Cover Image

PAPERBACK
$32.95



View/Hide Left Panel
Click for next page ( 2


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 1
Executive Summary Reproduction and neurodevelopment are processes on which the continuation of any species depends. For humans, repro- ductive processes carry substantial emo- tional weight. We want healthy children born without impairments that hinder their structural or functional development, and we want reproduction to be successful at the appropriate time in life. In the United States, approximately 250,000 babies are born with birth defects each year. Twenty percent of these birth defects are attributed to multiple causes, 15% to intrauterine infections, and 5% to a mutant gene. Environmental factors are identified as a cause with relative certainty in only 2-3% of the total number of cases. This leaves nearly 60% of birth defects for which the etiology is unknown but in which environmental exposure might play a role. For every 3,000,000 U.S. births annually, at least 600,000 embryos or fetuses are aborted spontaneously be- fore the 20th week, and some 24,000 fetuses die before birth. Of live births, nearly 8% are premature, and approximately 7% have low birthweight. Another 3-7% possess some type of malformation. Although the overall incidence of infer- tility remained stable between 1965 and 1982, infertility among married couples in which wives were ages 20 to 24 increased from 4% to 10%, and more than 2 million Amer- 1 lean couples who want to have a baby are unable to do so. This increase appears to be linked to several factors, including changes in the incidence of sexually trans- mitted diseases, but other factors, such as xenobiotic exposures, have not been well studied, and may contribute to repro- ductive impairment. The adverse effects on human reproduction of high doses of ionizing radiation, polychlorinated bi- phenyls, dibromochloropropane, cancer chemotherapy, and alcohol are well estab- lished, but the consequences of lower doses of these and other materials to which humans might be exposed environmentally have not been well studied. Despite the substantial expenditures that have been made on the study of basic reproductive and neurodevelopmental proc- esses, knowledge of these processes and their relationship to environmental exposures remains disappointing. How- ever, it is clear that the situation can be improved if science can identify bio- logic markers of the various steps in the processes. In light of these longstanding and con- tinuing concerns, the Board on Environmen- tal Studies and Toxicology (BEST) of the National Research Council's Commission on Life Sciences undertook a major inves- tigation of the use of biologic markers in environmental health research. At the

OCR for page 1
2 request of the Office of Health Research of the U.S. Environmental Protection Agen- cy, the National Institute of Environmen- tal Health Sciences, and the Agency for Toxic Substances and Disease Registry, the Markers Oversight Committee was formed to clarify the concepts and definitions of biologic markers that could be applied by two subcommittees: the Subcommittee on Markers of Pulmonary Toxicity to review and assess promising markers of the pul- monary system in a separate volume; the Subcommittee on Reproductive and Neuro- developmental Toxicology developed this report with individual panels on male re- production, female reproduction, pregnan- cy, and neurodevelopment. ORGANIZATION OF THIS REPORT This report has four major sections that correspond to the work of the panels on male, female, pregnancy, and neurode- velopmental toxicology. Each section ends with a summary on conclusions and recommen- dations. This executive summary encapsu- lates the major points of the four sections and is followed by an introductory chapter that presents concepts, definitions, and selected applications of biologic mark- ers, which reflects the efforts of the oversight committee to create a framework for the overall project. CONCEPTS AND DEFINITIONS The oversight committee's task was to clarify and define these concepts such that subsequent subcommittees could apply them to reviews of the status and potential use of biologic markers in specific areas of environmental health research. Biologic markers are indicators signal- ing events in biologic systems or samples. It is useful to classify biologic markers into three types-exposure, effect, and susceptibility-and to describe the events particular to each type. A biologic marker of exposure is an exogenous substance or its metabolite~s) or the product of an interaction between a xenobiotic agent and some target molecule or cell that is measured in a compartment within an organ- ism. A biologic marker of effect is a measur- BIOLOGIC MARKERS able biochemical, physiologic, or other alteration within an organism that, de- pending on magnitude, can be recognized as an established or potential health im- pairment or disease. A biologic marker of susceptibility is an indicator of an inher- ent or acquired limitation of an organism's ability to respond to the challenge of exposure to a specific xenobiotic sub- stance. Biologic markers of susceptibili- ty are discussed in the report only insofar as they also can serve as markers of expo- sure or effect. Once exposure has occurred, a continuum of biologic events may be detected. These events may serve as markers of the initial exposure, dose (half-life, circulating peak, or cumulative dose), biologically effective dose (dose at the site of toxic action, dose at the receptor site, or dose-to-target macromolecules), altered structure/function with no subsequent pathology, or potential or actual health impairment. Even before exposure occurs, biologic differences among humans might cause some individuals to be more suscep- tible to environmentally induced disease. Biologic markers, therefore, are tools that can be used to clarify the relation- ship, if any, between exposure to a xeno- biotic compound and health impairment. Markers of Exposure Exposure is the sum of xenobiotic materi- al presented to an organism, whereas dose is the amount of the xenobiotic compound that is actually absorbed into the organ- ism. Blood flow, capillary permeability, transport into an organ or tissue, the number of receptor sites, and route of administration (which determines the path of the parent compound or its metabo- lites in the body) all can influence ab- sorbed or biologically effective dose. An inhaled carcinogen might produce tumors in the lung, but if the same material were ingested and eliminated via the kidney, renal tumors might be Produced. If the parent compound is responsible for the observed toxicity, the amount of metabo- lite reaching the target may be of no con- sequence. If metabolites are responsible,

