|
|
Public Health Assessment
Ciba-Geigy Corporation
(Dover Township) Toms River, New Jersey
|
Public Health Implications
The public health implications of completed exposure pathways in the past will now be considered. Several contaminants were confirmed to be present in water from private wells and community water supply wells at levels above health-based Comparison Values.
(See Appendix for definitions and uses of Comparison Values.) The NJDHSS and the ATSDR have further evaluated the public health significance of past exposures to these contaminants through an examination of relevant toxicologic and epiderniologic
information. In addition, this section will include a brief summary of the findings of an analysis of childhood cancer incidence data for Dover Township.
Childhood Cancer Incidence in Dover Township
The NJDHSS and the ATSDR reviewed cancer incidence data in the period 1979 to 1995 for Dover Township as part of the Public Health Response Plan. Findings are described in a separate Public Health Consultation by the NJDHSS and the ATSDR (NJDHSS, 1997d).
Dover Township was the only municipality in Ocean County in which overall childhood cancer incidence (ages up to 19 years) was statistically elevated. Ninety cases were observed in the 17-year period, compared to 67 that would have been expected if childhood cancer rates were the same in the Township as in the entire State of New Jersey. Leukemia incidence was elevated in Dover Township, particularly in females under the age of five years. In the Toms River section of the township, overall childhood cancer was elevated (24 observed vs. 14 expected). Both leukemia and brain/central nervous system cancers were elevated, with the excess occurring primarily in female children under age five.
Toxicologic and Epidemiologic Evaluation for Adults and Children
Community Water Supply Well Pathway
The nature, magnitude and duration of exposure to CGC site-related contaminants through the distribution of water to the community from the Holly Street well field is not fully known. As noted before, diazotizable amines (measured as aniline) and nitrobenzene were measured in certain wells in 1965 and 1966, but concentrations in the distribution system have not been documented. Contaminant concentrations measured at the well are not necessarily accurate representations of levels to which individual households or sections of Dover Township may have been exposed.
The full nature of the contamination is not known. Dye manufacturing operations at the Toms River Plant involved the use and production of many classes of chemicals. Anthraquinone dye production involved the use of anthracene, aniline, benzene, nitrobenzenes, chlorobenzenes, chlorotoluenes, acids and metals, among others. Azo dye production involved the use of naphthalene, introbenzenes, aniline, phenol, benzidine, naphthylamine, o-toluidine, solvents and acids, among others. The following discussion describes the known toxicological characteristics of certain dyes and intermediates that are associated with Toms River Plant operations, and the results of epidemiologic follow-up of workers formerly employed at the facility.
Azo Dyes
Azo dyes are a family of synthetic chemicals of varying structure and complexity. Dyes may be metabolized into benzidine or other aromatic amines in the body. Studies of experimental animals exposed to certain azo dyes and to benzidine demonstrate the potential for carcinogenicity. Target organs for carcinogenicity appear to be the liver, kidney and bladder. Epidemiologic studies of workers exposed occupationally to benzidine and azo dyes have found increased risks of bladder cancer (ATSDR, 1997 a).
Benzidine is classified by the International Agency for Research on Cancer as a human carcinogen (class A), while certain benzidine-based dyes have been classified as probable human carcinogens (ATSDR, 1997a). No specific information is available regarding risks to children from exposure to these chemicals.
Aniline
The target organs of aniline toxicity in humans are reported to be the spleen and hematopoietic (blood) system. Methemoglobinemia may result from high levels of exposure. Animal studies (rats) have shown decreased mean pup weights and fetal hypoxia as a result of matemal methemoglobinemia from high doses (ATSDR, 1997a).
The USEPA has concluded that aniline is a probable human carcinogen (class B2) based upon the induction of spleen and body cavity tumors in rats. No adequate studies were located by the ATSDR regarding cancer in humans after exposure to aniline. One epidemiologic study (Ward, 1996) investigated the excess occurrence of bladder tumors in workers 'in the chemical dye industry, but results were confounded by exposure to other agents (o-toluidine and 4-wninobiphenyl). Based on the animal toxicology data, the ATSDR estimates that one additional case of cancer would be expected in a population of 100,000 persons from a lifetime of exposure to 60 µg/l of aniline in drinking water (ATSDR 1997a). However, because of uncertainty regarding several exposure factors, an individual's theoretical increased risk of cancer could be higher or lower.
Nitrobenzene
Little is known of the toxicologic effects of oral exposure to nitrobenzene. The primary systemic effect resulting from oral exposure to nitrobenzene is methemoglobin formation (resulting in decreased tissue oxygenation and a decrease in overall metabolism) (ATSDR, 1990). No other information is available regarding hepatic, renal, inununologic, developmental, reproductive, or genotoxic effects associated with oral exposure to nitrobenzene. The ATSDR reports no toxicological studies regarding carcinogenesis in animals or humans resulting from oral exposure to nitrobenzene.
