Click for next page ( 41


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 40
Evidence has long been available that certain polluted occupational or ambient environments can cause or contrib- ute to both acute and chronic degenerative diseases. More recently, as knowledge of the importance of environmental contaminants in health and disease has increased, it has also become apparent that exposures to tcxic agents in the environment may produce other effects, such as cancer, mutations, birth defects, and behavioral abnormalities (McCullough 1975). Government agencies charged *ith protecting health and the environment from effects cf toxic substances face a substantial challenge in identifying the risks involved in any given case. If the purpose of taking action is to prevent harm, decisions must te based on presumptive evidence, before disease has been become overt. The kinds cf evidence needed to establish a basis for action include: • identification of causative agents in the environ- ment; • demonstration of adverse effects, such as morbidity cr mortality, in exposed populations; • production of toxic effects ir laboratory animals; and, • reduction or eliminatior of harm when exposure is reduced or ceases, or when the population is protected. Such sequential chains of evidence have been established in scme cases of occupational exposures (for example, to beta naphthylamine and nickel [International Agency for Fesearch on Cancer 1971-1975]). The same general kinds of evidence are needed to evaluate the effects of contaminants in the ambient environment. - 40 -

OCR for page 40
Several research approaches are needed to establish or refute causal relationships between environmental pollutants and effects on health, Epidemiological studies can show patterns of disease or death that, when correlated with pat- terns of exposure to environmental ccntaminants and other factcrs, suggest possible causal relationships. Toxicolog- ical research can demonstrate qualitative similarities between effects observed in humars and effects in labora- tory animals. Additional important information can be gained from investigations of the biochemical and physio- logical mechanisms of action of environmental agents and cf the functional and structural changes a chemical or metabolite may produce in tissues and organs. No one research approach can, by itself, provide adequate infor- mation on the potential health effects cf a chemical in the environment; congruent results are needed from both human experience and laboratory studies. Ideally, cause- effect relationships suggested by epidemiological associa- tions could be confirmed by both toxicc.logical testing and the elucidation of metabolic mechanisms. In each specific case, research needs will vary according to the nature cf the effects involved, the kinds cf data already available, and the potential success of particular investigatory approaches. EPICEMICICGY Epidemiology is the study of the determinants cf the incidence and distribution of disease in populations. Observation .of associations between patterns of disease and patterns of exposure to environmental contaminants can be a powerful tool for generating hypotheses about cause-effect relationships. When supported by clinical cr experimental findings of similar effects, epidemiolog- ical studies can provide the most convincing evidence that an environmental agent has had an impact on the health of a population. There are, however, limits tc what can be learned from epidemiology. Spurious correlations may be found. The quality of available data is often poor; data on mortality are not always recorded in consistent forms, and there are gaps and inconsistencies in data on morbidity. Information on exposures to potential causative agents is generally inadequate. Few communities monitor the air or water for more than a small number of substances, at a few locations. Information en what was in the environment 20 or 30 years ago (the latent period for some diseases) is often impos- sible to obtain; furthermore, pecple mcve from place to place. There is no way to know exactly what mixture of contaminants and other factors (such as cigarette smoking

OCR for page 40
and diet) a person or a population encountered before its disease status was observed. Nevertheless, there is much rich grcund to be broker in environmental epidemiology, A great deal of informa- tion exists on the incidence of various diseases; some of it is in useful form, (such as the cancer atlas compiled by the National Cancer Institute (NCI) (Levin et al. 1976), although much has never been examined systematically. When information on ambient environmental exposures to an agent is unavailable, it is often possible tc study a small popu- lation with a history of high exposure, for instance in an occupational environment. The study of such populations can generate valuable qualitative data to guide further inves- tigations. The Data Ease ite_tc_eEideir research, imp r.gyed_SYSt€gs^arg_needed ~ ca u sat iye_f sot O£§ . The need for a national registry of death certificates is particularly critical, so that researchers can deter- mine which members of a population under study have died and where their death certificates are filed. Mobility cf individuals makes a national index essential. The importance of such a national death registry has long been recognized (U.S. Department cf Health Education and welfare 1970), and methods for its implementation are currently being reviewed by the Department of HEW. Nevertheless, this is such a fundamental need fcr epidemiological research that we believe it cannot be overemphasi2ed. National registries for major diseases are also needed. Progress is being made toward the development of a national cancer registry, an effort that should continue. There are no comparable data systems for. many other important diseases that may have environmental determinants, such as heart disease or diabetes. Collection cf mortality and morbidity data would be best coordinated through the National Center for Health Statistics. A variety of methods should be explored for obtaining data, such as census surveys or registration procedures for national health insurance programs (if and when the latter are implemented). A simple, effective pro- cedure is needed to correlate data on occupational history.

