Association between Acute Myelogenous Leukemia and Thrombopoietin Receptor Agonists in Patients with Immune Thrombocytopenia

作者：Oshima Yasuo^*; Yuji Koichiro; Tanimoto Tetsuya; Hinomura Yasushi; Tojo Arinobu

来源：Internal Medicine, 2013, 52(19): 2193-2201.

DOI：10.2169/internalmedicine.52.0324

• thrombopoietin receptor agonist
• myeloproliferative leukemia virus proto-oncogene product (MPL)
• immune thrombocytopenia (ITP)
• FDA adverse event reporting system (FAERS)
• spontaneous report

Objective The development of myeloid malignancies is a concern when administering thrombopoietin receptor (or the myeloproliferative leukemia virus proto-oncogene product MPL) agonists. Progression from myelodysplastic syndrome (MDS) to acute myelogenous leukemia [AML, 9 (6.12%) AML patients among 147 MDS subjects] was reported in a clinical trial. However only one (0.15%) case of AML among 653 immune thrombocytopenic purpura (ITP) subjects was reported. Our objective was to determine whether there is currently a safety signal in the FDA files termed Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) for AML in ITP patients who receive MPL agonists.
Methods We conducted a case-controlled study using the FAERS as a source of case and control data. We compared demographic characteristics, such as gender, age and exposure to MPL agonists between AML patients and others among ITP subjects registered between 2002 and 2011.
Results Total of 4,821 ITP subjects were identified including 62 AML patients. The number of patients treated with romiplostim and eltrombopag was 54 (1.74%) AML patients among 3,102 ITP subjects and nine (1.52%) AML patients among 594 ITP subjects respectively. It should be noted that all AML patients were exposed to one or more MPL agonists. Another factor associated with AML was male gender.
Conclusion We herein report an association between AML and MPL agonist use in ITP subjects. Due to various biases and the incompleteness of the FAERS data, further studies are warranted to determine whether the detected signal is a real risk. Physicians should not alter their prescribing behaviors based on this single preliminary analysis.

• 出版日期2013
• 单位河北医科大学