Carcinogenicity studies on halogenated hydrocarbons.

A series of halogenated compounds was tested by oral intubation in 200 Osborne-Mendel rats and 200 B6C3F1 mice of both sexes. Carbon tetrachloride, used as a positive control, induced liver and adrenal tumors in mice and neoplastic nodules in the livers of rats. 1,2-Dibromoethane and 1,2-dibromo-3-chloropropane caused stomach tumors with many metastases in both rats and mice. Chloroform, known to cause hepatocellular carcinomas in mice, led in addition to kidney tumors in male rats. 1,2-Dichloroethane was much weaker than the analog, 1,2-dibromoethane, and induced only a few stomach tumors in rats. It increased liver and lung tumors in mice. Most of the compounds, namely, trichloroethylene, 1,1-dichloroethane, 1,1,2-trichloroethane, hexachloroethane, and tetrachloroethylene, increased hepatocellular carcinomas in mice but had little or no action in rats. Iodoform tended to increase thyroid tumors in male rats and hepatocellular carcinomas in male mice. The action of 3-chloropropene was questionable. No tumors could be attributed to 1,1,1-trichloroethane (methyl-chloroform).


Introduction
In the 1960's, scientists associated with the Bioassay Program of NCI realized that although many halogenated aliphatic compounds had widespread environmental uses, they had not received adequate chronic toxicity studies. Furthermore, at that time there were fewer regulations on such substances since they apparently did not fall in the province of the FDA, then almost the sole agency involved with such matters. Therefore when the opportunity arose to test chemicals under the contract system, a study of halogenated compounds with varied uses was initiated.

Chemicals
The chemicals were generally purchased from laboratory supply houses. Purity of each was checked by gas chromatography and infrared spectroscopy. Gas chromatography-mass spectrometry *Carcinogen Metabolism and Toxicology Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20014. was used to identify any minor components found in each chemical. In many cases the minor components were stabilizers usually added to commercial formulations.

Animals and Conditions
Osborne-Mendel rats (Battelle Memorial Institute, Columbus, Ohio) were chosen because previous studies by FDA scientists (1) and by Reuber and Glover (2) had shown this strain was sensitive to various chlorinated compounds such as DDT or carbon tetrachloride. The B6C3FI mouse, a hybrid of the C57B11/6 female and C3H/He male (Charles River, Wilmington, Massachusetts) was selected because of previous extensive use in NCI bioassays (3). The 35-day-old rats and 25 day-old-mice, as received from the suppliers, were quarantined for 10 days and randomly assigned to treatment groups. Rats were individually housed in wire mesh galvanized steel cages; mice in groups of 10 of one sex in polypropylene cages (fitted with filter bonnets) on Sanichips bedding. Cages for rats were changed weekly; for mice twice each week. The temperature range was 20-24°C; air was changed 10-15 times per hour, while relative humidity was kept in the 45 -55% range. Feed (Wayne Lab-Blox meal) and water were freely available.

December 1977
Acute Study For each compound single dose range finding studies were done in two male rats and two female mice. Up to 10 individual dose levels in corn oil were given by gavage; the animals were then observed for 14 days. The lowest doses causing death were selected as the highest level for the 8-week subchronic study, the primary objective of which was to determine the maximum tolerated dose (MTD) for the chronic test. In the subchronic phase, six groups of five males and five females of each species were used. Five groups received the test compound at varying dosages; the remaining group received the corn oil vehicle only. Dosing was by gavage, 5 days weekly for 6 weeks. The weekly weights served as a guide for the dosage during the next week. After the 6-week dosing period, the animals were observed an additional 2 weeks to detect delayed toxicity and/or the rate of recovery. Animals were then killed and a gross necropsy was done to detect abnormalities.

Chronic Test
Based on the results of the subchronic study, two dose levels of each compound were chosen to be given lo both sexes of Osborne-Mendel rats and B6C3F1 mice, in groups of 50 animals each, a total of 400 animals in eight groups. There were also 99 male and 98 female rats, 77 male and 80 female mice as untreated colony controls. Each compound as a corn-oil solution was administered by gavage under a hood for 5 consecutive days a week. Carbon tetrachloride was administered as a positive control compound.
The animal weights and food consumption per cage were obtained weekly for the first 10 weeks and monthly thereafter. Animals were checked daily for mortality. In cases where mortality was high or weight gain was lower than within 10% of the controls, the dose levels of a compound were reduced or dosing was on an intermittent schedule to allow survival. Generally dosing of the animals was stopped after 78 weeks. After an additional 12 weeks on observation or a total of 90 weeks, the mice were terminated; at 110 weeks the rats were killed. A gross necropsy was performed on each animal that died during the experiment or was killed at the end. Specified organs* plus any tissue con-*Brain, pituitary, adrenal, thyroid, parathyroid, trachea, esophagus, thymus, salivary gland, lymph nodes, heart, nasal passages, lung, spleen, liver, kidney, stomach, small intestine, large intestine, pancreas, urinary bladder, prostate or uterus, seminal vesicles and testis with epididymis or ovary, skin with mammary gland, muscle, nerve, bone, and bone marrow. taining visible lesions were fixed in 10% buffered formalin embedded in Paraplast, and sectioned at 5 ,um for slides. Hematoxylin and eosin staining was used routinely, but other stains were employed when needed. Diagnoses of any tumors and other lesions were coded according to the Systematized Nomenclature of Pathology (SNOP) of the College of American Pathologists, 1965. Further information on the experiments is included in the individual report on each compound, as well as statistical analysis of the results.

