Registration Dossier

Administrative data

Description of key information

Key value for chemical safety assessment

Skin sensitisation

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not sensitising)
Additional information:

Based on these data, and according to the guidance provided by the European Centre for Ecotoxicology and Toxicology of Chemicals (ECETOC), the substance produced effects that could indicate it is a weak dermal sensitizer. However, the weight-of-evidence evaluation supports the conclusion that it should not be classified for dermal sensitisation.


The weight-of-evidence evaluation supports the conclusion that acetylated lecithin should not be classified as a dermal sensitizer (Xi, R43). Although the substance produced a weak positive response in the mouse local lymph node assay (LLNA) (DuPont, 2009), the following evidence supports the assessment that this is likely a false positive, and that the substance is not a human hazard for this endpoint and should not be classified.


The EC3value for the substance was 45% (DuPont, 2009). This value is well above the cut-off values used to classify a substance in the weakest ECB or ECETOC category: ECB moderate sensitizer (>2.0%) or ECETOC weak sensitizer (>10%) (Loveless et al., 2009). ECB does not recognize a cut-off EC3value for not classifying dermal sensitizers based on potency. However, it is recognized that false positives can occur with this assay, as with any assay (Basketter et al., 2009a). 


There are numerous examples of substances for which there are discordant results between the LLNA and guinea pig maximization test (GPMT) (Kreiling et al., 2008; Basketter et al, 2009b). These include examples of substances that have tested positive in the LLNA with a strong history of safe human exposure, such as sodium lauryl sulfate (Basketter, 2009a). A study was conducted in which the responses of eight lipids were compared between the LLNA and the GPMT (Kreiling et al., 2008). Three of the tested lipids (oleic, linolenic, and linolenic acids) are components of lecithins, including the acetylated lecithin substance tested to satisfy REACH data requirements. All tested substances in this study were naturally occurring biological substances, similar to lecithin. Seven of these lipids, including the three present in acetylated lecithin, gave positive responses in the LLNA. However, two of the three substances (oleic and linolenic acids; EC311% and 10%, respectively) were considered negative in the GPMT, while the third substance (linoleic acid; EC314%) produced inconclusive GPMT results. A literature search did not find any case reports of human skin sensitisation for any of these three substances, despite widespread exposure. No skin sensitisation was observed in human patch tests with oleic acid. The authors concluded that it would be scientifically inappropriate to label and classify these substances as dermal sensitizers, despite the positive LLNA response. 


In a separate study, oleic acid and oleic acid ester were tested in the GPMT and gave 0% positive responses, yet both produced positive responses in the LLNA, with EC3values lower than that of acetylated lecithin (21% and 27%, respectively). Both of these substances were considered by the authors to be non-classified for dermal sensitisation by the weight-of-evidence judgment (Basketter, 2009b). 


The fatty acids giving discordant results in the above studies are significant components of the acetylated lecithin tested to meet REACH data requirements, as well as the natural lecithin from which it is manufactured. Under REACH, additional animal testing to clarify the acetylated lecithin LLNA result should not be carried out to evaluate whether acetylated lecithin would produce a similar discordant response. However, the positive LLNA response observed with acetylated lecithin is considered to be a false positive, likely triggered by one or more of these fatty acids. Although the LLNA is a validated test for dermal sensitisation, the original validation study did not include any fatty acids (Basketter et al., 2009b). Therefore, these fatty acids may be outside the domain of applicability for the LLNA.


Natural lecithin (non-chemically modified; CAS# 8002-43-5) is a naturally occurring material found in living plants and animals and is a major component of cell membranes. Commercial lecithin is a complex material prepared from numerous sources, including soybeans. It consists of acetone-insoluble phosphatides (in particular phosphatidylcholine, phosphatidylethanolamine, and phosphatidylinositol), combined with triglycerides (primarily containing oleic, stearic, palmitic, linolenic, and linoleic acids), carbohydrates, and a minor amount of free fatty acids (Fiume, 2001).


Natural lecithin has been exempted from REACH legislation under Annex IV (Official Journal of the European Union, 2008) and has been considered to be not a skin sensitizer, based on expert judgment (European Commission, 2008). It is used in a wide variety of products with significant human dermal contact, including foods, cosmetics, personal care products, and transdermal drug delivery systems (e.g., liposomes, microemulsions). These products have a history of being well-tolerated following dermal exposure. Animal data to assess dermal sensitisation is not reported for these products, but there are reports of human patch testing with selected products, and results do not support the potential to produce dermal sensitisation (Fiume, 2001).


