Gene-directed enzyme prodrug therapy (GDEPT) involves the treatment concept of having maximal efficacy and minimal adverse effects. Several GDEPT strategies have been developed combining cytosine deaminase and 5-fluorocytosine, cytochrome P450 2B1 and cyclophosphamide, and carboxylesterase (CES) and irinotecan in experimental models. The active forms of these prodrugs, however, are not a frontline therapy for the treatment of ovarian cancer. It would be beneficial to develop a more effective prodrug-enzyme combination for the treatment of this disease. Paclitaxel (Taxol; TAX) is currently one of the most important anti-cancer drugs in chemotherapy of ovarian cancer. One of TAX prodrugs, 2'-ethylcarbonate-linked paclitaxel (TAX-2'-Et), was generated and examined regarding its pharmacological aspects. The prodrug of TAX-2'-Et converts into active form TAX by carboxylesterase (CES). TAX-2'-Et did not exhibit polarized transport in the Caco-2 cells expressing P-glycoprotein (P-gp) in the absence or presence of verapamil which is a inhibitor of P-gp, suggesting that TAX-2'-Et is not a target of P-gp like TAX and rhodamine123. Moreover, SKOV3/TAX60 cells which are overexpressing P-gp did not also exhibit any change in cellular uptake of TAX-2'-Et regardless of the absence or presence of verapamil. Consequently, the uptake of TAX-2'-Et into the TAX-resistant cells was quantitatively similar to that internalized in the parental SKOV3 cells which are P-gp-negative. In the CES-transfected SKOV3 cells, the EC50 value of TAX (10.6 nM) was approximately 4-fold higher than that of TAX-2'-Et (2.5 nM). We herein provide evidence that TAX-2'-Et could circumvent P-gp-associated cellular efflux of TAX, suggesting that this combination therapy is a potential GDEPT strategy for ovarian cancer in the future. Finally, this review focuses on the development, application and potential of various GDEPTs for treating ovarian cancer, and the scope and progress of new GDEPTs are discussed.