titanyl (IV) diascrobate

Endpoint summary

Key value for chemical safety assessment

Acute toxicity: via oral route

Link to relevant study records

Reference

Endpoint:

acute toxicity: oral

Type of information:

(Q)SAR

Adequacy of study:

key study

Study period:

Mai 2019

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:

The computational simulation was performed based on the read-across approach. The readacross is one of the so-called alternative test methods recommended by REACH, where the predictions are based on the experimental data available for the most similar compounds. The predictions were performed according to the Read-Across Assessment Framework (RAAF), which assumes six different risk assessment scenarios of chemical compounds.
Applied tool:
The OECD QSAR Toolbox, version 4.3
Procedure of analysis:
I. Profiling of the target substance in order to retrieve relevant information related to mechanism of action and observed or simulated metabolites
II. Analogue (source compound) search based on selected criteria:
a. analogue hydrolysis similarly like the target compound (hydrolysis simulator (neutral))
b. analogue has similar transformation products as the target compound (metabolism simulators, similarity >50%).
III. Data collection for the analogues (OECD Toolbox database/ECHA CHEM).
IV. Toxicity prediction for the target substance
V. Category consistency check in order to assess the quality of the prediction
Applied scenario:
Scenario 1
Toxicity prediction for the target substance:
This read-across is based on the hypothesis that source and target substances have similar toxicological properties like substrates because they hydrolysed to common products.
The target substance is an organometallic compound containing titanium (IV) centre, and ascorbate (Asc) ligands. The metallic centres of the substance are linked by oxygen
coordination bonds of the Asc ligands.
The only one sodium L-ascorbate (NaAsc) analogue has been found according to the assumed requirements; therefore, it was used as the source compound. The acute oral toxicity for analogue was measured according to the OECD 401 and this value was taken into account for the prediction.
The acute oral toxicity for the source compound was performed according to:
Test guideline: OECD 401
Endpoint: LD50
Test organism: rat
The read-across prediction of the acute oral toxicity for the target substance was performed based on the approach “one to one”.

Principles of method if other than guideline:

In order to meet regulatory needs, reliability of the predicted results should be assessed. In case of classic quantitative structure-activity relationships (QSAR) modelling, this idea can be realised by analysing, whether the predicted value is located within so-called applicability domain. The applicability domain is a theoretical region, defined by the range of toxicity values and structural descriptors for the training compounds, where the predictions may be considered as realistic ones. In a specific case of read-across, the assessment is performed based on the assessment of degree of similarity between the source and target compounds (in %). Moreover, the internal consistency of the group of source compounds (called „category” in OECD Toolbox nomenclature, independently which approach: analogue approach or category approach is used). The category consistency check could be based on the parameters describing the structural similarity and/or properties as well as mechanistic similarity of the tested compounds.
For example, all members of the category (analogues as well as target substance) need to have the same functional groups and endpoint specific alerts.
In the case of read-across-based prediction of the acute oral toxicity of the titanyl (IV) diascorbate dihydrate, the read-across hypothesis considers that source and target compounds have similar transformation products. Based on the Dice measure, the structural similarity between hydrolysis products of source and target substances was at least 50%. Therefore, using experimental data of sodium ascorbate for predicting biological activity for the target compound was justified.
Besides, the category consistency, the boundaries of the applicability domain are verified by the critical value of log KOW. In case of TiO(Asc)2 x2H2O, the log KOW value is not available. What is more, in case of “one to one” approach, this criterion would be met only if source and target compounds are the same substance. Thus, information that “domain is not defined” is not critical in this situation.
The structural similarity between the source (NaAsc) and the target compound (TiO(Asc)2 x2H2O) equals to 53.7%

Key result

Dose descriptor:

LD50

Effect level:

1 410 mg/kg bw

Clinical signs:

other:

Interpretation of results:

Category 4 based on GHS criteria

Remarks:

Migrated information Criteria used for interpretation of results: EU

Conclusions:

The acute oral toxicity for the target substance is predicted at level LD50 = 1 410 mg/kg bdwt

Executive summary:

The target compound undergoes a hydrolysis reaction. The analogues search was performed assuming at least 50% structural similarity between hydrolysis products of source and target substances. The toxicity prediction was performed based on the experimental data included in the OECD QSAR Toolbox. Sodium L-ascorbate would have similar hydrolysis products as well as the experimental data related to its acute oral toxicity was available. Therefore, the prediction is based only on the NaAsc.

