Registration Dossier

Administrative data

Description of key information

Key value for chemical safety assessment

Skin sensitisation

Endpoint conclusion
Endpoint conclusion:
adverse effect observed (sensitising)
Additional information:

The skin sensitising potential of (R)-3,7-dimethyloct-6-enal (CAS 2385-77-5) or the racemate citronellal (CAS 106-23-0) has been assessed in a weight of evidence approach based both on experimental data in test animals and human volunteers and on human experience from patch tests.

Animal data

In the only available guideline study, i.e. a mouse local lymph node assay performed according to OECD Guideline 429 and GLP, citronellal (7.5, 15 or 30 % in diethyl phthalate/ethanol (3:1)) showed no sensitising activity as indicated from the determination of cellular proliferation in the auricular lymph nodes in CBA mice (IFF 2004). The EC3 is considered to be above 30% or 7500 µg/cm2.


In a Maurer guinea pig optimisation test Pirbright-Hartley guinea pigs were treated with 9 intradermal injections of 0.1% citronellal in saline during 3 weeks with the last 6 injections supplemented with Freund's complete adjuvans for induction (Maurer 1985). For challenge treatment anintradermal injectionof 0.1% citronellal in salineon day 38 was followed by a 24-hour occlusive patch with 30% citronellal on day 50. Two of 20 animals showed dermal reactions after intradermal challenge and 0/20 reacted after epidermal challenge, indicating no skin sensitising capacity of citronellal under the chosen testing conditions.


However, the negative outcome of these two assays is in contrast to positive test results from a number of methods in guinea pigs using different testing protocols.

Kashima et al. evaluated three test systems that all applied Freund's complete adjuvans. High sensitisation rates of up to 90 or 100% were correspondingly found in the Guinea pig maximization test, in the CCET and the AP2 test.


In the Guinea pig maximization test similar to OECD Guideline 406, Hartley guinea pigs were induced intradermally with 3% citronellal (in liquid paraffin) and FCA on day 0 and a 48-h occlusive patch on day 7 (3 % citral in ethanol), followed by challenge with concentrations of 0.01, 0.03 and 0.1% in ethanol on day 21 by a 24-h occlusive patch (Kashima 1993a). Rechallenge was performed accordingly 3 weeks after the first challenge. All animals (10/10) showed skin reactions after challenge and rechallenge at all concentrations tested. Considering the sensitization rate of 100% and an intradermal induction concentration of 3%, the potency of citronellal to induce skin sensitization is found to be moderate according to the ECHA guidance document R8.


In a short-period contact hypersensitivity assay with Hartley guinea pigs (AP2 test), induction was performed by two applications of 24-h occlusive patches with 3% citronellal in ethanol after intradermal injection of FCA on day 0 and 4 (Kashima 1993b). Subsequent first and second challenges were performed on day 11 and 3 weeks later by open topical applications (30%, 10%, 3% in ethanol) and a third challenge by a 24-h occlusive patch one week later (0.01%, 0.03% and to 0.1% in ethanol). A concentration dependent increase of animals showing dermal reactions after first challenge was observed and 10/10 animals showed dermal reaction after administration of 30% citronellal. Upon third challenge, 9-10/10 animals showed dermal reaction at lower concentrations tested, i.e. 0.01-0.1% citronellal.


In the cumulative contact enhancement test (CCET) with Hartley guinea pigs, citronellal was topically applied (3% in ethanol) for induction (24h occlusive applications on day 0, 2, 7 and 9 including a intradermal injection of FCA on day 7, ; Kashima 1993c) . Animals were challenged on day 21 and 42 with open dermal application (3%, 10%, 30% in ethanol). A concentration dependent increase of animals showing dermal reactions has been observed after challenge and rechallenge, and 7-9/10 animals reacted after application of 30% citronellal.


In a modified Buehler test, patchy and confluent erythema were observed in Hartley Guinea pigs both in the sensitization test group (induction: 10% citronellal, challenge: 1.0 or 2.5% citronellal) and the challenge control group (only challenge with 1%, 2.5% or 10% citronellal) indicating that the challenge concentration was irritating. However, based on a significantly increased cutaneous basophile hypersensitivity (CBH) response found only in the sensitization test group, the authors concluded that skin grades of that test group represented a weak sensitising response rather than an irritant response (Method of CBH response not validated) (Robinson 1990).

