Safety and Efficacy of Long-Term Open-Label Dosing of Subcutaneous (SC) Romiplostim in Pediatric Patients with Immune Thrombocytopenia (ITP)

Background: Chronic ITP in children is an autoimmune disorder characterized by increased platelet destruction and suboptimal platelet production. Pediatric patients with chronic ITP that completed a romiplostim phase 1/2 or phase 3 study could enroll in this open‑label long‑term extension; 5.2 years of data are reported.

Methods: All patients received SC romiplostim once weekly. The initial dose was the final dose from the parent study or 1 µg/kg (for patients previously receiving placebo or who had not received romiplostim for >24 weeks), adjusted from 1–10 µg/kg to target platelet counts of 50–200×10⁹/L. The incidence of adverse events (AEs) was the 1° endpoint. If approved by investigators, patients who maintained platelet counts ≥50×10⁹/L for ≥4 consecutive weeks at a stable dose could subsequently administer doses by self or caregiver. Patients who turned 18 years of age could remain on study.

Results: A total of 66 patients (phase 1/2, n=12; phase 3, n=54) entered the extension; 65 received romiplostim for ≤269 weeks (5.2 years); 1 patient withdrew consent before treatment. At baseline, median (range) age was 11 (3–18) years; 56% were female; 61% were white, 14% African American, and 14% Hispanic/Latino; 9.1% had prior splenectomy. Median (range) treatment duration was 57.9 (1–269) weeks; median number of doses was 53.0 (1–268).  Median (range) average weekly romiplostim dose was 5.5 (0.1–10.0) µg/kg, including escalation to stable dose, median maximum dose was 8.0 (1.0–10.0) µg/kg, and median most frequent dose was 6.0 (0–10) µg/kg.  The median dose fell from 6 to 1 µg/kg; after ~week 140 (n≤9), the median dose fluctuated (Fig 1).  Thirteen patients discontinued treatment: consent withdrawn (n=6), noncompliance (2), administrative decision (2), nonresponse (2), and per protocol (1). Fifty-two (79%) patients continued in the study; none withdrew due to AEs. After the first study week median platelet counts remained >50×10⁹/L (Fig 2); for 15 patients (23%), the first study week was the first week receiving romiplostim (ie, previously received placebo).  Fifty‑six (86%) patients (or caregivers) self‑administered romiplostim. Twenty-one (32%) patients received rescue medications on 63 occasions (for low platelet counts [n=35], bleeding/bruising [17], pre- or post-procedure [9], and other [2]), including IVIG (n=10), prednisone (9), aminocaproic acid (3), tranexamic acid (2), methylprednisolone (2), and platelet transfusion (1). Patients required rescue treatment during the first 3 months (27/63 instances), >3–6 months (9), >6–9 months (6), >9–12 months (7), and after 1 year (14) in the extension. Five previous placebo recipients received rescue medication, mostly during the first 3 months (10/14 instances). Three patients achieved remission (platelet counts ≥50×10⁹/L for 24 weeks with no ITP treatments): 1) 9-year-old boy with ITP for 9 years, after 4.3 years of romiplostim, entered remission for the last 2.1 years as of this datacut; 2) 13-year-old boy with ITP for 8.5 years, after 3.3 years of romiplostim, entered remission for the last 1.1 years; and 3) 17-year-old girl with ITP for 8.2 years, after 5.9 years of romiplostim, entered remission for the last 44 weeks. Thirty‑four serious AEs occurred in 14 patients, including pyrexia (n=3), epistaxis (2), and thrombocytopenia (2); 3 were deemed treatment-related (anemia, epistaxis, and thrombocytopenia), and none led to discontinuation of romiplostim. Five patients had life‑threatening AEs, including thrombocytopenia (n=2) and infection, decreased platelet counts, and subcutaneous abscess (1 each); none were fatal or deemed treatment-related. Bleeding AEs occurred in 47 patients; 3 were deemed treatment-related by the investigator (gingival bleeding, petechiae, and epistaxis). Bleeding AEs occurring in ≥4 patients included contusion (n=25), epistaxis (21), petechiae (16), gingival bleeding (9), mouth hemorrhage (6), ecchymosis (5), and hemorrhage, injection-site bruising, and purpura (4 each). No thrombotic events were reported.  There were no peripheral blood abnormalities suggestive of malignancy to warrant a bone marrow examination in any patient.  There was no development of anti-TPO or anti-romiplostim antibodies.

Conclusion: In this ongoing open-label extension of children with chronic ITP, romiplostim for ≤5.2 years maintained platelet counts with a safety profile similar to that seen in past studies.