Document M4Zeej75amK97bRZBdw7gv1BL

THE TOXICOLOGY OF PCB*S An Overview with Emphasis on Human Health Effects and Occupational Exposures Prepared by the Hazard Evaluation System Epidemiological Studies Section State of California Department of Health Services/Department of Industrial Relations 2131 Berkeley Way Berkeley, California 94704 (413) 340-2113 January, 1981 HONS 209764 THE TOXICOLOGY OF PCB'S Table of Contents I. INTRODUCTION....................................................................................... ...... II. GENERAL BACKGROUND INFORMATION.............................................. 3 III. PHARMACOKINETICS ................................................................................. 5 Absorption......................................................................................................... Distribution, Accumulation (Mammals)..........................................................5 Transplacental Exposure, Secretion in Milk.............................................. 6 Metabolism........................................................................................................ 7 IV. ANIMAL TOXICOLOGY............................................................................... 10 Acute.............................................................................................................10 Sub-acute, Chronic..................................................................................... 11 Reproductive Effects..................................................................................... 12 Other - lmmunosuppresive, Endocrine........................................................ 14 V. CARCINOGENICITY/MUTAGENICITY........................................................ 16 Carcinogenicity............................................................................................ 16 Test Results................................................................ 16 Mutagenicity................................................................................................. 19 VI. BIOCHEMICAL EFFECTS................................................................ 21 Enzyme Induction........................................................................................... 21 Porphyria............................. 22 VII. HUMAN TOXICOLOGY AND EPIDEMIOLOGY....................................... 24 Dermatologic Effects.............................................. 24 Systemic Symptoms..................................................................................... 25 Liver Damage................................................................................................. 25 Yusho.............................................................................................................25 Neurotoxicity................................................................................................. 27 Cancer................................................................ 27 Ongoing Occupational Studies .... 25 VIII. MEDICAL SURVEILLANCE AND BIOLOGIC MONITORING ... 31 IX. SUMMARY AND CONCLUSIONS............................................................... 33 _ HONS 209765 1 Table of Contents REFERENCES............................................................................................................ 34 Table l. Dose-Response for Animal Toxicology...........................................39 Table II. Occupational Exposure to PCBs..................................................... 40 Table 111. Percent Distribution of Symptoms of Yusho Reported by 189 Patients Examined Before October 3, 1968 ............................. 44 Figure 1.................................................................................................................... 45 Figure 11.......................... 46 Figure III.................................................................. 47 Figure IV.....................................................................................................................48 HONS 209766 ii L INTRODUCTION As a consequence of the EPA ban on further manufacture of PCB's* in 1977, occupational exposures to these compounds have been drastically reduced. However, significant exposures may remain for particular occupational groups. Utility workers, for example, may experience sporadic but potentially massive exposures when cleaning up spills, or when servicing and dismantling transformers and capacitors that still contain PCB fluid. Electricians, appliance service workers and firefighters also may have continued occupational exposure. NIOSH estimates that 12,000 workers have potential exposure as a result of current uses of PCB's (NIOSH, 1977), Despite the vast scientific literature on the toxicology of PCB'S, the human health effects likely to result from such exposure remain ill-defined. The Hazard Evaluation System has reviewed the literature on PCB toxicology in response to inquiries about worker health. Requests for information have come from unions and workers who handle PCB fluids in clean-up of spills, in maintenance work, and in transportation, storage and disposal of used equipment. Toxicology information has also been requested by medical professionals evaluating the clinical significance of PCB exposures, and public health officials who are attempting to set standards for occupational and environmental exposures. Our primary goal has been to review the data relevant to the human health effects of PCB's especially those resulting from occupational exposures. Since the published epidemiologic evidence is limited we have utilized animal toxicology studies where appropriate in anticipating potential biologic effects in humans. We have not attempted to summarize the extensive literature on PCB toxicology, but the reader is referred to a number of recent reviews (DHEW, 1978; 1ARC, 1978; Fishbein, 1974} Kimbrough, 1974; EPA, 1977; Nelson, 1972; NIOSH, HONS 209767 1 1977). * PCSs: Polychlorinated Biphenyls HONS 209768 2 IL GENERAL BACKGROUND INFORMATION In all animal species that have been studied PCB's have a very low acute toxicity. They are readily absorbed across biological membranes, poorly metabolized and only very slowly eliminated. Because PCB's persist in the environment and accumulate in living tissue, they are concentrated ("biomagnified") in the food chain in a similar manner to other organochlorine compounds like DDT. Concern about exposure to PCB's has, therefore, focused on their resistance to biodegradation with the consequent potential for long-term or delayed health effects. A number of published reports have established "background" levels of PCB's in the blood and tissues of human populations with no previous history of exposure. Surveys in various geographical areas have found detectable residues in blood, fat and mothers' milk. Measurable levels of PCB's are typically found in greater than 50% of subjects tested with maximum blood levels generally less than 20 ppb (Finklea, 1972). The levels reported from adipose tissue are typically somewhat higher, in the range of 1-2 ppm (Kutz, 1975). Residues of PCB's in human milk have ranged from 40-100 ppb in whole milk (New York State Health Council, 1977X Two facts complicate the documentation of the human and animal toxicology of PCB's; 1. Commercial products are rarely single agents, but rather are complex mixtures of chlorinated biphenyls with different numbers and arrangements of attached chlorine atoms (see Figure I). The metabolism and toxicology of PCB's seem to vary with the percent of chlorination and with the isomeric structure of the PCB molecule. 2. All commercial products are potentially contaminated with chlori nated naphthalenes and polychlorinated dibenzofurans (PCDPs). The degree of this contamination varies with different commercial mixtures (see Figure in). HONS 209769 3 Contamination by dibenzof urans (PCDF's) is of particular concern because of the structural similarity of these compounds to the highly toxic dibenzodioxins (see Figure It). The pattern of observed effects in animals exposed to PCDF's closely resembles that seen following exposure to 2,3,7,8 tetrachlorodibenzodioxin (TCDD). In comparative animal studies the toxicity of PCDF's is much greater than the PCB's, particularly in the thymus, skin (acne), liver and hematopoetic system (Oishi et al., 1978; Moore et ah, 1979). In addition, PCDF's are 1000 times more potent than PCB's as enzyme inducers (see Section VI). Uncertainties in the analytic methods used for detection of PCB's must be considered when reviewing the published data on PCB toxicology; Monitoring PCB's in environmental or biological samples by gas-liquid chromatography/mass spectrometry (GC/MS) is made difficult by the presence of other chlorinated hydrocarbons (e.g.t pesticides) which are commonly present at similar concentrations (Stalling et al., 1979). Because of the difficulties in the interpretation of GC/MS spectra, the PCB levels reported from different laboratories may show considerable variation. HONS 209770 it CL PHARMACOKINETICS Absorption There is relatively little information on the rate or degree of absorption of PCB's by any route for any species of animal. Since similar systemic toxicity has been observed in rodents after dermal, oral and inhalational adminstration of comparable doses, it is likely that PCB's are easily absorbed by all routes. The few quantitative measurements of relative absorption rates indicate that most, if not all, PCB's which contain six or fewer chlorine atoms are efficiently absorbed from the Gl tract (Albro and Fishbein,. 