Objective: Bevacizumab has been challenging in the treatment of metastatic breast cancer. To investigate its efficacy, optimal partner to combine with and maintenance therapy, we performed a retrospective study based on Chinese patients with metastatic breast cancer (MBC). Methods: Patients with MBC treated with bevacizumab-contained regimens at the Sun Yat-sen University Cancer Center from 2006 to 2010 were recruited to the study. The primary endpoints were overall survival (OS), time to progression (TTP), objective response rate (ORR), and disease control rate (DCR). These endpoints were analyzed using the Kaplan-Meier and Chi-squared tests, respectively. Results: (1) A total of 229 cycles of bevacizumab with a median cycle of 7 (1-34) were administered among 25 patients. (2) In the whole group, ORR and DCR were 60% (15/25) and 76% (19/25), respectively. The mTTP was 5 months (1-21), mOS from diagnosis was 48 months (13-172), mOS from bevacizumab administration was 24 months (1-45). (3) Both ORR (73.7% vs. 16.7%, P = 0.023) and DCR (94.7% vs. 33.3%, P = 0.005) were significant higher once patients treated with the combination of taxanes-based regimen and bevacizamab when compared with the combination with non-taxanes-based regimens. (4) In the taxanes-based group, no matter bevacizumab used in first line or non-first line, the differences of ORR (P = 0.637) and DCR (P = 0.316) were insignificant. However, the maintenance therapy with bevacizumab will bring more longer TTP (P < 0.001) than those without maintenance therapy. Conclusion: Taxanes-based regimens were the optimal candidate to combine with bevacizumab regardless the timing in palliative setting, however, the maintenance therapy with bevacizumab should be considered once indicated. Objective: Bevacizumab has been challenging in the treatment of metastatic breast cancer. To investigate its efficacy, optimal partner to combine and maintain therapy, we performed a retrospective study based on Chinese patients with metastatic breast cancer (MBC). Methods: Patients with MBC treated with bevacizumab-contained regimens at the Sun Yat-sen University Cancer Center from 2006 to 2010 were recruited to the study. The primary endpoints were overall survival (OS), time to progression (TTP), objective response rate (ORR), and Disease control rate (DCR). These endpoints were analyzed using the Kaplan-Meier and Chi-squared tests, respectively. Results: (1) A total of 229 cycles of bevacizumab with a median cycle of 7 (1-34) were administered among 25 patients. (2) In the whole group, ORR and DCR were 60% (15/25) and 76% (19/25), respectively. The mTTP was 5 months (1-21), mOS from diagnosis was 48 months (13-172), mOS from bevacizumab administration was 24 months (1-45). (3) Both ORR (73.7% vs. 16.7%, P = 0.023) and DCR (94.7% vs. 33.3%, P = 0.005) were significantly higher once patients treated with the combination of taxanes-based regimen and bevacizamab when compared with the combination with non-taxanes-based regimens. (4) In the taxanes-based group, no matter bevacizumab used in first line or non-first line, the differences of ORR (P = 0.637) and DCR (P = 0.316) were However, the maintenance therapy with bevacizumab will bring more longer TTP (P <0.001) than those without maintenance therapy. Conclusion: Taxanes-based regimens were the optimal candidate to combine with bevacizumab regardless the timing in palliative setting, however, the maintenance therapy with bevacizumab should be considered once indicated.