OCR for page 1
EXECUTIVE SUMM4RY however, metabolism in the liver, the tar- get organ, or elsewhere as a result of meta- bolic cooperation between several tissues is an important determinant of absorbed and biologically effective dose. Markers of Effect For present purposes, the effects on or responses of an organism to an exposure are considered in the context of the rela- tionship of exposure to health impairment or the probability of health impairment. An effect is defined as an actual health impairment or recognized disease, an early precursor of a disease process that indi- cates a potential for impairment of health, or an event peripheral to any disease proc- ess but correlated with it and thus predic- tive of development of impaired health. A biologic marker of an effect or re- sponse, then, can be any change that is qualitatively or quantitatively predic- tive of health impairment or potential impairment resulting from exposure. Bio- logic markers are also useful to identify an endogenous component or a system func- tion that is considered to signify normal health, e.g., blood glucose. It is impor- tant to realize, however, that the concen- tration or presence of these markers repre- sent points on a continuum. Therefore, the boundaries between health and disease may change as knowledge increases. Markers of Susceptibility Some biologic markers indicate individ- ual or population differences that affect the biologically effective dose of or the response to environmental agents indepen- dent of the exposure under the study. An intrinsic genetic or other characteristic or a pre-existing disease that results in an increase in the absorbed dose, the biologically effective dose, or the target tissue response can be markers of increased susceptibility. Such markers may include inborn differences in metabolism, varia- tions in immunoglobulin levels, low organ- reserve capacity, or other identifiable genetically determined or environmentally induced variations in absorption, metabo- lism, and response to environmental agents. Other factors that may affect 3 individual susceptibilities include nu- tritional status of the organism, the role of the target site in overall body function, condition of the target tissue (present or prior disease), and compensa- tion by homeostatic mechanisms during and after exposure. The reserve capacity of an organ to recover from an insult at the time of exposure may also play an impor- tant role in determining the extent of an impairment. EXTRAPOLATION FROM ANIMALS TO HUMANS Extrapolations to humans are to be based on the most sensitive animal species tested, barring clear evidence that the species is toxicologically distinct from humans. Within the past 2 years, EPA issued guidelines for evaluating reproductive studies. These provide a means to estimate data quality and stipulate segment II de- velopmental toxicity requirements of two species and three treatment groups with 20 rodents or 10 nonrodent mammals. As this report went to press, EPA issued additional guidelines on evaluating re- productive studies. Laboratory animals and humans can differ in toxicokinetics. Thus, use of data from animals to determine health risks in humans must be assessed carefully, es- pecially when the data are derived from monitoring of external exposure. The toxicity of some chemicals is medi- ated either by activation or by detoxifica- tion biotransformation reactions. Inas- much as biotransformation differs among species, it is important to establish whether the routes and rates of human and animal metabolic pathways are similar. Health risks often are associated with combinations of effects in humans. For example, cardiovascular disease in humans can encompass atherosclerosis and hyper- tension. Although swine provide the most suitable animal model for studying spontaneous atherosclerosis, young rats might be most appropriate for studying hypertension. For humans, estimating the diseases necessarily would entail some appropriate combination of the rele- vant animal test systems.

OCR for page 1
4 A common source of uncertainty in risk assessment is the dose-response relation- ship at low doses or for rare effects. It is often impractical to conduct studies of effects at low doses, because large numbers of animals are required to detect a low incidence of effects. Demonstrable health effects in humans, given the limits of epidemiology, often are associated with high doses and hence high risk. Sensitive molecular markers being developed will permit study of the relationship between exposure to chemicals at low ambient con- centrations and the formation of a molecu- lar marker predictive of human risk. The development of biologic markers might enable scientists to make better use of laboratory animal data in estimating the effects of chemicals in humans. As a 1986 NRC study on drinking water and health observed, the timing of exposure and the patterns of dose response obtained in animal studies have important implica- tions for extrapolating data to humans. Most often neglected in such extrapola- tions in the reproductive arena is fetal growth retardation, which has obvious relevance to low birth weights in humans. In the absence of a significant reduction in maternal weight gain, fetal growth re- tardation can be an important event. The NRC also reported that good evidence of dose-response relationships exist for a number of well-studied developmental toxicants. An FDA review of the literature showed that of 38 compounds having demon- strated or suspected teratogenic activity in humans, all except one tested positive in at least one animal species. More than 80% were positive in more than one species. Despite this concordance of findings, qualifications must be placed on their direct application to risk assessment for humans. QUALITY AND QUANTITY OF DATA Quantity of data required can be deter- mined by statistical power considerations and resource considerations. Statistical power is related to the number of subjects in a group, rarity of the end point studied, and variability in the frequency of the end point occurrence. The greater the BIOLOGIC MARKERS expected relative risk, the smaller the population that will need to be studied. The study of reproduction and develop- ment poses major resource and logistic problems for those working with laboratory animals. For instance, manageable sample populations do not reveal increases in toxic events of less than 5 to 10%. For some health effects associated with reproduc- tion and neurodevelopment, such as muta- genesis and teratogenesis, incidences in a human population of 3 per 10,000 are significant. Obviously, these effects cannot be well defined in whole-body stud- ies of thousands of experimental animals at a time. Classic toxicology studies of rodents involve the exposure and patholog- ical analyses of 200 animals for 2 years. In such assays, each animal is a surrogate for 1,000,000 people. Birth defects un- detectable in this rodent population could be epidemic within ten generations, if they occurred in humans. BIOLOGIC MARKERS ASSOCIATED WITH REPRODUCTIVE AND NEURODEVELOPMENTAL TOXICOLOGY From the outset, the subcommittee grap- pled with the knowledge that the processes under its consideration are complex and normal, and therefore, markers of func- tioning might well provide the only mark- ers for study. (In contrast, a study of markers of disease would seek to iden- tify signals of adverse health or patho- physiology.) Reproduction and neurodevelopment are complex, stepwise processes that begin with gametogenesis; continue through gamete interaction, implanta- tion, embryonic development, growth, par- turition, and postnatal adaptation; and are completed with sexual and developmen- tal maturation of the newly formed indi- vidual. The study of these areas subsumes the disciplines of reproductive and devel- opmental biology, toxicology, terato- logy, and pharmacology, as well as epide- miology, occupational and environmental health, and medicine. Interest is growing in the use of biolog- ic markers to study the human health ef-

