Pregnancy in a Kidney Transplant Woman Under Treatment With Eculizumab for Atypical Hemolytic Uremic Syndrome: Is It Safe?

Introduction

Eculizumab, a humanized IgG2/4 chimeric monoclonal antibody against complement protein C5, is the first drug that specifically targets complement activation. Initially approved for paroxysmal nocturnal hemoglobinuria (PNH), its indications have been subsequently extended to the treatment of atypical hemolytic uremic syndrome (aHUS).¹ Currently, eculizumab also is being used for treating or preventing aHUS recurrences after kidney transplantation, with a significant positive impact on clinical outcomes and quality of life. Nonetheless, published data on pregnancy outcomes of women treated with eculizumab remain scarce. To our knowledge, isolated case reports,S1–S7 small case series,S8,S9 and a cohort study² have specifically described the pregnancy course of women with PNH treated with eculizumab. In general, the published literature has shown acceptable outcomes, with high fetal survival rates and rare maternal complications. However, the risk of adverse pregnancy outcomes may be higher in nontransplanted women with aHUS who are being treated with eculizumab. Here, we describe the first case of a kidney transplant recipient who developed aHUS and became pregnant while being treated with eculizumab. In addition, we reviewed the available literature on pregnancies in nontransplanted women suffering from aHUS who received antiC5 antibody, with a special focus on maternal and fetal outcomes.

Case Presentation

In 2007, a 24-year-old woman presented with nephrotic syndrome and acute renal failure. The results of kidney biopsy revealed an IgA nephropathy with extracapillary proliferation, which was treated for 12 months with steroids and inhibitors of the renin-angiotensin system. After 5 years, the patient developed a pyelonephritis complicated with sepsis, accompanied by biological signs of thrombotic microangiopathy (serum creatinine 450 μmol/l, haptoglobin <0.08 g/l, lactate dehydrogenase 965 UI/l, hemoglobin 8.5 g/dl, rare schizocytes). Biopsy findings confirmed an IgA nephropathy associated with features of thrombotic microangiopathy (Figure 1). There was evidence of arteriolar thrombosis without extracapillary proliferation. Signs of thrombotic microangiopathy regressed completely on resolution of infection, and renal function improved (serum creatinine at 200 μmol/l). Unfortunately, her renal function progressively deteriorated thereafter, and she was diagnosed with end-stage renal disease 3 years later. Phenotypic analyses of the alternative complement activation pathway did not reveal abnormal findings. Moreover, no mutations in the CFH, CFI, CFB, CFHR5, MCP, C3, DGKe, and thrombomodulin genes were identified. In 2015, the patient underwent preemptive kidney transplantation from a deceased donor. Donor-specific antibodies were absent. Immunosuppression was based on thymoglobulin, cyclosporin, mycophenolate mofetil, and steroids. During the first posttransplantation days, the patient showed signs of severe thrombotic microangiopathy characterized by hemolytic anemia with schizocytosis >1% requiring blood transfusions, marked thrombocytopenia, and acute kidney failure (Figure 2). Despite treatment with daily plasma exchange, progressive hemolysis was evident 3 days thereafter. The patient was therefore started on eculizumab on the sixth posttransplant day. After induction with 900 mg weekly for 4 weeks, eculizumab was maintained at a dosage of 1200 mg every 2 weeks. After 1 week of eculizumab treatment, signs of hemolysis started to resolve, and graft function gradually recovered. At 3 months after transplantation, the patient’s serum creatinine and estimated glomerular filtration rate were 100 μmol/l and 60 ml/min per 1.73 m², respectively. At 5 months, a routine graft biopsy was performed. The biopsy identified infra-clinical features of microangiopathy with arteriolar and glomerular lesions and borderline cellular rejection (Figure 3). The patient was given oral steroids (1 mg/kg per day), which were subsequently tapered off while continuing eculizumab.

Figure 1.

