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 175
--> 7 Risks to Reproduction and Offspring The preceding chapters have identified scientific, social, ethical, and legal issues that would need to be taken into account in implementing the committee's three fundamental principles of justice with regard to the inclusion of women in clinical studies (see Chapter 3). This chapter illustrates these issues, and potential mechanisms for resolving them, with regard to a particular subpopulation-subjects of reproductive age-in a particular category of clinical studies-clinical trials of drugs. Historically, concern for these risks has focused on women of reproductive potential, including pregnant and lactating women. Nevertheless, the possibility that certain drugs pose unique risks to the male reproductive system may also merit attention. People of reproductive age get sick and take medications, and drugs intended for use by this population should therefore be tested in this population. Some of these drugs, however, have potential risks to reproduction or for the development of offspring. These risks give added importance to informed consent and contraceptive options. Risk assessment for reproductive and developmental toxicity is complicated by the high background rates of infertility and birth defects, as well as the difficulty of identifying the specific effects of the drug under investigation. Techniques such as animal studies, in vitro analysis, and surveillance for developmental effects, among others, can provide some information on potential hazards to humans. Once the potential reproductive and developmental hazards of participation in a clinical study have been identified, investigators can attempt to design trials to minimize these hazards. In some cases, hazards may be
OCR for page 176
--> altogether avoidable; in others, it may be decided that hazards cannot be effectively minimized and that the trial should not proceed. The process of weighing potential risks against potential benefits is a complicated process with many players, including the investigator who proposes a study, the institutional review board (IRB) that assures that the protocol is consistent with human subjects regulations, and the potential study participants. This chapter describes some of the factors-such as toxicity data, subjects' understanding of risks, and contraception-that investigators must consider when they wish to study a drug in populations of reproductive age. It also describes the process of making risk-benefit assessments and the values that different people place on certain risks and benefits. The chapter concludes with the committee's recommendations for the conduct of research, including drug trials in persons of reproductive age and in lactating and pregnant women, and a discussion of the policy implications of these recommendations. SCIENTIFIC ISSUES: RISK ASSESSMENT For men and women of reproductive age, and for pregnant and lactating women, there are risks associated with taking experimental and nonexperimental drugs. In many cases, these risks are virtually impossible to detect before the drug is in widespread use because of the inherent limitations of animal studies and clinical trials. Nevertheless, concerns about reproductive and developmental toxicity do not override the need to improve the medical management of these populations. Our understanding of treatment options for all of these groups will only be advanced by their inclusion in clinical studies and by more systematic collection of empirical data on reproductive and developmental outcomes. Identifying Reproductive and Developmental Toxicants Reproductive and developmental toxicants are physical, chemical, and biological agents that produce a toxic effect in animals and/or humans. Reproductive toxicants alter fecundity, decreasing the ability or increasing the time needed to achieve pregnancy. Developmental toxicants alter the structure or function of offspring. Animal testing and clinical trials are the principal mechanisms for identifying drugs as either reproductive or developmental toxicants. The goal of these approaches is to determine the potential for reproductive and developmental toxicity prior to broader human exposure. In the absence of data from such experimentation, reports of adverse events in humans are also gathered in registries in an attempt to identify unrecognized toxicity. In addition, newborn infants are frequently
OCR for page 177
--> screened for birth defects so that drugs and other agents with developmental toxicity might be identified. To determine whether or not a drug is a potential human reproductive or developmental toxicant requires two steps: (1) hazard identification and (2) hazard characterization. Characterization of the magnitude of the risk for adverse outcome requires two additional steps: (3) exposure assessment and (4) qualitative or quantitative risk characterization (NRC, 1983). Together, these four steps provide the framework for the assessment of risk to reproduction or offspring during a clinical trial. Hazard Identification The first step in risk assessment explores the question: Does the drug produce adverse effects on reproduction or offspring in animals or humans? Because data on the effects of drugs on human reproduction or pregnancy outcome are generally not available prior to a clinical trial, it is necessary to utilize data from animal studies or other studies, such as in vitro assays. Reproductive studies in animals are generally useful and valid for determining whether a drug represents a potential human reproductive or developmental toxicant (Wilson and Fraser, 1979; Shepard, 1983; Schardein, 1985; Jelovsek et al., 1989; Jelovsek et al., 1990). Laboratory animals and humans can differ in toxicokinetics, however, and the use of data from animals to determine health risks in humans must be assessed carefully (NRC, 1989). Hazard Characterization If a drug represents a potential hazard for reproduction or development, it is necessary to determine the dose-response relationship, site of action, and mechanisms through which the adverse effects are produced. Some drugs that appear to be reproductive or developmental toxicants in animals may not produce adverse effects in humans during a clinical trial. For example, compounds metabolized to toxicants in animals may produce different metabolites in humans, or substantially smaller amounts of the same metabolites, and therefore may not represent a hazard. The converse situation also occurs: a chemical may be non-toxic in the conventional animal assays, but be a human developmental toxicant. Fortunately, this is unusual-all known human teratogens are also teratogens in at least one animal species (Schardein, 1985). In addition, genetic differences among humans in their response to toxicants, such as that observed with hydantoin developmental toxicity (Phelan et al., 1982), may also modify risk to reproduction and development.