Acetylated lecithin (CAS# 91053-50-8) is classified as an unknown or variable composition biological substance (UVCBS). It is manufactured by reacting lecithin with acetic anhydride, leading to acetylation of the amine group of the phosphatidylethanolamine component of lecithin. This modification is quantitative, rather than qualitative, as a small amount of naturally occurring N-acylphosphatidylethanolamine (APE) is present in the natural lecithin. Analysis of the acetylated lecithin (the batch tested to meet REACH data requirements) and the natural lecithin used in its manufacture demonstrated an increase in phosphatidylethanolamine from 2.22% to 6.02% in the content of APE. Acetic anhydride is not classified for skin sensitisation under EU or GHS classifications, and is not detected in the final product (<0.1%). Therefore, it would not be expected to contribute to dermal sensitisation by exposure to acetylated lecithin.


Acetylated lecithin is not used in personal care products and transdermal drug delivery systems. However, acetylated lecithin would be expected to behave similarly to natural lecithin. As discussed above, the difference between acetylated and natural lecithin is the level of N-acylphosphatidylethanolamine (APE) component of this substance. The relatively small increase in the level of this component from 2.22% to 6.02% would not be expected to change the dermal sensitisation response. 


Since acetylated lecithin is a complex mixture, and not a single chemical entity, it cannot be evaluated for structural alerts for dermal sensitisation. However, key components of the substance were evaluated. L-phosphatidylcholine, L-phosphatidylethanolamine and N-acylphosphatidylethanolamine were processed through two systems to investigate their skin sensitisation potential. Representative two-dimensional structures were drawn for these 3 substances using a C18 unsaturated acid as the nominal fatty acid in each case (specific structures are provided in the QPRFs which are attached in the sensitisation endpoint summary in IULCID5). The systems applied included the OECD Toolbox v2 (beta) which contains a protein binding profiler and the sensitisation model contained within the TIMES v2.26.3 platform (LMC, 2009). The OECD Toolbox provides the capability of grouping chemicals on the basis of their presumed similarity. This similarity can be characterised with respect to chemical mechanisms, modes of action as well as structural similarity. The rate determining step for sensitisation induction is assumed to be the stable association between the skin protein nucleophile and the chemical electrophile, the interaction being a covalent one (Roberts et al., 2008). Efforts to predict skin sensitizers rely on the identification of electrophilic features. The OECD Toolbox contains a profiler named “Protein binding by OASIS” which is an implementation of many of the known skin sensitisation structural alerts. Within the OECD Toolbox, there are 52 alerts for skin sensitisation. Applying the protein binding profiler to the three representative substances revealed all to be “non-binding” (i.e. none of the substances, the acetylated lecithin or lecithin surrogates, identified any structural features that would be indicative of skin sensitisation potential; documentation is attached in the sensitisation endpoint summary in IUCLID5). The consistency across the three substances processed is suggestive of their similarity with respect to the skin sensitisation endpoint.


The skin sensitisation model within the TIMES platform was also used to make predictions of skin sensitisation model. The hybrid expert system TIssue MEtabolism Simulator for Skin Sensitization (TIMES-SS) encodes structure toxicity and structure metabolism relationships through a number of transformations simulating skin metabolism and interaction of the generated reactive metabolites with skin proteins. The skin metabolism simulator mimics metabolism using two-dimensional structural information. Metabolic pathways are generated based on a set of 236 hierarchically ordered principal transformations including spontaneous reactions and enzyme catalyzed reactions (phase I and II). The covalent reactions with proteins are described by 47 alerting groups (structural alerts). Some of these alerts are additionally underpinned by mechanistically based 3D-QSARs to refine the predictions. These 3D-QSAR models depend on both the structural alert and factors that influence its reactivity - steric effects, molecular size, shape, solubility, lipophilicity and electronic properties. More information on the TIMES skin sensitisation is provided in the corresponding QMRF.


The three substances were processed through TIMES and all were found to be potential weak sensitizers by virtue of autoxidation across the double bond in the unsaturated fatty acid side chain (i.e. air oxidation), resulting in the formation of hydroperoxides. Hydroperoxides are known to easily form radicals by cleavage of their labile O–O bond. The activity of hydroperoxides is likely to involve radical reactions either by direct formation of covalent bonds with proteins by radical coupling or, as in the case of allylic hydroperoxides, radical rearrangements to form epoxides acting as electrophilic haptens. More information about autoxidation and its relevance for skin sensitisation is discussed in Karlsberg et al, 2008. All three representative structures identified the same alert for sensitisation. Accordingly on the basis of these two model estimates, one would not expect any differences between the acetylated and non acetylated lecithins with respect to the skin sensitisation endpoint.


