEC

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

DNEL related information

Local effects

Long term exposure

Hazard assessment conclusion:

no hazard identified

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

DNEL related information

General Population - Hazard via dermal route

Systemic effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

1.7 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

DNEL related information

Overall assessment factor (AF):

120

Modified dose descriptor starting point:

NOAEL

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

DNEL related information

Local effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

122.5 µg/cm²

Most sensitive endpoint:

sensitisation (skin)

DNEL related information

Overall assessment factor (AF):

10

Dose descriptor:

other: NESIL

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

General Population - Hazard via oral route

Systemic effects

Long term exposure

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

0.8 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

DNEL related information

Overall assessment factor (AF):

120

Modified dose descriptor starting point:

NOAEL

Acute/short term exposure

Hazard assessment conclusion:

no hazard identified

DNEL related information

General Population - Hazard for the eyes

Local effects

Hazard assessment conclusion:

low hazard (no threshold derived)

Additional information - General Population

For the derivation of DNELs for systemic effects after long term exposure, the oral combined repeated dose/ reproductive toxicity screening study were chosen (BASF 96R0466/09052).

The lowest NOAEL observed in this study, i.e for general maternal toxicity (1500 ppm representing a minimum dose of approx. 100 mg/kg bw/d during the premating period) was chosen as point of departure for derivation of the oral, inhalative and dermal long term systemic DNEL by route to route extrapolation.

On the basis of the toxicokinetic assumptions, a high rate of oral absorption in rats is considered and a 100% bioavailability has been set for the oral route. No data on dermal absorption are available for nerolidol, however, on the basis of its physicochemical characteristics, a low bioavailablility via the dermal route is to be expected and a dermal penetation rate of 50% has been chosen as a worst case for a route to route extrapolation.

Based on the availability of a sufficient toxicity dataset, the default assessment factors (acc. to ECHA GD R8) have been modified into substance specific assessment factors (AF), considering the intrinsic hazard properties of the registered substance.

The values used as point of departure for the respective systemic DNELs were based on general systemic effects of Nerolidol such as impaired body weight gains and liver effects mainly resulting from hepatic enzyme induction / metabolic activation. On the basis of the nature of these general adverse systemic effects observed after Nerolidol administration no difference in sensitivity (toxicodynamic and/or additional toxicokinetic differences) between test animals and humans is to be expected at these Nerolidol dose levels besides aspects covered by allometric scaling.This substance specific argumentation is supported by a probabilistic approach for interspecies extrapolation using the RepDose database (Escher et al. 2013), supporting that species are on average equally sensitive to equipotent doses, if doses are related to energy turnover (uptake via inhalation, drinking water or food). The remaining uncertainties observed in this study were rather low considering the included differences in study design and intraspecies variability. These findings are found to hold true for systemic as well as for local effects after inhalation. Based on this assessment, a default interspecies extrapolation distribution between animals and humans according to the respective allometric scaling factor for body doses (e.g. gavage studies) and a factor of 1 for doses in ppm or mg/m3 (e.g. uptake via food, drinking water or inhalation) is proposed.

Furthermore, due to the general systemic effects observed for Nerolidol such as impaired body weight gains and liver effects mainly resulting from hepatic enzyme induction / metabolic activation, the default assessment factors for intraspecies variability for systemic toxicity after repeated dosing have been amended to less conservative values. These findings are not considered to represent severe specific organ toxicity. The derivation of these Nerolidol specific assessment factors are guided by the following information. In an attempt to evaluate the intraspecies variability within the human population, the distribution of human data for various toxicokinetic and toxicodynamic parameters were examined (Hattis et al 1987, 1999; Hattis and Silver 1994; Renwick and Lazarus, 1998; see ECETOC TR No.86, 2003). These evaluations included data from ‘healthy adults’ of both sexes, as well as limited data from the young and elderly, mixed races and patients with various medical conditions such as cancer and hypertension. The data of Renwick and Lazarus (1998) and Hattis et al. (1999) were based exclusively on human data and similar values were obtained within each percentile. Considering that the data analysed by these authors included both sexes, a variety of disease states and ages, the use of the 95th percentile is considered sufficiently conservative to account for intraspecies variability in the general population. Thus, a default assessment factor of 5 was taken for the general population with a lower factor of 3 (i. e. closer to the 90th percentile) for the more homogenous worker population.

