Cancer induction following single and multiple exposures to a constant amount of vinyl chloride monomer.

Vinyl chloride monomer (VCM), already identified as a human animal carcinogen, was selected as a model agent to explore an area of concern for single and intermittent low level exposure. In traditional cancer bioassay, animals are repeatedly exposed over their lifespan to a dose of suspected chemical. In the current studies rats and mice were exposed in an inhalation chamber to single one-hour doses of VCM ranging from 50 to 50,000 ppm. A second group was given 10 one-hour exposures to 500 ppm or 100 one-hour exposures to 50 ppm of the same chemical. All animals were then observed for the remainder of their lives, generally 18-24 months. Moribund animals were euthanized, and survivors were sacrificed on schedule and their tissues examined for pathological changes. Specifically, the oncogenic study demonstrated dose related effects for single one-hour exposure of VCM at high levels, i.e., 5,000 and 50,000 ppm. These concentrations increased the incidence of pulmonary adenomas and carcinomas in mice. Repeated exposure of A/J mice to the same chemical at 500 ppm X 10 one-hour exposures also increased the incidence of pulmonary adenomas and carcinomas which are considered highly one-hour exposure, no significant increase in tumors was observed. Rats exposed to identical concentrations of VCM failed to elicit a tumorigenic response.


Introduction
In the last decade there have been concerted efforts by industry and the Federal government to identify carcinogenic substances in the workplace. However, carcinogens are not limited to occupational exposures. Many of these same chemicals are also used in the formulation of household products. Therefore, consumers in all walks of life may be exposed to suspect chemicals in their homes without *U.S. Consumer Product Safety Commission, Westwood Towers Building, Room  their knowledge. This type of consumer exposure is usually brief and at very low levels. The suspect chemical may be modified in the process of manufacturing the consumer product, i.e., as in the case of polymerization of vinyl chloride (VC) to poly(vinyl chloride) (PVC), or trace quantities of the unreacted monomer may be present in the product which may be leached out under normal conditions of use. There are other situations wherein a chemical like VC is used because it is chemically inert, for example as a propellant, and as an unreacted gaseous chemical it could become a potential health problem. While regulatory bodies try to establish safe exposure levels and industry attempts to control exposures in the production facilities, no such control measures would be of practical value in the October 1981 home environment. Another complication associated with making value judgements about potential risks of consumer exposure is that traditional cancer bioassays involve daily exposures to the test chemical over the animal's lifetime. There were no comparable studies on short-term exposure to a carcinogen like VC followed by a lifetime monitoring for toxicological symptoms with complete histopathological examination at death. Therefore, our approach was to explore what happens to laboratory animals following brief or intermittent exposure to a known carcinogen like VC. Vinyl chloride appeared to be the ideal chemical to use in these experiments because of its widespread use in industrial and consumer products. Furthermore, the relationship for inducing unique pathological lesions, i.e., angiosarcoma of the liver for high exposure levels in animal and man was already established by Viola et al. (1) Maltoni (2) and Creech et al. (3).
Quite apart from the need to establish a bench mark for acute short-term intermittent exposures to a carcinogen, such as vinyl chloride, was the question of "threshold effects" or "safety factor" for low level exposures. In setting up our experiments we were mindful of the many factors that could have an impact on the study design. Some of the more important factors are that (1) the substance may not reach a susceptible cell; (2) the substance may make noncarcinogenic biochemical combination with the cell; (3) the "initiated" lesion may not receive adequate "promotion"; (4) host factors may be unfavorable to carcinogenesis; (5) biochemical repair of the DNA lesion may occur; (6) morphological regression of tumorigenic proliferation may occur; (7) carcinogenic cells may be destroyed by the body's immune system. These items need to be addressed in any conventional cancer bioassay but are even more critical in searching for noncarcinogenic exposure levels.

