Sildenafil and bosentan are used to treat pulmonary arterial hypertension (PAH). Sildenafil (molecular weight: 475 g/mol) is an inhibitor of cyclic guanosine monophosphate specific phosphodiesterase type 5. Sildenafil is quickly absorbed (T_max: 60 min), with a mean absolute oral bioavailability of 41%. Sildenafil is 96% bound to plasma proteins, and metabolized primarily by the enzymes cytochrome P450 (CYP)3A4 and CYP2C9, followed by excretion in feces and urine. The terminal phase half-life is approximately 4 h. Bosentan (molecular weight: 552 g/mol) is an endothelin receptor antagonist, has an absolute bioavailability of 50%, and is highly bound to plasma proteins (>98%). It is metabolized by CYP2C9 and CYP3A4, followed by biliary excretion. The terminal elimination half-life is 5.4 h. The drug-drug interaction between bosentan (125 mg, twice-daily) and sildenafil (80 mg, three times daily) resulted among healthy volunteers in a reduction of 63% in the area under the plasma concentration-time curve (AUC) of sildenafil and an increase of 50% in the AUC of bosentan (Sandoz nv/sa, 2020; Accord Healthcare, 2021).

Pregnancy is contra-indicated for women with PAH (Luo et al., 2020). Consequently, there is little experience with the combined use of sildenafil and bosentan during pregnancy and lactation. Breastfeeding is not recommended during treatment with bosentan, since the extent of transfer into human milk is unknown (Bosentan, 2006). A case report of a 23-year old woman treated with sildenafil and bosentan during pregnancy and postpartum concluded that the infant was nursed with good outcome, although the doses and extent of breastfeeding were not specified (Molelekwa et al., 2005). Another case report of a woman treated with sildenafil during pregnancy and lactation suggested that secretion of sildenafil in human milk is limited (1.64—4.49 μg/L) (Wollein et al., 2016).

Breastfeeding plays an important role in the development and survival of infants, and is associated with positive health effects for mother and nursing infant (Binns et al., 2016). The World Health Organization ¹ recommends exclusive breastfeeding during the first 6 months of life, and continued breastfeeding in combination with safe and adequate complementary foods for up to 2 years and beyond. Although the transfer of most medicines into human milk is expected to be limited, this area has historically only received little attention. Off-label use of medicines during lactation exposes neonates and infants to unknown health risks. Simultaneously, women might decide to discontinue breastfeeding or to avoid initiation of their pharmacotherapy. The case report fits in the Innovative Medicines Initiative ConcePTION ² , which aims to reduce the uncertainty about the effects of medicines used during pregnancy and lactation.

The aim of this case report was to determine human milk concentrations of bosentan and sildenafil in a lactating mother, to estimate infant exposure, and to provide the reported general health of the nursing infant until the last sampling moment.

This case report was prepared along the CARE Guidelines (Gagnier et al., 2013).

This case report includes a single mother-infant pair. The mother was a 43-year old European woman, with a body mass index of 29.1 kg/m² (83 kg, 169 cm). Treatment with sildenafil and bosentan for PAH was initiated more than 6 months after delivery. Approval was obtained from the Ethics Committee Research UZ/KU Leuven (S64702), and the woman provided written informed consent. The mother did not smoke, nor used alcohol or was on a specific diet. She didn’t have renal or hepatic failure. The infant was male, born healthy and term. In the period of human milk sample collections, the infant received breastfeeding (± three times daily) in combination with adequate complementary foods. At that time, he was 21 months old, weighing 11 and 11.2 kg on the first and second sampling day respectively.

The sampling days are illustrated in Figure 1.

The mother was treated with Balcoga (sildenafil) 20 mg film-coated tablets (Sandoz nv/sa, Vilvoorde, Belgium) three times daily, and Bosentan Accord 125 mg film-coated tablets (Healthcare B. V., Utrecht, the Netherlands) twice-daily (Table 1). As co-medication, she used Vitamin D3 25 µg (1000 IE, oral spray, Holland and Berrett) each morning and occasionally Ibuprofen (400 mg). Dosing and timing of medicines used are provided in Table 1 and Figure 1.

This case report provides human milk levels of sildenafil and bosentan along with the reported health outcomes of the nursing infant. To our knowledge, there is no published information about the levels of bosentan in human milk. A single case report previously described the milk concentrations of sildenafil (1.64–4.49 μg/L) and N-desmethylsildenafil (1.18–1.82 μg/L) in a woman receiving multiple oral doses of 20 mg sildenafil daily for PAH (Wollein et al., 2016). We presently observed similar sildenafil concentrations ranging from 1.43 μg/L to 5.65 μg/L.

