Abstract Myelodysplastic syndromes and related myeloid neoplasms (myelodysplasia) are a heterogeneous group of clonal disorders showing deregulated blood cell production and a predisposition to acute myeloid leukemia, whose pathogenesis is only incompletely understood. So, to clarify the molecular pathogenesis of myelodysplasia, we performed whole-exome sequencing of paired tumor/control DNA from 29 patients with myelodysplasia, leading to the identification of novel pathway mutations of the splicing machinery in myelodysplasia (Yoshida et al., Nature, 2011). In addition to these pathway mutations, we also identified a number of previously unreported gene mutations. Among these are a missense and a nonsense mutation involving two cohesin components, STAG1 and STAG2 found in single cases, respectively. Cohesin is a multimeric protein complex and enables post-replicative DNA repair and chromosome segregation by holding sister chromatids together during mitosis. To extend the findings in the whole-exome sequencing, we investigated mutations of cohesin complex, including STAG2/STAG1, SMC1A, SMC3 and RAD21, in 370 cases of myeloid malignancy by deep sequencing of pooled DNA. In total, 38 mutations were identified in 36 out of the 370 cases, where STAG2 and RAD21 accounted for most of the mutations. These mutations occurred in a completely mutually exclusive manner, suggesting a common impact of these mutations on the pathogenesis of myeloid neoplasms. Most mutations of STAG2 and RAD21 were nonsense or frameshift changes, or splice site mutations and widely distributed along the entire coding region, causing loss-of-function of the proteins. On the other hand, all mutations detected in SMC1A, SMC3, and STAG1 were missense changes, indicating that their functions are essential for tumor survival, complete loss of functions of which could lead to cell death. In cytogenetics, 11 cohesin-mutated cases had normal karyotypes, and only 16 out of the 36 tumors with cohesion mutations showed abnormal karyotypes, where most cases had near-diploid with only 2 patients having complex karyotypes. So far, several lines of evidence suggest that cohesin plays an important role for genomic stability and mutational inactivation of STAG2 was shown to cause aneuploidy in human cells. However, our results raise the possibility that alterations of cohesin genes could be involved in carcinogenesis at least partly through mechanisms other than causing aneuploidy. In this context, it is of note that growing evidence have shown that cohesin forms long-range chromosomal interactions and regulate gene expression in association with CTCF, mediator, or transcription factors. Further functional study should be warranted to gain new insights into the role of cohesin in the pathogenesis of myeloid malignancies as well as other human cancers. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5117. doi:1538-7445.AM2012-5117