Experimental Conditions
Two strains of rats, Fischer 344 (Charles River Laboratory, Wilmington, Mass.) and Sprague-Dawley/Wistar (Chemical Systems Lab, Edgewood, Md.), and two strains of mice, A/J (Jackson Lab, Bar Harbor, Maine) and ICR (Charles River Laboratory) were treated in single or multiple intermittent inhalation exposures to VC. The chambers were Rochester type, stainless steel, 1000 liter, constructed to provide laminar air flow and insure uniform exposures to VC to test animals.
Chamber concentrations were established by proportioning the amounts of VC being dispersed with the air flow through the chamber. Airflow 64 through the chamber was created by a blower motor located in the flow pipe on the exhaust side of the chamber. A negative pressure was maintained in the chamber at all times when operational and the exhaust gas completely filtered (M6AI gasparticulate filter) before discharge to the environment. The concentration of gas in the inhalation chamber was monitored by using a Hewlett-Packard 5830A gas chromatograph with a dual flame ionization detector.

Exposure Procedures for VC Lifetime Cancer Studies
Male and female rats and Fischer 344 and ICR mice were totally exposed for 1 hr to 50, 500, 5000 and 50,000 ppm VC. Fischer rats and A/J mice equally divided by sex received ten 1 hr exposures to 500 ppm VC (1 hr/day, 5 days/wk for 2 weeks) or 100 1-hr exposures to 50 ppm (1 hr/day, 5 days/wk for 20 weeks). Male and female parents (Sprague-Dawley/Wistar) rats (obtained from Veterinary Medical Division, Edgewood Arsenal, Md.) from the reproduction study were also maintained and observed for 24 months post exposure to 50 to 500 ppm VC 1 hr per day, 5 days per week for 10 weeks (49 exposures).
Following exposure, all animals were air washed until the chamber concentration was less than 1 ppm VC. Animals of the same sex were segregated in stainless steel cages with a maximum number of 5 rats in each compartment and placed in the animal holding area. Mice were similarly treated, but smaller cages were used and the number of mice per compartment was limited to two. The animals were observed twice daily for general health, sores, masses, alertness, activity or mortality. All animals were weighed weekly for the first 8 weeks post exposure and monthly thereafter. The exposure schedules for the studies described above are shown in Tables 1-3. No blood chemistries or hematology studies were performed.

Pathology
Gross and Light Microscopy. A complete gross and microscropic examination was performed on most control and exposed animals which died or were sacrificed. Autolysis precluded such examination in a few cases.
All rodents were to be serially sacrificed at 8, 16 and 24 months post exposure. However, the life span of the mice forced some changes in the later times of sacrifice and termination of mouse experiments. For single exposure the planned 16 and 24 month sacrifice were replaced by 18 month sacrifices.
Environmental Health Perspectives

October 1981
This observation period was the minimum suggested by the National Cancer Institute (4) for cancer bioassays in small animals. For multiple dose studies in mice the final sacrifice was at 20 months rather than 24 months. The change was made in consideration of the risk of animal loss through death and possible cannibalism. Tissues examined were lung, trachea, heart, liver, stomach, small intestines, spleen, kidney, bladder, bone marrow (sternum), adrenals, pancreas, duodenum, brain, eye, zymbal gland, ear, nose, muscle and bone (femur).
Electron Microscopic Studies. Groups of five male and five female Fisher (344) rats from each of the single 1-hr exposure studies (50, 500, 5000 and 50,000 ppm) with equal numbers of their corresponding control groups were sacrificed at 8, 16, 24 months.
Groups of five male and five female Fischer (344) rats from the multiple 1-hr exposure studies at 50 and 500 ppm VC along with equal numbers of control animals were sacrificed at 16 and 24 months.

Toxicity During Exposure
Rats and mice exposed for 1 hr to concentrations of VC ranging from 50 to 50,000 ppm or for repeated exposures, i.e., 50 ppm or 500 ppm, produced no remarkable signs of toxicity with the exception of mice at the 50,000 ppm level. At 50,000 ppm VC 50% of the male mice exhibited hyperventilation at 45 minutes together with twitching and possible ataxia. Female mice became hyperactive after 40 min exposure, and respiratory difficulty and ataxia was observed in approximately 25% of the female mice after 55 min. Upon removal from the test atmosphere, all animals recovered to normal appearance within 24 hr. After exposure there were no consistent or dose-related differences between control and exposed (single or multiple) mice or rats in death rate, toxic signs or change in body weight.

