Evaluation of Azacitidine in Transfusion-Dependent, Epo-Refractory Patients with Lower-Risk Myelodysplastic Syndrome,

Transfusion-dependency (TD) in patients with low- and int-1-risk MDS is associated with increased morbidity and mortality (Malcovati, JCO 2007). Around 30% of such patients respond to Epo +/− G-CSF with a median duration of two years (Jadersten, Blood 2005). Azacitidine (Aza) has been reported to induce transfusion-independence (TI) in patients with lower-risk MDS (Lyons, JCO 2009). However, this treatment has not been systematically evaluated in patients with a well-documented resistance to Epo +/− G-CSF, neither has the benefit of the combination of Aza and Epo been evaluated in a prospective study.

This prospective open-label, non-randomized phase II study will include 30 consecutive patients with IPSS low- or int-1-risk MDS and a RBC transfusion need of ≥4 units q 4 weeks. Patients should be refractory to full-dose Epo + G-CSF for >8 weeks, or considered ineligible for such treatment according to a predictive model based on serum erythropoietin levels and transfusion need (Hellstrom-Lindberg, Br J Hem 2003). Included patients are treated with Aza 75 mg/m²/d for five days q 28 days for six cycles. Patients remaining TD after six cycles are treated with another three Aza cycles, with the addition of Erythropoietin â 60,000 units/week. Primary endpoint is number of patients achieving TI after six cycles and secondary endpoints are number of patients achieving TI after another three cycles of Aza+Epo, effect on bone marrow morphology, peripheral blood parameters, and safety profile.

Twenty-eight patients were enrolled from January 2010 until August 2011, another six were screen failures. See table 1 for patient characteristics. Seventeen patients were previously treated with and refractory to Epo (60,000 units/week) + G-CSF (300 μg/week), whereas remaining patients scored “low” in the predictive model and were thus considered refractory upfront. Median number of transfusion units needed was 7 (4–14) during the 8 weeks preceding inclusion. Fifteen patients are so far evaluable for treatment with Aza alone and 6 patients for Aza+Epo. TI was achieved in 2/15 patients (13%) after Aza alone. No additional responses were observed after the addition of EPO. Twenty-three serious adverse events were reported in 12 patients, with neutropenic fever (n=8), infections (n=4), and diarrhea (n=2) being most common. Three patients pre-terminated the study, two due to sustained neutropenia and one due to leukemic transformation. Nadir values after each cycle of Aza were seen at Week 3 for thrombocytes (147±34×10⁹) and at Week 4 for neutrophils (1.4±0.9 x10⁹), respectively, reflecting a cytotoxic effect on hematopoiesis.

Aza can induce TI in patients with TD lower-risk MDS, but response rate is lower in this group of documented EPO-G-CSF-refractory patients compared to previous reports of less well-controlled cohorts. Since toxicity is substantial, candidate patients for this treatment must be selected carefully. The combination of Aza and Epo remains to be further evaluated.

Off Label Use: Azacitidine for lower-risk MDS. Hellström-Lindberg:Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding.