Document V3JO2Yr6NpN51yOe57nk8Nb04

jj Jittaal. Vfli If. pp. J|? |o Ui IffI purtfit i Gtt*i Iron* ' fi-fltl-{&&> /ftI ooi.m-:m si owi'-itwcooo Pm* U4 R&S 006926 LIFESPAN ORAL TOXICITY STUDY OF VINYL CHLORIDE IN RATS* V, J. Feron, C. F. M. Hendriksen. A. J. Speer. H. P. Til and B. J. Sprr Institute CIVO--Toxicology and Nutrition TNO, PO Box 360. 3700 AJ, Zeist. The Netherlands (Received 8 July 1980) Abstract--A lifespan oral toxicity study of vinyl chloride monomer (VCM) was carried out in Wiitar rats, using five groups each of 60-80 males and 60-80 females. VCM was administered by incorporating polyvinyl chloride (PVC) powder with a high VCM content into the diet or by gastric intubation of a 10!'. VCM solution in soya-bean oiL The VCM doses (actual exposures) were 0 (control). 1-7. 5-0 and 14*1 mg/kg body weight/day provided by diets containing PVC powder, and 300 mg/kg body weight given by stomach tube as a solution of VCM in oil od 5day$/wk. The death rate was higher in all VCM-treated groups than in the controls and increased with increasing VCM doses. The 14-1* and 300-mg,"kg treatments were associated with shortened blood-clotting times, slightly increased 3-foetoprotein levels in the blood serum, liver enlargement and an increased haematopoietic activity in the spleen. A variety of neoplastic and non-neoplastic treatment-related liver lesions was found at each of the VCM levels. The changes varied from swollen and irregularly-shaped mitochondria in hepatocytes to hepato cellular carcinomas and hepatic angiosarcomas. The tumour response of the liver appeared to shift from a predominance of angiosarcomas at the highest dose level da a mixture of angiosarcomas and hepato cellular tumours at the intermediate levels to the exclusive development of hepatocellular tumours at the lowest VCM level. Tumours attributable to VCM exposure and found at other sites included pulmonary angiosarcomas, extrahepatic abdominal angiosarcomas and tumours of the Zymbal glands; these neo plasms occurred at VCM levels of 5 0 mg/kg and above. In addition, there was some evidence that VCM exposure enhanced the development of abdominal mesotheliomas and of adenocarcinomas of the mam mary glands. Thus, the present study showed that orally administered VCM is a carcinogen in rats, and that the "no-observed-adverse-effect level" in rats was lower than 1-7 mg/kg body weight/day under the rigorous conditions of continuous oral VCM exposure resulting from the release of VCM from PVC powder present in the gastro-intestinal tract. INTRODUCTION Industrial exposure to vinyl chloride monomer (VCM) has been associated with disorders such as acro-osteoiysis. non-malignant liver disease, angiosar coma and carcinoma of the liver, and tumours of the brain and lungs and of the lymphatic and haemato poietic systems (Haley. 1975; Hopkins. 1980; Monson. Peters & Johnson, 1974; Rawls. 1980; Thomas & Popper. 1975; Vale. Kipling & Walker. 1976: Waxweiler. Stringer, Wagoner. Jones. Falk & Carter, 1976). In addition, various types of tumours, such as hepatic and extrahepatic angiosarcomas, hepatocellu lar tumours, mammary-gland carcinomas. Zymbalgland tumours, nephroblastomas, brain tumours, pul monary adenomas and nasal olfactory carcinomas, as well as a series of non-neoplastic lesions in several organs, have been found in a number of animal spe cies after prolonged exposure to atmospheres contain ing VCM at sufficiently high concentrations (Basa laev. Vazin &. Kotchetkov. 1972: Feron & Kroes, 1979; Lee. Bhandari. Winston. House, Dixon & Woods. 1978: Maltoni & Lefeminc. 1975; Suzuki, 'The study was sponsored by a group of co-operating European industrial concerns, including Verband Kunststofferzeugende Industrie e.V. (FRG), Shell Nederland Chemie. Dutch Slate Mines. Akzo Zout Chemie Neder land B.V. and Dow Chemical Europe S.A. 1978; Torkelson. Oyen & Rowe. 1961: Williamson. 1976; WinelL Holmberg & Kronevi. 1976). Residual VCM present in extruded polyvinyl chlor ide (PVC) has been shown to be liable to migrate into PVC-packed foods and drinks (Daniels & Proctor. 1975; Fuchs. GawelL. Albanus & Slorach. 1975; Pot ter, 1976; Randolph. 1973; Williams & Miles. 1975). There is still only a small amount of data on the oral toxicity of VCM. In a 13-wk toxicity study, in which the monomer was dissolved in soya-bean oil and ad ministered to rats at levels of 0. 30. 100 or 300 mg/kg body weight once daily on 6 days. wk. the no-effcctlevei was conservatively placed at 30mg'kg body weight but was probably higher since the effects at the higher doses were of doubtful toxicological signifi cance (Feron. Speck. Willems, van Banum & de Groot 1975). From preliminary observations it appeared that a more practical method for chronic oral exposure of rats to VCM is the feeding of diets containing PVC powder with a high VCM content (Feron et al, 1975). Therefore, in the present lifespan study. PVC powder containing VCM w-as incorpor ated in the diet at levels to give planned daily intakes of 1. 3 or 10 mg VCM/kg body weight. This method does not allow dietary VCM intakes much higher than 10 mg/kg body weight/day, a dose that is, how ever, very high in comparison with a recent estimate of the maximum likely oral daily intake by man. i.e. 00017 pg VCM/kg body weight/day or Ol pg VCM/ r. n j. 317 318 V. J. Feuon. C. F. M, Henoai tests. A. J. SreE*. H, P. Tu. and B. J. Spit man'day (Ministry of Agriculture, Fisheries and Food. 1978). On the other hand. 10 mg VCM/kg body weight/day is low when compared with the doses used in the aforementioned 13-wk study (Feron al. 1975). Since a chronic toxicity study should include at least one effect-producing level a group of rats given VCM in soya-bean oil by gavage at a level as high as 300 mg/kg body weight/day (on 5 days/wk) was in cluded in the present experiment, despite the disad vantages of gastric intubation. EXPERIMENTAL Test materials. Vinyl chloride monomer (VCM) was obtained from Akzo Zout Chemie. Rotterdam, in pressurized stainless-steel cylinders. The product had the following standard specification: vinyl chloride monomer ^ 99-97,, w/w; acetylene <2. monovinylacetylene < 15. IJ-butadiene < 10. methyl chloride <75. ethyl chloride <50. chloroprene <1. 1.1-dichloroethane <1 and 12-dichloroethane <20j/l/litre (gasl: acetaldehyde <5. hydrochloric add <1, iron <05. water <100 and evaporation residue < 10 mg/kg. PVC powder (Carina S 65-02) was supplied by Shell Nederland Chemie. Pcrnis. The particle-size dis tribution (by weight), was max 01% > 300. max 4% > 200. max 90% > 88 and max 95% > 40 pm. The VCM content of a portion of the PVC powder was raised to approximately 4000 ppm by mixing the powder with a calculated amount of liquid VCM in a closed steel barrel. This PVC powder was stored in tightly closed steel containers in a refrigerator at 4`C until a few minutes before being mixed with the diet. Part of the PVC powder obtained from Shell was freed from VCM by being kept in thin layers in a vacuum oven at 60'C for a period of 3-4 days. After this treatment the VCM content of the PVC powder was found to be less than 0-3 ppm. A 10% solution of VCM in soya-bean oil was pre pared by injecting liquid VCM into the oil. The VCM concentration was checked by gas-liquid chroma tography according to the method described pre viously (Feron et al, 1975). The solution was stored at 4"C for a maximum period of 4 wk. Preparation and administration of PVC-containing diets. Each diet contained 10% PVC powder, with varying proportions of VCM-containing and `VCMfree" (i.e. <0-3 ppm VCM) powders. Thus the control diet contained 10% VCM-frce PVC powder and the high-dose diet 10% PVC powder containing 4000 ppm VCM. while the diets for the low- and mid dose groups contained, respectively. 1 and 3% VCMcontaining PVC powder with the balance of 9 and 7, PVC made up with powder without VCM. The various diets were prepared daily by mixing appro priate amounts of PVC powder (with or without VCM) with the Institute's rat stock diet just prior to their being offered to the rats. The diets were avail able to the rats each day for a period of four consecu tive hours (generally between 09.00 and 15 00 hr): thereafter, the rats had no food until the 4-hr feeding period on the next day. The rats had constant access to bottled, unfluoridated tap-water. Calculation of the oral VCM exposure levels. To cal culate the actual oral exposure levels of VCM pro vided by VCM-containing PVC powder in the diets, the following information was needed: (a) the amount of VCM evaporating from the diets during the 4-hr feeding period; (b) the speed with which the animals consumed the food during this period: and (c) the amount of VCM excreted in the faeces. The rate of evaporation of VCM from the diets was determined by measuring the VCM content of the diets at the beginning of the feeding period and after 1. 