Impaired fetal growth can lead to stillbirth or neonatal death, and babies that survive are still at higher risk of infections and long-term problems.

Cavan Social/Alamy Stock Photo

Trials of impotence drug to help poorly growing fetuses are halted after infant deaths

AMSTERDAM—When gynecologist Wessel Ganzevoort received a request for an urgent meeting from the independent committee watching over his clinical trial last week, he thought it might be good news.

Ganzevoort leads a study to find out whether sildenafil could help poorly growing fetuses if given to their pregnant mothers; the drug dilates blood vessels and in theory could bring fetuses more oxygen and nutrients. A data and safety monitoring committee, however, had looked at the unblinded evidence halfway through the study and wanted to talk. Ganzevoort, who’s at the Amsterdam University Medical Center (AUMC), knew that probably meant one of two things: Either the drug worked so well that continuing to give a placebo to half of the mothers in the trial would be unethical, or the study had to be stopped because sildenafil was causing serious complications. Because previous studies had not shown any major side effects, Ganzevoort expected the former.

He was wrong. At the 19 July meeting, the committee told him that a very specific complication, persistent pulmonary hypertension—in which blood vessels in the baby’s lungs fail to open after birth—had occurred 17 times in the sildenafil-treated group, leading to 11 deaths. There were only three cases of the condition in the placebo group and no deaths. (In total, there were 19 deaths in the treatment group versus nine in the placebo group—the latter not being a surprise given all the fetuses were at high risk because of their poor growth.) “It took us very little time to make the decision to halt the study,” Ganzevoort says. That same day, he and his colleagues began to call the women who participated in the study, including several who are still pregnant or whose babies are in the hospital.

Researchers at the University of British Columbia in Vancouver, Canada, have stopped a similar study, pending an in-depth investigation of what happened in the Netherlands. But the sudden halt likely means the end of the road for what many thought was a promising therapeutic approach, says AUMC gynecologist Ben Willem Mol, who co-initiated the study.

“Fetal growth restriction” is caused by a lack of blood flow from the placenta to the unborn child, resulting in undernourishment and stunted growth and development. It can lead to stillbirth or neonatal death, and babies that survive are still at higher risk for infections and often suffer from long-term problems such as obesity and cardiovascular disease. Growth restriction is diagnosed using ultrasound, but now the only medical treatment is closely monitoring the pregnancy and inducing birth when the risk of stillbirth is deemed high. That poses a difficult dilemma for doctors: Inducing birth too early increases the risk of complications, but waiting too long could lead to developmental anomalies or stillbirth.

Increasing placental blood flow could spur fetal growth, which is why a growing number of doctors around the world are prescribing sildenafil to pregnant women whose fetuses suffer from impaired growth. The drug is known to dilate some blood vessels, including those in the penis—which is how it became a blockbuster drug for erectile dysfunction—and a number of animal studies and small human trials suggested the drug might benefit unborn children. Yet it is not approved for such use in pregnant women by any regulatory body, so physicians have been using it in a so-called off-label manner.

Ganzevoort and a number of other experts in obstetrics wanted to firm up the evidence on the issue. In 2012, they started an initiative to conduct five separate but very similar trials that would culminate in a meta-analysis.

The studies began in 2015, and even before last week’s dramatic decision, the results so far had been disappointing. A study by a team in the United Kingdom, published online in The Lancet Child & Adolescent Health in December 2017, found sildenafil didn’t improve the duration of pregnancy, birth weight, or fetal and neonatal survival. The team did not observe any negative effects of the drug, except for a decrease in the blood flow from placenta to fetus through a shunt between the placenta and fetus called the ductus venosus. “This was an unexpected result and it was the first evidence for a potential adverse effect of the treatment,” says Gordon Smith of the University of Cambridge in the United Kingdom, who was not involved in the study.

A joint Australian–New Zealand team whose results from a similar trial were presented at a recent meeting didn’t find any benefits either, but also didn’t see complications, says study leader Katie Groom of the University of Auckland in New Zealand. A trial in Ireland has yet to start enrollment; researchers in that study did not respond to emails today, but Mol expects it will be canceled.

The Dutch team held out hope for a more positive result, until the call last week. “There were no signs of serious harm, so we continued with our plan,” Ganzevoort says.

The news was devastating to some study participants. “We were at peace with the death of our daughter, but now we no longer are,” one mother said on Dutch public television. It also raised questions about the information parents received about the study’s risks. The informed consent form they were asked to sign does not mention any potential negative effects for the child. “We did state that sildenafil normally is not prescribed during pregnancy, and that no strange things had been observed in this indication, which is a fact,” Ganzevoort says, “but we could have stated more specifically that we don’t know what we don’t know, as I did during all the counseling I did myself.” 

How the drug could have caused the complications is unclear. The Dutch team plans to investigate all of the cases of lung hypertension and the neonatal deaths it caused to check whether the diagnosis was correct and whether there were other specific characteristics among this group. One possible explanation is a “rebound effect,” Groom says. Some of the drug probably reached the fetus before birth, where it may have put dilating forces on the lung vessels, she says; that might have led to an increase of constricting signals from the fetus itself to prevent them from opening prematurely. “After birth, the child stopped receiving the dilating drug, potentially causing a constriction of the lung vessels,” Groom speculates.

It’s also unclear why the deaths happened only in the Dutch trial. Study inclusion criteria in the United Kingdom, Australia, New Zealand, and the Netherlands differed slightly, but not enough to explain the varying outcomes, Groom says. Researchers in the two other trials plan to check whether they may have missed cases of lung hypertension in the newborns.

Given the small numbers of deaths, it’s also possible that the outcome in the Netherlands is entirely due to chance, although Ganzevoort says the odds of that are lower than 5%. “Combining our results with those of the other trials and zooming in on subgroups might give more clarity,” he says.

The decision to halt the trial is a blow to the field, Mol says. “There were good reasons to believe this drug would work and it was responsible to run this trial,” he says. “I get emotional talking about it because we now have 11 grieving families and it does not help those people.”

There is always a risk in clinical trials, however, says Indira van der Zande, who recently finished her Ph.D. thesis on pregnant women in clinical research at the University Medical Center Utrecht in the Netherlands. “But we should always keep in mind that by not conducting this study, there would have been a risk as well. Doctors might have prescribed this drug for years off-label, assuming it was safe and effective.”