Registration Dossier

Administrative data

Description of key information

Key value for chemical safety assessment

Skin sensitisation

Link to relevant study records
Reference
Endpoint:
skin sensitisation: in vivo (LLNA)
Type of information:
experimental study
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: GLP status not known, data fromguideline studies, published in peer reviewed literature,no restrictions, fully adequate for assessment (although very limited data on evaluation of xylene in the LLNA included.
Qualifier:
equivalent or similar to
Guideline:
OECD Guideline 429 (Skin Sensitisation: Local Lymph Node Assay)
GLP compliance:
not specified
Type of study:
mouse local lymph node assay (LLNA)
Species:
mouse
Strain:
not specified
Sex:
not specified
Vehicle:
acetone/olive oil (4:1 v/v)
Concentration:
100%
Details on study design:
The assay was conducted using the standard protocol that contains the essential elements of the Guideline. Stimulation indices (SI) were derived by comparing cell proliferation in auricular lymph nodes from treated animals (quantified using thymidine incorporation) with that of the vehicle control group. The criterion for a positive response in the LLNA is that a SI of 3 or more is induced by one or more application concentrations of the test substance compared with the concurrent control group.
Parameter:
SI
Remarks on result:
other: SI = 3.1. This result indicates a very slightly positive result at the highest concentration tested 100%.
Parameter:
other: disintegrations per minute (DPM)
Remarks on result:
other: No data

Basketter et al. (1996) discussed the merits and methodology behind the LLNA assay, and presented tabulated results for a wide range of substances including mixed xylenes.

Basketter et al. (1999) conducted a statistical analysis (using Receiver Operating Characteristic curves) of data for 134 chemicals tested in the LLNA and in the guinea pig, or for which human information was available, to determine the specificity and sensitivity of the test.

Basketter and Kimber (2010) performed a weight of evidence evaluation of the sensitisation potential of mixed xylenes which included their own LLNA findings together with other chemical and biological considerations.

Interpretation of results:
not sensitising
Remarks:
Migrated information
Conclusions:
Xylene is not a skin sensitiser
Executive summary:

The sensitisation potential of mixed xylenes when tested in the LLNA was described by Basketter et al. (1996), who reported an SI = 3.1 following topical treatment of mice with neat (100%) substance. However while this finding is suggestive of a slightly positive result, it appears to be a false positive given the absence of structural alerts present in the substance, the very weak response obtained (recorded only at the very highest concentration tested), and the absence of any corroborative information from clinical experience, or from other human studies (Basketter and Kimber, 2010) A subsequent statistical analysis of LLNA data for 134 chemicals (including mixed xylenes) performed by Basketter et al. (1999) concluded that a SI “cut-off” value of 3.5 would lead to greater specificity in the interpretation of LLNA results, and also reduce the occurrence of occasional false positive results such as that reported for mixed xylenes (Basketter and Kimber, 2010).

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

Mixed xylene (CAS 1330-20-7) comprises individual xylene isomers (m-xylene, o-xylene, p-xylene) and <10% ethyl benzene. The following information is available to characterise the sensitisation potential of this substance.

Non-Human Information

The sensitisation potential of mixed xylene (comprising xylene isomers and ethyl benzene) when tested in the standard local lymph node assay was reported by Basketter et al. (1996), who obtained a stimulation index of 3.1 following topical treatment of mice with neat (100%) substance. However while this finding is suggestive of sensitisation (according to the strict interpretative criteria of the Guideline), it appears to be a false positive given the absence of structural alerts present, the very weak response obtained (recorded only at the very highest concentration tested), and a lack of any corroborative information from clinical experience, or from other human studies (discussed Basketter and Kimber, 2010). A subsequent statistical analysis of LLNA data for 134 chemicals (including mixed xylenes) performed by Basketter et al. (1999) concluded that a SI “cut-off” value of 3.5 would lead to greater specificity in the interpretation of LLNA results, and also reduce the occurrence of occasional false positive results such as the SI = 3.1 for undiluted mixed xylenes (Basketter and Kimber, 2010). Overall it is concluded that this publication does not provide convincing evidence that mixed xylene possesses a potential to induce or elicit skin sensitisation.

