Efficacy of TPO-mimetics in patients with immune thrombocytopenia

Background: Immune thrombocytopenic purpura (ITP) is an autoimmune disorder in which antibodies are produced to circulating platelets. The two currently available agents (Romiplostim and Eltrombopag) have similar efficacies and only slightly different safety profiles, being effective in restoring a safe platelet count in 70%-80% of cases with chronic ITP failing one or more lines of treatment, including splenectomy. Aims: We evaluated the efficacy of TPO-RAs in patients with ITP. Methods: From November 2008 and February 2017 65 patients (33 M; 32 F) were treated with a median follow-up of 29 months (1-96): 39 underwent therapy with Romiplostim and 26 to Eltrombopag. Median age was 69 years (range 39- 94 years). In the group of patients treated with Romiplostim, 21 had already received more than 4 lines of therapy, while 18 received 1-2 lines of therapy. 13/26 patients who received Eltrombopag were at the 3nd line of therapy, 1 at the second, and the others were at least at the 4th line. The median platelet count was 21×109/L (3-52) at the start of Romiplostim, with a median starting dose of 1 μg (1-2) and 17×109/L (1-53) in patients treated with Eltrombopag, with a median starting dose of 50mg (25-50). Results: Patients treated with Romiplostim we observed 22 complete responses and 10 responses, with a 82% response rate, while 7 patients were no responders. In our study 26 (66%) patients stopped Romiplostim after a median time of 16 months (1-93): 9 for stable response, 5 for no response, 3 for loss of response; 3 for adverse events (2 for bone marrow fibrosis, 1 for headache associated to visual disturbances and gastrointestinal disorders), 2 underwent splenectomy, and 3 patients interrupted the treatment for other causes (es. diagnosis of cancer). The median platelet count at suspension of Romiplostim was 91.5 x109/L (3-320). In patients treated with Eltrombopag 16 achieved a complete response, 5 a response, obtaining response in the 80% of cases; 5 were no responders. 14 (53%) patients stopped Eltrombopag after a median time of 1,5 months (1-12): 6 for adverse events (2 cases of major cardiovascular events, liver toxicity, skin rash, pharyngitis), 5 for no response, 1 for loss of response, 2 patients who achieved a CR interrupted Eltrombopag obtaining a sustained remission after discontinuation. The median platelet count at suspension was 97×109/L (2-739). Patients who did not interrupted treatment are still receiving therapy with a median of 29 months (3-96). Several studies reported Romiplostim and Eltrombopag to be highly effective against chronic ITP, with average immediate responses exceeding 80% in our study. We observed that therapeutic response was influenced by the starting platelet count. In particular platelets count before therapy influenced the first response observed. In particular in patients treated with Romiplostim PLT pre-treatment directly correlated with the first response and the mantainance of response during treatment at month 1°, 2° 3° and 6. Patients with a median starting platelet count of 15×109/L obtained a response (CR + R), while almost all patients who started therapy with PLT<15×109/L at baseline can obtain an initial response, but the majority is non-responder. Summary/Conclusions: TPO-mimetics have proved efficacy in patient with ITP and their use can be applied in several conditions (bridge to splenectomy; sustained response; switch and discontinuation). Future study on large series of patients are needed to best correlate baseline platelets with hematological response. The two currently available agents (Romiplostim and Eltrombopag) have similar efficacies and only slightly different safety profiles, being effective in restoring a safe platelet count in 70%-80% of cases with chronic ITP failing one or more lines of treatment, including splenectomy.