Registration Dossier

Administrative data

Endpoint:
carcinogenicity
Type of information:
(Q)SAR
Adequacy of study:
supporting study
Reliability:
2 (reliable with restrictions)
Justification for type of information:
QSAR prediction: migrated from IUCLID 5.6

Data source

Materials and methods

Principles of method if other than guideline:
Several computational tools are nowadays available for applying in silico approaches. Among them, for QSAR predictions the following were selected and used:
1. ACD/Percepta for the prediction of acute oral toxicity, skin and eye irritation, gene mutation (Ames test), micronucleus in vivo (rodent), carcinogenicity rat and mouse composite, acute aquatic toxicity (Daphnia), octanol-water partition coefficient (logKow)
2. Leadscope for the prediction of gene mutation (Ames test), chromosome aberration in vivo (rat and other rodent), micronucleus in vivo (mouse and rodent), carcinogenicity mouse and rat composite.
GLP compliance:
no

Test animals

Species:
mouse
Strain:
not specified
Sex:
not specified

Results and discussion

Effect levels

Dose descriptor:
NOAEC
Based on:
test mat.
Sex:
not specified
Basis for effect level:
other: Positive; Little Reliability assessment
Remarks on result:
not determinable
Remarks:
no NOAEC identified. Effect type:carcinogenicity (migrated information)

Any other information on results incl. tables

 In silico tool  Prediction  Positive Prediction Probability  Applicability domain Reliability assessment 
 ACD/Percepta  Undefined  0.29  Reliability index (RI) = 0.31  BORDERLINE
 Leadscope Model Applier  POSITIVE  0.62  Model Fragments Count = 27 30% Sim Training Neighbors Count = 2  LITTLE RELIABLE
 CONSENSUS  POSITIVE      

ACD/Percepta ACD/Percepta carcinogenicity on mouse prediction for 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol resulted to be undefined. In fact, the prediction is borderline reliable, being the reliability index equal to 0.31. Together with the prediction, Percepta displays up to 5 most structurally similar structures from the training set along with experimental carcinogenicity on mouse results for the corresponding compounds. The structural similarity is evaluated by a fragmental approach. The information on the structurally similar compounds in the training set is used to further assess the reliability of the prediction, since it illustrates how the test compound, i.e. 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol, is represented in the training set. This analysis can also help to better understand the reliability index value. The five most similar structures from the training set were identified along with their m carcinogenicity on mouse data, as illustrated belo. It has to be noted that they exhibit moderate similarity, with a similarity index lower than 0.7.

Salmeterol RN: 89365-50-4

Result: Negative

Similarity: 0.67

Bisopropol RN: 66722-44-9

Result: Negative

Similarity: 0.62

Bevantolol RN: 59170-23-9

Result: Negative

Similarity: 0.61

Betaxolol RN: 63659-18-7

Result: Negative

Similarity: 0.61

Estriol succinate RN: 514-68-1

Result: Positive

Similarity: 0.61

Leadscope Model Applier Leadscope FDA Model Applier prediction of carcinogenicity on mouse for 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol resulted to be positive since the positive prediction probability is equal to 0.62. The reliability of the prediction is evaluated by two parameters:  Model Fragment Count. Parameter used to verify that the test compound contains a significant number of fragments that are present in the prediction model. The prediction is reliable if at least one model fragment is present in the test compound.  30% Similarity Training Neighbours Count. Number of compound structurally similar to the test compound. In the case of 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol, 27 fragments were found, and two similar structures were identified in the training set as analogs to 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol. The similar training structure, i.e. Bisoprolol and Bosentan, are characterized by a little similarity with 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol. In addition, their experimental data are inconsistent with the prediction, being Bisoprolol negative and Bosentan positive. Thus it was concluded that 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol is predicted positive but the prediction has little reliability, since the similar training structures exhibit a little similarity with 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol, and exhibit controversial data.

Applicant's summary and conclusion

Conclusions:
The carcinogenicity on mouse of 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol was predicted employing two in silico predictors: the QSAR models provided by ACD/Percepta and Leadscope Model applier. The two predictors were employed in order to apply a consensus analysis to enhance the reliability of the prediction. Based on Leadscope prediction, it was concluded that 1-(4-(2-(benzyloxy)ethoxy)phenyl)-1,2-diphenylbutane-1,4-diol is positive on carcinogenicity on mouse.