4-Methyl-N-[[[3-(trifluoromethyl)phenyl]amino]carbonyl] benzenesulfonamide

Endpoint summary

Description of key information

1.      Introduction

The substance 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide (CAS: 1584-79-8) is to be registered under the REACH regulation at 1 to 10 tonnes per annum. In this case the sensitization properties of the substance should be evaluated. The REACH guidance recommends the use of in vitro assays that examine the key steps of the adverse outcome pathway (AOP) for skin sensitization. The key events (KE) of this skin sensitisation pathway are: 1) covalent binding of the electrophilic substance to skin proteins; 2) release of pro-inflammatory cytokines and induction of cyto-protective pathways in keratinocytes; 3) activation and maturation of dendritic cells, and their migration to the local lymph nodes; 4) presentation of the chemical allergen by the dendritic cells (allergen processed by the dendritic cell and displayed in its surface as an epitope) to naïve T-cells, which leads to their differentiation and proliferation into allergen-specific memory T-cells. Even though not considered as being part of the key events from one to four leading to the adverse outcome, dermal bioavailability (penetration and, if applicable, metabolism) is a prerequisite for a substance to cause skin sensitisation, i.e., the substance needs to reach the viable epidermis in its reactive form.

The in chemico/in vitro assays that are typically used to investigate the first 3 steps of the AOP are the direct peptide reactivity assay (DPRA), the Keratinosens assay, and the h-CLAT assay. In addition, there are QSAR models and in vivo assays that may be used to investigate the skin sensitizing potential of a substance. When data are available from more than one source then it is possible to do a WoE evaluation to determine the most relevant conclusion on skin sensitizing potential. This document is a WoE evaluation for 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide and is based on physicochemical properties, QSAR predictions, in chemico study data, and in vivo study data.

2.      QSAR Information

There are several QSAR models that have been used to evaluate the skin sensitizing potential of 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide.

         i.            DEREK Prediction

The Derek Nexus knowledge base system (version 6.1.0) was used to predict skin sensitization potential of 4-Methyl-N-{[3-(trifluoromethyl)phenyl] carbamoyl}benzenesulfonamide. In this case the structure of 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide did not match any structural alerts of examples for skin sensitization in Derek. In addition, the structure does not contain any unclassified or misclassified features and 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide was predicted to be a non-sensitizer. This prediction is true for several animal species, including guinea pig, mouse, and human. This prediction is considered to be both valid and robust.

       ii.            Leadscope Prediction

This report is a computational assessment based on Expert Alert and/or Quantitative Structure Activity Relationship predictions performed using the Leadscope Model Applier. Each model shows a Prediction (positive, negative, not in domain, classification category) and a confidence score (e.g., Positive Prediction Probability) when it is calculated. Certain models have defined a maximum negative probability and a minimum positive probability. If the Prediction indicates Indeterminate, then the probability fell between those cut-offs. If the Prediction indicates "Not in Domain" then the compound was not within the domain of applicability for that model and reliability of the prediction is lower. If the test structure is not at least 30% similar to one of the training set compounds using a fingerprint of Leadscope structural features and at least one model feature is present in the test structure, it is not in the domain. There are several statistics that can be used to help qualify the call. The "Model Features Count" can be used to determine if the test compound contains a significant number of features that are present in the prediction model. The "Training Neighbors count" indicates the number of training compounds that are relatively similar (30%) to the test structure using a fingerprint of Leadscope structural features. The results of the Leadscope prediction for skin sensitization are summarised in the table below.

Effect	Model	Prediction	QSAR Prediction	Expermiental Value	Positive Probability or Precision	Model Feature Count	Training Neighbors Count
Covalent interaction with skin proteins
Protein Reactivity	DPRA Modelv2	Not in Doamin	Not in Domain
Events in dendritic cells
Expression of co-stimulatory and adhesion molecules	h-CLAT Model v2	Negative	Negative		0.457
Events in keratinocytes
Activation of Nrf2-ARE	KeratinoSens Model v2	Negative	Negative		0.439	1	1
Events in rodent lymphocytes
Rodent local lymph node proliferation	LLNA Model V2 Skin Sensitization LLNA Alerts v2	Not in Domain Not in domain	Not in domain Not in Domain		0.173		0
Reaction Domain
Reaction domain	Skin sensitiastion Reaction domain Alerts v2	Not in Domain	Not in Domain				0

There are no positive predictions for skin sensitization potential, whilst some of the models show that the chemical structure of 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide is out of the domain of the model. However, for two of the key steps in the AOP (Keratinosens and h-CLAT) the models provide negative predictions. This prediction is considered to be both valid and robust.