1972; Van Miller et al., 1975). Distribution, Accumulation (Mammals) As with other heavily chlorinated chemicals the major storage tissue for PCB's is body fat. The concentration in adipose tissue is 10-1000 times that found in other tissues, both following single oral doses (Grant et al., 1971) and after chronic administration. (Curley et aU 1971) The lowest concentrations are found in whole blood and plasma where levels are usually several fold lower than in other tissues examined. This preferential distribution of PCB's into fat has been well documented after i.v. dosing in the rat (Lutz et al., 1977). Results are consistent with the high distribution coefficient of PCB's in fat and low perfusion of adipose tissue compared to skin, liver, muscle and blood (Anderson et al., 1977). Two important pharmacokinetic questions cannot row be resolved on the basis of available data: HONS 209771 5 1. Does the concentration of PCB's reach steady state with constant exposure? and ' 2. Will mobilization of adipose tissue after starvation or illness lead to a transient increase in PCB concentrations in blood and other tissues 7 Like most heavily chlorinated hydrocarbons the half-life of PCB's in animal tissue is quite long. In a chronic feeding study with Aroclor 1245 at 100 ppm in the diet of rats, a steady build-up of PCB's occurred in all tissues analyzed without a plateau level even after 240 days of treatment. By comparison, in a similar rat dietary study using DDT, plateaus in fat were attained after 90-140 days (Curley et al., 1971). In dairy cows a steady state was reached in 40-60 days (Fries, 1972), probably due to mobilization into fat micelles and secretion into milk. Thus, PCBs are unlikely to reach steady state levels in non-lactating animals; and fat mobilization or lactation may be expected to result in release of stored PCBs. Transplacental Exposure, Secretion in Milk Transplacental exposure of PCB's has been documented in mammals. Term fetuses taken from rats exposed to 10 mg/kg/day during days 7-15 gestation contained 0.63 ppm PCB, or about 1/60 the maternal dose. When the dose to the mother was increased five fold, the concentration in the fetuses increased two fold (Curley et ah, 1973). The hyper pigmented babies observed in the Yusho incident (see Section VII) represent additional circumstantial evidence of transplacental passage of PCB's. Secretion of PCB's into milk has also been observed. In mice, little passage of PCB's occurred across the placenta once PCB's had been sequestered into maternal adipose tissue, but they were readily transferred to suckling offspring through the milk (Vodicnik and Lech, 1980). These observations suggest that secretion of PCB's into milk may be quantitatively much more important as a source of exposure in newborns than is HONS 209772 6 transplacental passage. This has recently been documented in a prospective study of Japenese mothers and their infants. (Hirokadzu and Ota, 1980) Metabolism PCB's are metabolized primarily by hydroxylation and conjugation with glucuronic acid. The primary site of bio transformation is assumed to be the liver, although no data is available currently on the possible role of peripheral metabolism, e.g., skin. Many experimental feeding studies in both mammals and birds have shown an inverse relationship between percent chlorination and rates of metabolism. Less chlorinated PCB's are more readily metabolized than are more chlorinated ones, the rate of metabolism and excretion decreasing sharply as the number of chlorine atoms increases above five (EPA, 1977). The metabolism of the higher chlorinated biphenyls is also dependent on the position of chlorine atom substitution. The presence of two adjacent, unsubstituted carbon atoms is needed for the rapid enzymatic hydroxylation reaction (Jensen and Sundstrom, 1974). Since the more highly chlorinated biphenyls have a very much slower metabolic rate and longer half-life, they are more generally found as residues in human and animal tissues. This relationship alters with PCB's above 54% chlorination, presumably as a result of lower absorption from the gastrointestinal tract. Arene oxide intermediates have been described in a major pathway of the metabolic transformation of PCB's by hepatic mixed function oxidases (Safe et al., 1975; Gardner et al., 1973). These intermediates are of particular concern since they are capable of direct interaction with DNA and may be the active form of carcinogenic polyacyclic hydrocarbons. (Jerina and Daly, 1974) The PCB molecules which are more readily metabolized and excreted also are more likely to form these arene oxides. It does not necessarily follow, however, that those compounds which persist In tissue and are more 7 MONS 209773 likely to be measured In population sampling are less important in terms of their carcinogenic potential. From the limited data available, it appears that significant differences exist between non-hum an primates and rodents in the metabolism and pharmacokinetics of PCB's. The marked variation observed in PCB toxicity between rodents and primates may be explained by such differences. Primates appear to be more susceptible to the toxic effects of PCB's than are rats or mice (see Section IV and Figure IV). When single doses of radiolabeled PCB were administered by gastric intubation to infant monkeys, the metabolites measured in urine, tissue (liver) and serum included hydroxylation products derived from arene oxide intermediates (Allen et al., 1976); while in the rat, direct hydroxylation is the rule (Hsu et al., 1975), There is virtually no pharmacokinetic data in humans. A few generalizations can be made, however, based on studies reporting PCB blood levels: The higher the exposure levels, the higher blood concentration of PCB's (Hara et at., 1975; Inoue et al,, 1975; Karppanen and Lolho, 1973; Baker et al,, 1980); and the higher the environmental concentration and/or the longer the period of exposure, the longer the blood levels of PCB's remain elevated (Hara et al., 1975; Baker et al., 1980). However, there are a few reports which are inconsistent with this latter trend (Bumgarrer et al., 1975; Hasegawa et aL, 1972; Kitamura et al,, 1973). PCB levels have also been correlated with race and geographic residence (NIOSH, 1977) and with age and dietary intake of fish (Kimbrough, 1980) Comments Certain generalizations can be made from the limited pharmacokinetic data that is available: HONS 209774 8 1. Absorption occurs by all routes (skin, Gl, inhalation). 2. Distribution is primarily into fat. 3. Metabolism and excretion are dependent on specific molecular structure, varying inversely with percent chlorination. 4. Excretion is in general quite slow so that bio-accumulation occurs even at low exposure levels. 5. Transplacental transfer occurs but may be quantitatively less significant than secretion into milk, 6. Arene oxide metabolites are found in the metabolic transformation of PCBs. There compounds are highly reactive and may represent the active car cinogens. (see Section V) 7. The relationship between percent chlorination and.potency as carcinogen has not been established. 8. There are essentially no pharmacokinetic data in humans; it is not known, for example, if intermittent high doses are more or less hazardous than low level chronic exposures to the same total dose. HONS 209775 9 IV. ANIMAL TOXICOLOGY Acute Toxicity When given as a single dose, the acute oral LDjq of PCB's in rats, rabbits and mice ranges from 1-10 grams/kg of body weight. According to the American Industrial Hygiene Association classification system for acute toxicity, PCB's are classified as "slightly toxic" (0.5 - 5 g/kg), or "practically non-toxic" (5-15 g/kg). There is some evidence that young animals are more sensitive than adults, and that females are more susceptible than males to the acute effects of PCB's (Kimbrough et al., 1978). In rodents, the acute oral toxicity appears to decrease with increasing chlorine content of the administered PCB's. This may be secondary to decreased absorption of the higher chlorinated compounds or to the differences in metabolic transformation previously discussed. Although few clinical signs of toxicity have been reported in experimental animals, pathologic findings are extensively documented. CNS depression (decreased pain response and diminished exploratory behavior), anorexia and oliguria followed by ataxia, coma and death have been observed in rats following acute administration of large doses of PCB's (Brackner et al., 1973). Consistent pathologic findings associated with death in rats, rabbits and guinea pigs include liver damage with fatty infiltration, centrolobular atrophy, and in some cases necrosis. Pathologic changes in other organs in these species are not often described, except for chloracne-like lesions which occur at the site of skin or intradermal application. Comments The tow order of acute toxicity in experimental animals is consistent with the lack HONS 209776 10 of acute effects observed in workers exposed to PCB's. Reported symptoms after occupational exposures include mild irritation of the skin and eyes at levels above 0.1 mg/m3 with unbearable irritation occurring above 10 mg/m3 (ACGIH, 1976). Systemic symptoms of nausea and headache have been reported but may be secondary to the solvents (such as trichlorobenzene) in the PCB mixtures. Sub-acute and Chronic Toxicity In contrast to the low order of acute toxicity, effects from chronic exposures to relatively low doses of PCBs have been consistently observed and are of far greater concern. These sub-acute effects show appreciable variation among species, but liver damage is again the prominent finding. The major changes in rats fed Aroclor 1248, 1254, and 1262 at 100 ppm in their diet for six weeks included liver hypertrophy, marked fatty infiltration and degeneration of parenchymal ceils. As in acute toxicity studies PCB mixtures with lower chlorine content were more toxic (Allen and Abrahamson, 1973). In 8-12 month feeding studies increased serum lipids and focal areas of liver damage were observed (Allen et al., 1976; Kimbrough et al., 1972). Non-human primates are more sensitive than rodents to the toxic effects of PCB's (see Table I and Figure IV). Adult female monkeys exposed to dietary levels of 2.5 ppm for 12 months (*0.08 mg/kg/day) developed facial edema, alopecia, acne, gastritis with ulceration, anemia, hypoproteinemia and bone marrow hypoplasia. At 100 ppm (*I0 mg/kg/day) there was considerably more evidence of tissue damage than in rats, including marked hepatic hypertrophy with ultrastructural abnormalities (Allen, 1975). Based on extrapolation from the Yusho data, PCB's may cause symptoms to humans at levels (0.2 mg/kg/day) which are comparable to the lowest doses which produce effects in non-human primates (see Table I). 