Romiplostim for management of refractory immune thrombocytopenic purpura in the immediate postpartum period

Abstract

We present a case of a pregnant woman with chronic immune thrombocytopenic purpura and chronic hypertension who developed pre-eclampsia with severe features warranting delivery. Her overall clinical picture and liver enzymes improved in the immediate postpartum period, however, aggressively progressing thrombocytopenia posed a diagnostic dilemma to the interdisciplinary care team. After failing to respond to first-line therapies including high-dose corticosteroids and intravenous immunoglobulin, she was successfully managed with a trial of the thrombopoietin receptor agonist, Romiplostim.

Background

Thrombocytopenia in pregnancy, defined by a maternal platelet count <150×10⁹/L, is a common condition seen in 7%–12% of patients.^{1 2} Thrombocytopenia can result from a variety of physiological or pathological conditions, several of which are unique to pregnancy.

The most common cause of thrombocytopenia in pregnancy is gestational thrombocytopenia, which accounts for 80% of cases and does not impose an increased risk of bleeding to the mother or fetus.¹ Other causes include severe pre-eclampsia/HELLP (haemolysis, elevated liver enzymes and low platelets) syndrome, which occur in 5%–21% of cases and immune thrombocytopenic purpura (ITP), which occurs in 1%–4% of cases. Both of these are associated with increased risk of maternal and fetal bleeding.^{1 3} Bleeding risk and treatment vary depending on the cause, it is important to consider the overall clinical scenario and laboratory evaluation when making management decisions.

In some pregnant patients, thrombocytopenia <100×10⁹/L in the setting of additional laboratory abnormalities and pre-existing medical conditions poses a greater diagnostic challenge. In cases such as this, careful evaluation with an interdisciplinary approach may be helpful when trying to delineate the predominant cause and address management.

ITP during pregnancy can most often be managed conservatively with observation. When clinically necessary, first-line treatment therapies include corticosteroids and intravenous immunoglobulin (IVIG). We present a case of chronic ITP refractory to first-line management successfully treated with romiplostim in the immediate postpartum period.

Case presentation

A 34-year-old multipara with chronic ITP and well-controlled chronic hypertension on labetalol was found to have severe thrombocytopenia with a platelet count of 37×10⁹/L at 24 weeks of gestation. She was started on oral prednisone (initially 40 mg/day, increased gradually to 90 mg/day) with poor response and then treated with IVIG (0.4 g/kg daily × 5 doses) with good response and her platelet count rose to 144×10⁹/L.

At 28 weeks of gestation, the patient developed superimposed pre-eclampsia with severe features (elevation of liver enzymes). She was admitted to the labour and delivery unit, received antenatal corticosteroids for the acceleration of fetal lung maturity (intramuscular betamethasone 12 mg × 2 doses, 24 hours apart), and labour was induced resulting in a successful vaginal delivery. Labour course was uncomplicated with the exception of significant postpartum haemorrhage managed successfully with carboprost 250 μg intramuscularly (Hemabate), Bakri tamponade balloon placement and tranexamic acid (1000 mg intravenously). During the labour course, she had an elevation of liver enzymes (peak elevation alanine aminotransferase (ALT): 672 IU/L, aspartate transaminase (AST): 304 IU/L); however, her platelet count stayed stable ranging from 122×10⁹/L to 150×10⁹/L.

Postdelivery, her liver enzymes progressively improved (ALT: 438 IU/L and AST: 102 IU/L on second postpartum day). Her platelet count dropped from 112×10⁹/L to 32×10⁹/L 12 hours postdelivery, and to 9×10⁹/L 24 hours postdelivery.

Differential diagnosis

Thrombocytopenia seemed to be worsening in isolation. Differential diagnosis included pre-eclampsia/HELLP syndrome versus ITP.

Delivery is considered as the definitive management for HELLP syndrome and, typically, patients improve postdelivery. Our patient did not have evidence of haemolysis as the peripheral smear revealed no schistocytes, and the lactate dehydrogenase and haptoglobin were normal. Liver enzymes peaked intrapartum and progressively improved post partum. Her postdelivery blood pressure remained well controlled with oral labetalol 200 mg two times per day.