2 and 4 hr. Samples to be analysed were taken at random from the feeders in the cages, without hom ogenization of the diets in the feeders. Each test diet was sampled in this way on II or 12 different days and analysed for VCM content by gas-liquid chroma tography (Feron et al. 1975). The average VCM con tents of the various diets at the different times are depicted in Fig. 1. The eating-speed was determined by measuring the amount of residua) feed in the feeders after periods of 1. 2 and 4 hr. Since no appreciable variations were encountered in the rate of food consumption by the animals in the various groups, average Tates of food consumption were calculated for males and for females (Fig. 1). The VCM intake (mg/kg body weight/day) by male and female rats of each group was calculated from the graphs representing the rate of evaporation of VCM from the diets (Fig. 1) and the rate of food consump tion over the 4-hr feeding period. Both graphs were 33 09 </) o o CT) to ro --4 IS 10 3 Fig, I. Average VCM levels of the high-dose ( x |, mid-dose IA) and low-dose () diets and average amounts of food consumed by males (Q) and females !) at different times during the J-hr feeding period. Food consumption, j/r a t/d a y in t^^iets. I mg 4-hr the animals and (cl (he :he diets was ntent of the od and after rre taken at ithout homach lest diet ifferent days uidchroma; VCM connt times are teasuring the er periods of "rations were ption by the ates of food es and for Jay) by male .ted from the ion of VCM id consumpgraphs were 15 o \ 10 oA* \ V c <9 Cl aE -*3 uco UoO. 0 * l. mid-dose mms of food 'rcm times Chronic oral toxicity of vinyl chloride in rau 319 assumed to consist of three straight lines, one line for the first hour of the feeding period, one for the second hour and one for the last two hours. The VCM intake during each of these three periods was calculated by multiplying the amount of food eaten during a certain period with the average VCM content of the food in that period. The total VCM intake, the sum of the VCM intakes during these three periods, was subse quently expressed as a percentage of the theoretical intake, calculated from the average total amount of food consumed and the VCM content of the diet at the initiation of the feeding period. This figure was found to be 82-4, 291 and 791% for males, and 81-3, 79-3 and 79-0% for females of the low-, mid- and highdose groups, respectively. The overall average for males and females of the various test groups was cal culated to be 80%. To measure faecal excretion of VCM, freshly pro duced faeces from a representative number of rats of each test group were collected at 09.00 hr (1 hr before the start of the feeding period), at 14.00 hr (at the end of the feeding period) and at 18.00 and 23.00 hr (4 and 9 hr after termination of feeding). Fresh faecal samples were obtained by squeezing the lower part of the rat's abdomen. The droppings were weighed submerged in 10 ml ethyl acetate, and stored at 4:C in a closed vessel prior to analysis of the supernatant liquid by gas chromatography. Within a particular test group there were no appreciable differences in the VCM levels of faeces collected at the various tiroes, although the optimum value was invariably obtained from faeces collected 9 hr after termination of feeding (at 23.00 hr). The average amount of VCM found in the faeces, expressed as a percentage of the VCM intake, was found to be 8. 10 and 17% for the low-, mid- and high-dose groups, respectively. The VCM content of freshly prepared test diets was determined regularly, some twenty determinations for each dosage level being carried out during the study. Daily determinations were not considered necessary because the VCM content of the PVC powder stored at 4'C in a closed barrel was found to remain con stant. and the VCM content of the PVC powder used appeared to correspond very well with the VCM con tent of the various diets. From these VCM determi nations. the average VCM contents of the various diets were calculated as 46, 139 and 424 ppm for the low-, mid- and high-dose groups, respectively. Since both the loss of VCM from the diets before consump tion and the VCM content of the faeces were known, the actual oral exposure levels of VCM could be cal culated. They were found to be 1-7. 5-0 and 14-1 mg/kg body weight/day for the low-, mid- and high-dose groups (Table 1). VCM in oil administered by garage. One group of. rats received VCM in oil by gavage. The approximate dose was 300 mg/kg body weight, administered daily on 5 days/wk for 83 wk, in calculated volumes of a 10% VCM solution in soya-bean oil. The volumes were adapted to the mean body weights once every week if necessary. These rats were offered the Insti tute's stock diet for rats and bottled unfluoridated Up-water ad lib. Animals and housing. Newly weaned albino Wisiar rats (Cpb: WU; Wistar random!, obtained from the SPF colony of the Central Institute for the Breeding of Laboratory Animals TNO, Zeist, were allocated randomly over five groups of males and five groups of females in such a way that the mean body weights were virtually the same. The control group and the two highest dosage groups each consisted of 80 males and 80 females. The two lowest dosage groups each consisted of 60 males and 60 females. During the 5 days before the start of the experiment, the rats that were to be fed the PVC-containing diets received stock diet without PVC for 4-6 hr each day, to adapt them to the daily feeding period of 4 hr. At the start of the study the rats were 5 wk old. The study was ter minated when about 75% of the control rats were dead, a point reached for males in wk 135 and for females in wk 144. The rats receiving PVC powder in their diet were housed under conventional conditions, in groups of five in suspended stainless-steel cages with a wirescreen bottom in a well-ventilated, room maintained at" 24 + 10C. The rats given VCM in oil by gavage were housed in groups of two in suspended, tinned wire-screen cages in a well-ventilated cabinet main tained at 25-28C. Animals in bad condition were housed individually in separate cages until they died or were killed in extremis. Conduct of the experiment. The rats were individu ally weighed, initially, at wk 1, 2, 4. 6. 8, 10 and 12, and at 4-wk intervals thereafter. Food consumption of Table l. Designed and actual dosage of VCM in rats maintained on diets containing PVC po*dcr Designed VCM treatment Dietary level (ppm! Intake (mg-kg body weight/day) Actual initial dietary VCM level Ippm)' Oral intake of VCM Actual oral exposure (mg/kg body weight/day) level of VCM* ----------------------------------- (mg/kg body Theorelicalt Actual weight/day) 00 20 1 60 3 200 10 0 00 0 46 2-3 1-8 1-7 139 7-0 5-6 5-0 424 21-2 n-o 141 Average dietary VCM contents determined immediately after preparation of the diets, tAssuming no loss of VCM by evaporation from the diets (see Fig. 11. Oral intake of VCM diminished by the faecal VCM, which was found to be 8,10 and 17% or the actual oral VCM intake for the low-, mid- and high-dose groups, respectively. The VCM excreted in the faeces was considered to be still enclosed in the PVC granules and thus not to have been in contaci with the body. R&S 006928 R&S 006929 320 V. J. Feuon. C. F. M. Henduksen, a. J. SrsxK. H. P. Til and B. J. Sm 20 rats/sex/group was measured during wk 1-4. 10-11. 24-25. 36-37. 60-61. 72-73 and 84-85. Blood samples were collected from the lip of the tail of ten rats/sex/group in wk 13. 26. 52. 78 and 94. All samples were used for determinations of haemo globin concentration (cyanmethaemogiobin method of van Kampen & Zijlstra. 1961) and packed cell volume (microhaematocritL and for thrombocyte and red and white blood cell counts (Coulter Counter), and differential white cell counts by direct visual count of smears after Pappenheim staining (Gorier & De Graaff. 1955k Fasting blood glucose (Technicon AutoAnalyzer method N-9a) and blood-urea nitrogen (Ceriotti & Spandrio. 1965) were determined in wk 13. 26. 52 and 106. The analyses were conducted upon blood from the tip of the tail of ten rats/sex/group after the animals had been fasted overnight During wk 13. 26. 52 and 106. samples of blood were col lected from the orbital sinus of ten rats/sex/group. The following measurements were made in the serum after centrifugation at 3000rpm for 20min: alkaline phosphatase (Bessey. Lowry & Brock. 