Purpose: Cyclooxygenase-2 (COX-2) inhibitors hold promise for cancer chemoprevention; however, recent toxicity concerns suggest that new strategies are needed. One approach to overcome this limitation is to use lower doses of COX-2 inhibitors in combination with other established agents with complementary mechanisms. In this study, the effect of (-)epigallocate-ocatechin-3-gallate (EGCG), a promising chemopreventive agent from green tea, was tested alone and in combination with specific COX-2 inhibitors on the growth of human prostate cancer cells both in vitro and in vivo. Experimental Design: Human prostate cancer cells LNCaP, PC-3, and CWR22Rν1 were treated with EGCG and NS398 alone and in combination, and their effect on growth and apoptosis was evaluated. In vivo, athymic nude mice implanted with androgen-sensitive CWR22Rν1 cells were given green tea polyphenols (0.1% in drinking water) and celecoxib (5 mg/kg, i.p., daily, 5 days per week), alone and in combination, and their effect on tumor growth was evaluated. Results: Combination of EGCG (10-40 μmol/L) and NS-398 (10 μmol/L) resulted in enhanced (a) cell growth inhibition; (b) apoptosis induction; (c) expression of Bax, pro-caspase-6, and pro-caspase-9, and poly(ADP)ribose polymerase cleavage; (d) inhibition of peroxisome proliferator activated receptor γ; and (e) inhibition of nuclear factor-κB compared with the additive effects of the two agents alone, suggesting a possible synergism. In vivo, combination treatment with green tea polyphenols and celecoxib resulted in enhanced (a) tumor growth inhibition, (b) lowering of prostate-specific antigen levels, (c) lowering of insulin-like growth factor-I levels, and (d) circulating levels of serum insulin-like growth factor binding protein-3 compared with results of single-agent treatment. Conclusions: These data suggest synergistic and/or additive effects of combinatorial chemopreventive agents and underscore the need for rational design of human clinical trials. © 2007 American Association for Cancer Research.