Octanoic acid, monoester with 1,2-propanediol

Skin sensitisation

Administrative data

Endpoint:

skin sensitisation: in vitro

Type of information:

(Q)SAR

Adequacy of study:

key study

Study period:

2021

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:

1. SOFTWARE
Various tools, inter alia CASE Ultra, DEREK Nexus, OECD Toolbox, ToxRead, Toxtree, VEGA

2. MODEL (incl. version number)

Model Version Approach
CASE Ultra 1.8.0.2 Statistical/Expert
DEREK Nexus 6.0.1 Expert
OECD (Q)SAR Toolbox 4.4.1 Expert/Read-across
ToxRead 0.23beta Read-across
Toxtree 3.1.0 Expert
VEGA 1.1.5_b36 Hybrid

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
Main components of the substance:

1) CCCCCCCC(=O)OCC(C)O
2) CCCCCCCC(=O)OC(C)CO
3) CCCCCCCC(=O)OCC(C)OC(=O)CCCCCCC

Impurities (=educts) of the substance
4) CC(O)CO
5) OCCCO
6) CCCCCCCC(=O)O

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
See attached QMRF/ QPRF Report

5. APPLICABILITY DOMAIN
See attached QMRF/ QPRF Report

6. ADEQUACY OF THE RESULT
See attached QMRF/ QPRF Report

Cross-reference

Reason / purpose for cross-reference:

reference to same study

Reference

Endpoint:

skin irritation / corrosion, other

Remarks:

QSAR

Type of information:

(Q)SAR

Adequacy of study:

key study

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:

1. SOFTWARE
Various tools, inter alia CASE Ultra, DEREK Nexus, OECD Toolbox, ToxRead, Toxtree, VEGA

2. MODEL (incl. version number)

Model Version Approach
CASE Ultra 1.8.0.2 Statistical/Expert
DEREK Nexus 6.0.1 Expert
OECD (Q)SAR Toolbox 4.4.1 Expert/Read-across
ToxRead 0.23beta Read-across
Toxtree 3.1.0 Expert
VEGA 1.1.5_b36 Hybrid

3. SMILES OR OTHER IDENTIFIERS USED AS INPUT FOR THE MODEL
Main components of the substance:

1) CCCCCCCC(=O)OCC(C)O
2) CCCCCCCC(=O)OC(C)CO
3) CCCCCCCC(=O)OCC(C)OC(=O)CCCCCCC

Impurities (=educts) of the substance
4) CC(O)CO
5) OCCCO
6) CCCCCCCC(=O)O

4. SCIENTIFIC VALIDITY OF THE (Q)SAR MODEL
See attached QMRF/ QPRF Report

5. APPLICABILITY DOMAIN
See attached QMRF/ QPRF Report

6. ADEQUACY OF THE RESULT
See attached QMRF/ QPRF Report

Principles of method if other than guideline:

- Principle of test, short description of test conditions and Parameters analysed / observed
CASE Ultra Statistical/Expert
DEREK Nexus Expert
OECD (Q)SAR Toolbox Expert/Read-across
Toxtree Expert

GLP compliance:

no

Test system:

other: QSAR prediction

Remarks on result:

no indication of irritation

Remarks:

QSAR prediction

Compound	Method				Final Conclusion***
	Statistical / CASE Ultra		Rule-based
	Irritation	Corrosion	Toxtree	OECD Toolbox
2-hydroxypropyl octanoate	Inconclusive	Inconclusive	Irritating*	No alerts	Inconclusive
Propylene glycol 2-caprylate	Negative	Negative	Irritating*	No alerts	Inconclusive
Propylene-1,2-dioctanoate	Negative	Negative	Irritating*	No alerts	Inconclusive
1,2 propanediol	Inconclusive**	Inconclusive	Irritating*	No alerts	Negative
1,3 propanediol	Positive**	Inconclusive	Irritating*	No alerts	Negative
Caprylic acid	Positive	Positive	Irritating*	No alerts	Positive Corrosive

* Dismissed due to alerts out of context
** Negative study results in database

*** Expert opinion based on structures only, and not taking into account educt or other substance properties

Interpretation of results:

study cannot be used for classification

Conclusions:

PGMC is finally concluded to be NON SKIN IRRITANT and NOT SKIN CORROSIVE.

Executive summary:

Irritation and Corrosion

None of the query structures have an alert for skin and eye irritation from the rule-based expert system DEREK.