1946). glutamic-oxalacetic transaminase and glutamic-pyruvic transaminase (Reitman & FrankeL 19571 total pro tein (biuret reaction), albumin (De Leeuw-lsrael. ArpNeefjes &. Hollander. 19671 and serum-protein pattern (Wieme. 1965). Individual urine samples, collected from ten rats/ sex/group in wk 13. 26. 51 78 and 94. in each case during the last 16 hr of a 24-hr period of deprivation of food and water, were measured for volume (in cali brated tubes), specific gravity (by an Abbe-type refractometer). uric acid (Gorier & De Graaff. 1955) and glutamic-oxalacetic transaminase activity (Reitman & FrankeL 1957). Semi-quantitative measure ments were made of pH. protein, sugar, occult blood and ketones in pooled urine samples, using Labstix from Ames Laboratories. Deposits obtained by centri fugation at 3000 rpm for 3 min were examined micro scopically for erythrocytes, leucocytes, epithelial cells, amorphous substances, phosphate crystals, casts, bac teria. worm eggs and sperm cells. All males Still alive in wk 135 and all females sur viving to wk 144 were killed by decapitation, autopsied and subjected to a careful gross examination. A thorough autopsy was also performed on rats found dead or killed in extremis. Samples of the following organs were fixed in 10,, neutral formalin: heart, kid neys. liver, spleen, brain, testes, ovaries, pituitary, thy roid. adrenals, thymus, pancreas, epididymides, pros tate. coagulating glands, seminal vesicics. preputial glands, mammary glands, lungs, skeletal" muscle, spi nal cord, sciatic nerve, urinary bladder, parotid, sublingual and submaxillary salivary glands, axillary and mesenteric lymph nodes, nose, oesophagus. Stomach, duodenum, jejunum, ileum, caecum, colon, skin, femur with joint and bone marrow, trachea, aorta, cxorbital lachrymal glands. Zvmbal glands, cervix and uterus. The organs to be examined micro scopically were processed through paraffin wax. sec tioned at 5 frm and Stained with haematoxylin and The s-foetoprotein determinations were carried out in the Laboratory of Pathology of the National Institute of Public Health, Bihhoven. under the supervision of Dr R. Kroes. eosin. Microscopic examination of all organs pre served was carried out in 20 males and 20 females of the control group and of each of the two highest dosage groups. For the control groups these 20 males and 20 females comprised all animals killed at the end of the experimental period, supplemented by the ani mals that had lived longest before they had to be killed in a moribund condition. In the two highest dosage groups the rats subjected to detailed histopathology comprised the 20 males and 20 females that had lived longest before being killed m a moribund condition. Histopathological examination of all other rats was restricted to the liver. Zymbal glands, lungs, kidneys, spleen, pituitary, thyroid, adrenals, grossly visible tumours and organs containing gross lesions suspected of being tumours. Batches of ten males and ten females from the con trol group and the two highest dosage groups were killed by decapitation and subjected to a thorough autopsy after 26 and 52 wk (interim kills). At these times, the following were recorded: blood-clotting time (Normotest reagents from Nyogaard and Co_ Oslo. Norway), serum electrolytes Na. K. Ca, Mg (Paschen & Fuchs. 1971) and Cl (coulometric method), serum alkaline phosphatase, serum glutamic-oxalacetic transaminase, serum glutamic-pyruvic transaminase, total serum protein, albumin and serum-protein patterns (determined by the methods mentioned beforeL lactic dehydrogenase (Wroblewski & LaDue, 1955). serum x-foetoprotcin* (BoekesteinTjahjadi &. Kroes. 1976). liver function (bromosulphophthalein-extinction test and barbiturate sleepingtime method), kidney function (phenol-red excretion test: Sharratt & Frazer. 1963L activities of aminopyrine demethvlase (Gram. Wilson & Fouts. 1968) and aniline hydroxylase (Gilbert & Golberg. 1965) in liver preparations, and weights of liver and kidneys. Histopathology of the liver, kidneys and Zymbal glands was studied and the liver was also examined by light microscopy of semi-thin Paragon-stained plastic sections and by electron microscopy. For the latter the liver was fixed by perfusion with 1-5/; glutaraldehyde buffered with 0-067 M-sodium cacodylate con taining l?; sucrose (pH 7-4) at 4C for 17 hr and this was followed by post-fixation in cacodvlate-buffered 1% OsO*. dehydration in graded acetone,-water mix tures. and embedding in Epon 812. Additional control group. One extra control group of 60 male and 60 female rats was housed in a room separate from that used for the other rats fed PVCcontaining diets, thus preventing any possible contact of these control rats with VCM by inhalation. The diet containing I0o PVC containing no VCM was fed to the animals in this group ad lib. instead of for the restricted daily feeding period of 4 hr used for the real control group. Body weights and mortality were recorded, and all the animals were autopsied. The organs and tissues were preserved, but no slides were prepared. RESULTS Behat iour and appearance During yr 2 of the test the general condition of the rats in the 300-mg kg group gradually declined and administration of VCM by gavage became increas ingly difficult. Many rats grew lethargic and filthy gan^^rcfem^Vtf 0 n^Ksi : 20 males at the end y the anitad to be o highest led hist ta males that moribund f all other ids. lungs. 'S. grossly ss lesions 1 the contups were thorough At these d-clotting and Co_ Ca. Mg 'ulometric im glutac-pvruvic min and methods roblewski xkesteinTomaj^slet^B. excreron minopyr9681 and 5) in liver ys. Histoal glands I by light 1 plastic the latter iutaraldelate conand this >buffcrcd ater mix- ol group i a room ed PVCe contact tion. The CM was ad of for d for the ility were lied. The ides were >n of ned and increas ed filthv Chronic oral toxicity of vinyl chloride in rats 321 before they died or were killed in a moribund con dition. Since most of these rats had severe lesions, including tumours, of the liver and lungs, the VCM treatment of this group was discontinued at wk 84. After month 18 the number of unthrifty rats in the 5-0 and 14-1-mg/kg groups gradually increased, more rapidly in the latter (high-dose) group and more rapidly in females than in males. The poor condition started with a humpbacked position and slight ema ciation followed by pale eyes, lethargy, filthiness and often severe emaciation. Liver masses could be detected in many of these rats by abdominal palpa tion. Swollen abdomens were occasionally seen, generally associated with multiple intra-abdominal tissue masses detectable by abdominal palpation. In addition, animals with breathing difficulties were fairly common in these groups; their lungs were in variably found to contain multiple nodules, which appeared microscopically to be primary angiosar comas or metastases from either hepatic angiosar comas or hepatocellular carcinomas. External tissue masses occurred fairly often, mainly after month 18; most appeared to be mammary-gland tumours, but there were also other types of tumours originating from the skin, subcutis or dermal adnexa. Randomly distributed majoT abnormalities, not attributable to VCM. included staring coats, a bloody discharge from the nose, wet stools, focal alopecia, focal dermatitis, blood around the muzzle and eyes, paresis of the hind legs, loss of one or both eyes, and a white opaque cornea. Body weights and food consumption Body weights (Fig. 2) and food intakes of the rats fed the VCM-treatcd diets were very similar to those of the controls. The rats of the additional control group were much heavier than those of the other groups receiving diets containing PVC powder. This difference was undoubtedly due to the fact that the extra controls had constant access to their food, whereas diet was available to the Other rats only dur ing a period of four consecutive hours each day. The rats of the 300-mg/kg group had constant access to stock diet and had much higher body weights than the animals fed PVC-containing diets for a limited period each day. but their body weights were lower than those or the additional control group. females was only slightly higher than that of female controls. Haematology, biochemistry, urine analysis and organ function No relevant changes were found in any of the par ameters studied, except for a shorter blood-clotting time and an increased content of a-foetoprotein in the blood serum of animals in the group fed she highest VCM dose and in the intubated group (Table 3k Cross pathology Liver-to-bodv weight ratios were higher in the 14-l and 300-mg'kg groups than in the control group after both 26 and 52 wk (Table 3). Many