Gross Pathology
There was a suggestion of higher frequency of masses in the lungs and livers of mice exposed once or repeatedly to VC at the higher dose levels, i.e., 500, 5000 or 50,000 ppm.
Histological Examination of ICR Mice: Single Exposure Changes ascribable to vinyl chloride were apparent primarily in the lungs with the induction of pulmonary adenomas and pneumonitis. Pneumonitis was evident in all animal groups which were exposed to VC at 500 ppm and above.
The development of bronchio-alveolar adenomas increased with exposure to higher dose levels of VC. Tables 4 and 5 summarize the significant histological changes observed in ICR mice at 8 and 18 months following single exposure to graded doses of VC.
Mice were more susceptible than rats to pneumonitis following exposure to VC. The effect however was not incremental with dose level. The incidence of pneumonitis, adenoma and carcinoma in ICR mice following single exposure is presented in Table 6. No correlation was observed between the incidence of pneumonitis and adenoma or pneumonitis and carcinoma. However, males seemed more prone to the induction of pneumonitis, particularly at 50,000 ppm, i.e., 34% M vs. 13% F. At 5000 ppm and below, male and females were about equally sensitive.
There was an increase in bronchio-alveolar adenomas with exposure to higher doses of VC and the condition manifests itself more frequently in males, i.e., at 50,000 ppm: 51% M vs. 18% F. This trend continued at the 5000 ppm: 22% M vs. 13% F. Males and females were equally susceptible at dose level 500 ppm and below. The upper respiratory tract (nasal turbinates) and trachea revealed no unusual changes specifically attributable to VC. aPneumonitis was evident in all animal groups which were exposed to VCM at doses of 500 ppm or more.

66
Environmental Health Perspectives Histological Examination of A/J Mice at 8, 16 and 20 Months Following Multiple Exposure 500 ppm, 10 days, I hr. Changes ascribable to VC were apparent in the lung only, primarily the induction of pulmonary adenomas. Bronchio-alveolar adenomas were induced in the test group with approximately equal frequency in males and females, i.e., 73.7% M vs. 75.6% F, respectively. The increase in incidence as shown in Table 7 for the test group over that of the controls was substantial, by a factor of 2.2, i.e., controls, 34.4%; 500 ppm; 74.7.
Progression to malignancy (carcinoma) in the test group vs. controls was also highly significant: controls, 3/90 or 3.3%; 500 ppm, 22/166 or 13.3%. In animals scheduled for evaluation (survivors) pulmonary adenomas were observed as early as 8 months post exposure in 60% of males and females; at 20 months, 75% of animals were affected.
The incidence of pneumonitis, adenoma and carcinoma in A/J mice following multiple exposure to VC is presented in Table 8. Pneumonitis was observed more frequently in the control animal than Table 5. Overall summary: incidence of nonneoplastic changes and histologically proven neoplasms within the liver and lungs of ICR Swiss mice exposed to vinyl chloride in single inhalation exposures.  in either test group. No correlation was observed between the incidence of pneumonitis and adenoma or for pneumonitis and carcinoma. These results are similar to those observed in the single exposure study. 50ppm, 100 days, I hr. Again changes attributable to VC were apparent in the lung only, primarily pulmonary adenomas. However, the incidence in the induction of adenomas and progression to carcinoma are considered only marginal and not statistically significant.
Comparatively, the potential for development of pulmonary adenomas as shown in Table 7 was greater in A/J mice following multiple exposures at 500 ppm than at 50 ppm, i.e., incidence at 500 ppm, 74.7%; at 50 ppm, 44.1%, despite an equivalent total dose of5000 ppm VC. Also, pulnonary adenomas were induced earliest (8 months) following multiple exposures at 500 ppm. It is also of interest that the bNo significant difference for pulmonary adenomas (p < 0.14) and carcinomas (p 2 0.19) observed at the lower multiple exposure level. control groups of A/J mice for both multiple inhalation studies showed a baseline incidence for pulmonary adenomas which was nearly identical, i.e., at 500 ppm, 34.4%; at 50 ppm, 34.5%. Table 9 provides an overall summary of the incidence of histologically proven neoplasms observed in both strains of mice, i.e., ICR and A/J, following single and multiple exposures to VC. The neoplastic and nonneoplastic changes observed in the liver and/or lungs of A/J mice following multiple exposures to VC at 50 and 500 ppm are shown in Tables 10 and 11. Although other neoplasms and nonneoplastic changes occurred variously in all remaining organs and tissues, the response appeared either sporadically or was shared by all test groups including controls. Relationship by incidence and severity to test exposure was not evident. Furthermore, morphologic deviations were not unlike those normally observed in aging A/J mice maintained under standard laboratory conditions.