OECD Toolbox contains profiler for skin/eye irritation based on the exclusion/inclusion rules by BfR. The exclusion rules for eye/skin irritation/corrosion are based on physico-chemical cut-off values to identify chemicals that do not exhibit eye/skin irritation or corrosion potential. The parameters used for defining skin/eye irritation rules are: lipid solubility, surface tension (only skin), octanol water partition coefficient, aqueous solubility, melting point and molecular weight. Since none of the query structures met all the cut-off values Toolbox concluded the results to be undefined. The profiler with inclusion rules by BfR contains structural alerts which can be used for positive classification of chemicals causing irritation/corrosion. However, none of the query structures has such an alert or passed the exclusion rules. Additional read-across with OECD Toolbox failed because no structural similar chemicals with study results on irritation and corrosion could be obtained from the Toolbox database.

Almost the same physicochemical property limits and structural rules are used by the rule-based expert system Toxtree for estimating skin/eye irritation/corrosion (see,Table7). All query structures belong to Toxtree Group C (only molecules with CxHyOz) and have obviously not passed any the physicochemical limit values for being excluded from skin corrosion or skin irritation and are then considered to potentially cause skin corrosion or skin irritation (see,Table8)[1]. It has however to be noted that if any of the physicochemical parameter is missing, Toxtree will skip the physicochemical rules and only the structural alerts will be applied, which may result in a low quality prediction. This is in contrast to the inclusion and exclusion rules in OECD Toolbox, which consider physicochemical parameter and structural alert separately. Since no information were available for the query structures while all others were estimated using EpiSuite and US EPA Test software (see, appendix), the predictions by Toxtree are considered to be of low confidence. Moreover, the structural alerts between the OECD profiler and the structural classes of Toxtree are also different and therefore explaining the different results. 2-hydroxypropyl octanoate and propylene glycol 2-caprylate belong to the structural Toxtree class of C₁₀–C₂₀aliphatic alcohols which are considered potential skin irritants. This class is however not included in the inclusion rules of the OECD profiler. The Draft OECD SIDS Initial Assessment Report C6-22 primary aliphatic alcohols concludes that linear aliphatic alcohols in the range C6 – C11 are mild irritants, not anticipated to be corrosive. Aliphatic alcohols in the range C12 – C16 have a low degree of skin irritation potential; alcohols with chain lengths of C18 and above are non-irritant to skin. For the essentially linear alcohols, a differentiation in the irritation similar to that of the linear alcohols can be identified. For the alcohols in the C6- C11 range the skin irritation potential can be categorised as mild - irritant. The skin irritation potential for the higher members of the essentially linear alcohols in the range C12 – C16 is mild - essentially non-irritant. In addition, human data indicate that the irritation responses for the category of the linear alcohols are of a lower order than those observed in rabbits‎[12]. Therefore, the prediction by Toxtree does not validate well. Taking into account the absence of the structural class of aliphatic alcohols in the inclusion rules of the OECD Toolbox profiler, suggests no confidence in the prediction by Toxtree and the prediction are therefore dismissed. Taking into account the negative prediction by CASE Ultra for propylene glycol 2-caprylate and an inconclusive result for 2-hydroxypropyl octanoate (see,Table9), there no supporting evidence that skin irritation can be excluded, although the assessments in the OECD SIDS report suggests aliphatic alcohols to be only mild irritants.

Therefore, the outcome of the predictions for SKIN IRRITATIONof 2-hydroxypropyl octanoate and propylene glycol 2-caprylate can only be considered INCONCLUSIVE.

Toxtree assigned propylene-1,2-dioctanoate, 1,2 propanediol and 1,3 propanediol and caprylic acid to the structural class ethylene glycol ethers which are considered potential skin irritants. None of these structures is however a glycol ether. While an explanation including SMARTS definition could be obtained from the OECD Toolbox profiler (see,Figure8), the mapping in Toxtree indicates the structural alert not properly defined (see,Figure9). Overall, the structural alert of Toxtree is out of context and the predictions can therefore be dismissed.

However, the report conclusions for the educt substances 1,2/ 1,3-propanediol are clear negative and the for the caprylic acid clear positive. The latter is considered to be based on the short C-chain length and the pka of 4,89.

 

Taken into account the very low water solubility, the natural pH of the skin and the good metabolization of the substance PGMC, the substance is finally concluded to be NON SKIN IRRITANT and NOT SKIN CORROSIVE.