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Dose descriptor:

LD50

Value:

1 410 mg/kg bw

Acute toxicity: via inhalation route

Link to relevant study records

Reference

Endpoint:

acute toxicity: inhalation

Type of information:

(Q)SAR

Adequacy of study:

key study

Study period:

2019

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:

The computational simulation was performed based on the read-across approach. The readacross is one of the so-called alternative test methods recommended by REACH, where the predictions are based on the experimental data available for the most similar compounds. The predictions were performed according to the Read-Across Assessment Framework (RAAF), which assumes six different risk assessment scenarios of chemical compounds.
Applied tool:
The OECD QSAR Toolbox, version 4.3
Procedure of analysis:
I. Profiling of the target substance in order to retrieve relevant information related to mechanism of action and observed or simulated metabolites
II. Category (source compounds) search based on selected criteria:
a. analogues are structurally similar to the target compound (similarity >20%)
b. analogues have the same structural features by the Toxic hazard classification by Cramer profiler.
III. Data collection for the analogues (OECD Toolbox database/ECHA CHEM).
IV. Toxicity prediction for the target substance
V. Category consistency check in order to assess the quality of the prediction
Applied scenario:
Scenario 2
Toxicity prediction for the target substance:
This read-across is based on the fact that the organism is not exposed to common compounds but rather, as a result of similarity, to chemicals which have similar (eco)toxicological and fate properties.
The target substance is an organometallic compound containing titanium (IV) centres, ascorbate (Asc) ligands. The metallic centres of the substance are linked by oxygen
coordination bonds of the Asc ligands.
The target and source chemicals are classified as High (Class III) according to the Toxic hazard classification by Cramer profiler. Also, analogues are structurally similar to the target compound in more than 20%. Five compounds that met these requirements were found. Two compounds were excluded initially, due to no given information regarding used OECD guideline. Three remaining analogues had specified (OECD 403 guideline). Taking into the consideration the ‘worst-case scenario’ approach, out of three analogues that met all the criteria, one value from one analogue (2-benzofuran-1,3-dione) was selected. The acute toxicity by inhalation for analogue was measured according to the OECD 403 and this value was taken into account for the prediction.
Table 6. Compounds that met the assumed requirements for acute toxicity by inhalation predictions
Chemicals Acute Toxicity by inhalation [mg/L] OECD guideline
D-Glucose CAS 9050-36-6 >5.16 No information
4-(Hydroxymethyl)-1,3-dioxolan-2-one
CAS 931-40-8 >5.6 OECD 403
2-benzofuran-1,3-dione
CAS 85-44-9 >2.14 OECD 403
Tin (II) oxalate
CAS 814-94-8 2 No information
1,2,4-Benzenetricarboxylic anhydride
CAS 552-30-7 >2.33 OECD 403
The acute inhalation toxicity for the source compound was performed according to:
Test guideline: OECD 403
Endpoint: LC50
Test organism: rat
The read-across prediction of the acute inhalation toxicity for the target substance was performed based on the “one to one” approach.

Principles of method if other than guideline:

In order to meet regulatory needs, reliability of the predicted results should be assessed. In case of classic quantitative structure-activity relationships (QSAR) modelling, this idea can be realised by analysing, whether the predicted value is located within so-called applicability domain. The applicability domain is a theoretical region, defined by the range of toxicity values and structural descriptors for the training compounds, where the predictions may be considered as realistic ones. In a specific case of read-across, the assessment is performed based on the assessment of degree of similarity between the source and target compounds (in %). Moreover, the internal consistency of the group of source compounds (called „category” in OECD Toolbox nomenclature, independently which approach: analogue approach or category approach is used). The category consistency check could be based on the parameters describing the structural similarity and/or properties as well as mechanistic similarity of the tested compounds.
For example, all members of the category (analogues as well as target substance) need to have the same functional groups and endpoint specific alerts.
In the case of read-across-based prediction of the acute toxicity by inhalation of the titanyl (IV) diascorbate dihydrate, the read-across hypothesis considers that source and target compounds are classified as “High (Class III)” according to the Toxic hazard classification by Cramer profiler. Also, analogues are structurally similar to the target compound in more than 20%.
Besides, the category consistency, the boundaries of the applicability domain are verified by the critical value of log KOW. In case of titanyl (IV) diascorbate dihydrate, log KOW value is not available. What is more, in case of “one to one” approach, this criterion would be met only if source and target compounds are the same substance. Thus, information that “domain is not defined” is not critical in this situation.
The structural similarity between the source (2-benzofuran-1,3-dione) and the target compound (titanyl (IV) diascorbate dihydrate) equals to 20,5 %.

Dose descriptor:

LC50

Effect level:

2.14 mg/L air

Interpretation of results:

Category 4 based on GHS criteria

Conclusions:

The acute inhalation toxicity for the target substance is predicted at level LC50 = 2.14 mg/L air.

Executive summary:

The source and target compounds are classified as “High (Class III)” according to the Toxic hazard classification by Cramer profiler. Also, analogues are structurally similar to the target compound in more than 20%. The toxicity prediction was performed based on the experimental data included in the OECD QSAR Toolbox. Five chemicals met the assumed requirements, but due to availability of experimental data related to the acute toxicity by inhalation and the worstcase scenario, 2-benzofuran-1,3-dione was chosen as the source compound.

