Search  |  Index  |  Home  |  Glossary  |  Contact Us

Congressional Testimony
Public Health Implications of Dioxins

Testimony by
Barry L. Johnson, Ph.D.
Assistant Surgeon General
Vice Chairman
Public Health Service
Committee to Coordinate Environmental Health and Related Programs

Before the
Subcommittee on Human Resources and Intergovernmental Relations
Committee on Government Operations
House of Representatives

June 10, 1992

I am Barry Johnson, Ph.D., the Vice Chairman of the Public Health Service's Committee to Coordinate Environmental Health and Related Programs (CCEHRP). The Committee is chaired by the Assistant Secretary for Health, Dr. James O. Mason, and its membership consists of 8 of the 9 agencies of the Public Health Service. I am accompanied today by Dr. Marilyn Fingerhut of the National Institute for Occupational Safety and Health (NIOSH), Centers for Disease Control; Dr. George Lucier, National Institute of Environmental Health Sciences, National Institutes of Health; Dr. Robert Scheuplein, Center for Food Safety and Applied Nutrition, Food and Drug Administration; and Dr. Melvin Stratmeyer, Center for Devices and Radiologic Health, Food and Drug Administration. We welcome this opportunity to appear before you today to discuss the public health implications of dioxins and the efforts of the Public Health Service to measure exposure to dioxins, evaluate mechanisms of toxicity, and determine effects on human health.

The National Institutes of Health is organizing a consensus conference on dioxins to be held later this year under the auspices of the Public Health Service (PHS) Committee to Coordinate Environmental Health and Related Programs (CCEHRP). That committee has broad representation of PHS agency personnel involved in environmental health. The Environmental Protection Agency will participate in the conference. Scientists from the United States and other countries with expertise in dioxins will be invited to attend. One outcome of that conference could be a recommended position on dioxins. Such a recommendation would subsequently be considered by the CCEHRP, and we hope to be able to make a more definitive statement about dioxins subsequent to that conference. It is important to note at the outset of this testimony that the toxicity to laboratory animals and human health effects of dioxin remain the subject of active scientific investigation.

BACKGROUND

Dioxins is a generic name used to describe a family of compounds known as chlorinated dibenzo-p-dioxins. There are a total of 75 chemical congeners in the dioxin family. A dioxin molecule can have as few as one or as many as eight chlorine atoms attached to the dioxin molecule at any of the eight locations. The number of chlorine atoms and their position on the molecule determines the physical and chemical properties and the toxicity. The most notable, most studied, and most toxic chemical in this family is 2,3,7,8-tetrachlorodibenzo-p-dioxin, or 2,3,7,8-TCDD. For the purpose of this presentation, I will refer to this chemical as TCDD.

TCDD was first discovered as a by-product of chlorinated phenols in the 1950s. Studies with laboratory animals have shown TCDD to be extremely toxic and the most potent carcinogen ever tested under laboratory conditions for some species of animals. However, the effects in humans exposed to TCDD have been more difficult to pin down. Because of this, animal studies have been used as the basis of most risk assessments for dioxins. It is currently believed that all dioxins may act by a similar mechanism of action and that the number and location of the chlorine atoms on the dioxin molecule determines the relative potency of each dioxin. The potency is, in part, determined by the ability of the dioxin to bind to specific receptors in the body. For example, TCDD has the greatest ability to bind to these receptors.

The process by which dioxins exert their toxicity is complex and of prime importance in understanding and evaluating toxicity. The most recent proposed model for TCDD involves several steps. In this model: TCDD first must bind to a protein receptor in the liquid interior (cytosol) of the cell, a second cytosolic protein then activates the receptor/TCDD complex, the activated complex is then transported to the cell nucleus. Once in the nucleus, several activated complexes bind to specific DNA sequences, which initiates a cascade of potential events. This process and/or other receptor-mediated processes are believed to be responsible for the myriad of toxic effects caused by TCDD exposure.

Dioxins usually co-occur in the environment with other chemicals such as chlorinated dibenzofurans (furans). Furans contain 135 congeners, which are structurally similar to dioxins and which elicit a number of similar toxic and biochemical responses in animals. Dioxins, furans, and other closely related compounds are thought to act by a similar receptor-mediated mechanism. In general, dioxins and furans are highly persistent compounds with a strong affinity for soil and sediment and a high potential for accumulation in biological tissues. They are ubiquitous, and have been found in all media: air, water, soil, sediments, animals, and food.

