Cyclooxygenases (COXs) are a family of enzymes, which catalyze the rate-limiting step in prostaglandin biosynthesis. COX-2 is the inducible isoform, upregulated during inflammation and overexpressed in various cancers. There are evidences of a role for COX-2 in cell proliferation and apoptosis especially in solid tumors, whereas little is known for cancers of hematopoietic origin. In our study, we analyzed the effect of COX-2 inhibitors (nimesulide, NS-398 and celecoxib) on cell proliferation and apoptosis of a panel of leukemic and lymphoblastic cell lines, Hel, Jurkat, K562, K562, Raji and U937.We found that the different inhibitors slow down cell proliferation in the different hematologic cell lines tested. U937 cells appeared as the most sensitive, whereas K562 were the most resistant to this effect. We provide evidence that this modulation corresponds to an accumulation of the cells in G0/G1 paralleled by an early downregulation of c-Myc and the expression of cell type-specific differentiation markers in U937 (CD15) and Hel (CD41a and CD61). In the second part of our study, we investigate the effect of COX-2 inhibitors on apoptosis induced by chemotherapeutic agents in our cell models. We demonstrated that COX-2 inhibitors strongly prevent apoptosis induced by a panel of chemotherapeutic agents. We demonstrated an early prevention of apoptotic signaling, prior to Bax/Bak activation. The preventive effect is associated with an impairment of the ability of chemotherapeutic agents to trigger their apoptogenic stress. Altogether, our results demonstrate an anti-apoptotic effect of COX-2 inhibitors on intrinsic vs. extrinsic apoptosis at early steps of apoptosis commitment. These results suggest cautions in the use of COX-2 inhibitors with chemotherapy. In the third part of our project, we investigated the combination of COX-2 inhibitors with curcumin, a natural product known for its anti-tumor properties. Our findings show that curcumin alone leads to an accumulation of U937 cells in G2/M phase of cell cycle, followed by an induction of apoptosis. However, the pretreatment of U937 cells with celecoxib at non-apoptogenic concentrations, counteracted curcumin-induced apoptosis, thus showing that this combination is not a good anti-cancer strategy in our cell models.The chronic use of COX-2 inhibitors can be associated with severe side effects due to the inhibition of COX-2 enzyme. In the last part of our project, we demonstrated that 2,5 dimethyl-celecoxib (DMC), a structurally analogue of celecoxib, which is not able to inhibit COX-2 activity, induces an inhibition of cell proliferation and an induction of apoptosis in U937 and K562 cells. These effects are stronger than those observed with celecoxib. Thus, this compound demonstrated better anti-tumor properties and may represent a promising therapeutic approach against leukemia.Altogether, our study supports the idea that COX-2 inhibitors display anti-tumor effects in our cell models, but only when administrated alone. The effects observed with DMC suggest that this compound may represent an alternative approach to COX-2 inhibitors in cancer therapy.