11 HONS 209777 Most of the animal data are derived from feeding or oral intubation studies. There are relatively few reports of dermal or inhalation experiments. Inhalation studies again revealed liver damage to be the prominent finding in rodents. For summary of inhalation data see the NIOSH criteria document, page 125 (NIOSH, 1977). Dermal toxicity studies in rabbits have produced skin lesions at the site of application as well as systemic effects including liver and kidney damage, thymic atrophy, lymphopenia and increased fecal porphyrins (Vos and Beems, 1971). Comments Liver damage, documented histologically, is the consistent finding among various laboratory animals exposed to low chronic levels of PCB's. It has not been observed in the limited surveys of exposed workers. This may be an artifact, however, of the relative insensitivity of the standard liver function tests (such as serum levels of SCOT, SGPT) as compared to biopsy and histologic analysis. Reproductive Effects Adverse reproductive effects of PCB's have been noted in many mammalian and avian species. The pattern of reproductive effects include alterations In estrus cycles, failure of implantation, increased frequency of spontaneous abortions, low birth weight offspring and decreased post-natal survival. No specific teratogenic effects of PCB's have been observed in a variety of avian species. Transplacental effects, however, have been documented in both animals and humans (see Section VII). PCB's given to mice for 10 weeks at dosage of 1.0 mg/kg/day lengthened the estrus cycle by more than two days and decreased the number of successfully implanted ova (Orberg and Kihlstrom, 1973). Similarly, mice that received PCB's as sucklings in a HONS 209778 12 long-term transgenerational study showed subsequent alterations in estrus cycles, decrease in implantations, and when mated to each other (FI studies), reduced number of offspring per litter (Kihlstrom et al., 1975). In rats, studies suggest that reproductive effects of PCB's decrease as chlorination increases. No reproductive effects have been found with Arodor 1260 (60 percent chlorination) at 1, 10, 100 ppm, but significant effects have been noted with Arodor 1242 and 1254 (42 and 54 percent chlorination, respectively) at doses of 20 and 100 ppm. Arodor 1260 began to exert toxic effects at doses of 500 ppm. Rats chronically fed from 20 to 100 ppm Arodor 1242 and 1254 had reduced numbers of offspring. Surviving newborns showed increased mortality, with only 30 percent surviving to weaning. Five ppm of either Arodor 1242 or 1254 produced no effects over two generations. Thus, the minimum effective doses ranged from 20 ppm for the lower chlorination mixtures to 100-500 ppm of the more highly chlorinated compounds. (Keplinger et al., 1971; Linder, et al., 1974) Evidence of adverse reproductive effects is also available for non-human primates. Rhesus monkeys fed 2.5 and 5.0 ppm Arodor 1248 for 18 months in the diet showed changes in menstrual cycles in addition to other systemic signs of toxldty. Evidence was also obtained for frequent resorptions and spontaneous abortions following breeding to normal males. In ail, six infants were carried successfully to term out of 14 pregnandes. The offspring were of low birth weight and by two months began to show evidence of PCB toxidty, presumably from PCB's in the maternal milk; only three infants survived to six months. Behavioral tests in the three surviving animais showed marked defidts in several learning tasks, with increasing errors correlated with increasing body burdens of PCB's (Allen and Barsotti, 1976; Bowman et al., 1978; Barsotti et al., 1976). HONS 209779 13 The effect of PCB's on the male reproductive system is not known. There is one report of four male Rhesus monkeys exposed to 5.0 ppm Aroclor 1248 In the diet for 18 months. After 12 months, one of four animals developed clinical signs of PCB intoxication, showed marked sperm count depression and was functionally sterile. A testicular biopsy revealed an absence of spermatogonia. A second biopsy one year after exposure showed complete recovery (Allen et al., 1979). PCB's are negative in the mouse sperm morphology assay (Heddle and Bruce, 1977). Comments PCB's show significant effects on reproductive competence in a variety of species. These effects increase in intensity with increasing dosage and decrease with increasing chlorination of the PCB isomers. PCB's do not appear to be mammalian teratogens. A reasonable explanation for most of the reproductive effects of PCB's could be based on their estrogenic activity (see below). PCB's have been detected in human semen (Dougherty et al., 1980), but there have been no studies of semen quality in relation to PCB exposure in humans. The effects of PCBs on the male reproductive system in animals or humans has not been adequately studied. The only other evidence to date on reproductive toxicity in humans come from the Yusho incident and is summarized in Section VU. Other 1. Immunosuppressive Effects A number of reports implicate PCB's as immunosuppressants (Fishbein, 1974). Lymphoid atrophy has been observed in rabbits, chickens and guinea pigs. Suppression of humoral immune responses to several antigens was observed in rabbits and guinea pigs, and decreased cell-mediated immune response HONS 209780 14 followed PCB exposure in guinea pigs, A decreased tolerance to hepatitis virus was seen in ducklings without apparent intoxication. In monkeys exposed transplacentally and through contaminated milk, the lymph nodules of the spleen were extremely small and without germinal centers (Allen and Barsotti, 1976); morphologic changes were indicative of reduced immunologic compe tence. 2. Endocrine Effects Subcutaneous administration of Aroclor compounds with lower chlorination produced an estrogenic effect on the rat uterus which was not shown with Aroclor s of higher chlorination (Bitman and Ceal, 1970). Female primates fed Aroclor 124$ for six months showed an increase in concentration of urinary ketosteroids and a prolongation of their menstrual cycles with increased bleeding (Barsotti et al., 1976). Antiandrogenic effects have been described in birds although the mechanism is not clear. It may be secondary to an increased rate of androgen metabolism in the liver by induction of microsomal enzymes (see Section VI), or by virtue of PCB's exerting estrogenic effects. Comments The effect of PCB exposure on immune and endocrine system function has not been carefully studied in humans, so the relevance of these animal observations to human health remains unknown. There is one cross-sectional study of occupational exposure to PCB's which will include analysis of serum hormone levels and urinary metabolites, but results have not yet been published (Selikoff et al., in progress). HONS 209781 13 V. CARaNOGENICITY/MUTAGENICITY Carcinogenicity Several PCB mixtures are dearly cardnogenic in rodent bioassays, producing liver tumors (hepatocellular cardnomas). Kanechlor 500 and Arodor 1254 are cardnogenic in male mice (Ito et al., 1973; Kimbrough and Linder, 1974); and Arodor 1260 is cardnogenic in separate studies in two strains of female rats (Kimbrough et al., 1975; Norback et al., 1980). In addition, a purified component of a PCB mixture, 2,4,5,2',4',51 -hexachlorobiphenyl, has recently been found to be cardnogenic in female rats, causing hepatocellular cardnomas (Norback et al., 1980). Because high doses of PCB's are known to cause extensive injury to liver tissue it is important to consider the dose levels at which liver cardnomas were produced in the rodent bioassays. In two studies in rats, significant increases in hepatocellular carcinomas were present at doses which did not produce gross histologic changes. Hepatocytes were somewhat enlarged (probably due to microsomal enzyme induction), but no extensive fatty infiltration or necrosis occurred, as was characteristic of bioassays at higher dose levels (Kimbrough et al., 1975). Test Results 1.. Mice (Male) A. Kanechlor 300, 400, and 500 fed to groups of 12 eight-week-old male mice at 100, 250, and 500 ppm in the diet for 32 weeks produced hepatocellular cardnomas in 5 of 12 survivors in the high dose group fed Kanechlor 500. The remaining 7 mice in this group had nodular MONS 209782 16 hyperplasia (neoplastic nodules). No metastases or other tumors were present in this or other dosed groups. The control group (6 mice) was likewise tumor-free (lto et al., 1973). B. Arodor 1254 administered to groups of 50 five to six-week-old male BALBc/3 mice at dietary levels of 0 or 300 ppm (about 50 mg/kg body weight during the exposure) for 11 months produced neoplastic nodules (hepatomas or hyperplastic nodules) in 9 of 22 survivors in the dosed group. Other liver lesions (adenofibrosis) were present in all 22 survivors. Additional morphological changes in the livers of these animals included pleomorphism and areas of necrosis. Such changes and tumors were absent among survivors (24) in the control group (Kimbrough and Linder, 1974). A. Kanechlor 400 administered to ten-week-old Donryu rats (10 males and 10 females) at dietary levels which varied from 40-600 ppm during the 400-day study produced liver tumors (multiple adenomatous nodules) in 6/10 treated female rats. Such lesions were absent from the controls (5 males and 5 females) and the treated males (Kimura and Baba, 1973). B. Kanechlor 300, 400, or 500 administered to groups of 30 eight-week-old male Wistar rats at dietary levels of 0, 100, 500, or 1000 ppm produced increases in the incidence of cholangiofibrosis at the highest dose level of all Kanechlors (2/15, 2/10, and 4/13, respectively). All three compounds also produced hepatic nodular hyperplasia, the incidence of which increased with dose and extent of chlorination (Kanechlor 300 at 100 ppm: 1/22} Kanechlor 400 at 100 ppm: 2/16, and 1000 ppm: 3/10; Kanechlor 500 at 100 ppm: 3/25, at 500 ppm: 5/16, and at 1000 ppm: 5/13); (lto et al., 1974). HONS 209783 17 C. Aroclor 1260 administered to groins of 200 three to four-week-old female Sherman rats at 0 and 100 ppm in the diet (varying between 5-10 mg/kg body weight during the 21 month exposure) produced at 23 months among the dosed survivors clearly significant increases of hepatocellular carcinomas (controls 0/173; dosed group 146/184) as well as neoplastic nodules (hyperplastic nodules: controls 0/173; dosed group 26/184). The incidences of non-hepatic tumors did not differ between the dosed and control groups (Kimbrough et al., 1975). D- Aroclor 1254 administered to groups of 24 eight-week-old Fisher 344 rats of either sex at dietary levels of 0, 25, 50, or 100 ppm for 105 weeks was not carcinogenic to any of the treated groups under the test conditions. It is important to note that two of the dose levels used were lower than those which produced a positive response in Sherman rats. Rare adenocarcinomas and carcinomas of the gastrointestinal tract appeared in both sexes and may be related to the administration of the PCBs (males: historical controls 6/600, dosed group 2/24). In addition a high incidence of non-neoplastic liver hyperplasia was present among the dosed groups (males: controls 0/24, low-dose 5/24, mid-dose 8/24, high-dose 12/24; females: controls 0/23, low-dose 6/24, mid-dose 9/22, and high-dose 17/24); (NCI 1978). E. Aroclor 1260 administered to groups of 50 male and female SpragueDawley rats at dietary levels of 0 and 100 ppm for 105 weeks was carcinogenic in female rats, causing significant increases in liver hepatocellular carcinomas (Norback and Weltman, 1980). HONS 209784 18 F. A purified component of a PCB mixture, 2,4,5,2',4',5-hexachlorobiphenyl administered to groups of 50 male and female Sprague Dawley rats at dietary levels of 0 and 100 ppm for 105 weeks was carcinogenic in female rats, producing an increased incidence of liver hepatocellular carcinomas among the dosed animals (Norback and Weltman, 1980). Mutagenicity PCB mixtures have not been observed to have mutagenic activity nor to measurably affect chromosomes In repeated studies using a variety of in vitro or in vivo test systems Evidence of genetic damage from PCB's in laboratory test systems including chromosomal aberrations, non-disjunction, loss of sex chromosomes or increased frequency of sister chromatid exchange has not been observed. Report of a weak effect of Aroclor 1221 and of a stronger effect of 9-chlorobiphenyl in Salmonella using PCB-induced rabbit liver homogenate as a liver activation system appears unfounded (Wyndham et aU, 1976). Further attempts to repeat these results have been unsuccessful using a variety of Salmonella tester strains and liver activation systems (Katzeneilenbogen and Ames, 1980; Safe, 1978). However, PCB's belong to the class of heavily chlorinated animal carcinogens, most of which are not positive in short-term tests for mutagenicity. Examples in this class include dieldrin, chlordane, kepone, mirex, TCDD,# chloroform, and carbon tetrachloride. Whether this is because the in vitro metabolic activation systems do not produce the same spectrum of metabolites that occur in vivo or because heavily chlorinated compounds such as PCB's are carcinogenic by non-mutagenic mechanisms is not known at this time. TCDDt Tetrachlorodibenzodioxin 19 HONS 209785 Validation of the carcinogenic effects in rodents is provided by a positive cell transformation assay using C3H10T1/2 clone eight mouse fibroblast cells in culture by two separate PCB mixtures (Aroclor 1254 and 1260) and a purified component 2,4,5,2',4'5' -hexachlorobiphenyl (Norback and Weltman, 1980). Comments ' A wide variety of PCB mixtures have been subjected to rodent cancer bioassays and to numerous in vitro and in vivo short-term tests for mutagenicity. Several of these PCB mixtures are carcinogenic. None of the PCB mixtures are active in short-term tests for mutagenicity, a finding that holds true for most heavily chlorinated carcinogens. However, substantial confirming evidence for carcinogenicity is provided by positive cell transformation assays using these same PCB mixtures. Thus, under OSH A published criteria, PCB mixtures should be considered Category I* carcinogens. Both IARC (1ARC, 1978) and EPA (EPA, 1978) have concluded that based on available animal data PCB's should be considered as potential human carcinogens. * Category h Human evidence or^ two positive mammalian bioassays or 1 positive mammalian bioassay with supporting results in short term tests. Category 11; One positive mammalian bioassay. (Sources Occupational Health and Safety Letter Vol. 9, No. 24 November 8, 1979) HONS 209786 20 VL BIOCHEMICAL EFFECTS OF PCB'S Enzyme induction The principal biochemical effect of PCB's is the stimulation and induction of certain enzyme systems. Enzyme induction occurs in both the microsomal monooxygenase or cytochrome P-450 system and the aryl hydrocarbon hydroxylase or cytochrome P-448 system, and it has been observed in both man and experimental animals. Induction is not restricted to the liver. It occurs in numerous other organs including kidney, adrenal, lung, gut, skin, and testes. Fetal enzyme induction may occur via transplacental exposure, and induction may also occur by exposure to contaminated milk, (ref) Identification of structure-activity relationships for enzyme induction is difficult because of the large number of isomers in commercially prepared PCB's and because all commercial products contain trace amounts of polychlorinated dibenzofurans (PCDF's) which are orders of magnitude more potent as enzyme inducers than PCB's. (ref) In early studies using commercial Aroclors, potency for enzyme induction was found to be dependent on dtlorination of the PCB mixture. Later, when purified isomers were tested, potency was found to vary with the position of chlorine atom substitution (see Section Hi), (ref) Since rate of metabolism is also known to vary with isomeric configuration of the PCB molecule, it may be that potency for enzyme induction is simply a function of the relative rate of metabolism and excretion. The enzyme induction properties of PCB's are utilized in the metabolic activation system of in vitro bioassays for mutagenicity, it is unlikely, however, that enzyme HONS 209787 21 induction would consistently enhance the effects of carcinogens or pro-carcinogens; It might function synergistically to activate a d'lemical, but they also might function to deactivate reactive carcinogens. Both phenomena have been observed in rodent cancer bioassays. Porphyria Porphyria cutanea tarda (PCT) in humans is an acquired defect in hepatic porphyrin metabolism characterized by uropcrphcrinuria, photosensitivity and mechanical fragility of the skin. PCT can be produced experimentally by a number of drugs, including tetrachlorodibenzodioxins and PCB's. All of these agents have the ability to stimulate the activity of 2-aminolevulinic acid (ALA) synthetase which is the initial enzyme in the heme synthetic pathway. Experimental hepatic porphyria was observed in Sherman rats exposed to Aroclor 1254 in the diet. At doses of 100 ppm the animals became porphyric after a delay of approximately 2-4 months. The porphyria resembled hexachlorobenzene poisoning and human PCT (Goldstein et al., 1975). In chronic feeding studies ALA-synthetase induction occurs after rats have become porphyric, although with large single doses the enzyme induction is seen almost immediately after dosing the animals (Goldstein et al., 1975). It has not been established whether only certain isomers in the PCB mixtures or contamination with PCDFs is responsible for the production of hepatic PCT. Porphyria has not been reported in humans exposed to PCB's. Comments Enzyme induction has two important implications for human health: 22 MOHS 209788 1. The occurrence of disease secondary to the increased metabolism of endogenous or exogenous substances, and 2. The interference with medical therapy due to increased metabolism of administered drugs. PCB's are more potent enzyme inducers than phenobarbital, a drug that occasionally causes clinical problems due to its enzyme inducing effects. While the effects of phenobarbital decline after administration ceases, enzyme induction from PCB's persists long after cessation of exposure. HONS 209789 23 VIL HUMAN TOXICOLOGY AND EPIDEMIOLOGY Few good epidemiologic studies of the health effects of PCB's are available. Most studies reported in the literature have been characterized by one or more of the following shortcomings: 1. Small study populations. 