Tetrachloroethylene

Epidemiological data

Administrative data

Endpoint:

epidemiological data

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: A variety of published cancer epidemiology studies, generally assessed to be acceptable and peer-reviewed under the EU Existing Substances Regulation 793/93.

Cross-reference

Reason / purpose for cross-reference:

reference to same study

Data source

Reference

Reference Type:

other: European Union Risk Assessment report (EU-RAR, 2008)

Title:

European Risk Assessment Report Tetrachloroethylene, CAS No. 127-18-4, EINECS 204-825-9. Final draft human health report for publication.

Author:

Rapporteur: United Kingdom on behalf of the European Union

Year:

2008

Materials and methods

Study type:

other: summary of cancer mortality and cancer incidence studies in workers

Endpoint addressed:

carcinogenicity

Principles of method if other than guideline:

not applicable

GLP compliance:

not specified

Test material

Test material information

Constituent 1

Reference substance name:

Tetrachloroethylene

EC Number:

204-825-9

EC Name:

Tetrachloroethylene

Cas Number:

127-18-4

Molecular formula:

C2Cl4

IUPAC Name:

tetrachloroethene

Details on test material:

no information available

Method

Type of population:

occupational

Ethical approval:

not applicable

Details on study design:

limited information available

Exposure assessment:

not specified

Details on exposure:

limited information available

Statistical methods:

limited information available

Results and discussion

Results:

Several large-scale studies of cancer mortality and cancer incidence among workers exposed to tetrachloroethylene in dry-cleaning and laundering have been reported. In addition there are two general population drinking water studies, and a number of case-control studies. The latter have been uninformative with regard to the carcinogenic potential of tetrachloroethylene in humans.The occupational mortality studies have generally been limited by the lack of clear information on the levels of exposure to tetrachloroethylene, and by possible confounding due to exposure to other solvents among the subjects studied. In this regard, the widespread use of tetrachloroethylene in the dry cleaning industry did not begin until the 1960s, and assuming a latency for tumour development of 15-20 years, the cancer deaths observed in some of the studies for periods up to 1980-82, if occupationally related, would be attributable to exposure conditions prior to the widespread use of tetrachloroethylene. In those studies where cancer mortality could be distinguished among workers employed in establishments using tetrachloroethylene only, no increases in the risks for liver and renal cancers were observed. In this sub-cohort of tetrachloroethylene-exposed dry-cleaners, elevated mortality was seen only in relation to cancer of the oesophagus, buccal cavity and pharynx, and tongue. The anatomical location of these tumours suggests a role for a locally acting carcinogen, which is not supported by the available genotoxicity and animal carcinogenicity database. Latency considerations for this study also argue against a possible role for tetrachloroethylene.In cancer incidence studies, no increases in risks from exposure to tetrachloroethylene for any specific type of cancer, including risks for liver and renal cancers, have been observed. Furthermore, a recent critical review of the epidemiological literature on occupational exposure to tetrachloroethylene and cancer has considered that the current evidence does not support a conclusion that occupational exposure to tetrachloroethylene is a risk factor for cancer of any specific site, although a firm conclusion cannot be drawn based on the available data. Similarly, no convincing evidence is available for an increased risk of cancer (total cancer, or any specific type of cancer) arising from exposure to tetrachloroethylene in drinking water. Overall, a variety of epidemiological studies have shown no good evidence for an increased risk of carcinogenicity in humans resulting from exposure to tetrachloroethylene. 

Confounding factors:

limited information available

Strengths and weaknesses:

limited information available

Any other information on results incl. tables

Human carcinogenicity

Occupational mortality studies

Most of the useful information comes from mortality studies of dry-cleaners conducted in. Some studies suggest that use of tetrachloroethylene in the dry-cleaning industry became widespread there in the 1950s but most, drawing on National Institute for Occupational Safety and Health surveys of occupational hygiene in the industry, suggest that before 1960 petroleum-based solvents (carbon tetrachloride and Stoddard solvent) were in use in the majority of establishments. One study cites an estimate that by 1977 almost a quarter of dry-cleaners still used petroleum‑based solvents. Petroleum solvents have remained in use in some industrial cleaning premises and in 75% of commercial cleaners were still using them in 1984 (see below).

 

The latency for the majority of cancers under consideration is unlikely to be less than 10 years and is more likely up to 20 years. Since tetrachloroethylene was not the major solvent exposure for most cohorts until after 1960, the cancer mortality experience of cohorts prior to 1970 could not reasonably be ascribed to tetrachloroethylene exposure (except for subcohorts known to have worked only with tetrachloroethylene). Indeed, if tetrachloroethylene was causing cancer in workers, the true attributable mortality would not become apparent until 1980 or after. In several of the studies, therefore, it is not reasonable to attribute the excess mortality seen to tetrachloroethylene, even if the studies are otherwise of good quality.

 

Results from 4 American retrospective cohort mortality studies, conducted sequentially on essentially the same cohort, have been reported. The first explained the difficulties of assembling the cohort (Kaplan, 1980). Since it proved impossible to define a cohort from industry records, it was defined from trade union membership records. The construction of exposure histories was indirect, using several sources. Full information on the solvents used was available for less than half of the premises in which cohort members were employed. The total cohort studied was composed of workers who had been employed for at least 1 year prior to 1960 in an establishment that used tetrachloroethylene. Many of these workers may have worked before or after that year with other solvents. A sub-cohort that had worked only in premises where tetrachloroethylene was known to be the main solvent in use was also defined. Even these workers may have had some occupational exposure to other solvents, particularly in the early years of their working lives. Only where the exposure history was completely unknown or was known not to include tetrachloroethylene, were subjects excluded from follow-up.

 

The design was a retrospective cohort mortality study of 1597 subjects, giving 30 000 person.years of follow-up, with 285 deaths up to late 1977. Standardised mortality ratios (SMRs) were calculated using US national rates for expected deaths, for black and white subjects separately where the racial distribution of deaths was known. For all deaths the SMR was 0.93 and for all cancers it was 1.12. Elevations of mortality were seen for cancer of the colon (SMR 1.82, 11 deaths), pancreas (SMR 1.52, 5 deaths) and urinary organs (SMR 1.96, 5 deaths). Statistical significance tests were not performed. Patterns with latency and duration of exposure were consistent with exposure to tetrachloroethylene, but the author pointed out the possibility of serious misclassification biases in these analyses due to the incomplete nature of the exposure histories.

 

In the second study in the series the follow-up of the cohort was extended to the end of 1982. Numbers were now 1690 subjects, 42 267 person.years and 493 deaths (Brown and Kaplan, 1987). The authors do not explain the increase in the cohort size, but it is almost certainly due to deaths having been recorded between 1977 and 1982 for subjects originally excluded as untraced at follow-up. Separate analyses were now conducted for some diseases in the sub‑cohort (207 males and 408 females) who worked only in shops using tetrachloroethylene as the main solvent. Presumably these were not done in the previous study because the number of deaths in that group was too small to yield meaningful results.

The SMR for all deaths for the whole cohort was 0.86 and that for all cancer deaths was 1.16. The only statistically significant findings for the whole cohort were in respect of bladder cancer (SMR 2.96, confidence interval [CI] 1.28 - 5.86, 8 deaths) and all cancers of the urinary organs (SMR 2.55, CI 1.32 - 4.50, 12 deaths). Non-significant increases in mortality from colonic cancer (SMR 1.36, 16 deaths) and cancer of the pancreas (SMR 1.72, 11 deaths) persisted from the earlier study. There were no deaths from primary liver cancer. The excess in all urinary cancer deaths was largely accounted for by the bladder cancers. There was also an increase in mortality from kidney cancers, but the histological type is unknown. If they were transitional cell tumours, rather than the commoner clear cell carcinoma, the picture would be consistent with a shared carcinogenic mechanism for the kidney and urinary tumours. In the sub-cohort employed only in establishments using tetrachloroethylene there was only 1 death from urinary tract cancer, against 1.3 expected. This would count against the urinary cancer excess being related to tetrachloroethylene exposure.