Compounds
In some compounds small amounts of stabilizers or other minor impurities were detected, generally by gas chromatography. Details of the columns, oven temperature and the like are in the individual reports. As examples, for trichloroethylene the main impurity was 1,2-epoxybutane (0.19%) while in chloroform there was 2% of ethanol, added as a stabilizer. The 1,1,1-trichloroethane sample contained approximately 3% p-dioxane.

Animal Tests
Based on body weight gains and survival rates during the acute and subchronic studies, estimated maximum dose levels .for each compound were selected (Table 1). In the chronic study, due to signs of toxicity in some of the groups, the doses were decreased in order to maintain the animals. Conversely, if there was no effect on weight gain, doses were increased. The report for each compound gives the details along with the timeweighted average dose.  Female  1000  1000 2400  1800  180  180  300  500  1500  1500  1500  1500  120  60  100  100  40  40  120  120  30  30  260  260  100  100  200  300  900  900  3000  3600  100  100  100  100  500  500  500  500  55  55  200  300  1000  1000  1000  1000  100  150  2500 2500 Data on the primary tumors at the main anatomical sites, in comparison with findings in controls, for each compound are provided in the tables.
Reports have been issued by the Carcinogenesis Program of NCI on three compounds, trichloroethylene, chloroform, and 1,1 ,1-trichloroethane (methylchloroform, (4)(5)(6). The first of these, trichloroethylene, had little or no effect in rats but hepatocellular carcinomas appeared in mice ( Table   2).
As was expected, because of a previous study on chloroform (7), hepatocellular carcinomas occurred at a high incidence in both male and female mice given this compound. On the other hand, rats showed a different pattern. Kidney tumors were increased in males and thyroid tumors in females  (Table 3). Surprisingly, the analog 1,1,1-tri-had no significant action in mice. However, male chloroethane had no noticeable effect (Table 4) alrats at the lower dose level had a slight increase in though the relatively poor survival of treated ani-thyroid tumors (Table 5). In any event, the renal mals may have precluded any carcinogenic re-action of chloroform did not hold for the iodo sponse. Still another chloroform analog, iodoform, analog. A preliminary note on the carcinogenicity of 1,2-dibromoethane and 1,2-dibromo-3-chloropropane (DBCP) has appeared (8). 1,2-Dibromoethane led to many squamous cell carcinomas of the stomach with numerous metastases in rats and mice (Table 6). In addition to causing stomach carcinomas with metastases in both rats and mice, DBCP also caused many mammary tumors in female rats at both dose levels (Table 7). However, the chloro analog of 1,2-dibromoethane, namely 1,2-dichloroethane, did not have quite such a potent effect. Some stomach tumors were present in male rats on the high dose level. Male mice showed some increase in hepatocellular carcinomas and both males and females on the high dose level had a definite increase in lung tumors (Table 8). A different structural analog or 1,l-dichlo-( Table 9). Addition of a second chlorine in the roethane showed practically no effect in rats and led 2-position as in 1,1,2-trichloroethane, afforded a to only a slight increase in liver tumors in male mice compound with little significant action in rats. How- ever, both male and female mice had a decided inmice (Table 11). Introduction of a double bond, as in crease in the number of hepatocellular carcinomas 3-chloropropene, seemed to decrease the (Table 10). Addition of more chlorine atoms to the hepatomagenic action in mice. Rats were not afmolecule afforded hexachloroethane which showed a fected significantly (Table 12). like effect-minimal action in rats, liver tumors in Tetrachloroethylene, an analog of trichoroethylene, led to similar results as trichloroethyl-ene in animals. Liver tumors were increased in mice but there was no significant increase in rats (Table   13).  Carbon tetrachloride, used as a positive control, gave a high yield of hepatocellular carcinomas in male and female mice, as had been reported earlier (9). Also of interest was the increase in adrenal tumors in treated animals. Previous reports have not mentioned this type of tumor. In the rats, carbon tetrachloride caused neoplastic nodules and a few carcinomas of the liver. However, the incidence was lower than was anticipated (Table 14).

Discussion
From the results of these experiments, it is evident that oral administration of many halogenated aliphatics leads, in many cases, to hepatocellular carcinomas in mice and lesser effects in rats. Moreover, certain of the compounds such as 1,2-dibromoethane, DBCP, and chloroform had other unique carcinogenic actions in rats. It thus seems that such compounds could pose a hazard to humans on continued exposure. Of course, humans are more likely to receive these substances by inhalation, rather than by the oral route. Nevertheless, it seems prudent, in any environmental situation, to reduce exposure to such compounds to as low a level as is technologically feasible. The experimental work reported was carried out at Hazleton Laboratories America, Inc., Vienna, Va., originally under contract NOI-CP-23278 with NCI, and currently under a subcontract to Tracor-Jitco, Inc.
The secretarial and technical assistance of Frances M. Williams and Joan Chase is appreciated.