Registration Dossier

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

352 mg/m³

Most sensitive endpoint:

repeated dose toxicity

Overall assessment factor (AF):

3

Modified dose descriptor starting point:

NOAEC

Value:

1 050 mg/m³

AF for dose response relationship:

1

AF for differences in duration of exposure:

1

Justification:

chronic study

AF for interspecies differences (allometric scaling):

1

Justification:

not required for inhalation study

AF for other interspecies differences:

1

Justification:

see discussion

AF for intraspecies differences:

3

Justification:

see discussion

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

2 800 mg/m³

Most sensitive endpoint:

neurotoxicity

Overall assessment factor (AF):

3

Modified dose descriptor starting point:

NOAEC

Value:

8 400 mg/m³

AF for dose response relationship:

1

AF for interspecies differences (allometric scaling):

1

Justification:

not required for inhalation study

AF for other interspecies differences:

1

Justification:

see discussion

AF for intraspecies differences:

3

Justification:

see discussion

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

105 mg/m³

Most sensitive endpoint:

irritation (respiratory tract)

Overall assessment factor (AF):

1

Dose descriptor:

NOAEC

AF for dose response relationship:

1

AF for differences in duration of exposure:

1

AF for interspecies differences (allometric scaling):

1

AF for other interspecies differences:

1

AF for intraspecies differences:

1

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

6 767 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

Overall assessment factor (AF):

12

Modified dose descriptor starting point:

NOAEL

Value:

81 200 mg/kg bw/day

Explanation for the modification of the dose descriptor starting point:

From inhalation data. Assume rat respiratory volume 0.29m3/kgbw/6hr. Inhalation absorption efficiency 40%, dermal absorption 0.3%.

AF for dose response relationship:

1

AF for differences in duration of exposure:

1

Justification:

Chronic study used as a basis for DNEL derivation

AF for interspecies differences (allometric scaling):

4

Justification:

rat to human

AF for other interspecies differences:

1

Justification:

see discussion

AF for intraspecies differences:

3

Justification:

see discussion

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

no hazard identified

According to the REACH “Guidance on information requirements and chemical safety assessment”, a leading DN(M)EL needs to be derived for every relevant human population and every relevant route, duration and frequency of exposure, if feasible.

The DNELs are derived using the scientifically based assessment factors as reported by ECETOC (2010).

ECETOC (2010) Guidance on assessment factors to derive a DNEL. Technical report No. 110, ECETOC, Brussels, October 2010.

Kinetics

The inhalation and dermal absorption percentages used for DNEL derivation (in case of route-to-route extrapolation) are 40% and 0.3%, respectively.

Acute toxicity

ETBE is labelled for acute toxicity (R67 or Specific Target Organ toxicity – Single exposure, Cat. 3 (H336)) and therefore derivation of a DNELacute is necessary for situations where peak exposures may occur. The NOAEC for transient signs of CNS depression is 2000 ppm (8400 mg/m3). This NOAEC is derived from acute effects observed in a 28 day inhalation study conducted in rats.

ETBE is not irritating to the skin, eyes and respiratory tract and not sensitising to the skin. Therefore, no DNELs are derived for these endpoints.

Long-term toxicity

Based on the available 90-days studies, the overall NOAEC for rats and mice is 500 ppm based on changes in liver (female mice, 1750 and 5000 ppm (7350 and 21000 mg/m3)) over 90-days and reduced survival and serum enzyme alterations (rats, 1500 and 5000 ppm (6300 and 21000 mg/m3)) over 2-years.

No repeated dermal toxicity studies for ETBE are available, and appropriate DNELs will therefore be developed using route-to-route extrapolation from the available chronic inhalation toxicity data. This is considered acceptable given the negligible first-pass hepatic metabolism associated with both of these routes of exposure.

Regarding local effects, the human data on ETBE (Nihlen et al 1998b) indicate the occurrence of possible subjective symptoms of mild nasal discomfort with reports of unpleasant taste and odour experienced by volunteers exposed to 50 ppm ETBE. Based on these data a NOAEC of 25 ppm is established, however it is recognised that this is likely to be quite a conservative conclusion.

It can also be presumed that repeated skin exposure may result in skin fatigue (due to the effective lipid extraction properties of ETBE and consequent risk of toxic eczema), an effect common to a variety of organic solvents. No quantitative data on this effect are available. Therefore a qualitative risk characterisation will be performed.

ETBE is assessed as being non-mutagenic and not carcinogenic. Based on this, no separate risk characterisation for mutagenicity and carcinogenicity is needed.