Electron Microscopic Results
In general, these studies indicate that exposure to vinyl chloride increased organelle turnover as well as loss of volume control (bleb formation) and increased lysosomal activity in the liver of rats. These alterations progressively decreased as recovery after exposure increased.
Hepatocellular carcinoma was seen in one male Fischer rat which had received 10 exposures of 500 ppm. Lymphosarcoma was noted in one female Fischer rat which had received a single exposure at 500 ppm. Since these were individual cases and since no cancers were seen at 50,000 ppm, the lymphosarcoma and the hepatocellular carcinoma are not likely related to vinyl chloride exposure.

Reproduction Carcinogenesis Study
One hundred and fifty colony rats (Sprague-Dawley/Wistar) were divided into three groups with equal numbers of males and females. These specific-pathogen-free, random bred animals were obtained from the Animal Resources Branch at Edgewood Arsenal. The rats were 12 weeks old when initially exposed to VC. One hundred rats were exposed as described and the remaining 50 were carried as unexposed controls. The parental generations of S-D/W rats were maintained 24 months post exposure for carcinogenic evaluation.
A complete gross and microscopic pathological examination of all tissues was performed on each control and exposed animal. Particular emphasis was placed on examination of brain, lung, and liver tissues and zymbal gland in the rodent ear. The livers of randomly selected rats were prepared and examined by electron microscopic techniques.
Neoplastic and nonneoplastic lesions were observed in approximately equal frequency in control and Fo parent generation of Sprague-Dawley/Wistar rats exposed to 50 ppm or 500 ppm of VC, 1 hr per day, five days per week for 10 weeks (49 exposures). The only lesions that occurred in higher frequency in the exposed animals than in control rats were eosinophilic cellular alterations presented as foci/ or areas. The appearance of these foci was related to the dosage of VC. The nature of these lesions are of interest but as yet are controversial, so that no inference can be drawn.