OCR for page 40
siroking habits, and ether personal or environmental infcr- iraticn about individuals, with death and disease registries. Cne available method is to include Social Security numbers en death certificates and records; this approach is not fully satisfactory, however, and more effective ways to provide systematic linkage of data need to be developed. The most important limitation or the collection and use cf health statistics in epidemiological research is not a technical or economic consideration, but rather the ethical problem of invasion of privacy. We believe that, particu- larly in cases of individuals who have died, the benefit to society of research on environmental factors that may have contributed to those deaths outweighs any possible harm to individuals that might result frcm allowing investigators access to personal medical records. Nevertheless, privacy is a social issue of great concern today, and Congress has moved toward increased protection of the individual. We feel that the impacts of existing or proposed privacy leg- islation on epidemiological research should be carefully examined, so that balanced choices can b« made in cases where these values may be in conflict. Institutional Arrangements Epidemiological studies on ervircnmentally-caused diseases will be most effectively conducted through a coordinated effort involving several federal agencies, plus university and other research teams supported by grants and contracts. It is important that the comple- mentary strengths of assorted agencies be used in mutu- ally reinforcing rather than competitive ways. EPA is best equipped to determine what agents cccur or may occur in the environment, and has a strong program in some areas of environmental toxicology. On the other hand, the stron- gest capabilities for determining the health status of popu- lations and studying the causative processes of diseases are found in several of the National Institutes of Health, such as NCI, and in such agencies as the National Institute cf Occupational Safety and Health (NIOSH). EPA should con- tinue to draw on the expertise of such agencies, rather than establish parallel capabilities within its own research operations. In the past, EPA's needs for epidemiological support have not always been satisfied. More effective cooperative programs, which may reed to be structured to fit the problem being studied, must be devised to carry out the major epidemiolcgical studies needed in the future. Some suggestions for such programs are included in the sec- tions that follow.

OCR for page 40
Cancer te es^ah^ished_t9 examir^ the irc^depce The stateirer.t has been widely repeated that 70 percent cr more of all cancers are caused by environmental factcrs (including those in foods, drugs, cigarette smoke, and occu- pational settings), but this conclusion is not yet supported by adequate evidence on the specific agents and mechanisms responsible. Some very clear temporal and geographic pat- terns in the occurrence of cancers have established a case for environmental origins; however, dose-response relation- ships in humans have not been described for most carcino- gens. It is still not known, therefore, whether current low-level exposures to asbestos fibers or chloroform in water supplies, cigarette smoke in closed rooms, arsenic ar.d other trace metals in ambient air, or many other sub- stances known to be carcinogenic at high exposures pose significant risks to human populations. The possibility that a no-effect level may exist for each potential car- cinogen in the environment has not been resolved, and the combined impact on health of simultaneous exposures to many such agents at very low levels is unknown. The uncertainties inherent ir translating experimental data on carcinogens into environmental standards create a critical need for epidemiological research on cancer. Attention should be concentrated first or those areas for which the most complete environmental monitoring data are available. Data on the geographic distribution and inci- dence of many forms of cancer are being collected and ana- lyzed by the NCI (Levin et al. 1976); expansion cf tumor registries for the study of environmental cancers should build upon this existing base. For the study of environ- mental carcinogenesis to be effective, a stronger effort must be made to coordinate the complementary capabilities of EPA, NCI, the National Institute for Environmental Health Sciences (NIEHS), NIOSH, and other agencies. Because the information most urgently needed in this area today is on the identities and levels of potentially carcinogenic sub- stances in the environment, EPA should be given an important role in the program. The major biomedical components of the research, including collection and analysis of cancer mortality and morbidity data, shculd reside in NCI, NIEHS, or other appropriate specialized research agencies. A new five-year, five-millior-dcllar Program for the Assessment of Carcinogens in the Environment (PACE) has

OCR for page 40
been proposed by EPA, and $1.1 million has been appropriated to begin the effort. Sound design, adequate support, ard effective interagency cccrdinaticn are needed if this criti- cal research program is to succeed. Eirth Defects About 5 percent (150,000) of the babies born in the United States each year have birth defects. Of these defects, no more than one third have a known cause; cur- rent understanding suggests that the remainder may be caused by interactions between genetic ar.d environmental factors, or among multiple environmental influences (Wilson 1975). Extensive experimental and clinical evidence has shown that some drugs and certain chemical, physical, or biological environmental agents can produce specific birth defects. Examples include the drug thalidomide, ionizing radiation, rubella virus, and methyl mercury. Although most demonstrated cases of defects caused by each of these teratogenic agents involved unusually high exposures, the potential of these and similar agents at low levels in the ambient environment for causing birth defects needs to be investigated carefully. Eirth certificates are routinely monitored to compile data on occurrence of birth defects in only a few areas, such as the states of Washington and New York. The cost and effort involved in obtaining such records are saall, while the costs of not monitoring births could be great. Specific birth defects due to particular environmental agents are unlikely to cccur so often that they will easily be noticed, unless special efforts to discover them are made. Pesponsibility for establishing a national program for the surveillance of birth defects should most reasonably reside with the National Center for Health Statistics and the Center for Disease Control, since these agencies have experience in organizing such efforts. EPA is likely tc be a major user of the informaticn assembled under such a program, and should contribute tc the support, planning, and review of the surveillance system, sc that data of particular significance for the evaluation of teratcgenic effects of environmental pollutants would be included. - 45 -