ETBE did not show effects on fertility in an oral two-generation study with rats. No DNEL has to be derived for reproductive toxicity.

No developmental toxicity was observed in an oral prenatal developmental toxicity study in rats. No DNEL has to be derived for developmental toxicity.

There are no indications from the available data that dams are more sensitive regarding systemic effects compared to animals exposed in the repeated dose toxicity studies.

Acute – inhalation, systemic effects

The DNEL is based on findings of transient CNS depression recorded in a rat 28-day inhalation study, which returned a NOAEC of 2000 ppm (8400 mg/m3).

Description	Value	Remark
Step 1) Relevant dose-descriptor	NOAEC: 2000 ppm (8400 mg/m³)	Based on acute (transient) CNS effects in a 28-day inhalation study with rats: general sedation, reduced motor activity, mild to moderate ataxia.
Step 2) Modification of starting point		From the available data, it is clear that acute CNS effects occurred during the daily exposures, but it is not clear at which time point during exposure these effects occurred. In the REACH guidance (R.8, Appendix R. 8-8), it is mentioned: ‘If a DNEL for acute toxicity needs to be established, this should be derived only for a specified fraction of the daily exposure duration (usually 15 minutes) ’. The acute transient CNS effects were completely disappeared within 15 minutes after exposure for 6 hours to 4000 ppm. The half-life in the blood of rats has been estimated to be 0.4 hours (DeKant et al, 2001) which means that ETBE is cleared rapidly from the body. The acute CNS effects are assumed to be caused by filling of the brain compartment. As the acute DNEL is derived for only 15 minutes, filling of the brain compartment is considered less compared to 6 hours of exposure resulting a higher starting point. The REACH guidance prescribes a factor for activity driven differences of respiratory volumes in workers compared to workers in rest (6.7 m³/10 m³). The application of this factor results in a lower starting point. Based on a qualitative assessment of the above two subjects, it is considered acceptable that no modification of the starting point will be applied.
Step 3) Assessment factors
Interspecies	1	No allometric scaling is needed in case of inhalation exposure because extrapolation is based on toxicological equivalence of a concentration of a chemical in the air of experimental animals and humans; animal and humans breathe at a rate depending on their caloric requirements (ECETOC, 2010a). No additional assessment factor is required for “remaining differences” since inter- and intraspecies variability are not independent variables, with any remaining uncertainty accounted for already through the use of allometric scaling (where appropriate) and in the intraspecies assessment factor (ECETOC, 2010a).The systemic toxic effects have been thoroughly studied for ETBE and there are no indications that humans in comparison to animal test species would significantly differ in their responses, hence the additional adjustment factor of 2.5 is not warranted.
Intraspecies	3	An informed assessment factor of 3 is proposed (ECETOC, 2010a) based on an analysis of the available scientific literature rather than the adoption of standard defaults. Intraspecies sensitivity extrapolation from rat to man is done by extrapolation from the already interspecies extrapolated rodent EC05 (=NOAEL) to an assumed human EC05 (=NOAEL). Assuming the ratio between the EC95 and EC5 in a rat is 10, observed in many studies, and applying the ECHA intraspecies factor of 5 on the interspecies extrapolated rat EC05 in order to achieve the worker EC05 means that the ratio between the EC95 and EC05 in workers is assumed to be 250. A factor of 250 for the ratio between the worker EC95 and EC05 is far outside general clinical evidence. Therefore the ECETOC intraspecies assessment factors should be preferred.
Exposure duration	not applicable
Dose response	1
Quality of database	1
DNEL	Value
	8400 / (1 x 3 x 1 x 1) =2800 mg/m³

Long-term – inhalation, systemic effects

The DNEL is based upon a NOAEC of 500 ppm (2100 mg/m3) reflecting findings of reduced survival, slight haematological changes, altered clinical chemistry parameters and liver effects after chronic exposure to higher exposure concentrations.