OCR for page 178
--> Exposure Definition A critical step in characterizing risk is to define the exposure, including such variables as dose, duration, and timing of exposure with respect to reproduction or pregnancy. For example, exposures that occur before or after the development of a susceptible tissue or organ may carry no additional risk for fetal development. In the case of diethylstilbestrol (DES), the risk of vaginal adenosis is greater than 70 percent for female offspring exposed before the ninth week of gestation but less than 10 percent for exposure after the seventeenth week (see Appendix C). Similarly, in the sexually mature male, exposures to developmental toxicants that occur long before the development of the particular sperm cell that will fertilize an egg are unlikely to affect the development of the resulting offspring, due to the fact that sperm development occurs only in the few months prior to ejaculation. Risk Characterization Finally, after a reproductive or developmental hazard has been identified and characterized, and exposure has been defined, this information is analyzed using statistical models to characterize the risk to reproduction or pregnancy outcome. In this step it is important to define the amount and quality of the data available to characterize risk, including variability and uncertainty in the risk estimate. The risk characterization should also include the assessment of the background incidence of adverse reproductive or developmental outcomes. Challenges to Identification of Reproductive and Developmental Toxicants In order to isolate and evaluate observed drug effects, it is also necessary to consider the background incidence of adverse reproductive or developmental outcomes. This incidence is often higher than many potential study subjects would suppose. For example, about 15 percent of couples trying to conceive will not have done so after one year of trying (a common definition of infertility); 20 to 30 percent of recognized pregnancies end in miscarriage; 3 to 8 percent of all babies have birth defects; and 1 percent of liveborn children are born with severe mental retardation. Most of these adverse outcomes do not have recognized causes, and very few of those with known causes are the result of exposures to chemical, physical, or biologic agents. With such a high background rate of many different adverse outcomes, identification of adverse effects imposed by a specific drug exposure can be
OCR for page 179
--> difficult. For example, consider an exposure that increases the incidence of limb reduction defects tenfold. If the baseline incidence of such defects is I in 10,000 (0.01 percent), the impact of an increase to 0.1 percent will be undetectable against a background incidence of total birth defects of about 5 percent. It is only when there is some idea of the kind of defect associated with the exposure that studies can be targeted at detecting an increase in incidence. Determining whether a drug or other treatment is associated with an adverse reproductive or developmental effect also requires characterization of the endpoint of concern. Depending on the endpoint in question, sample sizes required to detect that endpoint can vary greatly (see Table 7-1). An uncommon endpoint such as a rare congenital malformation typically requires very large study populations: a malformation that occurs in only 1 of 10,000 births may not be detected in a trial involving 5,000 couples. When exposure produces a dramatic increase in the incidence of unusual abnormalities, however, it may not take very many cases before the association is recognized. This kind of recognition occurred, for example, with the birth defects associated with thalidomide and isotretinoin, with the cerebral-palsylike illness caused by methyl mercury, and with the severe testicular toxicity of the pesticide dibromochloropropane (DBCP). Determining whether or not a drug is associated with a relatively common adverse outcome such as miscarriage may be defined with a much smaller number of subjects. Listed in Table 7-1 are some adverse reproductive and developmental outcomes of interest, the background rate of these adverse outcomes, and the sample size needed to determine a doubling of that outcome in a clinical trial. Male-Mediated Developmental Toxicity Although traditional concerns about developmental toxicity have focused on exposures (or treatment) of the female during pregnancy, scientists have long suspected that the male may also contribute to adverse pregnancy outcomes such as spontaneous abortion, stillbirth, impaired growth, and structural and functional abnormalities (Olshein and Mattison, in press). Most systematic studies of male-mediated developmental toxicity, however, have been rodent studies, and the implications are inconclusive for human males. Scientists have postulated that the occurrence of developmental toxicity following exposure or treatment of the male may depend on a number of factors, including male reproductive status (fecundity), exposure (dose, duration), properties of the agent, pharmacokinetics (especially distribution to gonads and other endocrine organs), mechanism of action, stage of spermatogenesis affected, frequency and timing of intercourse with respect to
OCR for page 180
--> TABLE 7-1 Expected Frequency of Selected Endpoints of Reproductive or Developmental Failure and the Sample Size Needed to Detect a Doubling of that Endpoint in a Clinical Trial Reproductive or Developmental Endpoint Denominator Frequency (%)a Sample Size of Treatment and Control Groups to Detect a Doubling of the Endpointb Azospermia Men 1 3,300 Men Failure to conceive after one year of unprotected intercourse Couples 10-15 (12) 230 Couples Birth weight <2,500 g Live births 5-15 (10) 286 Live births Miscarriage Pregnancies 10-20 (15) 174 Pregnancies Chromosome aberration at miscarriage Miscarriages 40-50 (45) 25 Miscarriages Late fetal deaths (¬28 weeks) Late fetal deaths + live births 1-4 (3) 1,068 Late fetal deaths + live births Total birth defects at birth Live births 2-3 (3) 1,068 Live births Chromosome aberrations at birth Live births 0.6 5,533 Live births Neural tube defects Late fetal deaths + live births 0.005-1 (0.05) 66,936 Late fetal deaths + live births Severe mental retardation Children to age 15 0.4 8,324 Children to age 15 a Where a range is given, the background rate used in the determination of sample size is shown in parentheses. b The sample size indicated is the size required of each population (i.e., both the study population and the control population will need to be at least as large as the sample size indicated) to detect a doubling in the rate of the indicated endpoint. In these calculations it is assumed that the control and study populations are the same size, and that the investigator wants a 90 percent chance to detect a statistically significant difference in the study and control populations at a p ≤ 0.05. The calculations were performed with the sample size estimation modules for an unmatched comparison of proportions as implemented in True Epistat 4.0, Epistat Services, Richardson, Texas (1991). The function for the sample size calculation for unpaired comparison of proportions is described on page 399 of Zar, 1984.