Human data

Evidences for skin sensitizing properties of citronellal in humans can be derived from routine patch tests, indicating only a low sensitisation rate to citronellal due to the sporadic occurrence of dermal reactions observed.

In closed patch tests (24 to 48 hrs of contact, vehicle not specified) a total of 144 subjects were tested with different concentrations of citronellal. 1 of 28 subjects with healthy skin showed a positive reaction after application of 10% citronellal, and 1 of 86 subjects with dermatoses showed a questionable positive reaction after application of 0.5% citronellal. No reactions were observed in 30 subjects with healthy skin after application of 2% citronellal (Fujii 1972).In an occlusive patch test, reported as short database summary from a secondary source, application of citronellal (test concentration range 0.05-0.5% ,vehicle base cream or 99% ethanol) for 24 to 48 hours led to a sporadic positive skin reaction in 1 from 86 Asian patients (Takenaka 1986).In a patch test with 70 dermatitis patients none showed a positive reaction to 1% citronellal in petrolatum (Nethercott 1989). In patch tests on 42 patients (19 with cosmetic dermatitis, 23 with eczema and dermatitis) and 4 control subjects no skin reactions were observed after administration of 5% citronellal (Ishihara 1981). Positive patch test reactions to 1% citronellal were observed in 2 of 3 cases with eczematous contact-type hypersensitivity to oils of citronella (Keil 1947).

The sensitisation potential of citronellal was investigated in several controlled studies using human volunteers, i.e. a human repeated insult patch test (modified Draize test) and a human maximization tests (HMT).

In the human repeated insult patch test, the highest tolerable, non-irritating concentration used both for induction and challenge was 10% (no data for other concentrations) was applied either with petrolatum (N=104 subjects) or ethanol (N=73 subjects) as vehicle (Marzulli 1980a). Induction was performed by 10 consecutive 48-hr to 72-hr occlusive patches with 0.2 ml of 10% citronellal, followed by a 72 hr challenge patch after two weeks of rest. The findings were dependent on the vehicle used for application of citronellal. With the vehicle petrolatum, none of the 104 subjects showed a positive reaction. In contrast, application citronellal in ethanol resulted in a sensitisation index of 13.7% (10 from 73 subjects; estimated applied dose of approx. 5000 µg/cm2).

In the HMT with 25 volunteers, the test concentrations used for induction and challenge were 4% (3000 µg/cm2) in petrolatum. After pre-treatment with 1% SLS, induction was performed by 5 consecutive 48-hr semi-occlusive patches, followed by a 48 hr challenge patch after 10 to 14 days of rest. None of the 25 subjects showed a positive reaction.


In a further human maximisation test, only reported as short summary from secondary source, no sensitization reactions were observed in 25 volunteers after application of 4% citronellal in petrolatum (Kligmann 1971).




Overall in a weight of evidence, (R)-3,7-dimethyloct-6-enal (CAS 2385-77-5) and the racemate citronellal (CAS 106-23-0) are considered to be skin sensitizers. As the effect level described in a human repeated insult patch test (5000 µg/cm2) is of the same magnitude as the sensitization effect levels of the structurally related substance citral, i.e. 3900 µg/cm(Lalko and Api, 2008), a read-across with citral has been performed to derive a NOEL as starting point for local short/longterm dermal DNEL. However, dermal reactions upon application of citronellal at that dermal load is vehicle dependent. Furthermore, a higher potency of citral has been observed in local lymph node assays when compared to the LLNA using citronellal. Incorporation of the NOEL for citral (1400 µg/cm) is considered to be a worst case assumption when used as a NOEL for sensitization for citronellal and (R)-3,7-dimethyloct-6-enal (CAS 2385-77-5).

Justification for classification or non-classification

The present data on dermal sensitization fulfill the criteria laid down in 67/548/EEC and 1272/2008/EEC, and therefore, a classification with R43 and "Skin sensitisation" (Category 1) is warranted.