         i.            ToxTree Prediction

The Toxtree skin sensitization model does not provide a direct prediction for skin sensitization, contrary to the aforementioned models. Instead, it predicts skin sensitization indirectly in the form of protein reactivity alerts (electrophilic protein reactivity alerts; SNAr electrophiles, Schiff base formers, Michael acceptors, acylating agents and SN2 electrophiles) are predicted that are assumed to correlate with KE1 in the skin sensitization process (covalent interaction with proteins). Protein reactivity follows skin penetration in the skin sensitization process and, therefore, tends to correlate with the in vivo measurable endpoint contact hypersensitivity. However, as there are also chemicals that are reactive towards proteins and do not elicit skin sensitization, it is generally difficult to deduce skin sensitization potential directly from protein reactivity. The reactive chemistry as an absolute requirement for skin sensitization elicitation is doubted by experts.

For 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide, Toxtree predicts an alert for Michael Acceptor for protein binding, and an alert for Acyl Transfer agent for skin sensitivity reactivity domains. However, this prediction directly contradicts the predictions of the other two QSAR models described above, and also directly contradicts the results of the DPRA assay described below. This prediction is considered to be not valid and not robust.

       ii.            MechoA Prediction

KREATiS have developed a Mechanisms of Action (MechoA) scheme that uses the molecular structure of a substance to determine the toxicity MechoAs of the parent substance and its major metabolites, if they are identified, i.e., the first key events in the Adverse Outcome Pathway (AOP). Using a decision tree based on 2D and 3D structural alerts, the model classifies the substances into 6 major classes (e.g., pro-activity). These classes are further divided in a total of 25 subclasses (e.g., metabolism to reactive compounds). The model has been trained mainly with fish and rodent toxicity data, but also on other species. Therefore, the MechoA model can discern the differences found in toxicity between these species, uniting toxicology and eco-toxicology using the same classification. For mammals, the MechoA model predicts only non-polar narcosis and no toxic metabolites. In this case the potential for protein reactivity and skin sensitization is absent. This prediction is considered to be both valid and robust.

     iii.            Swiss ADME Prediction

The Swiss Institute of Bioinformatics QSAR tool called SwissADME was used to predict the physicochemical and pharmacokinetic properties of 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide. The tool predicted a skin permeation (Log Kp) value of -5.65 cm/s which is a low value and indicates that permeation through the skin is expected to be slow. In this case it seems unlikely that sufficient amounts of 4-Methyl-N-{[3-(trifluoromethyl)phenyl] carbamoyl}benzenesulfonamide can normally permeate the skin to cause skin sensitization.

3.      In Chemico Study (DPRA)

The OECD 442C test guideline describes the direct peptide reactivity assay (DPRA). The DPRA is proposed to address the molecular initiating event of the skin sensitisation AOP, namely protein reactivity, by quantifying the reactivity of test chemicals towards model synthetic peptides containing either lysine or cysteine. Cysteine and lysine percent peptide depletion values are then used to categorise a substance into one of four classes of reactivity (no or minimal reactivity, low, moderate, or high reactivity) for supporting the discrimination between skin sensitisers and non-sensitisers.

4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide was tested in the DPRA in December 2019 by CERI laboratory of Japan. 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide was dissolved in acetonitrile and mixed with cysteine or lysine peptide solution and incubated at 25˚C for at least 24 hours. The reaction solution was analyzed by high performance liquid chromatography and the peak area for each peptide was determined. The percent cysteine peptide depletion and percent lysine peptide depletion were calculated. The depletion values were 2.0% for cysteine peptide and 0.0% for the lysine peptide, and the mean depletion value was 1%. Therefore, the reactivity class of 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl} benzenesulfonamide was ‘No or minimal reactivity’ and the skin sensitization potential was predicted as ‘Negative’. This prediction is considered to be both valid and robust.

4.      In Vivo LLNA Study

The OECD 442B test guideline describes two non-radioactive modifications to the local lymph node assay (LLNA) test method, which utilise non-radiolabelled 5-bromo-2-deoxyuridine (BrdU) in an ELISA [Enzyme-Linked Immunosorbent Assay] - or FCM [Flow Cytometry Method]-based test system to measure lymphocyte proliferation. Similar to the LLNA, the LLNA: BrdU-ELISA investigates the induction phase of skin sensitisation and provides quantitative data suitable for dose-response assessment. The basic principle underlying the LLNA: BrdU-ELISA is that sensitisers induce proliferation of lymphocytes in the lymph nodes draining the site of test chemical application. This proliferation is proportional to the dose and to the potency of the applied allergen and provides a simple means of obtaining a quantitative measurement of sensitisation. Proliferation is measured by comparing the mean proliferation in each test group to the mean proliferation in the vehicle treated control group (VC). The ratio of the mean proliferation in each treated group to that in the concurrent VC group, termed the SI, is determined, and should be ≥1.6 before further evaluation of the test chemical as a potential skin sensitiser is warranted.

