Sargam Gupta for Vox
Semaglutide users, and even researchers, have to dig to learn about risks. Why is this information so hard to find?
After neuroscientist Martha Bagnall read a recent New Yorker article about Ozempic — the “buzzy” “celebrity weight loss jab” that’s upending obesity medicine — she emailed a colleague to dish on what she felt were gaps in the discussion about the drug’s effects on the brain. His reply blew her away.
“Another aspect that’s not talked about much,” he wrote to her, “is that Ozempic is not recommended for use during pregnancy.”
The colleague, an obesity researcher, was referring to studies in rats, rabbits, and monkeys, which were treated with the injectable drug and had higher rates of miscarriage. Their offspring were also born smaller and had more birth defects than would normally be expected.
The animal studies are the basis for Food and Drug Administration warnings that both Ozempic — approved only for diabetes but used off-label for weight loss — as well as the higher-dose obesity formulation, Wegovy, should be discontinued at least two months prior to pregnancy. The agency has also required the drugmaker, Novo Nordisk, to organize follow-up studies that investigate health outcomes in people exposed to Wegovy during pregnancy.
Bagnall had been following the news about the medicines for months, reading all about their promise for patients and how they were redefining beauty standards, especially for women, but she had not picked up on the pregnancy concerns, not in the media coverage, nor in online drug ads. “Given the prevalence of Ozempic ads with women in them,” she said, “you think that’s a pretty big [potential] side effect to draw attention to.”
The overwhelming majority of the dozen obesity, diabetes, obstetrical, gynecological, and regulatory experts Vox reached out to for this story said much the same: There’s not enough discussion of the pregnancy risks, and warnings are not featured prominently enough in advertisements or in the drug’s labeling and instructions. “We should be talking more about this,” said Diana Thiara, a professor and medical director of the UCSF Weight Management Clinic. Or as one obesity researcher, who has associations with the drugmaker and did not feel comfortable speaking on the record, put it: “Given the majority of people accessing these medications are women, and a significant portion will be of childbearing age, this needs more attention.”
It’s not yet clear why the birth complications in animals arise; it’s possible they’re driven only by the weight loss the drugs can bring on, not the medications — which both contain the active ingredient semaglutide — themselves. It’s also possible the side effects won’t manifest at all in humans. But like most drugs, semaglutide’s effects haven’t been studied yet in pregnant people, so the risks to human pregnancy are uncertain.
Even if it’s only the weight loss that led to the animal harm, the need for more awareness remains, said Daniel Drucker, a scientist and endocrinologist at the University of Toronto who helped discover GLP-1, the human hormone on which semaglutide is based. Weight gain is an important goal of a healthy pregnancy, he said. Given that nearly half of pregnancies in the US are unplanned, the time period before a pregnancy is detected may be particularly vulnerable.
“This pregnancy issue is a real issue. There’s no question about it,” Drucker said. “The drug manufacturers and most certainly prescribing health care providers should highlight this issue to [people] with potential for pregnancy.”
But in the Wild West of the Ozempic era, highlighting nuances about potential side effects is a massive challenge. Semaglutide is part of a new class of GLP-1-based prescription medicines, the first-ever that appear to safely cause substantial weight loss, and they have met a hungry market of millions who struggle with their body size. Some patients are circumventing doctor’s offices to get access — buying the drugs online, from other countries, or through compounding pharmacies, often without medical supervision. Meanwhile, telehealth companies that advertise directly to patients and operate with little regulatory oversight have been feeding on the frenzy, pushing the weight loss benefits of the medicines in their ads, as Stat recently reported, without appropriately warning patients about the drug’s risks.
The worst-case scenario for human pregnancies is alarming, said Joseph Ross, a professor at Yale School of Medicine who researches pharmaceutical regulation. “If a pregnant woman is taking these drugs and doesn’t realize the risks, and it does bear out the drugs cause human fetal harm,” he said, “we could end up in an awful mess — pregnancies ending in miscarriage or neonates born with birth defects.”
What animal studies have found
While seven in 10 people in the US take a prescription drug during pregnancy, only 10 percent of medicines recently approved by the FDA reach the market after they’ve been studied during human pregnancy, according to an analysis published in JAMA, because of safety and ethical concerns.
