Tetrachloroethylene

Carcinogenicity

Description of key information

The relevance to human health of the kidney tumours seen in male rats exposed to 400 ppm (2760 mg/m3) tetrachloroethylene for 2 years cannot be excluded. The most plausible mechanism of action involves the additional contribution over and above chronic nephrotoxicity of the genotoxic and cytotoxic activity of the reactive metabolite of the glutathione conjugation/beta-lyase pathway. However, as the potential genotoxicity of this metabolite is only expressed under conditions of sustained renal toxicity and associated increased cell proliferation, the threshold for renal toxicity is considered an appropriate starting point for DNEL derivation. Therefore, the exposure level which is considered without any effects in humans, 20 ppm (138 mg/m3) (8 hours TWA value) (see repeated dose toxicity), is considered also to be protective against carcinogenic effects.

Key value for chemical safety assessment

Carcinogenicity: via inhalation route

Endpoint conclusion

Dose descriptor:

NOAEC

138 mg/m³

Justification for classification or non-classification

Based on the available data, tetrachloroethylene has to be classified as Carc. Cat. 3; R40 according to Directive 67/548/EEC and as Carc. Cat. 2; H351 according to EU Classification, Labelling and Packaging of Substances and Mixtures (CLP) Regulation (EC) No. 1272/2008.

Additional information

Overall, although there are many epidemiological studies examining cancer mortality and incidence among dry-cleaners and certain other groups of workers, those that provide any useful information relating specifically to tetrachloroethylene are considerably fewer. When factors such as latency, confounding exposures and biological plausibility are taken into account, it becomes apparent that the evidence shows no increased risk of cancer in humans resulting from exposure to tetrachloroethylene. IARC (1995) concluded that there was an indication for an increased risk of cervical, oesophageal and non-Hodgin's lymphoma cancers from some studies that cannot be completely disregarded. However, in recently completed reevaluation (IARC, 2012) it was observed that a consistent pattern of human cancer associated with tetrachloroethylene exposures in dry cleaning workers was only evident for bladder cancer (limited evidence). These associations appear unlikely to be due to chance, although confounding factors cannot be excluded and the total numbers in the cohort studies combined are relatively small and the exposure-response relationship was weak. A similar conclusion was also reached by the EU ad-hoc expert meeting on carcinogenicity and mutagenicity of tetrachloroethylene held in Ispra in October 1998 (ECB4/21/99). It was agreed that the consistency across two studies for an increased incidence of oesophageal cancer, and across three studies for an increase in cervical cancer, among cohorts with potential occupational exposure to tetrachloroethylene suggested a trend that could not be disregarded. However, it was also agreed that based on the available data, it was not possible to conclusively attribute the observed excess risks of oesophageal and cervical cancer to tetrachloroethylene exposure. A recent study found no association between dry-cleaning work and cancer of the oesophagus and cervix in Denmark, Finland, Norway and Sweden (Lynge et al., 2006). The available human data do not allow a robust quantification because of a lack of quantitative exposure data and potential confounding by other risk factors. The epidemiologic study results are inconsistent, contradictory and do not suffice to identify a clear target organ for carcinogenicity.

 

In relation to animal data, tetrachloroethylene has been tested in two species (mice and rats) producing clearly different patterns of results (National Toxicology Programme, 1986). In the mouse, tetrachloroethylene induced liver tumours by the inhalation (from 100 ppm – 690 mg/m³) and oral (from 540 mg/kg bw/day) routes of exposure, but not kidney tumours; in the rat, no liver tumours were found but a non-statistically significant, low incidence (2/50) of kidney tumours (tubular cell adenocarcinomas) was observed in males only exposed by inhalation to 400 ppm (2760 mg/m³) tetrachloroethylene for 2 years. There were no tumours at 200 ppm (1380 mg/m³), although 3/49 male rats exhibited renal tubular hyperplasia. No kidney tumours were also seen in 3 oral bioassays conducted in SD (Maltoni and Cotti, 1986) and Osborne-Mendel rats (National Cancer Institute, 1977; Weisburger, 1977), although these studies exhibited a number of inadequacies.