Biopsy findings in the native kidneys. (a) Biopsy of the patient’s native kidney revealed mesangial IgA deposits. Granular IgA staining was observed in peripheral capillary loops in association with a proliferative form of IgA nephropathy (immunofluorescent staining, original magnification ×400). (b) Evidence of arteriolar thrombosis (immunofluorescent IgM staining, original magnification ×400).

Figure 2.

Evidence of thrombotic microangiopathy occurring after transplantation. Plasma exchange (PE) was performed on a daily basis from day 1 (D1) to day 6 (D6). Treatment with eculizumab was started on D6. Biomarkers of microangiopathy turned negative within 1 month of transplantation. Schizocytes were identified on D1 (1%−2%) and remained detectable until D30 after transplantation. LDH, lactate dehydrogenase.

Figure 3.

Biopsy findings in the kidney graft at 5 months of transplantation. (a) Arteriolar lesions indicating thrombotic microangiopathy characterized by the presence of fibrin thrombi (*) within the arteriolar lumen (hematoxylin-eosin staining, original magnification ×400). (b) Glomerular lesions indicating thrombotic microangiopathy characterized by the presence of fibrin thrombi within the glomerular capillary loop (arrows). Presence of hemorrhagic effusions (Masson’s trichrome, original magnification ×400).

At 1 year of transplantation, the patient expressed her desire to have a child. Owing to an optimal complement blockage and high trough levels (Table 1, Table 2), eculizumab dosing was reduced to 900 mg every 2 weeks. Graft function was stable (serum creatinine: 82 μmol/l, estimated glomerular filtration rate: 80 ml/min per 1.73 m²), with haptoglobin, platelet count, and red blood cells being all within the normal range. There was no elevation of lactate dehydrogenase and schizocytes were absent. The patient still showed an efficient complement blockade (total complement <14 U/ml) and therapeutic eculizumab trough levels (Table 1, Table 2). A novel routine graft biopsy revealed mesangial deposits of IgA without microangiopathy lesions. Under these circumstances, pregnancy was not medically contraindicated. Her immunosuppressive therapy was therefore modified by switching mycophenolate mofetil to azathioprine. At 21 months of transplantation and while still receiving eculizumab, the patient became pregnant. Eculizumab doses were increased from 900 mg to 1200 mg every 2 weeks during gestation and trough levels remained within the therapeutic range (Table 2). Biochemical screening for thrombotic microangiopathy (lactate dehydrogenase, haptoglobin, hemoglobin, platelet count, presence of schizocytes) was performed biweekly and yielded normal results throughout the entire pregnancy course (Figure 4). Total complement activity (CH50) was tested on a monthly basis and confirmed continuing blockade of the complement pathway (Table 1). Serum creatinine levels remained stable and no serious maternal complications occurred during pregnancy. Urinalysis was negative for proteinuria. The patient’s weight varied from 49 to 56 kg from the beginning of the pregnancy until the delivery. Fetal growth measured on ultrasound remained within the normal range until week 27 of gestation (Figure 5). On week 28, proteinuria increased from 0.1 g/24 h to 0.3 g/24 h and there were signs of moderate preeclampsia (maternal stage 1 hypertension) followed by fetal growth faltering. Because of signs of fetal distress (heart rhythm abnormalities) at 32 weeks of gestation, the patient underwent a Cesarean delivery and delivered a normotrophic preterm healthy boy (weight at birth: 1550 g). The newborn was not breastfed. On histological examination, signs of moderate placental insufficiency (focal infarctions accompanied by microthrombi) were evident. Treatment with eculizumab was maintained after delivery and the patient did not show further episodes of hemolysis in the postpartum period. At 10 months after delivery, the patient’s serum creatinine and estimated glomerular filtration rate were 70 μmol/l and 97 ml/min per 1.73 m², respectively. No growth delay was observed, and no serious infections or adverse events were noticed. Both the mother and her child are currently in good health. Eculizumab was stopped 6 months after delivery with no recurrence of thrombotic microangiopathy after a 6-month follow-up.

Table 1.