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Dose descriptor:

LC50

Value:

2 140 mg/m³ air

Acute toxicity: via dermal route

Link to relevant study records

Reference

Endpoint:

acute toxicity: dermal

Type of information:

(Q)SAR

Adequacy of study:

key study

Study period:

2019

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:

The computational simulation was performed based on the read-across approach. The readacross is one of the so-called alternative test methods recommended by REACH, where the predictions are based on the experimental data available for the most similar compounds. The predictions were performed according to the Read-Across Assessment Framework (RAAF), which assumes six different risk assessment scenarios of chemical compounds.
Applied tool:
The OECD QSAR Toolbox, version 4.3
Procedure of analysis:
I. Profiling of the target substance in order to retrieve relevant information related to mechanism of action and observed or simulated metabolites
II. Category (source compounds) search based on selected criteria:
a. analogue is structurally similar to the target compound (similarity >30%)
b. analogue has the same alert according to the Acute Oral Toxicity profiler
c. analogue has the same alert “High level” according to the Cramer profile.
III. Data collection for the analogues (OECD Toolbox database/ECHA CHEM).
IV. Toxicity prediction for the target substance
V. Category consistency check in order to assess the quality of the prediction
Applied scenario:
Scenario 2
Toxicity prediction for the target substance:
This read-across is based on the fact that the organism is not exposed to common compounds but rather, as a result of similarity, to chemicals which have similar (eco)toxicological and fate properties.
The target substance is an organometallic compound containing titanium (IV) centres, ascorbate (Asc) ligands. The metallic centres of the substance are linked by oxygen
coordination bonds of the Asc ligands. The target and source chemicals are classified as “High (Class III)” according to the Toxic hazard classification by Cramer profiler and they are not categorized according to the Acute Oral Toxicity profiler. Also, analogues are structurally similar to the target compound in more than 30%. Two compounds that met these requirements were found, but only one was tested with the recommended OECD 402 guideline (D-gluconic acid, CAS 526-95-4) and was chosen as the source compound and taken into account for the prediction.
Table. Identified analogues that met the assumed requirements for acute toxicity by dermal route predictions
Chemicals Acute Toxicity by dermal route [mg/kg bdwt] OECD guideline
Sodium erythorbate
CAS 6381-77-7 2 000 No information
D-gluconic acid
CAS 526-95-4 2 000 OECD 402
The acute dermal toxicity for the source compound was performed according to:
Test guideline: OECD 402
Endpoint: LD50
Test organism: rat
The read-across prediction of the acute dermal toxicity for the target substance was performed based on the approach “one to one”.

Principles of method if other than guideline:

In order to meet regulatory needs, reliability of the predicted results should be assessed. In case of classic quantitative structure-activity relationships (QSAR) modelling, this idea can be realised by analysing, whether the predicted value is located within so-called applicability domain. The applicability domain is a theoretical region, defined by the range of toxicity values and structural descriptors for the training compounds, where the predictions may be considered as realistic ones. In a specific case of read-across, the assessment is performed based on the assessment of degree of similarity between the source and target compounds (in %). Moreover, the internal consistency of the group of source compounds (called „category” in OECD Toolbox nomenclature, independently which approach: analogue approach or category approach is used). The category consistency check could be based on the parameters describing the structural similarity and/or properties as well as mechanistic similarity of the tested compounds.
For example, all members of the category (analogues as well as target substance) need to have the same functional groups and endpoint specific alerts.
In the case of read-across-based prediction of the acute toxicity by dermal route of the titanyl (IV) diascorbate dihydrate, the read-across hypothesis considers that source and target compounds are classified as “High (Class III)” according to the Toxic hazard classification by
Cramer profiler and they are not categorized according to the Acute Oral Toxicity profiler.
Also, analogues are structurally similar to the target compound in more than 30%.
Besides, the category consistency, the boundaries of the applicability domain are verified by the critical value of log KOW. In case of titanyl (IV) diascorbate dihydrate, log KOW value is not available. What is more, in case of “one to one” approach, this criterion would be met only if source and target compounds are the same substance. Thus, information that “domain is not defined” is not critical in this situation.
The structural similarity between the source (D-gluconic acid) and the target compound (titanyl (IV) diascorbate dihydrate) equals to 34,1 %.

GLP compliance:

no

Remarks:

QSAR

Dose descriptor:

LD50

Effect level:

2 000 mg/kg bw

Clinical signs:

other:

Interpretation of results:

Category 4 based on GHS criteria

Conclusions:

The acute dermal toxicity for the target substance is predicted at level LD50 = 2000 mg/kg bdwt

Executive summary:

The source and target compounds are classified as “High (Class III)” according to the Toxic hazard classification by Cramer profiler. Also, analogues are structurally similar to the target compound in more than 30%. The toxicity prediction was performed based on the experimental data included in the OECD QSAR Toolbox. Two chemicals met the assumed requirements, but due to the availability of experimental data obtained with the recommended OECD 402 guideline related to the acute toxicity by dermal route and the worst-case scenario, D-gluconic acid was chosen as the source compound.

Endpoint conclusion

Endpoint conclusion:

adverse effect observed

Dose descriptor:

LD50

Value:

2 000 mg/kg bw

Justification for classification or non-classification

Based on QSAR test the substance is classified as dangerous (category 4).