BACKGROUND: Prostate cancer is a major cause of cancer morbidity and mortality in men worldwide. Androgen deprivation therapy (ADT) has proven effective in early-stage androgen-sensitive disease, but prostate cancer gradually develops into an androgen-resistant metastatic state in the vast majority of patients. According to our oncogene-induced model for cancer development, senescence is a major tumor progression barrier. However, whether senescence is implicated in the progression of early-stage androgen-sensitive to highly aggressive castration-resistant prostate cancer (CRPC) remains poorly addressed. METHODS: Androgen-dependent (LNCaP) and -independent (C4-2B and PC-3) cells were treated or not with enzalutamide, an Androgen Receptor (AR) inhibitor. RNA sequencing and pathway analyses were carried out in LNCaP cells to identify potential senescence regulators upon treatment. Assessment of the invasive potential of cells and senescence status following enzalutamide treatment and/or RNAi-mediated silencing of selected targets was performed in all cell lines, complemented by bioinformatics analyses on a wide range of in vitro and in vivo datasets. Key observations were validated in LNCaP and C4-2B mouse xenografts. Senescence induction was assessed by state-of-the-art GL13 staining by immunocytochemistry and confocal microscopy. RESULTS: We demonstrate that enzalutamide treatment induces senescence in androgen-sensitive cells via reduction of the replication licensing factor CDC6. Mechanistically, we show that CDC6 downregulation is mediated through endogenous activation of the GATA2 transcription factor functioning as a CDC6 repressor. Intriguingly, GATA2 levels decrease in enzalutamide-resistant cells, leading to CDC6 stabilization accompanied by activation of Epithelial-To-Mesenchymal Transition (EMT) markers and absence of senescence. We show that CDC6 loss is sufficient to reverse oncogenic features and induce senescence regardless of treatment responsiveness, thereby identifying CDC6 as a critical determinant of prostate cancer progression. CONCLUSIONS: We identify a key GATA2-CDC6 signaling axis which is reciprocally regulated in enzalutamide-sensitive and -resistant prostate cancer environments. Upon acquired resistance, GATA2 repression leads to CDC6 stabilization, with detrimental effects in disease progression through exacerbation of EMT and abrogation of senescence. However, bypassing the GATA2-CDC6 axis by direct inhibition of CDC6 reverses oncogenic features and establishes senescence, thereby offering a therapeutic window even after acquiring resistance to therapy.