The hesitance to include pregnant people in studies is intended to protect them and their babies, but “the ultimate effect is to provide care without adequate evidence,” said Catherine Spong, professor and chair of the department of obstetrics and gynecology at UT Southwestern Medical Center. And even though many expert voices and even major task forces have called for the inclusion of pregnant women in clinical trials over the last several decades, she added, “the lack of inclusion persists. This is simply another example.”
So we don’t typically have data on the effects of most drugs in pregnant humans. But for some 90 percent of recently approved medicines, there are studies in animals — including the semaglutide products, Ozempic and Wegovy. And in these cases, the animal studies uncovered harms.
Studies of semaglutide in rats, rabbits, and monkeys found the animals experienced higher rates of miscarriage, and that their babies more often did not grow to full size. In addition, the babies were more often born with “structural abnormalities” in different organs, tissues, and parts of the skeleton: the heart, blood vessels, kidney, liver, cranial bones, vertebra, sternebra, and ribs. This data is outlined in the Ozempic and Wegovy drug labels — the big sheet of information that comes with all medicines.
But, again, it’s difficult to tease out whether the effects are driven by the medicines or the weight loss that the medicines can cause, Drucker explained. “During a period of rapid growth, if the fetus is in a caloric deficit and the mother is in a caloric deficit, that will clearly produce some impairment of fetal growth,” he said, and maybe the congenital disorders are a sign “the baby’s just not growing properly.”
An FDA spokesperson told Vox the agency takes the view that the pregnancy complications are likely caused by weight loss and poor nutrition. In the animals, the fetal complications happened when mothers did not gain weight or lost weight during their pregnancies. The spokesperson said that the agency is unaware of any humans harmed during pregnancy or in utero because of semaglutide exposure.
Because of the clear weight loss risks, the agency’s Wegovy pregnancy warning is slightly more stringent than the warning on Ozempic. Wegovy users are advised to stop the drug as soon as a pregnancy is detected, while Ozempic users, ostensibly taking the drug for its indicated diabetes use, may continue “only if the potential benefit justifies the potential risk to the fetus,” according to the Ozempic label.
In practice, experts said, the Ozempic warning may mislead or confuse people, since many use Ozempic off-label only for its weight loss benefits and both labels also advise discontinuing the drugs at least two months before pregnancy.
Still, based on the animal data, the spokesperson added, “taking semaglutide during pregnancy may increase the risk of birth defects and miscarriage above background for the US general population,” and the extent of increased risk has not yet been quantified.
When the effects of the drugs on pregnancy are better studied, we could learn about other contributing factors. When asked about potential mechanisms by which semaglutide itself — and not just the weight loss it causes — could do harm, Drucker said, “It’s possible [it] does have effects on, for example, placental blood flow or formation of the placenta. This is not an extensively studied area, and there are gaps in our knowledge.”
It may take years to understand the real impact of GLP-1 receptor agonists on pregnancy. At the FDA’s request, Novo Nordisk has organized a study to analyze complications related to Wegovy in pregnancies that already happened, and a registry that will prospectively follow the health outcomes of a group of people exposed to Wegovy during pregnancy compared to pregnant people who weren’t. The first will be completed in 2027, and the second in 2033. (No such studies are required for Ozempic.) Meanwhile, millions of people will take the drugs, including during pregnancy.
The ideal versus real world
Pharmaceutical companies and the FDA have an obligation to make patients and prescribers aware of the potential risks of medications. This typically happens through warnings on drug packages and drug labels, and in official drug ads, as well as patient encounters with physicians and pharmacists who have read the entire drug label.
But even in the best cases today, where patients actually see a specialist for a prescription, they might not get the pregnancy warnings, Yale’s Ross said. Rather than being featured prominently in the drug labels or side-of-package warnings, he said, the information appears in a section of the label called “use in special populations,” so prescribers are left to “dig through searching for the section on pregnancy where it says clearly the drug should be stopped.”
In the labels’ instructions for use, which explain how to store and administer Wegovy and Ozempic, there’s also no mention of pregnancy, said Steven Woloshin, who co-leads the Center for Medicine and Media at the Dartmouth Institute, “but there is a whole bit about how to protect the Wegovy pen. … That seems like an oversight. Protecting the pregnancy seems a lot more important. So why not repeat the warning?”