OCR for page 40
Mutaticrs damage Evidence on the incidence of genetic diseases is con- flicting, and the role of environmental agents in producing mutations in man is largely unknown. Methods for testing mutagenicity in experimental organisms are of uncertain relevance to man; therefore, it is important to monitor human populations to detect increases in mutations that may fce due to factors in the environment. A few effective tests for mutations in human populations have been devel- oped, but no organized surveillance programs for large populations have been established. A need exists, there- fore, to develop further effective techniques for detecting and monitoring mutations in man, and for maintaining sur- veillance and informati-cn systems. Research is also needed to determine the significance cf mutations that may occur because of environmental pol- lution. Very little is known abcut the health consequences cf either germinal mutations (which may be passed on to future generations) or somatic mutations (which are confined to the individual who is exposed to a mutagen) . Without more specific understanding of the imcacts of mutations on the lives of individuals, the social ccsts of genetic damage are difficult to assess. The importance attached tc muta- tions seems likely to increase as more is learned of their impact on health; population data on their incidence are therefore likely to be very significant for future decision makirg. Institutional and funding arrangements for the develop- ment cf testing methods and surveillance programs for muta- tions ought primarily tc include certain of the National Institutes of Health, such as NIEHS, NCI, or the CDC. EFA might participate in efforts to define the relationships between certain environmental contaminants and mutations, tut knowledge of this topic is still too rudimentary to expect early epidemiological surveys tc produce informa- tion of immediate regulatory significance.

OCR for page 40
Eehavioral Abnormalities Exj.stj.nq technique, § for. measuring ro2men.ta.l_a.gen.ts. Behavior is the product of the integrated function of the body's many organ systems and metabolic processes, and any number cf toxic effects may be expressed as behav- ioral changes. For example, heavy smog adversely affects the performance of athletes, and carbon- monoxide decreases alertness. Research on possible behavioral effects of most pollutants is at too early a stage of development tc per- mit extensive interpretation of the risk of such effects. Nevertheless, it is likely that behavioral effects can occur at levels of exposure below those that produce other, less equivocal toxic effects (see NRC 1975b) . Impacts on behav- ior may therefore be widespread; in the long run, such effects may represent a major social ccst of pollution. Eecause behavioral changes may be the most sensitive signs of toxic impacts cn organisms, research to advance understanding of this class of effects should have high priority in the environmental health field. The opportu- nity exists for behavioral assessment to play an important ancillary role in clinical medical evaluations of popula- tions exposed to toxic substances in the environment. 1c make the most of this opportunity, research is needed or the relationships between behavioral effects and other (often later) toxic consequences, Current clinical neurological tests can detect overt disease, but are inadequate for measuring many more sub- tle behavioral changes. Some techniques developed in experimental psychology and psychcpharmacology permit objective, quantitative measurements of human behavior; examples of such tests include those fcr physical strength and endurance, motor cocrdinatior. and control, vigilance, and performance on simple, monotonous intellectual tasks (Valzelli 1973). Additional techniques to test for behav- ioral changes in humans are needed for assessment of behavioral effects of environmental pollutants. The needed techniques include: - 47 -

OCR for page 40
• more precise ard complete measurements of neurc- logical dysfunction; • psychometric tests to detect early, subtle, non- specific, subjective behavioral effects, such as fatigue or mood changes; • quantitative measures of behavioral effects in populations, such as school absenteeism cn heavy smcg days. Behavioral assessments using existing techniques should be incorporated in examinations of individuals who are exposed both to high concentrations of toxic contaminants (as in an occupational setting) and to ambient levels of pollutants. Experience should suggest refinements needed to make tests more suitable for evaluaticns of effects cf environmental agents. Each cf the tests now available measures a different array of functional changes; it is important, therefore, to select a test that can answer a specific, relevant question. If, for example, a contami- nant is suspected of making people less responsive to mini- mal changes in their surroundings, an appropriate behavioral screen would be to measure performance on a test for vigi- lance, such as the ability to detect infrequent signals pre- sented in an1 experimental situation. If the likely nature of behavioral changes is unknown, a battery of tests may be required. Evidence of behavioral effects drawn from epidemio- logical studies ought tc be followed by controlled clini- cal studies of human subjects exposed to the suspected agents, where this is possible. Studies of behavioral effects should be combined with other clinical and epi- demiologcal investigations, such as those on chronic de- generative diseases, described elsewhere in this chapter. To carry cut the needed research, the various strengths of EPA, NIOSH, NIEHS, the National Center for Toxicological Research (NCTR), industry, private research institutes, and universities should be used in complementary fashion. The most critical need is to correlate behavioral assess- ments with other measures of toxic effects, rather than simply to conduct separate behavioral tests. Support should therefore be concentrated on integrated, multi- disciplinary epidemiological research prcgrams.