 

As tetrachloroethylene is concluded to be not genotoxic in conventional assays in vitro and in vivo, it can be concluded that the mechanism of tumorigenesis does not involve a direct mutagenic action of tetrachloroethylene itself. Although metabolites of tetrachloroethylene associated to both the cytochrom P450 oxidation pathway (tetrachloroethylene oxide) as well as to the gluthathione-S-transferase conjugation pathway (S-1,2,2-trichlorovinylglutathione, S-1,2,2-trichlorovinylcystein andN-acetyl-S-1,2,2-trichlorovinylcysteine) have been identified as genotoxic in vitro in bacterial test systems, their mutagenic potential is not expressed in vivo, as, generally, the pattern of the in vivo results by relevant routes of exposure is negative (see mutagenicity section).

 

For the mouse liver tumours, the mechanism has been shown to involve peroxisomal proliferation (binding to the nuclear receptor PPARα, peroxisome proliferation, oxidative stress, cytotoxicity/necrosis, regenerative cell proliferation, hyperplasia and tumours), an effect to which humans are not responsive. The oxidative metabolic pathway to trichloroacetic acid, an established peroxisomal proliferator in rodents, is particularly active in mouse liver, where effects indicative of such proliferation have been reported in repeated dose studies of tetrachloroethylene. Overall, the evidence demonstrates that the liver tumours obtained in mice are of no significance in relation to human health. This conclusion is in agreement with the opinion of the majority of the experts of an EU ad-hoc subgroup convened by ECB in 1998 to discuss the carcinogenicity and mutagencity of tetrachloroethylene (ECB4/21/99).

 

For the rat kidney tumours seen in F344 males but not in females or in mice of either sex following inhalation exposure to 400 ppm (2760 mg/m³) tetrachloroethylene, there are still some uncertainties about the underlying mechanism.

One plausible mechanism for the development of these tumours is via alpha-2u-globulin accumulation which is considered a male rat specific phenomenon and hence not relevant to humans. However, this putative mechanism has not been rigorously investigated and the available evidence shows a number of inconsistencies. No alpha-2u-globulin was found in male F344 rats following inhalation exposure up to 800 ppm (5520 mg/m³) tetrachloroethylene for 28 days (Green et al, 1990; Green, 1997). However, alpha-2u-globulin was found in 3 out of 3 male F344 rats exposed at 1000 ppm (6900 mg/m³) for 10 days (Green et al, 1990). These findings seem to suggest that if inhalation exposure to tetrachloroethylene induces the formation of alpha-2u-globulin, this is a high-dose phenomenon that occurs at levels of exposure (at least 1000 ppm, 6900 mg/m³, equivalent to an oral dose of 700 mg/kg/day) which are higher than those required to induce renal tumours. This observation is consistent with the findings of two high-dose oral investigations. Goldsworthy et al. (1988) detected alpha-2u-globulin and hyaline droplets formation in male F344 rats given by gavage the high dose of 1000 mg/kg/day for 10 days and Green et al. (1990) found hyaline droplets, tubular casts and areas of tubular regeneration in male F344 rats given by gavage the even higher dose of 1500 mg/kg/day for 42 days. However, no evidence of hyaline droplet nephropathy was found in the 13-week NTP inhalation study in F344 rats exposed up to the high concentration of 1600 ppm (11040 mg/m³). Furthermore, nephropathy including hyaline casts was seen but in both male and female Osborne-Mendel rats given up to 950 mg/kg/day tetrachloroethylene in the NCI oral cancer bioassay and cytomegaly/kariomegaly but not hyaline droplet nephropathy was noted in approximately 33% of male SD rats (but not in females) treated orally with 500 mg/kg/day tetrachloroethylene in the Maltoni & Cotti (1986) cancer bioassay. Overall, in view of the limitations, gaps and inconsistencies in the available data, the level of confidence in this mode of action is rather low. This conclusion is in agreement with the opinion of the majority of the experts of an EU ad-hoc subgroup convened by ECB in 1998 to discuss the carcinogenicity and mutagenicity of tetrachloroethylene (ECB4/21/99). The experts stated that although alpha-2u-globulin accumulation was a plausible mechanism for the development of these kidney tumours in male rats, the evidence available was not yet complete enough to accept this route to tumour formation.

Another postulated mechanism for the development of these kidney tumours in male F344 rats is via cytotoxicity (repeated cycles of cell injury, regenerative cell proliferation, hyperplasia and tumours). Chronic renal toxicity (nephropathy and karyomegaly) has been observed in mice and rats of both sexes at similar exposure levels (from 100 ppm-690 mg/m³by the inhalation route and from 390 mg/kg bw/day by the oral route). However, as tumours develop only in male rats, this mechanism alone does not appear to be a convincing explanation.