2. Lack of accurate exposure data, 3. Simultaneous exposure of workers to other potentially harmful chemicals. 4. Lack of control for confounding variables, such as alcohol consumption. 5. Inability to separate PCB's from contaminants and/or difficulty in comparing PCB's manufacture by different firms. In spite of these problems, some health effects have been consistently reported in studies of workers occupationally exposed to PCB's. In addition, a large-scale poisoning which resulted from ingestion of PCB-contaminated rice oil has been well documented and resulted in multiple signs and symptoms attributable to PCB's. The health effects Identified in a review of the epidemiologic literature are summarized below, and Table Q briefly describes the major epidemiological studies from 1934 through 1980. Dermatologic Changes Chloracne, contact or allergic dermatitis, and brown chromodermatosis have been consistently reported in studies of workers exposed to PCB's (Hara et aL, 1975; Hasegawa et aL, 1972; Inoue et al., 1975; Kitamura et al., 1973; Baker et al., 1980; Meigs et al., 1954; Ouw et al., 1976; Schwartz, 1936). . 24 HONS 209790 Systemic Symptoms Nausea, digestive disturbances, headaches, upper respiratory problems, and persistent body odor have been reported as a result of occupational exposures (Ouw et al., 1976; Schwartz, 1936; Warshaw et ai., 1979). Liver Damage This effect has been reported in some studies (Hasegawa et al., 1972; Higuchi, 1976; Meigs et al., 1954; Ouw et al., 1976). However, some investigators reporting abnormal liver function tests did not control for additional chemical exposures, previous medical problems, drinking patterns, etc. These confounding variables could explain some or all of the marginal differences encountered. Yus ho (Japenese word translated as "oil disease") Both dermal and systemic health effects are well documented in the epidemiologic study of a poisoning epidemic in Japan caused by ingestion of contaminated rice oil in 1968 (Higuchi, 1976; Kuratsune et al., 1972) It is not clear how much the health effects observed in Yusho victims can be extrapolated to occupational exposures for the following reasons; 1. The average amount of PCB (Kanechlor 400) ingested was estimated to be 2 grams and the minimum, 0.3 gram (Kuratsune et al., 1972). This is a higher dose than has been reported in most occupational exposures. In - addition, the PCB's were ingested as opposed to inhaled or skin-absorbed as is the case with occupational exposures. 2. The contaminated oil contained "used" Kanechlor 400, the exact chemical compositon of which is unknown. 3. Frying of foods with the rice oil could have produced new compounds which may have altered the toxidty of the PCB's or the toxicity of possible contaminants. HONS 209791 23 <*. Yusho oil was shown to contain high concentrations of dibenzofurans. 5. Reported concentration of PCB's in the oil may not have been accurate enough to permit a rigorous quantitative analysis since the methods for estimating PCB's in foods were not fully developed at the time. Clinical features of the Yusho patients are listed in Table HI. The Yusho incident is also important because it clearly documents the potential for reproductive and transplacental effects in humans: A study was made of the thirteen infants of 11 mothers affected by Yusho and of 2 unaffected wives of patients* Two of the Yusho mothers had stillbirths; ten of the babies had transient greyish or dark-brown pigmentation of the skin, and 5 had similar pigmentation of the gingiva and/or nails; increased ocular discharge was present in 9; and 12 of the 13 infants were small when compared with the national average (Funatsu et at., 1972; Kikuchi et al., 1969; Kuratsune, 1976; Taki et al,, 1969). Babies born to patients even 3 years after severe PCB exposure tended to show pigmentation of the skin on the back and the gingiva, although the degree of pigmentation was less than that of babies born to the same mothers up to one year after the poisoning (Kuratsune, 1976). Congenital abnormalities have also been observed in PCB-intoxicated infants. In the population of 13 offspring of Yusho mothers, premature eruption of teeth was observed in 2 cases, and larger than normal frontal and occipital fontanelles, exophthalmos and the persistence of an abnormally wide sagittal suture were observed in 3 others. Mo other gross malformations were reported nor was any relationship between dose and outcome considered. (Funtasu et al., 1972). HONS 209792 26 Mothers' milk contaminated with PCB's also appears to be a source of exposure for infants: one baby showed signs of poisoning even though the mother had ingested the contaminated rice oil only after the baby was delivered. The infant began to show jigns of PCB intoxication after 3-4 months of breast feeding (Kuratsune, 1972; Yoshimura, 1974). Neurotoxicity Paresthesias were reported In over 30% of Yusho patients (see Table HI). In the Yusho epidemic more detailed neurologic examinations were performed in 21 cases admitted to a University hospital in northern Japan. Ten of the patients complained of numbness or pain in the distal extremities, and in five cases decreased pain, touch and temperature sensation was observed. Sensory conduction velocity in sural and radial nerves was below normal in 6 of 10 individuals with neuropathic symptoms (Murai and Kuroiwa, 1971). Headache and peripheral nervous system symptoms were also reported in a poisoning episode which occurred in a Finnish paper company (Hakklnen et al., 1973). A decrease in amplitude of muscle action potential evoked by nerve stimulation, and a decrease in sciatic nerve conduction velocity has been reported in rats intoxicated with tetrachlorobiphenyl. Thus, PCB's can affect peripheral nerve function in both humans and experimental animals, but these have been reported only at doses which cause other systemic signs of poisoning. Cancer There is too little epidemiological evidence available yet to evaluate the potential of PCB's as human carcinogens (Bahn et al., 1980; Brown and Jones, 1980). A follow-up of the Yusho patients through 1977 has reported 51 deaths (31 with cause of death HONS 209793 27 confirmed) of the 1665 identified victims. There were 11 deaths from neoplasms, or 35.4% of the total While this rate is higher than the 21.1% in the population of the same prefecture in 1977, these data were not age-adjusted. No particular site was elevated, and there were no deaths from malignant melanoma, a tumor previously suspected to be linked to PCB exposure (Bahn et al., 1976). Two liver cancers and two lung cancers were reported but smoking and drinking patterns were not available (Urabe et al., 1979). A retrospective cohort mortality study of 2,567 workers in two capacitor manufacturing plants was recently completed by NIOSH. The reports, still in draft form, did not find any statistically significant excess mortality for any cause of death among the exposed workers. Deaths from liver cancer, cirrhosis of the liver, and rectal cancer were slightly higher than expected, but no information was available on medical histories, drinking patterns, etc. No correlation was observed between increased mortality and length of exposure, but the number of total deaths was small (163). NIOSH will continue to follow up the mortality experience of the cohort (Brown and Hones, 1980). Ongoing Occupational Studies Two additional cohort mortality studies are currently underway. The first is a mortality survey of the entire workforce employed between 1952-1957 at the largest U.S. facility that manufactured capacitors and transformers. There is detailed Information available on exposure levels in the plant. While the duration from onset of exposure Is shorter than optimal (only 25 years in some cases), the information will at least give data on the short-term mortality experience of a heavily exposed occupational group (Selikoff et al., in progress). HONS 209794 28 The second is a similar occupational mortality study, also of workers exposed in capacitor and. transformer manufacturing. Over 2,000 workers have been identified for this study but no further details are yet available (Bertazzi et al., in progress). One case control study is currently being conducted to assess whether there is excess risk of malignant melanoma among PCB-exposed workers. This data will not be available until March, 1981 (Bahn et al., 1976). There is also one cross-sectional clinical field survey of 326 capacitor manufacturing workers at two sites, encompassing a total workforce of 800 (Fischbein et al., 1979). Exposures were classified as none, low, medium, and high based on job description at the time of the survey (1973). Researchers were able to identify the PCB's used and had some data on environmental air levels in the plants. A number of parameters were measured, including complete history and physical exams, SMA panels and pulmonary function tests. Results have been published on respiratory function and general signs/symptoms, and results of serum lipids, endocrine function and dermatologic findings are forthcoming. To date, the only positive association involves dermatologic signs and symptoms. Further investigations of the effects of PCB exposure on serum lipids have been done in both occupational and general environmental exposure settings. Smith et al., (1978), reported some statistically significant differences between exposed and non-exposed workers at two sites. They reported higher serum triglycerides and lower levels of high density llpo-protelns in the exposed group. Whether the magnitude of the difference is biologically significant is not clear from this study. For