OCR for page 40
Chronic Degenerative Diseases Chronic degenerative diseases (exclusive of cancer) are the leading causes of death and disability. Many have known environmental associations, such as that between chronic bronchitis and particulate air pollu- tion (Storer 1975). Others, such as stroke, high blood pressure, bronchial asthma, and diabetes, have distinc- tive geographic patterns, clinical characteristics, or physiological mechanisms that suggest that they may have environmental causes or at least that there is a major environmental determinant of their frequency of occur- rence. Epidemiological studies are reeded tc identify popu- lations at high risk of such diseases, either because of intrinsic special sensitivities or because of excessive exposures to1 potential causative agerts. Once identified, such populations should be examined clinically, and toxi- cological tests should be undertaken to provide additional insights into pathogenesis and tc help identify causative agents. In some cases, substantial epidemiological inves- tigations may be needed to define approximate dose-respcnse relationships, especially when Icng-term, low-level expo- sures and long latency periods are involved. The need to include functional tests in epidemiological surveys deserves special emphasis. Most chronic degenera- tive diseases of such organs as the lung, liver, kidney, or circulatory system initially manifest themselves as phys- iological abnormalities, before morphological changes are apparent. These alterations in crgan function can be detected by clinical tests, usually at a stage where the course of the disease is still reversible. If epidemiolog- ical studies include, for example, tests of liver or pul- monary function, they can make it possible not only to iden- tify the early stages in the process of environmentally- induced disease, but also to halt or reverse that process in at least some cases. EFA's current efforts to identify populations at risk because of exposure to polluted environments are laudable, but much more needs to be done. Populations with special occupational exposures and residents of geographic areas with unusually high incidence of specific diseases offer

OCR for page 40
much potential for valuable studies. Tc avoid unproductive competition for the limited supply of both epidemiologists and funding, collaborative efforts among EPA, NIEHS, NICSH, and ether appropriate institutions should be pursued. As in the case of cancer epidemiology, EPA*s strongest contri- butions are likely to be in the area of characterization and measurement of environmental factors that may influ- ence particular diseases. The determination of the disease status of populations and the unraveling of causative pro- cesses should draw on the expertise cf seme of the more specialized units of the Department cf Health, Education and Welfare. TOXICOLOGY Texicological studies involving mammalian, avian, and aquatic species are the cornerstcne of research to deter- mine adverse effects of environmental agents. In this chap- ter, toxicology is discussed chiefly in terms of the use of animal (largely rodent) models as surrogates to test for potential effects on human health, Texicological studies cn other domestic animals and wildlife are important for the information they produce on environmental influences in diseases of species other than man (see Chapter 5). Toxi- cological experiments can demonstrate qualitative effects, such as cancer or birth defects, that might occur in humans as well as in test animals, and they can suggest quantita- tive (dcse-response) relationships. For some effects of some agents in some species, a nc-effect level can be dem- onstrated with a given degree of statistical reliability. Texicological research also includes study of mechanisms cf action leading to effects. Animal studies have some serious limitations, however. There are many uncertainties in attempts to relate data on cne species, such as a mouse, to possible effects in other species, such as man. Quantitative comparisons between species are especially difficult. Furthermore, it is vir- tually impossible to match the ccnditicns of exposure in the laboratory with those in the ambient environment, either in terms of the timing cf doses cr in terms of the mixture of pollutants and other variables that occurs in the "real world." When effects of very low-level, long-term exposures are the object of the investigation, statistically valid measurements require either that a very large population of animals be exposed or that doses,substantially higher than those encountered in the ambient environment be used. Tests for chronic toxicity involve elaborate, often cumbersome procedures, may take several years to complete, and can be costly in equipment, animals, and manpower. While much will continue to be learned through long-term - 50 -

