Eculizumab Treatment Improves Outcomes of Pregnancy in Patients with Paroxysmal Nocturnal Hemoglobinuria

Paroxysmal nocturnal hemoglobinuria (PNH) is a rare, acquired clonal stem cell disorder. Pregnancy in patients with PNH results in an increase in maternal and fetal complications and a consequent increase in maternal and fetal mortality. Maternal complications include thrombosis, cytopenias and infections while fetal complications relate to preterm delivery (de Guibert et al, 2011). Eculizumab is a humanized monoclonal antibody that blocks terminal complement activation, preventing intravascular hemolysis and its consequences. In PNH it reduces the need for transfusions, protects against thrombosis, and may improve long-term survival. We previously showed a benefit in using eculizumab in pregnancy for patients with PNH in a limited number of cases (Kelly et al, BJH 2011).

This study was of 70 pregnancies in 57 women. Patients were identified through physician willingness to participate as well as through the Global PNH Registry. A specific questionnaire was sent to physicians. Local IRBs approved the study.

Across the 70 pregnancies, 41 women were on eculizumab before conceiving and remained on the drug during pregnancy, 28 women started eculizumab in either the second or third trimester and 1 woman stopped eculizumab at 12 weeks gestation.

The median age at PNH diagnosis was 23 years (range 13-36). The median age at the start of each pregnancy was 29 years (range 18-40). Nineteen women (33%) had a prior documented history of aplastic anemia and 8 (14%) had a prior thrombosis. The median PNH granulocyte clone around the start of pregnancy was assessed in 51 cases and was 94.3% (range 24.3-100).

There were 62 live births, 2 stillbirths and 6 first trimester miscarriages. There were 3 twin pregnancies. In 1 of these, there was a fetal death due to intrauterine growth retardation (IUGR).

Anticoagulation (AC) with low molecular weight heparin (LMWH) was used in 60 pregnancies (86%) and with fondaparinux in 1 case: therapeutic doses were used in 26 pregnancies, prophylactic doses in 28 pregnancies and an intermediate dose in a further 7. Aspirin was used in 4 pregnancies. Folic acid and oral iron supplements were used in 62 (89%) and 26 (41%) pregnancies, respectively.

Transfusion requirements increased in pregnancy from a mean of 0.13 units per month in the 6 months before conception to a mean of 0.94 units per month whilst pregnant. This returned to pre-pregnancy levels after delivery. No platelet transfusions were needed in the 6 months before being pregnant compared with 99 platelet transfusions during the pregnancies.

Higher doses, or more frequent dosing, of eculizumab was used to treat recurrent intravascular hemolysis in over half the pregnancies (52%) that progressed past the first trimester. Eculizumab was stopped in 10 instances after the postpartum period, 9 due to funding issues and 1 as the patient underwent a bone marrow transplant. Two patients who stopped eculizumab 12 weeks after delivery experienced a thrombosis. The first experienced a mesenteric thrombosis 4 weeks after stopping eculizumab whilst on therapeutic dose of LMWH and the second suffered a Budd-Chiari 8 weeks after stopping eculizumab. Both were recommenced on eculizumab immediately.

The mean reduction in platelet count from the start of pregnancy to delivery was 37 x 10⁹/l. There were 10 significant bleeding episodes: 1 recurrent epistaxis, 1 antepartum bleed and 8 postpartum bleeds. Two patients experienced a postpartum thrombosis whilst on eculizumab, 1 a deep vein thrombosis and the other a mesenteric thrombosis, likely precipitated by a plasma infusion. Nineteen births were premature (31%), mainly due to pre-eclampsia (6 cases) and IUGR (5 cases). Four babies had complications due to prematurity: 1 had toxic megacolon and required a temporary ileostomy, and the other 3 had initial growth retardation due to prematurity.

Twenty-five babies were breast fed and in 10 cases, breast milk was tested for eculizumab but no drug was detected. Eculizumab was not detected in 13 cord blood samples and was found at very low levels in a further 7 samples (11.8-21.2µg/ml).

In conclusion, eculizumab appears safe to use in pregnancy in PNH and does not appear to cross the placenta in significant quantities to block complement or to be excreted in breast milk. Higher doses may be required later in pregnancy to prevent hemolysis. Overall pregnancy outcomes in this group are better than historical controls with supportive therapies alone without eculizumab.