OCR for page 1
EXECUTIVE SUMMARY fects of exposure to environmental toxi- cants in clinical medicine, epidemiology, toxicology, and related biomedical fields. Clinical medicine uses markers to allow early detection and treatment of disease; epidemiology uses markers as indicators of absorbed dose or of health effects; toxicology uses markers to help determine underlying mechanisms of dis- eases, develop better estimates of dose- response relationships, and improve the technical bases for assessment of risks at low levels of exposure. This report focuses on the identifica- tion of indicators of differences between individuals or between cells that might be related to the reproductive potential of adults or the development of children. Few such biologic markers have been demon- strated to identify early stages of health impairment or toxicologically relevant absorbed doses. The detection of in- creased alpha-fetoprotein in a pregnant woman's serum and in amniotic fluid has been used to identify fetuses at risk of neural tube defects. Concentrations of lead in serum have been correlated with necrologic changes. Those two examples of biologic markers predict adverse health effects if measured concentrations are extreme. But for only a few other biologic markers have particular values or ranges of values been demonstrated to be predic- tive of adverse health effects of specific toxic exposures. Therefore, this report discusses a broad range of biologic markers and their use in studies of reproductive and developmental toxicology. The sub- committee does not discuss their utility for predicting adverse health effects of toxic exposures, because research results relevant to that interpretation are not available. The use of biologic markers in reproduc- tive and developmental toxicology is changing quickly, which reflects the rapid incorporation of sophisticated measurement techniques at the subcellular levels. Markers now used in clinical stud- ies are sometimes also used in population studies merely because there is some knowl- edge of their utility and their shortcom- ings. Their use today does not imply that they will be used tomorrow in their present s form or even at all. The subcommittee de- termined that few reproductive and devel- opmental markers have been used in environ- mental health research. Therefore, in applying the oversight committee's defi- nitions of biologic markers, the subcom- mittee described biologic markers of re- productive function and neurodevelopment in general, and classified them according to their immediate and potential utility in environmental health research: Biologic markers that could be recom- mended today for assessment of an exposure and reproductive or developmental re- sponse to that exposure. Biologic markers found promising in laboratory studies that could be recom- mended for further research aimed at use in clinical and epidemiologic studies. Biologic markers that could be studied only in animal models or only in some specific situations in humans (e.g., at autopsy or in surgical specimens). For some assessments, the reproductive or neurodevelopmental health effects associated with an abnormal test result might not be defined, the range of normal values of a biologic marker might be poorly characterized, or information might be lacking. For example, if xenobiotic con- centrations in the exposed population are low, the population's risk of adverse reproductive or developmental health effects might be concluded to be small. However, such a conclusion might be based on the absence of studies of low-dose ex- posures, rather than on the presence of detailed epidemiologic research that demonstrates safety. These important distinctions must be explicitly taken into account in any assessment of exposure. Ideally, tests selected to detect the occurrence of xenobiotic exposure and resulting effects are sensitive, specif- ic, inexpensive, minimally invasive, attended by low risk, and easily used in large populations. It is sometimes pos- sible to use a multistage testing process to detect exposures: the earlier stages would use sensitive but easily applied tests to identify a group that is at high risk of exposure, and more definitive