OCR for page 40
formidable. This is a relatively new field of research; it is ore in which we feel a considerable exparision of effort is needed. In many occupational environments and in polluted urban areas, the most common route of exposure is likely to be inhalation. Inhalation toxicology therefore is one useful tool for study of exposures to mixtures of contaminants. Inhalation studies have been conducted or a wide variety of single toxicants, and on some combinations of contaminants, among which are NO2 and carbon particles (Boren 1961), S02 and carbon particles (Dalhamn and Strandberg 1963), SO2 and soot (Pattle and Burgess 1957), S02 and KaCl aerosols (Amdur 1959) and chromium-bearing ore ard diesel engine exhaust (Stuart et al. 1970). Seme studies have shown synergistic interactions of contaminants that produced reversible or irreversible disease. Seme combinations have been found to be less toxic than either of the compounds alone. A great deal of information cn inhalation toxicology has been developed, including considerable knowledge of the deposition and clearance of contaminants in the respiratory tract. Further refinements of the techniques of inhalation tcxicology will be required for investigations of complex mixtures of contaminants. Replication of occupational cr ambient environments for controlled studies is difficult because of the complexities of generating and monitoring multiple contaminants. Fange-finding tests are required to set the exposure levels for a chronic evaluation for a given test animal, and concentrations significantly above ambient conditions may have to be used. Improvements are needed in methods of generating and monitoring contaminants, and in the design of chambers and air-flew systems. The capability to conduct complex inhalation studies has been developed in a relatively small number cf govern- ment and private cr university laboratories. The most fruitful approach to accomplish reeded inhalation research would appear to be to continue tc support such programs. A more complete evaluation cf the effects of mixtures cf contaminants requires tests based on ether routes of administration, such as ingestion, as well as inhalation studies. Tests combining inhalation and ingestion of dif- ferent agents are also needed to mimic the actual condi- tions of human exposures. - 53 -

OCR for page 40
Bioassays for Mutagenicity developed. Many biological assays exist for testing the mutagenic properties of environmental contaminants. Most involve microorganisms or other simple systems such as or NeurgsESIS that allow rapid detection of mutations. The use of mammalian microsomal preparations in combination with micrcbial assays has increased the value of test results, but such tests are still of questionable relevance in evalu- ating the risk of mutagenic effects in man or other irammals. Mice offer a better human analog, but such studies are time consuming and require many animals. Jjj yjtr.o tests using cell cultures or tissue preparations (including soire derived from humans) have proved valuable. In a few instances, blood or urine from persons exposed to drugs or other agents have been tested for mutagenicity on microorganisms (for example, see U.S. DHEW 1976). Additional biological tests that are practical, accu- rate, and relevant are needed, bcth for screening chemicals for mutagenicity and for testing the anfcient environment for possible mutagenic hazards. Significant opportunities exist to develop additional useful techniques. Research on the mechanisms of mutations should continue to be pursued, both to facilitate the development of bicassays and to increase understanding of mutagenic risks; programs in this area are under way now at NIEHS. Research to accomplish this goal should probably be relatively unfocused, and would be best performed in a number of existing university or government laboratories with funding through certain of the National Institutes of Health, such as NIEHS, NCI, or through the Food and Drug Administration (FD£) , rather than through EPA. Behavioral Effects _behayj.orgl £h§aaes_aad_2ther_me§sures_2f .functional Eehavior depends on the functional integrity of a great many organ systems and metabolic processes. Toxic effects on any of the body's systems, such as the nervous, endo- crine, or immune systems, may be expressed in behavioral

OCR for page 40
changes. Behavioral effects may in turn involve a wide array of functional responses, such as nictor strength and coordination, learning and memory, speech and communica- tion, and emotional states. Examples of behavioral effects in humans include interference with senscry and motor inte- gration, and intellectual and emctional disabilities asso- ciated with methyl mercury and lead toxicity, respectively (Eryce-Smith 1972, Evans et al. 1975, Spyker 1976). As noted earlier, behavioral effects can be extremely deleterious to individual humans and to society. Behav- ioral changes in animals may have significant ecological consequences, through impacts on processes like feeding, reproduction, and predator-prey interactions. Current knowledge suggests that subtle behavioral changes may occur at extremely low dose levels, and thus may be sensitive indicators of toxicity that might be used to detect harmful effects before irreversible damage has been done (NPC 1975b). Assessments cf behavioral effects ir laboratory animals, therefore, are potentially very useful for evaluating a wide range of toxic responses. This advantage is partially cff- set by the complexity of behavior. So many different func- tions are involved in any given behavior that a battery of tests is usually required to screen for behavioral changes. Furthermore, compensatory mechanisms within the organism may mask a toxic process and its behavioral expression until overt physical damage has been dcne. A more fundamental difficulty, however, lies in the interpretation of results of behavioral tests. A great number, of techniques available from experimental psychol- ogy can reliably measure effects on behavior; but behavioral toxicology is still too new a field tc have developed unifying concepts or principles for determining the signif- icance of behavioral effects. Two approaches, epidemiology and toxicology, offer great promise to shed further light on the effects on behavior of a polluted environment. Epidemiclogical studies on behav- ioral responses, discussed above, must focus on the correla- tions between changes in behavior and other expressions of toxicity. The most productive approach in behavioral toxi- cology is also to emphasize investigations that can reveal relationships between behavioral changes and other measure- ments of functional impairment based on such techniques as neurophysiology, histochemistry, and morphology. The rela- tionship between a given behavioral change and underlying functional deficiencies must be understood before behavior can be used as a sensitive indicator of other, covert trx?.c effects, and before behavioral screens can be employed to - 55 -

