OBJECTIVES:   This study was performed to evaluate the cost-effectiveness of three kinds of combination therapies including class II anti-cancer drugs in patients with advanced or metastatic gastric cancer. METHODS:   A Markov model was simulated to assess the clinical and economic impact over 5 years from societal perspective. Life Years Gained(LYG) were measured as a clinical outcome. In the model, Docetaxel+cisplatin+5-FU(DCF), S-1+cisplatin(SP), Capecitabine+cisplatin(XP) were selected as 1st line chemotherapies. When the disease progressed in the second line therapy, Leucovorin + 5-FU + Irinotecan (FOLFIRI), it was assumed that best supportive care was performed. Transition probabilities and mortality were calculated by using adjusted parameter of “time to progression(TTP) or progression free survival(PFS), overall survival(OS)”, which were obtained by indirect comparison (control group : 5-FU + Cisplatin). Both direct medical costs and direct non-medical costs were calculated. Costs and outcomes were discounted at an annual rate of 5% and sensitivity analysis was performed to evaluate uncertainty in the results. RESULTS:   SP was dominated by XP because the total LYG per patient was higher and cost was lower for XP compared with SP. When DCF was compared with XP, incremental LYG was 0.045. However, incremental cost of DCF was also 10,719,975 KRW. Incremental cost-effectiveness ratio for DCF compared to XP was calculated over 200 million per LYG. The results of the sensitivity analysis showed no significant difference. CONCLUSIONS:   Although a threshold of ICER is not fixed in Korea, GDP per capita is usually used for reference. In that case, it is considered that XP is cost-effective compared with DCF. Therefore, XP is the more cost-effective than DCF and SP. Further research should be carried out about cost-utility by using utility weight according to the state of gastric cancer.