Lenalidomide Plus Erythropoietin Improves Erythroid Response in Low-Risk MDS Patients

Adding lenalidomide to erythropoietin treatment led to a significantly higher erythroid response rate than lenalidomide alone in patients with myelodysplastic syndromes (MDS) who were resistant to erythropoiesis-stimulating agents (ESAs), according to a randomized trial recently published in Leukemia. Unfortunately, this higher response rate failed to translate into a significantly higher rate of red blood cell (RBC) transfusion-independence.

“Treatment of anemia is the main therapeutic challenge for patients with MDS,” the authors, led by Andrea Toma, MD, from the hematology department at Hôpital Universitaire Henri Mondor in Creteil, France, wrote.

Lenalidomide increases transfusion-independence rates in patients with lower-risk MDS with del5q (but to a lesser degree in patients without del5q) resistant to ESAs alone, they noted, “however, our recent clinical experience suggested that the response rate to lenalidomide could be increased by the addition of ESA [in patients without del5q].”

To establish the benefit of adding erythropoietin to lenalidomide, Dr. Toma and colleagues conducted a randomized, intent-to-treat, phase III study of 131 patients with lower-risk, non-del5q, RBC transfusion-dependent MDS refractory to ESAs. Eighty-one patients (61.8%) had primary resistance to ESAs, and 50 patients (38.1%) relapsed after an erythroid response.

All patients in the study were classified as either “low-risk” (43.5%) or “intermediate-1” (56.5%), according to the International Prognostic Scoring System (IPSS), and required a median of six red blood cell units every eight weeks.

Patients were randomized to receive either:

  • Lenalidomide alone (10 mg per day for 21 days for 4 weeks; n=65)
  • Lenalidomide (10 mg per day for 21 days for 4 weeks) plus erythropoietin beta (60,000 units per week; n=66)

Thirty-two patients (16 in each study arm) did not receive the full four planned courses, due to adverse events (21 patients), early death (4), investigator’s or patient’s decision (5), suicide (1), and loss to follow-up (1).

In the intention-to-treat analysis, erythroid response after four treatment cycles (the study’s primary endpoint, defined as reduction of >4 units of RBC transfusion) was observed in 41 patients: 15 (23.1%) in the lenalidomide arm and 26 (39.4%) in the lenalidomide–erythropoietin arm (p=0.044). Median time to erythroid response was four months in both arms.

Median response duration was higher in the lenalidomide group compared with the lenalidomide–erythropoietin group, though the finding was not significant (18.1 months vs. 15.1 months; p=0.64). Seven of the 15 responders in the lenalidomide arm eventually relapsed, as did 13 of the 26 responders in the combination arm.

There was, however, no significant increase in RBC transfusion-independence rate with the combination arm (13.8% in the lenalidomide arm and 24.2% in the lenalidomide–erythropoietin arm; p=0.13), “possibly due to the insufficient size of our patient population or to the high transfusion burden in our patients,” the investigators noted. Transfusion-independence rates were significantly different, though, when the analysis was restricted to patients who did not receive more than four RBC units in the eight weeks before study inclusion.

Among the 58 non-responders, two patients (one in each study arm) experienced a late response – four weeks after treatment was discontinued in the lenalidomide arm and six weeks in the lenalidomide–erythropoietin arm.

After analyzing patient characteristics to identify factors that predicted a better response (including sex, age, cytogenetics, and type of ESA resistance), Dr. Toma and colleagues only found two:

  • baseline serum erythropoietin level <100 UI/I (odds ratio [OR] = 3.3; 95% CI 1.35-7.9; p=0.0087)
  • presence of the G allele at CBRN rs1672753 (OR=2.6; 95% CI 1.09-6.3; p=0.032)

Treatment-related adverse events were moderate and similar between the two treatment cohorts, the authors noted, with the most common grade 3 or 4 events being deep-vein thrombosis, neutropenia, and thrombocytopenia. “Still, about 20 percent of the patients had early protocol discontinuation due to side effects, mainly cytopenias, and the dose of lenalidomide had to be reduced in 31 percent and 36 percent of the patients in the lenalidomide and lenalidomide–erythropoietin arm, respectively,” they added. “These results suggest that a daily dose of lenalidomide of 10 mg during three weeks every four weeks may be too high in a substantial proportion of the patients. A lower starting dose may have to be considered in future studies in order to avoid early discontinuation in many patients.”

Limitations of the study, Dr. Toma and colleagues wrote, included the choice to use erythroid response, rather than RBC transfusion-independence, as its primary endpoint. The patient number was also underpowered to detect this outcome. “Achieving transfusion-independence, and not only a significant reduction of transfusion requirement may indeed be considered a better endpoint from a clinical perspective.”


Toma A, Kosmider O, Chevret S, et al. Lenalidomide with or without erythropoietin in transfusion dependent erythropoiesis-stimulating agent-refractory lower risk MDS without 5q deletion. Leukemia. 2015 October 26. [Epub ahead of print]