Results of complement activity and haptoglobin level obtained during the second and third trimesters of pregnancy and during the early postpartum period

Biologic markers	Gestational week			Postpartum week
	22	26	30	3	7	9	11	19
CH50 (U/ml) (reference range: 38−79 U/ml)	<14	<14	<14	<14	<14	<14	<14	<14
Haptoglobin (g/l)	0.33	0.31	0.72	0.55	1.51	0.93	0.70	0.81

Assays of complement hemolytic activity (CH50) were performed on a monthly basis in the Immunology Laboratory of the Strasbourg University Hospital, Strasbourg, France. CH50 activity remained undetectable, reflecting an efficient complement blockade (<14 UI/l is the lowest limit of detection of the liposome immunoassay).

Table 2.

Dosages and serum levels of eculizumab observed before, during, and after pregnancy

Eculizumab	18 mo before pregnancy	12 mo before pregnancy	Wk 0 of pregnancy	Wk 12 of pregnancy	Wk 20 of pregnancy	Wk 24 of pregnancy	5 mo after delivery
Eculizumab dosage, mg (every 2 wk)	1200	900	900	1200	1200	1200	900
Serum eculizumab levels, μg/ml	2478	691	744	423	396	364	571

Serum eculizumab levels were measured with an enzyme-linked immunosorbent assay at the Immunology Laboratory, Georges-Pompidou European Hospital, Hospital and Public Assistance of Paris (AP-HP), Paris, France.

Figure 4.

Biomarkers of thrombotic microangiopathy during pregnancy. Schizocytes were undetectable throughout pregnancy. LDH, lactate dehydrogenase; WG, weeks’ gestation; WPP, weeks postpartum.

Figure 5.

Assessment of fetal growth on ultrasound. EFW, estimated fetal weight.

Literature Review

We conducted a PubMed search for papers published until August 2018 using a combination of the following keywords: “atypical hemolytic uremic syndrome,” “eculizumab,” and “pregnancy.” Cases in which eculizumab was started after delivery were not included. We also excluded duplicate publications on the same case. We identified 3 case reports3, 4, 5 and 2 small case series6, 7 that specifically focused on the pregnancy outcomes of women with aHUS who were treated with eculizumab. Therefore, 13 cases of pregnancies occurring during eculizumab treatment have so far been reported (including the patient described in the current study; Table 3). Eculizumab was used as a preventive strategy in 6 cases (Huerta et al.⁷ and our current patient), whereas it was given with curative intent in the remaining 7 patients. In the latter cases, eculizumab was started between gestational week 10 and week 36. Doses varied from 900 mg to 1500 mg fortnightly, or 900 to 1200 mg per week when given in a weekly administration. Maternal outcomes were generally good (with the exception of a single case) without evidence of hemolysis breakthroughs while under eculizumab treatment. Kidney function remained either stable or improved (when eculizumab was given with curative intent). In 1 case, blood pressure, serum creatinine levels, proteinuria, and anemia worsened at 29 weeks, with the patient developing a HELLP (hemolysis, elevated liver enzymes, low platelet count) syndrome in the postpartum period.⁶ Preeclampsia was observed in 4 patients. One fetus died in utero at gestational week 24. The remaining 12 women delivered a live neonate. The rate of premature births was high (89%) with a mean gestational age of 31.5 weeks. Fetal distress occurred in 2 cases. Mean infant weight was 1750 g (range: 500−3650 g). Although 2 newborns showed signs of growth retardation, the clinical course following the postpartum period was uneventful in all cases.

Table 3.

