Outcomes of pregnancies in sickle cell patients treated with hydroxyurea : Findings from the escort-HU cohort studies

Abstract Background: Hydroxyurea (HU) is a widely used therapy for sickle cell disease (SCD), reducing the frequency of vaso-occlusive crises (VOCs), acute chest syndrome (ACS), hospitalizations, transfusions and mortality. HU is approved in the EU and USA for patients over two years old. SCD patients planning pregnancy are generally advised to discontinue HU 3 to 6 months before conception. However, for patients with a history of delayed hemolytic transfusion reactions (DHTR), for whom transfusion is not an option, stopping HU treatment can be particularly risky. Nevertheless, evidence regarding HU-associated teratogenicity in humans remains limited and inconclusive. Larger datasets are needed to better assess pregnancy outcomes following maternal and fetal HU exposure. Methods: ESCORT-HU (European Sickle Cell Disease COhoRT-HydroxyUrea) and ESCORT-HU Extension are prospective, multicenter, non-interventional cohort studies conducted in France, Greece, Germany, and Italy as part of a post-marketing commitment to the European Medicines Agency following the approval of HU in SCD. ESCORT-HU study (2009-2019) and ESCORT-HU Extension study (2020–2025) aims to characterize long-term safety in adults and children, focusing on poorly documented or unknown risks including pregnancies. In both studies, pregnancies were prospectively documented regardless of HU interruption, with outcomes recorded as live births, non-live births, or ongoing at data cutoff. Results: A total of 3145 patients were enrolled in 77 centers, with 1412 (45%) men and 1733 (55%) women, two-thirds aged 15 to 49. During the studies, 246 pregnancies occurred in 202 women (125 in Escort-HU and 121 in Escort-HU Extension). The mean maternal age at pregnancy was 31.3 +/- 5.6 years, with a median HU exposure duration before conception of 3.8 [1.6-6.6] years in ESCORT-HU and 10.0 [6.1-13.8] years in the extension study. Among 246 pregnancies, 213 (87%) of which occurred under HU at a mean dose of 16.5+/-7.5 mg/kg/day but were stopped in the majority of cases during the first trimester. Only in 30 cases, HU treatment was interrupted before pregnancy and not resumed during pregnancy (no information was available in 3 patients). Of the 213 pregnancies under HU, 18 were ongoing and 23 (11%) resulted in voluntary abortions. Of the remaining 172 pregnancies, 131 (76%) resulted in live births, 35 (27%) premature births, 28 (16%) miscarriages, 2 discontinuation for medical reason, and 2 fetal deaths (one in each study period). Nine had an unknown outcome. No maternal deaths were reported. Among patients who stopped their HU before pregnancy, outcomes included 19 (65%) normal births, one premature birth, nine miscarriages, and one ongoing pregnancy. No malformations of the newborns related to HU were reported. Transfusion support during pregnancy was reported in 41% of pregnancies. Conclusion Fertility in women treated with HU appears to be preserved, and no specific issues related to HU exposure have been reported to date in newborns. Most pregnancies occurred under HU treatment, suggesting a high rate of unplanned pregnancies. In the ESCORT-HU studies, 76% of pregnancies resulted in live births, with no maternal mortality despite severe SCD cases. No malformations in newborns were reported as being related to HU exposure in these studies. Although discontinuation of HU is recommended when pregnancy is planned. HU may be continued when transfusion is not an alternative, particularly in patients with a history of DHTR. Further studies with larger datasets are needed to better assess HU exposure during pregnancy and the long-term outcomes in exposed children