Dibutyl peroxydicarbonate

Short-term toxicity to fish

Link to relevant study record(s)

Reference

Endpoint:

short-term toxicity to fish

Type of information:

(Q)SAR

Adequacy of study:

supporting study

Study period:

2018-04-10

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:

For the endpoint short-term toxicity to fish four different, scientifically valid QSAR models were applied within a weight-of-evidence approach.

Qualifier:

according to guideline

Guideline:

other: Guidance on information requirements and chemical safety assessment Chapter R.6: QSARs and grouping of chemicals.

Principles of method if other than guideline:

QSAR prediction using VEGA.
Models used:
- Fish Acute (LC50) Toxicity classification (SarPy/IRFMN) 1.0.2
- Fish Acute (LC50) Toxicity model (KNN/Read-Across) 1.0.0
- Fish Acute (LC50) Toxicity model (NIC) 1.0.0
- Fathead Minnow LC50 96h (EPA) 1.0.7

GLP compliance:

no

Specific details on test material used for the study:

Compound SMILES: O=C(OOC(=O)OCCCC)OCCCC

Dose descriptor:

LC50

Effect conc.:

3.64 mg/L

Nominal / measured:

estimated

Basis for effect:

not specified

Remarks on result:

other: QSAR predicition

Remarks:

Fish Acute (LC50) Toxicity model (KNN/Read-Across) 1.0.0

Dose descriptor:

LC50

Effect conc.:

1.37 mg/L

Nominal / measured:

estimated

Basis for effect:

not specified

Remarks on result:

other: QSAR predicition

Remarks:

Fish Acute (LC50) Toxicity model (NIC) 1.0.0

Dose descriptor:

LC50

Effect conc.:

3.79 mg/L

Nominal / measured:

estimated

Basis for effect:

not specified

Remarks on result:

other: QSAR prediction

Remarks:

Fathead Minnow LC50 96h (EPA) 1.0.7

Dose descriptor:

other: predicted toxicity class

Effect conc.:

>= 1 - <= 10 mg/L

Remarks on result:

other: QSAR prediction

Remarks:

Fish Acute (LC50) Toxicity classification (SarPy/IRFMN) 1.0.2

Details on results:

See attachment.

Sublethal observations / clinical signs:

Fish Acute (LC50) Toxicity classification (SarPy/IRFMN) 1.0.2

Compound: Molecule 0

Compound SMILES: O=C(OOC(=O)OCCCC)OCCCC

Experimental value: -

Predicted toxicity class: Toxic-2 (between 1 and 10 mg/l)

Structural alerts: Toxicity class 2 alert no. 9

Reliability: the predicted compound is outside the Applicability Domain of the model

Remarks: none

Conclusion: The predicted compound is outside the applicability domain of the model, nevertheless, the read across assessment supports with a reasonable documentation that the QSAR prediction is realistic. There are chemicals with very good similarity in the training set. All similarity values (n=6) > 0.75 (similarity index= 0.799). Most of the similar chemicals of the training set have an experimental value between 1 and 10 mg/L.

Fish Acute (LC50) Toxicity model (KNN/Read-Across) 1.0.0

Compound: Molecule 0

Compound SMILES: O=C(OOC(=O)OCCCC)OCCCC

Experimental value [-log(mg/L)]: -

Predicted toxicity [-log(mg/L)]: -0.56

Predicted toxicity [mg/L]: 3.64

Molecules used for prediction: 4

Experimental value [mg/l]: -

Reliability: the predicted compound is outside the Applicability Domain of the model

Remarks: none

Conclusion: The predicted compound is outside the applicability domain of the model, nevertheless, the read across assessment supports with a reasonable documentation that the QSAR prediction is realistic. Strongly similar compounds with known experimental values in the training set have been found. All similarity values (n=6) > 0.75 (similarity index= 0.792). The experimantal values of the similar compounds (n=6) are between 2.93 and 8.05 mg/L. For compounds # 1 to 4 the experimantal values are in good accordance with the predicted LC50 value (for more details, please refer to attachment).

