Reaction mass of 1,4-bis(methylamino)anthraquinone and 1,4-bis[(2-ethylhexyl)amino]anthraquinone and 1-[(2-ethylhexyl)amino]-4-(methylamino)anthraquinone and 9,10-Anthracenedione, 1,4-bis(pentylamino)-, branched and linear and 9,10-Anthracenedione, 1-(methylamino)-4-(pentylamino)-, branched and linear and 9,10-Anthracenedione, 1-[(2-ethylhexyl)amino]-4-(pentylamino)-, branched and linear

Basic toxicokinetics

Administrative data

Endpoint:

basic toxicokinetics, other

Type of information:

(Q)SAR

Adequacy of study:

supporting study

Study period:

2017

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:

1. SOFTWARE: ADMET predictor and GastroPlus

2. MODEL (incl. version number): ADMET predictor (v7.2, Simulations Plus Inc, Lancaster, CA, USA) and GastroPlus (v9.0, Simulations Plus Inc, Lancaster, CA, USA).

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL:
O=C2c1ccccc1C(=O)c3c2c(ccc3NC)NC (Component A)
O=C2c1ccccc1C(=O)c3c2c(ccc3NCCCCC)NC (Component B)
CCC(CCCC)CNc3ccc(NC)c2C(=O)c1ccccc1C(=O)c23 (Component C)
O=C2c1ccccc1C(=O)c3c2c(ccc3NCCCCC)NCCCCC (Component D)
CCC(CCCC)CNc3ccc(NCCCCC)c2C(=O)c1ccccc1C(=O)c23 (Component E)
CCC(CCCC)CNc3ccc(NCC(CC)CCCC)c2c3C(=O)c1ccccc1C2=O (Component F)

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
Endpoint (OECD Principle 1):
a. Endpoint:
i. The absorption fraction (Fa%) of components A-F following oral, dermal, and inhalation exposures in humans.
ii. Systemic bioavailability (F%), Cmax, Tmax, and AUC0-168 of components A-F following oral, dermal, and inhalation exposures in humans.
iii. The plasma protein binding and volume of distribution (Vd) for components A-F.
iv. The potential metabolism and excretion of components A-F in human.