Description	Value	Remark
Step 1) Relevant dose-descriptor	NOAEC: 500 ppm (2100 mg/m³)	Based onreduced survival, slight haematological changes, altered clinical chemistry parameters and liver effectsafter chronic exposure to higher exposure concentrations.
Step 2) Modification of starting point	6/8 6.7 m³/10 m³	- Correction of exposure duration in study (6 hrs/day, 5 days/week) to default worker exposure (8 hrs/day, 5 days/week); - Correction for activity driven differences of respiratory volumes in workers compared to workers in rest (6.7 m³/10 m³).
Step 3) Assessment factors
Interspecies	1	No allometric scaling is needed in case of inhalation exposure because extrapolation is based on toxicological equivalence of a concentration of a chemical in the air of experimental animals and humans; animal and humans breathe at a rate depending on their caloric requirements (ECETOC, 2010a). No additional assessment factor is required for “remaining differences” since inter- and intraspecies variability are not independent variables, with any remaining uncertainty accounted for already through the use of allometric scaling (where appropriate) and in the intraspecies assessment factor (ECETOC, 2010a).The systemic toxic effects have been thoroughly studied for ETBE and there are no indications that humans in comparison to animal test species would significantly differ in their responses, hence the additional adjustment factor of 2.5 is not warranted.
Intraspecies	3	An informed assessment factor of 3 is proposed (ECETOC, 2010a) based on an analysis of the available scientific literature rather than the adoption of standard defaults. Intraspecies sensitivity extrapolation from rat to man is done by extrapolation from the already interspecies extrapolated rodent EC05 (=NOAEL) to an assumed human EC05 (=NOAEL). Assuming the ratio between the EC95 and EC5 in a rat is 10, observed in many studies, and applying the ECHA intraspecies factor of 5 on the interspecies extrapolated rat EC05 in order to achieve the worker EC05 means that the ratio between the EC95 and EC05 in workers is assumed to be 250. A factor of 250 for the ratio between the worker EC95 and EC05 is far outside general clinical evidence. Therefore the ECETOC intraspecies assessment factors should be preferred.
Exposure duration	1	A factor of 1 is applied since the results were obtained from a chronic study.
Dose response	1
Quality of database	1
DNEL	Value
	2100 x 6/8 x 6.7/10 / (1 x 3 x 1 x 1 x 1) =352 mg/m³

Long-term – inhalation, local effects

Eight male human subjects were exposed to 0, 5, 25, 50 ppm concentrations of ETBE for 2 hours (Nihlen et al 1998b). There were a total of 4 exposures to each subject to the 4 exposure concentrations (exposed once to each exposure concentration). Subjective discomfort of the throat and airways was increased significantly after exposure to 50 ppm ETBE, and remained elevated for the remainder of the 2 hr exposure period. Symptoms had resolved 80 min after exposure ceased. The taste and odour of the test atmosphere was unpleasant. The authors state that subjective symptoms of ocular and nasal discomfort, fatigue, nausea dizziness and intoxication were greater in the 50 ppm group, but no dose-response relationship or statistically significant difference was present. The intensity of the response at 50 ppm appeared to be very mild (stated to be between "not at all" and "hardly at all").

There was no statistically significant effect on eye redness, tear-film break-up, conjunctival damage or blinking frequency. Nasal volume was decreased by 6-15% (P = 0.001) in all groups, including controls, compared to pre-exposure values. Exposure to ETBE vapour was without effect on the composition of nasal lavage fluid (i. e. leukocytes, epithelial cells, inflammatory markers etc).

A small (3-4%) but significant (P<0.05) decrease in forced vital capacity (FVC) and vital capacity (VC) in subjects exposed to 25 or 50 ppm ETBE vapour, however short-term changes in lung function of <5% are likely to reflect normal variation and are therefore of no clinical significance. The authors also state that one-second forced expiratory volume (FEV1) and the transfer factor (TLco) were also decreased (approx. 2% and 2 -4%, respectively) after exposure, but this finding was not statistically significant. The slightly decreased lung function changes were mild in nature and minor in extent, and likely to reflect normal variation rather than an effect of ETBE vapour on lung function.

Subjective symptoms of mild nasal discomfort and reports of unpleasant taste and odour experienced by volunteers exposed to 50 ppm ETBE will be taken into account for risk characterisation of long-term local effects after inhalation, leading to a conservative NOAEC and DNEL of 25 ppm (105 mg/m3). This value will be used directly with no modification, and is lower than the systemic DNEL of 84 ppm (352 mg/m3) based on animal inhalation data (discussed above).

Long-term – dermal, systemic effects (based on the repeated dose inhalation NOAEC)

As dermal repeated dose toxicity studies with ETBE are not available, route to route extrapolation will be applied to derive a DNEL for the dermal route, based on the NOAEC of 500 ppm (2100 mg/m3) from the available inhalation repeated dose toxicity studies. This point of departure is used in preference to the oral NOAEL since ingestion leads to first-pass metabolism in the liver whereas more widespread systemic exposure and no first-pass hepatic metabolism occurs after inhalation and/or dermal exposure.