 

A further follow-up  study of the series extended the mortality analysis to the end of 1990 (Ruderet al, 1994). There were now 1701 traced subjects, 47 273 person.years at risk and 769 deaths. Male and female deaths were analysed separately for the whole cohort, and for the first time a detailed analysis was presented for subjects employed only in shops using tetrachloroethylene (although this still does not mean that they were exposed only to tetrachloroethylene).

 

For the whole cohort, the SMR for all deaths was 1.01. Significant elevations in mortality were seen due to cancer of the digestive organs (SMR 1.31, CI 1.01‑1.68, 63 deaths), oesophagus (SMR 2.14, CI 1.02‑3.94, 10 deaths), intestine except rectum (SMR 1.56, CI 1.02‑2.29, 26 deaths), urinary organs (SMR 2.07, CI 1.10‑3.54, 13 deaths), bladder (SMR 2.54, CI 1.16‑4.82, 9 deaths) and for all cancers (SMR 1.23, CI 1.07‑1.41, 209 deaths). For digestive organs, oesophagus and intestine the cancers occurred mostly in women, with SMRs of 1.48, 3.24 and 1.69 respectively. For urinary organs and bladder the effect was mostly in men, with SMRs of 2.21 and 3.27. For total cancers the increase was significant in both sexes. There was a non-significant increase in pancreatic cancers (SMR 1.66, 15 deaths).

 

The excess of all digestive cancers was new to this study. Numerically it was due to a combination of the cancers of the oesophagus, intestine and pancreas and was not accounted for by any individual cause. Cancer of the intestine effectively means cancer of the colon, since small intestinal malignancies are extremely rare. The excess in all urinary cancers was mostly accounted for by the bladder cancers. There were no new kidney cancer deaths since the previous study, and only 1 new bladder cancer death.

 

Analysis by latency and duration of exposure for all the significantly elevated cancer rates shows that the peak mortality is in the band with 20+ years since first exposure and > 5 years of presumed exposure. This fits with an occupational cause but is subject to the caveat about assumed tetrachloroethylene exposure noted above. Comparison of the mortality at end 1990 and end 1982 shows that the oesophageal cancer mortality is a relatively recent phenomenon (1982 SMR 1.17, 4 deaths; 1990 SMR 2.14, 10 deaths) and could therefore be related to more recent exposure patterns from, say, 1970. For the other causes most of the deaths had already occurred by 1982.

 

Analysis of mortality in the sub-cohort who worked only in shops using tetrachloroethylene (620 subjects, 222 deaths) shows no overall elevation of cancer mortality and no statistically significant elevation of any individual cancer SMR. The majority of the SMRs which were significantly raised in the whole cohort were close to 1 in this sub-cohort, suggesting strongly that if an occupational exposure was responsible for the increases in the whole cohort it was exposure to something other than tetrachloroethylene. The exception is oesophageal cancer, for which the SMR in the “tetrachloroethylene only” sub-cohort was 2.64 (CI 0.72‑6.76, 4 deaths). This sub-cohort also has high SMRs for cancer of the buccal cavity and pharynx (SMR 2.51, CI 0.52‑7.33, 3 deaths) and tongue (SMR 7.25, CI 0.88‑26.2, 2 deaths). This pattern of oesophageal, buccal and tongue cancers would appear consistent with oral ingestion of a locally acting carcinogen. Alcohol or smoking seem unlikely as the possible cause since there is no excess of respectively cancers of the liver and pancreas or lung and bladder in this group. However, there is insufficient evidence to conclude whether the development of oesophageal cancer is due to occupational exposure to tetrachloroethylene or a confounding factor. 

 

The most recent study of the series has extended the mortality analyses of the Brown and Kaplan (1987) and Ruderet al,(1994)cohorts to the end of 1996 (Ruderet al,2001). At this time, numbers are 1703 traced subjects, 53 622 person.years at risk and 995 deaths. As for the Ruderet al,(1994) study, male and female deaths were analysed separately and analyses were performed for men (207) and women (415) employed in dry-cleaning shops using tetrachloroethylene as the main cleaning solvent as opposed to establishments where this and other solvents were used.

 

For all deaths the SMR was 1.03 and that for all cancer deaths was 1.25. As observed in the previous study, statistically significant elevations in mortality were seen for cancer of the intestine (SMR 1.48, CI 1.01-2.09, 32 deaths), oesophagus (SMR 2.47, CI 1.35-4.14, 14 deaths) and bladder (SMR 2.22, CI 1.06-4.08, 10 deaths). Unlike the preceding studies, statistically significant elevations in mortality for the entire cohort were recorded for tongue (SMR 5.0, CI 1.62-11.68, 5 deaths), lung (SMR 1.36, CI 1.05-1.73, 65 cases) and cervical cancers (SMR 1.95, CI 1.0-3.40, 12 cases). Tongue cancers occurred exclusively in non‑white workers as did the statistically significant increases observed in the excess lung cancers recorded (men SMR 1.61, CI 1.0-2.4, 25 deaths; women SMR 1.87, CI 1.1-3.0, 16 deaths). For bladder (SMR 4.15, CI 1.1-10.6, 4 deaths) and intestinal (SMR 2.28, CI 1.0-4.5, 8 deaths) cancers the effect was predominantly observed in non-white men.

 

As was seen in the previous study, analysis by latency and duration of exposure showed that the peak mortality occurred in those with more than 5 years of presumed exposure and 20 or more years of latency since first exposure. This was true for oesophageal (SMR 5.03 CI 2.41‑9.47, 10 deaths) and bladder (SMR 4.31, CI 1.85-8.76, 8 deaths) cancers specifically.  The statistically significant elevations in SMRs recorded for tongue, intestine and lung cancer were associated with increasing time since first exposure. A significant association between duration of employment in dry-cleaning shops (i.e. those using tetrachloroethylene and/or other solvents) and cervical cancer rate was also recorded.

 

Analysis of mortality in the sub-cohort who worked only in shops using tetrachloroethylene (625 subjects, 302 deaths) compared to those in the corresponding sub-cohort of the Ruderet al, (1994) study, again showed no overall statistically significant elevation in cancer mortality, with the exception of tongue cancer (SMR 9.03, CI 1.86-26.39, 3 deaths).  Elevations were also seen in cervical cancer (SMR 1.89, CI 0.52-4.84, 4 deaths) and oesophageal cancer (SMR 2.65, CI 0.85-6.20, 5 deaths) but these did not reach statistical significance. Thus, increased SMR values consistent with the previous study were observed in the ‘tetrachloroethylene-only’ exposed sub-cohort for tongue and oesophageal cancers.

 

Overall, the pattern emerging throughout this series of studies suggests that although particularly in the early years there appears to be an occupationally associated increase in cancer of the bladder and colon in the cohort, this is not likely to be due to tetrachloroethylene exposure. However, it could well be due to other occupational exposures which the workers historically received. There is a more recent increase in oesophageal, oral and cervical cancers in the sub-cohort likely to have the purest exposures to tetrachloroethylene. This profile would fit statistically with an occupational cause however, it is also biologically plausible that it is in fact due to a confounding lifestyle factor not related to tetrachloroethylene exposure. Overall therefore, no clear view on the association of occupational exposure to tetrachloroethylene and the development of oesophageal, oral and/or cervical cancer can be reached. 

 

Some further information is available from an American proportional mortality study based on 330 death certificates available from the death benefits records of a trade union (Blairet al, 1979). Only 279 of the decedents were known to have worked exclusively in dry-cleaning premises; the other 51 could have worked in laundries. The deaths occurred in the period 1957-1977. Assuming a mean latency of 15 years for occupational cancers these deaths, if related to occupation, would reflect exposure conditions in 1942‑1962, a period largely prior to the widespread use of tetrachloroethylene.

 

Thus this study has very limited value as a study of mortality related to tetrachloroethylene exposure, and is best regarded as a hypothesis-generating study looking at dry-cleaners and laundry workers. Since the proportions of deaths due to all causes must add up to 1.00 in a study of this type, a deficiency of deaths due to one or more major causes must lead to an apparent excess due to other causes, and vice versa. The two major causes of death in this study were cancers and cardiovascular disease. All cancer deaths gave a proportional mortality ratio (PMR) of 1.28 and all cardiovascular deaths 0.79, both statistically significant. The actual numbers are such that these two statistics almost exactly balance each other. There is therefore a question as to whether the results indicate that the population primarily had a deficiency in cardiovascular deaths or an increase in cancer deaths across the whole spectrum of organs. The former explanation seems to be the more reasonable, particularly as it is what might be expected due to the healthy worker effect, although the authors prefer the alternative increased cancer hypothesis.

 

The specific cancers for which statistically significant excesses were found were lung, cervix and skin. The first two of these had the highest observedandexpected numbers and the third had the highest PMR, although the expected number was very small. They are therefore the three most likely to reach statistical significance. Nearly all cancers were elevated, and there was really little difference in risk between these three and all the rest. Overall, nothing useful to an assessment of tetrachloroethylene can be concluded from this study, given the limitations of its design and timing in relation to the introduction of tetrachloroethylene into dry-cleaning in.

 

The same authors went on to conduct a retrospective cohort mortality study, using trade union records, designed to follow up their PMR study (Blairet al, 1990). The cohort consisted of 5365 subjects who had been union members for at least one year during the period 1945 to 1978; a further 425 workers were excluded because the data held were inadequate for mortality analyses. The available data provided 98 818 person.years of follow‑up and 1222 deaths, with expected numbers being based on death rates from the general American population. For each subject a three-level classification of likely exposure to solvents was assigned, based on recorded job titles. It was thought that tetrachloroethylene was the most likely solvent exposure after 1960, with uncertain exposures before that year. Vital status was followed up from 1948 to 1979. Applying a 15-year estimated latency, it is likely that the majority of deaths in the study, if related to occupation, would be associated with exposure conditions prior to the widespread introduction of tetrachloroethylene to the industry.

 

The SMR from deaths from all causes was 0.9 (95%CI 0.9‑1.0, 1129 deaths). Significantly elevated SMRs were noted for all cancers (SMR 1.2, 95%CI 1.0‑1.3, 294 deaths), cancer of the oesophagus (SMR 2.1, 95%CI 1.1‑3.6, 13 deaths), cancer of the cervix (SMR 1.7, CI 1.0‑2.0, 21 deaths) and emphysema (SMR 2.0, 95%CI 1.1‑3.4, 14 deaths). Non-significantly elevated SMRs above 2 were found for thyroid cancer (3.3, 95%CI 0.7‑9.8, 3 deaths) and Hodgkin’s disease (2.1, 95%CI 0.6‑5.3, 4 deaths). Since 35 causes of death were examined it is likely that 2 causes would show significantly elevated risks due to chance alone. Ignoring the “all cancers” SMR, because it is secondary to the individual elevations of several other SMRs, there were in fact three significantly elevated SMRs. The results are not, therefore, a very convincing demonstration of a real elevation of cancer risk. The SMR for liver cancer was low at 0.7 (95%CI 0.2‑1.7, 5 deaths).

 

There were no trends apparent with increasing exposure level, i.e. no dose-response effects, for all cancers or for cancers of the oesophagus, cervix, bladder or lymphohaematopoietic systems, although for the last of these the risk was highest in the most highly exposed category. Analyses by duration of exposure, peak exposure, cumulative exposure and latency were stated to be uninformative. Perhaps the most striking finding was that 11 of the 13 deaths from oesophageal carcinoma occurred in black males, giving an SMR of 3.5 for that group (standardised for black male death rates). This pronounced racial pattern may suggest that an undetermined lifestyle factor was responsible for the excess. Overall, this study provides no convincing evidence of an occupational cancer risk in dry-cleaners, and more particularly no evidence of a risk associated with tetrachloroethylene exposure.

 

The mortality follow-up of the same cohort of dry-cleaners was extended up to December 1993 to further evaluate cancer risks associated with organic solvents (Blair et al., 2003). The cohort was composed of 11,062 dry-cleaners, members of Local No. 161 of the Textile Processors Service Trades Healthcare Professional and Technical Employees admitted to the union before 1978 (earliest entry 1948). Of these, 5,369 were included in the analysis after excluding 5,272 individuals who had worked less than one year and 421 who lacked important demographic information for the analysis. The available data provided 146,082 person×years of follow‑up and 2,351 deaths; expected numbers were based on 5-year age and calendar-time death rates from the general population. SMRs for selected causes of death were calculated after adjustment for age at death, year of death, race and gender. For each subject a three-level classification of likely exposure to solvents was assigned, based on recorded job titles at the time of entry. Cleaners were assigned to the high exposure group; pressers, sewers and those at the counter were assigned to the medium exposure group and those employed at pick-up stations were assigned to the little or no exposure group.

 

Significantly elevated SMRs were noted for cancer of the oesophagus (SMR 2.2, 95%CI 1.5‑3.3, 26 deaths), cancer of the cervix (SMR 1.6, 95%CI 1.0‑2.3, 27 deaths), lung cancer (SMR 1.4, 95%CI 1.1-1.6, 125 deaths) and emphysema (SMR 1.7, 95%CI 1.0‑2.5, 21 deaths). Non-significantly elevated SMRs were found for cancer of the larinx (SMR 1.7, 95%CI 0.6‑3.7, 6 deaths) and Hodgkin’s disease (SMR 2.0, 95%CI 0.6‑4.6, 5 deaths). The SMRs for liver and kidney tumours were low at 0.8 (95%CI 0.4-1.5, 10 deaths) and 1.0 (95%CI 0.4-2.0, 8 deaths) respectively. None of these causes of death showed strong relationships with duration or estimated level of exposure to dry cleaning solvents, although SMRs for cancer of the larinx (2.7 for medium/high exposure vs 0 for little or no exposure), lung (1.5 for medium/high exposure vs 1.0 for little or no exposure) and kidney (1.5 for medium/high exposure vs 0.3 for little or no exposure) were larger among subject estimated to have had higher levels of exposure.

 

The consistent excesses for cancers of the oesophagus, cervix, larinx and lung in the two follow-up periods indicate that these are unlikely to be chance findings. However, given the lack of information on potential confounding factors such as smoking, alcohol use and socioeconomic status, and the absence of detailed job histories while employed in the dry-cleaning industry which could lead to exposure misclassification, it is not possible to attribute these excesses to occupational exposure to organic solvents.

 

A standardised mortality odds ratio (SMOR) study, based on death certificates for 440 laundry and dry-cleaning workers in , provides very limited useful information (Duh and Asal, 1984). The reference used for the standardisation was either all cardiovascular deaths or all “other” causes not broken down in the analysis. The difference in the results was slight. The number of deaths comprised all death certificates from between 1975 and 1981 where the occupation could be identified as dry-cleaning or laundry work. It was not possible to separate the two because of the occupational coding system used on the death certificates. , at this time, was unusual in that more than half of the solvents used in dry-cleaning were still petroleum-based.

