A Multicenter Study Evaluating the Safety of Romiplostim at Maximal Dosage for Emergency Bleeding Situations in Immune Thrombocytopenia

Introduction: The serious consequences of severe, life-threatening bleeding in immune thrombocytopenia (ITP) justify the use of several regimens, including platelet transfusions, high-dose parenteral glucocorticoid and intravenous immunoglobulins. The goal of such treatments is to rapidly elevate the platelet count to a level where the risk of severe bleeding is minimized. We lack experience with the use of thrombopoetin receptor agonists (Tpo-RAs) in this situation of emergency. The aim of this study was to evaluate the safety of the use of romiplostim at maximal dosage in patients with severe bleeding manifestations, who have previously failed to respond to corticosteroids (CST) and intravenous immunoglobulin (IVIg).

Patients and Methods: We carried out a multicenter retrospective study in France. We included ITP patients with platelet count < 30 x 10⁹/L.and severe bleeding manifestations, who have failed to respond to corticosteroids and IVIg, and have received a rescue therapy with romiplostim at maximal dosage (ie., 10 mg/kg body weight per week). Failure to treatment was defined according to standardized international criteria and severe bleeding manifestations as a score >8 according to the bleeding score previously reported by our group that take into account cutaneous, mucosal and visceral bleeding. Physicians were interviewed and patients' medical charts collected using the standardized form of the Referral Center for Adult ITP. Complete response (CR) and Response (R) were defined according to standardized international criteria: platelet count > 30x10⁹/L with at least a doubling of the baseline value or >100 x10⁹/L. Non-response (NR) was defined as the absence of platelet count increase >30 x10⁹/L with at least a doubling of the baseline count.

Results: Fifteen patients (8 men/7 women) were included in the study. The median age was 68 ± 20.9 years [range 17-92]. Two patients presented secondary ITP according to the international definition criteria. Eleven were newly diagnosed ITP, and 4 chronic ITP. Severe bleeding symptoms were the manifestation of ITP in all cases, with a median bleeding score of 16 [9-28], including 4 intracranial hemorrhage, 2 metrorrhagia, 1 macroscopic hematuria, 3 gastrointestinal bleeding and 1 haemoperitoneum. The median platelet count was 1 [0-4]. All patients had received corticosteroids (1mg/bw/day) including 12 pulses of methylprednisolone and IVIg (median dose 2 g/bw). Twelve have also received 8 mg of vinblastine (1 to 3 injections). Nine received platelet transfusions and 4 received blood tranfusions. One patient had received anti-CD20 therapy (rituximab) the same day as TPO-RA therapy, and another hydroxychloroquine before starting TPO-RA therapy. Romiplostim was started 12 days [range 6-32] in median after the first bleeding episode. At that time bleeding symptoms were severe (>8), and the median platelet count was 4 [1-47]. All received 10 mg/kg body weight of romiplostim (including 3 escalating dose). At day 7 after TPO-RA initiation, 9 patients achieved a CR, 1 patient achieved a R and 5 were still non-responder. At days 14, 10 patients achieved a CR and 5 were still non-responder. During the first month of follow-up, the maximal platelets count among the responders was 590 [range: 312-1169], 9 patients experienced a platelet count > 500 x 10⁹/L. One 51 years old patient, bed bound for a muscular hematoma, experienced a deep vein thrombosis with asymptomatic pulmonary embolism, 13 days after IVIg and 5 days after TPO-RA initiation, with a platelet count of 629.10⁹/L. One patient presented superficial thromboembolism of the arm, 14 days after IVIg and 9 days after TPO-RA initiation with a platelet count of 162.10⁹/L. They were both treated with anticoagulation without complication. After a median follow-up of 10 months, no other TPO-RA-related side effects were observed, TPO-RA was discontinuated in 10 patients.

Conclusion: Our study demonstrated the relative safety and the interest of the use of romiplostim at the maximal dose in emergency bleeding situation in patients with failure of conventional therapy. However, risk of venous thromboembolism should be carrefully assess in this situation.