4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide was tested in the LLNA: BrdU-ELISA in September 2019 by CERI laboratory of Japan. 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide was dissolved in dimethylformamide (DMF), which is one of the recommended solvents for use in the LLNA, but it is not the primary choice, which is acetone:olive oil (4:1, v/v)(AOO). The laboratory evaluated AOO but concluded that it was not suitable because 4-Methyl-N-{[3-(trifluoromethyl)phenyl] carbamoyl}benzenesulfonamide precipitated from solution after 4 hours, which did not occur in DMF. The study results indicated that a dose-related increase BrdU labelling indices and stimulation indices (SI) was observed, such that all three tested dose levels (10, 25, and 50% w/v) had SI values greater than 2.0, which is the cut-off value for a clear positive response. The SI values for the 10, 25, and 50% concentrations were 2.13, 2.60, and 3.60 respectively, whilst the positive control substance (25% hexyl cinnamic aldehyde) induced an SI value of 5.33. The lymph nodes weights were also recorded and showed a dose related increase. However, the relationship between mean node weight and mean BrdU labelling index was not linear when the data of the positive control substance were included and the relationship between individual node weight and BrdU labelling showed some inconsistency, although perhaps not so much to invalidate the assay.

The vehicle used for 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl} benzenesulfonamide was DMF but investigations have shown that 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide is not fully stable in DMF and showed a slight level of degradation over time. The degradation products of 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide have not been identified and the influence of the mouse ear tissue on the rate of degradation is also not known. It cannot be excluded that the degradation products may have interfered with the LLNA assay. In addition, DMF has been shown to induce a false positive response in the h-CLAT in vitro assay (MB Labs internal report). DMF used as a vehicle in this study did not appear to induce any response in terms of BrdU labelling but it is difficult to be certain because there was no non-solvent control used in the study and there are no historical control data (HCD) included in the study report for comparison purposes. It cannot be excluded that 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl} benzenesulfonamide or its degradation products may have potentiated the false positive activation of dendritic cells (h-CLAT) that has been reported with DMF alone.

The absence of HCD in the study report is a clear deviation from the OECD 442B test guideline, which requires that HCD be included in the study report to demonstrate the proficiency of the laboratory with the OECD 442B assay methodology. The absence of HCD makes it difficult to evaluate the validity of the study and/or the competence of the laboratory. This result of this study is considered to be a suspected false positive.

5.      Conclusion

4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide is not a skin irritant and not acutely toxic via the oral route. The physicochemical properties suggest that absorption through the skin is possible but will be limited and very slow, particularly in the absence of an organic solvent (QSAR predicted Log Kp of -1.77). The consensus prediction of the various QSAR models suggest that 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide does not react with proteins and is not a skin sensitizer. The DPRA in chemico study provided a result that indicated no potential for 4-Methyl-N-{[3-(trifluoromethyl)phenyl] carbamoyl}benzenesulfonamide to react with proteins, which is in agreement with the consensus QSAR prediction. The in vivo LLNA-BrdU assay provided a positive result for skin sensitization. However, there are questions in relation to the validity of this result caused by the selection of DMF as the solvent because 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide is not stable in DMF and DMF itself may induce false positive results in some skin sensitization assays. On a weight of evidence basis, it is concluded that there is insufficient evidence to justify the classification of 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide as a skin sensitizer.

Key value for chemical safety assessment

Skin sensitisation

Link to relevant study records

Reference

Reference 1

Endpoint:

skin sensitisation, other

Type of information:

(Q)SAR

Adequacy of study:

weight of evidence

Study period:

2021

Reliability:

3 (not reliable)

Justification for type of information:

The Toxtree skin sensitization model ( v.3.1.0) does not provide a direct prediction for skin sensitization. Instead, it predicts skin sensitization indirectly in the form of protein reactivity alerts (electrophilic protein reactivity alerts; SNAr electrophiles, Schiff base formers, Michael acceptors, acylating agents and SN2 electrophiles) are predicted that are assumed to correlate with KE1 in the skin sensitization process (covalent interaction with proteins).