OCR for page 40
assess the potential toxicity of substances in a regulatory context. A secondary objective of research in behavioral toxi- cology should be the development of quantitative, practical, and reproducible test batteries that use behavioral changes to measure specific functional impairments. Until the mean- ing of particular behavioral changes is determined, however, uniform testing procedures cannot be developed (Spyker and Avery 1976), Coordinated multidisciplinary tcxicclogical research is required to determine the correspondence between various functional effects and their behavioral consequences. Pro- grams with the needed combinations of training and experi- ence to carry out such research can be found in a small number of universities and some government laboratories. Eecause of the fundamental nature of much of the research needed, universities should play a relatively large role in carrying out the recommended programs. Several federal agencies have an interest in applications of behavioral tests, and should be involved in the support and coordina- tion of this research; among these agencies are the National Institute of Mental Health, NIEHS, FDA, NIOSH, EPDA, and EPA (through NCTR)• Some research tc develop behavioral assess- ment techniques and to test specific substances before mak- ing regulatory decisions has beer conducted within these agencies and should probably continue. KIEHS should be best suited to coordinate research on behavioral effects of envi- ronmental pollutants. Teratology mphasise assessment s functional disorders gaused mental agents. Toxic substances, including environmental pollutants, can have marked adverse effects en the development process. Embryos are especially vulnerable to exposures during the organogenic phase of early development, but functional dis- orders produced prenatally by toxic action may appear at any point in the process from conception through postnatal matu- ration. In mammals, the nervous, endocrine, and immune sys- tems seem to be most susceptible to damage; all three con- tinue to differentiate well.into the postnatal period. Following the thalidcmide incident in 1960, tests for teratogenicity have been required for the approval of new - 56 -

OCR for page 40
drugs; corsequently, there is a substantial research base front which to conduct assessments of the teratogenic poten- tial of environmental contaminants. However, measurements of teratogenicity have sc far been largely based upon the detection of relatively gross defects, such as structural abnormalities, overt functional impairments at birth, or retarded growth. It is unlikely that mcst exposures to en- vironmental contaminants will produce such clinically obvi- ous defects; and many subclinical effects may be overtly expressed only later in life, as defects in subsequent matu- raticnal processes such as puberty. It is important, there- fore, that the evaluation of teratogenic effects of environ- mental agents include tests for subtle functional impair- ments and that evaluations be cortinued for an extended postnatal period. Although the assessment of minimal functional disorders is a relatively new direction for teratology, there are a number of useful tests that measure such effects. Nervous system damage as well as impairments of other systems can be detected with behavioral tests (see the preceeding section), and specific tests for endocrine, immune, and liver enzyme functions have been used successfully in a number of studies (Wilson et al. 1976) . An important related need is for research on mechanisms cf action of teratogens and the influence they may have upon biochemical processes of cellular differentiation. Because of the large number of chemicals and other agents (and com- binations of agents) to which an embryc may be exposed and the length of time required for complete post-natal evalua- tion cf potential teratcgenic effects, short-cuts are ur- gently needed to estimate the risk of teratogenicity for a particular substance or class of substances. At present, the best hope of finding such shcrt-cuts is probably to learn more about the mechanisms cf action of specific sub- stances. Knowledge on this subject is scanty, but oppor- tunities exist for significant progress through more inten- sive research (Wilson et al. 1976). Most teratological research at present is conducted by drug companies to meet FDA regulations on evaluation of new products. If this work were expanded to include more extensive assessments of subclinical functional deficits over the life of the organism, it might be of some general use for environmental decision making. More specific evalu- ations of teratcgenic effects of environmental contaminants are under way at NIEHS and NCTP. Some excellent: work is being done in a number cf universities, with support pri- marily from the National Institute cf Child Health and Human Development and the National Foundation—The March of Dimes. Increased support of those existing research programs that are moving in the directions described here - 57 -

