Registration Dossier

Toxicological information

Genetic toxicity: in vitro

Currently viewing:

Administrative data

Endpoint:
in vitro gene mutation study in bacteria
Type of information:
(Q)SAR
Adequacy of study:
key study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
results derived from a valid (Q)SAR model, but not (completely) falling into its applicability domain, with adequate and reliable documentation / justification
Justification for type of information:
QSAR prediction from an well known and acknowledged tool. See below under 'Overall remarks, attachments' for applicability domain and 'attached background material section' for methodology.

Data source

Referenceopen allclose all

Reference Type:
other: Guidance
Title:
Unnamed
Year:
2017
Report Date:
2017
Reference Type:
other: User Guide
Title:
User’s guide for T.E.S.T. (version 4.2) (Toxicity Estimation Software Tool) - A program to estimate toxicity from molecular structure.
Author:
United States Environment Protection Agency (US EPA)
Year:
2016
Bibliographic source:
Toxicity Estimation Software Tool for Microsoft® Windows, v 4.2.1, United States Environmental Protection Agency, Washington, DC, USA (https://www.epa.gov/chemical-research/toxicity-estimation-software-tool-test)
Report Date:
2016

Materials and methods

Test guideline
Qualifier:
according to
Guideline:
other: REACH guidance on QSARs: Chapter R.6. QSARs and grouping of chemicals
Principles of method if other than guideline:
Since the test substance is a UVCB, the mutagenicity potential was predicted for the all the constituents.
GLP compliance:
no
Type of assay:
other: QSAR prediction

Test material

Reference
Name:
Unnamed
Type:
Constituent

Results and discussion

Test results
Key result
Species / strain:
other: QSAR prediction from TEST v4.2.1
Remarks:
Ames Mutagenicity Test
Metabolic activation:
not specified
Genotoxicity:
negative
Remarks:
predicted for all the constituents
Remarks on result:
no mutagenic potential (based on QSAR/QSPR prediction)

Any other information on results incl. tables

Results

Constituents

Name

SMILES

TEST - Predicted Ames Mutagenicity value from Consensus method

TEST - Predicted Ames Mutagenicity result from Consensus method

Prediction statistics for similar chemicals - External and Training Dataset

Concordance

Sensitivity

Specificity

1

mono- C16 PSE, K+ and mono-C16 PSE

CCCCCCCCCCCCCCCCOP(O)(O)=O

0.14

Negative

0.89 and 0.90

0 and 0

1 and 1

2

di- C16 PSE, K+ and di- C16 PSE

CCCCCCCCCCCCCCCCOP(O)(=O)OCCCCCCCCCCCCCCCC

0.07

Negative

1 and 0.90

1 and 0

1 and 1

3

Cetyl alcohol

CCCCCCCCCCCCCCCCO

-0.06

Negative

1 and 1

N/A and NA

1 and 1

4

Isostearyl alcohol

CC(C)CCCCCCCCCCCCCCCO

-0.09

Negative

1 and 1

N/A and NA

1 and 1

5

Stearic acid/potassium stearate

CCCCCCCCCCCCCCCCCC(=O)O

-0.06

Negative

1 and 1

N/A and NA

1 and 1

6

Cetyl stearate

CCCCCCCCCCCCCCCCCC(=O)O(CCCCCCCCCCCCCCCC)

0

Negative*

-

-

-

7

Cetyl Isostearate

CC(C)CCCCCCCCCCCCCCC(=O)O(CCCCCCCCCCCCCCCC)

0.03

Negative

1 and 0.90

1 and 0

1 and 1

8

Isostearyl Isostearate**

CC(C)CCCCCCCCCCCCCCC(=O)O(CCCCCCCCCCCCCCCC(C)C)

0.06

Negative

1 and 0.90

1 and 0

1 and 1

9

Isostearyl stearate

CC(C)CCCCCCCCCCCCCCCOC(=O)CCCCCCCCCCCCCCCCC

0.06

Negative

1 and 0.90

1 and 0

1 and 1

N/A: Not available

*Experimental result available. So no need to perfprm domain evaluation.

In general, if the concordance is greater than or equal to 0.8, the model is considered to be valid. In addition both the leave-one-out sensitivity and specificity must be at least 0.5 to avoid using models which are heavily biased to predict either active or inactive scores.

**representative component of the 'esters of C18 (branched and linear) fatty acids with C18 (branched and linear alcohols)'

For more details on results, kindly refer the attached background material section of the IUCLID.

Applicant's summary and conclusion

Conclusions:
Based on the negative mutagenicity predictions for all the constituents using the Consensus method of the T.E.S.T. v4.2.1 program, the test substance is overall considered to be non-mutagenic
Executive summary:

The mutagenicity potential of the test substance 'Reaction products of hexadecyl dihydrogen phosphate, dihexadecyl hydrogen phosphate, hexadecan-1-ol, stearic acid, esters of C18 (branched and linear) fatty acids with C18 (branched and linear) alcohols, and potassium hydroxide' (UVCB) was predicted using the Consensus method of the T.E.S.T. v4.2.1 program. Since the test substance is a UVCB, the mutagenicity potential was predicted for all the individual constituents, using SMILES as the input parameter. All the constituents were predicted to be negative for mutagenicity (US EPA, 2019), indicating that the test substance can be overall considered to be non-mutagenic. Applicability domain evaluation was performed by checking the descriptor and structural fragment domains of the individual QSAR methods (i.e., FDA, hierarchical clustering and nearest neighbour methods) underlying Consensus model predictions. Since an experimental value could be identified for cetyl stearate, no domain evaluation was necessary for it. The domain evaluation of the remaining all constituents indicated that they were within both descriptor and structural fragment domains of the FDA and hierarchical clustering methods, but not completely within domain for the structural fragments identified for the three nearest neighbours. Further, the prediction accuracy of the binary toxicity endpoints, can be evaluated in terms of the fraction of compounds that are predicted accurately and are described based on three statistics parameters: concordance, sensitivity, and specificity. In general, if the concordance is greater than or equal to 0.8, the model is considered to be valid. And both the leave-one-out sensitivity and specificity must be at least 0.5 to avoid using models which are heavily biased to predict either active or inactive scores. Out of three statistical parameters, the concordance and specificity fractions of all the constituents are above the required cut-offs. For sensitivity, effectively only one chemical was identified to be experimentally active but not predicted correctly for constituent 1. Given the structural differences of this chemical compared to the target, the overall accuracy of the model and its prediction is not expected to be compromised. Therefore, the predictions can be overall considered to be accurate with medium to high confidence.