Even in the official Ozempic and Wegovy commercials, the advice to tell your health care provider if you’re pregnant or planning to be only flashes quickly in text on the screen.
(The FDA spokesperson said the agency would typically only feature a pregnancy-related risk more prominently if “teratogenic events have been observed in humans” — that is, when the DNA of a fetus has been altered by a medicine, causing congenital disorders. There’s no semaglutide data showing such harm.)
Yet many patients aren’t even getting the medicines through patient-specialist encounters. The rampant off-label use of semaglutide, Ross said, suggests patients are not seeing obesity or diabetes clinicians, which means their health care providers may be even less familiar with any contraindications or side effects.
“My guess is most physicians don’t know the concerns,” Ross said. (Bagnall wasn’t the only health researcher unaware of the pregnancy contradiction; even some of the GLP-1 researchers and obesity clinicians Vox reached out to about the pregnancy warnings didn’t know about them.)
Or patients, after being bombarded by online ads from telehealth marketers that don’t adhere to FDA marketing guidance, may not see a clinician at all.
“This is an unusual case,” Ross summed up. There aren’t a lot of drugs like semaglutide: ones originally intended for a narrow use with a desirable benefit for the broader public. On top of that, they’re “increasingly available through direct-to-consumer telehealth companies and for purchase over the internet,” he said. It’s a recipe for confusion. While it’s not clear whether the drugs pose harm to human pregnancy, Ross said, “widespread educational efforts are needed as a precaution.” This can happen through public awareness campaigns or extra messaging in the drug’s marketing materials, he added, including drug labels, ads, and packaging.
A Novo Nordisk spokesperson told Vox, “Novo Nordisk does our best to ensure patients and health care providers are educated about the appropriate and responsible use of our medicines. Our efforts are designed to ensure that health care providers are prescribing the right product for the right patient.”
What semaglutide users should do
So what are patients to do now?
Given the unknowns about its effects in pregnant humans, patients taking semaglutide who want to conceive should stop at least two months before a pregnancy, the FDA spokesperson said. They should also “know that appropriate weight gain based on pre-pregnancy weight is currently recommended for all pregnant patients, including those who are overweight or obese, because of the obligatory weight gain that occurs in maternal tissues during pregnancy.”
At the same time, semaglutide can potentially help prepare a person for a successful pregnancy in the years and months before they plan to have a baby. As Stephen O’Rahilly, an endocrinologist and director of the University of Cambridge MRC Metabolic Diseases Unit, pointed out, “obesity is also associated with poor fetal outcomes,” heightening a woman’s risk of everything from stillbirth to diabetes and high blood pressure. Neel Shah, an OB-GYN and chief medical officer of Maven Clinic, agreed, noting the drug may have particular benefits for people with weight-related infertility or conditions such as PCOS.
“For them, Ozempic may actually make them more likely to conceive and the risk-benefit trade-off may be different from the general population,” Shah said. They just have to be prepared to stop taking the drug before trying to get pregnant.
People with diabetes who are pregnant or thinking of becoming pregnant should switch to older, better-studied drugs to manage their blood sugar, such as insulin, said Samuel Klein, the director of the Center for Human Nutrition at Washington University in St. Louis — an approach that goes further than the FDA (which, again, says pregnant people using the drug for diabetes may continue if the benefits outweigh the risks). In Klein’s view, “there are safer ways to control blood glucose in pregnant women, and [GLP-1 receptor agonists] should not be used to lose weight during pregnancy.”
For her part, UCSF’s Thiara counsels patients of reproductive age who are on semaglutide to use contraception and to bring up the risks related to pregnancy at their regular appointments. (There’s no requirement for contraception with semaglutide, and such requirements for drugs are rare in general.) Like Klein, she also gets patients — even those with diabetes — off semaglutide before and during pregnancy.
But not every patient sees a specialist like Thiara, which means the onus is now on patients to learn about the risks — and that may not change anytime soon, she added.
“I suspect the drug companies are downplaying this risk because women are probably the biggest part of the market share,” she said. “The world doesn’t value women, and this is seen in women’s health as well. So perhaps the pessimist in me thinks this might just be another case of the medical system devaluing women.”