rats exposed to VCM had severe liver lesions. Pronounced swelling, discoloration and altered consistency of one or more lobes, often with varying numbers of cysts, were com mon findings, as were nodules and nodule-like pro cesses. varying widely in size (up to 4 cm in diameterL appearance and consistency. Many of the nodules were solid and pale; others were cystic and haemorr hagic. The larger firm and pale nodules with central necrosis appeared, upon microscopy, to be carci nomas. These were found mainly in completely dis torted livers. Angiosarcomas were most often seen as multiple soft dark cystic nodules, containing blood and granular necrotic material. Nodules, which were later classified as 'neoplastic nodules', were relatively smalL firm and compact: they were either pale or had the same colour as the adjacent normal liver tissue, and never contained necrotic material. These liver changes were most pronounced and occurred earliest 600r Mortality Mortality was low in males and females of the 4- hr-fcd control group during the first 2 yr of the test, amounting to 10% after 2 yr (Table 2). At this time the mortality among rats of the additional control group was some three times higher (about 30%). About 40% of the males and females of the 300-mg/kg group had already died by month 18 of treatment. Thereafter, mortality rapidly increased, and by just after month 24 all the animals in this group had died, mainly from pulmonary or hepatic insufficiency resulting from neoplastic or non-neo plastic lesions in these organs. A striking and doserelated increase in death-rate was also found in the Fig. 2. Average body weights of the extra controls fed the IQ",,-PVC diet uJ lib. (--I and of the rats given 300 mg 5- 0- and 14T -mg/kg gToups. with females dying earlier VCM kg body weight in oil by gavage i---- k The weight than males. In the low-dose group (1-7 mg VCM/kg curves of the rats receiving 0. 1-7. 5-0 or 14-1 mg VCM/kg body weight!, the mortality of males was comparable body weight-`day from the 10",,-PVC diets fed for a hr each to that of the male controls, and the death rate of day all he within the shaded area. R&S 006930 322 V. J. Fexon, C F. M, Henwuksen. A. I SretK. H. P. Til and B. J. Sm Table Z Cumulative mortality of rots exposed orally to VCM for up to 2-7 yr Treatment group Number of dcathst by end of wk: (mg VCM/kg body ------------------------------------------------------------------------------------- weight/day] 12 36 52 80 92 105 120 128 w: i43+* 0 1-7 5-0 14-1 300S 0- 0 1-7 50 141 3005 05 000 00 1 000 012 066 000 00 00 01 00 0- 3 00 0 1 2 l 7 1 0 1 2 8** 23 1 I 2 7* 7* 24 4 Males 2 3 7 22... 47 3 6 6 12 40*" 53 19 Females 5 4 16" 43... 47 10 6 13 31*** 60*** 58 17 ` 18 13 30* 56* ** 60 28 . 22 26 55*** 60"* 60 27 40 37 49 60*** 60 46 46 40 60**. 60** 60 46 27 32 60*** 60*" 60 42 32 34 60*** 60*** 60 43 41 55** 60*** 60*** 60 52 flnitial number of rats: 60/sea/group. JSurviving males were killed in wk 135 and surviving females in wk 144. $The figures for this group were not evaluated statistically, because no corresponding control group was included in the study. jAdditional control group housed in a separate room and having constant access to the diet containing 10% PVC without VCM. Values marked with asterisks differ significantly from those of the controls according to the chi-square test: *P < 005: "P < 001 ;**/>< 0001. and most frequently in the 300- and 14-1-mg/kg groups. Angiosarcomatous nodules were not seen at all in the low-dose group. Pulmonary alterations that could be ascribed to VCM treatment consisted of small haemorrhagic or greyish nodules, often located at the edge of a pul monary lobe. These changes were not observed in controls or low-dose animals. Rats with a swollen abdomen due to ascites and to numerous pale firm nodules (diameter 2-15 mm) in the peritoneum were encountered slightly more fre^ quently in each of the three lowest dosage groups (maximum incidence 15%) than in the control or 300-mg/kg groups (maximum incidence 5%). Micro scopically the nodules appeared to be mesotheliomas. Many other types of gross changes, both neoplastic and non-neoplaslic. were seen either frequently or just in a few animals, but there was no indication that any of these changes were related to VCM. Light microscopy of the liver The type and incidence of -treatment-related histopathological changes found in the liver are given in Table 4. The incidence of foci of cellular alteration Table 3. Mean prothrombin times, afoetoprotrin contents of the blood serum, and licer weights of rats exposed orally to VCM for 26 or 52 wk Treatment group (mg VCM/kg body weight/day) Prothrombin time (sec) at wk 26 52 a-Foetoprotein Oig/ml) at wk 26 52 Liver weight If 100 g body weight) at wk 26 <2 0 14-1 300t 0 141 300+ 41-3 38-9* 35-2 37-2 34-t*** 31-7 41-6 38-8* 37-3 32-8 307 305 Males 75 74 84 Females 99 91 93 30 51* 55 53 109** 56 2-54 2-91* 3-31 2-60 308" 3-86 2-63 3-00 3-38 2-57 3-31* 3-44 +The figures for this group were not evaluated statistically, because no corresponding control group was included in the study. Values marked with asterisks differ significantly from those of the controls according to the test of Wilcoxon (prothrombin time and a-foetoprotein) or Student's f test (liver weight): mP < 005: **P < 00): **/'< 0001. "n fio cn 0 0 O) ID 03 "* S'S-* SdJ S- 2 S3? ?= 8 2 B SJLS 3 w o~ .3? o uO A x[ P o3- Ak, p i -*< Chronic oral toxicity of vinyl chloride in rats Tiiblc 4. Trpc nm( fiirfift'tirr <>f treotim'iil-reloteil hlslopiilhitUnjU nl t'htiiiffei in the liter of rot* exposed orntly to I'Chf Incidence of change Type of cli an get Males Treatment group (mg VCM/kg/day)... 0 1-7 50 14 1 300J 0 1-7 Females 50 141 Clcar-cell foci Clear-cell foci Oasophilic foci Eosinophilic foci Neoplastic nodule 1 lepaloccllular carcinoma Cystic proliferation of bile duels Clear-cell foci llasophitrc foci Eosinophilic foci Neoplastic nodule Hepatocellular carcinoma Angiosarcoma Proliferation of atypical sinusoidal cells only Extensive necrosis Cysts Liver-cel! polymorphism Ccntrilobular degeneration Focal hnematopoicsis Animats killed after 16 wk No. of rats exit mined.. . 10 0 s -- -- 10 1 Animals kilted after Sink No. of rats examined.. . 9 l 0 0 0 0 0 - -- .... -- -- -- -- -- to 8** 0 2 1 1 0 Animals found dead or killed In extremis or terminally No. of rats examined . . . 55 0 8 3 0 0 0 5K IX 23*** 1 10 56 16*** 21* 2?*.. 7** 2 6* 59 21*** 22** 33*** 23*** 8** 27*** 9 ' 10 10 99 00 0 0 00 00 00 00 55 57 9 4 12 0 11 8 32 10 27 0 ---- -- -- .-- -- ---- ---- -- ---- 58 24*** 33*** 35*** 26** 4 0 59 22*** 17 20* 39*** 19*** 2 10 5** 10 8** 4 5** 2 1 4* 57 36*** 28*** 29*** 44*** 29*** 9** 2 4 2 4 .0 0 0 4 3 16* 0 1 4 8 4 28V* 0 0 7 23*** 16*** 42*** t 10** 64 6. 3 4 21 5 ' 6 '1 ' 19V* t 27*** 3 9 30***.i, 41*** , 49*** 36 34 t5*'AW 38 41 I 1 12 1, 3 1 8 1 ' 3 ,' 1 6 (Specific hepatocellular lesions were classified according to Sqtiirc & Levitt (197J|. j (The figures of lists, group were not evaluated xlnlistunUy. hccimsc no corresponding control group was included in the study. | ^Nol esumined. The initial number of animals was 60/sex/group. A ntintber of nils could not be examined because of cannibalism or advanced autolysis. Values marled with asterisks differ significantly from those of the controls according to the chi-st|i)are lest: */* < (105; '*/' < 001; ***/' < 0001. 