Although the focus of this discussion will be on the dioxins, it should be recognized that nearly all human exposures involve highly complex mixtures, and some involve mixtures consisting of structurally related compounds such as dioxins and furans. Because of the similarities between dioxins and furans, the Environmental Protection Agency (EPA) and many international environmental organizations have adopted Toxicity Equivalency Factors (TEFs) in assessing risk associated with exposure to material contaminated with dioxins and furans. This approach is based on a comparison of the toxicity of specific dioxins and furans to that of the most toxic dioxin, TCDD. TCDD is given a TEF of 1. The toxicity ratings of all other dioxins and furans are ascribed on a relative scale. For example, dioxins and furans with 5 chlorine atoms (penta) of which 4 occupy the 2,3,7, and 8 positions are slightly less biologically active and are given a TEF value of 0.5. The dioxin and furan isomers with 8 chlorine atoms (octa) are found in laboratory studies to be 1000 times less toxic than TCDD and therefore are assigned a TEF of 0.001. This weighted approach is an important step in the process of assessing the risk of exposure to dioxin and furan contaminated environmental media.

SOURCES OF DIOXINS

Dioxins do not occur naturally nor are they intentionally manufactured by industry, except in small amounts for research purposes. Dioxins occur as a contaminant in the manufacturing process of certain chlorinated organic intermediates and products, such as chlorinated phenols. TCDD is a by-product formed during the manufacture of 2,4,5-trichlorophenol. 2,4,5-Trichlorophenol was used in the manufacture of the bactericide, hexachlorophene, and the herbicide, 2,4,5-trichlorophenoxyacetic acid (2,4,5-T). Various formulations of 2,4,5-T have been used extensively for weed control on roads, crops, rangelands, and roadways throughout the world. Herbicide formulations of 2,4-dichlorophenoxyacetic acid (2,4-D) and 2,4,5-T were used for defoliation and crop destruction by the American military in the Vietnam War. In most industrialized countries the use of products contaminated with dioxins has been significantly reduced. 2,4,5-T can no longer be used legally in the United States and pentachlorophenol (a wood preservative) and hexachlorophene have been restricted in their uses. Dioxins have been released to the environment during the manufacturing, use, and disposal of these chemicals. Dioxins are also formed during the chlorine bleaching process used by pulp and paper mills and enter the environment in waste water effluent from these plants. A source of dioxin production in the past has been the burning of leaded gasoline.

There are numerous other sources that contribute to dioxins in the environment. Dioxins are known to form concurrently with furans during combustion processes such as: incineration of municipal solid waste and industrial waste, and are associated with ash generated in the incineration process. Emissions from these sources vary greatly and depend on management practices and the applied technologies. Combustion of many chlorine-containing materials (such as plastic material like polyvinyl chloride, paper, wood treated with pentachlorophenols, pesticide-treated waste, and PCBs) can produce dioxins and furans. Dioxins and furans have also been detected in emissions from coal-fired power plants, home-heating systems, exhaust from cars running on leaded gasoline, and cigarette smoke. Regulation of many of the sources of dioxins has been successful in reducing the amount of dioxins entering the ecosystem and in decreasing the potential for human exposure.

DETECTING DIOXINS

The instrumentation and methods used to detect dioxins are extremely precise and sensitive. Because of this, scientists are able to detect dioxins at concentrations in the parts per trillion (ppt) or parts per quadrillion (ppq) range for most environmental media and biological material (including blood, adipose tissue, and breast milk). The National Center for Environmental Health and Injury Control (NCEHIC) of the Centers for Disease Control (CDC) has been in the forefront in developing and improving methods for the detection of dioxins in biological samples. The analysis of dose-response relationships based on serum, adipose, or milk TCDD levels is a critical step in assessing the relationship between dioxin exposure and health outcome.

GENERAL POPULATION EXPOSURE

The general population is exposed to small amounts of dioxins, as exemplified by the fact that dioxins have been found in virtually all samples of adipose tissue and blood (serum lipids) from individuals with no known previous exposure. It is primarily the dioxins with chlorine atoms in the 2, 3, 7 and 8 positions that are retained in animals and humans and which selectively concentrate in body fat and lipids. The more highly chlorinated dioxins, such as octachlorinated dioxin (which has 8 chlorine atoms), are generally found at higher concentrations than the lower chlorinated dioxins. The dioxin contents of serum and adipose tissue is similar when expressed on a lipid basis, and both can be used to indicate body burden of dioxins. Serum lipid levels of TCDD from the general population typically range from 3 to 7 parts per trillion (ppt) and rarely exceed 20 ppt.

The average time it takes to remove one half of the TCDD from the body is around 7 years. The half-lives of other dioxins in the body are not known. About 98 % of the daily intake of dioxins for the general population comes from ingesting food and milk. Inhalation exposure to dioxins for the general population constitutes a minor portion of daily intake. Average intake of TCDD for adults has been calculated to be about 25 picograms (pg) per day or 0.35 pg per kilogram (kg) of body weight per day. If all dioxins and furans are included and TEFs are used, the total average daily intake of TCDD equivalents for adults is about 90 pg/day or 1.3 pg/kg body wt/day.