OCR for page 181
--> TABLE 7-2 Adverse Developmental Effects that May Occur Following Male Exposure Reproductive Process Toxicological Effect of Male Exposure Endpoint Observed in Epidemiological Studies Conception Preimplantation loss Decreased fertility Increased time to pregnancy Implantation Postimplantation loss Increased spontaneous abortion Embryo development Failure/disruption of embryonic development Increased spontaneous abortion Increased fetal death Increased malformation Increased growth retardation Increased functional deficit Increased premature birth Fetal growth and development Failure/disruption of fetal growth and development Increased fetal death Increased malformation Increased growth retardation Increased premature birth Increased functional deficit exposure, and maternal reproductive characteristics (fecundity). In addition, the type of endpoint that is observed is likely to vary with the process affected in the male (see Table 7-2). Studies conducted over the past five decades have identified three potential mechanisms of male-mediated developmental toxicity: Genetic: damage to the genetic material contained in the sperm through the creation of a mutation or chromosomal abnormality. Epigenetic: damage to processes that control the expression of the paternal genes after fertilization. Transport of toxicant: transport of an agent through the ejaculate during postconception sexual intercourse and subsequent exposure of the conceptus, embryo, or fetus. Although evidence is inconclusive concerning the role of the male in developmental toxicity, the possibility that he has a role provides reason enough for investigators to consider including discussion of developmental toxicity in the informed consent process for male subjects who may be exposed to developmental toxicants in the course of a clinical study. As discussed below, the provision of advice about contraceptive options may also be wise.
OCR for page 182
--> Evaluating Drugs for Use in Men and Women of Reproductive Age Women of reproductive age (defined as 15 to 44 years of age) constitute a significant proportion of the population. Men of reproductive age constitute an even greater proportion, because the window of potential fertility is much larger for males. Not surprisingly, members of this large segment of the population experience a myriad of diseases requiring medical treatment, and drugs intended for use in this population should be tested in this population. As discussed in earlier chapters, federal policies and practices designed to protect women of reproductive potential from risks associated with experimental treatment have hindered the collection of information about drug effects in this subpopulation. Investigators designing clinical studies need to be particularly concerned about the potential reproductive and developmental toxicity of the compounds they wish to study in this population. When a trial involves the exposure to a potential toxicant, issues of risk characterization, informed consent, and contraceptive options are paramount. Because people tend to overestimate the risks of harm to reproduction and development that are posed by drugs (Koren et al., 1989, 1990), enhanced subject education efforts may be required. Other characteristics of the population of reproductive age, such as use of hormonal contraceptives, must also be taken into account during trial design and recruitment. Discussed in greater detail below are the kinds of variables that investigators must consider when deciding how best to test new drugs in men and women of reproductive age. Use of Toxicity Testing Data If information is available on the reproductive and developmental effects of a given drug in animals, investigators must make several types of inferences in order to protect the reproductive and developmental health of their subjects (see Figure 7-1). First, are the animal data relevant to humans? Animal testing for developmental toxicants is generally relevant for human hazard identification (Jelovsek et al., 1989; Francis et al., 1990). While animal data may be less reliable as sources of information about human reproductive toxicity, these data have been used successfully to provide presumptive evidence of human reproductive toxicity (NRC, 1989). Next, it is necessary to determine how to utilize the dose-response information. For example, investigators must consider whether characterization of a benchmark dose, and subsequent calculation of a reference dose, with safety or uncertainty factors, is adequate to protect human health (Barnes and Dourson, 1986; Gaylor, 1989; Kimmel, 1990; Gaylor, 1991). Alternative interpretations must also be considered: in some cases, it may be more appropriate to assume a linear, nonthreshold, low-dose relationship
OCR for page 183
--> FIGURE 7-1 Considerations in the protection of reproductive and developmental health of research subjects.
OCR for page 184
--> between the chemical and the reproductive or developmental effect (Meistrich and Brown, 1983; Rai and Van Ryzin, 1985; Kimmel and Gaylor, 1988; Meistrich, 1989a,b; Kimmel, 1990; Mattison, 1991; Meistrich and Mattison, in press). Information from animal reproductive tests can be supplemented by other information. For example, data on the nonreproductive toxicity of an agent often provides clues about possible reproductive effects. If an agent is known to be toxic to the nervous system in adults, investigators testing this agent in pregnant animals know to examine the offspring for behavioral function as well as for gross birth defects. Investigators may also utilize what is known about the reproductive effects of certain agents to postulate the effects of related chemicals. In the rare situation when there is no information on the reproductive and developmental effects of the drug being tested, other approaches may be taken to protect the health of the subject. For example, data from in vitro testing or structure-activity relationships may provide useful information for counseling subjects in a clinical trial. Recent data using expert systems to explore the relationship between chemical structure and developmental toxicity suggest that a substantial amount of useful information could be extracted from existing developmental toxicity data sets (Takihi et al., in press). Finally, when available, human data on similar agents or classes of agents can be coupled with animal data to develop an overall assessment of an agent's potential for toxicity. In the case of fluoxetine, an antidepressant, the manufacturer collected several hundred cases of exposed pregnancies in which use of the drug was not associated with an increase in birth defects. This sort of evidence is reassuring, but by no means conclusive-an increase in birth defects of the kind and magnitude associated with, for example, valproic acid might not be detected with this number of patients. Arguments for the safety of fluoxetine for use in pregnant women are further strengthened, however, by the observation that rats and rabbits do not show an increase in developmental anomalies when exposed to the drug during pregnancy in standard protocols. Subjects' Understanding of Developmental and Reproductive Risks If men and women of reproductive potential are included in a trial in which they will be exposed to a potential reproductive or developmental toxicant, the potential risks must be characterized as accurately as possible so they can make an informed decision about whether or not to participate. If they decide to participate, they may also wish to consider measures to prevent pregnancy (see below). Finally, information about toxicity risk can help participants determine the likelihood that the baseline incidence of