Algorithm (OECD Principle 2):
a. Model or submodel name: For the prediction of Fa%, F%, Cmax, Tmax, and AUC 0-168 of components A-F, a PBPK model was utilized in GastroPlus (version 9) to simulate absorption parameters and systemic bioavailability of components A-F following a single oral (or inhalation or dermal) dose of 1 mg/kg.bw in a fed 30-year old human (70 kg). The oral dose formulation type was defined to be a suspension with particle size of 2.5 μm mean radius. This particle size was selected based on oral PSA (parameter sensitivity analysis) simulation results of Fa% and F% of the representative component A (Structure A2475-44-7) versus the particle size ranging from radius values of 2.5 μm to 250 μm with GastroPlus; for component A, particle size did affect absorption. Therefore, the particle size (2.5 μm radius) was finally used in simulations for all components AF. The oral absorption in GastroPlus utilizes the Advanced Compartmental
Absorption and Transit (ACAT) model to predict passive absorption across the gut and accounts for soluble and insoluble portions of the administered dose.
The inhalation dose formulation type was defined to be a powder (aerosol) with a particle size of 1.25 μm mean radius. This particle size was selected based on the Concawe report (Hext et al., 1999). Particle sizes of 1.25 μm or less (radius) are considered the fine fraction and are associated with a higher risk of health effects. These smaller particles are considered the highly respirable fraction of a particulate atmosphere and can reach the deep alveolar regions of the lung. In the current GastroPlus simulation, the inhalation dose was delivered over an 8-hr period. This inhalation model includes up to five (5) compartments: an optional nose, extrathoracic, thoracic, bronchiolar, and alveolar-interstitial. The deposition fractions for each compartment were generated with a built-in predictive model based on the International Commission for Radiological Protection Publication 66 (ICRP 66) deposition model described in GastroPlus.
The dermal dose formulation type was defined to be a water suspension with particle size of 25 μm mean radius. This particle size was selected based on the dermal PSA (parameter sensitivity analysis) simulation results of Fa% and F% of the representative component A versus the particle size ranging from radius values of 2.5 μm to 250 μm with GastroPlus. The PSA results showed that particle size did not affect absorption for component A across this particle size range. The dermal absorption simulation model in GastroPlus represents the skin as a
collection of the following compartments: stratum corneum, viable epidermis, dermis, subcutaneous tissue, sebum, hair lipid, and hair core.
The application surface is 1900 cm2 on human arm. The dose volume and exposure time were 19 mL, and 6 hrs, respectively. This surface area, dose volume, and exposure time were selected based on the US EPA dermal exposure assessment report (US EPA, 1992).
Bioavailability predictions for these three exposure routes were made by including metabolism by five major cytochrome (CYP) P450 enzymes (1A2, 2C9, 2C19, 2D6, and 3A4) in human. These QSAR predictions of metabolic clearance [(enzyme kinetics (Km and Vmax) based on recombinant CYP enzymes] were generated using ADMET Predictor (v7.2, Simulations Plus Inc, Lancaster, CA, USA) based on the structures of components A-F.
The plasma protein binding and volume of distribution (Vd) were predicted by ADMET Predictor (v7.5, Simulations Plus Inc, Lancaster, CA, USA).
The metabolism and excretion of components A-F were proposed based on the CYP metabolism in human predicted by ADMET predictor.
b. Model version: GastroPlus v9.0 (Simulations Plus Inc, Lancaster, CA, USA); ADMET Predictor v7.2 (Simulations Plus Inc, Lancaster, CA, USA).
GastroPlus is a physiologically based pharmacokinetic (PBPK) modelling and simulation software package that simulates intravenous, oral, oral cavity, ocular, inhalation, and dermal/subcutaneous absorption, pharmacokinetics, and pharmacodynamics in human and animals. It was developed for use by the pharmaceutical industry and is licensed for use by most top 25 pharmaceutical companies in the USA and Europe. Within GastroPlus, the ACAT (Advanced Compartmental Absorption and Transit) model has been refined numerous times since its inception in 1997 to provide accurate, flexible, and powerful simulations. ADMET Predictor is used for advanced predictive modelling of ADMET properties. The "ADMET" acronym is commonly used in the pharmaceutical industry to indicate all the phenomena associated with Absorption, Distribution, Metabolism, Elimination, and Toxicity of chemical substances in the human body.

5. APPLICABILITY DOMAIN
Descriptor values: Applicability domain (OECD principle 3):
a. Domains: Defined by GastroPlus and ADMET Predictor.
i. Descriptor domain: In general, ADMET Predictor and GastroPlus apply only to small organic molecules composed of the following elements: C, N, O, S, P, H, F, Cl, Br, I, B and their isotopes. Other elements (in particular metals) are not supported. In addition, the program limits the size and complexity of input molecules to no more than 256 bonds and no more than 20 ionizable groups. All components of C.I. Solvent Blue 98 meet these GastroPlus/ADMET predictor criteria.
ii. Structural fragment domain: ADMET Predictor and GastroPlus use calculated descriptors for each chemical structure as inputs to its predictive models; it does not use structural fragments
iii. Mechanism domain: ADMET Predictor and GastroPlus models use QSAR/QSPR (quantitative structure-activity relationship/ quantitative structure-property relationship) methodology, which is
a subset of statistical-correlative modelling. It does not consider mechanisms of action, at least not explicitly.
iv. Metabolic domain: Metabolism is considered relevant and is considered in the assessment as part of the GastroPlus/ADMET predictor modeling.
b. Structural analogues: n.a.
c. Considerations on structural analogues: n.a.

6. ADEQUACY OF THE RESULT
Regulatory purpose: The predicted information is adequate to support hazard characterization (classification and labeling) as well as chemical risk assessment.
Approach for regulatory interpretation of the model result: The oral, dermal, and inhalation Fa%, F%, and Cmax of components A-F of C.I. Solvent Blue 98 are predicted by the GastroPlus QSAR program. The plasma protein binding and volume of distribution (Vd) of components A-F are predicted by ADMET Predictor. The potential metabolism and excretion of components A-F are proposed according to human CYP metabolism predicted by ADMET Predictor.