Discussion
The fact that the severity of carcinogenic effects of VC had been described by various investigators like Maltoni (2,5) and Lee et al. (6) to coincide with dose and length of exposure implies that the total dosage (concentration x exposure time) may be an important factor in the carcinogenicity of VC. Therefore, the inhaled dose was approximated by the Haber (7) concept. In its simplest form this concept states that the dosage, Ct is the product of C (concentration in milligrams/cubic meters) and t (time in minutes). The total concentration can also be expressed in parts per million (ppm) and the time can be expressed in hours, producing Ct in ppm-hr. Factors for breathing rate and detoxication can be added. However, this simplified Ct approximation of total inhaled dose is relatively useful, as is, for comparative purposes.
In his experiments Maltoni (5, 10) exposed rats and mice as shown in Table 12 to a series of doses of VC ranging from 50 to 10,000 ppm for 4 hr/day, 5 days per week for 52 weeks. The results indicate a questionable carcinogenic effect at 50 ppm or total dosage of 52,000 ppm-hr after 135 weeks. In another experiment (BT3), Sprague-Dawley rats, treated in a similar fashion to BT1 but for 17 weeks only, show after 86 weeks a negative carcinogenic re- Table 11. Overall summary: incidence of nonneoplastic changes and histologically proven neoplasms within the liver and lungs of A/J mice exposed to vinyl chloride in multiple inhalation exposures.   (1), tumors were seen in rats which had been exposed to 30,000 ppm VC, 4 hr/day, 5 days/week, for 12 months. The total dose (Ct) of VC was 28,800,000 ppm-hr. In the studies of Caputo, Viola and Bigotti (8), rats and rabbits were exposed to VC for 4 hr/day, 5 days/week for 12 months at six dose levels. The exposure to 50 ppm VC or total dose of 48,000 ppm-hr produced no tumors. Carcinogenic effects were observed at total dosage of 320,000 ppm-hr and above for rats and at 7,800,000 ppm-hr for rabbits.
Keplinger et al. (9) exposed rats, hamsters and mice to VC. Only the data on mice were sufficiently complete for examination of total dose effect. Total dosage of 56,000 ppm-hr and above produced tumors in mice. The final report on the rats and hamsters is still unavailable for evaluation.
Lee et al. (6) exposed mice and rats to three levels of VC: 50, 250 and 1000 ppm, 5 hr/day, 5 days/week for up to 12 months. All tests with a total dose of 48,000 ppm-hr and below were essentially negative for mice. Positive carcinogenic effects were noted, however, in mice above 78,000 ppm-hr.
The results of our study indicate that a positive carcinogenic effect in rats may not appear until the total dose (Ct) of VC exceeds 50,000 ppm-hr. The single exposure study with ICR mice was negative at 50 and 500 ppm-hr, borderline at 5000 ppm-hr but did produce neoplastic lesions in the lung at 50,000 ppm-hr. The A/J mice exposed to multiple doses of VC, i.e., 50 ppm x 100 days x 1 hr or 500 ppm x 10 days x 1 hr for a total cumulative dose of VC of 5000 ppm-hr show a significant tumorigenic response, but only at the higher dose level.
In our studies Fischer rats exposed to a total dosage of 50, 500, 5000, and 50,000 ppm VC for 1 hr showed no chemically induc-ed tumor response. Neither did the Sprague-Dawley/Wistar rats that were exposed to 500 ppm VC 1 hr/ day, 5 days/week for 10 weeks (total dosage 24,500 ppm-hr). Table 13 summarizes various investigators' test results, including our own, on vinyl chloride. The tests with ICR mice, single exposure to VC, show an increased frequency of adenomas at total dosage of 5000 ppm-hr and above. For A/J mice, repeated exposures to 50 and 500 ppm VC for a total dosage of 5000 ppm-hr shows a similar dose response pattern for adenomas. In general, in terms of total dosage Ct, carcinogenic effects are seen in the two mice strains at Ct levels of 5000 to 50,000 ppm-hr. Thus total dosage for carcinogenicity in mice is in general agreement with Lee et al. (6) (78,000 ppm-hr) and Maltoni (5) (between 30,000 and 150,000 ppm-hr) and other investigators.
There is no doubt that risk is related to length of exposure and hence total dose. In our discussion much attention has been focused on the risk of cancer associated with total dosage of VC. However, the multiple exposure experiment at 50 and 500 ppm dose levels with A/J mice for an equivalent total exposure to 5000 ppm-hr appears inconsistent with the thesis of total dosage. Indeed, it appears from our own data that concentration may be the dominant factor for acute or low level intermittent exposures. This result may be explained on the basis of a number of factors, i.e., metabolism, detoxication, DNA repair or a time for tumor development for low level exposure beyond the animals' lifespan.
As is evident from Table 13, the two strains of rats, Fischer (344) and S-D/W, were more resistant to the adverse effects of single and multiple intermittent exposures of VC.
Our studies are in agreement as to the dose-time relationship for carcinogenesis related to the VC exposure. All of the continuous exposure studies considered collectively indicate that there is a lifetime total dose (Ct) for VC. However, from our own studies with single or intermittent low level exposures to VC we believe that concentration may be the most dominant factor in whether or not carcinogenic effects are observed. For single dose studies with VC in ICR mice, neoplastic lesions were produced at 5000 ppm. For multiple intermittent exposures studies with A/J mice the critical concentration for VC was 500 ppm, total dosage 5000 ppm-hr.
Approximate carcinogenic Ct levels of VC based on data for mice and rats are Ct = 5,000-50,000 ppm-hr, carcinogenic tendencies; 500,000 Ct > 50,000 ppm-hr, definite carcinogenicity; Ct > 500,000 ppm-hr, high incidence of carcinogenicity.

Considerations of Carcinogenicity of VC: Conclusions
Cancer seems dependent on total dose of vinyl chloride, especially in life-time exposure studies, but for short-term exposure the concentration may be the most critical factor.
One dose is sufficient if dose is high enough. The carcinogenic total dose was 5000 ppm-hr for mice and 50,000 ppm-hr for rats.
There were apparent noncarcinogenic doses in the study.
Mice were more sensitive indicators than rats for carcinogenic effects of vinyl chloride.