EMA's revised draft guideline links reproduction and breastfeeding data to product information

What happened

On 9 November 2026 EMA published Revision 1 of its draft guideline on assessing the risks of medicinal products in human reproduction and breastfeeding. According to the agency's description, the document covers the path "from data to product information". In other words, it deals with how evidence is assessed and how the conclusions are translated into what prescribers and patients read.

The records available to us give only that headline description and a link to the PDF. They do not say what changed from the earlier version, whether new data sources or new wording conventions are proposed, or when the consultation closes. We will not guess at any of these. Anyone planning a response should check the document itself and EMA's consultation page for the deadline.

Why it matters for pharmacovigilance

Pregnancy and lactation are among the hardest areas of safety communication. Pre-authorisation evidence is usually limited, because pregnant and breastfeeding women are generally not enrolled in pivotal trials. Much of what is eventually known comes from nonclinical studies, pregnancy registries, spontaneous reports and published literature. The quality of the product information depends on how consistently that mixed evidence is weighed.

A guideline that explicitly spans data through to product information touches both sides of a marketing authorisation holder's work. Safety scientists generate and appraise the evidence, and regulatory and labelling teams turn it into SmPC and package leaflet text. The SmPC section on fertility, pregnancy and lactation is where the output lands. Any change in expectations about how evidence is graded or worded can therefore affect routine variations, signal follow-up and periodic safety reporting.

The wider context: from evidence to label

Other items from the same day show how much attention the data-to-label step is getting. In the US, FDA is requiring application holders for every GLP-1 receptor agonist to add a Warnings and Precautions entry on a potential increased risk of NAION, a rare optic nerve condition. This is a class-wide labelling instruction, which shows a regulator turning accumulated safety evidence into uniform product text across a whole product class. It is not a reproductive-toxicity case, and the records do not link it to pregnancy or breastfeeding.

FDA's CBER also says postmarketing reports for Zolgensma include tumours with vector integration near the GLI1 oncogene. It is weighing regulatory action on Zolgensma and Itvisma. The outcome is not yet known. The case shows how postmarketing signals can reshape the risk picture long after approval, and why labels need a defensible route for incorporating new evidence.

EMA also published its 2026 pharmacovigilance recommendations for centrally authorised veterinary medicines. That is a separate framework from the human guideline, but it is a reminder that the agency is publishing guidance on several PV fronts at once.

Who is affected

The main audience is marketing authorisation holders with centrally and nationally authorised products in the EU. QPPVs, safety physicians, epidemiologists, nonclinical toxicologists and regulatory labelling staff will all have a stake. Teams that run pregnancy exposure registries or manage lactation data may find that the revision affects how they design and justify those activities.

Developers preparing new applications may also care. A guideline on assessing reproductive and breastfeeding risk can shape what evidence packages are expected and how conclusions are framed in draft product information. Whether Revision 1 changes those expectations is the first thing to establish when reading the text.

What to watch

First, the consultation timetable. Draft guidelines normally invite comments before they are finalised, but the records give no dates. Second, the comparison with the current text. A gap analysis against existing SmPC wording and internal procedures is the practical starting point. Third, any signs of alignment or divergence with US labelling practice for pregnancy and lactation, which matters for companies running global labels. The records do not address this, so it remains an open question.

Until the text has been read closely, the safest assumption is that the revision could change how evidence is weighed and worded, and that it is worth reviewing now rather than after finalisation.