Fish Acute (LC50) Toxicity model (NIC) 1.0.0

Compound: Molecule 0

Compound SMILES: O=C(OOC(=O)OCCCC)OCCCC

Experimental value [log(1/(mmol/L))]: -

Predicted fish toxicity [log(1/(mmol/L))]: 2.23

Predicted fish toxicity [mg/l]: 1.37

Molecular Weight: 233.25

Experimental value [mg/l]: -

Reliability: the predicted compound is outside the Applicability Domain of the model

Remarks: none

Conclusion: The predicted compound is outside the applicability domain of the model, nevertheless, the read across assessment supports with a reasonable documentation that the QSAR prediction is realistic. There are chemicals with very good similarity in the training set. All similarity values (n=6) > 0.75 (similarity index= 0.809). Most of the chemicals of the training set have a similar fish toxicity.

Fathead Minnow LC50 96h (EPA) 1.0.7

Compound: Molecule 0

Compound SMILES: O=C(OOC(=O)OCCCC)OCCCC

Experimental value [-log(mol/l)]: -

Predicted toxicity [-log(mol/l)]: 4.79

Predicted toxicity [mg/l]: 3.79

Molecular Weight: 233.25

Experimental value [mg/l]: -

Reliability: the predicted compound is outside the Applicability Domain of the model

Remarks: none

Conclusion: The predicted compound is outside the applicability domain of the model, nevertheless, the read across assessment supports with a reasonable documentation that the QSAR prediction is realistic. There are chemicals with very good similarity in the training set. All similarity values (n=6) > 0.75 (similarity index= 0.809). Most of the chemicals of the training set have a similar fish toxicity.

Validity criteria fulfilled:

not applicable

Conclusions:

The lowest estimated LC50 for dibutyl peroxydicarbonate is 1.37 mg/L using QSAR tool VEGA model Fish Acute (LC50) Toxicity model (NIC) 1.0.0.

Executive summary:

For the endpoint short-term toxicity to fish four different, scientifically valid QSAR models were applied within a weight-of-evidence approach.

The acute toxicity of dibutyl peroxydicarbonate to fish was estimated with the models Fish Acute (LC50) Toxicity classification (SarPy/IRFMN) 1.0.2, Fish Acute (LC50) Toxicity model (KNN/Read-Across) 1.0.0, Fish Acute (LC50) Toxicity model (NIC) 1.0.0 and Fathead Minnow LC50 96h (EPA) 1.0.7 implemented in the QSAR tool VEGA.

The following predictions were made:

- Fish Acute (LC50) Toxicity classification (SarPy/IRFMN) 1.0.2: Predicted toxicity [mg/l]: between 1 and 10 mg/L;

- Fish Acute (LC50) Toxicity model (KNN/Read-Across) 1.0.0: Predicted toxicity [mg/l]: 3.64;

- Fish Acute (LC50) Toxicity model (NIC) 1.0.0: Predicted toxicity [mg/l]: 1.37;

- Fathead Minnow LC50 96h (EPA) 1.0.7: Predicted toxicity [mg/l]: 3.79

The most sensitive LC50 value derived is the LC50 of 1.37 mg/L using the Acute (LC50) Toxicity model (NIC) 1.0.0, which is used as key value in a worst-case approach.

This prediction is reliable, it is generated by a scientifically valid model.

Description of key information

In a 96-h acute toxicity study according to OECD 203 (1992), zebra fish (Danio rerio) were exposed to dibutyl peroxydicarbonate at nominal test item loading rates of 0.0 (control) and 0.5 mg/L under semi-static conditions (limit test). The NOELR value, based on mortality and sublethal effects, was ≥ 0.5 mg/L (nominal test loading rate), respectively.  The LOELR was > 0.5 mg/L (nominal test loading rate). Sublethal effects were not observed.

The absence of any mortality indicates that, according to the Threshold Approach for Acute Fish Toxicity, fish is not the most sensitive group of test organisms after short-term exposure and no further testing (i.e. full LC50 Fish Acute Toxicity Test) is required.

Additional information

Furthermore, for the endpoint short-term toxicity to fish four different, scientifically valid QSAR models were applied within a weight-of-evidence approach. The lowest estimated LC50 for dibutyl peroxydicarbonate is 1.37 mg/L using QSAR tool VEGA model Fish Acute (LC50) Toxicity model (NIC) 1.0.0.