Registration Dossier

Diss Factsheets

Administrative data

Key value for chemical safety assessment

Additional information

The mutagenicity of trichloroethylene (TCE) has been extensively investigated in a number of in vitro and in vivo test systems. The most comprehensive reviews of the mutagenicity of TCE have been published by Moore and Harrington-Brock (2000), in the EU Risk Assessment Report (2004) and in the EPA-IRIS Document on TCE (2009). It should be noted that some older studies used TCE containing the epoxide stabilisers epichlorohydrin and/or 1,2-epoxybutane. These are known mutagens and consequently positive tests involving epoxide-stabilized TCE do not provide any information on the potential mutagenicity of pure TCE.

In vitro studies

TCE (with or without epoxide stabilizers) has been extensively tested in bacterial systems (Ames test). Exposures have involved either the vapour or liquid phase.There is no convincing evidence that pure TCE is mutagenic in these test systems. Apart from a single Ames test,where an equivocal (less than 2-fold) induction of revertant colonies was observed in Salmonella typhimurium TA100 in the presence of S9,all Ames tests with epoxide-free TCE showed no induction of the mutation frequency.

TCE has also been tested for mutations, gene conversions, or recombinations in fungal or yeast systems including host-mediated assays.TCE gave conflicting results in these test systems. All positive tests used TCE of unspecified purity, so it is possible that these tests demonstrated the presence of mutagenic activity that was related to epoxide stabilisers. Overall, because of the inconsistencies, it is not possible to draw any firm conclusions about the mutagenicity of TCE on the basis of these studies.

Epoxide-free TCE tested positive for mutagenicity in a mouse lymphoma gene mutation assay (MLA) at high dose levels, requiring the presence of an exogenous metabolising system (NTP, 1988). No colony sizing was performed and therefore, the mechanism of the induction of the mutation frequency is not known.

The key study for the in vitro chromosome aberration endpoint is the well-conducted study with epoxide-free TCE (containing an amine stabilizer) in Chinese hamster ovary cells by Galloway et al.(1987). This study was clearly negative, with and without metabolic activation (Galloway et al.,1987; NTP, 1988). The presence or absence of toxicity was not reported, but the highest concentration used was 14.9 mg/ml, which is in excess of the usual maximum dose level of 5 mg/ml.

TCE has also been tested in vitro for the induction of sister chromatid exchanges (SCEs), micronucleus formation, unscheduled DNA synthesis (UDS) and DNA fragmentation. Negative and positive results have been reported in these studies. However, due to inconsistent results and limitations in several of the available studies it is not possible to draw a clear conclusion concerning the in vitro genotoxicity of TCE in these assays.

To summarise the in vitro mammalian cell data, epoxide-free TCE was slightly positive in a mouse lymphoma gene mutation assay, requiring the presence of an exogenous metabolising system. It was not possible to draw firm conclusions from in vitro chromosome aberration, SCE and UDS assays, because the results were inconsistent and most tests had limitations. Overall, there is some weak evidence that TCE can induce mutations in mammalian cells.

In vivo studies

In vivo, TCE has been tested for micronuclei formation, chromosome aberrations and SCEs. Invivostudies are also available for UDS, DNA binding and damage, but these have not been discussed in detail here. For a review of these studies, see for example the EU Risk Assessment report on TCE.


Tests for micronuclei, chromosome aberrations and SCEs

The keyin vivostudy is the well-conductedmicronucleusstudy by Spencer et al., 2003. The in vivo genotoxic potential of TCE was evaluated by examining the incidence of micronucleated polychromatic erythrocytes (MN-PCE) in the bone marrow of male CD rats exposed by inhalation to targeted concentrations of 0 (negative control), 50, 500, 2500, or 5000 ppm for 6 consecutive hours on a single day. This study is the onlyin vivostudy conducted according to internationally accepted guidelines and under the conditions of GLP. There were nostatistically significant increases in the frequencies of MN-PCE in groups treated with the test material as compared to the negative controls. Since no increase in the incidence of MN was observed in any of the TCE exposure groups, kinetochore analyses (to detect possible aneugenic effects) were not performed. Under the experimental conditions used, TCE was considered to be negative in the mouse bone marrow micronucleus test.

Another well-conducted micronucleus study is the study in mice (male, B6C3F1 strain) by Shelby et al. (1993) involving intraperitoneal administration of TCE at 500, 1000 and 2000 mg/kg bw. TCE was negative in this assay.