300( 10 2 8 0 1 0 0 0 0 54 10 (9 6 2 0 ' 29 7 24 3 41 18 12 ZC6900 S'Sy R&S 006933 334 V. J. Feuon. C F. M. HenomkshW A. J. Smxk. H. P. Til and B. J. Sw was much higher in each of the three test groups receiving VCM-containing PVC powder (the 1*7-, 5-0and 14-1-mg/kg groups) than in the control group, and in addition, was nearly always higher than in the group receiving VCM in oil by gavage (300-mg/kg group). Similar differences also existed for neoplastic nodules and hepatocellular carcinomas. In the test groups receiving VCM-containing diets, the.incidence of both neoplastic nodules and hepatocellular carci nomas was positively related to the VCM dose, and was much higher in females than in males. Angiosarcomas of the liver (Figs 3 and 4) were found in males and females of the three highest dosage groups, but did not occur at all in controls and low-dose animals (Table 4). In both the 50- and 14-l-mg/kg groups the incidence of angiosarcomas was three times higher in males than in females. This difference between the sexes did not exist in the 300-mg/kg group, in which an angiosarcoma inci dence of about 50% was found in both males and females. In the 300-mg/kg group, the average latent period for the detection of liver tumours (almost exclusively angiosarcomas) at death was found to be 84 wk for males and 83 wk for females. In the 141-mg/kg group these average latent periods appeared to be 104 and 88 wk for males and females, respectively, clearly indi cating an earlier appearance of liver tumours in the former group especially in males. Several rats bearing angiosarcomas or liver-cell tumours also showed focal proliferation of atypical sinusoidal cells, often accompanied by distension of sinusoids. In addition, similar changes were observed in several rats not bearing a tumour in the liver. Large areas of necrosis were found in the livers of a fairly large number of rats of the three highest dosage groups. Cysts, very probably lined by proliferated bile-duct epithelium, were seen in a relatively large number of males of the 14-l-mg/kg group, and in females of the I-7-, 5-0- and 14-l-mg/kg groups. The size and multiplicity of the cysts varied widely in indi vidual animals. Cellular and nuclear polymorphism of hcpatocytes was much more common in the test groups than in controls. Centrilobular liver degener ation was a frequent finding in females, but not in males, of the 300-mg/kg group. Focal haematopoiesis was encountered more often in males and females in the 14-1- and 300-mg/kg groups than in those of the other groups. Electron microscopy of hepatic parenchyma In semi-thin sections, foci of hepatocytes and also isolated hepatocytes with a finely `vacuolized' cyto plasm were found in rats of the 14-1- and 300-mg'kg groups after both 26 and 52 wk. Ultrastructurally the `vacuoles' appeared to represent swollen mitochon dria with a pale matrix and short cristae (Fig. 5). The abnormal mitochondria closely resembled those found in rats following inhalation exposure to 5000 ppm VCM for periods of 4-52 wk and already described by Feron. Spit, lmmel & Kroes (1979b), Foci of hepatocytes. each containing numerous highly swollen, irregularly-shaped mitochondria with out cristae and a matrix widely varying in density, were encountered after 26 wk. An increased amount of tubular smooth endoplasmic reticulum (SER) was invariably present among the swollen mitochondria. Occasionally scattered individual hepatocytes were found to contain a few extremely swollen mitochon dria but were otherwise normal, as were their other mitochondria. After 52 wk, large areas of hepatocytes containing numerous swollen mitochondria were found, mainly in rats of the 141-mg/kg group. These hepatocytes had large nuclei with pronounced nuc leoli. Tubular SER and whorls of SER were occasion ally found together with mitochondria containing lucent areas in which a membranous or flocky material was often encountered. Single membranes of rough endoplasmic reticulum surrounding these mito chondria had partially lost their ribosomes. . Light microscopy of organs other than the liver A wide variety of alterations was found, nearly all apparently related to the normal ageing process. An exception may have been the very marked haemato poietic activity found in the spleen of six out of 40 males and ten out of 40 females from the two highest dosage groups, while in controls only slight to moder ate splenic haematopoiesis was observed. The site and type of tumours observed in organs other than the liver and their incidence in the different groups are presented in Table 5. Angiosarcomas were frequently found in the lungs at the two highest dose levels, and also occurred in a few rats of the 5-0-mg/kg group. They were most often seen as multiple small foci of tumour cells with an angiomatous growth pat tern. Their appearance was highly suggestive of metastases. On the other hand, in several cases, the histolo gical appearance of the neoplasms did not permit the exclusion of their being diagnosed as primary pul monary angiosarcomas. In addition, in three rats with a pulmonary angiosarcoma, no angiosarcoma was encountered outside the lungs. In four rats from the 300- and 141-mg/kg groups an angiosarcoma was .found in the abdominal cavity outside the liver, while the liver showed no signs of angiosarcoma formation. Therefore, these tumours were considered to be primary extrahepatic angiosarcomas. Pulmonary metastases of hepatocellular carci nomas were not uncommon. A total of five tumours of the Zymbal glands (ceruminous glandsl were found--two squamous-cell carcinomas in the 5-0-mg/kg group, and two squamous-cell carcinomas and one adenoma in the 300-mg'kg group. Abdomi nal mesotheliomas were observed in each of the groups, including the control group. In several groups their incidence was higher than in controls, but a positive dose-response relationship with respect to incidence was absent. On the other hand, the latent period for detection of abdominal mesothelioma at death was found to decrease with increasing dose levels for both males and females. The histological appearance of the peritoneal mesotheliomas varied, but two main types could be distinguished, a fibrous type in which sarcomatous areas predominated and an epithelial type consisting mainly of tubulo-papillarv formations. In several tumours both the sarcoma tous areas and the tubulo-papillary structures occurred to the same extent. Mitotic figures were never abundant Fibroadenomas of the mammary glands were much less frequent in females of the 5-0-, 14-1- and I lochondria. cvte^|CTC thci^other lepatocytes idria were oup. These meed nuct occasionc niaming or flocky nbranes of hese mito- rrr nearly all rocess. An haemalo^ out of 40 vo highest to moder- in organs e different amas were rhest dose 5-0-mg.kg iple small owth pat: of metaleh^fcx'n^^e nary pul rats with oma was from the oma was ver. while ormation. si to be ar carcitumours dsi were in the rci nomas Abdomih of the al groups is. but a rspect to he latent .`lioma at ing dose itoloeical s varied, a fibrous ated and lo-papillsarcomatruo^hi V :re much H- and ' Fig. 4. Higher magnification of the hepatic angiosarcoma depicted in Fig. 3. showing large irregularlyshaped cells with big pale nuclei often containing multiple conspicuous nucleoli. Hacmatoxylin and eosin x 400. 325 R&S 006934 1 - -> Fig. 5. Hepatocytes containing normal and highly swollen mitochondria with a pale matrix and short cristae from a male rat exposed orally to 300 mg VCM kg body weight on 5 days/wk for 26 wk. Uranvl acetate, lead citrate x 6400. ' 30 0) o o o> CD CO cn ^6 4 VlH* \*t tW* * /. ( Chronic oral toxicity of vinyl chloride in rats Table 5. 5ifc, type otnf inriJein-c of tumour* In oryon* uthrr than thr lirert In rufs exposei/ orti/fy to VC Af/or ruer 2*5 yr Incidence of tumours Site and type of til moil r Lungs Angiosarcoma Adenoma Zymbal glands Ss|uainoiis-ccll carcinoma Adenoma Abdomen Mesothelioma Angiosarcoma l"ibrosarconia Osteosarcoma Sarcoma Reticulum-cell sarcoma 'Schwann*ccll tumour' Unclassified Spleen llacmangioendolheliosarcoma Lymphosarcoma Nose Squamous-cell carcinoma drain Granular-cell myoblastoma Oligodendroglioma Plexus papilloma Glial-cell tumour Ependymoma Mesodermal tumour Pancreas Adenocarcinoma Thorax Mcsothcliomu Thyroid Parafollicular-cell adenoma Parafollicular-celt carcinoma lollicidar-ccll adenoma Treatment group (mg VCM/kg/day).,. -------0 Effective no. of rats... 55 No. of rats with primary tumours... 38 0 0 0 3 0 0 0 0 0 0 0 0 0 0 'I 1 0 0 0 0 0 I 4 I 0 Males 1-7 50 58 56 50 49 0 4* 00 02 00 17 00 00 00 03 10 00 00 I0 00 00 10 00 00 02 00 01 00 00 12* 10 01 00 14-1 300J 59 55 52 44 19*** 0 0 0 H 0 3 0 | 0 0 0 0 0 1 0 0 0 0 1 0 0 0. 3 0 1 19 1 I 0 1 t 0 '1 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 3 0 0 Fcniatcs 0 1*7 5-0 57 58 59 54 56 55 0 0 0 0 1 0 1 0 0 0 0 2 0 0 0 0 0 0 0 0 0 1 0 7* 0 1 0 0 0 0 6* 0 2 0 0 0 1 0 0 0 0 1 0 1 0 0 0 2 0 in 0 0 r 0 0 0 3 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 3 0 0 14 1 57 57 5* 0 0 0 3 2 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 t 300t 54 47 23 0 1 1 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 9C6900 S$U Site and type or in mour Adrenals Cortical mlcnonui I'hncnchromncyloma Pi I uitary Adenoma Carcinoma Blood leukaemia Heart `Endocardial disease's Haomangiocndotheliosarcoma Kidneys Nephroblastoma Clear-ccll tumour l.ipomalotrs tumour Unelassilied epithelial tumour Thymus Fibrosa reoma Reticulum-cell sarcoma . Mesenteric lymph nodes Rclieuliun-ccll sarcoma Skin St|iiamous-eell carcinoma Subcutis Fibroma Filu nsarcoma Mesenchymal tumour Skeletal muscle R hnbdomyosarcomn Skull Osteoma Mesenchymal tumour Ear region Adenocarcinoma of unknown origin Uiinaty hladilcr Unelassilied epithelial Inmntir t'rcpulial glands S^uamous-cell carcinumu Treatment group (mg VC'M/kg/dayk.. Tnhle S--conllmieif Incidence ol tumours Males --------------- --______________________ __________ 0 1-7 50 I4t 300t _______ 0 Females 1-7 5-0 14-1 300( ZS6900 IS 25 17 to 11 2k s 4 9 ' 26 30 20 17 14 62 1 202 12 25" 6 101 2** 0 14 16 to 0 0 3 0 2 5* 3 00 '01 1312101 I0 2 2 1 1 000 I I 00 0000000 1 00 1 000000 00 000 0 00 1 0 00 I 0000000 0 0 0 0 0. 