“Confirmatory trial” creates uncertainty about benefit of 17 alpha-hydroxyprogesterone for prevention of premature birth

EBM Focus - Volume 14, Issue 23

Reference: Am J Perinatol 2019 Oct 25;early online (level 2 [mid-level] evidence)

Approximately 1 in 8 pregnancies results in birth before 37 weeks in the United States. The Maternal-Fetal Medicine Units (MFMU) Network trial published in NEJM in 2003 demonstrated such high efficacy for 17 alpha-hydroxyprogesterone caproate (17-P) for prevention of recurrent preterm birth that it was considered unethical to continue enrollment and the trial was stopped early for benefit. After a different 2003 trial demonstrated similar efficacy for vaginal progesterone, the American College of Gynecologists and the Society for Maternal Fetal Medicine (SMFM) put out statements recommending routine use of progesterone for women with singleton pregnancy and history of preterm birth. Physicians began prescribing off-label 17-P and it became the standard of care. In 2011, the FDA granted conditional approval and orphan drug exclusivity for the brand-name product Makena and prescriptions for the costly drug skyrocketed. The conditional approval for the drug, however, required that a confirmatory trial be completed using a protocol that mirrored the original MFMU trial.

The confirmatory trial, called PROLONG, was a double-blind, randomized, placebo-controlled trial conducted primarily in Russia, Ukraine, and the United States. Notably, recruitment in the US was difficult compared to the MFMU trial. The authors hypothesize that the MFMU trial results changed the standard of care, and this discouraged enrollment in a trial where one arm would receive placebo. A total of 1,740 women (compared to only 463 in the MFMU trial) with a history of preterm birth between 20w0d and 36w6d were randomized 2:1 to weekly 17-P or placebo beginning at 16 weeks gestation. The sample size for the confirmatory trial was, as is usual fashion, determined based in part on expected event rates demonstrated in previous trials. Most of the participants were Caucasian (87%), married (90%), and non-smokers (92%), which represents a lower-risk cohort than in the MFMU trial. Event rates were lower than predicted, leading to an inadequately powered study. No difference was found in either of the co-primary outcomes. The rate of preterm birth before 35w0d was 11% in the 17-P arm and 11.5% in the placebo arm (relative risk 0.95, 95% CI 0.71-1.26, p = 0.72). There were also no significant differences between the groups in adverse maternal or neonatal outcomes.

What do we do with this information? Despite more than a 3-fold higher overall enrollment, the PROLONG trial was inadequately powered, suggesting that the lack of demonstrated efficacy might reflect type II error rather than a lack of benefit of 17-P. In other words, this FDA-approved medicine which costs upwards of $20,000/pregnancy in the United States is the current standard of care, but failed to demonstrate a benefit in a mandated confirmatory trial. Furthermore, the original MFMU trial was itself underpowered to detect neonatal morbidity and mortality because it was stopped early, so we cannot assure long-term medication safety. The PROLONG trial started recruitment in 2009, and potentially another decade could pass before more definitive trials are completed. We are left with more questions than answers.

For more information, see the topic Prevention of Preterm Labor and Preterm Birth in DynaMed.

DynaMed EBM Focus Editorial Team

This EBM Focus was
written by Carina Brown, MD, Assistant Professor at Cone Health Family Medicine
Residency, and Katharine DeGeorge, MD, MS, Associate Professor of Family
Medicine at the University of Virginia and Clinical Editor at DynaMed. Edited
by Alan Ehrlich, MD, Executive Editor at DynaMed and Associate Professor in
Family Medicine at the University of Massachusetts Medical School.