The SMOR for all cancer deaths was 0.9 (CI 0.7‑1.2, 97 deaths). Other findings were statistically significant elevations of mortality from lung cancer (SMOR 1.7, CI 1.2‑2.5, 37 deaths) and kidney cancer (SMOR 3.8, CI 1.9‑7.6, 7 deaths). Bladder cancer mortality was not elevated (SMOR 0.4, CI 0.1 - 2.8, 1 death). Overall, it is difficult to draw conclusions and any excess cause-specific mortality would in any case not relate to tetrachloroethylene exposure.

 

A retrospective case-control mortality study of 300 deaths of white men who had died of brain or CNS tumours has been described (Heinemanet al,1994). Controls comprised 320 white males who died of other causes, matched by age, year of death and study area. Next-of-kin were located for both groups and questioned by an examiner blind to the cause of death, to determine a history of past exposure to six chlorinated aliphatic hydrocarbons, including tetrachloroethylene.

Job-exposure matrices were used to indicate whether tetrachloroethylene and/or other organic solvents were likely to have been used in each industry and occupation and to provide estimates of the probability and intensity of exposure. The risk estimate of astrocytic brain cancer increased with probability, intensity and duration of exposure to tetrachloroethylene, but these trends were not always consistent and moreover were not statistically significantly different from matched controls. Given the limitations of this study in relation to a lack of direct exposure information and the similar associations apparent with most of the other solvents investigated, no conclusions can be drawn for tetrachloroethylene.

A proportional mortality study of 671 deaths of female laundry and dry-cleaning workers in in the period 1963-1977 has been reported (Katz and Jowett, 1981). To attempt to control for low socio-economic status, comparisons were made with proportional mortality in a group of other low wage service occupations as well as with US national data.

 

There was a significant excess mortality from cancer of the cervix (PMR 1.95, 10 deaths) using national reference rates, but this declined below significance (PMR 1.41) when low wage occupation rates were used. There were similar findings for ischaemic heart disease and diabetes. Two statistically significantly raised PMRs which did not change when referenced to low wage occupation rates were those for kidney cancer (2.57, 7 deaths) and unspecified cancers of the genital organs (4.95, 4 deaths). The PMR for bladder cancer was elevated (1.89, 5 deaths), but this was not statistically significant. There was no overall excess of cancer deaths when compared to either of the reference groups.

 

Overall, the findings of this study give some weak support to the hypothesis that there may be an excess mortality from urothelial cancers in dry-cleaning workers and launderers. However, given the lack of information on exposure histories, the likelihood of previous exposure to dry-cleaning solvents other than tetrachloroethylene and the inability to distinguish between the dry-cleaners and the laundry workers, no conclusions can be made regarding the effects of tetrachloroethylene.

 

A more recent proportional mortality study of 8163 deaths of male and female laundry and dry-cleaning workers in 28 American states in the period 1979-1990, has been described (Walkeret al,1997). The US National Occupational Mortality Surveillance system was used to make comparisons with decedents in all occupations. Statistically significant excess mortalities from cancers of the oesophagus (PMR 2.15, 12 deaths) and larynx (PMR 3.18, 6 deaths) were recorded in black and white men aged 15-64 years, respectively. Elevations in the PMR for oesophageal cancer were also recorded in black and white females of the same age range (1.84, 9 deaths and 1.89, 4 deaths, respectively), but these were not statistically significant. Similarly PMRs for lung cancer were increased somewhat in all groups in the 15-64 year age range. In those aged 65 years and above, a statistically significantly elevated PMR was observed for breast cancer in white men (12.75, 4 deaths), with non significant increases seen for lung cancer in black women. Overall, although this study indicates an excess mortality from certain cancers in dry‑cleaning workers and launderers, these findings are difficult to interpret as both groups were combined for analyses. Furthermore, the database used for these analyses lacks information on the duration of employment, estimates of workplace exposure levels, confounders such as alcohol and tobacco use and the possibility of exposure to other chemicals. Indeed, the direct relevance of the study to an evaluation of tetrachloroethylene is open to question, given that a substantial part of the data must have been derived from workers not predominantly exposed to tetrachloroethylene or even not exposed to it at all.

 

In a Japanese proportional mortality study, 1711 deaths in laundry and dry-cleaning workers between 1971 and 1980 were identified from union death benefit records (Nakamura, 1985).  Expected deaths were derived from Japanese national rates for 1975. Information on specific job, smoking and alcohol consumption was obtained for 294 subjects by means of questionnaires sent to relatives, and this permitted some limited stratified analyses, but for the majority of subjects even the type of work done (laundry or dry-cleaning) was unknown.

Significant elevations of non-cancer mortality are seen for “other forms of heart disease” and for “other diseases of the liver”. For cancer deaths, the most notable significant elevations seen are for small intestine in men of all ages, and bladder and kidney cancer in some age bands, also in men. The figures for women show some scattered increases in some age groups, but these are based on small numbers and there is no clear overall pattern. The stratified analyses by job, smoking and drinking habits are limited in scope and of no help in interpreting the overall picture. In conclusion, the study is inadequately reported and is of little value in examining mortality due to exposure to tetrachloroethylene, particularly since historically only about 30% of Japanese dry-cleaners have used tetrachloroethylene.

 

Finally, retrospective mortality studies of cohorts of metal polishers and platers (Blair, 1980), jewellery workers (Dubrow and Gute, 1987), chemical workers (Olsenet al., 1989) and aircraft maintenance and manufacturing staff (Spirtaset al, 1991; Boiceet al,1999) did not produce any useful information specifically regarding the carcinogenicity of tetrachloroethylene, largely because of the poorly-defined but undoubtedly multiple nature of the chemical exposures involved.

 

References

Blair A, Decoufle P, Grauman D (1979): Causes of death among laundry and dry cleaning workers.Am J Pub Health.69; 508-511

 

Blair A (1980): Mortality among workers in the metal polishing and plating industry, 1951-1969,J Occup Med.22; 158-162

 

Blair A, Stewart PA, Tolbert PEet al(1990): Cancer and other causes of death among a cohort of dry cleaners,Br JMed.47; 162-168

 

Boice JD, , Fryzek JP et al (1999): Mortality among aircraft manufacturing workers.Occup Environ Med. 56; 581-597

 

Brown DP and (1987): Retrospective cohort mortality study of dry cleaner workers using perchloroethylene,J Occup Med.29; 535-541

 

Dubrow R and Gute DM (1987): Cause-specific mortality among jewelry workers,Am JMed.12; 579-593

 

Duh RW and Asal NR (1980): Mortality among dry cleaners in ,Amer J Publ Health.74; 1278-1280

 

Heineman EF, Cocco P, Gomex MRet al.(1994): Occupational exposure to chlorinated aliphatic hydrocarbons and risk of astrocytic brain cancer.Am J.26; 155-169

 

Kaplan(1980): Dry cleaner workers exposed to perchloroethylene. A retrospective cohort mortality study,Report to NIOSH, 210-77-0094

 

Katz RM and Jowett D (1981): Female laundry and dry cleaning workers in : A mortality analysis,Am J Pub Health.71; 305-307

 

Nakamura K (1985): Mortality patterns among cleaning workers,Jap J Health.27; 24-37

 

Olsen GW, Hearn S, Cook RRet al(1989): Mortality experience of a cohort of chemical workers,J Occup Med.31; 32-34

 

Ruder AM, Ward EM and Brown DP (1994): Cancer mortality in female and male dry-cleaning

workers,J Occup Med.36; 867-874

 

Ruder AM, Ward EM and Brown DP (2001): Mortality in dry-celaning workers: an update.Am J.39; 121-132

 

