Why An Abortion Drug Approved 20 Years Ago Might Get Yanked From The Market
It's less about science, and more about philosophy — particularly that of the Trump-appointed judge poised to rule on the approval.
Approving an abortion drug is always going to come with a high level of scrutiny and pushback. Dr. Michael Greene knew this well back in 2000 when he was leading the Food and Drug Administration committee that first approved the abortion drug mifepristone for sale in the U.S.
To Greene, now a professor emeritus of obstetrics and gynecology at Harvard University, that meant his team and the agency as a whole would need to be extremely cautious about ensuring mifepristone’s safety. “Four-plus conservative,” he calls it; an approach that led to an approval process that took more than three times as long as other drugs approved that year. It also resulted in a series of stringent, post-approval regulations that tightly controlled who could prescribe the drug, where patients could take it, what tests doctors needed to perform and even where the drug could be kept. “It had to be stored quite literally under lock and key,” Greene said.
But to abortion opponents who didn’t want to see mifepristone approved at all, those extra regulations seem less like an abundance of caution, and more like an admission that the drug was extremely dangerous. “They needed to use that authority because the FDA at the time recognized the dangers with chemical abortion drugs,” said Erik Baptist, senior counsel for Alliance Defending Freedom, a conservative legal advocacy group that is representing a group of organizations challenging the FDA’s approval of the drug. By “shoehorning in” post-approval regulations, he said, the FDA was “recognizing the dangers [mifepristone] poses to women and girls who take it.”
Baptist’s clients are charging the FDA with ignoring sound scientific practice and its own guidelines to approve a drug that endangers women. They’re asking for mifepristone, which is currently approved for use during the first 10 weeks of pregnancy, to be taken off the market immediately. The lawsuit hinges on technical details — t’s crossed and i’s dotted more than two decades ago. But a closer inspection of the FDA’s approval process and its continued monitoring of mifepristone reveal flaws in the lawsuit’s argument. Legal experts told us that, under a different judge, this case would likely be thrown out as frivolous. But differences in philosophy — specifically about the value and availability of abortion — change that. There’s a good chance Baptist’s clients could get what they want. And if they do, the most commonly used form of abortion would no longer be available in the U.S., transforming the landscape of abortion access in ways that could be even more far-reaching than the Supreme Court’s decision in Dobbs v. Jackson Women’s Health Organization last summer.
Mifepristone — originally known as RU-486 — was first developed in France in the 1980s. The drug works by binding to receptors in cells that would normally bind to progesterone, a hormone that is crucial to maintaining pregnancy, especially in the early weeks. Cut off from access to progesterone, the uterus begins to behave as though it is having a naturally occurring miscarriage — the uterine lining begins to break down and the embryo detaches. When followed by misoprostol, a separate drug that softens the cervix and starts contractions, the combination medication abortion regimen is more than 90 percent effective in the first nine weeks of pregnancy. (While misoprostol would still be available for medication abortions even if Baptist’s lawsuit prevails, it’s not nearly as effective on its own.) A 2018 study of almost 5,000 Swedish women over eight years found that the most common serious complication is an incomplete abortion — when all the fetal tissue doesn’t get flushed out on its own, and the patient ends up needing a follow-up surgical abortion. In that same study, annual rates of complications following medication abortions ranged from about 4 to 8 percent.
It has also been shown to meet a level of safety that exceeds that of more common medications, including penicillin and the class of drugs that include Viagra. In fact, it’s safer than full-term pregnancy itself. In 2018, the FDA’s evaluation of 18 years of reported complications found 24 deaths among approximately 3.7 million women who had medication abortions, or 0.65 deaths per 100,000 abortions — a figure that included deaths the FDA concluded were unlikely to be related to the drugs themselves, such as homicide or suicide.
Today, the combination of mifepristone and misoprostol is standard practice for medication abortion in the U.S. and has been approved in 80 countries worldwide. But its approval in the U.S. was not a given. Opposition goes back to before mifepristone was even approved and has been rooted in opposition to abortion itself. In fact, getting the drug approved at all in this country required intervention of the Clinton administration, which helped arrange a nonprofit foundation to sponsor mifepristone’s FDA application after the European pharmaceutical company that owned the patent declined to apply under pressure of boycotts.
Mifepristone was ultimately approved under the FDA’s accelerated approval process. This system was originally designed to bring AIDS treatments to market quickly, said Supriya Munshaw, a lecturer at Johns Hopkins University’s business school whose expertise is on the approval and commercialization of medical products. Accelerated approval can either allow drugs to be approved with shorter trials or allow the FDA to impose strict regulations on their use post-approval. The lawsuit claims that mifepristone should not have been a candidate for this program. It also implies accelerated approval allowed the Clinton administration to make mifepristone available just before the 2000 election.
But Greene said that putting mifepristone through this process wasn’t an attempt to get it to market more quickly — rather, it was about showing that the agency was approaching the controversial drug with an abundance of caution. Despite the name, the accelerated approval process was used to subject mifepristone to heavier regulation and more monitoring after it was approved. It certainly did not speed up approval. From start to finish, approving mifepristone took 54 months. The average length of the process for all other drugs approved in 2000 was 15.6 months.
People with expertise in the FDA approval process told us that lawsuits against the agency are not abnormal, especially when the aggrieved party feels that the FDA has not properly handled public complaints made directly to the agency, as is the case here. But those lawsuits don’t typically request that a drug be entirely removed from the market, said Charles Seife, a professor of journalism at New York University who has investigated misconduct by the FDA.
