BACKGROUNDAplastic anaemia is a rare, life-threatening condition, characterised by pancytopenia with hypocellular bone marrow. Haematopoietic stem cells and most progenitor cells express thrombopoietin receptor (c-MPL). Romiplostim is a peptibody with c-MPL agonist activity that stimulates endogenous thrombopoietin production and leads to promoting the proliferation and differentiation of megakaryocytes in the bone marrow. In this phase 2 trial we aimed to assess the activity and safety of romiplostim in patients with aplastic anaemia who were previously treated with immunosuppressive therapy.METHODSWe did an open-label, phase 2 study including a randomised, parallel, dose-finding part followed by an extension part to evaluate long-term treatment at two clinical centres in Seoul, South Korea. Eligible patients were aged 19 years or older, and had aplastic anaemia confirmed by bone marrow and cytogenetic studies and thrombocytopenia (platelet count ≤30 × 109/L), an Eastern Cooperative Oncology Group performance status score of 2 or lower, and were previously treated with immunosuppressive therapy, including at least one course of antithymocyte globulin plus cyclosporin. In the dose-finding part, patients were randomly assigned to fixed dose cohorts (1, 3, 6, or 10 μg/kg) of subcutaneous romiplostim once weekly for 8 weeks, according to a static allocation procedure after stratification by platelet count. In the extension part of the study, patients continued romiplostim titrated every 4 weeks in single steps (1, 3, 6, 10, 13, 16, and 20 μg/kg once weekly), depending on platelet response and safety up to 1 year (weeks 9-52). Patients who had a platelet response during weeks 46-53 continued dose titration in single steps (3, 6, 10, 13, 16, and 20 μg/kg once weekly) for an additional 2 years (weeks 53-156). The primary endpoint was the proportion of patients achieving a platelet response at week 9 (after completion of the dose-finding part). Activity was assessed per-protocol in all patients evaluable for response at week 9 and safety was assessed in all patients who received at least one dose of romiplostim. This trial is registered with ClinicalTrials.gov, NCT02094417.FINDINGSBetween April 14 and Nov 24, 2014, 35 patients were enrolled and randomly assigned to one of four dose cohorts: romiplostim 1 μg/kg (n=7), 3 μg/kg (n=9), 6 μg/kg (n=9), and 10 μg/kg (n=10). Data cutoff for this final analysis was on April 14, 2018. The median duration of treatment for all patients was 53 weeks (IQR 35-155). Ten (30%) of 33 evaluable patients achieved a platelet response at week 9, including seven (70%) of ten patients in the 10 μg/kg cohort, three (33%) of nine patients in the 6 μg/kg cohort, and no patients in both the 3 μg/kg and 1 μg/kg cohorts. During the extension study, 18 (55%) of 33 evaluable patients had a platelet response during weeks 46-53 and were eligible for continued treatment. Ten (30%) patients maintained a platelet response at 2 and 3 years, of whom nine had an erythroid response and five a neutrophil response, and completed protocol treatment. Treatment-related adverse events occurred in three (9%) of 35 patients, including grade 1 or 2 myalgia, fatigue, and dizziness. 17 (49%) of 35 patients had adverse events of grade 3 or higher; seven (20%) had serious adverse events (one event of febrile neutropenia, cataract, retinal detachment, macular fibrosis, inguinal hernia, appendicitis, cellulitis, tendon injury, and transfusion reaction); and one patient died from sepsis during treatment; none of these events were related to treatment. No patients developed clonal evolution.INTERPRETATIONRomiplostim seems to be active and has a favourable safety profile in patients with refractory aplastic anaemia. 10 μg/kg once weekly might be used as a recommended starting dose in future studies. These findings warrant further investigation.FUNDINGKyowa Hakko Kirin Korea.