Spirtas R, Stewart PA, Lee JSet al(1991): Retrospective cohort mortality study of workers at an aircraft maintenance facility. I Epidemiological results,Br J Med.48;515-530

 

Walker JT, , Lalich NR, Sestito JP and Halperin WE (1997): Cancer mortality among laundry and dry cleaning workers.Am J.32; 614-619

 

 

Cancer incidence studies

A recent study measured the incidence of cancers detected up to 1992 in a cohort of 292 male and 557 female Finnish workers who were biologically monitored for tetrachloroethylene (Anttilaet al, 1995). For the period of monitoring (1974-1983), the median blood tetrachloroethylene level was about 115 µg/l in the men and 65 µg/l in the women. The 8‑hour TWA tetrachloroethylene exposure levels for the dry-cleaners, likely to have formed most of the cohort, were usually below 50 ppm (345 mg/m³). No overall excess of cancer was found in comparison with the Finnish national rate, with the standardised incidence ratio (SIR) for all sites being 0.90 (CI 0.61‑1.27, 31 cases). Several individual cancer sites had SIR values of greater than 2, namely pancreas (3.08, CI 0.63‑8.99, 3 cases), cervix (3.20, CI 0.39‑11.6, 2 cases) and non-Hodgkin’s lymphoma (3.76, CI 0.77‑11.0, 3 cases), but none of these findings were statistically significant. There was an apparent deficit in overall mortality, particularly among the women (SMR 0.5, CI 0.3‑0.9, 12 deaths; men, SMR 0.7, CI 0.3‑1.2, 10 deaths). Overall, this study provides some reassurance regarding the carcinogenic potential of tetrachloroethylene in humans.

 

Another recent study provides useful information regarding the incidence of certain cancers in dry-cleaners in , where tetrachloroethylene has been the predominant dry‑cleaning solvent since the late 1950s (Lyngeet al, 1995). The report builds on that of a previous study of cancer incidence (Lynge and Thygesen, 1990) by distinguishing between laundry work and dry-cleaning as the occupation for the cases of primary liver cancer identified in the earlier study. The cohort consisted of 10 600 laundry and dry-cleaning workers identified from 1970 census data; no specific exposure information was available for these subjects. Altogether 17 cases (14 women and 3 men) of primary liver cancer and 16 cases (9 women and 7 men) of renal cell carcinoma were observed during a follow-up period extending to 1987. For the updated study, it became possible to classify the launderers and the dry‑cleaners separately, and for each case 5 controls were randomly selected from the cohort. It was found that none of the 17 cases of liver cancer and only 3/16 (19%) of the kidney cancer cases (all men) worked in dry-cleaners, as compared to laundries, in 1970. Among the controls, respectively 20 (24%) and 20 (29%) were dry-cleaners. The relative risk for renal cell carcinoma among the dry-cleaners was 0.7 (CI 0.2‑2.6). In conclusion, this study provides good evidence that exposure to dry-cleaning solvents (predominantly tetrachloroethylene) did not lead to liver or renal cell cancer in a group of about 2500‑3000 workers followed-up over a 17-year period.

 

Another study focused on the occurrence of renal cancer using data from the Swedish Cancer-Environment Registry, which links 1960 census data and cancer incidence for the period 1961‑1979 with industry and occupation (McLaughlinet al, 1987). The standardised incidence ratios for renal cell cancer for the dry-cleaning and laundry industry were 0.99 (18 cases) among men and 0.86 (25 cases) among women, indicating no increased risk. No other useful information was given in the report.

 

A recent critical review of the epidemiological literature on occupational exposure to tetrachloroethylene and cancer has been conducted by Mundtet al. (2003). A comprehensive search was conducted to identify all available epidemiological literature pertaining to the carcinogenic effects of tetrachloroethylene. Two or more epidemiologists independently reviewed each paper according to a set of criteria (quality of the study, consistency of the study results within the context of the collective literature, strength of the evidence for an association) for final inclusion. Forty-four papers, which provided reasonable data on up to 17 cancer sites, remained for the critical review. These were analysed in the context of the available background literature for each cancer site and were assessed on the basis of specified methodological and scientific quality criteria.

 

While all the epidemiological studies selected for inclusion investigated similar exposure-health outcome relationships, there was a broad diversity of proxy measurements of exposure to tetrachloroethylene, as well as numerous specific cancer outcomes of interest. The widespread lack of valid exposure measurements or other adequate indicators of potential for exposure were consistent limitations. Although some of the published studies make a limited contribution to our understanding of the role of tetrachloroethylene exposure as a risk factor for cancer, none is adequately strong, nor is the body of evidence convincingly consistent to draw firm conclusions. Further, none of the cohort studies with sub-cohorts primarily exposed to tetrachloroethylene demonstrated any results different from the broader cohorts with mixed exposure. This argues against any tetrachloroethylene-specific association. A relationship between tetrachloroethylene and cancer of the following sites was considered unlikely, in part due to potential confounding: oral cavity, liver, pancreas, cervix and lung. Scientific evidence was inadequate for laryngeal, kidney, esophageal, and bladder cancer. Nevertheless, because of a number of positive findings suggested from some of these epidemiological studies (e.g. for esophageal and bladder cancer), the possibility that associations between tetrachloroethylene and some cancers exists in humans cannot be definitely ruled out. More evidence is needed to elucidate associations, if they exist, or to demonstrate with adequate power that they do not exist. However, until such additional evidence is available, the current epidemiological evidence does not support a conclusion that occupational exposure to tetrachloroethylene is a risk factor for cancer of any specific site, although a firm conclusion cannot be drawn based on the available data.

 

A series of case-control studies of selected cancers (oesophagus, gastric cardia, liver, pancreas, cervix, kidney, bladder and non-Hodgkin lymphoma) nested in a cohort of laundry and dry-cleaning workers identified from the 1970 censuses in Denmark, Norway, Sweden and Finland were carried out in order to investigate whether dry-cleaning work in the Nordic countries around 1970, when tetrachloroethylene was the dominant dry-cleaning solvent, was associated with an increased risk of cancer (Lynge et al., 2006). The cohort consisted of 46,768 people. Each person was followed up from 1971 until 1997-2001 for death, emigration and incident cancer based on linkage with the nationwide population, death and cancer registries. 1,616 cancer cases were identified. Controls (2,398) were randomly selected from the cohort using frequency match by country, sex, 5-year age group and 5-year calendar period at the time of diagnosis of the case. On the basis of information available on the census forms, cases and controls were categorised mainly in exposed dry-cleaners (878), unexposed laundry workers (2,420) and unclassifiable (716). Unfortunately, a high proportion (18%) of cases and controls were unclassifiable as to whether they were employed in dry-cleaning or laundry work in 1970. On the basis of information obtained through pension schemes and telephone interviews with a next-of-kin, exposed dry-cleaning cases and controls were further categorised by length of employment. The limited number of air measurements of tetrachloroethylene available did not allow subdivision of the study subjects by exposure level. However, as the data indicated a fairly stable exposure level (mean of 164 mg/m³), duration of employment was considered an acceptable proxy measure of relative cumulative dose. Information on smoking and drinking habits was obtained for the Norwegian and Swedish cases and controls through telephone interviews with a next-of-kin. Smoking was equally frequent among exposed (dry-cleaners) and non-exposed (laundry workers) and alcohol intake was rather limited. Cancer risks (calculated by logistic regression) of dry cleaners were compared with those of laundry workers, two groups with similar jobs apart from the use of solvents.