Another postulated mechanism for the development of these tumours is via a reactive, short-lived intermediate (dichlorodithioketene) formed in the kidney via the glutathione-conjugation/ β-lyase pathway. Due to its high reactivity with macromolecules such as DNA and proteins, this metabolite is likely to induce tumours via a combination of genotoxic and cytotoxic activity. There is some evidence to suggest that this intermediate may be produced, albeit at very low levels, in humans exposed to tetrachloroethylene. The available data have shown that there is little, if any, beta-lyase activity in humans and that the glutathione conjugation/beta-lyase pathway is at least 40-fold less active in humans (and probably mice) compared to rats. In view of this evidence, although there are large quantitative differences in the formation of this intermediate between rats and humans, the relevance to humans of this mechanism cannot be ruled out.

The activity of this metabolite as the underlying mechanism leading to tumour formation is supported by the available toxicokinetic data as the cancer findings clearly mirror the different activity of the glutathione-conjugation/ β-lyase pathway in different sexes and different species. The toxicokinetic data show that the conjugation/ β-lyase pathway is 2-3 fold more active in male rats (in which the tumours are seen) compared to female rats (in which no tumours are detected) and is at least 40-fold less active in mice (in which again no tumours are observed) and humans compared to rats. Overall, therefore, the genotoxic and cytotoxic activity of the reactive metabolite of the glutathione conjugation/beta-lyase pathway appears to be the most plausible underlying mechanism of the nephrocarcinogenicity of tetrachloroethylene in male rats. However, as chronic renal toxicity is also present in male rats, the genotoxic and cytotoxic activity of the reactive metabolite of the glutathione conjugation/beta-lyase pathway is most likely not the sole mechanism of action, but rather the additional contribution over and above chronic nephrotoxicity.

From the available data it is not possible to ascertain whether the genotoxicity of the metabolite of the conjugation/ β-lyase pathway is the driving force in the carcinogenic process. However, even if the potential genotoxicity of this metabolite were to play a role in tumour development, this would only be expressed under conditions of sustained renal toxicity and associated increased cell proliferation. From this it follows that below the threshold for renal toxicity the potential genotoxicity of the β-lyase metabolite is limited.

 

Overall, it can be concluded that, although the mechanism leading to the formation of kidney tumours in male rats has not been fully elucidated, on balance, the available evidence supports a threshold approach as follows:

-  the low incidence (2/50) of kidney tumours observed was not statistically significant;

- the kidney tumours were only seen in the male rat (sex-specificity and species-specificity of the carcinogenic response);

- the kidney tumours in the male rat occurred only at a concentration that caused significant renal toxicity;

-  there are a number of other plausible mechanisms of action that might explain the response in the male rat (e.g. alpha-2u-globulin accumulation), either alone or in combination;

-  it is well established that there are significant quantitative differences in the activity of the glutathione conjugation/ β-lyase pathway between the rat and human, which suggest that, the human is much less sensitive than the rat in its carcinogenic response to tetrachloroethylene;

- an in vivo UDS test performed in the kidney of rats treated orally with 1000 mg/kg bw/day tetrachloroethylene produced negative results (Goldsworthy et al., 1988);

- there is no convincing epidemiological evidence for renal tumours in humans in several large-scale studies of cancer mortality and cancer incidence among workers exposed to tetrachloroethylene in dry-cleaning and laundering.

Hence, the threshold for renal toxicity is considered an appropriate starting point for the cancer risk characterisation. Therefore, the exposure level which is considered without any effects in humans, 20 ppm (138 mg/m³) (8 hours TWA value) (see repeated dose toxicity), is considered also to be protective against carcinogenic effects.

Justification for selection of carcinogenicity via oral route endpoint:
See discussion below

Justification for selection of carcinogenicity via inhalation route endpoint:
The threshold for renal toxicity is considered an appropriate starting point for the cancer risk characterisation. Therefore, the exposure level which is considered without any effects in humans, 20 ppm (138 mg/m3) (8 hours TWA value) (see repeated dose toxicity), is considered also to be protective against carcinogenic effects.

Justification for selection of carcinogenicity via dermal route endpoint:
See discussion below

Carcinogenicity: via inhalation route (target organ): urogenital: kidneys