1 0 0 0 0 0 1 0 0000000 0 0. 0 0 1 0 1 I 0 0 0000 1 00000 2 3 3 1 0006 I 0 2 1 1 1 1 3 3 , 00 0 I 1 00 | | 000 00 0 00 1 0000 00 0 1 I 1 00 00 t 0 0 000 0000 00 0 000 1 000 00 1 0000000 00 1 00 0 0000 00 1 000 0000 V. ]. CFcron, F. M. H enchiiksen. A. J. Sfeek. H. P. T il and B. J. Spit Unclassified epithelial tumour Preputial glands .S^^oiiS'Ccll carcinoma 0 0 . t, , 0 1 000 1 000 .* * ,.,M <VI,. f ... - *a * 0 00O 0000 -t **-r>.* T .3-,. a, . Mammary glands Adenoma Fibroadenoma Adenocarcinoma Anaplastic carcinoma Testes Interstitial-cell tumour Uterus Adenocarcinoma Malignant fibroadenomatous tumour Leiomyoma Cervix Mesenchymal lype of tumour Adenoca rcinoma Ovaries Theca-cell tumour 00 000000 20 0 0 0 0 0 21 25 12** 4*** 7 ,o 1 0 2 0 3 2 4 7 7 00 00000 1 00 300 1 ' 63 o 1 0 0 1l 00 ' 0 0 0 0 t A small number of primary liver lumours unrelated to treatment were found in several groups. These tumours were one KupfTer-cell sarcoma, three reticulum-cell sarcomas, two fibrosarcomas, one hacmangiocndolheliorna and one mesenchymal tumour. tThe figures for this group were not evaluated statistically, because no corresponding control group was included in the study. Sin several cases the neoplastic character of the lesion was doubtful. Values marked with asterisks differ significantly from those of the controls according to the chi-square test: *P < 005; **P <0-01; ***P <0001. Chronic oral toxicity of vinyl chloride in rats \r 8E6900 S'Sd R&S 006939 330 V. J. Fekon. C F. M. Hendxiksen, A. J. Spoil H. P. Til and B. J. Sen 300-mg'Vg groups than in controls or low-dose ani mals (Table 5). The low incidence of this common tvpe of `spontaneous' tumour, which is associated with old age, at the three highest dose levels is un doubtedly connected with the much shorter survival time of the animals in these groups compared with that of the controls or low-dose animals. Despite the relatively short survival time of females in the 14-l and 300-mg/kg groups, the incidence of adenocarci nomas of the mammary glands in each of these groups was twice as high as in the control group. This may indicate that VCM enhances the development of mammary-gland carcinomas. The incidence of several other common types of tumour known to be associated with old age was found to decrease with increasing dose levels, as was to be expected from tbe dose-related reduction in sur vival time. Examples of these were cortical adenomas and phaeochromocytomas of the adrenals, pituitarygland adenomas, parafollicular-cell adenomas of the thyroid (the incidence of this tumour was exception ally low in males of the control group), and adenocar cinomas of the uterus. DISCUSSION The limited daily feeding period of 4 hr in the day time appeared to be associated with lower body weights but also with a higher survival rate. Appar ently. reduced body-weight gain caused by a restricted feeding period is a favourable rather than an un favourable effect. The much higher survival rate also points to the rats being healthy and having an ad equate nutritional status unaffected by the relatively short feeding period in the day-time. Only slight and often inconsistent differences were found between controls and test animals in several of the haemaiological and biochemical parameters stud ied. These slight deviations were regarded as loxicologically insignificant, with two possible exceptions, the shortening of the blood-clotting time and the in crease in a-foetoprotein levels in the blood serum of VCM-exposed rats. Failure of the blood to clot was noted in guinea-pigs that died during exposure to an atmosphere containing 40?; VCM (Mastromatteo. Fischer. Christie & Danziger. 1960) and in two workers in a PVC factory who died from acute VCM poisoning (Danziger. 1960). However, rats exposed to 5 or 3?; VCM for 19 or 92 days, respectively, had normal blood-clotting times (Lester. Greenberg & Adams.. 1963). In contrast, both in the present study and in a previous long-term inhalation study with VCM in rats (Feron, Kruysse &. Til 1979a), pro thrombin times were shorter in VCM-exposed rats than in controls. Because of the conflicting obser vations. it is difficult to assess the toxicological impor tance of the slight hypercoagulability of the blood seen in the rats exposed to VCM in our studies. Muller. Buchter. Gross & Bolt (1976) recently reported a disturbed thrombocytic function in nine of 17 patients suffering from `vinyl chloride disease'. Thrombocytopenia, which has been reported to be one of the signs of VCM intoxication in man (Jiihe. Lange. Stein & Veltman. 1973; Lange. Jiihe, Stein & Veltman. 1974; Muller et al. 1976). was observed neither in the present oral experiment nor in the pre vious 1-yr inhalation study (Feron er at. 1979aL In future studies, it would be desirable to pav speaaJ attention to the possible effects of VCM on thrombo cytes. Slight, though statistically significant, increases m the a-foetoprotein content of the blood serum were found in males and females of the 14-1-mg kg group at wk 52. At wk 26 a-foetoprotein levels were normal in animals of this group. Very similar results have been obtained in rats exposed to 5000 ppm VCM in air for periods varying from 4 to 52 wk t Feron et at. 1979a). These findings may indicate the presence or foeioglobulin-producing neoplastic or preneoplastic cells in the liver of rats exposed to VCM - for a pro longed period. However, since the increase in a-foetoprotein level was only slight, asperific liver damage, such as necrosis or degeneration of hepaiocytes. may have been responsible for this slight effect Although hepatocellular tumours have been reported in rats treated with VCM (MallonL 1977; Williamson, 1976) and some evidence exists as to the occurrence of hepatocellular carcinomas in persons exposed to VCM (Berk. Martin. Young. Creech, Selikoff. Falk. Watanabe. Popper &. Thomas. 1976; Popper, Thomas. Telles. Falk & Selikoff. 1978L the high incidence of neoplastic liver-cell nodules and hepatocellular carcinomas in test animals (especially females) receiving VCM orally in PVC powder was an unexpected finding in the present study. E'en at the lowest level (1-7 mg VCM/kg body weight 'dayl. VCM-related liver-cell tumours and an increased inci dence of foci of cellular alteration were noticeable in both males and females. It was remarkable that only a few hepatocellular neoplasms occurred in animals receiving VCM in oil by gavage at the very' high dose of 300-mg/kg body weight whereas nearly 50; of these rats had a hepatic angiosarcoma. The nature of the tumour response or the liver in the various test groups suggests a shift from an almost exclusive de velopment of angiosarcomas at the very high dose level, ria both angiosarcoma and hepatocellular tumours at the intermediate dose levels, to the ex clusive occurrence of liver-cell tumours at the lowest level. On the other hand, not only the dose but also the way of administering VCM may have been of significance for the difference in tumour response between the 300-mg/kg group and the lower dosage groups, because gastric intubation of VCM in oi) (daily on 5 days/wkl implies that the body tthe liver) has to deal with a large amount of VCM in a short time, whereas in the case of oral intake of VCMcontaining PVC powder the body (the liverl is con tinuously exposed (most probably for 24 hr day on 7days/wk) to `fresh* VCM released gradually and without interruption from the PVC powder during its transport through the gastro-intcstinal tract. One may also speculate that the unexpectedly high incidence of liver-cell tumours in the groups receiving VCM in PVC powder could be due to an unusual sensitivity of the liver cells to VCM. developing as a consequence of an altered metabolic state caused by the reduction of the daily period of food intake to 4 hr. This restric ted period of food intake may have caused a daily, transitory, depression of hepatic glutathione, which has been demonstrated to be of significance for the detoxification of VCM or its reacme metabolic 979al In reases in um were tg group i normal ills have VCM in on el ai sence of eoplastic r a proa-foetodamaee. ics. may c been tL 1977: is to the persons Creech, i 1976; >78L the lies and specially t was an a at ihe thl.dayl. >ed incieabj^fc t ori^ animals gh dose 50"; of ature of jus test sive deth dose icellular the ex: lowest >ut also been of esponse dosage in oil ic liven a shon VCIMis con* day on Jy and ring its nc may encc of CM in ivjiv of quencc Juctliioojn iredsat;ml which for the ibolites Chronic oral toxicity of vinyl chloride in rats 331 (VVatanabe. Hefner & Gehring. 