 

Dry-cleaning work in the Nordic countries was not associated with significantly increased risks of cancer of the gastric cardia (RR = 0.69; 95% CI, 0.31-1.53), pancreas (RR = 1.27; 95% CI, 0.90-1.80 and no association with length of employment), kidney (RR = 0.67; 95% CI, 0.43-1.05) or liver (RR = 0.76; 95% CI, 0.38-1.52) or with non-Hodgkin lymphoma (RR = 0.95; 95% CI, 0.65-1.41). Dry-cleaning work was not associated with an increased risk of oesophageal cancer (RR = 0.76; 95% CI, 0.34-1.69), but the study was hampered by several unclassifiable cases. The authors note that this result for the oesophageal cancer is in contrast to findings from the , but that the inconsistency can be explained by chance, confounding or differences in exposure levels. No association was seen between dry-cleaning work and increased risk of cervical cancer (RR = 0.98; 95% CI, 0.65-1.47), although there was a borderline significantly elevated risk among assistants in dry-cleaning shops (RR = 1.73; 95% CI, 1.00-2.97). Given that this increased risk of cervical cancer was not seen in those directly involved in the dry-cleaning process, it is unlikely that tetrachloroethylene caused it. There was an excess risk of bladder cancer (RR = 1.44; 95% CI, 1.07-1.93) among dry-cleaners from the Nordic countries, but this was not associated with length of employment.

 

References

 

Anttila A, Pukkala E, Sallmen Met al(1995): Cancer incidence among Finnish workers exposed to halogenated hydrocarbons,J Occup Environ Med.37; 797-806

 

Lynge E and Thygesen L (1990): Primary liver cancer among women in laundry and dry-cleaning work in ,Scand J Work Environ Health.16; 108-112

 

Lynge E, Carstensen B and Andersen O (1995): Primary liver cancer and renal cell carcinoma in laundry and dry-cleaning workers in ,Scand J Work Environ Health.21; 293-295

 

Lynge E, Andersen A, Rylander L, Tinnerberg H and Johansen K (2006): Cancer in persons working in dry cleaning in the Nordic countries.Environ Health Perspectives.114(2); 213-219 (funding: HSIA)

 

McLaughlin JK, Malker HSR, Stone BJet al(1987): Occupational risks for renal cancer in ,Br J Med.44; 119-123

 

Mundt, KA, Birk, T and , (2003): Critical review of the epidemiological literature on occupational exposure to perchloroethylene and cancer.Int. Arch. Occup. Environ.Health.76: 473-491 (funding: HSIA)

 

 

Drinking water studies

Cancer incidence in several American states has been examined in relation to contamination of the drinking water supply with a variety of chemicals, including tetrachloroethylene. The water supplies investigated were invariably contaminated with other volatile organic chemicals, notably trichloroethylene, 1,1,1-trichloroethane and dichloroethylenes, and these were often present at much higher concentrations than tetrachloroethylene itself. Thus no useful information regarding the specific carcinogenicity of tetrachloroethylene was generated in studies conducted in(Lagakoset al, 1986),(Isacsonetal,1985),(Faglianoet al, 1990; Cohnet al, 1994) or(Mallin, 1990). However, more recent investigations in focused on tetrachloroethylene, which for a number of years had leached from the linings of drinking water distribution pipes (Aschengrauet al, 1993, 1998; Pauluet al,1999). All these studies evaluated the relationship between specific cancers and the exposure to tetrachloroethylene estimated using an algorithm that took into account residential history, the presence, age and dimensions of lined pipe, and the flow characteristics of the distribution system. The exposure to tetrachloroethylene was expressed as the “relative delivered dose”, defined as the estimated amount of tetrachloroethylene (in mg) that entered a house as a solute in drinking water over a specified time period.

 

In the first study (Aschengrauet al,1993), the relationship between 61 bladder cancers, 35 kidney cancers and 34 leukaemia cases and exposure to tetrachloroethylene was evaluated.  Exposure was examined on an “ever” versus “never” basis, or as “unexposed”, “low” (up to the 90^thpercentile among the exposed) and “high” (above the 90^thpercentile among the exposed) relative delivered doses. The controls (852 for bladder cancer, 777 for kidney cancer and 737 for leukaemia) were selected to represent the population that gave rise to the cases, and so lived in the same towns. As such, a proportion had also potentially been exposed to tetrachloroethylene in their drinking water.

 

Only limited exposure data were provided in the report. A total of 5.7% of all subjects (i e cases and controls) were classified as being ever-exposed to tetrachloroethylene when latency was taken into account, rising to 13.5% when latency was ignored. Relative delivered dose estimates ranged from 0.01 to 90.6 mg with latency and from 0.01 to 209.4 without latency; the 90^thpercentiles among exposed controls were respectively 27.1 and 44.1 mg.

Analyses of morbidity were conducted with and without consideration of the latent period for cancer, taken to be 5 years for leukaemia and 15 years for bladder and kidney cancers. For leukaemia, 5 cases were considered ever exposed to tetrachloroethylene in drinking water when latency was taken into account. This gave a relative risk, adjusted for confounding factors, of 1.96 (CI 0.71‑5.37) compared to the control group. For exposed subjects whose relative delivered dose was “high”, the adjusted relative risk was 5.84 (CI 1.37‑24.91). Ignoring latency, there were 7 leukaemia cases, relative risks being 2.13 (CI 0.88‑5.19) for any exposure and 8.33 (CI 1.53‑45 .29) for the 2 “high” exposure cases. For bladder cancer, there was one ever-exposed case when latency was considered, and 13 cases when the latent period was ignored. This latter result gave an adjusted relative risk of 1.39 (CI 0.67‑2.91) among ever-exposed subjects, and 4.03 (CI 0.65‑25.10) for the cases exposed to “high” relative delivered doses. For kidney cancer, no effect of tetrachloroethylene exposure was apparent. There were no cases when latency was considered, and 6 cases (all “low” exposure) when latency was not taken into account, giving an adjusted relative risk of 1.08.

Overall, this study does not provide any convincing evidence for a carcinogenic effect for tetrachloroethylene. Only one of the 3 cancers showed an elevated relative risk when the latent period was taken into account, and even this increase (for leukaemia) was not statistically significant for the ever-exposed group. The significant effect obtained for “high” exposure was based on just 2 cases.

 

The second study by Aschengrauet al. (1998) evaluated the relationship between 258 cases of breast cancer and exposure to tetrachlorethylene. Exposure was examined on an ‘ever’ versus ‘never’ exposed basis and as never-exposed versus ‘low’ (up to and including the 50^thpercentile among the exposed) or systematically higher levels of exposure (above 50^th, 75^thand 90^thpercentile).  Several latent periods were considered:  5,7,9,11,13 and 15 years.  As in the previous study by this author the controls (686) were selected to represent the population that gave rise to the cases and therefore a proportion had also been exposed to tetrachloroethylene in the drinking/bathing water. As in the previous study, the reporting of exposure data was very limited in this study. A total of 14% of cases and 11.8% of controls were classified as ever having been exposed to tetrachcloroethylene when latency was ignored and from 0-9.7% of cases and 0.6-8.2% of controls were classified when 5-15 years latency was taken into account. Relative delivered dose estimates ranged from 0.004 –335.5 mg and estimates at the median, 75^thand 90^thpercentile were 9.1, 28.9 and 53.4 mg, respectively. Small statistically insignificant increases in relative risk, adjusted for confounding factors, were observed when latency was ignored and exposure was either above or below the median percentile (0.9-1.8). When 7 or 9 years latency were considered, the adjusted relative risks were calculated to be 1.5 (CI 0.5-4.7, 6 cases) and 2.3 (CI 0.6-8.8, 5 cases), and 2.7 (CI 0.4-15.8, 3 cases) and 7.6 (CI 0.9-161.3, 3 cases) in women exposed above the 75^thor 90^thpercentile, respectively. At 11 and 13 years latency no increases in relative risk were observed but the numbers of exposed cases were very small in these groups. In post-menopausal women when 7 and 9 years latency were taken into account, the adjusted relative risks were also increased but again these were statistically insignificant. It is difficult to interpret the results of this study as increases in relative risk were only seen when 7 and 9 years latency were taken into account and the numbers of exposed subjects was limited. Indeed the authors themselves stated that no firm conclusions could be drawn and have now undertaken a new study with a larger number of more recently diagnosed cases, the findings from which have yet to be published.

