Higher Dose, Longer Duration of Eltrombopag Improves Likelihood of Response in Refractory Severe Aplastic Anemia

The thrombopoietin receptor agonist eltrombopag was approved by the U.S. Food and Drug Administration for the treatment of adults with refractory severe aplastic anemia (rSAA) based on response data at 12 to 16 weeks, but a study published in Blood suggests that many patients need treatment for at least 24 weeks to achieve a response.

“Our original study that was used to support the current drug label for rSAA employed a slow dose escalation, from 50 mg/day to 150 mg/day, due to lack of experience at that time with doses above 75 mg/day,” lead author Cynthia Dunbar, MD, of the Translational Stem Cell Biology Branch at the National Heart, Lung, and Blood Institute told ASH Clinical News. “Given the lack of toxicity of 150 mg/day in our initial trial and multiple subsequent ongoing trials – including the trial described in this paper – we believe that initiating treatment with 150 mg/day in adults is indicated to improve the likelihood of a faster response.”

In this report, Dr. Dunbar and colleagues reviewed data from this initial phase II trial (“study 1”), which enrolled 43 patients (median age = 45 years; range = 17-77 years), with data from a subsequent phase II trial (“study 2”) of 40 patients (median age = 50 years; range = 4-79 years) whose disease did not respond to rabbit or horse antithymocyte globulin with cyclosporine.

Demographics and clinical characteristics were similar between the two study groups, although patients in study 1 had received more immunosuppressive therapies and more were receiving platelet transfusions at baseline.

In study 2, eltrombopag dosing was stratified by age:

  • ≥12 years: 150 mg/day
  • 6-11 years: 75 mg/day
  • 2-5 years: 2.5 mg/kg/day

Unlike the earlier trial, responses (defined as platelets >100×109/L, hemoglobin 10 g/dL, absolute neutrophil count >1,000/uL) were assessed at 12 weeks and 24 weeks. Patients whose disease responded to eltrombopag continued with therapy on an extension phase until their blood counts stabilized (defined as no further improvement for >6 months) or until achievement of transfusion independence. The researchers also performed a bone marrow aspiration and biopsy with cytogenetics at baseline, 12 weeks, and 24 weeks of treatment, and every six months in responders who remained on eltrombopag.

Results from study 2 were evaluated on an intention-to-treat basis to identify whether a fixed dose of eltrombopag 150 mg for 24 weeks was associated with improvements in response rates among patients with rSAA.

During a median follow-up of 27.5 months (range = 3.0-46.7 months), twenty patients (50%) in study 2 achieved hematologic response at 24 weeks (primary endpoint). Of the 20 responders, five (25%) who did not achieve a response at 12 weeks did by 24 weeks; they would have been deemed nonresponders following study 1 criteria.

Between three and six months of follow-up, several patients improved their hematologic response: Most responses (65%) were unilineage at 12 weeks, but many became bilineage (45%) and even trilineage (35%) at 24 weeks. This translated to an approximately twofold increase in the number of multilineage responses with an additional three months of treatment (41% vs. 79%; p=0.04).

Fifteen responders discontinued eltrombopag because of a robust response; however, five of these required eltrombopag re-initiation following disease relapse. All recovered their responses.

“Although the response rate was higher in study 2 compared with study 1, the initial null hypothesis of at least a 20% increase in response … was rejected, and thus accrual was halted based on futility after enrollment of 40 evaluable patients,” the authors wrote.

Because patients with SAA are at risk for clonal evolution, development of myelo-dysplastic syndromes, paroxysmal nocturnal hemoglobinuria, and acute myeloid leukemia, the researchers also evaluated evolution of cytogenetic abnormalities and acquired somatic mutations in all 83 participants treated with eltrombopag. Sixteen patients (19%) experienced evolution from a normal karyotype to an abnormal karyotype, mostly within six months of eltrombopag initiation.

Dr. Dunbar expressed surprise that patients who developed del7 or del7q while
on eltrombopag had abnormal metaphases after only three to six months of treatment, while other cytogenetic abnormalities were identified at later timepoints. “This pattern suggested that eltrombopag might selectively stimulate hematopoietic stem and progenitor cells harboring del7 already present in the marrow before starting the drug,” she explained.

According to these findings, the lack of dose escalation and the prolonged duration of eltrombopag treatment in study 2 improved the likelihood of response in patients with rSAA, without affecting the drug’s toxicity profile. Therefore, continuing treatment for at least six months is recommended for rSAA patients before discontinuation for nonresponse.

Limitations of the study include its single-arm design and the heterogenous patient population, which may have precluded the researchers’ ability to determine whether eltrombopag accelerates cytogenetic progression and whether all cytogenetic progression is associated with adverse clinical outcomes.

The study does not indicate whether extending treatment over a six-month period will result in further responses in patients who never responded, Dr. Dunbar noted. “From the pattern of responses observed in this study, with no nonresponding patients showing even a suggestion of a response at the primary endpoint of six months, we believe this is unlikely,” she commented.

The authors report relationships with GlaxoSmithKline and Novartis, which provided funding for the study.


Winkler T, Fan X, Cooper J, et al. Eltrombopag for refractory severe aplastic anemia: dosing, duration, long term outcomes and clonal evolution. Blood. 2019 April 16. [Epub ahead of print]