OCR for page 185
--> adverse pregnancy outcome will have been increased by study participation, should a pregnancy occur during the trial. In this regard, the use of reproductive and developmental toxicity information is similar to the use of information about toxicity to other organs and organ systems. Just as investigational agents can be evaluated in preclinical test phases for reproductive and developmental effects, they can be evaluated for carcinogenicity and for toxicity to liver, kidney, and bone marrow. Results from these evaluations are incorporated into the informed consent process, just as reproductive and developmental toxicity information is incorporated. Contraception in Clinical Trials When considering participation in clinical trials of agents with potential reproductive or developmental toxicity, men and women of reproductive age need to be made aware of the contraceptive options available to them as study participants and of what is known about the effectiveness of each method. Table 7-3 lists low and high reported failure rates for the range of contraceptive methods currently available in this country. The investigator must be concerned with contraception for two major reasons. First, hormonal contraceptives, widely used by women in this country, may introduce complexities into the evaluation of drug effects (see Chapter 4). Second, demographic characteristics of a population can play a role in contraceptive use by that population, with implications for recruitment and the informed consent process. An investigator may have difficulty assessing the independent effect of a drug in a woman who is using hormonal contraceptives, which can alter the pharmacokinetics (absorption, distribution, metabolism, and excretion) or pharmacodynamics (mechanism of action) of the drug under study. Accurate evaluation of drug effects can also be complicated by the need to clearly isolate the specific side effects of the hormonal contraceptive from those of the drug under study. Where scientific validity or contraceptive effectiveness may be compromised, investigators might consider excluding women using hormonal contraceptives, or recommending that an alternate, nonhormonal method of contraception be used. There are important differences among population subgroups in the use of contraception, and contraceptive failure rates tend to vary according to age, education, race, ethnicity, marital status, and socioeconomic status. For example, with increasing age, contraceptive failure rates decline as a result of decreased frequency of intercourse and decreased ability to conceive. Non-use of contraceptives is higher among the unmarried, poor, and less educated—a phenomenon that may be associated with access to contraceptive services (Planned Parenthood Federation of America, 1991). Cultural
OCR for page 192
--> Women's values can differ significantly from those of scientists (whether male or female) in assessing risk-benefit ratios. For example, women's health advocates tend to define the "safety" of contraceptive methods in terms different from those typically employed by biomedical scientists. According to one report: Scientists' concern is to establish safety of methods according to specific measurable parameters. They assess toxicity, first in animals and then in carefully controlled studies in human volunteers. Subsequent studies address efficacy and short- to medium-term safety. . . Women's health advocates . . . give more priority to methods that have fewer side effects and that protect against sexually transmitted diseases and their consequences such as infertility. While scientists have tended to give priority to methods which minimize users' control, women's health advocates prefer methods controlled by the users. [World Health Organization, 1991:11.] New policies to encourage the inclusion of women, and in particular women of childbearing age, in clinical studies evidence a greater acknowledgment of individual values and a respect of individual autonomy (see, for example, Merkatz et al., 1993). These policies will affect the responsibilities of IRBs and potential participants. The changes will be most evident in the communication of risks to participants and in IRB risk-benefit assessments. As with any potential participant, a thorough discussion of the risks and potential benefits of participation is a prerequisite for an individual's ability to make an informed decision to enroll in a clinical study. For men and women of reproductive age, reproductive issues affect the type of information included in the informed consent process. It will be the IRBs' obligation, as with all research involving presumptively competent adults, to continue to ensure that: (1) the selection of potential participants is fair; (2) the informed consent process is adequate; and (3) the risks to participants are outweighed by the potential benefits. This first duty-fair selection-is the subject of the committee's report and thus requires no additional comment. The other two duties will be discussed below. Women (Not Pregnant or Lactating) and Men of Reproductive Age Significant changes have occurred during the committee's tenure, in policies that govern the inclusion of women of childbearing potential in clinical studies, particularly studies of FDA-regulated products (see Chapter 6). FDA issued new guidelines permitting the participation of women of childbearing potential in the early phases of drug trials, and offered three reasons for this decision: (1) scientific gains in study design related to the early identification of gender differences in trials; (2) the ability to reduce
OCR for page 193
