Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Ecotoxicological information

Short-term toxicity to aquatic invertebrates

Administrative data

Endpoint:
short-term toxicity to aquatic invertebrates
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 and falling into its applicability domain, with adequate and reliable documentation / justification
Justification for type of information:
The target compound is a salt. The procedure commonly used to generate in silico predictions of salt substances consists in the removal of the counter ion (or the inorganic part, or the lowest MW part of the salt molecule) and, if necessary, the subsequent neutralization. This procedure is needed to calculate the descriptors used to make the predictions. This procedure was discussed and confirmed by several experts, both academic and regulators (European Agency for chemicals, ECHA). In the case of the target 2-aminoethanol hydrobromide, for modelling purposes, the chemical structure has been processed by removal of the inorganic part (HBr).

Data source

Reference
Reference Type:
study report
Title:
Unnamed
Year:
2018
Report date:
2018

Materials and methods

Principles of method if other than guideline:
QSAR model: ACD/Percepta model for acute aquatic toxicity in Daphnia magna (ACD Labs/Percepta (2017 Release).

This tool was selected according to the OECD Guidance Document on the validation of (Q)SAR models that describes generally accepted guidelines to evaluate the validity of a (Q)SAR model, which is the first condition to be fulfilled to use a (Q)SAR result or regulatory purposes.

ACD/Percepta (Advanced Chemistry Development, Inc., Pharma Algorithms, Inc.) (release 2017) is a suite of comprehensive tools for the prediction of basic toxicity endpoints, including hERG Inhibition, CYP3A4 Inhibition, Genotoxicity, Acute Toxicity, Aquatic Toxicity, Eye/Skin Irritation, Endocrine System Disruption, and Health Effects. Predictions are made from chemical structure and based upon large validated databases and QSAR models, in combination with expert knowledge of organic chemistry and toxicology. The majority of ACD/Percepta models were developed using the GALAS modelling methodology (Global, Adjusted Locally According to Similarity), which consists in two parts: 1) a global (baseline) statistical model based on PLS with multiple bootstrapping, using a predefined set of fragmental descriptors; 2) local correction to baseline prediction based on the analysis of model performance for similar compounds from the training set. ACD/Percepta allows to evaluate the robustness of the prediction by examining compounds similar to the target from the training set, together with literature data and reference. For the majority of ACD/Percepta models, similarity toward training set analogues is measured in terms of “property-specific” similarity. Quantitatively, the similarity between the target compound and each training set analogue is expressed in terms of their individual Similarity Index (SIi) calculated from the correlation of the corresponding two predicted property value vectors (made of multiple estimated values from bootstrapping PLS models). ACD/Percepta models based on GALAS modelling methodology also provide an estimation of the reliability of the prediction, by a reliability index (RI). This index provides values in a range from 0 to 1 and gives an evaluation of whether a submitted compound falls within the Model Applicability Domain. In particular: RI < 0.3 (Not Reliable), RI in range 0.3-0.5 (Borderline Reliability), RI in range 0.5-0.75 (Moderate Reliability), RI ≥ 0.75 (High Reliability). Estimation of the RI takes into account the following two aspects: similarity of the tested compound to the training set and the consistency of experimental values for similar compounds. The target compound 2-aminoethanol is included in the model applicability domain since the RI is greater than 0.3.

ACD/Aquatic Toxicity: ACD/Labs Aquatic toxicity module is an accurate and reliable predictive tool that serves as a valuable first estimate of the compounds’ toxicity to aquatic organisms. The module calculates short-term toxicity to three species that are typically used in aquatic toxicity assays. The predicted parameters include lethal concentration (LC50, mg/L) to fathead minnows (Pimephales promelas) and water fleas (Daphnia magna), as well as inhibitory growth concentration (IGC50, mg/L) to ciliate protozoa (Tetrahymena pyriformis). Experimental data that was used for model development was collected from the EPA and original scientific publications. After thorough verification of the obtained values the final data sets contained about 900 compounds with quantitative LC50 values characterizing acute toxicity to fishes (Pimephales promelas), 600 compounds - to water fleas (Daphnia magna), and 1100 compounds - to ciliate protozoa (Tetrahymena pyriformis). The predictive models were derived using GALAS modeling methodology.
ACD/Aquatic toxicity module estimates the reliability of predictions in terms of reliability index (RI). Together with the prediction, ACD/Percepta displays up to 5 most similar structures from the training set along with experimental results and references. The similarity is measured in terms of “property-specific” similarity.
GLP compliance:
no

Test material

Reference
Name:
Unnamed
Type:
Constituent

Results and discussion

Effect concentrations
Key result
Duration:
48 h
Dose descriptor:
LC50
Effect conc.:
ca. 160 mg/L
Details on results:
The prediction has a reliability index of 0.59 and is considered moderately reliable (i.e. target compound is included in the applicability domain of the model and prediction is assessed as moderately accurate (rather limited uncertainty)).