Due to the diagnostic dilemma for severe thrombocytopenia, the patient underwent a bone marrow biopsy, which revealed normal trilineage haematopoiesis with normal to an increased number of normal-appearing megakaryocytes and no evidence of malignancy.

The consulting haematology team and maternal–fetal medicine team felt that the severe thrombocytopenia was likely related to the ITP since the patient’s overall clinical picture including her blood pressures and liver enzymes had improved, and the thrombocytopenia seemed to be worsening in isolation.

Additionally, the patient’s poor response to platelet transfusions further supported the suspicion of ITP being the predominant aetiology (pretransfusion platelet count 9×10⁹/L, post-transfusion platelets count 8×10⁹/L).

Treatment

The patient was not actively bleeding, but due to her proximity to the immediate postpartum period, she was transfused with 5 units of platelets with no response. Following the transfusion, the platelet count was 8×10⁹/L. The patient was started on high-dose intravenous methylprednisolone 500 mg/day for 2 days and was given IVIG 1 g/kg with poor response.

The following morning, her platelet count had further dropped to 5×10⁹/L (~60 hours after delivery). She received another 5 units of platelets and the platelet count increased to 12×10⁹/L.

After extensive discussion between the patient and the care teams (maternal fetal medicine and haematology) regarding the patient’s poor response to the first-line therapies, her proximity to the immediate postpartum period and the risks associated with extremely low platelet count, the decision was made to use romiplostim.

The patient received romiplostim 2 μg/kg on the evening of postpartum day 2 (~68 hours postdelivery) along with 90 mg oral prednisone. She received the second dose of 2 μg/kg romiplostim on the third postpartum day. On the fourth postpartum day, her platelet count rose to 16×10⁹/L.

Outcome and follow-up

By the fifth postpartum day, her platelet count had increased to 40×10⁹/L and the patient was discharged home with instructions to follow-up with haematology and her obstetrician as an outpatient. Platelet count on the ninth postpartum day was 786×10⁹/L. Platelet count 6 weeks post partum was 184×10⁹/L.

Discussion

Thrombocytopenia can result from a variety of physiological or pathological conditions, several of which are unique to pregnancy.

Differential diagnosis is broad. The top three causes include gestational thrombocytopenia, pre-eclampsia/HELLP syndrome and immune thrombocytopenia.

Additional causes include pseudothrombocytopenia, viral infection, drug-induced thrombocytopenia, thrombotic thrombocytopenic purpura, haemolytic uraemic syndrome, disseminated intravascular coagulation, systemic lupus erythematosus, antiphospholipid syndrome and congenital thrombocytopenias.¹

Pre-eclampsia is a hypertensive disorder of pregnancy with evidence of end-organ injury. It affects 2%–3% of pregnancies and of these patients 15%–50% will have concurrent thrombocytopenia.³ By definition, platelet counts <100×10⁹/L reclassifies the disease as having severe features. If there is transaminitis and haemolysis in the setting of severe thrombocytopenia, then consider HELLP syndrome, which is a severe systemic complication of pregnancy seen in 0.5%–0.9% of patients. Maternal and perinatal mortality rates may reach 24% and 37%, respectively. Hence, prompt diagnosis is key.^{3 4}

The primary treatment of maternal thrombocytopenia (platelet count <100×10⁹/L) associated with severe features of pre-eclampsia or HELLP syndrome is delivery. In general, these disorders improve on delivery, but may also present in the postpartum period. Platelet counts can decrease for 24–48 hours after delivery followed by rapid recovery.¹

Primary ITP is an autoimmune condition characterised by isolated destruction of platelets. Maternal IgG antiplatelet antibodies sensitised to native maternal platelet antigens have the potential to cross the placenta placing the fetus and neonate at risk of thrombocytopenia.¹ Although the differential for thrombocytopenia in pregnancy is broad, persistent counts <50×10⁹/L is almost certainly due to ITP.^{5 6} The risk of fetal intracranial haemorrhage and severe maternal haemorrhagic complications is very low. Treatment is largely reserved for symptomatic bleeding and observation is generally appropriate.