Other Dye Intermediates
Literature searches by USEPA contractors of several chemical and toxicological databases located toxicity information on a number of the TICs in surface and sub-surface soils, and toxicity profiles have been prepared for Ciba-Geigy on 14 of these surface soil TICs (CDM, 1994b; CDM, 1995). In general, toxicological information on these species is sparse.
Toms River Plant Epidemiologic Study
A study of mortality of former Toms River Plant workers found excess deaths from bladder, kidney, and central nervous system cancers in those workers who had previously been employed by the Cincinnati Chemical Works (CCW), which had produced and used the known bladder carcinogens benzidine and ß-naphthylamine (Delzell ct al., 1989). Increased mortality from lung cancer was also observed among former maintenance workers, and stomach and central nervous system cancers among azo dye production workers. A recent update of this study (Sathiakumar and Delzell, 1998), with additional years of follow-up of former workers, generally confirmed the earlier findings.
Private Well Pathway: Cardinal Drive/Oak Ridge Parkway Area
A completed human exposure pathway to CGC site-related VOCs (benzene, chloroform, trichloroethylene, and tetrachloroethylene) likely existed through the use of private wells for irrigation and possibly potable purposes in the past, in the Cardinal Drive/Oak Ridge Parkway area. Actual exposure levels to VOCs would be dependent on the specific uses of contaminated water. The following is a discussion of toxicologic and epidemiologic studies of exposure to these VOCS, with emphasis on weight-of-evidence considerations related to childhood cancer.
Chloroform
Chloroform occurred at concentrations in the groundwater significantly above the respective ATSDR cancer risk evaluation guide (CREG) value of 6 µg/l. The liver is the primary target of chloroform toxicity in humans, a well as laboratory animals, who have been exposed orally, as indicated by blood biochemistry tests of liver enzymes. The kidneys are also major targets of chloroform toxicity in humans, as indicated by increased blood urea nitrogen and creatinine levels. Epidemiological studies have suggested an association between bladder and possibly colorectal cancers in humans and consumption of chlorinated drinking water from surface sources (which may contain chloroform and other trihalomethanes at concentrations ranging from tens to hundreds of micrograms per liter), but the results are not conclusive (Cantor, 1997). Chloroform
is the predominant trihalomethane found in chlorinated drinking water.
Although the data on human subjects are equivocal, chloroform has been classified as a probable human carcinogen on the basis of experimental animal studies (ATSDR, 1998). Mice that are exposed to high doses of chloroform can develop liver cancer, but the cancer appears to be secondary to chloroform-induced tissue damage and repair. Based upon a large number of genotoxicity assays, neither chloroform nor its metabolites appear to have a direct mutagenic activity, implying that low dose exposures to chloroform that do not cause direct liver toxicity, would not produce a carcinogenic effect.
Benzene
Occupational exposure to benzene and benzene-containing mixtures can result in damage to the blood-forming system (ATSDR, 1997d). Several studies of rubber workers have shown an increased risk of acute myelogenous leukemia and possibly other cancers. Experimental animal studies also indicate that high-level benzene exposure can lead to the development of multiple tumor types. It is not known what effects exposure to benzene might have on the developing human fetus, Studies with pregnant animals show that breathing benzene has harmful effects on the developing fetus. These effects include low birth weight, delayed bone formation and bone marrow damage (ATSDR, 1997d). Although some of the wells in the Cardinal Drive/Oak Ridge Parkway area contained benzene well above the MCL of 1 µg/l, the exposure levels in the occupational epidemiologic studies were much higher than those that could have been experienced by residents.
Trichloroethylene and Tetrachloroethylene
The effects of exposure to trichloroethylene (TCE) and tetrachloroethylene (perchloroethylene, or PCE) have been evaluated in scientific studies for their possible impact upon human health. TCE and PCE are classified as probable human carcinogens by the International Agency for Research on Cancer (LARC, 1995) based on the weight of evidence from laboratory animal experiments and limited human epidemiologic studies. Laboratory animals have been exposed to these chemicals via contaminated air, drinking water, and food. The results of these studies indicate that the nervous system and liver, and to a lesser degree the kidney and heart, arc the primary organs of adult animals affected by these VOCs (ATSDR, 1997f-, ATSDR, 1997g). Following long-term, high level exposure, TCE has been shown to produce liver cancer in mice and kidney and testicular tumors in rats (ATSDR, 1997f-, LARC, 1995). Chronic, high level PCE exposure produces liver cancer in mice and kidney tumors and mononuclear cell leukemia in rats (ATSDR, 1997g; LARC, 1995). The exposure levels needed to cause these adverse impacts in laboratory animals are many times higher than exposure levels that could have occurred through the use of contaminated drinking water (ATSDR, 1997f; ATSDR, 1997g).
Extrapolating animal toxicity data to predict human risk is often controversial and is especially so in the case of TCE, since some of the mechanisms implicated in its effects on animals do not exist in humans. For instance, TCE-induced peroxisome proliferation, a potential precursor to liver cancer, is common in rodents but not in humans. In addition, kidney tumors seen in male rats following exposure to TCE and other chlorinated VOCs are believed to be caused by the accumulation of alpha-2F-globulin in intracellular lysosomes. This histopathological alteration does not occur in humans, so the relevance of this pathological mechanism to human carcinogenesis is uncertain. Because of these significant interspecies differences in the metabolism and toxicity of TCE, the use of animal data to predict cancer risks in humans is problematic.
