Registration Dossier

Administrative data

Description of key information

Two in vivo skin sensitisation studies (Toxikon, 2013 and Harlan, 2009) in guinea pigs conducted according to the Buehler Test protocol together with information on workers form a weight of evidence for the skin sensitising potential of trimethoxy(methyl)silane. In the first test conducted by Harlan (2009) the test substance was found to be sensitising. However, in a more recent test the result was negative (Toxikon, 2013).

Two trimethoxy(methyl)silane manufacturers have provided statements which confirm that there have been no cases of skin sensitisation amongst workers during more than 20 years of manufacture (see IUCLID Section 7.10.4) (Wacker, 2017; Momentive, 2017).

Key value for chemical safety assessment

Skin sensitisation

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

Two recent studies conducted according to the appropriate OECD guideline were available for review for skin sensitisation (Harlan 2009, Toxikon 2013). Methodological problems were evident in the study by Harlan (2009). The weight of evidence from the skin sensitisation study by Toxikon (2013) and Harlan (2009) support the conclusion of the registered substance not being a sensitiser.

The study by Toxikon (2013) was conducted according to an appropriate OECD test guideline, with acceptable restrictions. The restriction was that the starting concentration selected for induction was not the highest to cause mild-to-moderate skin irritation as required by OECD test guideline 406. The highest, non-irritating concentration used in the preliminary test (test article and PEG at 50% dilution) was selected for use in the induction and challenge phases. No positive skin reactions were seen in any of the test animals, and appropriate responses were evident in all the control animals. The study concluded that the test material was not sensitising to guinea pig skin.

A second study for skin sensitisation found the test material sensitising in a study conducted according to an OECD 406 and in compliance with GLP (Harlan 2009). The results of the study are deemed equivocal as a result of methodological problems; the irritation potential reported is inconsistent between experiments and the negative control animals were seen to have positive skin reactions. The study is assigned reliability 2 on the basis that the methodological issues reduce its reliability.

The details of the study by Harlan (2009) are summarised as follows; trimethoxy(methyl)silane at 25% in PEG 300 was found non-irritating in two prescreen experiments as well as in the second control group. Skin reactions were observed in the test and control group in the first challenge when treated at 25% in PEG 300. At second challenge, local skin reactions were observed in test animals when the test substance was applied at the concentration of 25% in PEG 300 but not at the item concentration of 15% in PEG 300. The presence of skin reactions of grade 1 in 30% and 20% of the test animals after 24 and 48h respectively in the second challenge and absence of any evidence of irritation in the second control group support the conclusion that the skin reactions in the test animals are indicative of sensitization. The equivocal first challenge data followed by positive re-challenge data, lead the study authors to the overall conclusion that the test material is sensitising. However, the reviewer considers the result ambiguous due to the methodological issues in the study.

It is also noted that the the study by Toxikon (2013) which found the test material not sensitising, used a higher concentration of test material (50% at induction and challenge doses), than the study by Harlan (2009) which concluded the test material to be a sensitiser. This adds to the weight of evidence that the result from Toxikon (2013) is a more accurate representation of the true sensitisation potential of the registered substance.

In addition to the in vivo experimental animal data, two producers of trimethoxy(methyl)silane have provided statements which confirm that there have been no cases of skin sensitisation amongst workers during more than 20 years of manufacture (Wacker, 2017; Momentive, 2017).

These data included internal information from: (1) relevant plants (including the number of employees and exposure descriptions as well as medical surveillance and regular health checks (especially concerning skin status) already performed on employees; (2) information from the Network of Departments of Dermatology for the surveillance and scientific evaluation of contact allergies; (3) information from an Employer's liability insurance association (BG Bau); (4) information from customers and (5) a comprehensive literature search. This information demonstrates that there have been no cases of skin sensitisation during more than 20 years of production, handling and use of trimethoxy(methyl)silane.

There is therefore no evidence that this substance causes skin sensitisation under relevant exposures in workers.

There is also a question regarding the reliability of the LLNA for silicon-based substances, which means that conducting such a study will not add weight to the database for a positive or negative outcome and therefore will not make any contribution to the protection of human health. The current accepted and preferred method for skin sensitisation testing according to the REACH legislation (EC no 1907/2006) and CLP Regulation (EC No 1272/2008) is the murine local lymph node assay (LLNA). A validated test method, OECD TG 429 (OECD 2002) is available for the LLNA. The guideline acknowledges the limits of the LLNA, and states that there are instances where ‘test substance classes or substances containing functional groups shown to act as potential confounders’ make the use of guinea pig tests more appropriate. It is concluded that the LLNA is not applicable where the properties of the test material cause interference in the accuracy of the LLNA (OECD 2002). The statement in the OECD TG 429 is given with reference to the findings of Basketter et al. (2009a), who demonstrated false positives in silicone based substances which had previously been demonstrated to be non-sensitisers in the guinea pig maximisation test (GPMT). The importance of available evidence from guinea pig results, consideration of chemical reactivity, epidermal bioavailability and clinical and experimental human data are emphasised as central to reaching appropriate regulatory decisions for substances which have been shown to fall outside the specificity of the LLNA (Basketter et al., 2009b). The non-applicability of the LLNA for silicone based substances has also been demonstrated by Petry et al. (2012). The sensitisation potential of polyfunctional silicone materials was tested in a comparative study investigating the GPMT and the LLNA assays, which found the five tested substances to be negative in the GPMT whereas they were concluded to be weak to moderate skin sensitisers in the LLNA (Petry et al., 2012).


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


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


Petry, T., Bosch, A., Coste, X., Dupuis, V., Eigler, D., Germain, P. (2012). An assessment of the skin sensitisation hazard of a group of polyfunctional silicones using a weight of evidence approach. Regulatory Toxicology and Pharmacology, 64, 305-314.

Respiratory sensitisation

Endpoint conclusion
Endpoint conclusion:
no study available

Justification for classification or non-classification

The weight of evidence from two in vivo skin sensitisation studies in guinea pigs conducted according to the Buehler Test protocol, together with over 20 years of occupational medical data, literature search results and data from an insurance association, suggests that trimethoxy(methyl)silane does not require classification according to Regulation (EC) 1272/2008.