Eltrombopag Reduces Thrombocytopenic Events in Patients With AML and MDS, But Not as Much as Expected

Compared with placebo, the oral thrombopoietic receptor agonist eltrombopag significantly lowered the incidence of thrombocytopenic events in patients with acute myeloid leukemia (AML) or myelodysplastic syndromes (MDS) who had thrombocytopenia, according to a phase III, placebo-controlled study. However, the difference between the two treatment groups was less than 30 percent, which was lower than expected, the researchers noted.

The three-part, randomized, placebo-controlled ASPIRE (A Study of eltromboPag In myelodysplastic syndRomes and acutE myeloid leukemia) trial enrolled patients from 61 international hospitals and medical centers. Lead author Moshe Mittelman, MD, of Tel Aviv Sourasky Medical Center and Tel Aviv University in Israel, and colleagues published the findings from part one (open-label phase) and part two (randomized phase) in The Lancet Haematology.

Patients with intermediate-2 or high-risk (per International Prognostic Scoring System) MDS or AML who had grade 4 thrombocytopenia because of bone marrow (BM) insufficiency (defined as <25×109/L), or grade 4 thrombocytopenia before platelet transfusion, with platelet counts of ≥25×109/L after transfusion were included in the study. Participants were also required to have at least one of the following within four weeks of the screening period: platelet transfusion, symptomatic bleeding, or platelet count of <10×109/L. People with thrombocytopenia for reasons other than BM insufficiency, leukocyte count ≥25×109/L, or previous treatment with a thrombopoietin receptor agonist were excluded.

In part one, 17 patients (mean age = 71.5 years; standard deviation [SD] = 10.8 years; 9 with MDS and 8 with AML) received eltrombopag for eight weeks, with 11 completing all planned treatment.

In the randomized, double-blind, placebo-controlled, multicenter second part of the study, 145 patients were stratified based on baseline platelet count (<10×109/L vs. ≥10×109/L) and disease (MDS vs. AML) to receive 12 weeks of either:

  • eltrombopag 100 to 300 mg/day (n=98; mean age = 72.3 years; SD=8.9 years)
  • placebo (n=47; mean age = 70.6 years; SD=10.7 years)

Forty-three and 27 patients, respectively, completed all planned treatment. Discontinuation of eltrombopag was primarily related to adverse events (AEs; n=31; 32%), while discontinuation of placebo was mainly due to physician’s decision (n=8; 17%).

In the overall study population, patients received a median eltrombopag dose of 108.5 mg/day (interquartile range [IQR] = 50-122 mg/day) over a mean duration of 8.1 weeks (SD=4.2 weeks); mean duration of placebo was 9.6 weeks (SD=3.7 weeks).

During weeks five to 12 of treatment, eltrombopag-treated patients experienced significantly fewer clinically relevant thrombocytopenic events (CRTEs; primary endpoint; defined as one of the following, either alone or in combination: grade ≥3 hemorrhagic AEs, platelet counts of <10×10⁹/L, and platelet transfusions), compared with placebo: 54 percent versus 69 percent (odds ratio [OR] = 0.20; 95% CI 0.05-0.87; p=0.032). Although the weekly average proportion of CRTE was lower in the eltrombopag group for all 12 study weeks, the difference did not meet criteria for clinically meaningful efficacy (defined as 30% absolute difference between the groups).

The average proportion of patients with weekly platelet counts <10×109/L during treatment was significantly lower in the eltrombopag group (27% vs. 50%; p=0.0013); however, patients in each treatment group still required weekly platelet transfusions at a similar rate (51% vs. 52%; p=0.83).

Maximum mean platelet transfusion independence duration during weeks five through 12 also did not differ significantly between the eltrombopag and placebo cohorts: 26.3 days (SD=21.47 days) versus 25.4 days (SD=15.5 days).

Few responses were reported: one marrow complete response (CR) in the eltrombopag group and one morphologic CR in the placebo group (OR=0.47; 95% CI 0.03-7.75; p=0.59). Eighteen patients in the eltrombopag group (18%) and 10 in the placebo group (21%) experienced stable disease, although 36 and 98 patients in each cohort, respectively, were non-evaluable, “which could have influenced the lower disease progression rate observed in eltrombopag recipients,” the authors noted.

Sixty-one eltrombopag-treated patients (62%) and 36 placebo-treated patients (77%) experienced disease progression.

At all weekly evaluations following treatment, the proportion of patients with no bleeds was higher in the eltrombopag group (n=21/40; 53%), compared with the placebo group (n=6/27; 22%; p values not reported).

Median BM blast counts and median peripheral blood blast counts did not differ between treatment groups. Any hematologic improvement also was similar between groups (n=10 [10%] and n=4 [9%], respectively; OR=1.26; 95% CI 0.37-4.30).

“The bleeding and CRTE reductions in ASPIRE were not accompanied by hematologic improvement or improved platelet transfusion independence,” the authors noted. “This, and that only 48 percent of the 145 patients [receiving eltrombopag completed] the study, could be explained by the advanced disease stage of enrolled patients.”

During part two, 47 deaths (48%) were reported in the eltrombopag arm and 18 (38%) in the placebo arm, with 35 (36%) and 13 (28%) deaths, respectively, within 30 days of final treatment dose (p values not reported). The primary cause of death in both groups was underlying disease (n=41 [87%] and n=17 [94%], respectively).

In part two, the median overall survival did not differ between groups: 4.3 months (IQR=1.5-12.0 months) with eltrombopag and 4.6 months (IQR=2.4-8.5 months) with placebo (hazard ratio [HR] =0.97; 95% CI 0.64-1.48; p=0.89). Median progression-free survival was also similar between both groups: 1.08 (IQR=0.48-2.73 months) and 0.94 months (IQR=0.49-2.79 months), respectively (HR=0.99; 95% CI 0.68-1.43; p=0.94).

“Similar overall AE rates were observed with each treatment, and no new safety signals were identified,” the researchers reported. Forty-four (45%) and 12 (26%) AEs were suspected to be related to study treatment in the eltrombopag and placebo cohorts, respectively (see TABLE).

Thirty-three eltrombopag-treated (34%) and seven placebo-treated (15%) patients had one or more AEs that led to study discontinuation, most often because of sepsis (n=5) and increased alanine transaminase (n=3) in the eltrombopag group and general physical health deterioration (n=2) in the placebo group. Among patients who experienced serious AEs, two in the eltrombopag group were fatal and suspected to be related to treatment; no serious, fatal AEs were deemed related to placebo.

“Despite the modest results with respect to platelet transfusion–independence and hematologic improvement, the ASPIRE study suggests that eltrombopag might represent an appropriate monotherapy treatment option for some patients with MDS or AML and thrombocytopenia who have limited therapeutic alternatives,” the researchers concluded. “Given the requirement for an effective thrombocytopenia treatment for this high-risk population with expected poor outcomes, ASPIRE and other clinical studies might support the potential for oral eltrombopag monotherapy, … but further research is needed into alternative options.”

The study is limited by its small patient cohorts and limited duration of follow-up. The open-label design also may have introduced bias.

Novartis provided funding for the study.

The corresponding authors report financial support from Amgen, Celgene, GlaxoSmithKline, Janssen, Novartis, and Roche. Novartis provided editorial support.


Mittelman M, Platzbecker U, Afanasyev B, et al. Eltrombopag for advanced myelodysplastic syndromes or acute myeloid leukaemia and severe thrombocytopenia (ASPIRE): a randomised, placebo-controlled, phase 2 trial. Lancet Haematol. 2017 December 11. [Epub ahead of print]