Qualifier:

no guideline followed

Principles of method if other than guideline:

The Toxtree skin sensitization model does not provide a direct prediction for skin sensitization, contrary to the aforementioned models. Instead, it predicts skin sensitization indirectly in the form of protein reactivity alerts (electrophilic protein reactivity alerts; SNAr electrophiles, Schiff base formers, Michael acceptors, acylating agents and SN2 electrophiles) are predicted that are assumed to correlate with KE1 in the skin sensitization process (covalent interaction with proteins). Protein reactivity follows skin penetration in the skin sensitization process and, therefore, tends to correlate with the in vivo measurable endpoint contact hypersensitivity. However, as there are also chemicals that are reactive towards proteins and do not elicit skin sensitization, it is generally difficult to deduce skin sensitization potential directly from protein reactivity. The reactive chemistry as an absolute requirement for skin sensitization elicitation is doubted by experts .

Specific details on test material used for the study:

CAS:
1584-79-8
SMILES:
CC1=CC=C(C=C1)S(=O)(=O)NC(=O)NC2=CC=CC(=C2)C(F)(F)F

Remarks on result:

not determinable

Outcome of the prediction model:

data inconclusive

Interpretation of results:

other: prediction considered to be not valid and not robust

Conclusions:

Toxtree predicts an alert for Michael Acceptor for protein binding, and an alert for Acyl Transfer agent for skin sensitivity reactivity domains. However, this prediction directly contradicts the predictions of the other QSAR models, and also directly contradicts the results of the DPRA assay. This prediction is considered to be not valid and not robust.

Reference 2

Endpoint:

skin sensitisation: in vivo (LLNA)

Type of information:

experimental study

Adequacy of study:

other information

Study period:

16 August 2019 to 30 September 2019

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

guideline study without detailed documentation

Qualifier:

according to guideline

Guideline:

OECD Guideline 442B (Skin sensitisation: Local Lymph Node Assay: BrdU-ELISA or –FCM)

GLP compliance:

yes (incl. QA statement)

Type of study:

mouse local lymph node assay (LLNA): BrdU-ELISA

Species:

mouse

Strain:

CBA:J

Sex:

female

Details on test animals and environmental conditions:

TEST ANIMALS:
Twenty-four female CBA/J mice (SPF, 7 weeks old) were quarantined and acclimatized for 6 days. Body weights were measured, no abnormalities were observed in clinical signs, excretions, or bod weight changes.
The animals were separated into 6 groups of four groups and housed in barrier-system animal rooms
ENVIRONMENTAL CONDITIONS:
Temperature 21-25 °C
rel. Humidity: 40-70%
10-15 air changes per hour
photoperiod of 12 h light per day
pelleted diet: ad libitum
water: ad libitum

Vehicle:

dimethylformamide

Concentration:

Pre-screening: 50.0, 25.0 and 5.0 w/v% of test substance formulation
Main study: 50.0, 25.0 and 10.0 w/v%

No. of animals per dose:

pre-screening: one per does: 5.0 w/v%, 10. 0 w/v%, 25.0 w/v% and 50 w/v%.
Main tests: 4 for each test concentration (10. 0 w/v%, 25.0 w/v% and 50 w/v%), the positive and the negative control

Details on study design:

The pre-screen test was performed to determine the appropriate dose levels for skin sensitisation potential in the main study. Twenty-five µL of each formulation was applied to the dorsum of each ear of the animals using a micropipette once a day for 3 consecutive days. All animals were observed daily from the fist application day to terminal day. Body weights of all the animals were measured on Day 1 and Day 6. The ear thickness was measured on Day 1, Day 3 and Day 6 in duplicate.
Main Study: In the pre-screen test, no abnormal changes which suggested systemic toxicity or excessive irritation were noted in any animal. Therefore, the dose levels of the main study were set tat 50.0, 25.0 and 10.0 w/v%. The vehicle, test substance formulations and positive control substance solution were applied in the same way as the pre-screen test. Approximately 48 hours after the final sensitisation, 0.5 mL of BrdU solution was administrated intraperitoneally to each animal using a syringe and a needle. The clinical observations were performed in the same way as the pr-screen test except for the erythema scoring.

