A Lot of Attention on Aplastic Anemia

Within the past several years, important work showed that hematopoiesis in the majority of idiopathic acquired aplastic anemia patients is clonal, potentially reflecting selective expansion of cells with inhibited differentiation, increased proliferative capacity, and/or that are capable of escaping cytokine-mediated suppression (e.g., via loss-of function mutations in HLA class I genes).^1,2  In 2017, we continued our fast-paced journey of discovery towards improving outcomes in patients with acquired aplastic anemia and related bone marrow failure disorders.

Building on findings from a phase II study demonstrating trilineage hematopoietic improvement in patients with refractory severe aplastic anemia treated with the thrombopoietin mimetic eltrombopag,³  this year saw publication of Dr. Danielle M. Townsley and colleagues’ study on the use of eltrombopag in combination with antithymocyte globulin (ATG) and cyclosporine (CSA) in treatment-naïve patients.⁴  The Hematologist previously highlighted this prominent study in a Diffusion article from the July/August 2017 issue. Sequential patient cohorts received daily eltrombopag from day 14 to six months (cohort 1), from day 14 to three months (cohort 2), and from day 1 to six months (cohort 3). All patients received horse ATG on days 1 to 4, and daily CSA from day 1 to six months. A complete response was defined as an absolute neutrophil count of 1,000/μL or greater, hemoglobin 10 g/dL or greater, and platelet count 100,000/μL or greater. Outcomes were best in cohort 3, in whom complete and hematologic responses at six months occurred in 58 percent and 94 percent of patients, respectively. This compared to 10 percent and 66 percent rates observed in a historical cohort treated with immunosuppression alone. The improved blood counts were accompanied by increased marrow cellularity and hematopoietic progenitor numbers. Also, the addition of eltrombopag to standard immunosuppressive therapy seemed to increase the speed and robustness of the hematologic recovery in responding patients.

A concern persists, however. Data from a phase II study of refractory severe aplastic anemia patients treated for six months with eltrombopag (150 mg daily), presented at the 2017 ASH Annual Meeting by Dr. Thomas Winkler and colleagues, showed that 15 percent of eltrombopag-treated patients (6 of 39) demonstrated cytogenetic clonal evolution. This is comparable to the 19 percent rate (8 of 43 patients, within 3-13 months of starting eltrombopag therapy) reported in an earlier study of eltrombopag in the refractory setting.⁵  In a combined analysis of patients in both studies, clonal cytogeneic evolution occurred within six months of initiating therapy in 13 of 16 evolvers and in six of six evolvers with high-risk chromosome 7 abnormalities. Further work is needed to discern whether eltrombopag affects the frequency and/or temporal development of cytogenetic clonal evolution in aplastic anemia.

Eltrombopag binds to a domain of the thrombopoietin receptor, MPL, distinct from the binding site of thrombopoietin (TPO). After binding, TPO (or eltrombopag) activates the JAK-STAT pathway and leads to cell survival and proliferation, making it not only critical for the production of platelets by megakaryocytes, but also for the maintenance of hematopoietic stem and multipotent progenitor cells. Studies have shown that interferon-gamma (IFNγ) blocks thrombopoietin signaling and suppresses hematopoiesis.⁶  IFNγ is overexpressed in aplastic anemia^6-8  and recent work presented in abstract form, has shown that in the presence of IFNγ, eltrombopag and not TPO, can maintain hematopoietic progenitor cell growth in vitro.⁹  Notably, TPO levels are markedly elevated in patients with acquired aplastic anemia.¹⁰  During the Plenary Abstract Session at the 2017 ASH Annual Meeting, Dr. Luigi J. Alvarado presented exciting work that delineates the molecular mechanism by which IFNγ inhibits TPO signaling and by which eltrombopag bypasses this inhibition. Under chronic inflammatory conditions, IFNγ specifically heterodimerizes with TPO, preventing binding of TPO to its low-affinity binding site on MPL. IFNγ negatively affects MPL dimerization in cells cultured with TPO. Alternatively, IFNγ does not heterodimerize with eltrombopag, and IFNγ does not impair c-MPL dimerization in cells cultured with eltrombopag. This mechanism predicts that eltrombopag could ameliorate pancytopenia in other inflammatory states characterized by elevations in IFNγ.

We look forward to what 2018 brings to the field.