It’s also not unreasonable to criticize the FDA and, in particular, the accelerated approval process. That system has come under fire in recent years, Seife and Munshaw said, because of growing problems with the agency’s failure to enforce post-approval evaluations. In 2021, for example, the agency approved the sale of Aduhelm, an Alzheimer’s drug that critics argued has not been adequately shown to work. But both Seife and Munshaw said the flaws in accelerated approval are centered on the agency’s failure to ensure efficacy, not safety.
But it’s the safety of mifepristone that has been a major point of contention in this case, and the conflict seems to center around which studies one is willing to accept as valid. To Baptist, the studies used to prove the safety of the drug were inadequate, at least partly because they relied on controlled trial conditions and not the conditions of real-world use — something that matters because medical practitioners have changed some of the protocols of use over the years in response to evidence showing that changes could be safely made. For example, a lower dosage of mifepristone turned out to be just as effective as the higher dosage stipulated by the FDA.
The FDA’s 2018 post-approval review of safety and efficacy was based on real-world usage. Still, Baptist does not consider this to be good data, either, saying that the lax reporting in the United States is likely missing many instances of serious complications. In an interview with FiveThirtyEight, he said that he preferred to look at data from Finland, which has a national health service that collects data on every abortion performed. A study there looking at data from 2000 to 2006 found higher rates of hemorrhage (15.4 percent in adults) and incomplete abortion (10.2 percent in adults) than what turn up in the FDA surveillance. But this data is out of step with larger-scale systematic reviews of data from multiple countries, including others with national health system registries. For example, a 2013 systematic review that includes data from, among other countries, Finland, the U.K., China, Canada and the U.S., found an incomplete abortion rate of 4.8 percent, a hospitalization rate of 0.3 percent, and a blood transfusion rate of 0.1 percent.1
What’s more, nearly all the complaints made in the lawsuit have been addressed in two separate investigations by the Government Accountability Office, the independent nonpartisan watchdog that investigates federal malfeasance for Congress — one from 2008, which dealt with complaints about safety and the initial approval process, and a separate one published in 2018 that addressed complaints about post-approval monitoring and changes to the original safety regulations. The earlier investigation found that the FDA had followed its own norms and scientific evidence. The second investigation found that the FDA’s subsequent changes were consistent with the available scientific evidence, and documented that stakeholders — including both anti-abortion groups and abortion-rights groups — had a wide range of sometimes conflicting opinions on the drug’s safety and the quality of evidence. “This is a question that’s been asked and answered multiple times over more than two decades, and the only reason it’s being asked again is because the parties bringing the case believe they have a judge [who] will be sympathetic to it,” said Rachel Rebouché, a law professor at Temple University who studies abortion.
And it’s difficult to tell whether any evidence would be considered good enough to convince the groups who keep bringing complaints about mifepristone. Baptist doesn’t accept the GAO reports, for example, either. “When one federal agency supports the conclusion of another federal agency, that doesn’t mean that’s the end of the story,” he said.
Evaluating safety, meanwhile, always contains an element of subjectivity, Seife told us, because no drug can be 100 percent safe. It’s more of a cost-benefit analysis: What you’re really evaluating is whether the benefits outweigh the risks. From Seife’s perspective, mifepristone is a drug that has been on the market for more than 20 years and its benefit-to-risk balance doesn’t justify pulling it from the market. Pulling drugs off the market after approval is rare, usually happens much sooner after approval, and is prompted by more severe outcomes. For example, the FDA approved the painkiller Vioxx in 1999 and the drug was removed from the market five years later, after a study found that out of 1,287 people assigned to take Vioxx in a randomized trial, 46 of them ended up having serious medical events involving blood clots. Later research suggested that Vioxx may have killed upwards of 55,000 Americans.
But for those who believe abortion is fundamentally wrong, even small risks would not be outweighed by benefits because they see no benefits. That’s the issue happening here, Seife told us. And it’s also why the plaintiffs have a much better chance of winning than they would normally — they filed their lawsuit in front of one of the most conservative judges in the country. Matthew Kacsmaryk, a Trump appointee based in Amarillo, Texas, worked as a lawyer for a Christian conservative group before he was confirmed in 2019. Many experts think there’s a good chance he’ll rule in favor of the plaintiffs because of that background. David Cohen, a law professor at Drexel University who studies abortion, said that federal judges have a lot of freedom — even if under normal circumstances, this case wouldn’t have legs. “We assume they operate based on a certain set of rules, because that’s a more comforting way of thinking about how courts and judges operate in this world,” Cohen said. “But they are human beings. Some of them are politically motivated. And the beauty of being a smart lawyer is, you can reason to any conclusion you want.”
If Kacsmaryk rules that mifepristone should be taken off the market, it won’t be the end of the story. The case will be appealed, and mifepristone’s approval may ultimately end up surviving. But this lawsuit is an example of how easily something that seems nonpartisan — the mundane process of approving drugs — can become drawn into debate about whether an abortion drug can ever be safe. And Seife believes a win would open the door for similar groups to target other drugs that affect reproductive health. “I would wager you if this wins, you’re gonna get [new lawsuits going] right for the Pill. You can point to blood clots and stuff like that with the contraceptive pill,” Seife said. “So you can come up with the same sort of argument, just cut and paste it.”