The CSC population is differentially modulated according to abexinostatresponse prole.B, cell differentiation was monitored by measuring expression of differentiation markers using immunouorescent staining; nuclei were counterstained using, diamidinophenylindole. Our ndings suggest that abexinostat could be used as a novel therapeutic strategy for breast cancer through the induction of CSC differentiation.However, a biomarker is needed to predict drug response of patients with breast cancer.None of the classical molecular parameters tested could predict BCL drug response. Xist expression level was signicantly correlated between both techniques. We observed a strong correlation between X chromosomes SDMA number and drug response to abexinostat.We injected single cancer cells into fat pads of NODSCID mice and monitored tumor growth.When the tumor size was approximately mm, we started treatment with abexinostat or docetaxel.Tumor growth was compared with that of placebotreated controls.For each treatment condition, cells isolated from treated tumors were reimplanted.In sharp contrast, cells isolated from docetaxeltreated tumors showed a tumor regrowth comparable with placebotreated tumors.Interestingly, for CRCMX, cells isolated from abexinostattreated tumors presented a higher regrowth kinetic compared with cells isolated from docetaxel and placebotreated tumors. Both inhibiting selfrenewal and promoting differentiation would deplete the CSC pool and allow more differentiated tumor cells to be targeted by conventional treatments.These proles were associated with opposite effects on the breast CSC population.In addition, cellcycle progression was transiently stopped with an accumulation in G cellcycle phase.This checkpoint before entering S phase, also called R point, has been dened as an important cellcycle stage controlling stem cell fate allowing equilibrium between selfrenewal and committed cell fate decision. Indeed, a low dose of decitabine on epithelial and leukemic cancer cells had no immediate toxicity, induced memory response with cell differentiation and CSC depletion in serially transplanted mice, but a high dose triggered rapid DNA damages and cytotoxicity. If the molecular reason SDMA explaining the dual effect of epidrug treatment is unclear, we can postulate that the abexinostat effect is mediated at the cellular level through the modulation of the CSC pool.Xist is responsible for X dosage compensation of X genes between males and females. A wider link between X chromosome inactivation and oncogenesis has been made in a number of studies observing a gain or loss of X chromosomes in tumor cells. Further studies are needed to decipher the precise underlying mechanism.For the experiments, abexinostat was prepared in a. molL stock solution, in dimethyl sulfoxide and stored at C.Valproic acid was prepared in a molL stock solution in PBS and stored at C.For experiments, cells were treated with respective IC.Then blots were washed, incubated with appropriate secondary antibodies. Antibodies used were antiacetylated histone, antiP. After hours, cells were treated with respective IC or vehicle.Caspase activity induction was measured, and hours later according to manufacturers guidelines.Tumorspheres were grown in a serumfree mammary epithelium basal medium.We injected, per fat pads of NODSCID mice and monitored tumor growth.When tumor size was approximately mm, we initiated treatment with abexinostat alone or placebo injected with cyclodextrin were injected for each PDX and for each group.After, and weeks of treatment, two mice from each group were sacriced according to ethic statements.