Positive control substance(s):

hexyl cinnamic aldehyde (CAS No 101-86-0)

Parameter:

SI

Value:

1.03

Variability:

+/- 0.14

Test group / Remarks:

Vehicle control

Parameter:

SI

Value:

5.33

Variability:

+/- 0.65

Test group / Remarks:

postive control

Parameter:

SI

Value:

3.6

Variability:

+/-0.32

Test group / Remarks:

50.0 (w/v%) of the test substance group

Parameter:

SI

Value:

2.6

Variability:

+/- 0.21

Test group / Remarks:

25.0 (w/v%) test substance group

Parameter:

SI

Value:

2.13

Variability:

+/- 0.51

Test group / Remarks:

10.0 (w/v%) test substance group

Cellular proliferation data / Observations:

The mean lymph node weights were:
vehicle control: 4.93 (mg) +/- 0.62
positive control: 12.05 (mg) +/-1.10
10.0 /w/v% test substance group: 5.37 +/- 0.72
25.0 /w/v% test substance group: 6.18 +/- 0.96
50.0 /w/v% test substance group: 8.73 +/- 0.87

Interpretation of results:

study cannot be used for classification

Conclusions:

As a result the SIs of the 50.0, 25.0 and 10.0 w/v% test substance groups were 3.60, 2.60 and 2.13, thus the under the conditions tested, the test item was considered to be a sensitiser.
The result of the study is considered to be a suspected false positive for different reasons:
The substance was dissolved in dimethylformamide (DMF), which is one of the recommended solvents for use in the LLNA, but it is not the primary choice, which is acetone:olive oil (4:1, v/v)(AOO).
Investigations have shown that the test item is not fully stable in DMF and showed a slight level of degradation over time. The degradation products have not been identified and the influence of the mouse ear tissue on the rate of degradation is also not known.
The lymph nodes weights were also recorded and showed a dose related increase. However, the relationship between mean node weight and mean BrdU labelling index was not linear when the data of the positive control substance were included and the relationship between individual node weight and BrdU labelling showed some inconsistency, although perhaps not so much to invalidate the assay.
The absence of HCD in the study report is a clear deviation from the OECD 442B test guideline, which requires that HCD be included in the study report to demonstrate the proficiency of the laboratory with the OECD 442B assay methodology. The absence of HCD makes it difficult to evaluate the validity of the study and/or the competence of the laboratory.

Executive summary:

As a result the SIs of the 50.0, 25.0 and 10.0 w/v% test substance groups were 3.60, 2.60 and 2.13, thus the under the conditions tested, the test item was considered to be a sensitiser.
The result of the study is considered to be a suspected false positive for different reasons:
The substance was dissolved in dimethylformamide (DMF), which is one of the recommended solvents for use in the LLNA, but it is not the primary choice, which is acetone:olive oil (4:1, v/v)(AOO).
Investigations have shown that the test item is not fully stable in DMF and showed a slight level of degradation over time. The degradation products have not been identified and the influence of the mouse ear tissue on the rate of degradation is also not known.
The lymph nodes weights were also recorded and showed a dose related increase. However, the relationship between mean node weight and mean BrdU labelling index was not linear when the data of the positive control substance were included and the relationship between individual node weight and BrdU labelling showed some inconsistency, although perhaps not so much to invalidate the assay.
The absence of HCD in the study report is a clear deviation from the OECD 442B test guideline, which requires that HCD be included in the study report to demonstrate the proficiency of the laboratory with the OECD 442B assay methodology. The absence of HCD makes it difficult to evaluate the validity of the study and/or the competence of the laboratory.

Reference 3

Endpoint:

skin sensitisation: in chemico

Type of information:

experimental study

Adequacy of study:

key study

Study period:

15 November 2019 to 9 December 2019

Reliability:

1 (reliable without restriction)

Qualifier:

according to guideline

Guideline:

OECD Guideline 442C (In Chemico Skin Sensitisation: Direct Peptide Reactivity Assay (DPRA))

Version / remarks:

2015

GLP compliance:

yes (incl. QA statement)

Type of study:

direct peptide reactivity assay (DPRA)

Details of test system:

cysteine peptide, (Ac-RFAACAA-COOH)

lysine peptide (Ac-RFAAKAACOOH)

Details on the study design:

16.09 mg of the cysteine peptide was weighed, and 30.4 mL of 100 mM phosphate buffer (pH 7.5) was added to prepare 0.667 mM cysteine peptide standard stock solution. The 0.667 mM standard stock solutions were diluted with acetonitrile to prepare 0.534 mM standard solution. The 0.534 mM standard solution was serially diluted with the solution of phosphate buffer/acetonitrile (8/2 v/v) to prepare 0.267, 0.134, 0.0667, 0.0334 and 0.0167 mM standard solution.