 

In the most recent of this series of Masachusettes drinking water studies, the relative risk of colo-rectal (311 cases, 1158 controls) and lung cancers (243 cases, 1206 controls) has been examined (Pauluet al.1999). Exposure was examined as previously described (Aschengrauet al.1998) and controls were similarly selected to represent the population that gave rise to the cases and therefore a proportion had also been exposed to tetrachloroethylene in the drinking/bathing water. A total of 14% of colon-rectal cancer cases, 13.9% of lung cancer cases and 13% of controls were classified as ever having been exposed to tetrachcloroethylene when latency was ignored and the frequency of exposed subjects diminished with increasing empirical latency period. Relative delivered dose estimates for colon-rectal cancer cases ranged from 0.002 –356.7 mg and estimates at the median, 75^thand 90^thpercentile were 9.1, 28.9 and 53.4 mg, respectively. Those for lung cancer cases ranged from 0.002 –703.5 mg and estimates at the median, 75^thand 90^thpercentile were 7.0, 25.7 and 49.4 mg, respectively. Analyses of lung and colon-rectal cancer morbidity were conducted with and without consideration of latency periods. Small but insignificant increases in adjusted relative risk of 2.0 (CI 0.6-5.8, 5 lung cancer cases; 0.6-5.8, 6 colon-rectum cases) were observed when 13 years latency was considered. When latency periods of 5 and 7 years were considered, adjusted colon-rectal relative risks were 1.7 (CI 0.5- 5.7, 5 cases) and 1.5 (CI 0.3-6.1, 3 cases) in subjects exposed above the 90^thpercentile, again these were statistically insignificant elevations. Those for lung cancer were 3.3 (CI 0.6-13.4, 3 cases) and 6.2 (1.1-31.6, 3 cases) although several confounders contributed to the increase in relative risk in these groups.

 

Overall, this study does not provide any convincing evidence for a carcinogenic effect for tetrachloroethylene. Elevations in relative risk where observed when the latent period was taken into account, however these were not statistically significant for any of the exposed groups which consisted, in many cases, of small numbers of subjects.

 

Given the nature of the exposures described in the three Masachusettes drinking water studies described above, there is considerable uncertainty regarding the tetrachloroethylene levels to which the subjects of these studies were actually exposed. The estimated relative delivered doses, taken in over many years in drinking water, are typically less than could be inhaled in one shift at a dry-cleaning shop working well within the occupational exposure limit.

A further recent drinking water study investigated the incidences of various cancers amongst   villagers in whose water supply was found to be contaminated with both tetrachloroethylene and trichloroethylene, at concentrations of each of up to about 200 micrograms/l (Vartiainenet al., 1993). The period of exposure to the contaminated water is uncertain, but is most likely to have been several decades. A sample of 116 villagers using the contaminated water was shown to have significantly elevated urinary excretions of a metabolite common to tetrachloroethylene and trichloroethylene (trichloroacetic acid), compared to a control group of 60 subjects that did not drink such water. No increased incidences were found for any of the cancers  (total cancer, liver cancer, non-Hodgkin’s lymphoma, Hodgkin’s disease, multiple myeloma and leukaemia) evaluated for the total period 1953-1991 or for the separate periods of 1953-1971, 1972-1981 and 1982-1991.   

 

References

 

Aschengrau A, Ozonoff D, Paulu Cet al(1993): Cancer risk and tetrachloroethylene-contaminated drinking water inMassachusetts,Arch Environ Health.48; 284-292

 

Aschengrau A, Paulu C and Ozonoff D (1998): Tetrachloroethylene-contaminated drinking water and the risk of breast cancer.Environ Health Perspect.106; 947-953

 

Cohn P, Klotz J, Bove Fet al(1994): Drinking water contamination and the incidence of leukemia and non-Hodgkin’s lymphoma,Environ Health Perspect.102; 556-561

 

Fagliano J, Berry M, Bove F and Burke T (1990): Drinking water contamination and the incidence of leukaemia. An ecologic study.Am J Pub Health.80; 1209-1212

 

Isacson, Bean JA, Splinter Ret al (1985): Drinking water and cancer incidence in. III. Association of cancer with indices of contamination,Am J Epidem.121; 856-869

 

Lagakos, S. W., B. J. Wessen and M. Zelen (1986): An Analysis of Contaminated Well Water and Health Effects in,,J Am Stat Assoc.81; 583-596

 

Mallin K (1990): Investigation of a bladder cancer cluster inNorthwestern,Am J Epidem.132; (suppl 1) S96-S106

 

Paulu C, Aschengrau A and Ozonoff D (1999): Tetrachloroethylene-contaminated drinking water in and the risk of colon-rectum, lung, and other cancers.Environ Health Perspect.107; 265-271

 

 

 

 

Case - control studies

There have been a considerable number of case-control studies examining the occupations of subjects with a specific cancer, and comparing the findings with those from a group of matched but healthy controls. These studies invariably suffer from the deficiencies found in most of the mortality studies. Thus there is little or no account of exposure histories to allow distinction between exposure to tetrachloroethylene, to other (chlorinated or petroleum-based) solvents and / or to other chemicals. Usually, laundry workers were grouped with dry-cleaners, and matching of the controls did not take into account the generally low socio-economic status of the cases. Another problem is that many of the studies were based on relatively small numbers of cases, so that in turn the number of dry-cleaners identified was also very small. 

 

Against this background of sparse information, no convincing evidence has been forthcoming from a number of studies to indicate an association between exposure to tetrachloroethylene and primary liver cancer / hepatocellular carcinoma (Stemhagenet al., 1983; Hardellet al., 1984; Hernberget al., 1984; Austinet al., 1987; Hernberget al., 1988; Suarezetal., 1989; Bondet al., 1990).  Similarly, various studies have not produced any good evidence for an association between exposure to tetrachloroethylene and bladder cancer (Smithet al., 1985), pancreatic cancer (Lin and Kessler, 1981), oral cavity, laryngeal and oesophageal cancer (Vaughanet al. 1997) or renal cell cancer (Asalet al., 1988; Harringtonet al., 1989; Sharpeet al., 1989; McCredie and Stewart, 1993; Mellemgaardet al., 1994; Mandelet al., 1995).  In a recent study, the relative risk of renal cell carcinoma has been shown not to be elevated significantly in men or women exposed to tetrachloroethylene, furthermore any observed gender differences in this risk cannot be associated with exposure to organic solvents (Dosemeciet al.1999).  Furthermore a critical review of a number of epidemiologic studies of tetrachloroethylene has concluded that a causal association cannot be made between this material and the risk of renal cell carcinoma (McLaughlinet al. 1997).

