Purpose. Using a genome-wide array-based comparative genomic hybridization (array-CGH), DNA copy number changes in uterine leiomyosarcoma were analyzed. Materials and methods. We analyzed 4 cases of uterine leiomyoma and 7 cases of uterine leiomyosarcoma. The paraffin-fixed tissue samples were microdissected under microscope and DNA was extracted. Array-based CGH and fluorescence in situ hybridization (FISH) were carried out with Genome database (Gene Ontology). Results. Uterine leiomyoma showed no genetic alterations, while all of 7 cases of uterine leiomyosarcoma showed specific gains and losses. The percentage of average gains and losses were 4.86%and 15.1%, respectively. The regions of high level of gain were 7q36.3, 7q33-q35, 12q13-12q15, and 12q23.3. And the regions of homozygous loss were 1p21.1, 2p22.2, 6p11.2, 9p21.1, 9p21.3, 9p22.1, 14q32.33, and 14q32.33 qter. There were no recurrent regions of gain, but recurrent regions of loss were 1p21.1-p21.2, 1p22.3-p31.1,9p21.2-p22.2,10q25-q25.2, 11q24.2-q25, 13q12-q12.13, 14q31.1-q31.3, 14q32.32-q32.33, 15q11-q12, 15q13-q14, 18q12.1-q12.2, 18q22.1-q22.3, 20p12.1, and 21q22.12-q22.13. In the high level of gain regions, BAC clones encoded HMGIC, SAS, MDM2, TIM1 genes. Frequently gained BAC clone-encoded genes were TIM1, PDGFR-β, REC Q4, VAV2, FGF4, KLK2, PNUTL1, GDNF, FLG, EXT1, WISP1, HER-2, and SOX18. The genes encoded by frequently lost BAC clones were LEU1, ERCC5, THBS1, DCC, MBD2, SCCA1, FVT1, CYB5, and ETS2/E2. A subset of cellular processes from each gene was clustered by Gene Ontology database. Conclusion. Using array-CGH, chromosomal aberrations related to uterine leiomyosarcoma were identified. The high resolution of array-CGH combined with human genome database would give a chance to find out possible target genes present in the gained or lost clones. Using a genome-wide array-based comparative genomic hybridization (array-CGH), DNA copy number changes in uterine leiomyosarcoma were analyzed. Materials and methods. We analyzed 4 cases of uterine leiomyoma and 7 cases of uterine leiomyosarcoma. The paraffin- Fixed-tissue samples were microdissected under microscope and DNA was extracted. Array-based CGH and fluorescence in situ hybridization (FISH) were carried out with Genome database (Gene Ontology). Results. Uterine leiomyoma showed no genetic alterations, while all of 7 cases of The percentage of average gains and losses were 4.86% and 15.1%, respectively. The regions of high level of gain were 7q36.3, 7q33-q35, 12q13-12q15, and 12q23.3. And the regions of homozygous loss were 1p21.1, 2p22.2, 6p11.2, 9p21.1, 9p21.3, 9p22.1, 14q32.33, and 14q32.33 qter. There were no recurrent regions of gain, but recurrent regions of loss were 1p21.1-p21.2, 1p22.3-p31.1, 9p21.2-p22.2, 10q25-q25.2, 11q24.2-q25, 13q12-q12.13, 14q31.1-q31.3, 14q32.32-q32.33, 15q11- q12, 15q13- q14, 18q12.1- q12.2, 18q22.1-q22.3, 20p12.1, and 21q22.12-q22.13. In the high level of gain regions, BAC clones encoded HMGIC, SAS, MDM2, TIM1 genes. Frequently acquired BAC clone-encoded genes were TIM1 The genes encoded by frequently lost BAC clones were LEU1, ERCC5, THBS1, DCC, MBD2, FGF4, KLK2, PNUTL1, GDNF, FLG, EXT1, WISP1, HER- A subset of cellular processes from each gene was clustered by Gene Ontology database. Conclusion. Using array-CGH, ​​chromosomal aberrations related to uterine leiomyosarcoma were identified. The high resolution of array-CGH combined with human genome database would give a chance to find out possible target genes present in the gained or lost clones.