 

 

 

Data source

Reference

Reference Type:

study report

Title:

Unnamed

Year:

2021

Report date:

2021

Materials and methods

Principles of method if other than guideline:

- Principle of test:, Short description of test conditions, Parameters analysed / observed:
CASE Ultra LLNA
CASE Ultra Non- LLNA
CAESAR
IRFMN/JRC

Consensus score for being sensitizer and non-sensitizer

GLP compliance:

no

Justification for non-LLNA method:

QSAR study

Test material

Test material information

Constituent 1

Reference substance name:

Octanoic acid, monoester with 1,2-propanediol

EC Number:

953-046-7

Cas Number:

31565-12-5

Molecular formula:

not applicable to MCS

IUPAC Name:

Octanoic acid, monoester with 1,2-propanediol

Test material form:

liquid: viscous

In vitro test system

Details of test system:

other: QSAR modelling

Details on the study design:

- Prediction model used:
CASE Ultra LLNA
CASE Ultra Non- LLNA
CAESAR
IRFMN/JRC

Results and discussion

In vitro / in chemico

Results

Remarks on result:

no indication of skin sensitisation

Remarks:

QSAR prediction

Outcome of the prediction model:

negative [in vitro/in chemico]

Any other information on results incl. tables

Consensus results for skin sensitisation from statistical-based models with reliability score s^+/-and weighted reliability score s^+/-·w (see, appendix) and available study results from training sets of models

Model	Pred result (+/-)	s^+/-	Exp result (+/-)	s^+/-·w	Pred result (+/-)	s^+/-	Exp result (+/-)	s^+/-·w
	2-hydroxypropyl octanoate				1,2 propanediol
CASE Ultra LLNA	-	3		2.7	-	3	-	3
CASE Ultra Non- LLNA	-	2		1.2	-	2	-	3
CAESAR	+	2		1.2	-	3	-	3
IRFMN/JRC	-	2		1.2	-	3	-	3
Consensus score Sensitiser				0.10				0.00
Consensus score Non-Sensitiser				0.43				1.00
	Propylene glycol 2-caprylate				1,3 propanediol
CASE Ultra LLNA	-	3		2.7	-	1		0.2
CASE Ultra Non- LLNA	Inconclusive				-	2		1.2
CAESAR	+	2		1.2	-	3		2.7
IRFMN/JRC	-	2		1.2	+	1		0.2
Consensus score Sensitiser				0.13				0.02
Consensus score Non-Sensitiser				0.43				0.34
	Propylene-1,2-dioctanoate				Caprylic acid
CASE Ultra LLNA	-	2		1.2	Inconclusive
CASE Ultra Non- LLNA	-	1		0.2	-	2	-	3
CAESAR	+	3		2.7	-	3	-	3
IRFMN/JRC	+	3		2.7	-	3	-	3
Consensus score Sensitiser				0.45				0.00
Consensus score Non-Sensitiser				0.12				1.00

Applicant's summary and conclusion

Interpretation of results:

GHS criteria not met

Conclusions:

All three main consitutents of PGMC, namely 2-hydroxypropyl octanoate, propylene glycol 2-caprylate, propylene-1,2-dioctanoate, 1,2 propanediol, 1,3 propanediol and caprylic acid are considered NON-SENSITISER. Also all impurities are considered NON-SENSITISER.

Executive summary:

Skin Sensitisation

No alert for skin sensitisation was identified for the six query compounds with DEREK and the expert system concluded skin sensitisation in mammal is NON-SENSITISER (Table2). Moreover, no misclassified or unclassified features were detected which indicates confidence in the predictions.

None of the six query compounds triggered an alert from the protein binding profiler for skin sensitisation in OECD Toolbox. Negative study results for skin sensitisation of 1,2 propanediol, 1,3 propanediol and caprylic acid were included in the OECD Toolbox database. Read-across was performed for 2-hydroxypropyl octanoate, propylene glycol 2-caprylate and propylene-1,2-dioctanoate. According to the profiling result, 2-hydroxypropyl octanoate may act as prohapten with one metabolite triggering a structural alert for ketones which could interact with proteins via nucleophilic addition (see,Figure3). The profiler refers to a 3rd party expert stating that simple ketones are usually too weakly reactive to sensitise unless log K_owis very high and which was taken into account in the TIMES SS model by defining a threshold of log K_ow> 4 for weak skin sensitizers. This is substantiated by the category members used for read-across. All four analogues are likely to be subject of biotic transformation to simple ketones and have a log K_ow< 4, except 2-undecanol which is reported to be a weak sensitiser, while another study result indicates the substance to be not sensitising. This indicates that the log K_owof 2-hydroxypropyl octanoate and its ketone metabolite respectively, is too low to sensitise.