Detailed characteristics of the 13 pregnancies occurring in 11 women with aHUS under eculizumab treatment

Reference	Start of eculizumab treatment	Maximum eculizumab dosage	Maternal complications	Pregnancy outcome	Birth weight, g	Newborn complications	Evolution (FU)
Current case	From the beginning	1200 mg biweekly	Mild preeclampsia	Cesarean delivery at wk 32	1550	Mild growth retardation	Good (9 mo)
Ardissino et al.3	From wk 26	900 mg biweekly	No	Cesarean delivery at wk 38	3650	Healthy	Good (1 wk)
Demir et al.4	From wk 17	Biweekly	No	Cesarean delivery at wk 31	NA	Respiratory fetal distress	Good (2 wk)
Andries et al.5	From wk 10	1200 mg biweekly	No	Cesarean delivery at wk 36	NA	Healthy	Good (20 mo)
Servais et al.6
Pregnancy 1	From the beginning	1200 mg biweekly	Preeclampsia and postpartum HELLP syndrome	Cesarean delivery at wk 29	1550	Healthy	Good (3 yr)
Pregnancy 2	From the beginning	1200 mg weekly	Metrorrhagia	Cesarean delivery at wk 34	2500	Healthy	Good (2 yr)
Pregnancy 3	From the beginning	1500 mg biweekly	Preeclampsia	Cesarean delivery at wk 30	1410	Growth retardation	Good (6 mo)
Pregnancy 4	From the beginning	900 mg weekly	No	Intrauterine fetal death	NA	NA	NA
Pregnancy 5	From the beginning	1800 mg biweekly	Preeclampsia	Cesarean delivery at wk 30	1070	Growth retardation	Good (8 mo)
Huerta et al.7
Pregnancy 1	From wk 19	NA	No	Cesarean delivery at wk 24	500	Fetal distress	Discharged at 5 mo, healthy
Pregnancy 2	From wk 36	NA	NA	NA	NA	NA	NA
Pregnancy 3	From wk 36	NA	NA	NA	NA	NA	NA
Pregnancy 4	From wk 32	NA	NA	NA	NA	NA	NA

aHUS, atypical hemolytic uremic syndrome; FU, follow-up; HELLP, hemolysis, elevated liver enzymes, low platelet count; NA, unavailable data.

Discussion

Flares of aHUS are known to occur commonly during pregnancy and even more frequently in the postpartum period.S10 As a consequence, pregnancy has not traditionally been encouraged in women with a history of aHUS; however, acceptable pregnancy outcomes have been recently reported in women with PNH who received eculizumab during the course of pregnancy.² As far as aHUS is concerned, data from women who had undergone eculizumab treatment remain limited. Here, we report for the first time a case of successful pregnancy in a kidney transplant recipient who was being treated with eculizumab. Despite the absence of known aHUS genetic mutations in our patient, the risk of aHUS relapses during pregnancy seemed not negligible. Based on the available evidence and after adequate counseling with the couple, it was decided to maintain eculizumab therapy throughout pregnancy.

The gestational course of our patient was uneventful. Biological markers of microangiopathy were monitored on a monthly basis and no relapse of hemolysis was observed. Importantly, graft function remained constantly stable both during pregnancy and after delivery. In the published literature on pregnant women affected by aHUS who received eculizumab, 1 case of postpartum HELLP syndrome has been reported.⁶ Although the occurrence of preeclampsia has also been described in this patient group, this complication was likely related to the underlying condition (i.e., hypertension, chronic kidney disease) rather than to a relapsing aHUS per se. An insufficient complement blockade has been considered as the main risk factor for aHUS breakthroughs.⁶ Notably, previous pharmacokinetic and pharmacodynamic studies have shown that regular monitoring of both eculizumab levels and biomarkers of hemolysis is paramount to prevent relapses.⁸ Insufficient complement blockade during pregnancy requires an adjustment of eculizumab dosing.^6,S1 In the report by Sharma et al.,S1 a patient with PNH showed breakthrough hemolysis during the third trimester of pregnancy; notably, circulating levels of eculizumab and complement blockage were both suboptimal. Dose escalation was capable of increasing eculizumab concentrations and was effective in controlling hemolysis. In the cohort described by Kelly et al.,² breakthrough intravascular hemolysis and transfusions of blood and platelets, which occurred in 36 of the 67 pregnancies (54%) that progressed to delivery, required a more frequent use and/or higher doses of eculizumab. In a case series published by Servais and coworkers,⁶ 2 patients required an increase in eculizumab dosing (1 to 1500 mg every 2 weeks and the remaining 1 to 1800 mg fortnightly) to ensure a complete complement blockade. In our patient, eculizumab dosage was increased to 1200 mg on a biweekly schedule (from the beginning of pregnancy to delivery). This dosage allowed achieving therapeutically effective eculizumab levels and ensured an optimal complement blockage throughout the entire pregnancy. The need for higher eculizumab doses during the third trimester may be explained by increased complement activation as well as by higher water content. The question as to whether pregnancy can specifically affect the pharmacokinetics of eculizumab remains open and needs to be addressed in a specific study.

