Review of experimental carcinogenesis by compounds related to vinyl chloride.

The experimental carcinogenesis results in six compounds related to vinyl chloride are reported. Vinylidene chloride, given by inhalation, was carcinogenic in male CD-1 mice, male CD rats, Sprague-Dawley rats and male Swiss mice. Trichloroethylene, given by gavage and inhalation, was carcinogenic in the B6C3F1 mice. When given by gavage, perchloroethylene was carcinogenic in the B6C3F1 mice, and dichloroethane was carcinogenic in Osborne-Mendel rats and B6C3F1 mice. Dibromoethane, given by gavage and inhalation, was carcinogenic in B6C3F1 mice, F344 rats and Osborne-Mendel rats. Finally, epichlorohydrin was carcinogenic in male Sprague-Dawley rats and B6C3F1 mice.


Introduction
Since the carcinogenicity of vinyl chloride (VC) was demonstrated in experimental (1) and epidemiologic studies (2), data on the potential carcinogenicity of vinyl chloride structural analogs and related compounds have been increasing. This paper will attempt to review the experimental carcinogenicity data on a number of these compounds, while another paper (3)  A summary of some of the carcinogen bioassays of vinylidene chloride is given in Table 1. The strongest evidence of a carcinogenic effect was in a bioassay involving Swiss mice performed by Maltoni et al. (4,5). In this study, 24/150 dosed male mice had kidney tumors as compared to 0/190 in the controls. In the study by Lee et al. (6,7), dosed CD-1 mice and CD rats had two or three liver angiosarcomas. Since this type of tumor was induced by vinyl chloride in humans and rodents and the study was terminated after only one year, the results are viewed as biologically significant. In an inhalation study in Sprague-Dawley rats by Maltoni et al. (4,5), increases in mammary gland tumors in the dosed animals were reported. However, these incidences did not follow a dose-response relationship.
In contrast to these positive studies, a number of nonpositive two-year carcinogen bioassay studies have been reported (6)(7)(8)(9). Most notable is this initial report that preliminary evaluations of vinylidene chloride in a two-year National Toxicology Program (NTP)/NCI bioassay of Fischer 344 rats, given at 2-10 mg/kg of body weight and of B6C3F1 mice, given at 1-5 mg/kg of body weight, did not indicate any significant increased incidences of tumors in the dosed animals as compared to controls.

Trichloroethylene (TCE)
A summary of some of the carcinogen bioassays of trichloroethylene is reported in Table 2. In B6C3F1 mice, this compound, given by gavage, was found to be carcinogenic, inducing hepatocellular tumors (10). In confirmation of this finding, an inhalation study in B6C3F1 mice, performed by Biotest Laboratories, gave similar results (11).
Bioassays of TCE were nonpositive in two strains of rats, the Sprague-Dawley and Osborne-Mendel (10)(11)(12). Additional NTP/NCI carcinogen bioassays are under way in five strains of rat and the B6C3F1 mice to examine species and strain differences.

Tetrachloroethylene (Perchloroethylene) PCE
The carcinogen bioassay results of testing PCE are given in Table 3. In B6C3F1 mice, this compound, given by gavage, was carcinogenic inducing hepatocellular tumors.
Studies in two strains of rats, Osborne-Mendel and Sprague-Dawley, produced no carcinogenic effects (13)(14). Additional carcinogen bioassays in four strains of rats and the female B6C3F1 mice are currently underway at NTP/NCI.

1,2-Dichloroethane (DCE)
The experimental carcinogenicity results of testing DCE are given in Table 4. In B6C3F1 mice, the  Rat, Spr-Daw compound, given by gavage, was carcinogenic, inducing lung tumors in both sexes and mammary gland and uterus tumors in dosed female animals (15). The chemical was also carcinogenic in Osborne-Mendel rats, giving forestomach tumors and extrahepatic hemangiosarcomas in both sexes, sub-cutaneous fibromas in the males and mammary gland tumors in the females (15). In contrast, a preliminary report by Maltoni (12) indicated that DCE was not shown to be carcinogenic in an inhalation study with Swiss mice and Sprague-Dawley rats. Since there were strain and  Assay still in progress Environmental Health Perspectives (11) (10) dose differences as well as route of administration differences between the NTP/NCI and Maltoni studies, the causes of the differences in the experimental results are not readily apparent.