16.96 mg of the lysine peptide was weighed, and 31.0 mL of 100 mM ammonium acetate buffer (pH 10.2) was added to prepare 0.667 mM lysine peptide standard stock solution. The 0.667 mM standard stock solution was diluted with acetonitrile to prepare 0.534 mM standard solution. The 0.534 mM standard solution was serially diluted with the solution of ammonium acetate buffer/acetonitrile (8/2 v/v) to prepare 0.267, 0.134, 0.0667, 0.0334 and 0.0167 mM standard solution.

Positive Control substance solution:
Cinnamaldehyde (26.27 mg for cysteine peptide; 24.65 mg for lysine peptide) was weighed and dissolved to 2 mL of acetonitrile to prepare 100 mM positive control substance solution.

Preparation of test substance solution:
On the experiment day, test substance (68.82 mg for cysteine peptide; 68.62 mg for lysine peptide) was weighed and dissolved to 2 mL of acetonitrile to prepare 100 mM test substance solution.

Creation of calibration curve
Standard solutions of each peptide and the solution of each buffer/acetonitrile (8/2 v/v) were analyzed according to the condition described in 14.2 and a calibration curve for each peptide was made by the concentration of standard solution and the peak area. The coefficients of determination (r2) of both cysteine peptide and lysine peptide were 1.000, which satisfied the acceptance criteria. The concentrations for each peptide described below were calculated by the calibration curves.

Verification of the suitability
For each peptide, 750 μL of the 0.667 mM standard stock solution was mixed with 250 μL of acetonitrile to prepare reference control A (n=3). Reference control A was analyzed. Consequently, the mean concentrations of
reference control A were 0.514 and 0.500 mM for cysteine peptide and lysine peptide, respectively, which satisfied the acceptance criteria.

Verification of retention time of test substance
Cysteine peptide:
To prepare co-elution control, 750 μL of the phosphate buffer was mixed with 200 μL of acetonitrile, and then mixed with 50 μL of the test substance solution prepared in the examination of the solvent. The co-elution control was left at 25°C for 24 hours, and then analyzed. Consequently, no peaks derived from the test substance were detected at the retention time of the peptide.

Lysine peptide
To prepare co-elution control, 750 μL of the ammonium acetate buffer was mixed with 250 μL of the lest substance solution prepared in the examination of the solvent. The coelution control was left at 25°C for 24 hours or more, and then analyzed. Consequently, no peaks derived from the test substance
were detected at the retention time of the peptide.

On the experiment day, 100mM test substance solution (68,82 mg for cysteine peptide and 68.62 mg for lysine peptide) was prepared. Peptide depletion was using HPLC. A calibration curve using the standard solutions for both peptides and suitability of the test method was verified with a reference control (A) for each peptide.

Preparation of reference control B and C, and each reaction solution
For each peptide, reference control B (n=6), reference control C (n=3), positive control reaction solution (n=3) and test substance reaction solution (n=3) were prepared and left at 25°C.

Preparation of reference control B
For each peptide, 750 μL of the 0.667 mM standard stock solution was mixed with 250 μL of acetonitrile to prepare reference control B.

Preparation of reference control C
For each peptide, 750 μL of the 0.667 mM standard stock solution was mixed with 250μL of acetonitrile to prepare reference control C.

Preparation of positive control reaction solution
1) Cysteine peptide
To prepare positive control reaction solution, 750 μL of the 0.667 mM standard stock solution was mixed with 200 μL of acetonitrile, and then mixed with 50 μL of the positive control substance solution.
2) Lysine peptide
To prepare positive control reaction solution, 750 μL of the 0.667 mM standard stock solution was mixed with 250 μL of the positive control substance solution.

Preparation of test substance reaction solution
1) Cysteine peptide
To prepare test substance reaction solution, 750 μL of the 0.667 mM standard stock solution was mixed with 200 μL of acetonitrile, and then mixed with 50 μL of the test substance solution.
Test substance reaction solution was visually inspected immediately after the preparation and 23 hours after the preparation. Consequently, no suspension or precipitation were observed.
2) Lysine peptide
To prepare test substance reaction solution, 750 μL of the 0.667 mM standard stock solution was mixed with 250 μL ofthe test substance solution.
Test substance reaction solution was visually inspected immediately a:fter the preparation and 23 hours after the preparation. Consequently, no suspension or precipitation were observed.
Analysis of reference control B and C, and each reaction solution
For each peptide, the reference control B and C, positive control reaction solution and test substance reaction solution were analyzed.
The analysis of the positive control reaction solution and test substance reaction solution was conducted 24 hours after the preparation or after.
Consequently, the coefficients of variation (CV) of the peak area of the reference control B and C were 1.9% and 0.3% for cysteine peptide and lysine peptide, respectively, which satisfied the acceptance criteria . The mean concentrations of the reference control C were 0.453 mM and 0.501 mM for cysteine peptide and lysine peptide, respectively, which satisfied the acceptance criteria.

