Interferon beta babies.

Many women with multiple sclerosis (MS) ask about the risks of pregnancy before starting (or while on) disease-modifying therapies (DMTs). At least twothirds of patients with MS are women, and the majority are in their childbearing years at the time of clinical onset and diagnosis. DMTs taken during pregnancy could have direct or indirect toxicity (through immunomodulating properties) on pregnancy outcome. High doses of interferon beta (IFN ) are abortifacient in monkeys, but studies of teratogenicity in animals are limited by the rapid appearance of antibodies to human IFN (product information). Therefore, pregnancy outcome after in utero exposure to DMTs is important, particularly because DMTs are increasingly initiated early in the course of MS. In this issue of Neurology, two articles provide data on pregnancy outcome after in utero exposure to IFN .1,2 Sandberg-Wollheim et al.1 provide the first systematic and comprehensive analysis of pregnancy frequency and outcome, combining data from several randomized clinical trials of IFN -1a. Because allocation of treatment (IFN -1a at any dose vs placebo) was done randomly, there should be little bias introduced by patient assignment to treatment vs placebo groups for specific factors that could affect pregnancy outcome. Based on their study, there do not appear to be fertility issues for women exposed to IFN : The number of pregnancies is similar (or even slightly higher) in the IFN groups vs placebo. Furthermore, compared with patients exposed to placebo, immediate pregnancy outcomes (focusing on pregnancy loss and congenital malformations) were not different after in utero exposure to IFN -1a or after having discontinued IFN -1a at least 2 weeks before becoming pregnant vs placebo. However, the rate of spontaneous abortions in IFN -treated women was at the upper limit of the expected range for normal subjects. The data are reassuring and similar to data derived from patients exposed to glatiramer acetate.3 One concern is that the study groups in the article are a special subset of women (i.e., those on clinical trials) and may not be representative of the general population now given DMTs as part of “routine” MS management. Moreover, when assessing pregnancy outcome in relation to a maternal disease or teratogenic exposure, it is essential to control for factors such as maternal age, weeks of gestation at recognition of pregnancy, previous pregnancy history, and family history. One must therefore be careful in assigning relationships between treatment and outcome. Thus, the Sandberg-Wollheim et al.1 article is good news but has the limitations of relatively small sample size and the fact that all subjects were enrolled in clinical trials. In addition, there are no longitudinal data on live births. Finally, although spontaneous abortions were reported only in patients receiving IFN -1a at the time of conception (or within 2 weeks of discontinuing), this rate remains within expected range for subjects. Also in this issue of Neurology is a small prospective study by Boskovic et al.2 In this study, pregnancy outcomes were compared in three groups: patients (mostly with MS) who conceived while on IFN ; patients (mostly with MS) who conceived after discontinuing IFN ; and healthy control subjects. All subjects contacted “Motherisk help-lines,” a service that provides counseling to the public and to health professionals regarding the reproductive safety of drugs during pregnancy and lactation. After the expected date of confinement, interviewers follow up on pregnancy outcome. In this cohort, no woman elected to discontinue her pregnancy. There was an increased risk of spontaneous abortion in patients conceiving while on IFN and smaller birth weights in exposed infants. There was no difference for fetal death or prematurity. There was also a suggestion that IFN exposure during pregnancy may increase the risk of malformations. These findings remain after controlling for potential confounders such as age and cigarette smoking. It is impossible to be certain that the differences reported relate only to IFN exposure as the groups were not randomized and not all potential confounders were considered. As with the study by Sandberg-Wollheim et al.,1 patients with MS in the study by Boskovic et al.2