example, the non-exposed group at site #1 compared to the non-exposed group at site #2 showed a greater difference than the exposed and non-exposed comparison at either site. In another HONS 209795 29 study (Baker et al., 1980) workers and community residents with exposure to fertilizer made from sewage sludge contaminated with PCB's were studied. Plasma triglyceride levels were found to increase significantly with serum PCB concentration (both in drinkers and non-drinkers), and the authors concluded that PCB's may alter lipid metabolism at levels of exposure and bio-accumulation insufficient to produce other identifiable signs of toxicity. Comments Although many problems have been identified In the studies evaluating the health effects of PCB's, it is dear that occupational exposure, at a minimum, can produce dermatologic effects. The long half-life of PCB's and their bio-accumulation in various human tissues leaves open the possibility of substantial chronic and delayed effects analogous to those seen in animal bioassays. These effects have only recently begun to be studied in a rigorous manner, and although the epidemiological evidence is neither complete nor entirely consistent there can be no question of the necessity to protect the worker from exposure. HONS 209796 30 VHL MEDICAL SURVEILLANCE AND BIOLOGIC MONITORING Medical surveillance and biologic monitoring are of limited usefulness in predicting health hazards if dose-response relationships are not known. This certainly is the case with PCB's. Based on animal toxicology, there are many suspected adverse effects of PCB's which might result from exposure in occupational settings, but very few have been documented well enough to give even rough estimates of "no-effect" or "safe'1 levels. A large percentage of norvoccupationally exposed people have detectable PCB levels in body fat, blood and milk. However, any attempt to estimate an adverse health effect associated with increases above this background level necessarily involves extrapolation from animal data and therefore is subject to considerable error, especially when the marked variation in sensitivity of various animal species is appreciated. Furthermore, not enough is known regarding the relative dose-response characteristics of the various documented effects (e.g., liver damage, skin changes) to state that in the absence of a particular sign, symptom or laboratory abnormality, the risk of long-term effect (cancer, reproductive toxicity) will be negligible (see Table IX For the clinician confronted with a worker who has a history of exposure to PCB's the approach to management cannot be easily outlined. Given the current analytic methodology, residues can be measured in blood or tissue in the ppb range and compared to background; but assigning a health risk to a given level is virtually impossible, especially given the lack of pharmacokinetic data. Often patients are being evaluated after a considerable lag period (years) since last exposure occurred and extrapolation to peak blood levels is not possible. In fact, it may be that residue levels bear little relationship to the health risk. For example, the lower chlorinated compounds may be more toxic but they are more rapidly metabolized and excreted and therefore less HONS 209797 31 likely to persist in blood or fat. Further, with the possible exception of chloracne, the presence of specific signs, symptoms or laboratory abnormalities is very difficult to definitely relate to PCB exposure in any given patient. Given these uncertainties and the potential for serious health effects, the approach to monitoring should emphasize environmental sampling and every attempt should be made to minimize exposure by engineering controls or personal protective measures in those settings where occupational exposure still occurs (e.g., utility repair workers). Biologic monitoring may be used to assess the effectiveness of environmental control, but it is really best utilized within a specific research protocol and probably has little value in the routine work-up of individual patients. HONS 209798 32 SUMMARY AND CONCLUSIONS PCB's have low acute toxicity but are of public health concern because of their persistence in the environment and in human tissues and their demonstrated potential for chronic or delayed toxicity. They are potent inhibitors of reproductive function in both rodents and non-human primates and are positive in animal cancer bioassays. As potent inducers of hepatic enzyme systems, PCB's may have additional unpredictable long-term health effects. Some of the conflicting reports In the toxicology literature are undoubtedly related to the variable composition and trace chemical contamination of the tested mixtures. Occupational and environmental exposure is usually to those mixtures; but if we are to accurately assess the associated health hazards, further animal studies are needed which carefully define the toxicology of the individual agents. Epidemiologic studies of occupational exposures to PCB's to date have failed to detect serious adverse effects but are considered insufficient, and further studies are clearly needed. Of particular interest is the continued exposure among utility workers. Because of the potential ability to cause cancer and other long-term adverse effects such as infertility and hepatic injury, human exposure to PCB's should be kept to the lowest level technically possible. The persistence of PCB's in the body and the irreversibility of some of Its effects make it necessary to act now, rather than to wait until more definitive data are available. HONS 209799 33 REFERENCES A.C.G.I.H. Chlorodiphenyl - 42% dilorine, Documentation of the TLV's for substances in Workroom Air. 3rd ed., pp51-2. Cincinnati, 1976. Albro, PW and Fishbein, L. Intestinal absorption of PCB's in Rats. Bull Environ Contam Toxicol 8*26, 1972. Allen, 3R. and Abrahamson, L3. Morphologic and biochemical changes in the liver of rats fed PCB's. Arch Environ Contam Toxicol ,1:265, 1973. Allen, 3R Response of the non-human primate to PCB exposure. Fed Proc 34: 1675, 1975. -- Allen, 3R and Barsotti, DA. The effects of transplacental and mammary movement of PCB's on infant rhesus monkeys. Toxicol 6:331, 1976. Allen, JR et at. Response of rats exposed to PCB's for 53 weeks. Arch Environ Toxicol 4:404, 1976. Allen, 3R et al. Reproductive effects of halogenated aromatic hydrocarbons on non-human primates. Ann NY AS 320:419, 1979, Anderson, MW et al. The construction of a pharmacokinetic model for the disposition of PCB's in the rat. Clin Pharm Therap 22: 765, 1977. Bahn, AK et al. Melanoma after exposure to PCB's (letter to the editor) N Engl 3 Med 295: 450, 1976. Baker, E et al. Metabolic consequences of exposure to polychlorinated biphenyls (PCB) in sewage sludge. Amer 3 Epidem 112: 553, 1980. Barsotti, DA et al. Reproductive dysfunction in rhesus monkeys exposed to low levels of PCB's (Aroclor 1248) Food Cosmet Toxicol ^4: 99, 1976. Bertazzi, PA et al. Mortality experience among PCB workers. Univ di Milano, Inst Occ Health, Milan, Italy, in progress. Bitman, ] and Ceal, HC. Estrogenic activity of DDT analogs and PCBs. 3 Ag Food Chem IZi 1108, 1970. Bowman, R et al. Correlation of PCB body burden with behavioral toxicology in monkeys. Pharmocol Biochem Behav 9: 49-56, 1978. Brackner, 3V et al. Biologic response of the rat to PCB's. Tox Appl Pharm 24; 434, 1973. Goldstein, JA et al. A comparative etudy of two PCB mixtures on induction of hepatic porphyria and drug metabolizing enzymee. Tox Appl Pham 32:461, 1975. HONS 209800 34 Brown, DP and Jones, M. Mortality and industrial hygiene study of workers exposed to polychlorinated biphenyls. Draft report, NIOSH Division of Surveillance, Hazard Evaluation and Field Studies 1980. Bumgarner, 3E et ai. Polychlorinated biphenyl residues in refuse workers. National Institute of Environmental Health Sciences (as reported in NIOSH Criteria Document), 1973. Burse, VW et al. PCB's-storage, distribution, excretion and recovery: Liver morphology after prolonged dietary ingestion. Arch Environ Health 29% 30l, 1974. Curley, A et al. PCBs - Distribution and Storage in body fluids and tissues of Sherman rats. Environ Res 9: 481 1971. Curley, A et al. PCB's evidence of transplacental passage in the Sherman rat. Food Cosm Toxicol Lh 471, 1973. DHEW: Subcommittee on health effects of PCB's and PBB's - series of articles appearing in Env Health Persp 146-198, 1978. Dougherty, RC et_al. Sperm density and toxic substances: A potential key to environmental health hazard, submitted to Env Health Chem, 1980. EPA: Halogenated polyaromatics, in: L Fishbein, Potential industrial carcinogens and mutagens, pub #360/3-77-003, Office of Toxic Substances EPA, Washington, DC pp 173-197, 1977. Finklea, 3 et al. PCB residues in human plasma expose a major urban pollution problem. Amer 3 Public Health 62: 643,1972. Fishbein, A et_al. Clinical findings among PCB-exposed capacitor manufacturing workers, Ann NY AS 320: 203, 1979, Fries, GF. PCB residues in milk of environmentally and experimentally contaminated cows. Env Health Persp. 1_: 33, April 1972. Funatsu, I et al. Polychlorobiphenyls (PCB) induced fetopathy. 