OCR for page 1
6 tests-which might be more expensive, com- plicated, and invasive-would be used with the group identified by earlier tests as likely to have been exposed. The subcom- mittee encourages the development of im- proved tests. When protocols are being designed, bio- logic markers that are still being devel- oped should be included. New markers can then be validated against those in use. In addition, long-term storage of biologic and environmental samples always should be considered a resource for future stud- ies; they can be analyzed retrospectively for markers that will be developed later. Not all biologic markers are relevant to therapeutic intervention. Some might identify an exposure or the presence of disease but not provide information rele- vant to a therapeutic decision. Thus, in some circumstances, diseased or other- wise affected persons would not benefit from the information yielded but would function as biologic markers for the com- munity, allowing public health measures to prevent further exposure or disease. BIOLOGIC MARKERS ASSOCIATED WITH MALE REPRODUCTION Information on the frequency of natural- ly occurring reproductive disorders in the human male and reproductive abnor- malities induced by toxic chemicals is sparse. Reproductive health outcomes of two major classes are of concern when human males are exposed to toxic chemicals: pathophysiologic changes that might be associated with alterations in fertility and genetic damage that will be conveyed to future generations. Several markers are used to assess wheth- er pathophysiologic changes in human males have occurred in response to toxic chemicals. Such markers are measures of potency, measures of fertility, testicu- lar size, serum concentrations of gonadal and pituitary steroids, and semen charac- teristics, such as spermatozoa! concen- tration, motility, and structure. Those markers are of varied utility for identify- ing toxic effects. For instance, fertility is unlikely to be a sensitive indicator of moderate effects of toxic insult in BIOLOGIC ~= humans, because average daily sperm pro- duction exceeds what is required. In addi- tion, many elements of the male reproduc- tive system (e.g., epididymal function) can be altered in response to toxic chemi- cals without affecting fertility. Markers of genetic damage in the human male (or female) genome include pregnancy outcome, presence of sentinel phenotypes (e.g., achondroplasia), and changes in macromolecules in offspring, such as elec- trophoretic variants of red-cell enzymes. Changes in rates of heritable mutation have not been demonstrated in human pop- ulations exposed to agents that are known to induce heritable mutations in animal models. Clearly, a battery of biologic markers is needed that reflects a wide array of pathophysiologic changes and genetic damage. Research on or development of the following areas are important. Markers of Physiologic Damage In vitro sperm assays that measure the extent of the capacity of animal and human sperm to fertilize oocytes. Sperm function assays that use mono- clonal antibodies of specific domains on spermatozoa! plasma membranes. Computer-based and automated tech- niques for measuring and sorting spermato- zoa on the basis of such characteristics as counts, motility, structure, domains, and enzyme function. Specific markers and their cDNA probes for the testes and individual acces- sory sex organs. Markers that reflect epididymal func- tion, particularly sperm motility and the acquisition of fertilizing ability. Probes for testicular RNA, DNA, and other macromolecules, to use in developing assays for quantifying specific steps in the development of germ cells. Assays for inhibin, androgen-binding protein, and Mullerian-inhibiting factor and other polypeptides that constitute markers of Sertoli cell function. Assays that allow enumeration of sper- matogenic stem cells. Comparative analyses of human and laboratory-animal germ cell responses

OCR for page 1
EXECUTIVE SUMMARY to toxicants and identification of useful laboratory animal models for studying effects on human beings. Mechanisms of toxicity in laboratory animals and in vitro culture systems. Markers of Genetic Damage and Heritable Mutations Human semen markers of genetic toxici- ty and induced mutations, including meth- ods for detecting gene mutations, aneu- ploidy, chromosomal aberrations, and DNA adducts in mature sperm and in immature germ cells. Techniques to use DNA markers to moni- tor human heritable mutations such as Ler- ner gels, restriction-fragment-length polymorphisms, subtractive hybridiza- tion, and RNase digestion. Improved genetic and cytogenetic methods for detecting induced heritable aneuploidy and molecular methods for de- tecting heritable DNA changes in labora- tory mammals. Basic mechanisms of induction and molecular nature of heritable mutations in laboratory mammals. Identifying biologic markers to assess toxic agents or effects of toxic agents on specific male reproductive functions and heritable genetic damage is only the first step. Biologic markers to assess exposure must be distinguished from mark- ers to assess effect. Variations in mark- ers must be correlated with doses of spe- cific toxicants and with health outcomes, mechanisms of toxicity must be elucidated, and strategies for risk assessment must be developed. BIOLOGIC MARKERS ASSOCIATED WITH FEMALE REPRODUCTION Information on female reproductive toxicology is even more sparse than that on the male, because of differences in gametogenesis and accessibility of ger- minal cells and because of the cyclic na- ture of female reproductive function. The differences are found in experimental animals, as well as humans. Among possible biologic markers of fe- 7 male reproduction, some clinical measures are available for assessing human female sexual development and maturation and cyclic ovarian function. Those measures, based on serum concentrations of ovarian steroids and pituitary hormones, have not yet been used in population-based studies of reproductive toxicology. Development and improvement of hormonal markers in easily obtained biologic specimens, such as urine and saliva, are needed for appli- cation in field studies. In light of diur- nal and cyclic fluctuations in most repro- ductive hormones, sampling schemes that are valid and practical have to be worked out. Most applications to date have in- volved markers of ovulation (such as lu- teinizing hormone-LH) in the context of fertility research. The results have sug- gested that it is feasible to incorporate hormonal assessments of cyclic function into population studies. Conventional self-reported epidemiologic measures of female reproduction are also appli- cable, including pubertal development, cycle length and characteristics, fer- tility, and age at menopause. Results of work with conventional epidemiologic measures of reproductive performance sug- gest that the processes of sexual matur- ation and ovarian function are vulnerable to toxic insult. The use of objective and more precise biologic markers should pro- vide additional data on sites of action and mechanisms of toxicity. Appropriately sensitive and specific urinary human chorionic gonadotropin (hCG) assays of pregnancy are available for assaying fertility and have been field tested. However, the best of the current assays are labor-intensive, so further refinements will be needed to allow large- scale applications. hCG is a valid marker of pregnancy following implantation, but a marker of conception in the preimplanta- tion period also is needed, so that the frequency and fate of human conception can be estimated accurately and the deter- minants of embryonic loss in humans fully elucidated. As to genetic damage, the relative inac- cessibility of ovarian material makes it necessary to take advantage of special clinical opportunities to obtain tissues

OCR for page 1
8 or make observations. Materials from spon- taneous abortions, in vitro fertiliza- tion/embryo transfer centers and related technologies could be made available to researchers and would lend themselves to systematic study. Such specimens have been used to generate valuable data on germinal exposure to pollutants and to estimate the frequency of chromosomal anomalies in the oocyte and conceptus. The issue of female germ cell damage is of the utmost importance, because most aneu- ploidy in humans originates in the oocyte. Measurement of oocyte genetic damage now depends on human samples of convenience, such as material from surgical or infer- tility patients, and on research with la- boratory animals. Heritable damage can be assessed in offspring, but research has not yet identified an environmental influence on heritable damage, at least among the common exposures that have been studied in humans. Laboratory investigation of female reproductive toxicology, as well as female reproductive biology in general has suffered from the lack of a practical animal model appropriate for all aspects of reproduction. For example, the mouse is a good model of primate oogenesis, but lacks a functional corpus luteum, except during pregnancy, and thus is not useful for evaluating the effects of toxicants on luteal function. A systematic assess- ment might identify relevant animal models for different aspects of the female reproductive process that could be simul- taneously studied to evaluate environmen- tal health effects. Germ Cell Damage Use of new reproductive technologies could provide human materials for examin- ing relationships among oocyte cytogenet- ics and follicular fluid, serum, and tissue concentrations of pollutants. Estimates of oocyte chromosomal ano- malies need to be validated based on com- parison of infertility patients with nor- mal females. Mechanisms of aneuploidy and assays that identify agents that cause it need study and development. BlOL~IC ~M Markers of genotoxicity (e.g., DNA- adduct formation) should be applied to ovarian materials (oocytes and granulosa cells). Molecular techniques to measure al- terations in ovarian function, such as DNA probes for follicular regulatory fac- tors, should be developed. Development and Aging Critical periods in sexual differen- tiation should be identified further. Methods of applying markers of puber- tal onset in epidemiologic studies should be evaluated. Menstrual cycle-length changes as indicators of maturation and aging should be assessed. Oocyte depletion should be assessed (perhaps through the use of gonadotropins, imaging analysis, and inhibin assays) as a potential indicator of exposure to xeno- biotics. Neurotransmitters in cerebrospinal fluid could be measured in relation to reproductive changes. Hypothalamic dysfunction should be assessed in field studies (perhaps through the use of gonadotropin hormones). Cyclic Ovarian Function The use of urinary progesterone metab- olite assays as markers of ovulation in exposed populations should be encouraged. Assays of salivary progesterone and other steroid hormones need validation. Gonadotropin assays with increased sensitivity and the ability to measure bioactivity should be developed. Data on cycle regularity should be correlated with data on markers of ovula- tory function. Endometrialdevelopment in the preim- plantation period should be assessed (perhaps through the use of uterine wash- ings). Fertilization, Implantation, and Early Loss The role of tubal motility in gamete and conceptus transport needs to be evaluated.

