CC-486 Improves Transfusion Independence in MDS, but Adverse Events Are Common

For patients with lower-risk myelodysplastic syndromes (MDS) and both red blood cell (RBC) transfusion-dependent anemia and thrombocytopenia, CC-486, an oral formulation of the hypomethylating agent (HMA) azacitidine, significantly improved RBC transfusion independence compared with placebo, according to findings from a phase III study presented at EHA25 Virtual, the remotely held 25th European Hematology Association Annual Congress. However, the study was closed early at the recommendation of the trial’s independent Data Safety Monitoring Committee (DSMC) due to a higher incidence of deaths in the active therapy arm.

Lead investigator Guillermo Garcia-Manero, MD, of the University of Texas MD Anderson Cancer Center, told ASH Clinical News that he believes, based on these findings, that an oral HMA will eventually become standard in both lower- and high-risk MDS – conditions where treatment options are currently limited.

The study enrolled adults with International Prognostic Scoring System lower-risk MDS and anemia that required an average of ≥2 units of RBCs every 28 days for ≥56 days. Only those patients with 2 platelet counts of ≤75×109/L that were taken ≥21 days apart were considered eligible for study entry. Additional inclusion criteria were:

  • thrombocytopenia (sustained for ≥21 days) within 14 days of randomization
  • Eastern Cooperative Oncology Group performance status of 0-2
  • A total of 216 patients were randomly assigned to receive:
  • oral CC-486 at a starting dose of 300 mg daily (n=107)
  • identically matching placebo (n=108)

Both the active treatment and placebo were administered with best supportive care on days 1 through 21 of each 28-day treatment cycle.

The baseline characteristics were well balanced between the treatment arms, Dr. Garcia-Manero said. The median age of the cohort was 74 years (range = 30-89), and the median hemoglobin was 8.1 g/dL (range = 5.4-10.9). Participants had median platelet counts of 25×109/L (range = 5-73) and the median transfusion requirement was 6.7 units over 56 days.

He reported that all randomized patients either completed ≥12 months of treatment or discontinued the study before 12 months at time of data cutoff. The median treatment durations were 5 cycles for CC-486 (range = 1-70) and 6 cycles for placebo (range = 1-69).

Outcomes were assessed at cycle 6. Per study protocol, patients who achieved transfusion independence or a ≥50% reduction in transfusion requirements, had hematologic improvement in erythroid and platelet lineages, and had no evidence of progressive disease stopped treatment and were followed to measure survival, progression to acute myeloid leukemia, development of second primary malignancies, and need for other MDS treatments. Participants who remained transfusion dependent or had inadequate response at cycle 6 discontinued treatment.

At cycle 6, a significantly higher proportion of patients who received CC-486 achieved the primary outcome of RBC transfusion independence lasting ≥56 days: 30.8% versus 11.1% (p=0.0002). The investigators found that 28% of CC-486–treated patients were transfusion independent for ≥84 days, compared with 5.6% of placebo-treated patients (p<0.001).

Platelet transfusion independence rates were similar between the CC-486 and placebo groups, although the median duration of transfusion independence was numerically longer among patients treated with CC-486 (12.1 vs. 4.4 months).

Improvement in hemoglobin response was greater in the CC-486 arm than the placebo arm, with more patients experiencing a ≥1.5 g/dL hemoglobin increase from baseline (p<0.0001). Mean hemoglobin levels in the treatment arm increased by approximately 2 g/dL from baseline to cycle 6. Dr. Garcia-Manero pointed out that platelet counts increased by an average of 38×109/L by cycle 3 in the active therapy group, and the improvements in hemoglobin and platelet counts were sustained during therapy. However, changes in hemoglobin and platelets were considered negligible in the placebo group.

In the safety analysis, the researchers noted that approximately 30% of patients in the CC-486 group and 28% in the placebo group discontinued treatment because of AEs, the most common of which were grades 1 and 2 gastrointestinal events. The investigators reported that 90% of patients in the CC-486 arm and 73% of patients in the placebo arm experienced a grade 3/4 AE. Treatment-related AEs occurred more frequently with CC-486, particularly during cycles 1 and 2, Dr. Garcia-Manero noted.

The overall mortality rate was similar for both arms, and median survival was similar (17.3 vs. 16.7 months, respectively; p=0.88). However, more deaths occurred in patients treated with CC-486 (n=16) than with placebo (n=6) between days 1 and 56. Most deaths were associated with infection. This resulted in early termination of the study at the recommendation of the independent DSMC.

The authors report relationships with Celgene, the manufacturer of CC-486.


Garcia-Manero G, Santini V, Almeida A, et al. A phase III placebo-controlled trial of CC-486 in patients with red blood cell transfusion-dependent (RBC-TD) anemia and thrombocytopenia due to IPSS lower-risk myelodysplastic syndromes (LR-MDS). Abstract S180. Presented as part of EHA25 Virtual, June 12, 2020.