OCR for page 40
appears to promise results most useful fcr evaluating the teratogenic risks of pollutants. MECHANISMS CF ACTICK more complete _j,nf ormat 3,25-2" -t&§ Du _ € f f got s . Information' on the mechanisms of action of toxic sub- stances can provide the crucial link between evidence of environmental exposures and observed effects (Miller and Miller 1974). The term roechanisirs of action encompasses the full range of biochemical and physiological processes that cccur in living systems in response to environmental contaminants. From initial exposure to some end point cr effect, processes at levels from. the sub-molecule to the organism as a whole determine the ways in which substances are absorbed, translocated, and transf ormed. A similar array of processes governs the responses of living systems to the presence of contaminants. Research on mechanisms seeks to illuminate, in biochemical and physiological terms for the most part, the details of the processes and changes that occur. Critical areas of information related to mechanisms include: Biological availability at the host/ervi- ronment interface and at the surface of the target cell. ^£ • Identity of target cells; rates and extents of translocation, accumulation, storage, mobiliza- tion, and excretion of chemicals and metabolites. • Mgt§bolic_Fgte: Pathways, products, by-products, and rates of the metabolism of the substance; toxicatior and detoxication. Functional and structural changes in target cells, tissues, and organs. Knowledge of mechanisms of action is extremely impor- tant for the inference cf causal relationships. It can also provide early indications of biological effects that could make it possible to detect the initiation of a pathogenic process before overt disease appears, when preventive action or early treatment are possible. Information on mechanisms - 58 -

OCR for page 40
also offers a basis for deducing the potential effects of substances that have not been extensively tested. When the mechanism of action of a known carcinogen, mutagen, or tera- tcger is understood, a demonstration that another substance produces analogous biochemical or metabolic changes might lead to a presumption that the second substance could have similar toxic effects. Such extrapolations require further proof, but knowledge of irechanisms is the most valuable presumptive basis for choosing additional tests and for estimating risks in the absence cf results of more time- consuming studies. Examples of' the need for information on mechanisms can be found in almost all of the problem areas discussed earlier in this chapter. In research on mutations, for instance, both the metabolism of mutagens and the molecu- lar events associated with mutation ne£d to be understood. Such knowledge could help to answer the difficult question of whether a threshold exists for mutagenic hazards. The study of environmental carcinogenesis requires information on the metabolic degradation of substances and the inter- actions between products of metabolism and the genetic material of cells. Because of the long latent period until tumors appear, knowledge of changes in cell or tissue func- tion that occur early in the onccgenic process is very much needed. It has been suggested that a small number of meta- bolic pathways, perhaps fewer than a dczen, are involved in the metabolism of environmental contaminants by livirg systems (Gehring 1975). Nevertheless, the processes may be extremely complex; for example, different biochemical pathways may be activated at different dcse levels cf a single substance (Miller and Miller 1974). Because of this complexity, it seems unlikely that study of limited aspects cf the mechanisms of action of a large number of substances would lead to useful generalizations within the near future. A sounder research strategy would be to emphasize concentra- ted efforts to fill in gaps in knowledge of the mechanisms cf action.of a few well-studied agents (e.g., chlorinated hydrocarbons or heavy metals). Detailed understanding cf the molecular kinetics of processes that occur in response to a range of doses and under different modifying conditions will be most useful both for decision making and as a guide to future research. Vital as it is to understanding effects, research on mechanisms is subject tc some important limitations. Varia- tions in the ways a single organism or different species respond to a given agent make the meaning of many findings uncertain. Demonstration of a mechanism in a tissue cul- ture, or even in a live animal, is not evidence that the same process occurs in ether animals or man; and elucida- - 59 -

OCR for page 40
tion of the mechanisms cf action of single agents does rot necessarily allow estimation of the effects of combinations cf agents. Understanding a mechanism of action may require years, not months, of careful research (Killer and Miller 1974). Nevertheless, the answers such studies can provide are very important, and this field should continue to receive substantial emphasis in research planning. Several excel- lent existing programs, such as those at NIEHS and NCI, have been mentioned in earlier porticns cf this chapter. EPA should interact closely with NIEHS ard other agencies involved in research on mechanisms, to ensure that the work is useful for the evaluation of potential environ- mental hazards; EPA need not, however, develop an exten- sive program of its own to do such work. - 60 -