Description	Value	Remark
Step 1) Relevant dose-descriptor	NOAEC: 500 ppm (2100 mg/m³)	Based onreduced survival, slight haematological changes, altered clinical chemistry parameters and liver effectsafter chronic exposure to higher exposure concentrations.
Step 2) Modification of starting point	0.29 40 / 0.3	Conversion into dermal NAEL (in mg/kg bw/day) by using a 6 h respiratory volume of 0.29 m³/kg bw for the rat. Correction for absorption: 40% inhalation absorption and 0.3% dermal absorption.
Step 3) Assessment factors
Interspecies	4	Assessment factor for allometric scaling from rats to humans as proposed by ECETOC (2010a). No additional assessment factor is required for “remaining differences” since inter- and intraspecies variability are not independent variables, with any remaining uncertainty accounted for already through the use of allometric scaling (where appropriate) and in the intraspecies assessment factor (ECETOC, 2010a). The systemic toxic effects have been thoroughly studied for ETBE and there are no indications that humans in comparison to animal test species would significantly differ in their responses, hence the additional adjustment factor of 2.5 is not warranted.
Intraspecies	3	An informed assessment factor of 3 is proposed (ECETOC, 2010a) based on an analysis of the available scientific literature rather than the adoption of standard defaults. Intraspecies sensitivity extrapolation from rat to man is done by extrapolation from the already interspecies extrapolated rodent EC05 (=NOAEL) to an assumed human EC05 (=NOAEL). Assuming the ratio between the EC95 and EC5 in a rat is 10, observed in many studies, and applying the ECHA intraspecies factor of 5 on the interspecies extrapolated rat EC05 in order to achieve the worker EC05 means that the ratio between the EC95 and EC05 in workers is assumed to be 250. A factor of 250 for the ratio between the worker EC95 and EC05 is far outside general clinical evidence. Therefore the ECETOC intraspecies assessment factors should be preferred.
Exposure duration	1	A factor of 1 is applied since the results were obtained from a chronic study.
Dose response	1
Quality of database	1
DNEL	Value
	2100 x 0.29 x 40/0.3 / (4 x 3 x 1 x 1 x 1) =6767 mg/kg bw/day

Explanation of ECETOC Assessment Factors

The ECHA default interspecies assessment factors are useful for an isolated study for only one toxicological endpoint. In all other cases the ECETOC factors should be preferred for the following reasons: In case of existing high production volume chemicals there is at least one study for each toxicological endpoint and many times more studies for the same toxicological endpoint. So the confidence in the real No Observed Adverse Effect Level (NOAEL) is much higher than in case of an isolated study for only one toxicological endpoint. Also developmental and reproductive toxicity studies and studies focused on neurotoxicity contribute to the evidence of the true NOAEL.

ECETOC interspecies extrapolation to human is based on allometric scaling. Comparison of tolerable repeated dose levels in rodents and humans for many substances has revealed, that allometric scaling provided the appropriate adjustment of rodent tolerable repeated dose levels to human repeated dose levels.

Intraspecies sensitivity extrapolation from rat to man is done by extrapolation from the already interspecies extrapolated rodent EC05 (= EC05 interspecies adjusted rat to human) to an assumed human EC05 (= NOAEL human). It is assumed, that the rodent EC05 is a NOAEL and is conservatively 3.16 times smaller than the rodent EC50 and 10 times smaller than the rodent EC95. So the ratio between the rat EC95 and EC05 is assumed to be a factor of 10. This is in agreement with the applied dose range in a rodent study being at least a factor of 10 (see the long term NTP rodent studies).

* The ECETOC intraspecies factor of 3, applied to the interspecies extrapolated rodent EC05 (= EC05 interspecies adjusted rat to human) means that the ratio between worker EC95 and EC05 is assumed to be 90.

* The ECETOC intraspecies factor of 5, applied to the interspecies extrapolated rodent EC05 (= EC05 interspecies adjusted rat to human) means that the ratio between consumer EC95 and EC05 is assumed to be 250.

* The ECHA intraspecies factor of 5, applied to the interspecies extrapolated rodent EC05 (= EC05 interspecies adjusted rat to human) means that the ratio between worker EC95 and EC05 is assumed to be 250.

* The ECHA intraspecies factor of 10, applied to the interspecies extrapolated rodent EC05 (= EC05 interspecies adjusted rat to human) means that the ratio between consumer EC95 and EC05 is assumed to be 1000.