In a recent study, investigators reported that the body burden of TCDD in the general population appears to be decreasing. Mean background levels of TCDD in human adipose tissue collected in 1970 were 2 times higher than the more recent samples of adipose tissue. Although several factors may be responsible for this, it is worthwhile to note that similar results have been reported for other countries.

Two well documented incidents resulting in TCDD contamination took place in Missouri and Seveso, Italy. The Missouri incident, which occurred in the early 1970s, resulted from the spraying of waste oils and chemical waste from the manufacture of 2,4,5- trichlorophenol on roads and in horse arenas for dust control. Although spraying and disposal of the waste occurred in many areas of the state, the most noted contaminated areas included a horse arena, Quail Run trailer park, and Times Beach. In 1986 CDC carried out an adipose tissue study of 51 persons who lived or played in areas with TCDD-contaminated soil and 128 persons who did not. For the unexposed population the median adipose TCDD level was 6.1 ppt (range ND-20.2 ppt), whereas a median of 24 ppt (range 3.7-750 ppt) was reported for the exposed population. Twenty-two of the persons previously classified as exposed had adipose tissue TCDD levels of 20.2 ppt or less.

The most extensive known residential contamination of TCDD occurred in Seveso, Italy, when a 2,4,5-trichlorophenol reactor exploded in 1976. The contaminated area was divided into three zones based on the concentration of TCDD in the soil. The 211 families in Zone A, the most heavily contaminated area, were evacuated within 20 days of the explosion and measures were taken to minimize exposure to residents in nearby zones. In a preliminary study, CDC evaluated serum samples from 5 residents of Zone A who had chloracne, 4 from Zone A without chloracne, and 3 from outside the contaminated area. All samples had been collected and stored shortly after the accident. TCDD was detected in only one sample from the unexposed group at 137 ppt. The high level was thought to be due to misclassification or sample contamination. In Zone A, serum TCDD levels ranged from 1772 to 10,439 ppt for persons without chloracne and from 828 to 56,000 ppt for persons with chloracne. The TCDD levels detected are the highest ever reported in humans. CDC is in the process of evaluating several hundred historical blood samples taken from Seveso residents for TCDD. There are plans to determine half-life estimates and to evaluate serum TCDD levels for participants in the Seveso cancer registry.

Background levels of TCDD and TCDD equivalents in breast milk are around 3 ppt and 39 ppt (lipid basis), respectively. In general, infants who are breast fed are exposed to higher levels of dioxins on a body weight basis than adults. Average daily uptake of breast-fed infants is 20 pg/kg body wt/day for TCDD and 180 pg/kg body wt/day for TCDD equivalents. The World Health Organization has concluded that this risk to infants does not outweigh the positive biological and psychological aspects of breast feeding.

HEALTH EFFECTS AND CANCER IN ANIMALS

The health effects of dioxins have been extensively studied in animals. In these studies, the toxicity of specific congeners has been found to vary widely and to depend on the number and position of the chlorine atoms on the molecule. TCDD, and to a lesser extent, dioxins with five or six chlorine atoms substituted in the 2, 3, 7, and 8 positions are extremely toxic to animals. The focus of this discussion will be on TCDD.

A considerable range of sensitivity to TCDD exists among species. Guinea pigs have been shown to be highly susceptible to the lethal effects of 2,3,7,8-TCDD. The hamster is 3000 to 6000 times less sensitive. The signs of toxicity also vary considerably from species to species. Most animals exhibit a wasting-type syndrome characterized by progressive and profound loss in body weight. In addition to differences among species, there are considerable differences among strains of the same species. The differences in tissue-specific responses and species and strain susceptibility to TCDD may, in part, be due to the nature of the TCDD receptor.

Exposure to dioxins by any route is known to cause various systemic effects in exposed animals. The most characteristic effect of TCDD exposure in animals is the wasting syndrome. The mechanism of the wasting syndrome is not well understood, but some studies indicated that it may be related to an effect of TCDD on the thyroid gland. The liver was one of the target organs of TCDD-induced toxicity in several species. Reported effects included increased activities in liver enzymes indicative of pathological changes, changes in liver weight, and necrosis. Although chloracne is a characteristic effect of humans exposed to dioxins, it is not that typical in animals. However, hair loss, thickening of the skin, and a development of acne-like lesions were reported in some studies. The most severe systemic effects were found in monkeys.

