Pregnant and breastfeeding patients routinely need medicines, yet they are largely excluded from the clinical trials designed to establish whether those treatments are safe and effective. Around 90% of pregnant people must take at least one medicine, including over-the-counter drugs or those for chronic conditions, be it hypertension, asthma, epilepsy, depression, or autoimmune diseases. Yet 71% of available medicines have no information on their use in pregnancy, and only 5% have adequate safety information for use during pregnancy and breastfeeding, according to data from the European Medicines Agency (EMA) and the European Institute of Women’s Health.

Now, regulators are seeking to change the presumption that pregnancy or breastfeeding should automatically exclude someone from clinical research.

The E21 guidelines from the International Council for Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) would establish a framework for the earlier and more systematic inclusion of pregnant and breastfeeding individuals in clinical trials when the potential benefits justify the risks.

The proposals received broad support from European regulators, healthcare professionals, academics, and industry representatives at a recent EMA meeting in Amsterdam on evidence gaps in women’s health.

“This draft guidance marks an important step towards addressing a long-standing evidence gap that has too often left pregnant and lactating individuals, their babies and their healthcare professionals making treatment decisions without sufficient data on the safe and effective use of medicines during pregnancy and breastfeeding,” Fionnuala McAuliffe, MD, a consultant ob/gyn at the National Maternity Hospital in Dublin, Ireland, and division director of maternal and newborn health at the International Federation of Gynecology and Obstetrics, told Medscape News Europe.

An Evidence Gap Decades in the Making

Pregnant and breastfeeding individuals have traditionally been excluded from clinical trials because of concerns about potential harm to the fetus or infant, unless the medicine is specifically intended for obstetric use.

The legacy of the thalidomide disaster, in which use of the drug during pregnancy caused severe birth defects, contributed to a highly cautious approach to exposing pregnant people to investigational medicines.

But Anna Oleárová, PhD, MBA, vice-chair in the Ethics Committee of the Ministry of Health of the Slovak Republic, argued that exclusion from trials has risks of its own. People who rely on medicines for chronic illnesses do not cease to need treatment when they become pregnant, she said at the EMA meeting.

Specific research on pregnant and lactating individuals tends to be carried out after the drug is already on the market. They are often carried out by clinicians in the form of case reports or series that appear in the medical literature, as well as pregnancy pharmacokinetics studies, lactation studies, and observational studies to assess safety in pregnancy. The problem is that it can take well over 20 years to establish a more precise pregnancy risk, according to EMA. “When [people] become pregnant, their need for effective treatment doesn’t simply disappear; it stays. If a woman stops her treatment, her uncontrolled disease can pose a direct risk to both her and her baby.”

The Illusion of Caution

Corinne de Vries, PhD, EMA’s lead on medicines in pregnancy and breastfeeding and an ICH E21 rapporteur, said the guidelines are intended to replace automatic exclusion with more pragmatic assessment of risks and benefits. “People may find it counter-intuitive, but excluding pregnant and breastfeeding women from clinical trials by default creates an illusion of caution,” she said. “And we are not proposing a reckless approach by any means — on the contrary. What we are saying is that we need to reconsider how we generate evidence for this population earlier on.”

E21 does not propose that pregnant and breastfeeding people should participate in every clinical trial. Rather, it establishes principles for determining when their inclusion is appropriate and what safeguards are required. The aim is to move away from exclusion by default toward consideration of whether, when, and how evidence can be safely generated in these patients. Drug developers would need to consider the issue early in a medicine’s development rather than waiting until after market authorization.

The draft includes recommendations for safety monitoring and follow-up of infants, as well as proactive planning for the types of data to be collected.

The approach would also be informed by emerging safety signals. Findings from nonclinical studies, including reproductive and developmental toxicity studies, could, for example, indicate that exposing pregnant participants to a medicine would pose an unacceptable risk.

One of the central challenges is determining the point in a medicine’s development at which there is sufficient evidence to consider exposure during pregnancy or breastfeeding. That decision would be made on a case-by-case basis.

A medicine might first be evaluated in nonclinical studies and then in nonpregnant participants before researchers determine whether there is sufficient evidence to consider including pregnant people. But the calculation may be very different depending on what the medicine is intended to treat.

For example, for a treatment with little or no medical benefit, such as one used purely for cosmetic purposes, even a relatively small potential fetal risk might make inclusion difficult to justify. The balance could shift for a medicine intended to treat a serious condition, particularly where the existing standard of care carries known risks during pregnancy. In such circumstances, generating evidence on an alternative treatment may have greater potential benefit. Researchers would also need to consider when during pregnancy exposure should occur because fetal risks and maternal physiology differ across trimesters.

For many medicines, developers have not conducted studies to establish whether a drug passes into breast milk and, if it does, the level of exposure in a breastfeeding infant. “Only very rarely do developers conduct the studies to determine as to whether a medicine passes into the breast milk, and if it does, what impact it might have if the baby drinks the milk,” said de Vries. Without those data, patients may decide not to breastfeed or may stop a medicine because they and their clinicians cannot adequately assess the risk to the infant.

If researchers establish that a medicine enters breast milk, for example, data might help determine whether infant exposure could be reduced by adjusting dosing or breastfeeding timing. For some medicines, it may be possible to establish an interval after treatment when breastfeeding can safely resume.

The principle, said experts at the EMA meeting, is not to eliminate risk but to assess it explicitly against the risks for untreated or inadequately treated disease and the limitations of existing therapies.

Ethical Questions Remain

Changing the traditional approach to clinical trials in pregnancy and while breastfeeding raises difficult ethical and practical questions.

Public consultation on the draft E21 guideline generated around 2500 comments, which are being reviewed as the guidance moves through the ICH process.

Among the questions under discussion is how pregnancy should be handled in trials that currently impose contraceptive requirements. In some circumstances, should those restrictions be lifted or should dedicated studies involving pregnant participants be conducted instead?

Consent is another area of debate.

Should a pregnant person’s partner have any role in consenting to trial participation? The Nordic Alliance of Clinical Trials in Obstetrics and Gynecology, a group of professors and associate professors in ob/gyn from Denmark, Sweden, and Norway that responded to the consultation, argued that they should not. But experts at the EMA meeting acknowledged that views on consent and participation may differ across countries and cultures, posing a challenge for guidance intended to be global.

The next ICH-level discussion incorporating consultation comments is expected in the first quarter of 2027. The final guidelines are anticipated around the end of 2027 or beginning of 2028.

E21 is part of a wider effort to address evidence gaps affecting women’s health. At the EMA meeting, participants prioritized 10 areas for action. Alongside supporting the inclusion of pregnant and breastfeeding people in trials, these included improving the enrolment of women in clinical trials; reporting high-quality sex-disaggregated data; generating evidence with women of different ages and life stages in mind; and developing sustainable infrastructure for collecting and analyzing real-world data.

De Vries and McAuliffe reported no relevant financial relationships.

Tatum Anderson is a global health journalist with almost 30 years of experience. Based in London, UK, Tatum specializes in technology, public health, infectious and non-communicable diseases, drug and vaccine development, medical research, and policy. As well as Medscape, her bylines have also appeared in The Economist, BBC News, SciDev.net, Guardian Weekly, The BMJ, Nature, the Bulletin of the World Health Organization, The Lancet, and a range of medical conference newspapers. She is also an editor, proofreader, and writer.