The sensitisation potential of mixed xylene was also assessed in a modified local lymph node assay where increased ear-draining lymph node weight and cell counts were used to quantify lymph node cell proliferation by 9 laboratories participating in a ring-trial. [3H]TdR incorporation in vitro by lymph node cells was also assessed in a single laboratory (assay not conducted by other laboratories). Acute skin inflammation (a potential confounder that can lead to "false positive" results) was also determined based on changes in ear tissue weight. Treatment with mixed xylenes was associated with a statistically significant increase in ear-draining lymph node weight and cell count in 7 of the 9 laboratories, and an increase in ear tissue weight in 3 of 9 laboratories. An increased (but not quantified) increase in 3[H]TdR incorporation by lymph node cells was also reported by one laboratory. Based on these findings, mixed xylene was considered an allergen by a majority of laboratories participating in the trial. However the methodology used was evaluated by an ECVAM Workshop (reported by Basketter et al., 2008) which concluded the assay methods deviated from the Guideline in terms of the strain of mouse, the choice of vehicle, the time of lymph node/lymph node cell collection, and the magnitude of the SI used to differentiate positive and negative samples. Overall, the ECVAM Workshop concluded that these differences represented a major change in methodology that required further validation. The use of [3H]TdR incorporation by lymph node cells in vitro (as opposed to incorporation of [3H]TdR following i.v. injection in vivo, as required by the Guideline) was not discussed by the ECVAM Workshop but is another change relative to the Guideline. When conducting a weight of evidence evaluation of the sensitisation potential of mixed xylenes, Basketter and Kimber (2010) also noted that these deviations meant that deviations meant that the findings were of limited toxicological relevance. Overall it is concluded that this publication does not provide convincing evidence that mixed xylene possesses a potential to induce or elicit skin sensitisation.

There is no relevant non-human information for the individual isomers or for ethyl benzene.

Human information

From widespread use, there have been no reports of skin sensitising properties of mixed xylene as a result of skin contact. In a human maximization test (Kligman, 1966), mixed xylene was tested at 100% and subjects were challenged at 25%. No skin sensitization resulted. This test is considered to be a highly sensitive detector of skin sensitizing potential; the results for more than 80 chemicals were presented in the publication which demonstrates the very considerable predictive sensitivity of this assay.

As cited in Opdyke (1975), Kligman (1974) reported that ethyl benzene produced no sensitisation reactions in 25 volunteers subjected to a maximisation test with 10% ethylbenzene (no data on purity) in petrolatum.

Conclusion

Mixed xylene gave a marginal (SI = 3.1) response in a standard LLNA at the highest concentration tested (100%; Basketter et al., 1996) however an absence of chemical structural alerts, the very weak response obtained, and a lack of any corroborative human information indicates that this is a false positive (Basketter and Kimber, 2010). Results suggestive of a positive response obtained from a non-standard LLNA are considered unreliable given the extent of the methodological deviations present (Basketter et al., 2008; Basketter and Kimber, 2010). Results from a human maximisation test on mixed xylene (Kligman, 1966) using an induction concentration of 100% were negative. Overall there is no convincing evidence that mixed xylene possesses a potential to induce or elicit skin sensitisation. Xylene gave a very slightly positive result in the LLNA when tested undiluted, but the result fits well with the criteria for a false positive. It is an unreactive chemical that would not be identified on the basis of chemical structure as being a potential skin sensitizer. Added to which is the clinical evidence demonstrating that xylene does not cause skin sensitization in humans, even when tested in a very rigorous human predictive assay. Thus, the overwhelming is that xylene (including mixed xylene and the individual isomers) should not be classified as a skin sensitizer. The RAR (2008) for ethyl benzene concluded that there are no reports on skin sensitisation caused by the substance at the workplace and that no sensitisation potential is expected.


Migrated from Short description of key information:
A human maximization test, a sensitive detector of the skin sensitising properties of chemicals, found that mixed xylene did not induce skin sensitisation. The individual xylene isomers and ethylbenzene are not considered to be skin sensitizers.

Justification for selection of skin sensitisation endpoint:
Mixed xylene gave a marginal (SI = 3.1) response in a standard LLNA at the highest concentration tested (100%; Basketter et al., 1996) however an absence of chemical structural alerts, the very weak response obtained, and a lack of any corroborative human information indicates that this is a false positive (Basketter and Kimber, 2010). Results suggestive of a positive response obtained from a non-standard LLNA are considered unreliable given the extent of the methodological deviations present (Basketter et al., 2008; Basketter and Kimber, 2010). Results from a human maximisation test on mixed xylene (Kligman, 1966) using an induction concentration of 100% were negative. Overall there is no convincing evidence that mixed xylene possesses a potential to induce or elicit skin sensitisation.

Respiratory sensitisation

Endpoint conclusion
Endpoint conclusion:
no study available
Additional information:

Migrated from Short description of key information:
No data available, but no evidence of respiratory sensitisation reported from worker exposure to xylenes.
The RAR (2008) for ethyl benzene concluded that there are no reports on inhalation allergy caused by the substance at the workplace and there no sensitisation potential is expected.

Justification for classification or non-classification

Xylene is an unreactive chemical that would not be identified on the basis of chemical structure as being a potential skin sensitizer. In addition, there is no clinical evidence demonstrating that xylene causes skin sensitization in humans, even when tested in a very rigorous human predictive assay. Therefore, mixed xylene does not warrant classification as a skin sensitizer under CLP. In conclusion, mixed xylene streams do not warrant classification as a respiratory sensitizer under CLP.