No Benefit With Adding Eltrombopag to Azacitidine in Patients With Higher-Risk MDS

In the phase III SUPPORT trial, researchers tested the hypothesis that concomitant administration of the thrombopoietin receptor agonist eltrombopag could ameliorate the thrombocytopenic effects of azacitidine, a standard frontline treatment for patients with intermediate- or high-risk myelodysplastic syndromes (MDS). While earlier trials showed “an acceptable safety profile of eltrombopag in patients with low-risk and high-risk MDS and positive outcomes on thrombocytopenia,” the authors of an analysis published in Blood found no benefit when it was added to azacitidine treatment.

Michael Dickinson, MD, from the Peter MacCallum Cancer Centre and Royal Melbourne Hospital in Australia, and authors enrolled adults diagnosed with intermediate- or high-risk MDS (defined as Int-1, Int-2, or high risk per the International Prognostic Scoring System [IPSS]) between June 2014 and December 2015. Participants also were required to have platelet counts <75×109/L within 28 days before the first dose of azacitidine.

Patients were randomized based on IPSS risk score, baseline platelet count, and platelet transfusion dependence to receive azacitidine with either eltrombopag (n=179) or placebo (n=177). Azacitidine 75 mg/m2 was administered as a once-daily subcutaneous injection for a total of seven days during 28-day cycles. Continuous doses of eltrombopag or placebo were administered at 200 to 300 mg/day starting on day 1 of azacitidine therapy.

Mean platelet counts in patients randomized to eltrombopag increased gradually during the study and to a greater degree than observed for those receiving placebo, though p values for this comparison were not reported. Patients received eltrombopag for a median of 83 days (range = 1-477 days), and 68 patients (38%) were able to receive the recommended six or more cycles of azacitidine, compared with 91 patients (51%) in the placebo group (p value not provided).

At a planned interim analysis, 147 patients (79 in the eltrombopag group and 68 in the placebo group) were evaluable for the primary endpoint of platelet transfusion independence during the first four cycles of azacitidine treatment. While fewer patients in the eltrombopag group than in the placebo group were transfusion independent, the difference was not statistically significant (primary endpoint; 16% [n=13/79] vs. 40% [n=27/68]; odds ratio [OR] = 0.25; 95% CI 0.11-0.61; one-sided p=1.0).

The authors also reported that, “while [the committee] found no difference in overall deaths that would indicate harm (14% eltrombopag vs. 12% placebo),” local review of disease progression and progression to acute myeloid leukemia (AML) indicated higher rates of progression for those receiving eltrombopag (14% vs. 5% for progression and 9% vs. 2% for progression to AML, respectively; p values not reported). Based on these results, an independent data monitoring committee recommended terminating the study prematurely, “primarily for futility and secondarily for safety.”

The authors conducted a final analysis from the point of study termination, finding that results were consistent with those from the interim analysis. Again, fewer patients were transfusion-independent in the eltrombopag group than in the placebo group (16% vs. 31%; OR=0.37; 95% CI 0.21-0.65; p=0.001).

There also were no significant differences observed between the two groups in terms of overall survival: 108 patients died, including 57 patients (32%) in the eltrombopag group and 51 patients (29%) in the placebo group (hazard ratio [HR] = 1.42; 95% CI 0.97-2.08; p=0.16). Most of the deaths were attributed to patients’ MDS or sepsis, and other causes of death in the eltrombopag group included cardiovascular death (n=2), hemorrhage (n=1), other cancer (n=1), and other noncardiovascular death (n=7). The risk of disease progression or death also appeared to be higher with eltrombopag (HR=1.47; 95% CI 1.05-2.07; p=0.06).

The final analysis also revealed that patients in the placebo group, compared with those in the eltrombopag arm, had:

  • higher overall response rates, including complete responses [CRs], marrow CRs, and partial responses (35% vs. 20%; OR=0.51; 95% CI 0.30-0.86; p=0.005)
  • lower rates of progression to AML (9% vs. 15%; OR=1.59; 95% CI 0.81-3.14; p=0.08)

The incidence of serious adverse events (AEs) and AEs leading to discontinuation appeared to be higher in the eltrombopag arm, compared with the placebo arm (60% vs. 42% and 20% vs. 8%, respectively; p values not provided).

A greater proportion of patients in the eltrombopag group discontinued study treatment due to AEs (25% vs. 14%) in the final assessment. Serious AEs appeared to be more common in the eltrombopag group than the placebo group (72% [n=128/177] and 56% [n=100/177]); more of these were considered related to treatment in the eltrombopag group than the placebo group (15% [n=26] and 2% [n=4]). AEs with a >5 percent difference between the treatment groups included febrile neutropenia (31% for eltrombopag vs. 21% for placebo), neutropenia (31% vs. 26%), nausea (31% vs. 26%), and diarrhea (25% vs. 14%), though no p values were reported for these comparisons.

“One hypothesis for our findings is a potential inhibition of the effects of azacitidine by eltrombopag when given concomitantly,” the researchers concluded. “This question remains unanswered and is the subject of further, ongoing research.”

The study’s short follow-up and the relatively small number of participants included in the entire cohort represent potential limitations of the analysis.

The authors report financial relationships with Novartis, which provided financial support and editorial assistance for this study.


Dickinson M, Cherif H, Fenaux P, et al. Azacitidine with or without eltrombopag for first-line treatment of intermediate- or high-risk MDS with thrombocytopenia. Blood. 2018 October 10. [Epub ahead of print]