Epidemiology and Ethics, Scientific Integrity, Regulatory Control Working Groups

The Health Outcomes in Pregnancy and Epilepsy (HOPE) project approached the issues of research concerning women with epilepsy, pregnancy, and antiepileptic drugs (AEDs) by asking expert working groups to take on broad areas of responsibility with respect to the current situation and to provide suggestions for future studies. Two working groups found overlap in their discussions and concerns, which led to collaboration between the groups. Articles based on their work will be presented in this section.

Epidemiology working group members

The epidemiology working group was chaired by W. Allen Hauser MD, Professor of Neurology and Epidemiology at Columbia University College of Physicians and Surgeons, New York, NY. Other members of the group were Anne T. Berg, PhD (Northern Illinois University), Dale Hesdorffer, PhD (Columbia University), and David Thurman, MD (Centers for Disease Control). This group concentrated on the ideal approaches to studying the health outcomes of pregnancy in offspring of women with epilepsy and touched on the problems with the current epilepsy pregnancy registries. They also researched potential populations for future research and made some suggestions of alternatives to registries for studying the effects on offspring of antiepileptic drugs taken during pregnancy.

The ethics, scientific integrity, and regulatory control working group

The ethics, scientific integrity, and regulatory control working group was chaired by Jacqueline French, MD, Professor of Neurology and Director of the Penn Epilepsy Center at the University of Pennsylvania, Philadelphia, PA. Other members of the group were Kimford Meador, MD (University of Florida), Tuuli Cnaan, PhD (Children’s Hospital of Philadelphia), Frank Gilliam, MD, MPH (Columbia University), Jill Conway MD, PhD (University of Pennsylvania), Richardae Araojo, PharmD (U.S. Food and Drug Administration), and Karen Feibus, MD (U.S. Food and Drug Administration). This group addressed important ethical principles surrounding outcomes research and reporting related to women with epilepsy and their offspring and the development of pregnancy registries.

Background information: epidemiology group

The epidemiology working group presented background information for the HOPE project. It is estimated that in the United States each year there are 500,000 – 1,100,000 women with epilepsy of childbearing age. There are about 25,000 births annually to these women, the majority of whom are taking AEDs. Comparing childbearing rates for women with epilepsy to those for women in the general population has produced contradictory results. For women treated at epilepsy specialty centers, childbearing rates are one-third to one-fourth those of women without epilepsy.(1) These rates reflect women who have been referred by a neurologist for specialized treatment and they generally have severe or refractory epilepsy. The difference in childbearing rates is not found in population-based studies from Iceland and from the United States.

The risks for fetal malformations and developmental delays in babies exposed to AEDs in utero will be discussed in detail in the section reporting the work of the anatomical and cognitive teratogenesis work groups. The epidemiology work group noted that a recent study reported odds ratios for risk of anatomic malformations between 2.8 (monotherapy) and 4.2 (polytherapy) for the most commonly prescribed AEDs: phenytoin, phenobarbital, carbamazepine, and valproate.(2) Thus there may be 1200-3800 annually babies born in the United States with malformations associated with AED use. Risks associated with the newer AEDS are not yet known but are under investigation. At this point, strategies for prevention of fetal effects of maternal AED use include changing from polytherapy to monotherapy where possible, using the lowest effective dose of medication, considering a change in AED, and tapering off AEDs for mild epilepsy only prior to pregnancy and under the supervision of a neurologist. As more information becomes available, particularly concerning the new AEDs, other strategies will be proposed.

The remainder of the work of the epidemiology working group concerned the ideal approaches for studying health outcomes in the offspring of women with epilepsy taking AEDs. Articles in this section will be concerned with general study design and methodology principles, problems with designs currently in use and with existing pregnancy registries, and suggestions for possible populations appropriate for future research.

Balance between ethics and scientific integrity

Much of the work of the group devoted to ethics, scientific integrity, and regulatory control of studies was concerned with the balance necessary between ethics and scientific integrity for any trial involving humans. One issue discussed was the timing of the release of negative data and the role of regulatory agencies in determining when labeling should be changed. The group also considered the conditions that might explain some differing results among the different pregnancy registries. Variables differ from study to study, which impacts outcomes. There is a question regarding how to make comparisons with the variety of compounds and different combinations of compounds prescribed, different exposure rates, and possible selection biases in terms of registry populations, as well as other variable factors. Care needs to be taken when making comparisons, also when comparing registry data with trial data due to these confounders.

In the literature on pregnancy and epilepsy, there is at most one Class 1 paper, with the majority of papers being Class 3 or Class 4 (see box below). Much clinical research excludes women who are pregnant, so that registries provide one of the best ways to examine the outcomes for offspring of pregnant women with epilepsy, and those exposed in utero to AEDs. The United States Food and Drug Administration believes that when a product is marketed that will be used during pregnancy, a company-sponsored pregnancy registry may be needed. “When the purpose of the pregnancy registry is to assess margins of safety, to monitor for potential harm, or to detect safety signals, it is appropriate to initiate the registry as soon as possible, such as at the time of initial marketing, when a new indication is approved, or when patterns of use reveal that the product is used by women in reproductive age.”(3)

Some issues for registries include the difficulty of matching a control group for risk factors and determining what confounders should be considered. Timing of data collection in registries is important. For anatomical defects, collating data at birth is not sufficient. Data should be collected again at one year of age to ensure that certain malformations are not missed. For cognitive/behavioral issues, age 6 or 8 may be appropriate. Another issue involves determining inclusion/exclusion criteria. For example, if women with an abnormal ultrasound are excluded from a registry database, the risk for early defects will be underestimated. Follow-up is another issue for consideration. Registry size may also be a factor when considering confounders.

Ethical issues

The working group examined the ethical issues surrounding the requirement for women to initiate participation in the North American registry; their doctors may not enroll them. This policy has had an adverse effect in terms of enrollment, and thus may be seen as being overly conservative and causing a delay in accumulating the data needed to provide urgent information to the epilepsy community. Many of these issues will be addressed in the articles in this section.