OCR for page 1
EXECUTIVE SUAfMARY hCG assay methods (including, but not limited to, methods based on radio- activity) should be further refined to reduce laboratory time, and quantities of urine and antibodies required and thus maximize utility in field studies. Markers ofpregnancyin the preimplan- tation period need to be developed. New reproductive technologies to evaluate relationships among follicular fluid, blood, and tissue concentrations of pollutants, and fertilization, early embryo transplant, implantation (as meas- ured by hCG production), clinically ap- parent pregnancy, and pregnancy outcome need to be studied. In viva and in vitro fertilization assessments should be extended to oocyte function (both oocyte fertilizability and control of post-fusion events). BIOLOGIC MARKERS ASSOCIATED WITH PREGNANCY Identifying biologic markers of toxic exposures that take place during pregnancy is a multidisciplinary challenge that requires integration of basic and clinical sciences. We need to learn how an adverse state (such as premature labor and delivery or spine bifida) is produced and how a xeno- biotic agent can influence biologic proc- esses to produce such a state. When persons in their reproductive years are exposed to a xenobiotic agent that is a teratogen, an important concern should be whether they intend to conceive a child in the near future or whether the woman is already pregnant. Various periods of fetal development are differentially sensitive to insult. They include the particularly vulnerable times before and around implantation and the times of development of various organ systems throughout gestation, during which exposure to xenobiotic agents can cause specific functional and structural impairment. The critical postimplanta- tion windows of developmental sensitivity have been associated with specific de- fects, such as phocomelia from the maternal ingestion of thalidomide and reproductive tract anomalies from the maternal inges- tion of diethylstilbestrol. As our knowl- 9 edge base increases, it is important to reevaluate our understanding of critical periods for functional and structural development. Not only the conceptus is at risk. The mother's health can be compromised by preg- nancy itself and by exposure to a xenobiot- ic. A detailed clinical history (genetic, reproductive, and menstrual) is essen- tial, not only to obtain an exact chronol- ogy of the pregnancy, but also to determine potential risks for both mother and concep- tus. The timing of known exposures must be established, including the time of the most recent exposure. All the above infor- mation is used to interpret markers of exposure, which could be reflected in the concentrations of xenobiotic agents and their metabolites in maternal blood. , urine, hair, or fat at various times during gestation. Each gestational stage is associated with its own problems of assessment and physiologic evaluation. Inaccessibility of fetal and placental tissues and fluids is one of the most important difficulties, especially in the embryonic stage of a continuing pregnancy. Because of that inaccessibility, much effort has been expended on the evaluation of more readily available fluids and tissues from the moth- er, especially before 8 weeks of gestation. In the fetal period, more invasive tech- niques have been used to demonstrate (usu- ally after the fact) damage to the concep- tus. A goal for the study of markers is to discover a point at which therapy might reverse or prevent damage to the various fetal organ systems. Markers of exposure and effect at three periods during pregnancy have been evalu- ated: around implantation, during organo- genesis, and during the fetal and neonatal periods. The current application of biologic markers during the pert-implantation period includes the use of sensitive hCG assays to document implantation. Other markers should be developed to document the differentiation of the inner cell mass to the embryo and of the trophectoderm to the fetal placenta. Early pregnancy factor assays and immunologic assays also are potentially useful in limited studies to

OCR for page 1
10 identify women at risk of spontaneous abor- tion. It is hoped that results of those tests will be integrated into the set of pert-implantation markers to document some of the mechanisms of spontaneous abor- tion. hCG concentrations -widely used to document normal, ectopic, and failing pregnancies-are not particularly useful until after implantation. The specificity of several early pregnancy factors needs to be evaluated. Concentration of hCG continues to be a useful marker during organogenesis to document pregnancy. Sonography has pro- vided noninvasive measures of growth, location, size, and movement of the concep- tus; increasing resolution will provide additional information on embryonic or- ganization. Although these markers are nonspecific and insensitive to physiolog- ic changes, they are the only biologic markers that now can be used to assess re- productive toxicity to the conceptus dur- . ng organogenesls. Many more biologic markers have been used routinely during the fetal period after 8 weeks of gestation. However, they do not assess variation specific for particular xenobiotic exposures. Sonography can be used to detect dysmor- phisms (such as limb defects, anencephaly, and renal agenesis), central nervous sys- tem (CNS) function, and fetal chest-wall movement. Other kinds of biophysical monitoring can be used to assess fetal cardiovascular function, e.g., Doppler monitoring for uterine blood flow velo- city, umbilical blood flow velocity, and fetal cardiac function. Plasma markers, such as concentrations of human placental lactogen and alpha-fetoprotein may be used to assess placental growth and the poten- tial for neural tube defects, respective- ly. Invasive tests, such as amniocentesis and chorionic villus sampling, yield addi- tional substantive information concerning genetics (through karyotyping), exposure (through measurement of xenobiotic agents with biologic assays or direct analysis), and markers of effect (e.g., through meas- urement of alpha-fetoprotein or cholines- terase in the amniotic fluid to document a neural tube defect). The continued use of specific assays of selected organs is BIOLOGIC MARKERS encouraged, although they are not agent- ~ SpeC11 1C. Thus, important biologic markers re- lated to pregnancy have not been linked with exposures to specific xenobiotic agents. The only biologic marker that has diagnostic utility after exposure to a xenobiotic agent is the concentration of alpha-fetoprotein in maternal serum and amniotic fluid after exposure to the drug valproate, for which increased concentra- tion of alpha-fetoprotein is evidence of spine bifida in the fetus resulting from the exposure. Research is needed on the following im- portant subjects: Knowledge on endometrial-trophoblas- tic attachment in animals should be extend- ed to humans, to identify measurable mark- ers of implantation and placentation. The use of products of conception obtained after spontaneous or induced abortion to document exposure should be developed. Embryonic or fetal tissues should be analyzed for genetic or environmental factors that alter development. Fluorescence-activated flow cyto- metry to isolate embryonic (fetal) cells in maternal circulation at various criti- cal periods of gestation should be inves- tigated, to determine the early impact of environmental exposure without inva- sive techniques. The utility and risk of new noninvasive means of detecting defects-such as mag- netic resonance imaging-should be deter- mined. The application of current molecular biologic techniques, such as use of DNA probes or detection of DNA adducts, to assess xenobiotic interactions with the conceptus should be studied. (Some DNA adducts, for example, appear to provide specific fingerprints of exposure in se- lected fetal tissues in animals and in human placenta; other adducts are not spe- cific, but reflect exposure to general classes of toxicants. Molecular genetic technology combined with earlier chorion- ic villus sampling might provide powerful assessments not only of exposure, but also of effect, by examining enzyme induction,