Evidence regarding the specific platelet threshold at which pregnant patients with ITP should be treated is lacking.⁷ Treatment should be initiated when the patient has symptomatic bleeding, when platelet counts fall below 30×10⁹/L, or to increase platelet counts to a level considered safe for procedures.^7–9 At the time of delivery, the management of ITP is based on an assessment of maternal bleeding risks associated with the delivery and epidural anaesthesia.¹

The first-line treatments for thrombocytopenia secondary to ITP remain corticosteroids and IVIG. They have been used separately as well as in synergy with success. Prednisone is safe to use in pregnancy with daily dosing starting at 10–20 mg and titrated to effect; it is important to weigh the benefits of improving platelet count with consequences of maternal and fetal hyperglycaemic. IVIG is appropriate when the response to steroids is suboptimal or when side effects outweigh the benefits of treatment. It is acceptable to give 0.4 g/kg daily for 2–5 days; however, transient responses in patients with persistent platelet counts <20×10⁹/L have been reported.¹⁰

When patients with severe chronic ITP are not responding to first-line treatments and splenectomy is not preferable, thrombopoietin receptor agonists (TPAs) pose a viable second-line therapeutic option. They work by increasing the endogenous production of platelets. Romiplostim (Nplate, Amgen) and eltrombopag (Revolade, Novartis) are TPAs approved by the Food and Drug Administration for patients with ITP. Both agents increase the platelet level in patients with ITP and healthy volunteers.¹¹ In non-pregnant patients, the usual dosage of romiplostim is 1–10 µg/kg given subcutaneously once a week with a treatment goal of >50×10⁹/L.^{12 13}

Authorisation of romiplostim and eltrombopag within Europe between 2009 and 2010 resulted in response rates up to 70%–80% in patient’s refractory to first-line therapy.¹⁴

In a long-term open-label study evaluating romiplostim in patients with platelet level <50×10⁹/L (NCT00116688), a positive platelet response was observed in 87% of the 142 patients with ITP enrolled in the study.¹⁵ Treatment-related serious adverse effects were reported in 13 patients (9%). Available data indicate that romiplostim increased platelet counts in the majority of patients with ITP and is well tolerated.¹³

The obstetric literature has scant information in the form of case reports regarding romiplostim use during pregnancy. A dose range of 2–6 μg/kg weekly has been reported in the obstetric literature.¹⁶ For our patient, the haematology team recommended a dose of 4 μg/kg as the initial dose; however, the pharmacy was only able to give a 2 μg/kg dose on the first day and then gave an additional dose of 2 μg/kg on the second day.

Romiplostim is usually well tolerated and side effects are mostly mild to moderate. Known side effects of romiplostim include headaches, dizziness, insomnia, skin rash, diarrhoea, abdominal pain, arthralgia and myalgia. Serious side effects that may occur are bone marrow fibrosis and rebound thrombocytopenia. Thromboembolism or thrombotic complications may occur with increased platelet counts. Our patient consented to the use of romiplostim after being informed of these side effects and the concern regarding thromboembolic risk especially given her postpartum state. She tolerated the medication well without any side effects.

The effect of romiplostim on a nursing infant is unknown. Its high molecular weight may limit excretion into breast milk, but the long half-life (median 3.5 days) may assist excretion. If excretion were to occur, the protein might be digested in the infant’s gut.¹⁷ Our patient chose to breastfeed the infant after she was informed of the limited information available. The neonate was closely monitored by the neonatology team and did well during the hospital stay.

It is not known if romiplostim crosses the placenta. Controlled studies of this drug in pregnant patients are not yet available; however, the manufacturer states romiplostim may be indicated when benefits outweigh the risks.¹⁸ Though more data are necessary to appreciate the long-term effects of TPAs through surveillance initiatives, this case demonstrated the efficacy of treating chronic primary ITP with a TPA agent refractory to first-line therapy in a postpartum patient.

Learning points.

• The most common cause of thrombocytopenia in a pregnant patient is gestational thrombocytopenia.

• HELLP (haemolysis, elevated liver enzymes and low platelets) syndrome and pre-eclampsia can masquerade other conditions posing unique clinical challenges.

• First-line treatments of immune thrombocytopenic purpura (ITP) in pregnancy are corticosteroids and intravenous immunoglobulin.

• Thrombopoietin receptor agonists such as romiplostim pose a viable second-line therapeutic option in chronic ITP refractory to first-line therapies.