FDA Pregnancy Categories: Why the Letters Are Gone

FDA pregnancy categories were the old A, B, C, D, and X letters on prescription medicines. FDA replaced them with explanations of pregnancy and breastfeeding evidence, starting June 30, 2015. Labels changed in stages.

The letters looked like quick answers in prescribing information, the detailed document written for clinicians. Removing them did not establish that a medicine was suitable during pregnancy.

Why FDA pregnancy categories gave way to words

The letters came from FDA's 1979 labeling rules. They mixed human and animal evidence with treatment benefits, but often looked like a safety ranking.

After a public hearing, two advisory committees and focus groups, FDA proposed replacing the letters on May 29, 2008. FDA received 72 comments from manufacturers, clinicians, consumers, and others. Many supported clearer information while questioning details.

The final rule, published December 4, 2014, removed the letters and required narrative summaries, meaning explanations in words. FDA's resource page dates the rule's release December 3, 2014; the Federal Register printed it the next day.

Comments helped shape the result. FDA dropped proposed standardized risk statements because they could oversimplify things too. It also added a place for pregnancy testing, contraception, and fertility information, as its December 2014 Q&A explains.

June 30, 2015 started a transition

The rule's effective date wasn't a deadline for every label to change. Its implementation table set different submission deadlines:

Here, applications cover new drugs and biological medicines, such as vaccines, plus applications adding new uses. Products approved before June 30, 2001, which never moved to FDA's 2006 label layout, had to lose their letter by June 30, 2018, without adopting the whole format. Over-the-counter medicines were excluded.

Egrifta shows what replaced a letter

The Pregnancy and Lactation Labeling Rule, or PLLR, separates pregnancy, breastfeeding, and reproductive questions. Lactation means breastfeeding. Reproductive potential covers pregnancy testing, contraception, and possible fertility effects.

As of November 5, 2026, the day this article was written, the Egrifta WR record says “Updated July 29, 2026,” but its prescribing information says “Revised: 3/2025.” These are different dates.

Egrifta WR is branded tesamorelin. Its section 8.1 explains that changing deep abdominal fat offers no pregnancy benefit and could harm a developing baby. Pregnancy is also a contraindication, a situation in which the medicine must not be used, in section 4. The explanation matters more than assigning a letter.

Better labeling did not fill every evidence gap

A clearer document doesn't create missing studies. Byrne and colleagues' 2020 JAMA Network Open study examined 290 new medicines approved during 2010–2019. Only about 1 in 9 labels included human pregnancy data: 31 of 290, or 10.7%. That was a study of labels, not a safety estimate for any medicine.

The transition was uneven too. Among 138 submissions from before the effective date in that study, 45 still lacked the new format at the June 2019 deadline. That number fell to 26 by January 2020. These are historical findings, not a count of labels today.

Letters also had a lasting pull. Robinson and colleagues' randomized survey assigned clinicians by chance to review identical case and drug information, with or without pregnancy letters. It appeared online in 2022 in the Journal of Perinatal Medicine. Without the letter, they were less willing to prescribe medicines formerly classified B or C. The study measured stated choices, not pregnancy outcomes.

Vyleesi shows why section 8.3 matters

Vyleesi's label, marked “Revised: 3/2024,” separates several questions about bremelanotide, also called PT-141.

Section 8.1 says the few human pregnancies recorded in trials cannot establish the risk, while animal findings raise concern about harm to a developing baby. Section 8.2 reports missing human milk and infant-effect information. Section 8.3 repeats that use during pregnancy is not recommended and describes effective contraception and stopping treatment if pregnancy is suspected.

These aren't interchangeable statements. Limited pregnancy observations don't establish breastfeeding safety. Contraception information belongs before an exposure happens. None of those distinctions fit neatly inside a letter. Vyleesi's document does not become the label for compounded PT-141, a medication prepared by a pharmacy for an individual prescription.

Section 8.3 isn't required when none of its topics apply. Egrifta WR's contents, for example, jump from 8.2 to 8.4. A missing 8.3 doesn't mean pregnancy is impossible, and it isn't the only place fertility appears: Egrifta WR's section 13.1 reports no effect on fertility in rats, an animal finding rather than a human one.

What the narrative headings add

The practical lesson is that removing shorthand still leaves work for the clinician: interpreting the evidence for the person in front of them.

The newer layout makes that work easier to follow. A Risk Summary states the overall picture. Clinical Considerations explains treatment questions, including risks from an untreated illness. Data describes the studies behind the summary. These headings distinguish the conclusion from the evidence supporting it.

Geref's old Category C belongs to its history

Sermorelin offers a different example: a discontinued branded label. A California government document dated March 19, 1999 reproduces Geref's pregnancy labeling. It says Category C, describes animal findings, and notes the lack of adequate controlled studies, those with a comparison group, in pregnant women.

FDA's March 4, 2013 notice records that Geref's approvals were withdrawn effective June 18, 2009. FDA determined the products had not been withdrawn from sale for safety or effectiveness reasons.

That finding doesn't establish pregnancy safety. The old Geref wording describes that discontinued branded product; it doesn't supply a current label for compounded sermorelin. An old Category C listing isn't a present-day permission slip.

It also wasn't a percentage chance of harm. The narrative beside the letter carried the animal findings and human evidence gaps. Reading only the C discarded the very information needed to understand it.

What a provider reads now

Sections 4, 5, and 8 answer different questions: when a medicine must not be used, which warnings need attention, and what is known for particular populations. A boxed warning highlights a serious risk; it doesn't replace those sections.

For medicine-specific pregnancy questions, the separate guides to semaglutide and pregnancy and tirzepatide and pregnancy cover those labels and studies.

Tesamorelin and PT-141 prescribed through Promise are dispensed by a licensed U.S. compounding pharmacy as compounded medications, which are different from FDA-approved products such as Egrifta WR and Vyleesi: the formulations offered here are not FDA-approved. Sermorelin prescribed through Promise is dispensed the same way, but its history differs: no FDA-approved sermorelin product is currently marketed — Geref's approvals, from 1990 and 1997, were withdrawn in 2009 — and the compounded formulation offered here is not FDA-approved. FDA explains the distinction.

A licensed provider may still prescribe — that decision is between you and your doctor. At Promise, a licensed provider reviews every request, and not everyone qualifies.

The useful discussion concerns the exact medicine, the reason for treatment, pregnancy plans, breastfeeding, and what the evidence actually covers. A missing letter answers none of those questions.