In the general population the ratio between the EC95 and EC05 is about a factor of 25. That is larger than that in the rat. The intraspecies assessment factors of ECETOC account for this factor of 25 and, in addition, for statistical uncertainty in the comparison between human and animal studies on repeated tolerable dose levels. Therefore, the ECETOC intraspecies assessment factors are conservative for deriving a DNEL for workers and for consumers (=general population).

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

105 mg/m³

Most sensitive endpoint:

repeated dose toxicity

Overall assessment factor (AF):

5

Modified dose descriptor starting point:

NOAEC

Value:

525 mg/m³

AF for dose response relationship:

1

AF for differences in duration of exposure:

1

Justification:

chronic study

AF for interspecies differences (allometric scaling):

1

Justification:

not required for inhalation study

AF for other interspecies differences:

1

Justification:

see discussion

AF for intraspecies differences:

5

Justification:

see discussion

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

1 680 mg/m³

Most sensitive endpoint:

neurotoxicity

Overall assessment factor (AF):

5

Modified dose descriptor starting point:

NOAEC

Value:

8 400 mg/m³

AF for dose response relationship:

1

AF for interspecies differences (allometric scaling):

1

Justification:

not required for inhalation study

AF for other interspecies differences:

1

AF for intraspecies differences:

5

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

63 mg/m³

Most sensitive endpoint:

irritation (respiratory tract)

Overall assessment factor (AF):

1.7

Dose descriptor:

NOAEC

AF for dose response relationship:

1

AF for differences in duration of exposure:

1

AF for interspecies differences (allometric scaling):

1

AF for other interspecies differences:

1

AF for intraspecies differences:

1.7

Justification:

see discussion

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

4 060 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

Overall assessment factor (AF):

20

Modified dose descriptor starting point:

NOAEL

Value:

81 200 mg/kg bw/day

Explanation for the modification of the dose descriptor starting point:

From inhalation data. Assume rat respiratory volume 0.29m3/kgbw/6hr. Inhalation absorption efficiency 40%, dermal absorption 0.3%

AF for dose response relationship:

1

AF for differences in duration of exposure:

1

Justification:

chronic study

AF for interspecies differences (allometric scaling):

4

Justification:

rat to human

AF for other interspecies differences:

1

Justification:

see discussion

AF for intraspecies differences:

5

Justification:

see discussion

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

DNEL (Derived No Effect Level)

Value:

6 mg/kg bw/day

Most sensitive endpoint:

repeated dose toxicity

Overall assessment factor (AF):

20

Modified dose descriptor starting point:

NOAEL

Value:

121 mg/kg bw/day

AF for dose response relationship:

1

AF for differences in duration of exposure:

1

Justification:

chronic study

AF for interspecies differences (allometric scaling):

4

Justification:

rat to human

AF for other interspecies differences:

1

Justification:

see discussion

AF for intraspecies differences:

5

Justification:

see discussion

AF for the quality of the whole database:

1

AF for remaining uncertainties:

1

Hazard assessment conclusion:

no hazard identified

Hazard assessment conclusion:

no hazard identified

According to the REACH “Guidance on information requirements and chemical safety assessment”, a leading DN(M)EL needs to be derived for every relevant human population and every relevant route, duration and frequency of exposure, if feasible.

The DNELs are derived using the scientifically based assessment factors as reported by ECETOC (2010).

ECETOC (2010) Guidance on assessment factors to derive a DNEL. Technical report No. 110, ECETOC, Brussels, October 2010.

Kinetics

The inhalation and dermal absorption percentages used for DNEL derivation (in case of route-to-route extrapolation) are 40% and 0.3%, respectively.

Acute toxicity

ETBE is labelled for acute toxicity (R67 or Specific Target Organ toxicity – Single exposure, Cat. 3 (H336)) and therefore derivation of a DNELacute is necessary for situations where peak exposures may occur. The NOAEC for transient signs of CNS depression is 2000 ppm (8400 mg/m3). This NOAEC is derived from acute effects observed in a 28 day inhalation study conducted in rats.

ETBE is not irritating to the skin, eyes and respiratory tract and not sensitising to the skin. Therefore, no DNELs are derived for these endpoints.

Long-term toxicity

Based on the available 90-days studies, the overall NOAEC for rats and mice is 500 ppm based on changes in liver (female mice, 1750 and 5000 ppm (7350 and 21000 mg/m3)) over 90-days and reduced survival and serum enzyme alterations (rats, 1500 and 5000 ppm (6300 and 21000 mg/m3)) over 2-years.