Vehicle / solvent:

acetonitrile

Positive control:

cinnamic aldehyde

Positive control results:

Peptide depletion (percent cysteine peptide): 71.5 % (SD 0.4)
Peptide depletion (percent lysine peptide): 55.8 % (SD 0.7)

Key result

Group:

test chemical

Run / experiment:

run/experiment 2

Parameter:

mean lysine depletion

Value:

0 %

At concentration:

100 mM

Positive controls validity:

valid

Key result

Group:

test chemical

Run / experiment:

run/experiment 1

Parameter:

mean cystein depletion

Value:

2 %

At concentration:

100 mM

Positive controls validity:

valid

Interpretation of results:

GHS criteria not met

Conclusions:

The mean value of the percent cystein and lysine depletion was 1.0% the reactivity class of the test substance was classified to "No or Minimal reactivity" and the skin sensitivity was predicted as "negative".

Executive summary:

The mean value of the percent cystein and lysine depletion was 1.0% the reactivity class of the test substance was classified to "No or Minimal reactivity" and the skin sensitivity was predicted as "negative".

Reference 4

Endpoint:

skin sensitisation, other

Type of information:

(Q)SAR

Adequacy of study:

weight of evidence

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 Derek Nexus knowledge base system (version 6.1.0) was used to predict skin sensitization potential of the test item. In this case the structure of the test item did not match any structural alerts of examples for skin sensitization in Derek. In addition, the structure does not contain any unclassified or misclassified features and the test item was predicted to be a non-sensitizer. This prediction is true for several animal species, including guinea pig, mouse, and human.

Qualifier:

no guideline followed

Principles of method if other than guideline:

Principle of test: QSAR prediction of skin sensitisation
Parameters analysed/observed: skin sensitisation

Specific details on test material used for the study:

Smiles: CC1=CC=C(C=C1)S(=O)(=O)NC(=O)NC2=CC=CC(=C2)C(F)(F)F

Run / experiment:

other: QSAR

Parameter:

other: Prediction

Remarks on result:

other: No indication of skin sensitisation based on QSAR /QSPR prediction

Outcome of the prediction model:

negative [in vitro/in chemico]

The Derek Nexus knowledge base system (version 6.1.0) was used to predict skin sensitization potential of the test item. In this case the structure of the test item did not match any structural alerts of examples for skin sensitization in Derek. In addition, the structure does not contain any unclassified or misclassified features and the test item was predicted to be a non-sensitizer. This prediction is true for several animal species, including guinea pig, mouse, and human.

Interpretation of results:

other: predicted non-sensitiser

Conclusions:

In this case the structure of the test item did not match any structural alerts of examples for skin sensitization in Derek. In addition, the structure does not contain any unclassified or misclassified features and was predicted to be a non-sensitizer. This prediction is true for several animal species, including guinea pig, mouse, and human.

Reference 5

Endpoint:

skin sensitisation, other

Type of information:

(Q)SAR

Adequacy of study:

weight of evidence

Study period:

2021

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:

KREATiS have developed a Mechanisms of Action (MechoA) scheme that uses the molecular structure of a substance to determine the toxicity MechoAs of the parent substance and its major metabolites, if they are identified, i.e., the first key events in the Adverse Outcome Pathway (AOP).

Qualifier:

no guideline followed

Principles of method if other than guideline:

KREATiS have developed a Mechanisms of Action (MechoA) scheme that uses the molecular structure of a substance to determine the toxicity MechoAs of the parent substance and its major metabolites, if they are identified, i.e., the first key events in the Adverse Outcome Pathway (AOP).

Remarks on result:

other: no indication of skin sensitisation based on QSAR/QSPR prediction

Outcome of the prediction model:

negative [in vitro/in chemico]

KREATiS have developed a Mechanisms of Action (MechoA) scheme that uses the molecular structure of a substance to determine the toxicity MechoAs of the parent substance and its major metabolites, if they are identified, i.e., the first key events in the Adverse Outcome Pathway (AOP). Using a decision tree based on 2D and 3D structural alerts, the model classifies the substances into 6 major classes (e.g., pro-activity). These classes are further divided in a total of 25 subclasses (e.g., metabolism to reactive compounds). The model has been trained mainly with fish and rodent toxicity data, but also on other species. Therefore, the MechoA model can discern the differences found in toxicity between these species, uniting toxicology and eco-toxicology using the same classification. For mammals, the MechoA model predicts only non-polar narcosis and no toxic metabolites. In this case the potential for protein reactivity and skin sensitization is absent. This prediction is considered to be both valid and robust.