Overall, it needs to be pointed out, that several factors of conservatism have been applied to the systemic DNEL derivation made. Next to the use of the 90-95th percentile concerning intraspecies variability, the multiplicatory principle of AF provide a sufficient degree of conservatism.

 

For the general population, the following DNELs were derived:

For derivation of the long-term systemic oral DNEL, the chosen NOAEL (100 mg/kg bw/d) was divided by the following assessment factors (AF):

 

·       allometric scaling = 4 (according toR8 ECHA 2008)

·       remaining differences = 1 (On the basis of the general adverse systemic effects observed, i.e. body weight changes and liver effects mainly resulting from hepatic enzyme induction / metabolic activation at high doses of Nerolidol, no difference in sensitivity (toxicodynamic and/or additional toxicokinetic differences) between test animals and humans is to be expected besides aspects already covered by allometric scaling).

·       intraspecies = 5 (based on the main substance related adverse effects observed after repeated oral dosing and general considerations)

·     exposure duration = 6 (subchronic to chronic);

·     quality of whole database = 1 (based on validity of studies performed).

 

AF = 4 x 1 x 5 x 6 x 1 = 120. Consequently, the oral long-term systemic DNEL derived was 0.8 mg/kg bw/d.

 

 

For derivation of the inhalative DNEL, the oral NOAEL was converted into a corrected inhalative NOAEC of 87 mg/m3 according to the procedure, recommended in the current guidance document (R8, ECHA 2008).

 

NOAECinhal corrected= 100*(1/1.15)*(100/100)= 87 mg/m3

 

The following assessment factors (AF) were applied:

·       allometric scaling = 1 (not applicable according toR8 ECHA 2008)

·       remaining differences = 1(On the basis of the general adverse systemic effects observed, i.e. body weight changes and liver effects mainly resulting from hepatic enzyme induction / metabolic activation at high doses of Nerolidol, no difference in sensitivity (toxicodynamic and/or additional toxicokinetic differences) between test animals and humans is to be expected besides aspects already covered in the route to route extrapolation performed above).

·       intraspecies = 5 (based on the main substance related adverse effects observed after repeated oral dosing and general considerations)

·       exposure duration = 6 (subacute to chronic);

·       quality of whole database = 1 (based on validity of studies performed).

 

 

AF = 1 x 1 x 5 x 6 x 1 = 30.Consequently, the inhalative long-term systemic DNEL was set at 2.9 mg/m3.

 

For derivation of the long-term systemic dermal DNEL, the oral NOAEL was converted into a corrected dermal NOAEL of 200 mg/kg bw/d according to the procedure, recommended in the current guidance document (R8, ECHA 2008).

 

NOAEL corrected dermal = 100*(100/50) = 200 mg/kg bw/d

 

The following assessment factors (AF) were applied:

·       allometric scaling = 4 (according toR8 ECHA 2008)

·       remaining differences = 1 (On the basis of the general adverse systemic effects observed, i.e. body weight changes and liver effects mainly resulting from hepatic enzyme induction / metabolic activation at high doses of Nerolidol, no difference in sensitivity (toxicodynamic and/or additional toxicokinetic differences) between test animals and humans is to be expected besides aspects already covered by allometric scaling).

·       intraspecies = 5 (based on the main substance related adverse effects observed after repeated oral dosing and general considerations)

·     exposure duration = 6 (subchronic to chronic);

·     quality of whole database = 1 (based on validity of studies performed).

 

AF = 4 x 1 x 5 x 6 x 1 = 120. Consequently, the dermal long-term systemic DNEL derived was 1.7 mg/kg bw/d.

 

No DNELs were derived for local effects after short term or after long term inhalative exposure and no DNELs were also derived for systemic effects after short term dermal, inhalative or oral exposure, as the substance exhibits no hazardous potential in terms of these endpoints. Hence at these points there is no need for classification and therefore, there is consequently no need for deriving additional DNELs for the general population.

 

No hazard concerning dermal irritation exist. To assess the DNEL for local effects after long term dermal exposure, data for skin sensitization were considered.