No repeated dermal toxicity studies for ETBE are available, and appropriate DNELs will therefore be developed using route-to-route extrapolation from the available chronic inhalation toxicity data. This is considered acceptable given the negligible first-pass hepatic metabolism associated with both of these routes of exposure.

Regarding oral exposure, the most critical NOAEL, 121 mg/kg bw/day, is derived from a rat 2-year drinking water study where changes in clinical chemistry parameters were observed at received doses of 540-560 mg/kg bw/day.

Regarding local effects, the human data on ETBE (Nihlen et al 1998b) indicate the occurrence of possible subjective symptoms of mild nasal discomfort with reports of unpleasant taste and odour experienced by volunteers exposed to 50 ppm ETBE. Based on these data a NOAEC of 25 ppm is established, however it is recognised that this is likely to be quite a conservative conclusion.

It can also be presumed that repeated skin exposure may result in skin fatigue (due to the effective lipid extraction properties of ETBE and consequent risk of toxic eczema), an effect common to a variety of organic solvents. No quantitative data on this effect are available. Therefore a qualitative risk characterisation will be performed.

ETBE is assessed as being non-mutagenic and not carcinogenic. Based on this, no separate risk characterisation for mutagenicity and carcinogenicity is needed.

ETBE did not show effects on fertility in an oral two-generation study with rats. No DNEL has to be derived for reproductive toxicity.

No developmental toxicity was observed in an oral prenatal developmental toxicity study in rats. No DNEL has to be derived for developmental toxicity.

There are no indications from the available data that dams are more sensitive regarding systemic effects compared to animals exposed in the repeated dose toxicity studies.

In the available oral gavage studies, ptyalism (excessive salivation) was observed at all dose levels (i. e., from 250 mg/kg bw/day onwards). This is considered related to unpalatable nature of the dosing solutions / to the unpalatable 'taste' of ETBE. In the draft EU RAR this effect is not taken into account as an adverse effect, therefore, this effect is not considered a significant health effect.

Acute – inhalation, systemic effects

The DNEL is based on findings of transient CNS depression recorded in a rat 28-day inhalation study, which returned a NOAEC of 2000 ppm (8400 mg/m3).

Description	Value	Remark
Step 1) Relevant dose-descriptor	NOAEC: 2000 ppm (8400 mg/m³)	Based on acute (transient) CNS effects in a 28-day inhalation study with rats: general sedation, reduced motor activity, mild to moderate ataxia.
Step 2) Modification of starting point	-	From the available data, it is clear that acute CNS effects occurred during the daily exposures, but it is not clear at which time point during exposure these effects occurred. In the REACH guidance (R.8, Appendix R. 8-8), it is mentioned: ‘If a DNEL for acute toxicity needs to be established, this should be derived only for a specified fraction of the daily exposure duration (usually 15 minutes) ’. The acute transient CNS effects were completely disappeared within 15 minutes after exposure for 6 hours to 4000 ppm. The half-life in the blood of rats has been estimated to be 0.4 hours (DeKant et al, 2001) which means that ETBE is cleared rapidly from the body. The acute CNS effects are assumed to be caused by filling of the brain compartment. As the acute DNEL is derived for only 15 minutes, filling of the brain compartment is considered less compared to 6 hours of exposure resulting a higher starting point. Therefore, no modification of the starting point will be applied.
Step 3) Assessment factors
Interspecies	1	No allometric scaling is needed in case of inhalation exposure because extrapolation is based on toxicological equivalence of a concentration of a chemical in the air of experimental animals and humans; animal and humans breathe at a rate depending on their caloric requirements (ECETOC, 2010a). No additional assessment factor is required for “remaining differences” since inter- and intraspecies variability are not independent variables, with any remaining uncertainty accounted for already in the intraspecies assessment factor (ECETOC, 2010a). The systemic toxic effects have been thoroughly studied for ETBE and there are no indications that humans in comparison to animal test species would significantly differ in their responses, hence the additional adjustment factor of 2.5 is not warranted.
Intraspecies	5	An informed assessment factor of 5 is proposed (ECETOC, 2010a) based on an analysis of the available scientific literature rather than the adoption of standard defaults. This is considered adequate to protect the general population as well as the elderly and young. Intraspecies sensitivity extrapolation from rat to man is done by extrapolation from the already interspecies extrapolated rodent EC05 (=NOAEL) to an assumed human EC05 (=NOAEL). Assuming the ratio between the EC95 and EC5 in a rat is 10, observed in many studies, and applying the ECHA intraspecies factor of 10 on the interspecies extrapolated rat EC05 in order to achieve the consumer EC05 means that the ratio between the EC95 and EC05 in consumers is assumed to be 1000. A factor of 1000 for the ratio between the consumer EC95 and EC05 is far outside general clinical evidence. Therefore the ECETOC intraspecies assessment factors should be preferred.
Exposure duration	not applicable
Dose response	1
Quality of database	1
DNEL	Value
	8400 / (1 x 5 x 1 x 1) =1680 mg/m³

Long-term – inhalation, systemic effects

The DNEL is based upon a NOAEC of 500 ppm (2100 mg/m3 ) reflecting findings of reduced survival, slight haematological changes, altered clinical chemistry parameters and liver effects after chronic exposure to higher exposure concentrations.

Description	Value	Remark
Step 1) Relevant dose-descriptor	NOAEC: 500 ppm (2100 mg/m³)	Based onreduced survival, slight haematological changes, altered clinical chemistry parameters and liver effectsafter chronic exposure.
Step 2) Modification of starting point	6/24	Correction of exposure duration in study (6 hrs/day) to default general population exposure (24 hrs/day) [1].
Step 3) Assessment factors
Interspecies	1	No allometric scaling is needed in case of inhalation exposure because extrapolation is based on toxicological equivalence of a concentration of a chemical in the air of experimental animals and humans; animal and humans breathe at a rate depending on their caloric requirements (ECETOC, 2010). No additional assessment factor is required for “remaining differences” since inter- and intraspecies variability are not independent variables, with any remaining uncertainty accounted for in the intraspecies assessment factor (ECETOC, 2010a). The systemic toxic effects have been thoroughly studied for ETBE and there are no indications that humans in comparison to animal test species would significantly differ in their responses, hence the additional adjustment factor of 2.5 is not warranted.
Intraspecies	5	An informed assessment factor of 5 is proposed (ECETOC, 2010a) based on an analysis of the available scientific literature rather than the adoption of standard defaults. This is considered adequate to protect the general population as well as the elderly and young. Intraspecies sensitivity extrapolation from rat to man is done by extrapolation from the already interspecies extrapolated rodent EC05 (=NOAEL) to an assumed human EC05 (=NOAEL). Assuming the ratio between the EC95 and EC5 in a rat is 10, observed in many studies, and applying the ECHA intraspecies factor of 10 on the interspecies extrapolated rat EC05 in order to achieve the consumer EC05 means that the ratio between the EC95 and EC05 in consumers is assumed to be 1000. A factor of 1000 for the ratio between the consumer EC95 and EC05 is far outside general clinical evidence. Therefore the ECETOC intraspecies assessment factors should be preferred.
Exposure duration	1	A factor of 1 is applied since the results were obtained from a chronic study.
Dose response	1
Quality of database	1
DNEL	Value
	2100 x 6/24 / (1 x 5 x 1 x 1 x 1) =105 mg/m³

Long-term – inhalation, local effects

Based on the results of Nihlen et al (1998b), a conservative NOAEC of 25 ppm (105 mg/m3) is established. This level is considered a DNEL for local effects for workers. The taste and odour of ETBE and slight irritation of the respiratory tract are more critical for the worker DNEL than prevention against systemic effects in humans. Application of an assessment factor for intraspecies differences (5/3, i. e., intraspecies factor for the general population divided by the one for workers) results in a DNEL for the general population of 25 ppm (105 mg/m3) / (5/3) = 15 ppm (63 mg/m3).

Long-term – dermal, systemic effects (based on the repeated dose inhalation NOAEC)

As dermal repeated dose toxicity studies with ETBE are not available, route to route extrapolation will be applied to derive a DNEL for the dermal route, based on the NOAEC of 500 ppm (2100 mg/m3) from the available inhalation repeated dose toxicity studies. This point of departure is used in preference to the oral NOAEL since ingestion leads to first-pass metabolism in the liver whereas more widespread systemic exposure occurs after inhalation and dermal exposure.

Description	Value	Remark
Step 1) Relevant dose-descriptor	NOAEC: 500 ppm (2100 mg/m³)	Based on reduced survival, slight haematological changes, altered clinical chemistry parameters and liver effectsafter chronic exposure.
Step 2) Modification of starting point	0.29 40 / 0.3	Conversion into dermal NAEL (in mg/kg bw/day) by using a 6 h respiratory volume of 0.29 m³/kg bw for the rat. Correction for absorption: 40% inhalation absorption and 0.3% dermal absorption.
Step 3) Assessment factors
Interspecies	4	Assessment factor for allometric scaling from rats to humans as proposed by ECETOC (2010a). No additional assessment factor is required for “remaining differences” since inter- and intraspecies variability are not independent variables, with any remaining uncertainty accounted for already through the use of allometric scaling (where appropriate) and in the intraspecies assessment factor (ECETOC, 2010a).The systemic toxic effects have been thoroughly studied for ETBE and there are no indications that humans in comparison to animal test species would significantly differ in their responses, hence the additional adjustment factor of 2.5 is not warranted.
Intraspecies	5	An informed assessment factor of 5 is proposed (ECETOC, 2010a) based on an analysis of the available scientific literature rather than the adoption of standard defaults. This is considered adequate to protect the general population as well as the elderly and young. Intraspecies sensitivity extrapolation from rat to man is done by extrapolation from the already interspecies extrapolated rodent EC05 (=NOAEL) to an assumed human EC05 (=NOAEL). Assuming the ratio between the EC95 and EC5 in a rat is 10, observed in many studies, and applying the ECHA intraspecies factor of 10 on the interspecies extrapolated rat EC05 in order to achieve the consumer EC05 means that the ratio between the EC95 and EC05 in consumers is assumed to be 1000. A factor of 1000 for the ratio between the consumer EC95 and EC05 is far outside general clinical evidence. Therefore the ECETOC intraspecies assessment factors should be preferred.
Exposure duration	1	A factor of 1 is applied since the results were obtained from a chronic study.
Dose response	1
Quality of database	1
DNEL	Value
	2100 x 0.29 x 40/0.3 / (4 x 5 x 1 x 1 x 1) =4060 mg/kg bw/day

Long-term – oral, systemic effects

The DNEL is based on a NOAEL of 121 mg/kg bw/day that was obtained from a 2-year drinking water study where alterations in clinical chemistry parameters were noted in male and female rats. This NOAEL is quite conservative given the occurrence of reduced water consumption (and probable dehydration) in the higher treatment groups, an observation considered related to the unpleasant taste of ETBE.

Description	Value	Remark
Step 1) Relevant dose-descriptor	NOAEL: 121 mg/kg bw/day	Based on alterations in clinical chemistry parameters in male and female rats. This NOAEL is quite conservative given the occurrence of reduced water consumption (and probable dehydration) noted in the higher treatment groups, an observation considered related to the unpleasant taste of ETBE.
Step 2) Modification of starting point	Not applicable
Step 3) Assessment factors
Interspecies	4	Assessment factor for allometric scaling from rats to humans as proposed by ECETOC (2010a). No additional assessment factor is required for “remaining differences” since inter- and intraspecies variability are not independent variables, with any remaining uncertainty accounted for already through the use of allometric scaling (where appropriate) and in the intraspecies assessment factor (ECETOC, 2010a). The systemic toxic effects have been thoroughly studied for ETBE and there are no indications that humans in comparison to animal test species would significantly differ in their responses, hence the additional adjustment factor of 2.5 is not warranted.
Intraspecies	5	An informed assessment factor of 5 is proposed (ECETOC, 2010a) based on an analysis of the available scientific literature rather than the adoption of standard defaults. This is considered adequate to protect the general population as well as the elderly and young. Intraspecies sensitivity extrapolation from rat to man is done by extrapolation from the already interspecies extrapolated rodent EC05 (=NOAEL) to an assumed human EC05 (=NOAEL). Assuming the ratio between the EC95 and EC5 in a rat is 10, observed in many studies, and applying the ECHA intraspecies factor of 10 on the interspecies extrapolated rat EC05 in order to achieve the consumer EC05 means that the ratio between the EC95 and EC05 in consumers is assumed to be 1000. A factor of 1000 for the ratio between the consumer EC95 and EC05 is far outside general clinical evidence. Therefore the ECETOC intraspecies assessment factors should be preferred.
Exposure duration	1	A factor of 1 is applied since the results were obtained from a chronic drinking water study. This duration and route of exposure is highly appropriate for humans exposed via drinking water.
Dose response	1
Quality of database	1
DNEL	Value
	121 / (4 x 5 x 1 x 1 x 1) =6.0 mg/kg bw/day