A major obstacle to curing chronic myeloid leukaemia (CML) is residual disease maintained by tyrosine kinase inhibitor (TKI)-persistent leukemic stem cells (LSCs). These are BCR-ABL1 kinase-independent (1, 2), refractory to apoptosis (3, 4) and serve as a reservoir to drive relapse or TKI-resistance. We demonstrate that Polycomb Repressive Complex 2 (PRC2) is mis-regulated in chronic phase (CP) CML LSCs. This is associated with extensive reprogramming of H3K27me3 targets in LSCs, thus sensitizing them to apoptosis upon treatment with an EZH2-specific inhibitor (EZH2i). EZH2i does not impair normal hematopoietic stem cell (HSC) survival. Strikingly, treatment of primary CML cells with either EZH2i or TKI alone caused significant up-regulation of H3K27me3 targets, and combined treatment further potentiated these effects and resulted in significant loss of LSCs compared to TKI alone, in vitro, and in long-term bone marrow murine xenografts. Our findings point to a promising epigenetic-based therapeutic strategy to more effectively target LSCs in CML patients receiving TKIs.