1,2-Dibromoethane (DBE)
The results of some carcinogen bioassays on DBE are given in Table 5. In bioassays on B6C3F1 mice and Osborne-Mendel rats, DBE given by gavage was carcinogenic, inducing multiple tumors in each species and sex. There were not only chemically induced tumors at the site of application (forestomach tumors) but also tumors distant from the site of application (lung tumors in mice and extrahepatic hemangiosarcomas in male rats and hepatocellular tumors in female rats) (16).
In addition, a preliminary analysis of an inhalation study of DBE performed by NTP/NCI indicates that nasal cavity tumors were found in dosed rats and dosed female mice. Furthermore, elevated incidences of lung tumors were found in the dosed mice as well as elevated incidences of mammary gland tumors in dosed females of each test species. Mesotheliomas were also induced by DBE in male rats.

Epichlorohydrin (EPC)
The bioassay results on epichlorohydrin are given in Table 6. Recently, this compound, given by October 1981 inhalation, was found to be carcinogenic inducing nasal cavity tumors in male rats (17). In addition, experiments by Van Duuren (18)(19) indicated that EPC was an initiator in a two-stage study and induced local sarcomas by subcutaneous injection. addition more comprehensive epidemiological studies involving the manufacture, production and use of these compounds should be undertaken.
A summary of some of the bioassay results for each compound by species/strain is given in Table 7.

Discussion
The data on these compounds can generate some interesting points for discussions, such as the probable mechanism of some of these compounds, the controversy over TCE results and suggestions for future action.
In the first case, studying the summary table indicates that the compounds (DBE, DCE, and EPC) which are proposed to alkylate directly, gave forestomach tumors when given by gavage or nasal cavity tumors when given by inhalation while the compounds which are proposed to require metabolic activation did not produce these types of tumors. One can envision that these compounds, due to their alkylating ability and their ability to induce site of application tumors, are direct-acting carcinogens. Two of these former compounds, DBE and    the TCE. The doses of TCE were about 2000-1000 mg/kg of body weight, corresponding to less than 2 mg/kg of body weight of epichlorohydrin. If epichlorohydrin is the active agent, one might expect forestomach tumors from a gavage study because of its alkylating ability. However, the target site in the mice is the liver. In addition, if epichlorohydrin is the carcinogen, then why are the results negative in the rat? Direct-acting carcinogens, like epichlorohydrin, are less likely to show such a species difference than compounds which would require metabolic activation, such as TCE itself. Thus, an alternative explanation is that there is a species difference reflecting some type of metabolic or target site sensitivity difference between the species. The answer may be forthcoming. NTP/NCI is testing TCE without EPC in the B6C3F1 mice and five strains of rat. If EPC is the active compound, the mice study should not show a carcinogenic effect. However, if TCE is the active agent then the B6C3F1 mouse study should be positive. The results in the rats will also give insight into whether there is a species or a species/strain difference for TCE. Similar types of studies are also underway with PCE. In addition, Maltoni has indicated that he has started bioassays on TCE with B6C3F1 mice and Swiss mice.
With the possible resolution of the TCE issue, there still remains the differences in experimental results involving vinylidine chloride (VDC). The results in VDC may indicate that there are important species and strain differences. As a consequence, this compound may be an ideal candidate for studying species/strain differences systematical-October 1981   ly. In addition, the differences in the NCI and Maltoni studies on DCE could be the source for additional research.
From an examination of Table 7, the compounds with positive results in multiple species yielding multiple tumor sites are the direct acting carcinogens, DBE and DCE. (EPC has not been tested