AML has a dismal prognosis. It was previously shown that the expression of gene coding for the hyperfusogenic gibbon ape leukemia virus envelope glycoprotein (GALV.fus) can efficiently kill leukemic cells. However, target killing effect of GALV.fus on leukemia cells may be limited. Viral vectors, such as retroviruses and adenoviruses, have been developed to deliver heterologous genes into tumors in vivo, but these vectors have some limitations for gene therapy of leukemia. Another virus that has drawn interest as a gene transfer vector is the Sindbis virus. Sindbis virus efficiently infects human tumor cells through the high-affinity 67 kDa Laminin receptor. We found that Laminin-R was obviously expressed in HL-60 and primary human AML cells, but weakly expressed in K562 cells and blood samples of normal human. So we reasoned that Sindbis-virus-based vectors might be ideal for target gene transfer of GALV.fus to acute myeloid leukemic cell. It was shown that Sindbis virus efficiently transduced human acute myeloid leukemic cells with high expression of Laminin-R and exhibited potent cytopathic potential. What is more, we found that CFU-GMs were significantly reduced after Sindbis virus carrying GALV.fus transduced human primary AML cells. Sindbis virus carrying GALV.fus was active against human AML xenografts in vivo. Taken together, we concluded that Sindbis virus carrying GALV.fus may be an useful strategy for gene therapy of acute myeloid leukemia.