Senescence is a tumor suppressive mechanism that induces a permanent proliferative arrest in response to an oncogenic insult or to the genotoxic stress induced by chemotherapy. We have recently described that some cells can escape this arrest, either because senescence was incomplete or as a consequence of a phenotypic adaptation. Malignant cells which resisted senescence emerged as more transformed cells that resist anoikis and rely on survival pathways activated by Akt and Mcl-1. In this study, we further characterize senescence escape, investigating how emergent cells could reproliferate. During the initial step of chemotherapy-induced senescence (CIS), we found that cyclin D1 was upregulated and that cell emergence was prevented when its main partner cdk4 was inactivated. Results indicate that this kinase induced the upregulation of the EZH2 methylase, a component of the polycomb PRC2 complex. Downregulated during the early step of treatment, the methylase was reactivated in clones that escaped senescence. The inactivation of EZH2, either by siRNA or by specific inhibitors, led to a specific inhibition of cell emergence. We used quantitative proteomic analysis to identify new targets of the methylase involved in senescence escape. We identified proteins involved in receptor endocytosis and described new functions for the AP2M1 protein in the control of chemotherapy-mediated senescence. Our results indicate that AP2M1 is involved in the transmission of secreted signals produced by senescent cells, suggesting that this pathway might regulate specific receptors involved in the control of CIS escape. In light of these results, we therefore propose that the cdk4-EZH2-AP2M1 pathway plays an important role during chemotherapy resistance and senescence escape. Since targeted therapies are available against these proteins, we propose that they should be tested in the treatment of colorectal or breast cancers that become resistant to first-line genotoxic therapies.