Higher Erythroid Response Rates with Epoetin Alfa Versus Other Agents in Treatment-Naïve Low/Int-1 Myelodysplastic Syndrome Patients: A Comparative Meta-Analysis

Background: The treatment of newly diagnosed low/intermediate-1 (low/int-1) myelodysplastic syndromes (MDS) varies significantly from observation only (until the patient develops transfusion dependence) to using specific agents like erythropoiesis-stimulating (ESA), demethylating, or immunomodulatory (IMiD) agents. To date no prospective comparative clinical trial has been conducted to evaluate relative efficacy of these agents. In the present literature meta-analysis, we compared the erythroid response (ER) rates achieved with epoetin alfa (EPO)-based therapy (monotherapy or in combination with other growth factors or non-ESAs) versus single agent/combination non-ESA therapies for the treatment of anemia in patients with MDS and specifically in low/int-1 risk category.

Methods: Data extraction was performed on studies from PubMed and ASCO/ASH proceedings in MDS patients treated with EPO ± granulocyte-/granulocyte-macrophage colony stimulating factor (G/GM-CSF) or other non-ESAs (search cutoff date 9/30/07). To allow cross comparison, only studies using IWG or IWG-like ER criteria and treatment-naïve patients were selected. Patients were further stratified according to risk and non-ESA therapy. Pooled estimates of ER rates were calculated using random-effect (R-E) meta-analysis methods.

Results: Of 790 studies identified, 37 were included in the present analysis. Among the 23 studies that included low/int-1 MDS patients, a majority of the patients in both the EPO-based (82.4%) and non-ESA (74.4%) groups had refractory anemia or refractory anemia with ringed sideroblasts, and baseline transfusion dependency rates were lower in the EPO-based (52.4%) than in the non-ESA (91.7%) groups. As shown in the Table, among all MDS patients regardless of risk category, EPO-based regimens had a higher ER rate when compared to IMiD-based therapies (p=0.0328), or demethylating agent-based therapies (p=0.1660). In the patients with low/int-1 risk disease, EPO-based therapy had a significantly better ER rate when compared to single-agent/combination non-ESA-based therapies (p=0.0015) and specifically, IMiD-based therapies (p=0.0231). The studies with demethylating agent-based therapies in the present analysis did not include low/int-1 MDS patients.

Conclusions: These results suggest that, for appropriately selected treatment-naïve patients with low/int-1 MDS, among the currently available therapies, EPO-based regimens may yield a higher erythroid response than non-ESA therapies, with lenalidomide in del 5q patients being the only exception. Further prospective clinical trials are needed to determine relative clinical benefits with different agents in different MDS risk groups.