Another case report described the use of sildenafil and bosentan during pregnancy and postpartum in a 23-year old woman with Eisenmenger’s syndrome (Molelekwa et al., 2005). The doses or human milk concentrations were not reported. The authors stated that the infant was nursed for 11 weeks with good outcome. Unfortunately, the infant died from a respiratory syncytial virus infection at 26 weeks post-delivery (Molelekwa et al., 2005). In our case study, the mother did not report any adverse effects, health problem or hospitalization of the infant from birth until the last sampling day. Treatment with sildenafil and bosentan was initiated more than 6 months after delivery.

The RID in our case study for both sildenafil and bosentan were well below 1%, amounting to 0.06 and 0.24% for sildenafil and bosentan, respectively. Although questions can be raised about the use of an empirical cut-off value (typically 10%), it is highly unlikely that these very low RID’s would result in toxic exposure to the infant, especially for sildenafil as there are reassuring safety data in infants (Samiee-Zafarghandy et al., 2014). This was also confirmed by comparing the calculated DID to usual doses administered to infants for therapeutic reasons. Finally, when estimating infant systemic concentrations based on reported clearance values for these medicines in infants, negligible systemic exposure in the nursing infant through breastmilk (Equation (4)) was predicted. For sildenafil, the exposure might even be lower due to the drug-drug interaction with bosentan in the infant (Sandoz nv/sa, 2020). For bosentan, this interaction might lead to an increase in exposure, although still very low, also because clearance values in infants were obtained from studies in very young infants and therefore most likely underestimated (Steinhorn et al., 2016; Accord Healthcare B.V., 2021). Other medicines taken by the mother are not expected to interfere with the concentrations of sildenafil and bosentan.

This study has several limitations. First, it includes only a single mother-infant pair. In addition, the 21-months old infant was not exclusively breastfed. Human milk volume and composition in this setting and period of lactation also differs from mature milk during exclusive breastfeeding (Casey, 1989). During the first days postpartum, the immature blood-milk barrier might result in higher concentrations of medicines in colostrum (Nguyen and Neville, 1998). At the same time, the volume of colostrum ingested by the infant is much lower compared to mature human milk. Furthermore, infants have immature pharmacokinetic and pharmacodynamic processes in the first 6 months (van den Anker et al., 2018). Both sildenafil and bosentan are metabolized by CYP enzymes to (partially) active metabolites. Genetic polymorphisms in CYP3A4 and CYP2C9 are expected to contribute to interindividual variabilities in exposure to these medicines and their metabolites (Chen et al., 2014; Tang et al., 2020). Physiologically-based pharmacokinetic modelling and simulation could be applied to predict the possible impact of such polymorphisms on pharmacokinetics of these medicines. The genotypes of the mother-infant pair for these enzymes were not known in our case report. Consequently, it is not recommended to extrapolate these data to the entire population of lactating mothers and their infants. However, as there is currently almost no data available about the use of these two medicines during breastfeeding, this case contributes to the current knowledge supporting clinicians to make evidence-based decisions in practice. Second, blood samples from the mother-infant pair for medicine quantification have not been collected. Although using a structured (sampling) approach for human milk, the number of samples collected remains limited. However, samples were collected under steady-state conditions, limiting fluctuations in concentrations. Finally, a LC-MS/MS method was developed and able to quantify bosentan and sildenafil in human milk, although their main metabolites were not determined in the present study (Supplementary Data). The contribution of N-desmethylsildenafil to the pharmacological activity is 36% compared to the parent compound (Sandoz nv/sa, 2020). N-desmethylsildenafil is highly bound to plasma proteins and is further metabolized. Similarly, the active metabolite of bosentan is expected to contribute less than 20% to the total pharmacological activity (Accord Healthcare B.V., 2021). The metabolite is excreted into the bile. Overall, based on their lower lipophilicity compared to the parent compounds, metabolites are expected to show lower transfer rates across the blood-milk barrier.

In summary, the concentrations of both sildenafil and bosentan measured in human milk collected at 21 months after delivery were very low, especially when compared to the maternal and commonly used therapeutic infant doses. In addition, in this case study no negative effect on the general health of the nursing infant was reported by the mother.