Results from the phase 3 SUPPORT trial suggest there is no role for the combination of eltrombopag and azacitidine in patients with intermediate- or high-risk myelodysplastic syndromes (MDS), according to investigators.
Adding eltrombopag to azacitidine worsened platelet recovery, reduced the overall response rate, and did not improve overall or progression-free survival when compared to azacitidine plus placebo.
Investigators had hypothesized that eltrombopag would reduce the thrombocytopenia exacerbated by azacitidine treatment.
Not only was this not the case, but the investigators observed a trend toward increased progression to acute myeloid leukemia (AML) in eltrombopag-treated patients.
An independent monitoring committee recommended the trial be terminated early.
Michael Dickinson, MBBS, of Peter MacCallum Cancer Centre and Royal Melbourne Hospital in Melbourne, Australia, and his colleagues reported results from the trial in Blood.
The investigators conducted the SUPPORT study (NCT02158936) to investigate the efficacy and safety of eltrombopag as platelet supportive care in patients with intermediate- to high-risk MDS and thrombocytopenia who were receiving azacitidine.
Investigators randomized patients 1:1 to eltrombopag or placebo in combination with azacitidine. Patients had to have at least one platelet count less than 75 x 109/L within 28 days prior to the first azacitidine dose.
Patients received azacitidine subcutaneously at a dose of 75 mg/m2 once daily on 7 consecutive days for 28 days.
Patients received eltrombopag or placebo at a starting dose of 200 mg/day (100 mg/day for East Asians), adjusted by 100 mg increments (50 mg increments for East Asians) to a maximum of 300 mg/day (150 mg/day for East Asians).
Investigators enrolled 356 patients from 30 countries between June 2014 and December 2015, randomizing 179 patients to eltrombopag and 177 to placebo.
Patients were a median age of 70, 66% were male, 83% were white, and 47% had platelet counts between 20 and 50 x 109/L.
At the start of the trial, 19% of patients (n=66) were platelet transfusion-dependent—16% (29/179) in the eltrombopag group and 21% (37/177) in the placebo group.
Patients received eltrombopag for a median of 83 days (range, 60 to 148) and placebo for a median of 149 days (range, 8 to 503).
For non-East Asian patients, the mean dose of eltrombopag was 205 mg/day (range, 65 to 293), and the mean dose of placebo was 245 mg/day (range, 107 to 316).
Sixty-eight patients (38%) in the eltrombopag arm received the recommended number of six or more azacitidine cycles, compared to 91 (51%) in the placebo arm.
Two patients in the eltrombopag arm did not receive treatment. Therefore, the safety analyses were conducted on the 177 treated patients in each arm.
The most common reason for treatment discontinuation was termination of the study. Thirty-two percent of patients in the eltrombopag cohort and 44% in the placebo cohort discontinued for this reason.
Patients had to discontinue eltrombopag or placebo if azacitidine was discontinued, and this occurred in 30% of patients in the eltrombopag group and 26% in the placebo group.
Twenty-two percent of eltrombopag-treated patients discontinued due to adverse events (AEs), compared with 14% in the placebo group.
The most common reasons for azacitidine discontinuation in the eltrombopag and placebo arms, respectively, were study termination (35% and 46%), AEs (25% and 19%), disease progression (15% and 14%), and patient decision (11% and 9%).
AEs with the greatest difference between the eltrombopag and placebo arms, respectively, were febrile neutropenia (31% and 21%), neutropenia (31% and 26%), nausea (31% and 26%), and diarrhea (25% and 14%).
Sixty-three of the 177 AEs in the eltrombopag group were suspected to be treatment-related, compared to 42 of 173 AEs in the placebo group.
One hundred twenty-eight (72%) serious AEs occurred in the eltrombopag arm, compared to 100 (56%) in the placebo arm.
Fifteen percent of the serious AEs were suspected to be related to eltrombopag treatment, compared to 4% of the serious AEs in the placebo group.
At the final analysis, 108 patients had died—57 (32%) in the eltrombopag group and 51 (29%) in the placebo group (hazard ratio [HR] 1.42; 95% CI 0.97, 2.08; nominal P=0.164).
The main causes of death were disease progression (28 in the eltrombopag group and 21 in the placebo group) and sepsis (18 in the eltrombopag group and 13 in the placebo group).
The study’s primary endpoint was the proportion of patients free of platelet transfusions during cycles one to four of azacitidine therapy.
At the final analysis, there were fewer patients in the eltrombopag arm than in the placebo arm who had achieved platelet transfusion independence—16% (28/179) and 31% (55/177), respectively. The odds ratio (OR) was 0.37 (95% CI 0.21, 0.65; two-sided P value=0.001).
Secondary efficacy endpoints included overall survival, disease response, duration of response, progression to AML, progression-free survival, and hematologic improvement.
The investigators pointed out that no formal statistical tests were performed for the secondary endpoints. Therefore, “statistical interpretation should be made with caution,” they wrote.
Overall response by IWG criteria occurred in 20% of patients in the eltrombopag group and 35% of those in the placebo group, according to investigator assessment (OR=0.51; 95% CI 0.30, 0.86; nominal P=0.005).
There was no significant difference in hematologic improvement, overall survival, or progression-free survival between the treatment arms.
The investigators found a higher rate of progression to AML in the eltrombopag arm than in the placebo arm—15% and 9%, respectively (OR=1.59; 95% CI 0.81, 3.14; nominal P=0.079).
They pointed out that these results contrasted with those of recent clinical studies of eltrombopag monotherapy1,2,3,4 in patients with MDS.
“[T]he findings of this trial were unexpected,” the investigators wrote.
They hypothesized that eltrombopag inhibits the effects of azacitidine when the drugs are given concomitantly. The issue is being studied further with ongoing research, they said.
The authors acknowledged financial support for medical editorial assistance provided by Novartis Pharmaceuticals Corporation.
Dr. Dickinson disclosed relationships with Celgene, Novartis, Janssen, Pfizer, and Roche. Senior study author Uwe Platzbecker, MD, disclosed relationships with Amgen, Celgene, GlaxoSmithKline, and Novartis.
1. Platzbecker U, et al Lancet Haematol. 2015;2(10):e417-e426. DOI: https://doi.org/10.1016/S2352-3026(15)00149-0
2. Oliva EN, et al. Lancet Haematol. 2017;4(3):e127-e136. DOI: https://doi.org/10.1016/S2352-3026(17)30012-1
3. Mittelman M, et al. Lancet Haematol. 2017;5(1):e34-e43. DOI: https://doi.org/10.1016/S2352-3026(17)30228-4
4. Buckstein R. Lancet Haematol. 2015;2(10):e396-e397. DOI: https://doi.org/10.1016/S2352-3026(15)00200-8