Contrary to these negative findings, both Kligerman et al.(1994) and Robbiano et al.(2004) identified evidence of genotoxicity in micronucleus assays. In the study by Kligerman et al.(1994), groups of five male rats or C57BL/6J mice received a single 6 hour exposure to reagent grade TCE (purity >99%) by the inhalation route at nominal concentrations of 0, 5, 500 or 5000 ppm. Additionally, groups of five male rats received four consecutive daily 6 hour exposures at nominal concentrations of 0, 5, 50 or 500 ppm. A single, 6 h TCE exposure induced a statistically significant and dose-related increase in the numbers of micronucleated PCEs in rats. In contrast, TCE had no effect in rats on the numbers of micronucleated PCEs following repeated exposure or on the numbers of micronucleated PBLs, following either a single or repeated exposure. Similarly, TCE had no effect on the numbers of micronuclei in mice splenocytes or PCEs. No changes in the frequency of chromosome aberrations and SCEs were reported in the TCE-exposed rats (PBLs) or mice (splenocytes).

In a single oral dose study by Robbiano et al.(2004), a statistically significant increase in the average micronuclei frequency and a statistically significant increase in the average frequency of DNA breaks and/or alkali labile sites were observed in the kidney of rats that had undergone unilateral nephrectomy and folic acid injection to stimulate cell proliferation in the remaining kidney. This study can only be considered to be of limited value in contributing to an assessment of the genotoxicity of TCE in the intact animal due tohigh oral dose and the non-standard methodology that was used.

Kligerman et al. (1994), as part of the previously described inhalation study, conducted chromosome aberration and SCE analyses in rat PBLs and mouse splenocytes in parallel to the previously described micronucleus analysis. No changes were reported in the TCE exposed rats or mice. In another in vivo cytogenetic study following inhalation exposure, TCE had no effect on the frequency of chromosome aberrations (NIOSH, 1980). Other negative in vivo chromosome aberration tests have been reported, using the oral (Loprieno and Abbondandolo, 1980; Sbrana et al.1985 (cited in ECETOC, 1994)) and intraperitoneal (Cerna and Kypenova, 1977) routes.

Tests for UDS, DNA binding and damage

A Comet assay in the kidney was performed by Clay (2008) to investigatethe effects of TCE and its metabolite S-1,2-dichlorovinylcysteine (DCVC). Comet formation was assessed in kidney proximal tubule cells from groups of five rats exposed to trichloroethylene (purity 99.5%)by inhalation or DCVC by the oral route. Inhalation exposures were by whole body to TCE concentrations of 0, 500, 1000 or 2000 ppm, 6 h/day for 5 days. DCVC was administered as a single dose of 0, 1 or 10 mg/kg, and the animals killed at 2 and 16 hours after dosing. This study did not show an increase in the tail length of comets in the rat kidney.In the latter study, and contrary to the findings of Robbiano et al.(2004), TCE exposure of rats by inhalation did not induce DNA damage in the rat kidney,a target organ relevant for carcinogenicity, even at a high dose of 2000 ppm.

Trichloroethylene did not induce UDS in rat or mouse hepatocytes Mirsalis et al.(1989). However, there was evidence of hepatocellular proliferation, observed as an increased percentage of cells in S-phase; this effect was more marked in males than females. Negative results were also reported in a similar UDS assay in mice (Doolittle et al., 1987).

Stott et al.(1982) found little evidence of hepatic DNA alkylation in mice. Parchman and Magee (1982) found no clear evidence of trichloroethylene binding in the liver of rats following administration of single intraperitoneal doses of between 10 and 1000 mg/kg. Nelson and Bull (1988) found some evidence of induction of single strand breaks (SSBs) in DNA in the liver of rats and mice, following high doses of 3 to 4 mg/kg (22 to 30 mmol/kg) in the rat and 1.5 mg/kg (11.4 mmol/kg) in the mouse. The biological significance of such effects is uncertain.

Other in vivo studies

Furthermore, an inhalation transgenic (LacZ) mice study showed no mutagenic response in all investigated organs, i. e. liver, lung and bone marrow, spleen, kidney and testicular germ cells (Douglas et al.,1999). Taken into account the high dose positive in vitro MLA test, for which an induction of gene mutations is assumed, the negative results as obtained in the transgenic mice study, including the liver, is indicative for the non-genotoxicity of TCE in vivo.

Germ cells

The mutagenic activity of trichloroethylene in germ cells has been investigated using the dominant lethal assay. No evidence of the induction of dominant lethal mutations was observed. However, this study did not involve exposure to toxic concentrations of trichloroethylene, and consequently the significance that can be attached to this negative result is limited.

The ability of trichloroethylene to induce micronuclei in spermatids of mice, exposed by inhalation at concentrations of 0, 5, 50 or 500 ppm,for 6 h/day for 5 days, has been investigated by Allen et al.(1994). The frequency of micronuclei in spermatids harvested 14 days after the initial exposure was not affected by treatment. No information on the presence or absence of general toxicity was presented, which raises a question mark against the appropriateness of the chosen exposure concentrations.

Studies in humans

The investigation of possible trichloroethylene-related genotoxic activity in humans has been investigated in several studies. Three have looked for SCE induction (Gu et al., 1981; Nagaya et al., 1989; Seij i et al., 1990), one examined clastogenicity (Rasmussen et al., 1988). In addition, two studies have investigated the presence of mutations in the von Hippel-Lindau (VHL) tumour suppressor gene in renal cell cancer patients with occupational exposure to trichloroethylene.