1976: Watanabc, McGowan & Gehring. 1976; Watanabc, McGowan. Madrid &. Gehring. 1976). The lesions of the hepatic parenchyma that could be attributed to VCM included not only foci of cellu lar alterations', neoplastic nodules and carcinomas, but also necrosis, csntrjlobular degeneration and mitochondrial damage. In addition, the slight in creases in the a-foetoprotein content of the blood and in relative liver weight may also have been indications of VCM injury to the hepatic parenchymal cells. Since it is known that tissue injury followed by repair (hy perplasia) may stimulate tumour formation, the possi bility exists that the 'promoting' properties of this chronic process of continuous damage and repair play a major role in the development of hepatocellular tumours following VCM exposure. The association between VCM and angiosarcoma in man was recognized some years ago (Creech. & Johnson. 1974: Lloyd, 1975) and several investigators have produced angiosarcomas in experimental ani mals by exposure to VCM either by inhalation or oral administration (Feron et ai 1979b; Holmberg. Kronevi & Wincll, 1976; Keplinger. Goode. Cordon & Calandra, 1975: Lee et al. 1978; MaltonL 1975 & \977y. The results of the present experiment show that oral exposure of rats to VCM at levels of 5-0 mg/kg body weight/day or more caused hepatic angiosar comas. pulmonary angiosarcomas (probably both primary tumours and metastases) and at the higher levels, also a few primary extrahepatic abdominal angiosarcomas. No angiosarcomas were found at the lowest level of 1-7 mg VCM/kg body weight/day. Uhrastructural studies of the liver showed a scat tering of foci of parenchymal cells affected by VCM. Swelling of mitochondria was the earliest change ob served and mitochondrial alterations were also characteristic of the VCM-induced damage seen at later stages. Swelling of mitochondria has been found after exposure to other carcinogens (Bannasch. 1975) and also under various pathological conditions, such as ethanol damage (Dobbins. Rollins. Brooks & Fal lon. 1972) and riboflavin deficiency (Tandler. Erlandson & Wynder. 1968). In addition to mitochondrial swelling, prolonged treatment with VCM for 26 or 52 wk appeared to cause a decrease in and loss of ribosomes from the RER. and most importantly an increase in SER. These findings suggest an enhancing effect of VCM on the so-called mixed-function oxi dase system (m.f.o. system) which has been shown to be involved in the detoxification of VCM (Boll. Kappus, Bruchtcr & Boh. 1976: Reynolds. Moslen. Szabo. Jaeger & Murphy. 1975). However, the deter minations of aminopyrine-dcmethylase and anilinehydroxylase activities in the liver did not produce evi dence of m.f.o. induction by VCM. Full development of VCM-induced angiosarcoma has been found to be preceded by nodular hypertro phy and hyperplasia of liver cells, which may proceed to hepatocellular carcinoma (Feron et al, 1979b: Popper. Selikoff. Mahoni. Squire & Thomas, 1977). The liver changes found in the three lowest dosage groups are well in line with this observation, but in this connection it is difficult to understand the high incidence of hepatic angiosarcomas in the 300-tng.Lg group, because focal hypertrophy and hyperplasia of the parenchyma occurred in only a very limited number of rats of this group. Several investigators (Lee et ai 1978: MaltonL 1975) have found extrahepatic angiosarcomas in rats exposed to VCM. In the present oral study, extrahepatic angiosarcomas were also seen, although it should be stressed that most of the pulmonary angio sarcomas were considered to be metastases from hepatic angiosarcomas. Only a few test animals bore a tumour of the Zymbal glands, but since spontaneous Zymbal-gland tumours are rare and tumours of this organ are known to be induced in rats by VCM, it seems justifi able to ascribe to the VCM treatment the few Zyrobal-gland neoplasms found. VCM-induced carcinomas of the mammary glands have been observed in mice (Lee et ai 1978; MaltonL 1975) and presumably also in rats (MaltonL 1975). The incidence of mammary-gland carcinomas in several of the test groups was only slightly higher than that in the controls and was not considered to be conclusive evidence that VCM can induce this type of neoplasm in rats. At most it may be regarded as an indication that VCM may enhance carcinoma forma tion in this organ. To our knowledge. VCM-induced mesotheliomas have-not been reported, Ln the present study, the number of rats with abdominal mesothelioma was higher in each of the test groups receiving PVCpowder containing VCM than in controls fed PVC powder free of VCM. This may indicate that the ingestion of VCM has a potentiating effect on the development of mesothelioma in rats. The high incidence of liver-cell tumours in females of the low-dose group (28/58) clearly demonstrates the absence of both a `no-observed-adverse-effect level" and a `minimum-effect level' in our experimental sy$,lem. These levels must be known before any attempt can be made to extrapolate the animal data to man and to assess the carcinogenic risk of VCM. There fore. a similar lifespan oral carcinogenicity study with VCM in rats has been initiated, using three different dose levels (0-017. 017 and 1-7 mg VCM.-kg body weight/day) and two control groups: this study had been running for 14 months in October 19S0. Conclusions This study shows that the incorporation of VCMcontaining PVC powder into the diets of rats is an effective method for studying the long-term effects of oral administration of VCM. VCM is a carcinogen when administered to rats by the oral route and the tumour response of the rat liver to the oral intake of VCM seems to shift from an almost exclusive devel opment of angiosarcomas at very high levels to the exclusive induction of hepatocellular tumours at lowlevels of exposure. The ingestion of VCM in PVC pow-der may enhance in rats the development or ab dominal mesotheliomas and of adenocarcinomas of the mammary glands. The `no-observed-adversc-effect level' was lower than 1-7 mg VCM/kg body weight.day under the rigorous conditions of continuous oral VCM exposure resulting from the gradual release of VCM from PVC powder present in the gastro intestinal tract. R&S 006940 R&S 006941 332 V. J. Feros. C. F. M. Hendwksen. a. J. Speeic H. P. To. and B. J. Sm Acknowledgements--The authors thank Drs A. P, de Groot and D. van Battum for advice and continuous interest dur ing the planning and execution of the study and for help with the preparation of the manuscript. They are indebted to Dr R. Kroes for his critical review of the manuscript. REFERENCES Ministry of Agriculture. Fisheries and Food (1978). The Surveillance of Food Contamination in the United Kingdom: Survey of Vinyl Chloride Content of Poly vinyl chloride' for Food Contact and of Foods. The Second Report of the Steering Group on Food Surveil lance--The Working Party on Vinyl Chloride. Food Surveillance Paper No. 2. p. 6. HMSO. London. Bannasch. P. (1975). Die Cyloplasmaorganellen der Hepatocyten wahrend der Carcinogenesc. 1. Mitochondrion. In Handhuch der ullgtmeinen Pathologic. Stchster Band, siehenter TeiL Geschwulsle-Tumors 111. Edited by E. Grundmann. p. 211. Springer.Verlag: Berlin. Basalaev. A. V_ Vazin. A. N. & Kotcheikov. A. G. (1972). On the pathogenesis of changes developing due to a long-term exposure to the effect of vinyl chloride. Gig. Trudu prof. Zohol. 16, 24. Berk. P. D_ Martin. J. F_ Young. R. S, Creech. J_ Selikoff. I, J.. Falk, H.. Waianabe, P. Popper. H. & Thomas. L. (1976). Vinyl chloride-associated liver disease. Ann. intern. Med. 84, 717. Bessey. O. A. Lowry, 0. H. & Brock. M. J. (1946). A method for the rapid determination of alkaline phospha tase with five cubic millimetres of serum. J. biol. Chem. 164. 321. Boekesiein-Tjahjadi, H. M. & Kroes. R. (1976). s.-Fetoproteine (AFP): Radioimmunologische Bepaling en de Vroege Verschijning van APF na Aflatoxine B| Toediening. Bcrichten uit het R1V-197J. p. 136. Staatsuitgeverij. Den Haag. Bolt. H. M,, Kappus, A. Bruchter. A. & Bolt. W. (19761 Disposition of (1.2-1'`C) vinyl chloride in the rat. Archs Toxicol. 35, 153. Ceriotti, G. & Spandrio. L. (1965). Catalytic acceleration of the urea-diacetyimonoxime-phcnazone reaction and its application 10 automated analysis. Ciinica chim. Acta 11, 519. Creech. J. L~ Jr & Johnson. M. N. (1974). Angiosarcoma of the liver in the manufacture of polyvinyl chloride. J. occup. Med. 16, 150. Daniels. A. G. & *Proctor. D. E. (1975). VCM extraction from PVC bottles. Modern Packaging April, p. 45. Danziger. H, (I960|, Accidental poisoning by vinyl chlor ide. Can. Med. Ass. J. 82. 