Thus the weight of evidence supports that acetylated lecithin should not be classified as a dermal sensitizer, despite the weak positive LLNA response for the following reasons:


  • The minimal difference between acetylated lecithin and the starting material, natural lecithin.
  • The lack of structural alerts for dermal sensitisation with N-acylphosphatidylethanolamine, the only significantly different component in the acetylated substance.
  • The presence of oleic acid and linolenic acid (substances that produced LLNA responses considered likely to be false positives) as components of the acetylated lecithin.
  • The history of safe use of the starting material (natural lecithin) in products with significant human dermal exposure.




Basketter DA, McFadden JF, Gerberick F, Cockshott A, and Kimber I (2009a). Nothing is perfect, not even the local lymph node assay: a commentary and the implications of REACH. Contact Derm., 60:65-69.


Basketter D, Ball N, Cagen S, Carrillo J-C, Certa H, Eigler D, Garcia C, Esch H, Graham C, Haux C, Kreiling R, and Mehling A (2009b). Application of a weight of evidence approach to assessing discordant sensitisation datasets: Implications for REACH. Regul. Toxicol. Pharmacol., 55:90-96.


DuPont Haskell (2009). H-29216: Local Lymph Node Assay in Mice. Unpublished data. DuPont-18511-1234.


FiumeMZ (2001). Final report on the safety assessment of lecithin and hydrogenated lecithin. Int. J. Toxicol., 20(Suppl. 1):21-45.


Karlberg A-T, Bergström MA, Börje A, Luthman K, and Nilsson JLG (2008). Allergic Contact Dermatitis––Formation, Structural Requirements, and Reactivity of Skin Sensitizers. Chem. Res. Toxicol., 21(1):53-69.


Kreiling R, Hollnagel HM, Hareng L, Eigler D, Lee MS, Griem P, Dreeβen B, Kleber M, Albrecht A, Garcia C, and Wendel A (2008). Comparison of the skin sensitizing potential of unsaturated compounds as assessed by the murine local lymph node assay (LLNA) and the guinea pig maximization test (GPMT). Food Chem. Toxicol., 46:1896-1904.


Loveless SE, Api A-M, Crevel RWR, Debruyne E, Gamer A, Jowsey IR, Kern P, Kimber I, Lea L, Lloyd P, Mehmood Z, Steiling W, Veenstra G, Woolhiser M, and Hennes C (2010). Potency values from the local lymph node assay: Application to classification, labeling, and risk assessment. Regul. Toxicol. Pharmacol., 56:54-66.


Official Journal of the European Union. (2008) L 268/13, 2008; Commission Regulation (EC) No. 987/2008


European Commission. (2008). DG Environmental Report, Review of Annex IV of the Regulation No. 1907/2006 (REACH) Appendix 2, Evaluation of existing entries, pp. 135-136.


Roberts DW, Aptula AO, Patlewicz GY, and Pease C (2008). Chemical Reactivity Indices and Mechanism-based read across for non-animal based assessment of skin sensitization potential. J. Appl. Toxicol., 28(4):443-454.

Migrated from Short description of key information:
EC = 45%. Stimulation index >3 observed with 50% and 100% dosing formulations. Reliability = 1.
Based on these data, and according to the guidance provided by the European Centre for Ecotoxicology and Toxicology of Chemicals (ECETOC), the substance produced effects that could indicate it is a weak dermal sensitiser. However, the weight-of-evidence evaluation supports that it should not be classified for dermal sensitisation.

Respiratory sensitisation

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed (not sensitising)
Additional information:

There is no specific information available for the respiratory tract sensitisation of this material. However, acute inhalation exposures in rats to this material did not show signs of respiratory sensitisation.

Migrated from Short description of key information:
Not specifically tested for this endpoint. However, no adverse respiratory tissue effects observed in acute inhalation study, and unlikely to have exposure to respirable atmosphere of substance due to its viscosity.

Justification for classification or non-classification

The test substance was not considered to be a dermal sensitizer based on a weight-of-evidence evaluation, and there was no evidence of respiratory sensitisation during inhalation exposures. The substance does not need to be classified for sensitisation according the EU Directive 67/548/EEC and EU Classification, Labelling and Packaging of Substances and Mixtures (CLP) Regulation (EC) No. 1272/2008.