Studies in animals suggest that the immune system may be the earliest and most sensitive target of toxic effects caused by dioxins exposure. Organ changes include thymic and lymph node atrophy, and/or degenerative changes in bone marrow of treated animals. In addition, functional alterations in the immune response affecting both humoral and cell-mediated immunity were reported in numerous studies. Recent studies indicated that TCDDinduced suppression of humoral immunity can be several times higher following subchronic exposure versus acute exposure to the same cumulative dose. This finding may be important for humans who are likely to have prolonged exposure to very low levels of dioxins.

There have been 17 studies designed to determine if TCDD is a carcinogen in experimental animals. All of these studies demonstrate that TCDD administration leads to increased tumor occurrence in multiple sites. It is a carcinogen in both sexes and in several species, including the hamster, and in tissues remote from the site of treatment. TCDD administration increases cancer at doses well below the maximum tolerated dose. There is increasing agreement that the mechanism whereby TCDD causes cancer is not genotoxicity. For example, TCDD does not interact with DNA in such a way as to produce cellular mutations. Instead, it appears to promote cancer by triggering cells to divide, most likely those cells which already contain a mutation (caused by other chemicals, u.v. radiation, etc.). This process gives the genetically altered cells a selective growth advantage over the normal cells, leading to an increased risk of tumor development. Similar effects on cell division and differentiation may be involved in the reproductive and developmental effects of TCDD. Much of what we know about the effects of TCDD comes from experimental animal and cell culture data.

Experimental studies provide strong evidence that dioxins may cause developmental effects in animals. Oral exposure to TCDD during pregnancy caused high incidences of cleft palate and skeletal anomalies in the offspring of exposed dams. Other frequently seen anomalies included kidney malformations and thymic atrophy. Chronic exposure of monkeys to extremely low doses of TCDD caused decreased survival in the offspring and changes in their social behavior and learning ability.

Reproductive studies demonstrate that oral exposure to TCDD causes preand/or postimplantation losses of fertilized eggs in rodents. Exposure of monkeys to TCDD during pregnancy caused spontaneous abortions of the fetuses. It has been proposed that dioxins block the estrous cycle by antagonizing the estrogen-induced uterine response to the egg. Other plausible mechanisms for adverse reproductive effects include effects on growth factor pathways.

HUMAN HEALTH EFFECTS

Early human health studies in the 1970s and 1980s had many shortcomings, such as the small size of the group studied, coexposure to other chemicals, inadequate follow-up time, and inability to document TCDD exposure. Those shortcomings made it difficult for scientists to establish a clear association between TCDD and human health outcomes. The early studies on TCDD-exposed populations had to rely on indirect indices of exposure, such as: work history, duration of exposure, location relative to the contaminated media, and frequency of contact with contaminated media in defining the exposed population or classifying groups within the population. During the 1980s, a major breakthrough was the development of methods to measure TCDD in adipose and human serum. This has enabled subsequent studies to estimate participant exposure to dioxins more accurately.

A well known health effect observed in human populations exposed to relatively large amounts of TCDD is chloracne. Chloracne is a severe skin disease characterized by follicular hyperkeratosis (comedones) occurring with or without cysts and pustules. Chloracne has been reported in some workers involved in the production of 2,4,5-trichlorophenol and/or subsequent products. No evidence of current or past chloracne was found in Vietnam veterans involved in Operation Ranch Hand. In 1976, chloracne was reported in some Seveso children after the trichlorophenol reactor explosion. No cases of chloracne have been reported in Missouri residents of Times Beach or Quail Run trailer park.

A recent study on the health status of Vietnam veterans who participated in Operation Ranch Hand did not find any signs of liver disease, but did report increased levels of triglycerides and cholesterol in the blood (a second report does not support these increases). In addition, an increase in body fat, diabetes, and blood pressure were also noted. These effects were strongly associated with TCDD levels in the serum. Ranch Hand veterans also had changes in blood (increased white blood cells, platelet, IgA, and sedimentation rates) which suggest a chronic inflammatory response, although no immunologic system diseases were identified. These immune system changes were also strongly associated with serum TCDD levels. These results differ from those reported in previous analyses of the Ranch Hand group in 1982 and 1985. The earlier analyses did not include an assessment of serum TCDD levels. A physical examination of Ranch Hand veterans is currently under way.

STUDIES OF CANCER IN WORKERS EXPOSED TO DIOXINS

The National Institute for Occupational Safety and Health (NIOSH) of CDC is conducting health studies of U.S. chemical workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and higher chlorinated dioxins.

TCDD Mortality Study

In 1991, the New England Journal of Medicine published results of a NIOSH study of cancer mortality among workers exposed to TCDD. This study analyzed the mortality experience of all U.S. workers assigned between 1942 and 1984 to the production of chemicals contaminated with TCDD. The cohort included 5172 male workers at 12 U.S. plants.