--> the risk of fetal exposure through protocol design; and (3) recent social changes indicating respect for women's autonomy and decisionmaking capacity in reproductive issues. NIH guidelines are currently under revision. These policy changes should have the effect of including more women of reproductive age in clinical studies, with implications for risk-benefit assessments. In a study that poses risks to potential offspring, women who are not pregnant at the outset of the investigation may become pregnant while they are still participants. The committee believes that the informed consent process for these women should include information about contraception and the alternatives of voluntarily withdrawing from the study and terminating a pregnancy should conception take place. Similar discussions should be held with men who could father a child while participating in the study. As in all research involving human participants, every effort should be made to ensure that the consent decision is fully voluntary. An example of language for consent forms proposed by Moreno (1994) in his presentation to the committee appears below, as modified by the committee: It is possible that your participation in this study may cause damage to children if you choose to have them. You have already been told what is known about this possibility, and you are encouraged to ask further questions. (Include as appropriate: We urge you or your partner not to become pregnant while you are part of this study.) You may want to discuss this with others before you agree to take part in this study. If you wish, we will arrange for a doctor, nurse or counselor who is not part of this study to discuss this possibility with you and anyone else you want to have present. The committee recommends that investigators and IRBs not exclude persons of reproductive age from participation in clinical studies. In the case of women of reproductive age, the potential or prospect of becoming pregnant during the study may not be used as a justification for precluding or limiting participation. Risks to the reproductive system should be considered in the same manner as risks to other organ systems. Risks to possible offspring of both men and women who are not pregnant or lactating should not be considered in the risk-benefit calculation. It is the responsibility of investigators and IRBs to assure that the informed consent process include an adequate discussion of risks to reproduction and potential offspring, including, where appropriate, an adequate discussion of relevant considerations of birth control. The committee recommends that the participant be permitted to select voluntarily the contraceptive method of his or her choice where there are no relevant study-dependent, scientific reasons to require
OCR for page 194
--> the exclusion of use of certain contraceptives (e.g., drug interaction). The committee recommends that pregnancy termination options be discussed as part of the consent process in clinical studies that pose unknown or foreseeable risks to potential offspring. Lactating Women The possible transmission of drugs to nursing infants is a risk that must be considered when including lactating women in clinical studies. This additional consideration must be thoroughly discussed in the informed consent process. The committee recommends that investigators and IRBs not exclude women who are lactating from participation in clinical studies. It is the responsibility of investigators and IRBs to ensure that the informed consent process includes, wherever appropriate, an advisory to potential participants that there may be special risks to their children if nursing mothers participate. No nursing mother should be permitted to agree to participate without first receiving additional information about these special risks. Pregnant Women As reflected in the recommendation presented earlier in this chapter, the committee wishes to encourage clinical research to advance the medical management of pregnant women who are or may become ill. In this context, the committee reviewed the current DHHS regulations concerning the involvement of pregnant women as research subjects. The committee's review was limited to situations in which the pregnant woman is the subject of the research (see Chapter 6). It did not include situations in which the fetus is the subject of the research (currently covered by the same regulation); fetal research was outside of the committee's charge. The DHHS regulations begin with a presumption of exclusion-that is, "no pregnant woman may be a research subject" except under certain conditions. The regulations also require that IRBs ensure during their review of research protocols that the exclusionary standard enunciated in the regulations is met. In addition, the regulations classify pregnant women as a "vulnerable population" deserving of special protection. For the reasons discussed below, the committee concluded that the current regulatory scheme should be revised. The committee acknowledges that the current regulations (45 C.F.R. 46
OCR for page 195
--> Subpart B) may reflect an inadvertent attribution of the vulnerability of the fetus (which obviously lacks autonomy) to the pregnant woman. Nonetheless, it is inappropriate for the regulations to retain a presumption of exclusion on the basis that pregnant women are a "vulnerable population" in need of special protection. In this context, "vulnerable" suggests that pregnant women are less autonomous or more easily exploited than other persons an inference that the committee has found no evidence to support. The labeling of pregnant women as a vulnerable population also might be viewed as suggesting that they cannot weigh the risks to a fetus or potential child in deciding whether to enroll in a clinical study; that pregnant women do not care sufficiently about the health or well-being of their future children to make sound decisions; and that the prevention of all potentially harmful outcomes of pregnancy is a goal that warrants governmental, regulatory, or other official intervention into the lives and free choices of women. The committee rejects these inferences as well. Removal of pregnant women from the regulatory category of "vulnerable" potential subjects would avoid any possible inference that pregnant women are less capable of making informed decisions by virtue of their pregnancy, than are other potential research participants. For all potential research participants, risk-benefit assessment is a complex and difficult task. Nevertheless, it is no more difficult for pregnant women than it is for nonpregnant women or for men. Virtually all women desire healthy infants, even when their pregnancies are unplanned. While occasionally there may be pregnant women who are incapable of acting in the interests of their future children, it would be inappropriate to base a public policy on an atypical case, rather than a normative case. There also is little public support for the proposition that the prevention of all potentially harmful outcomes of pregnancy is a goal that warrants governmental, regulatory, or other official intervention into the lives and free choices of women. Pregnant women may choose to work in stressful jobs, engage in recreational activities, drive automobiles, and do other things that could place their own or their fetuses' health or life in jeopardy. The committee recommends that pregnant women be presumed to be eligible for participation in clinical studies. It is the responsibility of investigators and IRBs to ensure that pregnant women are provided with adequate information about the risks and benefits to themselves, their pregnancies and their potential offspring. Even when evidence concerning risks is unknown or ambiguous,1 the decision about acceptability of risk to the pregnancy or to offspring should be made by the woman as part of the informed consent process.