As far as fetal and obstetrical outcomes in women treated with eculizumab are concerned, a large multicenter study conducted in the United Kingdom has shown that the rate of fetal deaths (4%) in female PNH patients was similar to that observed in the general population.² The authors also reported a lower miscarriage rate (9%) but a higher risk of premature birth (29%) compared with women without PNH.² It is conceivable that the observed changes were caused by PNH per se, with the potential effects of eculizumab administration being unclear. Among the 12 reported pregnancies in women with aHUS who were treated with eculizumab, one in utero fetal death occurring at week 24 of gestation has been described. The newborns were premature and delivered by Cesarean in most of the published cases, suggesting that obstetrical risk in women with chronic kidney disease is not negligible. Our patient delivered at week 32 of gestation and her child’s weight was 1550 g. The mean weight at birth in the large case series² of women with PNH who received eculizumab was 2692 g (i.e., markedly higher than those reported for women with aHUS who received the same drug [1750 g]). Taken together, these findings indicate that fetal outcomes seem to be chiefly driven by the underlying disease rather than the use of eculizumab per se.

In our case, the newborn did not show any complications and/or signs of infection. This observation is in line with previous reports showing that eculizumab does not affect complement activity in the newborn,⁹ despite being detectable in cord blood.² Even though eculizumab is able to cross the placenta at least in some cases, its cord blood levels may be too low to impair complement activation in the newborn.² In addition, Servais et al.⁶ did not detect eculizumab in neonatal blood. Unfortunately, we did not measure eculizumab concentrations in the cord blood of our patient.

Previous studies demonstrated that eculizumab is not excreted in breast milk.2, 6 In the cohort study by Kelly et al.,² a total of 25 infants were breastfed. Of them, measurements of eculizumab were conducted in 10 breast milk samples, and drug was undetectable in all cases. In our report, the woman did not breastfeed and her infant’s outcomes at 10 months of age were unremarkable. These data are in line with the results of Kelly and coworkers,² who showed no complications at 12 months (the only exception being a case of spontaneously resolving neutropenia).

Conclusion

Here, we describe the first successful pregnancy in a kidney transplant recipient who received eculizumab for aHUS and reviewed the published literature of the 12 cases occurring in nontransplanted women. Based on the literature, we propose recommendations aimed at optimizing the management of pregnancies occurring in transplanted women suffering from aHUS who require eculizumab treatment (Table 4).The successful outcome observed in our patient was likely the result of (i) a thorough adjustment of eculizumab dosing during pregnancy, (ii) a close monitoring of hemolysis biomarkers, and (iii) the ability to reach optimal circulating eculizumab levels that ensured an effective complement blockade. Women with a history of kidney transplantation who express their desire to have a child while on eculizumab treatment should be counseled by a multidisciplinary team comprising both nephrologists and obstetricians.

Table 4.

Teaching points: recommendations for the counseling and monitoring of pregnant women with a kidney transplant under eculizumab therapy

1.Closely monitor biomarkers of hemolysis and kidney function

2.Periodically monitor complement activity and ensure complement hemolytic activity (CH50) target level <10 %

3.Increase eculizumab dosing in case of insufficient complement blockade or low trough levels

4.Inform parents about the risk of preeclampsia and premature birth

5.Closely monitor the newborn

6.Breastfeeding is allowed

Disclosure

All the authors declared no competing interests.