OBJECTIVE： To evaluate the cost-effectiveness of SOX regimen （tegafur+oxaliplatin） vs. CapeOX regimen （capecitabine+oxaliplatin） in the treatment of metastatic colorectal cancer， and to provide reference for exploring more economical first-line regimen of metastatic colorectal cancer. METHODS： Based on published high-quality Ⅲ-phase randomized controlled trial， Markov model was established according to the process of disease development in patients with metastatic colorectal cancer. The model was divided into progression-free survival state， progressive disease state and death state. Combined with relevant data of our hospital， pharmacoeconomic cost-effectiveness analysis was conducted for SOX regimen and CapeOX regimen. Sensitivity analysis validation model was used to analyze the stability of the model. RESULTS： According to the results of Markov model operation， compared to standard CapeOX regimen， SOX regimen could increase 0.14 QALYs， and cost increased by 35 493.45 yuan； incremental cost-effectiveness ratio was 253 524.64 yuan/QALYs， which was higher than willingness-to-pay （WTP） threshold （168 201.201 yuan/QALYs）. Single factor sensitivity analysis showed that cost of oxaliplatin had the most important impact on the result of cost-effectiveness analysis. Probabilistic sensitivity analysis depicted that with the increase of GDP per capita， the probability of SOX regimen with cost-effectiveness would increase. CONCLUSIONS： At present， compared with standard CapeOX regimen， SOX regimen has no cost-effectiveness for metastatic colorectal cancer， which is not recommended as the first choice for first-line treatment of metastatic colorectal cancer.