Data source

Reference

Reference Type:

study report

Title:

Unnamed

Year:

2017

Report date:

2017

Materials and methods

Objective of study:

absorption

distribution

excretion

metabolism

Test guideline

Qualifier:

no guideline followed

Version / remarks:

QSAR Prediction Reporting Format (QPRF) v 1.1

Principles of method if other than guideline:

To assess the ADME potential of the reaction mass in humans, the toxicokinetics of the major representative components (Components A-F, Text Table 1) of the mixture were estimated via the widely accepted QSAR programs, ADMET predictor (v7.2, Simulations Plus Inc, Lancaster, CA, USA) and GastroPlus (v9.0, Simulations Plus Inc, Lancaster, CA, USA).

GLP compliance:

no

Test material

Test material information

Constituent 1

Reference substance name:

Reaction mass of 1,4-bis(methylamino)anthraquinone and 1,4-bis[(2-ethylhexyl)amino]anthraquinone and 1-[(2-ethylhexyl)amino]-4-(methylamino)anthraquinone and 9,10-Anthracenedione, 1,4-bis(pentylamino)-, branched and linear and 9,10-Anthracenedione, 1-(methylamino)-4-(pentylamino)-, branched and linear and 9,10-Anthracenedione, 1-[(2-ethylhexyl)amino]-4-(pentylamino)-, branched and linear

EC Number:

911-360-1

Molecular formula:

variable structures

IUPAC Name:

Reaction mass of 1,4-bis(methylamino)anthraquinone and 1,4-bis[(2-ethylhexyl)amino]anthraquinone and 1-[(2-ethylhexyl)amino]-4-(methylamino)anthraquinone and 9,10-Anthracenedione, 1,4-bis(pentylamino)-, branched and linear and 9,10-Anthracenedione, 1-(methylamino)-4-(pentylamino)-, branched and linear and 9,10-Anthracenedione, 1-[(2-ethylhexyl)amino]-4-(pentylamino)-, branched and linear

Specific details on test material used for the study:

Chemical name: C.I. Solvent Blue 98 (3Amine)
Input for prediction: SMILES codes:
O=C2c1ccccc1C(=O)c3c2c(ccc3NC)NC (Component A)
O=C2c1ccccc1C(=O)c3c2c(ccc3NCCCCC)NC (Component B)
CCC(CCCC)CNc3ccc(NC)c2C(=O)c1ccccc1C(=O)c23 (Component C)
O=C2c1ccccc1C(=O)c3c2c(ccc3NCCCCC)NCCCCC (Component D)
CCC(CCCC)CNc3ccc(NCCCCC)c2C(=O)c1ccccc1C(=O)c23 (Component E)
CCC(CCCC)CNc3ccc(NCC(CC)CCCC)c2c3C(=O)c1ccccc1C2=O (Component F)

Results and discussion

Main ADME results

Main ADME results 1

Type:

absorption

Results:

The predicted (Fa%), (F%), and (Cmax ) values for oral, dermal, and inhalation exposures to components A-F by GastroPlus are summarized in Text Table 1. The predicted (Vd) for components A-F are also summarized in Text Table 1.

Main ADME results 2

Type:

distribution

Results:

Similar human protein binding (94.0-99.3%) and Vd values (5.5-7.74 L/kg) are predicted for all components.

Main ADME results 3

Type:

metabolism

Results:

The ADMET predicts all components will be metabolized to hydroxylated metabolites and formed aldehyde will be further metabolized to the corresponding acid by aldehyde dehydrogenase. The metabolites can be further metabolized to water soluble metabolites.

Main ADME results 4

Type:

excretion

Results:

The various conjugate forms (such as sulfates, glucuronides) of the above proposed metabolites would be more water-soluble than the parent compound; therefore, these metabolites would be excreted in urine and faeces.

Main ADME results 5

Type:

other: Accumulation

Results:

On the basis of low volume of distribution, components A-F are not expected to bioaccumulate.

Any other information on results incl. tables

Predicted value (model result):

Absorption:

The oral PSA simulation results showed that both Fa% and F% of the representative component A (Structure A2475-44-7) were impacted by particle sizes ranging from radius values of 2.5 μm to 250 μm in oral exposure. The highest Fa% and F% values were reached at the particle size of 2.5 μm in radius. Therefore, a particle size of 2.5 μm radius was selected for oral simulations of all components A-F to maximize F2% and F%. In contrast to oral PSA simulation results, the dermal PSA simulation results for Fa% and F% for the representative component A were not impact by particle sizes ranging from radius values of 2.5 μm to 250 μm. Based on the PSA simulation results on the representative component A, 2.5 μm and 25 μm radius particle sizes were applied for oral and dermal simulations, respectively, for all components A-F (Table 1).

With exposure to 1 mg/kg (oral, inhalation, and dermal) or 5 mg/kg (dermal only) exposure dose level in a fed 30-year old human (70 kg), the predicted fractional absorption (Fa%), bioavailability (F%), and maximum plasma concentration (Cmax ) values for oral, dermal, and inhalation exposures to components A-F by GastroPlus are summarized in Text Table 1. The predicted human plasma protein binding upon absorption and volume of distribution (Vd) for components A-F are also summarized in Text Table 1.

Overall, at the same exposure dose level (1 mg/kg), the predicted Cmax value from oral exposure for each component is higher than that from either dermal exposure or inhalation exposure. These data indicate that higher than 1 mg/kg dermal exposure level of each component will be needed to produce the corresponding oral Cmax level. At 1 mg/kg exposure dose level, similar Cmax values were predicted from inhalation and dermal exposure for component A-D; for components E and F, a 1 mg/kg dermal exposure had a lower Cmax than 1mg/kg by inhalation. At the dermal exposure level of 5 mg/kg, the predicted Cmax values for components B-D is almost equal to the corresponding Cmax value at 1 mg/kg oral exposure level. For compound A, a 5 mg/kg dermal exposure yields a much higher Cmax than a 1mg/kg oral exposure, whereas for compounds E and F, 5 mg/kg dermal exposure yields a much lower Cmax value than a 1 mg/kg oral exposure. Over the 168 hour exposure period, the dermal AUC values depended on the specific component as some dermal AUC values at 5 mg/kg were greater (compounds A-D), less (compounds E-F) than oral AUC values at 1 mg/kg. Inhalation AUC values at 1 mg/kg showed the same relationship to the 5 mg/kg dermal ADUC values except for component E which had a lower inhalation AUC.

Distribution:

The predicted human plasma protein binding upon absorption and volume of distribution (Vd) for components A-F are also summarized in Table 1. Similar human protein binding (94.0-99.3%) and Vd values (5.5-7.74 L/kg) are predicted for all components. Higher protein binding values generally indicate lower bioavailability to interact with other target sites. The volume of distribution for components A-F in humans was estimated to is low (5.50 L/kg to 7.74 L/kg), which indicates low distribution to body tissues.

Accumulation:

On the basis of low volume of distribution, components A-F are not expected to bioaccumulate.

Metabolism:

Based on the metabolism prediction by ADMET predictor, all components will be metabolized to hydroxylated metabolites by human CYP 1A2 or

CYP 3A4 or by both human CYP 1A2 and CYP 3A4 and the formed aldehyde will be further metabolized to the corresponding acid by aldehyde dehydrogenase (ALDH). The formed metabolites can also be further metabolized to water soluble metabolites (such as glucuronides and sulfates), which will be mainly excreted into urine and feces.

Excretion:

The various conjugate forms (such as sulfates, glucuronides) of the above proposed metabolites would be more water-soluble than the parent compound; therefore, these metabolites would be excreted in urine and faeces.

Applicant's summary and conclusion

Conclusions:

Overall, at the same exposure dose level (1 mg/kg), the predicted Cmax value from oral exposure for each component is higher than that from either dermal exposure or inhalation exposure. These data indicate that higher than 1 mg/kg dermal exposure level of each component will be needed to produce the corresponding
oral Cmax level. At 1 mg/kg exposure dose level, similar Cmax values were predicted from inhalation and dermal exposure for component A-D; for components E and F, a 1 mg/kg dermal exposure had a lower Cmax than 1mg/kg by inhalation. At the dermal exposure level of 5 mg/kg, the predicted Cmax values for components B-D is almost equal to the corresponding Cmax value at 1 mg/kg oral exposure level. For compound A, a 5 mg/kg dermal exposure yields a much higher Cmax than a 1mg/kg oral exposure, whereas for compounds E and F, 5 mg/kg dermal exposure yields a much lower Cmax value than a 1 mg/kg oral exposure. Over the 168 hour exposure period, the dermal AUC values depended on the specific component as some dermal AUC values at 5 mg/kg were greater (compounds A-D), less (compounds E-F) than oral AUC values at 1 mg/kg. Inhalation AUC values at 1 mg/kg showed the same relationship to the 5 mg/kg dermal ADUC values except for component E which had a lower inhalation AUC. Similar human protein binding (94.0-99.3%) and volume distribution (Vd) values (5.5-7.74 L/kg) are predicted for all components.
Based on the metabolism prediction by ADMET predictor, all components will be metabolized to hydroxylated metabolites by human CYP 1A2 or CYP 3A4 or by
the combination of human CYP 1A2 and CYP 3A4. The formed aldehyde will be further metabolized to the corresponding acid by aldehyde dehydrogenase (ALDH). The formed metabolites can also be further metabolized to water soluble metabolites (such as glucuronides and sulfates), which will be mainly excreted into urine and feces.
On the basis of low volume of distribution, and predicted metabolism and excretion of all components, it can be predicted that C.I. Solvent Blue 98 is not expected to bioaccumulate in humans.

Executive summary:

C.I. Solvent Blue 98 is a reaction mass containing six major components (components A-F). Experimental data on absorption, distribution, metabolism and excretion (ADME) are not available for this mixture. To assess the ADME potential of the reaction mass in humans, the toxicokinetics of the major representative components (Components A-F, Text Table 1) of the mixture were estimated via the widely accepted QSAR programs, ADMET predictor (v7.2, Simulations Plus Inc, Lancaster, CA, USA) and GastroPlus (v9.0, Simulations Plus Inc, Lancaster, CA, USA).

With exposure to 1 mg/kg (oral, inhalation, and dermal) or 5 mg/kg (dermal only) exposure dose level in a fed 30-year old human (70 kg), the predicted fractional absorption (Fa%), bioavailability (F%), and maximum plasma concentration (Cmax ) values for oral, dermal, and inhalation exposures to components A-F by GastroPlus are summarized in Text Table 1. The predicted human plasma protein binding upon absorption and volume of distribution (Vd) for components A-F are also summarized in Text Table 1.

Overall, at the same exposure dose level (1 mg/kg), the predicted Cmax value from oral exposure for each component is higher than that from either dermal exposure or inhalation exposure. These data indicate that higher than 1 mg/kg dermal exposure level of each component will be needed to produce the corresponding oral Cmax level. At 1 mg/kg exposure dose level, similar Cmax values were predicted from inhalation and dermal exposure for component A-D; for components E and F, a 1 mg/kg dermal exposure had a lower Cmax than 1mg/kg by inhalation. At the dermal exposure level of 5 mg/kg, the

predicted Cmax values for components B-D is almost equal to the corresponding Cmax value at 1 mg/kg oral exposure level. For compound A, a 5 mg/kg dermal exposure yields a much higher Cmax than a 1mg/kg oral exposure, whereas for compounds E and F, 5 mg/kg dermal exposure yields a much lower Cmax value than a 1 mg/kg oral exposure. Over the 168 hour exposure period, the dermal AUC values depended on the specific component as some dermal AUC values at 5 mg/kg were greater (compounds A-D), less (compounds EF) than oral AUC values at 1 mg/kg. Inhalation AUC values at 1 mg/kg showed the same relationship to the 5 mg/kg dermal ADUC values except for component E which had a lower inhalation AUC. Similar human protein binding (94.0-99.3%) and volume distribution (Vd) values (5.5-7.74 L/kg) are predicted for all components.

Based on the metabolism prediction by ADMET predictor, all components will be metabolized to hydroxylated metabolites by human CYP 1A2 or CYP 3A4 or combination of human CYP 1A2 and CYP 3A4. The formed aldehyde will be further metabolized to the corresponding acid by aldehyde dehydrogenase (ALDH). The formed metabolites can also be further metabolized to water soluble metabolites (such as glucuronides and sulfates), which will be mainly excreted into urine and feces.

On the basis of low volume of distribution, and predicted metabolism and excretion of all components, it can be predicted that C.I. Solvent Blue 98 is not expected to bioaccumulate in humans.