Any other information on results incl. tables

Prediction: 48-h LC50 (Daphnia magna) equal to 160 mg/L.

Applicability Domain (AD): Detailed structural and/or response limits of the applicability domain are not defined. The applicability domain of predictions is assessed using the Reliability Index (RI). This index provides values in a range from 0 to 1 and gives an evaluation of whether a target compound falls within the model applicability domain. Estimation of the RI takes into account the following two aspects: similarity of the tested compound to the training set and the consistency of experimental values for similar compounds. For RI values lower than 0.3 the target compound is considered outside the model AD and the prediction is assessed as not reliable. The target compound 2-aminoethanol is included in the model applicability domain since the RI is greater than 0.3.

Structural analogues: ACD/Percepta displays up to 5 most structurally similar structures from the training set along with their experimental test results. The similarity of the target compound with respect to the training set compounds is analysed in terms of “property-specific” and structural similarity.

Four training compounds were identified as analogues of the target 2-aminoethanol; these training analogues exhibit moderate similarity toward the target (similarity index in the range 0.51-0.78) and experimental LC50 values ranging from 26 mg/L (Ethylenediamine) to 13000 mg/L (Ethylene glycol monoethyl ether). The four mostly similar training set analogues are illustrated in Table below.

 Name of analogue  Experimental LC50 (mg/L)  Similarity
 Ethylenediamine  26  0.78
 Diethanolamine 120  0.67 
 Allylamine 40   0.52

Ethylene glycol

monoethyl ether

7700;13000 

0.51 

Prediction uncertainty & reliability: The reliability of the prediction is evaluated by ACD/Percepta by means of the reliability index (RI). In particular: RI < 0.3 (Not Reliable), RI in range 0.3-0.5 (Borderline Reliability), RI in range 0.5-0.75 (Moderate Reliability), RI >= 0.75 (High Reliability). In the case of the target 2-aminoethanol, the prediction was assessed as moderately reliable,

based on the reliability index equal to 0.59 and the following considerations:

- Training set analogues similarity: moderate.

- Experimental data of training set analogues: wide range of LC50 values. In more detail, it is highlighted that: i) the two mostly similar analogues showed a “narrower” range of LC50 values (26 - 120 mg/l) compared to Ethylene glycol monoethyl ether (7700; 13000); ii) experimental LC50 values for the two mostly similar analogues are lower than the target prediction, suggesting a possible slight

underestimation of the target acute toxicity on Daphnia; however, predicted LC50 value for the target and experimental LC50 values for the analogues are all greater than the acute aquatic toxicity classification threshold of 1 mg/L.

Adequacy: The acute toxicity on daphnia QSAR prediction was assessed as adequate for regulatory purposes (Klimisch 2 - result derived from a valid QSAR model and falling into its applicability domain, with adequate and reliable documentation/justification).

Applicant's summary and conclusion

Validity criteria fulfilled:
yes
Conclusions:
This study was designed to generate estimated in silico (non-testing) data for 2-aminoethanol hydrobromide (CAS 23382-12-9) to be used for its hazard assessment. For modelling purposes, the chemical structure of 2-aminoethanol hydrobromide has been processed in silico by removal of the inorganic part (HBr). Thus, in the current report the in silico assessment was performed for 2-aminoethanol.

Based on QSAR ACD/Percepta model for aquatic toxicity module (LC50 Daphnia magna QSAR model), the 48-h LC50 (Daphnia magna) for 2-aminoethanol hydrobromide is predicted to be equal to 160 mg/L.
The prediction has a reliability index of 0.59 and is considered moderately reliable (i.e. target compound is included in the applicability domain of the model and prediction is assessed as moderately accurate (rather limited uncertainty)).