OCR for page 196
--> The committee was unanimous in the view that pregnant women should be presumed to be eligible for participation in clinical studies. The committee also unanimously endorsed the importance of recognizing, in public policy as well as in the deliberations of IRBs and investigators, that pregnant women should be treated as competent adults capable of making their own decisions about participation in research. It should be emphasized that the committee is not recommending that NIH impose an affirmative obligation on investigators to recruit pregnant women into every clinical study. What follows is further explication of the committee's intent with respect to the implementation of this recommendation. Adequate Information With respect to the obligation to ensure that pregnant women are provided with adequate information about the risks and benefits to their pregnancy and potential offspring, the committee recommends the following strengthened informed consent procedure. The disclosure statement of consent forms for all studies that pose a risk to pregnancy or potential offspring should include, highlighted in bold type, a statement such as: If you are pregnant or contemplating pregnancy, we urge you to consult your obstetrical care provider before deciding about participation in this study. Participation in this study may (does) pose a risk of (significant) harm to your pregnancy and/or your potential baby. Investigators should ask all potential participants if they are pregnant as part of the initial screening phase of recruitment. If a woman is pregnant, her attention should be drawn to this bolded statement. This process should include a special disclosure statement that details in easily understood lay language what is known about the risks and potential benefits to her pregnancy and potential offspring, resulting from participation in the study. This statement should be reviewed with the pregnant woman, and she should be encouraged to consult with her obstetrical care provider before proceeding further in the consent process. It is important for a pregnant woman to have benefit of the advice of her obstetrical care provider in deciding whether to participate in a study. (In the case where the woman's own obstetrical care provider is the study investigator, the pregnant woman should be offered the opportunity to discuss her participation with a similarly qualified individual who is not associated with the study.) If the pregnant woman does not wish to consult with her obstetrical care provider, and even if she has had such a consultation, specific procedures should be instituted to ensure that she understands the relevant risks and benefits. For example, the potential participant could be asked to describe in her own words what the risks and benefits are. It should be clear that the pregnant woman understands
OCR for page 197
--> that no drug or other intervention can improve on normal pregnancy in a healthy woman. Alternatively, the pregnant woman could be asked to complete a knowledge test. Deficiencies in understanding discovered through either method should be addressed through continued discussion and education. Only after the woman demonstrates an adequate understanding should consent be solicited. These are procedures that are generally advocated to improve the quality and the meaningfulness of the informed consent process (Faden and Beauchamp, 1986; Appelbaum et al., 1987). They are particularly important when the stakes associated with participation are high, as is the case for pregnant women if participation entails significant risks to pregnancy or potential offspring. Paternal Consent It is appropriate for investigators to encourage a potential participant who is pregnant to discuss her participation in clinical studies and risks to potential offspring with the potential baby's father, but the committee rejects any requirement that the consent of the potential baby's father be a condition of the participation of a pregnant woman in research. The committee recognizes that the husbands of pregnant women, as well as future fathers who are not husbands, have an interest in the health of their children and that these men may have a deep emotional attachment toward their offspring prior to birth. Until a child is born, however, the future father can only protect the health of the potential child by controlling the decisions and actions of the woman. To give men the authority to veto the decisions of their wives or partners to participate in research grants men unacceptable power over women. It also would accord greater protection to fetuses than to children; only one parent's permission is required to enroll an infant or child in clinical research. Scientific Criteria for Exclusion The committee recognizes that, as in all clinical studies, there may be scientifically and medically valid reasons for excluding pregnant women from a particular study. A pregnant woman would be excluded if the medical condition of pregnancy disqualifies her as a subject in the same sense that anyone else, pregnant or nonpregnant, male or female, would be disqualified based on medical conditions that would interfere scientifically with the study. For example, a pregnant woman would be excluded from a study of hormone replacement or contraception. A pregnant woman also would be excluded from a study of weight loss, as would any person who, for example, was already very underweight; scientifically, it would not make sense to include either type of person in such a study. Similarly, a pregnant
OCR for page 198
--> woman would be excluded from a study when the condition of pregnancy places the woman in a risk category (because pregnancy increases the risk of harm to the woman) that would exclude others due to an unacceptable risk/benefit ratio. Other Criteria for Exclusion There was considerable discussion within the committee about whether there are any exceptional instances in which IRBs can be given the discretion to exclude pregnant women from participation for other than scientific reasons. Most committee members ultimately endorsed the following recommendation: Investigators and IRBs may exclude pregnant women from participation only when the IRB finds, and records its finding in writing, that the following standard has been met: (1) there is no prospect of medical benefit to the pregnant woman, and (2) a risk of significant harm to potential offspring is known or can be plausibly inferred. A finding that a risk of significant harm to potential offspring is ''known or can be plausibly inferred" may be based on evidence from animal studies, in vitro studies, structure-activity relationship data, or previous clinical experience. Under this standard, IRBs may exclude pregnant women from the earliest phases of many drug trials, but most clinical studies would remain open to pregnant women. Committee members adopting this standard were motivated by a desire to be true to the underlying principle that pregnant women should be treated no differently than other presumptively competent adults in the context of IRB deliberations. In addition, these committee members were particularly concerned that if the exceptive case was not narrowly constructed, variation in interpretation could open the door to widespread exclusions of pregnant women. A few members of the committee, however, were not able to endorse the above mentioned standard. They wished to reserve for the IRB the discretion to exclude pregnant women from participation not only when there is no prospect of medical benefit to the women but also when there is only potential for benefit to them that could be characterized as minimal or insignificant. The intent here is to allow the IRB more room for judgment about the appropriateness of exclusion. An example of a situation in which these members believed that IRBs should have the discretion to exclude pregnant women was that of a clinical trial of a medication thought to be helpful in the management of severe acne but known to cause malformations in offspring if taken during pregnancy. The standard endorsed by most
OCR for page 199
--> committee members would not permit a blanket exclusion of pregnant women from such a study, as it could not be claimed that there is no prospect of medical benefit to the pregnant participant. The committee also struggled with how to accommodate within its support for the shift of the presumption to inclusion of pregnant women (from that of exclusion) a role for conscience and an individual investigator's moral commitments. It was agreed that, at a minimum, such a mechanism would require that the investigator provide the IRB with a written explanation of his or her concerns of conscience and that the IRB review any such requests in light of a presumption that favors the inclusion of pregnant women in clinical studies. It also would require the IRB to guard against any abuse of conscience claims, and, in particular, against circumstances in which a request for exemption on the basis of conscience is offered in lieu of other reasons not based in moral commitment. It is precisely because of the potential for abuse of a "conscience" exemption that the committee could not resolve whether or under what conditions such an exemption should be constructed. Appeals to conscience are in many respects unassailable; in some contexts, the force of such appeals has had a chilling effect on public policy. Documentation and Monitoring of Exclusions An IRB must record in writing both its reasons for permitting any exception to the general presumption of inclusion of pregnant women and the frequency with which it grants such exceptions. It is anticipated that IRBs would record such information in the minutes of their meetings and that the act of documentation would help the IRBs to properly implement the standard. Such record keeping also would provide a source of information should OPRR desire to evaluate the performance of an IRB on this issue. Conclusion The committee recognizes that its recommendation concerning the participation of pregnant women in clinical studies cannot ensure the prevention of a small, theoretical risk of harm to offspring. Pregnancy and the controversial moral and legal standing of the fetus or potential child raise unique considerations. We do not wish to dismiss or evade these important considerations. However, the committee was persuaded of the overriding value of ensuring that all women-pregnant or otherwise-be treated justly with respect to the opportunity to derive the benefits of research. The shifting of the presumption to one of inclusion of pregnant women in clinical studies from one of exclusion is an important step in that direction.
