At the end of October, an FDA advisory committee voted massively against the continued FDA approval of 17α-hydroxyprogesterone caproate (17-OHPC; Makena), a drug indicated to reduce the risk of preterm labor in pregnant women. In documents prepared by FDA scientists prior to the committee hearing, the conclusion was that the drug “should be taken off the market”. Similarly, an FDA advisory committee voted 3 years ago to withdraw the drug in the United States, citing a clinical trial 2019 showing that the drug offered no benefit over placebo. Still, the manufacturer has so far been unwilling to voluntarily withdraw the drug, so it remains unclear when (or even if) the FDA will follow the recommendations of its scientists and advisers.
Much of the current regulatory concern about 17-OHPC centers on the drug’s initial approval by the FDA under the fast-track approval program and how subsequent clinical trials of 17-OHPC, under program, cast doubt on the drug’s safety and efficacy. Under expedited approval, the FDA can approve drugs based on surrogate measurement changes that are reasonably likely to predict real-world, meaningful clinical parameters, such as how a patient feels, performs, or survives. Manufacturers are required to conduct a post-approval study to confirm the drug’s effect on clinical parameters. However, recent studies have shown that confirmatory studies are often delayed (about 70% were delayed between 2013 and 2016), frequently evaluate alternative measures and may not lead to timely drug withdrawal when confirmatory trials are negative.
The history of 17-OHPC
The case of 17-OHPC illustrates many ways that fast-track approval can go wrong. The drug, 17 alpha-hydroxyprogesterone, was first approved by the FDA in 1956 under the brand name Delalutin for the prevention of “habitual and recurrent abortion.” The drug fell out of favor for this indication, but received new FDA approval 55 years later in 2011 to prevent premature births, based on a 2003 trial. Although the trial did not shown clinical benefit for neonatal health, the drug was approved via accelerated approval due to substantial public health need for preterm labor treatments and promising surrogate results.
As part of the accelerated approval pathway, the manufacturer of 17-OHPC was required to perform a confirmatory trial to assess its safety and efficacy with respect to neonatal health. The confirmatory trial was slow to unfold and 17-OHPC gained wide use. The test, finally completed in 2019, recruited 1,708 women and found no effect on neonatal morbidity index. Interestingly, this also directly contradicted the main findings of the 2003 study, showing no difference in the frequency of preterm labor at less than 35 weeks.
In addition to the new trial results, meta-analyses revealed that 17-OHPC was no better than a placebo to prevent premature births at less than 32, 35 and 37 weeks of gestation, and was lower to another care alternative, vaginal progesterone, to reduce the rate of preterm birth in mothers at less than 32, 34 and 37 weeks of gestation. (Meta-analysis research on the use of vaginal progesterone is still ongoing, with a recent study muddying the waters on the effectiveness of vaginal progesterone in preventing preterm birth.) Additionally, 17-OHPC was also linked to rising cancer rates in neonates exposed in utero.
Further investigation suggests that the differences between the results of the 2019 and 2003 studies may be related to fatal weaknesses in the design of the previous trial. Randomization in the 2003 study appears to have been flawed, as participants in the placebo group had a higher risk of preterm labor than those in the 17-OHPC group, and the placebo group had a much higher rate of premature births than expected. at 55% (only around 10% of live births in the United States are premature).
The manufacturer rejects
Covis Pharma insists that the drug should stay on the market partly because the 2003 trial had a higher percentage of black participants than the confirmatory trial and because of post hoc analysis of patient data. There were major differences in the patient populations and maternal risk factors of those enrolled in the 2003 trial and the 2019 study. The smaller 2003 trial recruited a high percentage of non-Hispanic black patients (59%) compared to the 2019 trial, which recruited only 7%. The majority of patients in the 2019 trial were Caucasian (88%), with most coming from trial sites in Eastern Europe. The manufacturer claims that these population differences are sufficient to explain the differences in test results. As the claimed manufacturerthe 2019 trial is “not a negative study, but rather, due to major differences in study population recruitment, was inherently unable to confirm the [2003] conclusions of the trial.”
It is worth wondering why the manufacturer would pursue a study which, by its own admission, now claims to be unable to confirm the drug’s safety and benefits. More importantly, there are analytical reasons to believe that the demographics of the 2019 study do not invalidate its negative conclusion. For example, the FDA has performed subgroup analyzes to assess whether demographic or socioeconomic characteristics were effect modifiers, and concluded that race was unlikely to explain the differences.
Covis Pharma also cited routine clinical use data for the drug showing an association between its use and benefits for black mothers who have already given birth at less than 34 weeks. However, we must evaluate this claim with great caution. The 2019 study recruited a patient population in which 46% had had at least one spontaneous preterm birth at less than 34 weeks – an incredibly high percentage of women represented with this birth history – suggesting that race is probably not not the factor that predisposed these mothers to preterm birth. Moreover, there are no physiological mechanisms or other biological reasons to believe that 17-OHPC would have a different impact depending on the race. emerging population health literature showed that there is wide and varied genomic variability within American racial categories that generally makes race a bad power of attorney for genetic similarity.
Perhaps most important in evaluating the manufacturer’s claim, we must consider the precedent that would be set if a manufacturer could overturn a negative clinical trial finding by citing their own analysis of observational data. This approach goes against the very basis of FDA drug approval.
Economic incentives are likely at play here. While Covis Pharma fights to keep the drug on the market because of its purported clinical benefit for the underserved, the company’s pricing decisions appear to run counter to that goal. Prior to FDA approval of 17-OHPC in 2011, the active ingredient 17 alpha-hydroxyprogesterone was available as a compound product sold for around $15 (the active ingredient was off-patent, having been approved 55 years earlier). However, upon discovery of the potential use of 17-OHPC for the prevention of preterm labor, the drug was granted brand name status in 2011 and the manufacturer raised the price to over $15,000. It is particularly telling that prices remain high despite health equity rhetoric. In November 2022, according to IBM Micromedex RED BOOK and pharmaceutical retail websitesthe annual price of the Makena brand was still around $10,000 per year.
A call for greater FDA authority
Today, as we rapidly approach the end of 2022, 17-OHPC remains on the market. The FDA needs to take the necessary steps to change that. The case highlights areas for reform in the FDA’s fast-track approval pathway, particularly its reliance on confirmatory testing as a regulatory tool. An enforceable requirement for manufacturers to fulfill their confirmatory testing obligations is essential, and if the trials are not completed or the drugs are found to be unsafe or ineffective, the FDA must be able to withdraw the drugs more effectively. of the market. Similarly, there must be mechanisms to ensure that all confirmatory trials are properly designed to test real clinical parameters and have patient populations comparable to the populations of the original trial.
Achieving equity in health care delivery and the pharmaceutical market is a necessary goal. But, as the ongoing experiment with 17-OHPC shows, pharmacoequity is not achieved by specious claims about racial equity or by prices that only exacerbate existing disparities in access to drugs. care.
Frazer Tessema is a third-year medical student at the University of Chicago Pritzker School of Medicine and a member of the Program On Regulation, Therapeutics, And Law (PORTAL). Walls of Melinica is a third-year medical student at the University of Chicago Pritzker School of Medicine. Aaron S. Kesselheim, MD, JD, MPH, is the Program Director of Regulation, Therapeutics, and Law (PORTAL) in the Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine at Brigham and Women’s Hospital/Harvard Medical School.