Mortality from all cancers combined was slightly but significantly elevated in the overall cohort (SMR 115, 95% CI 102-130) and was higher in a subcohort of 1520 workers with over one year of exposure and at least 20 years of latency (SMR 146, 95% CI 121- 176). In this subcohort, mortality was also significantly increased from soft tissue sarcoma (3 deaths, SMR 922, 95% CI 190- 2695) and from cancers of the respiratory system (SMR 142, 95% CI 103-192). Excess mortality from all cancers combined, cancers of the respiratory tract, and soft tissue sarcoma, especially in the subcohort with more than one year of exposure, supports the classification of TCDD as a carcinogen. Mortality was not significantly elevated in the overall cohort from several cancers previously associated with TCDD (stomach, liver and nasal cancers, Hodgkin's disease, non-Hodgkin's lymphoma, and soft tissue sarcoma).

NIOSH is currently working on a second analysis of the cancer mortality experience of this cohort using a more precise measure of TCDD exposure.

Two additional recent studies support the NIOSH study finding that TCDD may cause cancer in humans at very high exposure levels. These studies also suggest that TCDD may induce a variety of cancers rather than a single type. (This is consistent with the animal carcinogenicity data.) One study of German chemical workers found a two-fold excess of total cancer in 114 workers who had chloracne and who had been exposed to TCDD following an accident 20 years earlier. A second study of German chemical workers found a 24% excess of total cancer in a group of about 1500 workers. The cancer excesses in these two studies were accounted for by cancers of various sites.

NIOSH STUDIES OF OTHER HEALTH EFFECTS

NIOSH researchers conducted a cross-sectional medical study of 281 workers from two of the 12 plants included in the dioxin mortality study and 260 unexposed referents. The study was designed to evaluate many of the non-fatal potential health effects related to occupational exposure to TCDD. The potential health effects examined include dermatologic, neurologic, hepatic, endocrine, gastrointestinal, immunologic, reproductive, and other effects previously reported to be associated with exposure to TCDD.

The overall health status of all study subjects was assessed by medical and psychological examination and interviewer-administered occupational, demographic, and medical history questionnaires. Blood samples were collected from each participant for analysis of clinical chemistries and serum TCDD levels.

Several reports have been generated from this study. No association was found between TCDD level and liver disease, lung disease, peripheral neuropathy, or depression. Significant differences were observed for two hormones. Workers with high serum dioxin levels (over 75 ppt) were two to three times more likely to have abnormally low testosterone and/or abnormally high luteinizing hormone. These endocrine effects are consistent with research showing that dioxin has the ability to disrupt the reproductive hormones of animals. Among the measures of liver function, a statistically significant elevation of GGT was found with increasing serum TCDD levels, but only among heavy drinkers. The clinical importance of this finding is uncertain. Additional analyses of this group of workers are under way, including analyses of reproductive outcomes, diabetes, chloracne, and other effects.

Pentachlorophenol Production Workers Mortality Study

This study will evaluate whether occupational exposure to pentachlorophenol (which is contaminated with predominantly higher chlorinated dioxins than TCDD) is associated with increased mortality. The study population includes about 2100 workers involved in the production of the wood preservative, pentachlorophenol (PCP). NIOSH has collected personnel, medical, and industrial hygiene records and ascertained vital status for the cohort. NIOSH is preparing these data for analyses.

Serum Levels of Dioxin and Furans in Currently Exposed Workers

The goal of this project is to assess the level of current exposure to dioxin among workers. The groups of interest include paper and pulp workers, pentachlorophenol producers and users, workers producing brominated fire retardants, and municipal incinerator workers. NIOSH and the National Center for Environmental Health and Injury Control of CDC are presently collaborating with the Environmental Protection Agency to assess exposure in a group of paper and pulp workers.

