Mutations in genes encoding splicing factors (which we refer to as spliceosomal genes) are commonly found in patients with myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML). These mutations recurrently affect specific amino acid residues, leading to perturbed normal splice site and exon recognition. Spliceosomal gene mutations are always heterozygous and rarely occur together with one another, suggesting that cells may tolerate only a partial deviation from normal splicing activity. To test this hypothesis, we engineered mice to express a mutated allele of serine/arginine-rich splicing factor 2 (Srsf2(P95H))-which commonly occurs in individuals with MDS and AML-in an inducible, hemizygous manner in hematopoietic cells. These mice rapidly succumbed to fatal bone marrow failure, demonstrating that Srsf2-mutated cells depend on the wild-type Srsf2 allele for survival. In the context of leukemia, treatment with the spliceosome inhibitor E7107 (refs. 7,8) resulted in substantial reductions in leukemic burden, specifically in isogenic mouse leukemias and patient-derived xenograft AMLs carrying spliceosomal mutations. Whereas E7107 treatment of mice resulted in widespread intron retention and cassette exon skipping in leukemic cells regardless of Srsf2 genotype, the magnitude of splicing inhibition following E7107 treatment was greater in Srsf2-mutated than in Srsf2-wild-type leukemia, consistent with the differential effect of E7107 on survival. Collectively, these data provide genetic and pharmacologic evidence that leukemias with spliceosomal gene mutations are preferentially susceptible to additional splicing perturbations in vivo as compared to leukemias without such mutations. Modulation of spliceosome function may thus provide a new therapeutic avenue in genetically defined subsets of individuals with MDS or AML.