In the chosen key study, i.e. a murine LLNA according to OECD TG 429 and GLP, (BASF SE; 58V0466/09A141), an EC3 for 3H-thymidine incorporation was determined to be 4.9%. According to the ECHA guidance document R8, this EC3 is converted as follows:

EC3 [μg/cm2] = EC3 [%]*250 [μg/cm2/% ] = 4.9 * 250 = 1225 µg/cm2

Furthermore, human data for Nerolidol need to be taken into account, such as a human maximization test conducted on 25 healthy volunteers, showing no dermal effects at a concentration of 4 % (2760 µg/cm2) tetrahydrolinalool (Kligman 1973). Different patch tests in patients revealed sporadic postive skin reactions, however, these data do not provide a sound basis to derive a skin sensitization induction threshold level relevant for the DNEL derivation. 

The human data available for Nerolidol, i.e. HMT, indicate the absences of a skin sensitization potential at a concentration above the EC3 (µg/cm2) determined in the murine LLNA determined (2760 µg/cm2 versus 1225 µg/cm2, respectively). Since no evident species specific differences in potencies in terms of a higher human susceptibility are indicated between the murine test and human subjects, the use of an additional interspecies assessment factor is not plausible. Although the human data are considered reliable, the point of departure for the local long term dermal DNEL is based on the murine LLNA EC3 as a conservative approach, due to limitations in study design of the available HMT, such as the number of participants.

It is recognized that a general DNEL must take into account that the threshold for skin sensitization varies between individuals. This may be due to differences in parameters such as genetic effects, sensitive subpopulations, inherent barrier function, age, gender, and ethnicity (Api et al., 2008). Whereas the latter three are recognized to have some effect on the sensitization threshold, it is generally recognized that genetic differences, the inherent barrier function and especially sensitive subpopulations play a major role (Api et al., 2008). The barrier function of the skin may be compromised which in turn may lead to a greater susceptibility of the individual. At the same time the barrier function is thought to be very similar from infancy to adulthood. The influence of the genetic setting is not well understood, however, may be plausible in the light of the immunological effect under consideration. The term sensitive subpopulations refers mostly to individuals who have previously been sensitized to other substances which may increase the susceptibility to further sensitizers (Api et al., 2006, Api et al., 2008). Overall, an assessment factor of 10 for intraspecies differences is applied to adequately address the combined influence of these effects.

The relevant parameters, i. e. induction of skin sensitization, are considered to depend on threshold concentrations and not on exposure duration. Therefore no assessment factor concerning exposure duration is deemed necessary for the derivation of the long term local dermal DNEL. The concept of threshold concentrations for the induction of these effects is generally well accepted (see e.g. Api et al. 2006; or Api et al. 2008). Furthermore, the ECHA guidance document does not indicate at all, that an assessment factor for exposure duration needs to be taken into account for the derivation of a DNEL for skin sensitization.

Therefore, a DNEL for skin sensitization was set at 122.5 µg/cm2. Using this DNEL allows for a quantitative risk characterization for skin sensitization. This DNEL is considered sufficient to cover also local effects after short term dermal exposure.

Furthermore, Nerolidol is to be classified as eye irritant (category 2) according to 1272/2008/EEC. Since no quantitative data addressing the hazard of eye irritation are available, a respective no effect concentration cannot be derived and included in the derivation of the short term/long term local dermal DNEL. However, due to the low final concentrations of Nerolidol in consumer products, the risk for eye irritation is considered to be low.

 

·   Api AM, Basketter DA, Cadby PA, Cano M-F, Graham E, Gerberick F, Griem P, McNamee P, Ryan CA, Safford B (2006). Dermal Sensitization Quantitative Risk Assessment (QRA) for fragrance ingredients. Technical dossier. March 15, 2006 (revised May 2006).

·   Api AM, Basketter, DA, Cadby PA, Cano M-F, Ellis G, Gerberick GF, Griem P, McNamee PM, Ryan CA, Safford R (2008). Dermal sensitization quantitative risk assessment (QRA) for fragrance ingredients. Reg Toxicol Pharmacol 52: 3-23.

·   ECETOC (2003). Contact Sensitization: classification according to potency. Technical Report No. 87, April 2003.

·   Escher S, Batke M, Hoffmann-Doerr S, Messinger H, Mangelsdorf I (2013). Interspecies extrapolation based on the RepDose database—A probabilistic approach. Toxicology Letters 218: 159– 165