In the study by Nagaya et al.(1989), no effects of TCE were observed on the frequency of SCE in peripheral blood lymphocytes of exposedworkers. From the other two SCE studies (Gu et al., 1981; Seiji et al., 1990), and from the chromosome aberration study by Rasmussen et al.(1988), no firm conclusions could be drawn because of a number of study design deficiencies, such as small group size, poor characterisation of exposure and lack of consideration of potential confounding factors such as smoking.

Twenty-three patients with histologically verified renal cell carcinoma and with a history of occupational exposure to trichloroethylene were analysed for the presence of somatic mutations within the Von Hindel-Lippau (VHL) tumour suppressor gene (Brüning et al., 1997). All 23 patients were found tohave mutations in the VHL gene. In a further study by the same team, the presence of VHL mutations was investigated in 44 renal cell cancer patients who had a history of occupational exposure to trichloroethylene (Brauch et al., 1999). VHL mutations were seen in tumour DNA of 75% of trichloroethylene exposed patients. These results suggest that there is a type of VHL gene mutation present in the renal cells of renal cell carcinoma patients that is not present in the tumours of renal cell carcinoma patients who have no occupational exposure to trichloroethylene. This association between trichloroethylene exposure and unique pattern of mutations in the VHL gene is difficult to interpret in the context of an assessment of the hazardous properties of trichloroethylene. Possibly the VHL mutations were an unspecific late stage development due to the interaction of trichloroethylene and cancer cells, and not part of the chain of events leading to cancer. Although the methods used in this study are considered to be state-of-the-art, this investigation is considered pioneering in nature. Consequently, without replication in another group of patients, the weight that can be given to these results is limited.

A second team has studied VHL mutations in another group of renal cell cancer patients (12) with a history of occupational exposure to trichloroethylene (Schraml et al., 1999). In contrast to the findings of Brauch et al., this investigation revealed no differences between solvent exposed and “spontaneous” tumour tissue in the phenotype, genotype or mutation pattern in the VHL gene. However, this study can contribute little to an assessment of the mutagenic properties of trichloroethylene because solvent exposure histories and other relevant information (such as age) on the study subjects are not available.

Summary of in vivo mutagenicity data

In vivo, most of the mutagenicity tests performed showed a negative result. Trichloroethylene was unequivocally negative in two well-conducted micronucleus tests, one in rats by inhalation (male, CD strain; Spencer et al., 2003) and another one in mice (male, B6C3F1 strain; Shelby et al., 1993) involving intraperitoneal administration. Considerable weight is given to these negative results because these studies were conducted according to internationally recognised guidelines, employing dose levels and a route of administration that ensured that exposure of the target tissue to the test substance (and its metabolites) was maximised.

A single micronucleus test was positive, whereas in the same experimental set-up no chromosome aberrations or SCE’s were found (Kligerman et al.,1994). This positive result might well reflect aneuploidy instead of clastogenicity, although the presence of epoxides cannot be excluded considering that reagent grade trichloroethylene was used. A study by Robbiano et al.(2004) showed an induction in the Comet assay and micronucleus formation, for which the relevance is not clear considering the use of reagent grade trichloroethylene and the protocol used.

Negative results were observed in theinhalation transgenic (LacZ) mice study by Douglas et al.(1999), in several chromosome aberration tests (Cerna and Kypenova, 1977; Loprieno and Abbondandolo, 1980), and in the comet assay by Clay (2008).

Although there does appear to be an association between the presence of mutations in VHL gene kidney tumours in humans, there is no clear evidence that it was TCE that caused it and whether it led to the development of the tumours.


In conclusion, there does not appear to be a consistent, reproducible profile of activity emerging from the many mutagenicity/genotoxicity assays that have been conducted on TCE in the past. Positive findings from in vitro and in vivo assays are typically associated either with high dose levels, presence of epoxide stabilizers or non-standard methodologies. In the available well-conducted in vitro and in vivo assays the results indicate an absence of genotoxicity. Therefore, based on a weight-of-evidence approach, the data available suggest that trichloroethylene should be regarded as non-mutagenic and non-genotoxic.

Short description of key information:
Using a weight-of-evidence approach, the available data indicate that trichloroethylene should be regarded as non-mutagenic/non-genotoxic.

Endpoint Conclusion: No adverse effect observed (negative)

Justification for classification or non-classification

Using a weight-of-evidence approach, the available data indicate that trichloroethylene should be regarded as non-mutagenic/non-genotoxic.

Based on these results, trichloroethylene does not need to be classified as mutagenic according to EU Directive 67/548/EEC and the EU Classification, Labelling and Packaging of Substances and Mixtures (CLP) Regulation (EC) No. 1272/2008.

Following the EU Risk assessment trichloroethylene was officially classified as a Category 2 mutagen according to CLP.