(4E)-4-(3,3,4-trimethylcyclopentylidene)butanal

Skin sensitisation

Administrative data

Endpoint:

skin sensitisation, other

Remarks:

in silico prediction model

Type of information:

(Q)SAR

Adequacy of study:

key study

Study period:

09 July 2020

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

results derived from a valid (Q)SAR model and falling into its applicability domain, with adequate and reliable documentation / justification

Justification for type of information:

1. SOFTWARE
Derek Nexus v6.1

2. MODELS (incl. version number)
Derek Nexus - skin sensitisation
Derek EC3 Model - 1.2.0


3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
CC1CC(=CCCC=O)CC1(C)C

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
(See QMRF for details)
- Defined endpoint: Skin sensitisation in mammal and LLNA EC3 (%)
- Unambiguous algorithm: yes
- Defined domain of applicability: yes
- Appropriate measures of predictivity: The software reports the number of positive and negative compounds from the validation data sets that activate each alert and calculates positive predictivity using this data.
- Mechanistic interpretation: yes

5. APPLICABILITY DOMAIN
- Descriptor domain: inside
- Structural domain: inside
- Mechanistic domain: inside
- Similarity with analogues in the training set: Yes

6. ADEQUACY OF THE RESULT
Derek Nexus skin sensitisation model is based on the activity (EC3 values) for nearest neighbours derived from a local lymph node assay data set of chemicals. This model gives quantitative EC3 predictions (given in % units) for alerting compounds which can be used for the purpose of classification and labelling and/or risk assessment.

Data source

Reference

Reference Type:

study report

Title:

Unnamed

Year:

2020

Report date:

2020

Materials and methods

Principles of method if other than guideline:

Model description: Derek Nexus gives quantitative EC3 predictions (in % units) for alerting compounds for nearest neighbours derived from a local lymph node assay data set and gives negative predictions when no alert is fired for the skin sensitisation endpoint.

Alert writers design the patterns to describe the activating and deactivating features found during expert assessment of the alert training set. Structural fingerprints are generated using Derek EC3 model 2020 from the 685 compounds (comprising 595 sensitisers and 90 non-sensitisers) in the local lymph node data set [Canipa et. al., 2017]. Misclassified and unclassified features are generated by processing a large Skin Sensitisation reference set (comprising 1467 sensitisers and 1674 non-sensitisers) against Derek Nexus (v6.1) and fragmenting [Chilton et. al., 2018].

GLP compliance:

no

Remarks:

Not applicable (in silico model)

Test material

Test material information

Constituent 1

Reference substance name:

(4E)-4-(3,3,4R-trimethylcyclopentylidene)butanal

Molecular formula:

C12 H20 O

IUPAC Name:

(4E)-4-(3,3,4R-trimethylcyclopentylidene)butanal

Constituent 2

Reference substance name:

(4E)-4-(3,3,4S-trimethylcyclopentylidene)butanal

Molecular formula:

C12 H20 O

IUPAC Name:

(4E)-4-(3,3,4S-trimethylcyclopentylidene)butanal

Test material form:

liquid

Details on test material:

- SMILES: CC1CC(=CCCC=O)CC1(C)C
- Molecular formula: C12H20O
- Molecular weight: 180.29 g/mol

Results and discussion

In vitro / in chemico

Results

Key result

Group:

test chemical

Run / experiment:

run/experiment 1

Parameter:

other: EC3

Value:

14 %

Vehicle controls validity:

not applicable

Negative controls validity:

not applicable

Positive controls validity:

not applicable

Remarks on result:

positive indication of skin sensitisation

Outcome of the prediction model:

positive [in vitro/in chemico]

Applicant's summary and conclusion

Interpretation of results:

Category 1B (indication of skin sensitising potential) based on GHS criteria

Remarks:

LLNA EC3 value > 2% (predicted EC3 = 14%)

Conclusions:

Using DEREK Nexus skin sensitization model, the test item is inside the applicability domain of the model triggering 2 structural alerts: "alert 419 Aldehyde” and "alert 712 Terpenoid". The "alert 419 Aldehyde” has a predictive performance between 23% - 100% depending on the source of the data set.
Using Derek EC3 Model - 1.2.0, the test item generated an LLNA EC3 value of 14% via the nearest neighbors with "alert 419 Aldehyde” which is considered reliable and fit for purpose.

Executive summary:

The test item is predicted as weak skin sensitizer (Category 1B) with an LLNA EC3 value of 14% using the Derek EC3 Model - 1.2.0 in silico model. Based on the structural similarity verification of the data set used to generate the EC3 prediction and on the "alert 419 Aldehyde" predictive performance up to 100%, the prediction is considered reliable and fit for purpose.