Objective: To select the antisense oligonucleotides (asONs) which hybridize with the mRNA of vascular endothelial growth factor receptor2 (VEGFR2, also named as kinase insert domain-containing receptor:KDR) in an effective and specific way, and to investigate their antitumor activity in MCF-7 cells. Methods: The effective antisense oligonucleotides were chosen by computer prediction combined with oligonucleotide microarrays. The inhibition effect on MCF-7 cells proliferation was measured by MTT; and VEGFR2 expression was surveyed by Western-blotting and RT- PCR. Results:Using predicting secondary structure of VEGFR2 mRNA with RNA folding program, computer prediction designed 30 antisense oligonucleotide probes that were directed to local loose regions of RNA structure. In 30 probes, 4(4/30, 13.33%)antisense oligonucleotides showed strong hybridization intensities in oligonucleotide microarrays test and were selected. All these antisense oligonucleotides targeting 4 different sites of VEGFR2 mRNA lowered the level of VEGFR2 mRNA and protein present in MCF-7 cells. Proliferation of MCF-7 cells was reduced by 4 antisense oligonucleotides, respectively, in which asONl was the most effective, with the inhibitory rates being 53.06% at 0.8 μmol/L. Conclusion: Combination of computer prediction with oligonucleotide microarrays is an effective way in selecting optimal antisense oligonucleotides. The antisense oligonucleotides showed good correlation between their antitumor activity and the hybridization intensities. The antisense oligonucleotides targeting VEGFR2 mRNA demonstrated prominent antitumor role in vitro.