Inadequate evaluation of potential side effects of sildenafil in preterm stopped Dutch trial

1. Introduction

A small proportion (est. <0,5%) of pregnancies are complicated by severe early onset growth restriction (FGR) caused by placental insufficiency [1,2]. Placental insufficiency is the result of abnormal formation and function of the placenta (placentation) with inadequate remodelling of the maternal spiral (utero-placental) arteries. Fetal growth restriction is associated with a high risk of perinatal morbidity with long-lasting sequelae and mortality. There is currently no therapy available with demonstrated effectiveness.

The Dutch Sildenafil TheRapy In Dismal prognosis Early-onset fetal growth Restriction (STRIDER) study investigated the efficacy of oral sildenafil compared to placebo in pregnant women with placental insufficiency bearing foetuses with severe, early-onset fetal growth restriction [3]. In this paper we describe the study rationale and course, unaddressed concerns, discourse and finish with a conclusion.

2. Study rationale and course

From Doppler waveform analysis of pregnancies complicated by severe fetal growth retardation it was concluded that the utero-placental circulation was compromised and that the placenta was hypo-perfused [4,5]. It was hypothesized that sildenafil, a phosphodiesterase inhibitor that potentiates the action of nitric oxide and is licensed for the treatment of erectile dysfunction in men [6], causes vasodilatation and might improve utero-placental circulation and perfusion. Through improved gaseous and nutrient exchange this might improve fetal growth and well-being.

There is some evidence that suggests that sildenafil improves utero-placental blood flow, growth, and meaningful outcomes [2]. The available evidence is rather limited as only in a small observational study with 27 patients there was a tendency towards more live-born children with survival intact to primary discharge for women treated with sildenafil when compared to women with pregnancies at similar risk but not receiving sildenafil [2]. STRIDER-researchers state that sildenafil is increasingly used in pregnancy for maternal cardiac conditions with no reports of adverse maternal or fetal effects [1].

The study started in 2015 and 183 women were included of which 93 women were allocated to sildenafil oral 25 mg 3 times daily. The study was stopped early in July 2018 because 19 babies died whose mothers had received sildenafil compared to 9 in the placebo-group. The babies died in the first weeks of their lives.

3. Unaddressed concerns

The discontinuation of STRIDER gained a lot of media and scientific attention [7–11]. The role and the conduct of the Medical Ethical Commission (METC) and the Central Commission on Medical Studies (CCMO) in the evaluation of the study protocol and the informed consent procedure did not gain the attention it deserved. These relate to inadequate evaluation of the literature and relying solely on the references the researchers provided to them.

Patients participating in an experimental drug study have the right to be informed adequately about possible gains and risks (informed consent). The researchers in this study wrote in the patient information leaflet that there is only very limited information available on what effects sildenafil exerts on pregnant woman, the placenta and the unborn but that there are some signs that it may be beneficial [12,13].

However, the manner in which side effects were dealt with in that letter was problematic. The researchers solely describe the side effects listed in the Summary of Product Characteristics (SmPC) also known as the product information or package insert of sildenafil licensed for the treatment of erectile dysfunction in men [6]. Patients read that the most frequent reported side effects of sildenafil are headache, dizziness, flushing and rhinitis. When reading this mild side effect-profile many might think that this drug is quite harmless and if it might result in a greater chance that their baby will survive, it is worth trying: If it does not work, it surely does not harm them or their babies.

Sildenafil, however, at the time of the study had been on the market for almost 20 years and many studies have been performed and many patients and researchers have used and studied the side effects of this drug. Most of this data is freely available. For instance on the website RxISK.org [13] and in the journals of the International Society of Drug Bulletins (ISDB) [14] many information on the side effects of drugs are available. RxISK.org mentions more than 100,000 accounts of side effects of sildenafil that were registered by the American regulating authority Food and Drug Administration (FDA). They include myocardial infarction, serious heart problems and death. From the literature it is well known that the information available in the summary of product characteristics does not reflect the scientific knowledge that is available in the community [15]. The information from RxISK.org and ISDB is not available in the summary of product characteristics of sildenafil because fetal growth retardation is an off-label indication and information on these indications are not listed.

Sildenafil has also been studied in children with pulmonary arterial hypertension. In a randomized double-blind and placebo-controlled trial 243 patients with pulmonary arterial hypertension (STARTS-1) were studied in a dose-ranging trial of oral sildenafil during 16 weeks with a follow-up of three years [16]. The authors concluded that sixteen-week sildenafil monotherapy is well tolerated in paediatric patients with pulmonary arterial hypertension. The follow-up up to three years (STARTS-2) however showed that quite a lot of patients died and the explanation given was that the drug was administered in too high dosages (hazard ratio for death of high dose compared to low dose HR 3,5). The Food and Drug Administration advised not to recommend sildenafil for children [17].

In 32 neonatal animals (piglets) the effect of increasing dosages of sildenafil given intravenously on hemodynamics and oxygenation was studied [18]. 17 piglets had meconium-induced pulmonary hypertension and lung injury, 15 had not. In all animals sildenafil caused a reduction in mean pulmonary artery pressure and pulmonary vascular resistance at all doses. In the lung-injured group sildenafil also caused a dose-related increase in the oxygenation index. In this group it also reduced the systemic vascular resistance at the lowest dose. The authors conclude that sildenafil is a potent vasodilator and should be used with caution in the presence of lung injury [18].

Patients participating in an experimental drug trial might want to learn more about this before deciding about informed consent. Yet, in the case of STRIDER, they were not informed about this.

4. Discourse

After we had informed ourselves about the study and contacted the board of the Amsterdam Medical Centre and shared our above-mentioned concerns with them, we received a letter stating that the study was approved, that the researchers cohered to all standards and requirements and that there was sufficient independent control on the safety of the participants. As we felt this response was inadequate we contacted the Dutch Health Inspectorate (Inspectie Gezondheidszorg en Jeugd, IGJ) and informed them about our concerns regarding patient-safety. They started an inquiry into STRIDER and they wrote that they would incorporate our writings about STRIDER in their inquiry. The Health Inspectorate finished their inquiry in 2019. Their report is not in the public domain but we received a letter from the Health Inspectorate in which they explained their findings and conclusions [19]. They stated that they doubted, with the knowledge of today, whether the patients that participated in STRIDER were well enough informed that sildenafil could also have adverse effects that were not known at the start of STRIDER. They also stated that the written patient information regarding the possible side effects of sildenafil could have been more inclusive by pointing out the possible side effects and by explicitly referring to the package insert. But the Health Inspectorate also stated that they could not conclude that patients were informed incompletely. The Health Inspectorate furthermore wrote that the manner in which patients were informed about the aim of the study and side effects was documented inadequately [19].

5. Conclusion

The conclusion therefore should be that at the time the trial started in 2015 it was already known that sildenafil can cause major serious and life-threatening side effects. The STRIDER-researchers did not perform an adequate literature search, otherwise they would have found additional data regarding side effects and patient safety. Relying solely on the summary of product characteristics as the researchers and the Ethical Commissions did is inappropriate and scientifically inadequate and certainly not in the best interest of patients. Denying patients scientific trustworthy information in matters of life and death is unacceptable. For Ethical Commissions it is necessary that they perform their own literature search and make their own interpretations in order to fulfil their task properly and not to rely solely on the documents researchers present to them.

The Ethical Commissions and the researchers have risked weakening the trust patients have in the scientific medical community. The role of these Commissions should be clarified in order to prevent a repeat of such a disaster.