The specific regimen was determined based on guideline recommendations and the doctor's choice. A total of 40 patients received treatment including eribulin, with 22 patients receiving monotherapy with eribulin and 18 patients receiving eribulin in combination with anti-angiogenic medicine (bevacizumab or anlotinib). Combination therapy with anti-human epidermal growth factor receptor-2(HER-2) targeted treatment was allowed for HER-2-positive patients. The patient received 1.4 mg/m 2 of eribulin intravenous chemotherapy on the 1 st and 8 th day of each cycle, with bevacizumab (7.5 mg/kg, administered on the 1 st day of each cycle) or anlotinib (12 mg once daily for 14 consecutive days followed by a 7-day break) as an anti-angiogenic agent. The treatment regimen consisted of a 21-day cycle and continued until the disease progressed or intolerable toxicity occurred. After every two treatment cycles, the efficacy of the anti-tumour treatment was evaluated according to the response evaluation criteria in solid tumours (RECIST) version 1.1. 12 Ten clinical variables were recorded: Patient age, Eastern Cooperative Oncology Group (ECOG) performance status, hormone receptor (HR) (includes oestrogen receptor, ER, and / or progesterone receptor, PR) expression status, HER-2 expression status, Ki-67 levels, number of metastases, number of lines of eribulin treatment, whether combined with anti-angiogenic medicine therapy, history of taxane resistance, and the occurrence of grade 3-4 neutropenia after eribulin treatment.
Statistical analysis was performed using IBM SPSS 19.0 software. Qualitative variables were described by frequencies and percentages, continuous and ordinal variables by mean ± SD and median and interquartile ranges. The Kaplan-Meier method was used to calculate the median PFS (the time from the start of eribulin treatment to the time of disease progression) and the corresponding 95% confidence interval (CI). The log-rank test was used to compare differences in PFS between different groups. The Cox regression model was used for multivariate analysis. The Fisher exact probability test was used to compare the difference in adverse reactions between the two groups, with a level of significance set at p-value <0.05.