 

References

 

Asal NR, Lee ET, Geyer JRet al(1988): Risk factors in renal cell carcinoma. II Medical history, occupation, multivariate analysis and conclusions,Cancer Detect Prevent.13; 263-279

 

Austin H, Delzell E, Grufferman Set al(1987): Case-control study of hepatocellular carcinoma, occupation and chemical exposures,J Occup Med.29; 665-669

 

Bond GG, McLaren EA,et al(1990): Liver and biliary tract cancer among chemical workers,Am J Med.18; 19-24

 

Dosemeci M, Cocco P and Chow WH (1999): Gender differences in risk of renal cell carcinoma and occupational exposures to chlorinated aliphatic hydrocarbons.Am J.36; 54-59

 

Hardell L, Bengtsson NO, Jonsson Uet al(1984): Aetiological aspects on primary liver cancer with special regard to alcohol, organic solvents and acute intermittent porphyria - an epidemiological investigation,Br J Cancer.50; 389-397

 

Harrington, Whitby H,et al(1989): Renal disease and occupational exposure to organic solvents : a case referent approach,Br J Med.46; 643-650

 

Hernberg S, Korkala ML, Asikainen U and Riala R (1984): Primary liver cancer and exposure to solvents,Int Arch Occup Environ Health.54; 147-153

 

Hernberg S, Kauppinen T, Riala Ret al(1988): Increased risk for primary liver cancer among women exposed to solvents,Scand J Work Environ Health.14; 356-365

 

Lin RS and Kessler II (1981): A multifactorial model for pancreatic cancer in man,J Am Med Assoc.245; 147-152

 

Mandel JS, McLaughlin JK, Schlehofer Bet al(1995): International renal-cell cancer study. IV. Occupation,Int J Cancer.61; 601-605

 

McCredie M and Stewart JH (1993): Risk factors for kidney cancer inNew. IV Occupation,Br J Med.50; 349-354

 

McLaughlin JK and Blot WJ (1997): A critical review of epidemiology studies of trichloroethylene and perchloroethylene and risk of renal-cell cancer.Int Arch Occup Health.70; 222-231

 

Mellemgaard A, Engholm G, McLaughlin JK and Olsen JH (1994): Occupational risk factors for renal-cell carcinoma in,Scand J Work Environ Health.20; 160-165

 

Sharpe CR, Rochon JE, Adam JM and Suissa S (1989): Case-control study of hydrocarbon exposures in patients with renal cell carcinoma,Can Med Assoc J.140; 1309-1318

 

Smith EM, Miller ER, Woolson RF and Brown CK (1985): Bladder cancer risk among laundry workers, dry cleaners and others in chemically related occupations,J Occup Med.27; 295-297

 

Stemhagen A, Slade J, Altman R and Bill J (1983): Occupational risk factors and liver cancer. A retrospective case - control study of primary liver cancer in,Am J Epidem.117; 443-454

 

Suarez L, and Martin J (1989): Primary liver cancer death and occupation in,Am JMed.15; 167-175

 

Vaughan TL, Stewart PA, Davis S and Thomas DB (1997): Work in dry cleaning and the incidence of cancer of the oral cavity, larynx and oesophagus.Occup Environ Med.54; 692-695

 

 

Case reports

There is one case report, describing a dry-cleaner exposed to trichloroethylene for 8 years, then to tetrachloroethylene for 24 years, who developed acute myeloid leukaemia (Jalihal and Barlow, 1984). It is clearly impossible to ascribe the occurrence of this disease to tetrachloroethylene.

 

References

 

Jalihal SS and Barlow AM (1984): Leukemia in dry cleaners,J R Soc Health.104; 42

 

 

Summary of human carcinogenicity

 

Several large-scale studies of cancer mortality and cancer incidence among workers exposed to tetrachloroethylene in dry-cleaning and laundering have been reported.  In addition there are two general population drinking water studies, and a number of case-control studies.  The latter have been uninformative with regard to the carcinogenic potential of tetrachloroethylene in humans.

 

The occupational mortality studies have generally been limited by the lack of clear information on the levels of exposure to tetrachloroethylene, and by possible confounding due to exposure to other solvents among the subjects studied. In this regard, the widespread use of tetrachloroethylene in the dry cleaning industry did not begin until the 1960s, and assuming a latency for tumour development of 15-20 years, the cancer deaths observed in some of the studies for periods up to 1980-82, if occupationally related, would be attributable to exposure conditions prior to the widespread use of tetrachloroethylene.

 

In those studies where cancer mortality could be distinguished among workers employed in establishments using tetrachloroethylene only, no increases in the risks for liver and renal cancers were observed.

In this sub-cohort of tetrachloroethylene-exposed dry-cleaners, elevated mortality was seen only in relation to cancer of the oesophagus, buccal cavity and pharynx, and tongue. The anatomical location of these tumours suggests a role for a locally acting carcinogen, which is not supported by the available genotoxicity and animal carcinogenicity database. Latency considerations for this study also argue against a possible role for tetrachloroethylene.

 

In cancer incidence studies, no increases in risks from exposure to tetrachloroethylene for any specific type of cancer, including risks for liver and renal cancers, have been observed.

 

Furthermore, a recent critical review of the epidemiological literature on occupational exposure to tetrachloroethylene and cancer has considered that the current evidence does not support a conclusion that occupational exposure to tetrachloroethylene is a risk factor for cancer of any specific site, although a firm conclusion cannot be drawn based on the available data. Similarly, no convincing evidence is available for an increased risk of cancer (total cancer, or any specific type of cancer) arising from exposure to tetrachloroethylene in drinking water. Overall, a variety of epidemiological studies have shown no good evidence for an increased risk of carcinogenicity in humans resulting from exposure to tetrachloroethylene. 

Applicant's summary and conclusion

Conclusions:

Several large-scale studies of cancer mortality and cancer incidence among workers exposed to tetrachloroethylene in dry-cleaning and laundering have been reported. In addition there are two general population drinking water studies, and a number of case-control studies. The latter have been uninformative with regard to the carcinogenic potential of tetrachloroethylene in humans. Overall, a variety of epidemiological studies have shown no good evidence for an increased risk of carcinogenicity in humans resulting from exposure to tetrachloroethylene.

Executive summary:

Several large-scale studies of cancer mortality and cancer incidence among workers exposed to tetrachloroethylene in dry-cleaning and laundering have been reported. In addition there are two general population drinking water studies, and a number of case-control studies. The latter have been uninformative with regard to the carcinogenic potential of tetrachloroethylene in humans.

The occupational mortality studies have generally been limited by the lack of clear information on the levels of exposure to tetrachloroethylene, and by possible confounding due to exposure to other solvents among the subjects studied. In this regard, the widespread use of tetrachloroethylene in the dry cleaning industry did not begin until the 1960s, and assuming a latency for tumour development of 15-20 years, the cancer deaths observed in some of the studies for periods up to 1980-82, if occupationally related, would be attributable to exposure conditions prior to the widespread use of tetrachloroethylene. In those studies where cancer mortality could be distinguished among workers employed in establishments using tetrachloroethylene only, no increases in the risks for liver and renal cancers were observed. In this sub-cohort of tetrachloroethylene-exposed dry-cleaners, elevated mortality was seen only in relation to cancer of the oesophagus, buccal cavity and pharynx, and tongue. The anatomical location of these tumours suggests a role for a locally acting carcinogen, which is not supported by the available genotoxicity and animal carcinogenicity database. Latency considerations for this study also argue against a possible role for tetrachloroethylene.

In cancer incidence studies, no increases in risks from exposure to tetrachloroethylene for any specific type of cancer, including risks for liver and renal cancers, have been observed. Furthermore, a recent critical review of the epidemiological literature on occupational exposure to tetrachloroethylene and cancer has considered that the current evidence does not support a conclusion that occupational exposure to tetrachloroethylene is a risk factor for cancer of any specific site, although a firm conclusion cannot be drawn based on the available data. Similarly, no convincing evidence is available for an increased risk of cancer (total cancer, or any specific type of cancer) arising from exposure to tetrachloroethylene in drinking water.

Overall, a variety of epidemiological studies have shown no good evidence for an increased risk of carcinogenicity in humans resulting from exposure to tetrachloroethylene.