OCR for page 200
--> The committee believes that given the safeguards described above, holding IRBs and investigators to the presumption of including pregnant women in research is not a significant threat to the health of future generations. Except for studies specifically designed to investigate outcomes in pregnant women-which the committee strongly endorses-it is exceedingly unlikely that investigators will seek out pregnant women for recruitment. It should be emphasized that the committee is not recommending that NIH impose an affirmative obligation on investigators to recruit pregnant women into every clinical study. Moreover, the committee's conclusions are consistent with the position that women who are or who might become pregnant have a moral obligation to weigh risks to a future child when deciding whether to participate as research subjects. It is unlikely that pregnant women will seek admission into studies that pose a significant risk of harm to their offspring, unless there is some offsetting benefit to the health of the pregnant woman that in turn advances the interests of the potential child by its having a healthy mother. A policy of presuming that pregnant women are eligible to participate in clinical research, although introducing a possibility of harm to a potential child, is in fact likely to produce health dividends for mothers that will inure to their children. Although the committee is not indifferent to the risk of harm to even one potential child, the committee felt compelled to consider as primary the interests of all women in being treated justly and with dignity. The committee recommends that OPRR revise and reissue Subpart B of the DHHS regulations for the Protection of Human Subjects, titled "Additional Protections Pertaining to Research, Development, and Related Activities Involving Fetuses, Pregnant Women, and Human In vitro Fertilization [45 C.F.R. 46, subpart B] in accordance with the committee's recommendations. At least a technical amendment to Subpart A, sec. 46.111(a)(3), eliminating the reference to pregnant women as a "vulnerable population" will be required by this revision to Subpart B. NOTE 1. There is historical precedent for classification of unknown or ambiguous risks to the fetus as more than minimal. This policy was developed with respect to fetoscopy in a decision by the Department of Health, Education, and Welfare Ethics Advisory Board in 1979 (DHEW, 1979) and by the NIH with respect to chorion villi sampling in the 1980s (C. McCarthy, former director of NIH Office of Protection from Research Risks, personal communication, October 1993). In both cases, it proved to be an appropriate
OCR for page 201
--> presumption, since both procedures were subsequently determined to increase the risk of spontaneous abortion. IRBs and investigators may find it helpful to use this convention in discussing the level of risk with potential participants. REFERENCES Appelbaum, P.S., Roth, L.H., Lidz, C.W., Benson, P., and Winslade, W. 1987. False hopes and best data: Consent to research and the therapeutic misconception. Hastings Center Report 17(April):20-24. Barnes, D.G., and Dourson, M. 1986. Reference dose (RfD): Description and use in health risk assessments. Regulatory Toxicology Pharmacology 8:471-486. Briggs, G.G., Freeman, R.K., and Yaffe, F.J. 1986. Drugs in Pregnancy and Lactation: A Reference Guide to Fetal and Neonatal Risk. 2nd ed. Baltimore, Md.: Williams & Wilkins. DHEW (Department of Health, Education, and Welfare). 1979. DHEW Support of Fetoscopy. Report and Recommendations of the DHEW Secretary's Ethics Advisory Board, 23 February. Faden, R.R., and Beauchamp, T.L. 1986. A History and Theory of Informed Consent. New York: Oxford University Press. Fletcher, J.C., and Ryan, K.J. Federal regulations for fetal research: A case for reform. Law, Medicine and Health Care 15(3):126-138. Francis, E.Z., Kimmel, C.A., and Rees, D.C. 1990. Workshop on the qualitative and quantitative comparability of human and animal developmental neurotoxicity: Summary and implications. Neurotoxicology and Teratology 12(3):285-292. Gaylor, D.W. 1991. Comparison of the properties of reference doses based on the NOAEL and benchmark doses. Proceedings, Air and Waste Management Association 71: paper no. 91 173.6. Gaylor, D.W. 1989. Quantitative risk analysis for quantal reproductive and developmental effects. Environmental Health Perspectives 79:243-246. Harlap, S. 1991. Preventing pregnancy, protecting health. New York: Alan Guttmacher Institute. Heinonen, O.P., Sloan, E.D., and Shapiro, S. 1983. Birth Defects and Drugs in Pregnancy. Boston: John Wright. Holtzman, N.A., and Khoury, M.J. 1986. Monitoring for congenital malformations. Annual Review of Public Health 7:237-266. Jelovsek, F.R., Mattison, D.R., and Chen, J.J. 1989. Prediction of risk for human developmental toxicity: How important are animal studies for hazard identification? Obstetrics and Gynecology 74:624-635. Jelovsek, F.R., Mattison, D.R., and Young, J.F. 1990. Eliciting principles of hazard identification from experts. Teratology 42:521-533. Kimmel, C.A. 1990. Quantitative approaches to human risk assessment for noncancer health effects. Neurotoxicology 11:189-198. Kimmel, C.A., and Gaylor, D.W. 1988. Issues and qualitative and quantitative risk analysis for developmental toxicology. Risk Analysis 8:15-20. Kline, J., Stein, Z., and Susser, M. 1989. Conception to Birth: Epidemiology of Prenatal Development. New York: Oxford University Press. Koren, G., Bologna, M., and Pastyszak, A. 1990. The way women perceive teratogenic risk: The decision to terminate pregnancy. Pp. 373-381 in Maternal-Fetal Toxicology: A Clinician's Guide, G. Koren, ed. New York: Marcel Dekker. Koren, G., Bologna, M., Long, D., Feldman, Y., and Shear, N.H. 1989. Perception of teratogenic risk by pregnant women exposed to drugs and chemicals during the first trimester. American Journal of Obstetrics and Gynecology 160:1190-1194.
