We evaluated the efficacy and safety of romiplostim, a thrombopoietin mimetic, in patients with low-or intermediate-risk myelodysplastic syndromes (MDS) receiving azacitidine therapy. Forty patients with low-or intermediate-risk MDS were stratified by baseline platelet counts (< 50 vs > 50 ؋ 10 9 /L) and randomized to romiplostim 500 g or 750 g or placebo subcutaneously once weekly during 4 cycles of azacitidine. The primary endpoint was the incidence of clinically significant thrombocytopenic events, defined by grade 3 or 4 thrombocytopenia starting on day 15 of the first cycle or platelet transfusion at any time during the 4-cycle treatment period. No formal hypothesis testing was planned. The incidence of clinically significant thrombocytopenic events in patients receiving romiplostim 500 g, romiplostim 750 g, or placebo was 62%, 71%, and 85%, respectively. The incidence of platelet transfusions was 46%, 36%, and 69%, respectively. These differences were not statistically significant with the small numbers in each group. Romiplostim 750 g significantly raised median platelet counts during cycle 3 on day 1 (P ؍ .0373) and at the nadir (P ؍ .0035) compared with placebo. Grade 3 rash and arthralgia each were reported in 1 romiplostim-treated patient (4%