Abstract: Objective To explore factors forecasting clinical efficacy of oxaliplatin（L-OHP）combined with capecitabine（CAP） as the firstline therapy for metastatic colorectal cancer（mCRC） and their correlation. Methods From March 2006 to December 2010，80 mCRC patients were treated by L-OHP combined with CAP（CAP 1000mg／m2，d2-d15；L-OHP 130mg／m2，d1. Twenty-one days was a cycle）. Classification and regression tree（CART）method was used to analyze the factors affecting objective response and predicting progressing-free survival（PFS）.Results There were 368 cycles of chemotherapy in 80 mCRC patients（average 4-6 cycles）. Twelve patients achieved complete remission，25 cases partial response，27 cases stable disease and 16 cases progressive disease. The response rate was 46.2％ and the median PFS was 9.6 months（95％CI：8.2-11.0 months），which were evaluated according to RECIST 1.0 criteria or follow-up. Through CART analysis，objective response and histological differentiation were first nodes，tumor-free interval and liver metastasis were second nodes，and there were four end subgroups. Response rate of high，and median differentiation with liver metastasis subgroup was the highest（77.8％）；low differentiation and undifferentiation with live metastasis subgroup was the lowest（8.0％）. The CART analysis of PFS showed that objective response was the first node and liver metastasis was the second node，and there were three end subgroups. The mean PFS of response rate subgroup was longest（11.1 months）；non response rate with liver metastasis subgroup was shortest（8.0 months）. There was a significant difference between response rate subgroup and non response rate with or without liver metastasis subgroup by Log-rank test（P＜005）. Conclusion Objective response of L-OHP combined with CAP regimen as the first-line therapy in mCRC may be relevant to tumor-free interval and liver metastasis，while PFS of them may be relevant to objective response and liver metastasis. More accurate forecast results are still worth discussing.