1 Clinical observation (Abstract No, 72-2360), Kurume Med 3 J_9i 43-31. Gardner, AM et al. PCB's Hydroxylated urinary metabolites of 2,3,2',3'-tetrachlorobiphenyl identified in rabbits. Biochem Biophys Res Comm 55% 1377, 1973. Grant, DL et al. Metabolism of a PCB (Aroclor 1234) mixture in the rat. Bull Environ Contam Tox 6:102, 1971. *{Gold*tLn, JA et el* bottom p. 34) Hakkinen, 3 et al. Diphenyl poisoning in fruit paper production. Arch Environ Health 26: 70; 1973. Hara, I et al. Follow-up study of condenser factory after use of PCB discontinued. (As reported NIOSH Criteria Document) 3pn 3 Ind Health J_7* 371-372, 1973. Hasegawa, H et al. Report on survey of work area environment where PCB is handled and of the health of workers handling PCB, in Special research report on prevention of environmental pollution by PCB-like substances. (As reported in NIOSH Criteria Document) 3apan, Research Coordination Bureau, Science and Technology Agency, pp. 141-99, 1972. HONS 209801 33 Heddle, 3A and Bruce, WR. Comparison of tests for mutagenicity or carcinogenicity using assays for sperm abnormalities, formation of micronuclei and mutations in Salmonella, In, Origins of Human Cancer, Cold Spring Harbor Lab,, pp 1549, 1975. Higuchi, K ed. PCB poisoning and pollution. Academic Press, NY 1976. Hirotodzu, D and Ota, H. Transfer of PCBs to infants from their mothers. Arch Environ Health 35: (2) 95, 1980. Hsu, 1C et al. Metabol.ic fate .if 3H2,5,2',5,-tetrachlorobiphenyl in infant non-human primates. Bull Environ Contam Toxicol 14:233, 1975. IARC: Working Group on the evaluation of the carcinogenic risk of chemicals to humans - Polychlorinated Biphenyls, Vol 18, 1978. Inoue, Y et al. Discovery of PCB pollution in textile factory--I. PCB level in blood serum of laborers and results of physical examination (As reported in NIOSH Criteria Document) 3pn. 3 Public Health 22: 1637, 1973. Ito, N et al. Histopathological studies on liver tumorigenesis in rats treated with PCB's Gann 65T745, 1974. * 3ensen, 5 and Sundstrom, G. Structure and levels of most chlorobiphenyls in two technical PCB products and in human adipose tissue. Ambio 3: 70, 1974. Jerina, DM and Daly, 3W. Arene Oxides: A New Aspect of drug metabolism. Science 1 85: 573 1974. Karppanen, E and Kolho, L. The concentration of PCB in human blood and adipose tissue in three different research groups. In, PCB Conference 11, Stockholm 1972. (As reported in NIOSH Criteria Document) Solna, Sweden National Swedish Environment Protection/Publications, 4E, pp 124-28, 1973. Kataenellenbogen, 3 and Ames, BN. Personal Communication 1980. Keplinger, M et al. Toxicologic Studies with PCBs (abstract) Tox Appl Pharmocol T9: 402-403, 1971. Kihlstrom, 3E et al. Sexual function of mice neonatally exposed to DDT or PCB. Environ Phys Biochem 5: 54, 1975. Kikuchi, M et al. An autopsy case of stillborn of chlorobiphenyls poisoning. Fukuoka Acta Med 6(0*9, 1969. Kimbrough, RD et al. Morphologic changes in liver of rats fed PCB's, Arch Ind Health 25: 35<f, 1972. Kimbrough, RD. The toxicity of polychlorinated polycyclic compounds and related chemicals. Crit Rev Toxicol 2: 445, 1974. - Kimbrough, RD and Linder, RE. Induction of adenofibrosis and hepatomas of the liver in BALB/c3 mice by PCB's (Aroclor 1254). 3 Natl Cancer Inst 53; 547, 1974. Kimbrough, RD et al. Induction of liver tumors in Sherman strain female rats by PCB (Aroclor 1260). 3 Natl Cancer Inst 55: 1453, 1975. 36 HONS 209802 Kimbrough, R et al. Animal toxicology, in DHEW Subcommittee on Health Effects of PCB's and PBB's. Env Health Persp 24: 173, 1978. Kimbrough, R Chronic toxicity of halogenated biphenyls and related compounds in animals and health effects in humans. CDC report, 1980. Kimura, NT and Baba, T. Neoplastic changes in the rat liver induced by PCB. Gann 64:105, 1973. Kitamura, M et al. PCB in blood of workers employed in an electrical parts manufacturing plant. (As reported in NIOSH Criteria Document) Jpn J Ind Health 15: 539, 1973 (Jap). Kuratsune, M et al. Epidmeiologic study on Yusho, A poisoning caused by ingestion of rice oil contaminated with a commercial brand of PCB's. Environ Health Persp 1: 119, 1972. " Kutz, FW and Strassman, SC. Residues of PCB's in the general population of the US. Jn: Proceedings of the natural conference on PCB, Chicago, EPA - 560/6-75-004, Washington DC, pp 139. Linder, R et al. The effect of PCB's on rat reproduction. Food Cosmet Toxicol J2i 63, 1974. Lutz, RJ et al. Preliminary pharmacokinetic model for several chlorinated biphenyls in the rat. Drug Metab Dis 5: 386, 1977. Mathews, HB and Anderson, NW. Effect of chlorination on the distribution and excretion of PCB's. Drug Metab Dis 3; 371, 1975. Meigs, JW et al. Chloracne from and unusual exposure to Aroclor. JAMA 154: 1417, 1954. Miller, JW. Pathologic changes in animals exposed to a commercial chlorinated biphenyl. Public Health Rep 59: 1083, 1944. Moore, JA et al. Comparative toxicity of three halogenated dibenzofurans in guinea pigs, mice and rhesus monkeys. Ann NY AS 320: 151, 1979. Murai, Y and Yoshigoro, K. Peripheral neuropathy in chiorobiphenyl poisoning. Neurol 2h 1173, 1971. NCI: Carcinogenesis technical report series #38, DHEW publication #(NIH) 78-838, 1978. Nelson, N et aL PCB's - environmental impact. Environ Res 5; 249, 1972. NIOSH: Criteria for a recommended standard - Occupational exposure to PCB's NIOSH, Cincinnati, 1977. Nor back, DH and Weltman, R. Personal communication 1980. New York State Health Planning Commission. Report of the ad hoc committee on the health implications of PCB's in mothers' milk. Albany Health Advisory Council, 1977. Oishi, 5 et al. Comparative toxicity of PCB's and dibenzofurans in rats. Tox Appl Pharm 4JTT3^ 1978. 37 HONS 209803 Orberg, 3 and Kihlstrom, 3E. Effects of long-term feeding of PCB, Clophen A-60 on length of estrus cycle and frequency of implanted ova in the mouse. Environ Res 6: 176, 1973. " Ouw, HK et al. The use and health effects of Aroclor 1242, a polychlorinated biphenyl, in an electrical industry. Arch Environ Health _3l: 189, 1976. Sato, M and Hasegawa, H. Amount of PCB in blood of laborers. (As reported in NIOSH Criteria Document) 3pn 3 Ind Health _16s 363, 1974 (3pn), Safe, S et al. The metabolism of 4-chiorobiphenyl in the pig. Can 3 Phys Pharmacol 53: 392,-T9T5. Schwartz, L. Dermatitis from synthetic resins and waxes. Dermatitis 26: 586, 1936. Selikoff, 13 et al. Mortality experience of factory workers exposed to PCB's in the manufacture of transformers and capacitors. Mt. Sinai School of Med, Env Sciences Lab, New York, NY, in progress, 1980. Shiotajl Postnatal behavioral effects of prenatal treatment with PCB's in rats. (As reported in 1ARC, 1978). Okajimas Fol Anat 3pn 53t 103, 1976. Smith, AB et al. Lipid and Lipoprotein alteration: Occupational exposures to PCB. Clinical Res ^6: 549, 1978. Stalling, DL et al. An expanded approach to the study and measurement of PCB's and selected planar halogenated aromatic environmental pollutants. Ann NY AS 320: 48, 1979. Taki, I et al. Report on Yusho (chlorobiphenyls poisoning): Pregnant women and their fetuses. Fukuoka Acta Med 60: 471, 1969. Treon, 3F et al. The toxicity of the vapors of Aroclor 1242 and Aroclor 1254. Am Ind Hyg Q _1_7: 204, 1956. Urabe, H et al. Present state of Yusho patients. Ann NYA5 320: 273, 1979. Vodicnik, MJ and Lech, 33. The transfer of 2,4,5,2',4',5',-hexachlorobiphenyl to fetuses and nursing offspring. Tox Appl Pharm 54; 293, 1980. Van Miller, 3P et al. Distribution and metabolism of H-2,5,2*,5',-tetrachlorobiphenyl in rats. Proc Soc Exp Biol Med 148: 682, 1975. . Vos, 3G and Beems, RB. Dermal toxicity studies of technical PCB's and fractions thereof in rabbits. Tox Appl Pharm J_9s 617, 1971. Warshaw, R et al. Decrease in vital capacity in PCB-exposed workers in a capacitor manufacturing facility. Ann NY AS 320: 277, 1979. Wyndham, C et al. The in vitro metabolism, macromolecular binding and bacterial mutagenicity of 4-chlorobiphenyl, a model PCB substrate. Res Commun Chem Pathol Pharmacol Ih 563, 1976. Yoshimura, T. Epidemiological study on Yusho babies born to mothers who had consumed oil contaminated by PCB. Fukuoka Acta Med 65: 74, 1974. HONS 209804 38 TABLE I owt.mfONsi rot uiimi toiscoloct Jpmc L Ut *** Adult Ut Uaenllng hi Adult Mt VnUn| JUT A*ull Ut Adult Ut Adult Ut Adult nice Adult Hie* Honker Monkey Honker Monkey Adult Adult Adult "yaung" Adult Mi Guinea rift Adult Adult Ubblc Ut, beuat, Cuifico pig, end cal Htee Adult Adult H 1 HU* U( 3 vtiha 4 mha Kata Uta t vn|a 8 vafl# Agent Route Doe* kratlm Tatel Dose Iffact lafarancaa Arecter 1*54, i:40 Aroelor 1254, 1J60 Aroelor 1243, i:S4. t!&2 erector 1242, 1014 At-odor Or- 1 Orel 0r*t Orel oTl 1 4,^0 10,000 i,?oo i.soo 50 3.4-t.t too Aroelor 1740 0t*1 13.0 Areclor 1734 Orel 3.0 (imilir 500 Orel 1.0 'fCl'e1* Orel 1.0 Clophon aW 0r! Aroelor lJ48 Oral Aroelor (248 Orel Aroelor 1742 Orel 1.0 10 3.3 0.1.10 SlnftU tre*ur Slnale Upoaure 6 alcka A fwitho 7-13 d*y liitaitan 2 wntha prlor to hc!h| 2 writhe prior to ling Oer 8*14 r 13-21 lietium 4 week* iUft(8| at birth "throfllt" 4 veela 1 omOi 4 aanthe ^50 ........... `'so 2100 700.1100 L|v*r Ky|p*rtnyhy, fettr liciiuitiiA Utty Llvr 700 1500 Survival - offspring no trret*. 3llttar alt* fiatTlnl at(ape In, 300 feartallty * effaprlng 4'netlftft behavior * ef feprlng 4.0 4 T*rlue learning aaaiyi effaprlng too Unknown *0 100 27-2700 4ln*1*ntattone, 1 liter Bl-offiprlng; change* in eetroue cycle-offspring lengthened tetrov# cycle# i^l*ipl*Atatlei* weight loaa, alopecia, facial ape dlecharga liver abnormal!tie* (hlet*l*i,lejl) death, gastric ulceration, facial ede*at thymic atrophy Kimbrough et it 1478 Kimbrough el it 147ft All** A AbrehaMoft, 19 ut** t *t,, 1174 1474 Unden at *l.a 1474 Linden et ei.,. 1174 Jhlata, 1376 Klhlatroa at *1 1173 Oberg 4 Uhlme, 1473 Allan, 1175 Allen, 1475 Allan, 1373 Aroelot 1748 Orel O.Od.0.17 1 year 24-42 facial eden, eloped* acne, ifartilhr. !rr*ftu(ar vna#e, pearly abortion*- offspring: pwelftht, heed tlTCumferenee, (dataatabla ftl'a In tlaau,) Allen, 1173 IricUfi ArecXpr* Aroelor 1240 Aroelor LISA N Kanaklw 300 Dinal Dtmt attOO *70 0*r**l SnHel, " Orel =10 1.3 * 0.7 * 30 23 daya Single Cepe*ore 38 day* 17 onoke 31 make 32 vila 1300 -- 740 *30 117 11.Id* light A'a* 11 ear hlctalagy death - delayed up to 21 day# with liver atrophy akinp liver and kidney danage nlcraeaoplc liver A*a ravoraibl# U*ar f hapotaeal luler carcinoma* {contra] 0/4: doaad 3/22) preopleatie nodule* (control 0/e; doetd 7/12) Hiller* 144A Hiller, 1444 Voa t fteewo r 1471 Traon, 1134 Triefi, 1434 1 to at el-* 1473 Aroelor 12SA Orel Araalar 11W Oral 30 10 Araelof 1240 Oral 1,*.!.!.