NIEHS RESEARCH

Researchers at the National Institute of Environmental Health Sciences (NIEHS) have been actively working for several years on some of the issues relevant to dioxin risk assessments. They have addressed the issue of dose-response relationships by designing experiments to determine if dioxin's effects exhibit a threshold or whether there is a proportional relationship between exposure and response in the low-dose region of the dose-response curve. These studies are examining biochemical and biological responses in female rats exposed chronically to TCDD over a wide dose range (100 pg/kg/day to 125 ng/kg/day). Female rats were selected because nearly all dioxin risk assessments are based on tumor incidence in female rat liver. NIEHS's results showed that for a number of biochemical responses (cytochrome P-450 induction and effects on growth factor receptors) there was no deviation from linearity over this dose range and that the best fit to the data was inconsistent with the presence of a threshold. Analysis of data from singledose experiments, using an even wider dose range, gave similar results. NIEHS's single-dose studies included an exposure level which produced tissue concentrations of dioxin similar to those seen in humans as a consequence of dioxin exposure encountered in day-to-day living (1-3 pg/kg/day). However, some biological responses, such as TCDD-mediated increases in cell proliferation, appear to be less sensitive and NIEHS data cannot rule out a threshold for those effects. Evaluation of cell proliferation data is complicated by the high amount of interindividual variation in response. These findings are consistent with the idea that the shape of the dose-response curve cannot be predicted solely on the basis that a response is receptor mediated. It is known from basic endocrinology that two responses, both utilizing the same receptor system, may exhibit qualitatively and quantitatively different dose-response relationships. Some receptor-mediated responses may exhibit threshold behavior whereas others may not. The mechanisms responsible for diversity in receptor-mediated responses is an active area of research. It is now important to gain an understanding of the mechanistic link between TCDD-mediated changes in gene expression and toxic effects.

The observation that TCDD acts like a potent and persistent hormone is consistent with the findings that it produces a wide variety of effects. Of particular concern is the finding that children of women exposed to rice oil contaminated with high levels of PCBs and polychlorinated dibenzofurans (structural analogs of TCDD) exhibited impaired brain development. These children had lower IQs, delayed memory development, and increased hyperactivity. It is not known whether these effects were caused by the PCBs, polychlorinated dibenzofurans, or a combination of both. There is also no information on dose-response relationships for these effects.

The NIEHS laboratory, as well as several other non-NIH laboratories, has generated data on the relevance to humans of animal models for TCDD's effects. These studies have shown that humans have a fully functional aryl hydrocarbon (Ah) receptor. Also, TCDD and its structural analogs produce many of the same changes in gene expression in humans as seen in experimental animals, and humans seem to be at least as sensitive as animals to these effects. For example, placentas of the same Taiwanese women exposed to rice oil contaminated with structural analogs of dioxin exhibited markedly elevated levels of cytochrome P-450, and growth factor pathways were also altered in these women. Also, human and animal cells in culture respond to TCDD in a similar way. Recent epidemiological studies conducted by NIOSH and European scientists have shown an association between increased cancer risks and dioxin exposure. The sites of human cancer were, in general, consistent with tumor data from experimental animals. When this information is taken together, it seems reasonable to assume that TCDD is a human carcinogen at high exposure levels and that animal models are appropriate to use along with human data for estimating human risks. However, there is legitimate scientific debate concerning the shape of the dose-response curve for the carcinogenic effects of TCDD as well as other toxic effects such as developmental toxicity, reproductive toxicity, endocrine toxicity, and neurotoxicity. It is possible that these effects require relatively high doses of TCDD.

It is becoming apparent that there is considerable interindividual variation for certain of TCDD's effects on humans. For example, some individuals accidentally exposed to a given level of dioxin develop chloracne whereas other individuals exposed to the same amount of dioxin do not. Moreover, some individuals are much more sensitive to the enzyme induction effects of dioxins than are others. In collaboration with the National Cancer Institute and Italian scientists, NIEHS researchers are attempting to determine the mechanism responsible for sensitivity or resistance. Preliminary findings suggest a genetic basis but considerably more work needs to be done. It is important to determine if there are sensitive subpopulations for each or all of the toxic effects of dioxin, as there are for chloracne.

REGULATORY ACTIVITIES AT THE FOOD AND DRUG ADMINISTRATION

FDA is concerned about dioxin primarily because food (including the potential contamination of food from food contact articles like coffee filters, milk cartons, and paper plates) accounts for a large majority of human exposure to dioxin.

Several FDA centers have been involved with the dioxin issue, as some of their regulated products were thought to have the potential to contribute to dioxin exposure. However, the most important exposure to dioxin for the general population is now believed to occur through food ingestion and through articles that may come in contact with food, and FDA's Center for Food Safety and Applied Nutrition is the lead Center on the risk assessment of dioxin.

The average human daily intake of dioxin from all food-related sources is between 1-3 pg/kg TEQ or 0.2-0.6 pg/kg TCDD. FDA currently uses 0.057 pg/kg/day as the intake of dioxin that will result in at most a one in a million life-time risk of cancer (1 x 10-6). This equates to a carcinogenic risk of 17 x 10-6 for an intake of 1-3 pg/kg/day of dioxin equivalents from food sources; our hypothetical "worst case" cancer risk based on these calculations can be estimated at 17-51 x 10-6, or 17-51 cancer cases per million.