Interpretation of results:

other: For mammals, the model predicts only non-polar narcosis and no toxic metabolites. The potential for protein reactivity and skin sensitization is absent.

Conclusions:

For mammals, the MechoA model predicts only non-polar narcosis and no toxic metabolites. This prediction is considered to be both valid and robust.

Reference 6

Endpoint:

skin sensitisation, other

Type of information:

(Q)SAR

Adequacy of study:

weight of evidence

Study period:

2021

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:

This report is a computational assessment based on Expert Alert and/or Quantitative Structure Activity Relationship predictions performed using the Leadscope Model Applier. Each model shows a Prediction (positive, negative, not in domain, classification category) and a confidence score (e.g., Positive Prediction Probability) when it is calculated.

Qualifier:

no guideline followed

Principles of method if other than guideline:

QSAR estimation using Leadscope Model

Remarks on result:

other: no indication of skin sensitisation based on QSAR/QSPR prediction

Outcome of the prediction model:

negative [in vitro/in chemico]

This report is a computational assessment based on Expert Alert and/or Quantitative Structure Activity Relationship predictions performed using the Leadscope Model Applier. Each model shows a Prediction (positive, negative, not in domain, classification category) and a confidence score (e.g., Positive Prediction Probability) when it is calculated. Certain models have defined a maximum negative probability and a minimum positive probability. If the Prediction indicates Indeterminate, then the probability fell between those cut-offs. If the Prediction indicates "Not in Domain" then the compound was not within the domain of applicability for that model and reliability of the prediction is lower. If the test structure is not at least 30% similar to one of the training set compounds using a fingerprint of Leadscope structural features and at least one model feature is present in the test structure, it is not in the domain. There are several statistics that can be used to help qualify the call. The "Model Features Count" can be used to determine if the test compound contains a significant number of features that are present in the prediction model. The "Training Neighbors count" indicates the number of training compounds that are relatively similar (30%) to the test structure using a fingerprint of Leadscope structural features. The results of the Leadscope prediction for skin sensitization are summarised in the table below.

 

Effect	Model	Prediction	QSAR Prediction	Expermiental Value	Positive Probability or Precision	Model Feature Count	Training Neighbors Count
Covalent interaction with skin proteins
Protein Reactivity	DPRA Modelv2	Not in Doamin	Not in Domain
Events in dendritic cells
Expression of co-stimulatory and adhesion molecules	h-CLAT Model v2	Negative	Negative		0.457
Events in keratinocytes
Activation of Nrf2-ARE	KeratinoSens Model v2	Negative	Negative		0.439	1	1
Events in rodent lymphocytes
Rodent local lymph node proliferation	LLNA Model V2 Skin Sensitization LLNA Alerts v2	Not in Domain Not in domain	Not in domain Not in Domain		0.173		0
Reaction Domain
Reaction domain	Skin sensitiastion Reaction domain Alerts v2	Not in Domain	Not in Domain				0

There are no positive predictions for skin sensitization potential, whilst some of the models show that the chemical structure of the test item is out of the domain of the model. However, for two of the key steps in the AOP (Keratinosens and h-CLAT) the models provide negative predictions.

Interpretation of results:

other: predicted non-sensitiser

Conclusions:

For two of the key steps in the AOP (Keratinosens and h-CLAT) the models provide negative predictions.

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (not sensitising)

Respiratory sensitisation

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (not sensitising)

Justification for classification or non-classification

4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide is not a skin irritant and not acutely toxic via the oral route. The physicochemical properties suggest that absorption through the skin is possible but will be limited and very slow, particularly in the absence of an organic solvent (QSAR predicted Log Kp of -1.77). The consensus prediction of the various QSAR models suggest that 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide does not react with proteins and is not a skin sensitizer. The DPRA in chemico study provided a result that indicated no potential for 4-Methyl-N-{[3-(trifluoromethyl)phenyl] carbamoyl}benzenesulfonamide to react with proteins, which is in agreement with the consensus QSAR prediction. The in vivo LLNA-BrdU assay provided a positive result for skin sensitization. However, there are questions in relation to the validity of this result caused by the selection of DMF as the solvent because 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide is not stable in DMF and DMF itself may induce false positive results in some skin sensitization assays. On a weight of evidence basis, it is concluded that there is insufficient evidence to justify the classification of 4-Methyl-N-{[3-(trifluoromethyl)phenyl]carbamoyl}benzenesulfonamide as a skin sensitizer.