OCR for page 202
--> Mattison, D.R. 1991. An overview of biological markers in reproductive and developmental toxicology: Concepts, definitions and use in risk assessment. Biomedical and Environmental Sciences 4:8-34. Merkatz, R., Temple, R., Sobel, S., Feiden, K., Kessler, D., and the Working Group on Women in Clinical Trials. 1993. New England Journal of Medicine 329(4):292-296. Meistrich, M. 1989a. Calculations of the incidence of infertility in human populations from sperm measures using the two-distribution model. Pp. 275-290 in Institute for Health Policy Analysis Forum on Science, Health, and Environmental Risk Assessment. Alan R. Liss . Meistrich, M. 1989b. Interspecies comparison and quantitative extrapolation of toxicity to the human male reproductive system. In Toxicology of the Male and Female Reproductive Systems, P.K. Working, ed. Bristol, Penn.: Hemisphere. Meistrich, M., and Brown, C.C. 1983. Estimation of the increased risk of human infertility from alterations in semen characteristics. Fertility and Sterility 40(2):220-230. Meistrich, M., and Mattison, D.R. In press. Methods for quantitative assessment of reproductive risks. In: Assessing the Risks of Adverse Reproductive Outcomes, Monograph No. 4 of the Conte Institute for Environmental Health, D. Brenner and A. Bloom, eds. March of Dimes. Moreno, J.D. 1994. Ethical issues related to the inclusion of women of childbearing potential in clinical trials. In: Women and Health Research: Ethical and Legal Issues of Including Women in Clinical Studies, Volume 2, A. Mastroianni, R. Faden, and D. Federman, eds. Washington, D.C.: National Academy Press. NRC (National Research Council) 1983. Risk Assessment in the Federal Government: Managing the Process. Washington, D.C.: National Academy Press. NRC. 1989. Biologic Markers in Reproductive Toxicology. Washington, D.C.: National Academy Press. Olshein, A.R., and Mattison, D.R., eds. In press. Male-Mediated Developmental Toxicity. New York: Plenum. Phelan, M.C., Pelloch, J.M., and Nance, W.E. 1982. Discordant expression of fetal hydantoin syndrome in heteropaternal twins. New England Journal of Medicine 307:99-101. Planned Parenthood Federation of America. 1991. New Birth Control Conferences Report. Nine Conferences in Nine Cities: A Report, January 1990-January 1991. New York: Planned Parenthood Federation of America. Quirk, J.G. 1986. Use and misuse of drugs in pregnancy. In: Drug and Chemical Action in Pregnancy: Pharmacologic and Toxicologic Principles, S. Fabro and A.R. Scialli, eds. New York: Marcel Dekker. Rai, K., and Van Ryzin, J. 1985. A dose-response model for teratological experiments involving quantal responses. Biometrics 41:1-9. Schardein, J.L. 1985. Chemically Induced Birth Defects. New York: Marcel Dekker. Shepard, T.H. 1983. Catalog of Teratogenic Agents, 4th ed. Baltimore, Md.: Johns Hopkins University Press. Sherwin, S. 1992. No Longer Patient. Philadelphia: Temple University Press. Takihi, N., Rosenkranz, H.S., Klopman, G., and Mattison, D.R. In press. Structural determinants of developmental toxicity. Submitted to the Conference Proceedings for Workshop on Quantitative Methods in Developmental Toxicology, National Research Council, Washington, D.C. Wilson, J.G., and Fraser, F.C., eds. 1979. Handbook of Teratology. New York: Plenum. World Health Organization Special Programme of Research, Development and Research Training in Human Reproduction and International Women's Health Coalition. 1991. Creating Common Ground: Women's Perspectives on the Selection and Introduction of Fertility Regulation Technologies. Geneva: World Health Organization. Zar, J.H. 1984. Biostatistical Analysis. Englewood Cliffs, N.J.: Prentice-Hall.
Representative terms from entire chapter: