Registration Dossier

Administrative data

Endpoint:
eye irritation
Remarks:
other: QSAR
Type of information:
(Q)SAR
Adequacy of study:
key study
Study period:
1 September 2009
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: see 'Remark'
Remarks:
The QSAR prediction is fully adequate. A QSAR is allowed methodology and the uncertainty of the prediction is within the OECD test method of eye irritation, and the potential mechanisms of eye irritation are covered too. The prediction and validity of the model fulfil the 5 OECD principles of the validation of (Q)SARs (REACH Guidance on Information Requirements and Chemical Safety Assessment Chapter R6.1 Guidance on (Q)SARs). A Klimisch rating of 2 was assigned based on Practical guidance #5 on the reporting of (Q)SARs in IUCLID (ref: ECHA-10-B-10-EN).
Justification for type of information:
The adequacy of the information is presented in the Endpoint summary

Data source

Reference
Reference Type:
publication
Title:
The use of discriminant analysis, logistic regression and classification tree analysis in the development of classification models for human health effects
Author:
Worth AP, Cronin MTD
Year:
2009
Bibliographic source:
Journal of Molecular Structure (TheoChem) 622 (2003) 97-111

Materials and methods

Test guideline
Guideline:
other: Guidance on information requirements and chemical safety assessment, Chapter R.6: QSARs and grouping of chemicals, May 2008.
Principles of method if other than guideline:
General model for the presence or absence of eye irritation according to EU C&L for organic liquids.
GLP compliance:
no
Remarks:
N/A

Test material

Reference
Name:
Unnamed
Type:
Constituent
Type:
Constituent
Type:
Constituent
Test material form:
liquid

Test system

Controls:
no
Number of animals or in vitro replicates:
Not relevant
Details on study design:
The presence or absence of eye irritation following EU C&L rules was predicted based on molecular weight according to the method of Worth and Cronin (2003) and on the basis of structural analogue substances. The experimental data of substances in the training set used usually originates from tests carried out according to OECD guideline 405.

TYPE OF MODEL
- Structure Activity Relationship (SAR) using Classification tree analysis
- Descriptor values: molecular weight (calculated with ACDChemsketch) and molecular structure

Results and discussion

In vivo

Results
Irritation parameter:
other: N/A
Time point:
other: 0
Score:
0
Irritant / corrosive response data:
The substance has a MW of 234 g/mol, which is higher than the models' threshold value (137 g/mol) for prediction of eye irritation indicating that the substance is not an eye irritant. Additionally, two structural analogues (2,6-dimethyl-4-heptanone, 6,10,14-trimethyl-2-pentadecanone) of OTNE are classified as not irritating to the eyes.

Any other information on results incl. tables

Test material is within the applicability domain of the model (see details on test material):

- Model is about organic liquids

- Training set has MW range of 30-268.5 g/mol

- Training set of the model contains 5/119 substances with ketons

- Model hypothesis: smaller molecules are more likely to penetrate into eye and cause irritation than larger chemicals

- Model does not discuss the metabolic domain (parent compound is expected to cause the effect)

Considerations on structural analogues (2,6-dimethyl-4-heptanone and 6,10,14-trimethyl-2-pentadecanone):

The presented analogues are aliphatic organics containing C, H, O atoms, and the difference with the predicted chemical OTNE is that the chemical cyclic features and has one unsaturated bond. This dissimilarity is not expected to cause other properties, because the longer aliphatic chain is expected to bend and may have similar cyclic features as OTNE. The bioavailability of the chemicals is expected to be the highest for the smaller molecule with the lowest log Kow. The bioavailability of OTNE and the other analogue are suspected to be similar. The cyclic structure may be somewhat less bioavailable than the aliphatic chains. At the other hand the log Kow of the aliphatic chain is somewhat higher. The keton group of OTNE is expected to be somewhat less electrophilic than the two analogues presented due to some hindrance of a methyl group.

The uncertainty of the prediction:

The uncertainty of the prediction is within the uncertainty of the model: 98% correct predictions for negatives. It is unlikely that this prediction is a false negative because of additional applicability domain reasoning considering physical state, bioavailability, structural alert and reactivity considerations are strengthening the prediction besides OTNE being within the applicability domain of the model. Also the false negative chemical, polyethylene glycol butyl ether, is not similar to OTNE.

The chemical and biological mechanisms according to the model underpinning the predicted result:

The model is based on the hypothesis that smaller chemicals are more likely to penetrate into the eye and cause irritation than larger chemicals. Consistent with this hypothesis was the finding of Rosenkranz that the mean MW of irritants is significant lower than the mean MW of non-irritants (Worth and Cronin, 2003).

Applicant's summary and conclusion

Interpretation of results:
other: Not an eye irritant
Remarks:
according to EU CLP (EC 1272/2008 and its amendments)
Conclusions:
Based on the predicted values in the model, the substance OTNE does not need to be classified as irritant (molecular weight higher than threshold value and non-irritating property of structural analogues).
Executive summary:

The model of AP Worth and MTD Cronin as published in Journal of Molecular Structure (TheoChem) 622 (2003) 97-111 ("The use of discriminant analysis, logistic regression and classification tree analysis in the development of classification models for human health effects") was used to predict the irritating potential of OTNE based on its molecular weight and structural analogues.

The substance has a MW of 234 g/mol, which is higher than the model's threshold value (137 g/mol) for prediction of eye irritation. Additionally, two structural analogues (2,6-dimethyl-4-heptanone, 6,10,14-trimethyl-2-pentadecanone) of OTNE are classified as not irritating to the eyes.

Based on the predicted values in the model, the substance OTNE does not need to be classified as irritant (molecular weight higher than threshold value and non-irritating property of structural analogues)

Tabulated summary on the QSAR prediction, the (Q)SAR methodology and the relevance for REACH:

 

Reliability of prediction

Scientific validity of

 method

Relevance for REACH C&L

Defined Endpoint

QSAR based on OECD TG 405 information and DSD criteria

QSAR based on OECD TG 405 information and DSD criteria

REACH requires information similar to OECD TG 405

The algorithm

Prediction according to algorithm: OTNE has MW of 234, which is much > 137 and therefore not an eye irritant

Liquid organic substances with MW >137 are not eye irritants

(Q)SARs are allowed methodology when used according to Annex XI

Applicability domain

OTNE is within the domain, considering i) being a liquid; ii) MW range used in the model and; iii) other cyclic ketons are present in the training set

See substances in training set, cyclic ketons are included

Eye irritation hazards for OTNE has to be assessed under REACH

Uncertainty

Limited, because OTNE does not have additional groups which may lead to higher eye irritation e.g. higher electrophilicity

Limited, see paper of Worth and Cronin, 2000

The uncertainty of the result is limited, because OTNE is within the domain

Mode of action explanation

OTNE has insufficient reactive properties to cause eye irritation or corrosion

Not explained, but liquids with MW > 137 are expected to limitedly penetrate the eye barriers

The QSAR result of OTNE covers the mode of actions in OECD TG 405 because the training set is based on this type of tests

KC

1

Sufficiently valid

Sufficiently relevant

Translation into

Not an eye irritant according to DSD

Uses DSD criteria

Result is based on DSD criteria

Actual use into

Not an eye irritant

Adequacy

Sufficiently adequate, no further information needed