FDA believes these risk estimates are conservative. Regulators in Europe and other western countries also believe the risks from typical dietary exposures are small and have established their Tolerable Daily Intakes (TDIs) for TCDD in the 1-10 pg/kg/day range. These are approximately 16-160 times less conservative than ours. However, since it is not known for certain whether there is any risk at all from these small levels, or the extent of risk, other agencies outside the U.S. and FDA believe that reasonable efforts should be taken to reduce the dioxin contamination of the environment to as low a level as feasible in order to achieve an eventual reduction of the human body burden and a lessening of any potential human health risks.

FDA has established a policy to reduce avoidable dioxin exposures in regulated products when they appear to contribute significantly to the ubiquitous dioxin background. FDA was concerned two years ago when it was found that bleached, kraft paper used in milk cartons contained TCDD and TCDF. The average amounts leaching into milk were only 0.02 ppt TCDD and 0.15 ppt TCDF, or 0.28 pg TEQ/kg/day, but these constituted approximately 10-30 % of the daily dietary background intake, and the added risk exceeded one in a million. Today, with the cooperation of the paper industry, these levels have been reduced. The industry has supplied us with data that shows that added dioxin levels in milk are now indistinguishable from background levels. We are presently conducting monitoring studies to evaluate the industry's claims. We expect these studies to be completed this summer.

FDA'S VIEW OF THE HEALTH RISK OF DIOXIN

Average levels of dioxin TEQ exposure are between 1-3 pg/kg/day. We have not, at this time, directly identified the actual cancer risk, if any, posed by this small amount of dioxin. Essentially this is because the experimental studies upon which the risks have to be based (in the absence of any reliable human data) need to be conducted at high dose levels in order to overcome the insensitivity of animal bioassays. A typical animal study for dioxin (Kociba, 1978) used doses of 1000, 10,000, and 100,000 pg/kg/day; and tumors were definitively obtained only at the highest dose. The necessary extrapolation from the high test doses to the much lower human use levels requires a knowledge of dioxin's mechanism of action that we do not have.

This leaves the field wide open to speculation and to different interpretations of the limited mechanistic data that we do have.

One school of thought emphasizes the fact that TCDD is not genotoxic, does not bind itself directly to DNA, shows strong promotional rather than initiator activity, appears to cause liver toxicity prior to liver cancer induction, and apparently acts through a receptor mediated mechanism--and concludes from this that linear risk assessment is not appropriate. This group includes most of the world's regulatory agencies (except the federal agencies in the United States and some state agencies), and they believe a threshold-safety factor approach is appropriate. These regulators have established TDIs for TCDD in the range of 1-10 pg/kg/day.

These same data also lead to another interpretation, which is held by the U.S. federal agencies. It appears true that dioxin is not genotoxic and does not bind DNA directly. But, it is now widely acknowledged in the scientific community that dioxin binds to a specific intracellular protein (the Ah receptor) which, after some further modification, and interaction with other proteins does bind to a specific part of DNA. Through this mechanism the dioxin receptor is known to regulate the transcription of genes encoding metabolizing enzymes. A substantial body of evidence, both genetic and biochemical, indicates that the Ah receptor mediates the biological effects of TCDD. The observation that dioxin action is receptor dependent implies there may be a threshold below which no detectable biological effects occur, but we presently do not know what this concentration is. Recent scientific studies have observed dioxin-induced enzyme induction in animals at levels comparable to typical human dietary exposures. Enzyme induction is not cancer and the significance of exceedingly low levels of enzyme induction to human risk is not clear. But these studies suggest that cancer may possibly occur at lower concentrations than implied by the detectable organ responses in the limited number of animals used in rodent bioassays. However, cancer is a far more complex process than enzyme induction and requires multiple, coordinated steps which may be expected to increase the concentrations required. In addition, both environmental and genetic factors influence TCDD-induced neoplasia in animals and, by analogy, we may envision highly inter-individual responses in humans.

The receptor dependence of dioxin action does indeed impose some constraints upon the assumptions used in the risk assessment process. However, these constraints do not necessarily mean that one is fully justified in using the lowest experimentally observed level with a safety factor to obtain the safe dose. This alternative assessment may seem more justifiable biologically than conventional linear risk extrapolation because it recognizes that dioxin is not a typical genotoxic compound, but it may underestimate the actual risk because dioxin is not a conventional toxicant either. There may be a threshold for dioxin action, but any particular threshold procedure will only become convincing as the mechanisms of dioxin action and its dose response become understood in more detail.

It should be understood that despite the apparent differences in the way risk assessments of dioxin are viewed by different countries, the regulatory consequences are similar. While the Europeans have regulatory levels in the 1-10 pg TEQ/kg/day range while ours are between 0.057-0.006 (FDA and EPA, respectively), we are both taking similar steps to reduce dioxin exposures. For example, WHO has recommended reduction of dioxin release into the environment to the extent possible consistent with sound engineering practices. Its targeted list includes virtually all known dioxin sources: incinerators, the metal industry, motor vehicles, sewage sludge, the pulp and paper industries, pentachlorophenol, and flame retardants, among other sources and contamination pathways.