*.3hezecMerobl* phenyl Orel 10 10 44 wall 16.1 ** 21 oofltha t.l t+ 10* ."k* 7.1 ** 104 vitbi 1.1** PneopSeetlc nodulae (central 0/24; dated 1/22) Kt*reufth & linden, 19/4 4hepatocellular carcineTM*# (control 1/i7J; doaed 146/184) dnaeplaetlc nodolra (control 0/17); do*d 24/184) Kimbrough et * 1173 fhepotoce 1 luler clrllMHi (unpubllahod. 1480) Norbeth at el. 1980 4heptoceI lular iirilnam (unpubllihvd, 1140) Horbech et el* 1980 * AaimIi iri tipoiwl to 1*4 at 1*5 of 1154 for 7 hmri/<l*ri 5 dr/^rk for IT #uk*, j To d-*M In ih falitwtnii wre Applied' 0*] (wtluna |[ air VramhaJ m by enlnol per Hiylj 0*3 (weight of mlMt U bn); 24/7 (correction fur 24 hour/doy iipeir)t V) (eoireetlon for 7 day/vcah aapeaura), laputtlnf vilu* aiiuii*a 10tf& abeorptlon* , 'Wa. iehn. 39 HONS 209805 Study Exposm Laval & Tima Meigs 1934 - Out break ot Otirm-, litis in chemicaj plant 0.1 mg/M* Arochlor - 5 to 19 moths inter mittent expo sure through vapor laaksga Hesagewa et al, 197? Study ol i Industrial plants includ ing PCB manu- Iractura, cap acitor, msnu- lacture, and biphenyl recovery Vapory 13-965 yg/M Particulates 4-4M (.270 In a spill) ' <1 to 20 years Hera al si 1973-197* Level ot exposure not reported In NIOSH Criteria Document Kitaroura 1*7* Study of workers medical earns to a taiin jtor manu'-iCtur.ntj p!xnl Exposure level not reported in NIOSH Criteria Document Time 2.3 ye&rs TABLE n - OCCUPATIONAL EXPOSURE TO PC8s Study Population Exposed Controls , Dermal Effort* Liver Function Findings Blood Concentrations 1* 0 7/14 mild to 6 normal, 1 moderate borderline Chloracnet (in chlorscne cases) Cancer/ Mortal tty r 99 32 ' Various Slightly Exposed* 370 abnormal . Ppm (alevated liver non-exposeek enzymes) 20 ppb 11B (study concentr ated on 17 immsrssion pro cess workers) *3% black, heads, 37% acne, 13% Irritation not reported Exposed 7-300 ppb 13 Various: Normal 920 ppb acne, seb- avenge (320- - orrhea 2100 ppb) adiposa, folliculitis Comments Since blood concentra tions ol POBs could not be measured, no cor relation ol individual dose and skin effects was possible. , Dermal arlmenu were unrelated lo blood concentrations. . Based on 3 plants. Ihfre was no rel ationship of expo sure to blood con cent ration; fat metabolism was apparently affected Blood concentrations closely related to years of exposure; Follow-up study alter exposure ceased allowed calculation at serum half life; The longer the duration oT exposure, the longer the PCB half-life. (range 3-30 months) No relationship was found between concentration in blood and duration of expo sure. HONS 209606 OCCUPATIONAL EXPOSURE TO PCBi stud/ Exposure Study Population Level A Time , Expaaed Controls Dermal Effects Liver function Findings Blood Concentrations Cancer/ Mortality Comments Ouw et al 1974 Study of capacitor manufacturer Arochlor 1242 , 1.00-1.44 mg/MJ (19 fillers! .32 mg/MJ (IS assemblers) Workers wore no protective clothing Time 1 month to 23 years 34 JO Mild burninqp ifritation of face, eyes and skin; .5 had rashes, 1 chloracne. ; several dermatitis Bromsul phothalium tests elevated 4 of 7 fillers with blood levels > 500 ppb Exposed 100602 ppb (mean 400 ppb) Not detected in non-exposed Systemic effects reported such as nausee and persistent body odor. There was no adverse response at blood concentrations below 200 ppb Bahn, at al 1974 Study of worker* In a refinery Arochlor 1254 over a 9 year period 31 researchers end development; 41 refinery workers ' ' 2 malignant These were pre- . melanomas li min ary results observed, reported in a letter .04 expected to the editor. (based on Workers were also TNCS data) exposed to other chemical*. Study is in progress* Brown and Jones I960 cohort morlality study of capacitor manufacturers! {draft raport) Plant 01; 1.24 pg/NC- 3.93 ug/M ; , Plant 02: LJO yg/MJ 1240 ug/M Plant #1-940 Plant #2-1599 1 1 1 All cause mortality was lower than expected (141 obs. vs 174 exp) All cancer mortality was lower than expacted (39 obs. vs 40.4 exp.) Rectal and liver cancer were slightly elevated but not significantly. Lower observed mortality may be attributable to the "health worker" effect. NIOSH wilt continue to followup mortality expert ence MOWS 209807 Study none el si 975 Study of a amily iilk thread llosung jperatian iata and Hass)awa 1974 Pressure senitlive carboneu paper nanufacturer . <arppanen and <olho 1972 Study of ) groups: 1) no exposute I) analytical lab J) capacitor impregnal ion Bumgarner et al 1973 Study of refuse workers exposed to PCB in incinsration of waste occupational exposure to PC8* Exposure Study Population Level &t Time ' Exposed Controls - Dermal Effects Liver Function Not reported in NIOSH Criteria Document 0.15-1.2 u/M^ 0.13-4,4 pg/MJ Number studied not reported in NIOSH document (Later ' 54 more * persons ware "j studied) Mild skin Unknown skin lesions includ- ing comedones measured 2 yrs, past PCB use Not reported in NIOSH Criteria Document High exposure (capacitor manu facturing); low exposure (ana lytical Lab) High - 12 Low - 6 4 males 5 females Incinerated waste 37 36 lumber yard workers ' ' . Findings Blood Concentrations 130-520 ppb Cancer/ Mortality Comments Goad correlation between degree of exposure and blood PCS levels. Exposed 75 ppb nan-exposed 20 ppb ' Unexposed 5.6-12 ppb Medium exposure 36-63 ppb High exposure 74-1,900 ppb Control; max. 4.2 ppb Exposed: max. 14 ppb (4-14 ppb) . Unable to detect any "Biologic Eflects"i type at monitoring not reported in NIOSH review Concentrations not well correlated with duration of exposure, age or race HONS 209808 Study Baker, Land. rig an el al 1900 Study of. exposrua to PCBi In sewage lodge ' OCCUPATIONAL EXPOSURE TO PCS* Exposure Level Tima Study Population Expoaad Controls ` Dermal Effects Liquid sewage entering plant 30-470 ppb. Up stream sewage 1290-5200 ppb (Aroclor 1014) Concentrations in sludges (Aroclor 1242) were as high as 1700 ppm (mean 479.1 ppb) and 107.3 ppm in treated soil (mean 17.1 ppm) 09 sludge users, 10 workers exposed to PCBs. 19 member* worker* families 22 com munity members 4 s. Acne, increased pigmenta tion in 4 workers ; Liver Function Findings Blood Concentrations Cancer/ Mortality Comments Sludge users 17.4 ppb, workers 75.1 ppb Families 33.6 ppb, community 24.2 ppb Plasma triglyceride levels increased significantly with serum PCS con centrations. Data indicate that PCQs may alter lipjd metabolism. MONS 209809 TABLE ni PERCENT DISTRIBUTION OF SYMPTOMS-OF YUSHO REPORTED BY 189 PATIENTS EXAMINED BEFORE OCTOBER 31, 196*. Symptoms Dark brown pigmentation of nails Distinctive hair follicles increased sweating at palms Acnelike skin eruptions Red plaques on limbs Itching Pigmentation of skin Swelling of limbs .- Stiffened soles in feet and palms o{ hands Pigmented mucous membrane Increased eye discharge Hyperemia of conjunctiva Transient visual disturbance Jaundice Swelling of upper eyelids Feeling of weakness Numbness in limbs Fever Hearing difficulties Spasm of limbs Headache Vomiting Diarrhea -- - .. _ Males (N-S9) Females (N- 100) 83.1 64.0 50.6 87.6 20.2 42.7 75.3 20.2 24.7 X>.2 88.8 70.8 56.2 11.2 71.9 58.4 32.6 16.9 18.0 7.9 30.3 23.6 19.1 75.0 56.0 55.0 82.0 16.0 52.0 72.0 41.0 ----29.0 47.0 83.0 71.0 55.0 11.0 74.0 52.0 39.0 19.0 19.0 8.0 39.0 28.0 17.0 Sources Kuratsune et al, 1972 HONS 209810 FICURE I 32 6 5' 5 2' 3' BIPHENYL HOLECULE AND RING NUMBERING SYSTEM 3-chloroMpt.B.Tl 2.2 \ 3.4' .J-ptt.chlorc,blph.n7l EXAMPLES OP- HOMEHCUOTK STSTEM OP CKLOROSIPHEim. COMPOUNDS NUMBER OF ISOMERS AND PERCENT CHLORINE FOR THE 10 CHLOROBIPHENYL (PCB) CLASSES Chlorobiphenyl Empirical Formula No. of Isomers Weight % Cl mono di tri - tetra penta hexa hepta octa nona deca C iH,CI C 1HIC1 2 C iHtC1 3 C 2 HfCI * C 2HSC1 5 C6 c 2H1CI 7 c 2HzC) 2 c iHCl 9 c Cl 10 Source: NIOSH, 1977 3 12.79 12 31.77 24 41.30 42 42.56 46 54.30 42 52.93 24 62.77 12 65.92 3 62.73 I 71.12 HONS 209811 FIGURE H i FIGURE III CHLORODJBENZOFURAN TYPES AND CONCENTRATIONS (ug g) IN COMMERCIAL PCB PREPARATIONS Ch Lorodibenzolu rans Mixture* di tri tetra penta hexa hepta Total CL) 1016 (1) 1016 (1) 1248 (1) 1234 (1) 1234 (1) 1260 Cl) 1260 (2) A-60 ' C3) DP-6 (4) K300 (4) K400 (4) K500 (4) K600 0.5 (c)*** (e) <0.0001 0.3 0.1 0.2 0.1 0.2 1.4 0.7 (a) (e) (a) <0.0001 1.2 0.2 0.4 0.4 0.3 3.0 10.0 (a) (c) (a) (a) <0.0001 0.3 1.4 0.9 0.5 0.3 2.2 2.9 (c) (b) 0.5 2.0 1.7 1.3 1.0 0.8 8.4 13.6 1-1.5 17-L 8 (a) 2,5-4 Cb) 3-3 *(1) Aroclor, (2) Clophen, (3) phenoclor, (4) Kanechlor **(a), (b), (c), Cd), (e) represent relative amounts in increasing order Source NI"SH, 1977 ' 46 HONS 209812 FIGURE IV Responses of primates and rats to PCB*s* Response Man Monkey Rat Susceptibility to toxicity Acne Hyperpigmentation of skin Alopecia Hyperactive Meibomian glands Conjunctivitis Oedema of eyelids Subcutaneous oedema Keratin cysts in hair follicles Hyperplasia of hair follicle epithelium Gastric hyperplasia Thymic atrophy Hepatic hypertrophy Liver enzyme change Decreased no. of red-blood cells Decreased haemoglobin Serum hyperlipidaemia Leucocy tosis High Yes Yes NA Yes Yes . Yes Yes Yes Yes NA NA Yes NA Yes Yes Yes Yes High Yes Only infants Yes Yes Yes Yes Yes Yes Yes Yes NA Yes Yes Yes Yes Hypolipidaemia Yes Moderate No No No No No No No No No No' " Yes Yes Yes No No Yes No Source: 1ARC, 197* * This table summarizes acute and subacute clinical effects but does not Include chronic or delayed effects such as reproductive effects or cancer. HONS 209813