Epidemiological studies of occupationally-exposed workers suggest an association between long-term inhalation exposure to high levels of TCE and increased risk of liver and biliary tract cancer and non-Hodgkin's lymphoma (IARC, 1995; ATSDR, 1997f. Increased risk of esophageal cancer, cervical cancer, and non-Hodgkin's lymphoma have been observed in workers exposed to high levels of PCE (IARC, 1995; ATSDR, 1997g). Participants in the ATSDR TCE Exposure Sub registry (approximately 4,300 individuals with exposure to TCE in drinking at levels ranging from 2 to 19,000 µg/l for as long as 18 years) have reported a variety of health problems at rates above national averages (ATSDR, 1993b). However, only the rate for strokes was reported to increase with increasing concentration of TCE in drinking water. Results from the Sub registry have not documented any increased occurrence of cancer in the study population (ATSDR, 1993b).
Children may be particularly susceptible to the toxic effects of chemicals; fetuses may also be sensitive to toxic effects if the chemicals can cross the placental barrier. Recent epidemiologic studies suggest that fetal exposure to VOCs in drinking water could result in adverse health effects. The NJDHSS evaluated the effects of VOCs in drinking water on birth outcomes in an area of northern New Jersey (Bove et al., 1995). This exploratory study found that maternal residence during pregnancy in areas with TCE- contaminated drinking water was associated with an increased risk of birth defects of the neural tube and oral cleft. Exposure to PCE during pregnancy was associated with an increased risk of oral cleft defects. The authors concluded that their study by itself cannot determine whether the drinking water contaminants caused the reported adverse birth outcomes. A recent ATSDR study of exposure to VOCs in drinking water and occurrence of adverse
pregnancy outcomes was conducted at the U.S. Marine Corps Base at Camp LeJeune, North Carolina (ATSDR, 1997e). Decreased mean birth weight and increased small for gestational age babies were reported for two potentially susceptible subgroups: infants of mothers older than 35 years of age and infants of mothers with histories of fetal death. This study provides limited evidence for a causal relationship between exposure to VOCs and the reproductive and developmental effects evaluated.
A study of childhood leukemia conducted in Woburn, Massachusetts, concluded that the incidence of childhood leukemia was associated with the mother's potential for exposure to water from specific wells contaminated with TCE and PCE, particularly for exposure during pregnancy (MDPH, 1997). The study did not find any association between the development of childhood leukemia and the child's exposure to contaminated water after birth. The Woburn study should be interpreted with caution, however, since small numbers of study subjects led to imprecise estimates of risk. A study by the NJDHSS found a statistically elevated rate of childhood leukemia in towns served by community water supplies contaminated with TCE and PCE in the years 1979 to 1987 (before current drinking water regulations had been implemented), compared to towns without a history of such contamination (Cohn et al., 1994). Overall, the associations drawn from these limited epidemiological
data in humans are suggestive, yet inconclusive, that exposure to these VOCs through drinking water may cause birth defects or childhood leukemia in children exposed while a fetus. ATSDR and others are conducting or sponsoring research to clarify this possible relationship.
Private Well Pathways in Other Areas (Not CGC-Site-related)
Lead and mercury were detected in several private residential wells as described above, although these contaminants are unlikely to be CGC site-related. Lead is toxic to the nervous system, particularly in the fetus and young children whose nervous systems are undergoing rapid development (ATSDR, 1997b). To protect against the neurotoxic effects of lead, the USEPA has promulgated an Action Level of 15 µg/l for lead in drinking water for community water systems. If lead concentrations in well water were sustained above the action level throughout the day, they could pose a public health hazard, particularly to a developing fetus or a young child. Carcinogenicity studies of rodents exposed to high levels of lead are equivocal. The available data on the carcinogenicity of lead following ingestion by laboratory animals indicate that lead acetate and lead phosphate are carcinogenic, and that the most common tumor response is renal tumors.
However, the animal studies are limited in their usefulness since they involved small group sizes (ATSDR, 1997b). As a result, it is not possible to extrapolate from the carcinogenic effects of high level acute oral exposure in rats to low level chronic exposure in humans (ATSDR, 1997b).
The health effects of mercury are variable and dependent upon its form: metallic mercury, inorganic mercuric compounds, or organic mercury. The target organs for mercury toxicity are the kidney and nervous system. The ATSDR Toxicological Profile for Mercury presented no studies regarding carcinogenesis in humans with respect to oral exposure to inorganic or organic mercury (ATSDR, 1997c). Animal research studies have shown a significant increase of renal tumors in rats exposed orally to organic mercury (methyl mercuric chloride or phenylmercuric acetate). Oral exposure to mercuric chloride has also shown an increase in renal tumors in rats.
|
 BACK
||
CONTENTS
||
NEXT
||
|
|
|