828. De Leeuw-lsrael. F. R. Arp-Neefjes. J. M. & Hollander. C. F. (19671. Quantitative determination of albumin in mic roliter amounts of rat serum. Expl Geront. 2, 255. Dobbins.. W. O- Rollins. E. L- Brooks. S. G. & Fallon. H. J. (1972). A quantitative morphological analysis of etha nol effect upon rat liver. Gastroenterology 62. 1020. Feron. V. J, & Kroes, R. (1979). One-year time-sequence inhalation toxicity study of vinyl chloride in rats. II. Morphological changes in the respiratory tract, cer uminous glands, brain, kidneys, heart and spleen. Toxicoloyy 13. 131. Feron. V. J_ Kruysse. A. & TiL H. P. (1979a). One-year time sequence inhalation toxicity study of vinyl chloride in rats. I. Growth, mortality, haematology, clinical chem istry and organ weights. Toxicology 13, 25. Feron. V. J_ Speek, A. J.. Willems. M. 1_ van Battum. D. & de Groot, A. P. (1975). Observations on the oral admin istration and toxicity of vinyl chloride in rats. Fd Cosmet. Toxicol. 13. 633. Feron. V. J_ Spit, B. J,, lmmel. H. R. &. Kroes. R. (1979b). One-year time-sequence inhalation toxicity study of vinyl chloride in rau. 111. Morphological changes in ibe liver. Toxicology 13, 143. Fuchs. G- GawalL B. M_ Albanus. L. & Slorach. 5. (1975k Vinyl chloride monomer levels in edible fats. Vdr Poda 27. 134. Gilbert. D. & Golberg. L. (1965). Liver response tests. ML Liver enlargement and stimulation of microsomal pro cessing enzyme activity. Fd Casmet. Toxicol. 3, 417, Gorier, E. &. De Graaff. W. C. (19551 Klinische Diagnostiek. 1. 7th Ed. p. 267. Edited by H. E Slenfert. Krocsc NV, Leiden. Gram. T. E_ Wilson. J, T. &. Fouls. J. R. (19681 Some characteristics of hepatic microsomal systems which metabolize aminopyrine in the rat and rabbit. J. Pharmoc. exp. Ther. 159, 17i Haley. T. J. (19751 Vinyl chloride: how many unknown problems? J. Toxicol, encir. Hlth 1, 47. Hojmberg. B- Kronevi, T. & WioelL W. (19761 The path ology of vinyl chloride exposed mice. Acta cet. scand. 17, 328." Hopkins. J. (1980). Vinyl chloride--Part 4: Carcinogenicity in man. Fd Cosmet. ToxicoL 18, 94. Jiihe. S.. Lange. C. E Stein. G. & Veltman. G. (19731 Ober die sogenannte Vinylchloride-Krankheit. Dt. tned. Wschr. 98. 2034. Keplinger. M. L. Goode, J. W_ Cordon, D. E. &. Calandra. J. C. (1975). Interim results of exposure of rats hamsters, and mice to vinyl chloride. Ann. N.Y. Acad. Sci. 246. 219. Lange. C. E_ Jiihe. S.. Stein. G. & Veltman. G. (1974). Die sogenannte Vinylchloride-Krenkheit--cine berufsbedingte Systemsklerosc? Ini. Arch. Arbeilsmed. 32. 1. Lee, C. C- Bhandari. J. C,, Winston. J. M- House. W. B, Dixon. R. L. & Woods. J. S. (19781 Carcinogenicity of vinvl chloride and vinylidene chloride. J. Toxicol, encir. Hlth. 4. 15. Lester. D.. Greenberg. L, A. &. Adams, W. A. (1963). Effects of single and repeated exposures of humans and rats to vinyl chloride. Am. ind. Hyg. Ass. J. 24. 265. Lloyd. J, W. (1975). Angiosarcoma of the liver in vinyl chloride/polyvinyl workers. J. occup. Med. 17, 333. Malioni. C. (1975). The value of predictive experimental bioassays in occupational and environmental carcino geneses. An example: vinyl chloride. Amhio 4. 18. Malioni. C. &. Lefemine. G. 119751 Carcinogenicity bioas says of vinyl chloride: current results. Ann. H.Y. Acad. Sci. 246. 195. Malioni. C. (19771 Vinyl chloride carcinogenicity: an ex perimental model for carcinogenesis studies. In Origins of Human Cancer. Book A. Edited by H. H. Hiatt. J. D, Watson & J. A. Winsten. p. 119. Cold Spring Harbor Laboratory. NY. Mastromatteo. E_ Fischer. A. M- Christie. H. &. Danziger. H. 11960). Acute inhalation toxicity of vinyl chloride to laboratory animals. Am. ind. H\g. Ass. J. 21. 394. Monson, R. R.. Peters. J. M. & Johnson, M. N. M914L Proportional mortality among vinyl-chloride workers. Lancet II, 397. Muller. K. T Buchter. A_ Gross. R. & Bolt, W. (19761 Ergebnisse einer Sludie an 17 Fallen von Langzeitexposition gegeniiber Vinylchlorid. Medsche Welt. Bert. 27. 21. Paschen. K. & Fuchs. C. (19711 A new micro-mtthod for Na. K. Ca and Mg determinations in a single serum dilution by atomic absorption spectrophoiometry. CHnica chim. Acta 35. 401. Popper. H.. Selikoff, 1. J, Malioni. C_ Squire. R. A. & Thomas. L. B. (1977). Comparison of neoplastic hepatic lesions in man and experimental animals. In Origins o} Human Cancer. Edited by H. H. Hiatt. J. D. Watson & J. A. Winsten. p. 1359. Cold Spring Harbor Laboratory. NY. Popper. H- Thomas. L. B- Telles. N. C- Falk. H. i'Selik- ar Mgtewtinuth*e ch. #5 s. Vir Foda tie tests. III. osomal pro3.417. -'he Diagnostfert Kroese 19681. Some leros which bit. J. Pkar- tv unknown ik The pathti. xond. 17, rcinogenicity (1973). Dber mtd. Wschr. & Calandra. ts. hamsters. id. Sci. 246. . (1974). Die bcrufsbed- 32, 1. ouse. W. B_ iQgenicity of oxic^^^cir. 9635). Effects and rats to ver in vinyl 7. 333. txperimcntal tal earcino4. 18. nicitv bioasN.Y. Acad. icity: an exi. In Origins HiatL J. D. ring Harbor A Danziger. chloride to . 394. 1. N. (19741. dc workers. - W. (1976). LangzeitexW/. Bert. 27, -method for .ingle serum ometry. eli te. R. A. & astic^Matic n of Wats3^: J. Laboratory. H. A Selik- V 9 5* %el . I 4 t J Chronic oral toxicity of vinyl chloride in rats 333 off. I. J. (1978). Development of hepatic angiosarcoma in man induced by vinyl chloride, thorotrast and arsenic: comparison with cases of unknown etiology. Am. J. Path. 92. 349. Potter. H. R. (1976). Vinyl Chloride--Part II. Fd Cosmet. Toxicol. 14. 498. Randolph. W. F. (1973). Prior-sanctioned polyvinyl chlor ide resin. Federal Register 38. 12934. Rawls, R. (1980). Studies update vinyl chloride hazards. Chem. Engng News 58 (14k 27. Reitman. S. St Frankek S. (1937). A colorimetric method for the determination of serum gluiamic-oxalacctic and glutamic-pyruvic transaminases. Am. J. din. Path. 28, 56. Reynolds. E. S,, Moslen. M. T. Szabo. S. Jaeger, R. J. A Murphy, S. D. (1975).. Hepatotoxicity of vinyl chloride and 1.1-dichloroethylene: role of mixed function oxidase system, ,4m. J. Path. 81, 219. Sharratt, M. & Frazer. A. G (1963k The sensitivity of func tion tests in detecting renal damage in the rat. Toxic, appl. pharmoc. 5, 36. Squire. R. A. & Levitt M. H. (1975). Report of a workshop on classification of specific hepatocellular lesions in rats. Cancer Res. 35, 3214. Suzuki. Y. (1978). Pulmonary tumors induced in mice by vinyl chloride monomer. Envir. Res. 16, 285. Tandler. B. Erlandson, R. A. St Wynder. E. L. (1968k Riboflavin and mouse hepatic cell structure and func tion. 1. Ultrastructural alterations in simple deficiency. Am. J. Path. 52, 69. Thomas. L. B. & Popper. H. (1975). Pathology of angiosar coma of the liver among vinyl chloride-polyvinyl chlor ide workers. Ann. N.Y. Acad. $ci. 246, 268. Torkelson. T. R,, Oven, F. St Rowe. V. K. (1961k The toxicity of vinyl chloride as determined by repeated ex posure of laboratory animals. Am. ind. Hyg. Ass. J. 22, 354. Vale. P. T, Kipling. M. D. & Walker. A. E. (19761 Miscel laneous symptoms occurring in workers engaged in the manufacture of PVC. J. Soc. oecup. Med. 26. 95. van Kampen, E. J. A Zijlstra. W. C (1961L Standardiz ation of hcmoglobinometry. II. The hemoflobincyanide method. Clinica ehim. Acta 6, 538. Watanabe. P. G,, Hefner. R. E, Jr A Gehring. P. J. (1976k Vinyl chloride-induced depression of hepatic non-pro tein suUhydryl content and effects on bromosulphakin (BSP) clearance in rats. Toxicology 6, 1. Watanabe. P. G_ McGowan, G. R. St Gehring. P. J. (1976k Fate of [uC]vinyl chloride after single oral administra tion in rats. Toxic, appl. Pharmac. 36, 339. Watanabe, P. G- McGowan, G. R- Madrid. E. O. St Gehr ing. P, J. (1976k Fate of [uC]vinyl chloride following inhalation exposure in - rats. Toxic appl. Pharmoc. 37, 49. Waxweiler. R. J- Stringer. W, Wagoner. J. K. Jones, J. Falk. H. St Carter. C (1976k Neoplastic risk among workers exposed to vinyl chloride. Ann. N.Y. Acad. Set 271,40. Wieme, R. J. (1965k Agar Cel Electrophoresis, p. 425. Else vier Publishing Corp, Amsterdam. . Williams. D. T. St Miles. W. F. (1975). Gas-liquid chroma tographic determination of vinyl chloride in alcoholic beverages, vegetable oils and vinegars. J. Ass. off. anolyt. Chem. 58, 271 Williamson. K. S. (1976k Review of animal studies. Proc. R. Soc. Mtd. 69, 281. WinelL M, Holmberg. B. St Kronevi. T. (1976). Biological effects of vinyl chloride: an experimental study. Envir. Hlth Ptrspect. 17, 211. Wroblewski. F. & LaDue, J. S. (1955). Lactic dehydro genase activity in blood. Proc. Soc. exp. Biol. Mtd. 90. 210. R&S 006942