OCR for page 40
REFERENCES Amdur, M.O. (1959) The physiological response of guinea pigs to atmospheric pollutants. International Journal of Air and Water Pollution 1:170-183. Ames, B.N. (1976) Carcinogens are mutagens: a simple system for detection. Seminar for Science Writers, American Cancer Society, March 26-30, 1976, St. Petersburg, Florida. New York: American Cancer Society. Eoren, H. G. (196<») Carbon as a carrier mechanism for irri- tant gases. Archives of Environmental Health 1:119-132. Eryce-Smith, D. (1972) Eehavioral effects of lead and other heavy metal pollutants. Chemistry in Britain 8:24O-243. Ealhamn, T. and L. Strandberg (1963) Synergism between sulphur oxide and carbon particles. Studies cn the adsorption and on ciliary movements in the rabbit trachea in yiyo. International Journal of Air and Water Pollution 7 (6/7):517-529. Evans, H.L., V.G. Laties and B. Vieiss (1975) Behavioral effects of mercury and methylmercury. Federation Proceedings 34 (9): 1858-1 867. Gehring, P.J. (1975) Concerns of industry related tc carcinogenic hazards. Presented to National Meeting of Comprehensive Cancer Centers cf the United States, Euke University Medical Center, April 9-10, 1975. Curham, North Carolina. International Agency for Research on Cancer (1971-1975) IARC Monographs on the Evaluation cf Carcinogenic Risk cf Chemicals to Man. Volumes 1-9. Geneva: World Health Organization. Levin, D.I., S.S. Devesa, J.C. Godwin, II, and D.T. Silverman (1974) Cancer Pates and Risks. Second Edition. National Cancer Institute, Eiometry Branch, Publication No. 76-691. Washington, E.G.: U.S. Eepartment of Health, Education and Welfare. - 61 -

OCR for page 40
KcCann, J. and E.N. Ames (1976) P. simple irethod for detecting environmental carcinogens as mutagens. Symposium on Occupational Carcinogens. Annals of the New York Academy of Scierces 271:5-13. McCullough, J.M. (1975) Human health and low-level chronic pollutarts. Pages 53-171, Effects of Chronic Exposure to Low-Level Pollutants in the Environment. Background document by the Congressional Research Service for Hearings presented before the Subccmmittee on the Environment and the Atmosphere, Committee on Science and Technology, D.S. Congress, House. 94th Congress, 1st Session. ' Miller, E.G. and J.A. Miller (1974) Eicchemical mechanisms of chemical carcinogenesis. Pages 377-402, The Molecular Eiolcgy of Cancer, edited by H. EuSch. New York: Academic Press. National Research Council (1975a) Decision Making fcr Regu- lating Chemicals in the Environment. Environmental Studies Board, Commission on Natural Resources. Washington, D.C.: National Academy of Sciences. National Research Council (1975b) Principles for Evaluating Chemicals in the Environment. Environmental Studies Ecard, Commission or Natural Resources and Committee cn Toxiciology, Assembly of life Sciences. Washington, E.G.: National Academy of Sciences. Pattle, R.E. and F. Burgess (1957) Tcxic effects of mix- tures of sulphur dioxide and smoke with air. Journal of Pathology and Bacteriology 73:411-419. Spyker, J.M. (1976) Assessing the impact of low level chemicals on development: behavioral and latent effects. Pages 161-180, Behavioral Pharmacology: The Current Status, edited by B. Weiss ard V. Laties. New York: Plenum Press. Spyker, J.M. and E.L. Avery (1976) Eehavioral tcxicclogy and the developing organism. Proceedings of the Workshop cr Behavioral Toxicclogy, edited by R.S. McCutcheon. Washington, D.C.: National Institutes of Health. Storer, J.E. (1975) Testimony. Pages 150-157, Costs and Effects of Chronic Exposure to Low-Level Pollutants in the Environment. Hearings before the Subcommittee on the Environment and the Atmosphere, Committee on Science and Technology, U.S. Congress, House, 94th Congress, 1st Session, - 62 -

OCR for page 40
Stuart, E.O,, D.H. Willard, and E.B. Howard (1970) Uranium mine air contaminants in dogs and mice. Pages 413-427, Inhalation Carcinogenesis; Proceedings of a Conference held in Gatlinburg, Tennessee, Octcber 8-11, 1969. Eiology Division, Oak Ridge Rational Laboratory. Oak Ridge, Tenn.: U.S. Atomic Energy Commission. U.S. Department of Health, Education and Welfare (1970) Man*s Health and the Environment—Some Research Needs. Report of the Task Force on Research Planning in Environmental Health Sciences, National Institute of Environmental Health Sciences. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Health, Education and Welfare (1976) Occupational Exposure to Epichlorohydron. National Institute of Occupational Safety arid Health, Publica- tion No, 76-206. Washington, D,C.: U.S. Department cf Health, Education and welfare. Valzelli, L. (1973) Methods in psychopharmacology. Pages 59-88, Psychopharmacology, edited by W.B. Essman. New Ycrk: John Wiley and sons. Wilson, J.G. (1975) Environment and birth defects. Pages 29-48, Methods for Detection cf Environmental Agents that Produce Congenital Defects, edited by T. Shepard, J. Miller, and M. Marois. Amsterdam: North Holland Publishing Company. Wilson, J.G., J.M. Spyker, and P. Vergieva (1976) Reproduc- tion, teratology, and postnatal function. Manual on the Evaluation of the Tcxicity cf Chemicals. Geneva: World Health Organization. - 63 -