OCR for page 1
EXECUTIVE SUMMARY genetic abnormalities, and general metab- olic function of embryonic tissue, es- pecially the trophoblast.) The application of new testing proce- dures during the first 8 weeks of gestation should be studied. (In spite of current limitations, assessment of pregnancies beyond 8 weeks has been substantially im- proved; however, many new techniques might be especially applicable during the first 8 weeks of gestation, a period that lacks thorough evaluation.) Detailed epidemiologic data bases with exacting quality-assurance standards need to be developed. Markers related to pregnancy that can be assessed using or with in vitro stud- ies of human material or in animals should be investigated. The use of chorionic villus sampling material for cytogenetic research should be investigated. BIOLOGIC MARKERS ASSOCIATED WITH NEURODEVELOPMENT Neurodevelopmental toxicity includes any detrimental neurological effect pro- duced by exposures during embryonic or fetal stages of development. Complex in- teractions of lifestyles, pre- and post- natal environmental factors, education, peers, and social class all profoundly effect the ultimate development and well being of children. To decrease the per- sonal and social burdens of reproductive losses, developmental impairment, and mental retardation, scientific progress must be made in unraveling the complex- ities of development, and articulating the factors that affect it. At present, even careful neuropathologic studies cannot account for most causes of neurolog- ic dysfunction. Although a few pathologic correlates of mental retardation have been identified, they seldom have been studied systematically. The interactions among neurodevel- opmental, ecologic, medical, and socio- political factors are complex and not well understood. Little information has been published on the effect of particular environmental chemicals on reproduction or development in experimental animals, 11 and less is available on effects in humans. Where data are available on experimental animals, questions abound concerning extrapolation of results across species or from high to low doses. The process of development affects the absorption, distribution, metabolism, excretion, and effects of toxicants. Age- dependent changes in body composition and xenobiotic clearance have profound ef- fects on interpretation of markers of ab- sorbed dose and of biologic effect. Unique physiologic variables in the new- born and during puberty also can alter markers quantitatively and qualitatively. For compounds with very long half- lives (e.g., polychlorinated biphenyls), effects of in utero exposure noted later in life might be due either to persistence of the xenobiotic or to effects secondary to physiologic changes that occur during earlier critical periods of development. Such considerations have important im- plications for potential therapeutic interventions designed to alter xenobiot- ic clearance or otherwise modify toxic outcome. Establishing causal relationships in neurotoxicology requires both experi- mental and epidemiologic studies; neither is sufficient in itself. Studies of animal responsiveness are needed to validate epidemiologic data, to determine mechan- isms, and to establish dose-response rela- tionships. Three approaches to the appli- cation of animal data to human observations are used: investigation of the underlying mechanisms of functional changes observed in animals, examination of comparable end points in humans and animals and a search for homologous mechanisms. In this report, the CNS is discussed as a model of developing biologic systems. The CNS develops through multiple morpho- genetic processes, including cell death and cell migration. Two toxic agents ionizing radiation and lead-have been shown to influence cell death and cell migration and show dose and stage specifi- city with respect to particular biologic end points. If granule cells of the cere- bellum or hippocampus are killed before their final migration, the deficit in cell number produces highly specific behavior-

OCR for page 1
12 al aberrations, especially hyperactivity. Therefore, these behaviors serve as bio- logic markers that are correlated with lead and irradiation exposures. Minor physical abnormalities (MPAs), which are primarily of ectodermal origin, have been associated with various behav- ioral aberrations. Increased MPAs are found among schizophrenic patients, au- tistic children, and boys. Boys with mul- tiple MPAs tend to be hyperactive; girls tend to be behaviorally inhibited and in- tractable. A dose-dependent relationship between lead concentration in the umbili- cal cord blood and MPAs has been reported. And an inverse association between number of MPAs and verbal IQ score has been re- ported. These findings suggest that the presence of MPA indicates impaired CNS development. Neuronal communication is determined by neurochemical factors. Most of the neurotransmitters released in the CNS are unknown or poorly characterized. In addi- tion to their role as chemical messengers, many neurotransmitters have a role in CNS development. Access to CNS neurotransmit- ters is constrained; only a small portion of neurotransmitters measured peripheral- ly in the blood are related to CNS activity. Cerebrospinal fluid analysis offers a closer view of CNS metabolism, but this material is not readily available. Few data are available on neurochemical markers of toxicant exposure. In children with lead intoxication, high 24-hour ex- cretion of homovanillic acid has been re- ported. Experimental administration of N-methyl-4-phenyltetrahydropyridine (MPTP), an agent known to produce a syn- drome similar to parkinsonism, causes decreased excretion of dopamine metabo- lites. Investigators of behavioral neurotoxi- city can use a broad repertoire of behav- ioral assessments in evaluating subjects exposed to toxicants-from measures of psychophysical functions (such as reac- tion time, hearing thresholds, and nerve conduction time) to measures of more com- plex functions (such as psychometric in- telligence, visual motor integration, and social behavior). Research is needed on the following im- portant subjects: BIOLOGIC AfAR~ERS Neurotoxic exposures should be cor- related with behavioral measures and with physiologic assessments that use such techniques as magnetic resonance imaging, positron emission tomography (PET), and brainstem evoked potentials. The specificity and sensitivity of markers of exposure and effect need to be estimated. The use of such tissues as cerebrospin- al fluid and CNS cells to correlate expo- sure and effect markers with tissue con- centrations of neurochemicals needs to be studied. Interactions between neuroendocrine and neuroimmune functions and their rela- tion to pollutant exposure and effect should be examined. Integrated assessments of exposure and outcome are needed for estimating toxic effects on development. The use of early developmental markers of exposure and effect needs to be adapted to the study of brain plasticity and the prediction of later effects. Relationships among minor physical abnormalities, toxicant exposure, and behavioral effects should be established. GENERAL RECOMMENDATIONS AND CONCLUSIONS Assessment of reproductive and develop- mental health is a task that cuts across several scientific disciplines and inter- national boundaries. In the United States, this complex field comes under the purview of several federal agencies with respon- sibilities for regulation, research, or monitoring. It is not within the scope or expertise of this committee to evaluate the administration of scientific re- search, but broad international collabor- ation in addressing the scientific issues related to environmental hazards to repro- duction probably would be advantageous and appropriate. International collabor- ations could increase the statistical power of particular studies by combining data and diversifying the assessments performed, as well as providing a broader base of funding and permitting interna- tional health-registry data to be used. Several structures for international scientific cooperation have been success-