These actions to reduce exposure are justified on the basis that while their tolerable dioxin level (i.e., 1-10 pg TCDD/kg/day) is usually not exceeded, dioxin is neither useful nor necessary and is clearly toxic at some level, and therefore all feasible steps should be taken to minimize unnecessary human exposure. We arrive at the same practical steps to reduce dioxin exposures despite disagreement over the regulatory weight to be given to the state of the science.

CCEHRP INTERIM STATEMENT ON DIOXINS

An interim statement on the health risk of chlorinated dibenzodioxins and chlorinated dibenzofurans expressed as 2,3,7,8-TCDD equivalents has been adopted by the CCEHRP. The following statement is considered interim until on-going risk assessment and risk characterization reviews have been completed by the EPA and the CCEHRP:

"Dioxin (i.e., 2,3,7,8-TCDD) and structurally-related compounds have been of concern to health officials and the public for several years as compounds with potential to harm human health. This concern was originally based on results from laboratory animal studies. More recent evidence from epidemiologic studies of workers occupationally exposed to dioxin, and from toxicologic studies of the mechanism of toxicity of dioxin, has raised questions about dioxin's degree of toxicity to human populations.

"Both the U.S. Public Health Service (PHS) and the U.S. Environmental Protection Agency (EPA) are currently reviewing the data on risks of dioxin for human health. Until these reviews have been completed, the following interim statement on the health risk of dioxin and related compounds represents the views of the PHS Committee to Coordinate Environmental Health and Related Programs (CCEHRP).

"The appropriate risk assessment methodology for 2,3,7,8-TCDD and the other dioxins and furans is complex, and consensus across the CCEHRP agencies is not sought at this time for several reasons: (1) Several important and relevant epidemiologic and toxicologic studies have not been completed, (2) the issues require consideration and expertise by scientists in addition to those who are immediately available to the CCEHRP, and (3) the bulk of the disagreement in the risk assessment methodology appears to exist between the U.S. health agencies and the health agencies of other countries, and its resolution requires international participation. "Because FDA must act on the regulatory issues concerning dioxins and furans confronting it, an interim risk assessment is necessary. The CCEHRP recommends that the current FDA risk-specific dose for 2,3,7,8-TCDD (0.057 picograms per kilogram body weight per day [pg/kg/day] for a one in a million risk) be adopted as the risk of the calculated '2,3,7,8-TCDD equivalents.' This is a conservative estimate and represents an upper bound estimate. The real risk, when compared to the estimated risk from this dose, may be lower. For example, CCEHRP acknowledges the existence of the World Health Organization's recommendation that the acceptable daily intake of dioxin not exceed 10 pg/kg/day (the current EPA value is 0.006 pg/kg/day). However, FDA's risk-specific dose is appropriate for regulatory purposes while the issue is still under debate and the data are under evaluation. "Because 2,3,7,8-TCDD is known to be toxic to humans and laboratory animals at sufficiently high doses, CCEHRP agencies will continue with research to elucidate the substance's toxicity at lower doses to human populations and to participate in risk assessment forums that provide an estimate of harm to humans posed by contact with dioxin and related compounds. Until deliberative processes of risk assessment and scientific review have been completed, it is premature to make definitive statements regarding the health risk of dioxin."

CONCLUSION

Mr. Chairman, in summary, the agencies of the Public Health Service remain concerned about the human health implications of dioxins and related compounds. Over the years, PHS agencies have been centrally involved in assessing the health risk of dioxin, developing methods to measure human exposure, conducting studies of workers and community residents exposed to dioxins, and performing laboratory studies of the toxicity of dioxin. Although much has been learned, there remain some critical gaps in our knowledge that are currently the subject of research. These new scientific data are being used by EPA and PHS agencies in assessing the risk of dioxin and by the PHS as it proceeds, with active EPA participation, with a consensus conference on the risk characterization of dioxin. As stated at the beginning of this testimony, we look forward to considering the outcome of the CCEHRP consensus conference later this year. It is critically important that scientists knowledgeable of the animal and human data come together and develop consensus on interpretation of that data.

That concludes my testimony. My colleagues and I shall be glad to respond to any questions you and other members of the subcommittee have.

ATSDR Home  |  Search  |  Index  |  Glossary  |  Contact Us
About ATSDR  |  News Archive  |  ToxFAQs  |  Public Health Assessments
Privacy Policy  |  External Links Disclaimer  |  Accessibility
U.S. Department of Health and Human Services