OCR for page 1
EXECUTIVE SU~fM~4RY ful (e.g., the World Health Organization and the International Agency for Research on Cancer), and they might be considered as a means for coordinating and promoting research in reproductive and developmen- tal toxicology through multidisciplinary symposia, research, and public education. Another substantial need in reproduc- tive and developmental toxicology is for normative data and better methods to follow the developmental and reproductive health of specific persons throughout their lives. Scandinavian countries have as- signed a personal identification number (similar to a Social Security number) at birth and use it on hospital, employment, and other records. This has allowed the development of data bases that can be used to explore relationships among environ- mental or occupational exposures and many facets of reproductive and developmental health. This approach might not be accept- able in the United States, but alternative approaches, such as exposure registries and selective pre-employment sampling, should be explored. Collection of data on individual and population exposure to xenobiotics and health outcomes is vital. Major issues of ethics, confidentiality, and impact on the legal system and insurance industry must be addressed, and interdisciplinary discussion of such issues is urgent. Consideration should be given to perfor- mance of more detailed cross-sectional and longitudinal studies of reproduction and development. Models for such studies include the National Health and Nutrition Examination Survey and the Collaborative Perinatal Project in the United States and the Birthday Trust in the United King- dom. Lifetime studies of a selected cohort should be considered for the collection of data on the development, expression of mature function, and performance of the reproductive system. It is important to remember that nonin- vasive techniques might themselves have reproductive or developmental effects, that physiologic changes in pregnancy might alter maternal toxic responses, and that existing invasive diagnostic or ther- apeutic techniques (e.g., amniocentesis, follicular aspiration, chorionic villus biopsy, and therapeutic abortion) might 13 provide unique materials for the develop- ment and validation of biologic markers. The use of batteries of tests for popu- lation investigations is encouraged. Given the complex multiorgan nature of reproduction and development, the use of tests to assess only a single facet of re- productive or developmental health is likely to fail more often than succeed in the identification of reproductive or developmental toxicants. A series of general guidelines warrant attention when a previously identified, potentially useful biologic marker is to be validated: Establish normal baseline values and distribution for the marker in labora- tory animals and humans. Evaluate the sensitivity and specifi- city of the markers in predicting a health outcome (e.g., infertility) or genetic damage. Understand in detail the time course of response of the marker to a toxic chemi- cal, with special attention to the recovery process. Develop a strategy for and a consensus on the use of multiple species in toxico- logic studies. Develop human assays that use semen, saliva, or urine, rather than tissue or blood, whenever possible. Use noninvasive techniques, such as ultrasound or magnetic resonance imag- ing, whenever possible. Consider a battery of markers that reflects a wide array of physiologic func- tions and genetic damage and then relate the marker in question to others in the battery. Identify populations at high risk for reproductive or developmental health impairment (perhaps populations exposed to drugs with reproductive or developmen- tal toxicity, aging populations, or off- spring of women treated with specific drugs during pregnancy), to serve as test sub- jects for the initial assessment and vali- dation of biologic markers. Include among high-exposure popula- tions those with special or unique occupa- tional exposures (e.g., agricultural groups).

OCR for page 1
14 Encourage and support institutions in the development of sample banks, to speed the identification and validation of markers. Establish a task force to develop and coordinate strategy. Few areas have changed in medicine as rapidly as those that are the subject of this report. The problems of reproductive and developmental toxicology remain com- plex, because advances in our understand- ing of toxic chemicals have not kept pace with the introduction of new materials BIOLOGIC MARS into the environment. A significant pro- portion of human reproductive wastage remains of unknown etiology. Important research opportunities identified in this volume likely will provide significant improvements in promoting reproductive and developmental health. The subcommit- tee recommends specific research be pur- sued as outlined above and offers one gen- eral observation: those concerned with the complex ethical questions of protect- ing life at its origin need to appreciate the importance of scientific research in furthering their aims.