Objective To explore the prognostic value of primary lesion resection (PLR) in patients with stage Ⅳ triple positive breast cancer (TPBC), to analyze the difference of survival outcome between systemic therapy (ST) treated with PLR combined with system and ST alone, and to further explore the effect of PLR on the survival outcome of stage Ⅳ TPBC patients.

Methods The clinical characteristics, treatment and survival outcome of 993 female patients with breast cancer diagnosed as stage Ⅳ TPBC from 2010 to 2015 were extracted from SEER database. According to whether the primary tumor was operated or not, the patients were divided into PLR group and non-PLR (NPLR) group. The 1:1 tendency score was used to match the propensity score matching (PSM) balance mixed bias. COX proportional hazard model was used to analyze the independent prognostic factors affecting the survival of patients with stage Ⅳ TPBC. Kaplan-Meier method and Log-rank method were used for survival analysis, and the survival differences of two treatment modes (PLR+ST; ST) under different metastatic conditions (simple bone metastasis; simple visceral metastasis (lung, liver, brain); bone metastasis + visceral metastasis; unusual site metastasis) were compared. And further analyze the influence of PLR (breast conserving surgery/total mastectomy) on the survival outcome of patients.

Results Before PSM, there were more patients with younger age, lower metastatic load, earlier T stage, later N stage and lower degree of differentiation in PLR group than in NPLR group. After PSM, only the difference between brain metastasis (P=0.049) and N stage (P=0.027) was statistically significant. The median follow-up time of 648 patients after PSM was 41 months (0~83 months). The median overall survival (OS) was 57 months, the median OS in the PLR group was 73 months, and the NPLR group was 41 months (χ²=44.970, P < 0.001). The median breast cancer specific survival rate (BCSS) in the PLR group was 74 months, and 45 months in the NPLR group (χ²=43.373, P < 0.001). COX regression analysis showed that age, T stage, PLR, chemotherapy, and brain metastasis are independent prognostic factors that affect OS and BCSS in patients with stage Ⅳ TPBC. Survival analysis showed that before and after PSM, the PLR group had a significant survival benefit compared to the NPLR group. Subgroup analysis of patients after PSM showed that among patients with brain metastases, the PLR group did not benefit from survival. A subgroup analysis based on the metastasis status found that in the bone metastasis group, the simple visceral metastasis group, and the bone metastasis with visceral metastasis group, the PLR+ST group had a significant survival benefit compared with the simple ST group, but metastasis in the unusual site Except for bone, liver, lung, and brain metastases), stage Ⅳ TPBC patients had no statistically significant difference in survival outcomes compared with ST group. Survival analysis found that before PSM, the survival outcome of the breast-conserving group was better than that of the total mastectomy group, and after PSM, there was no significant difference in the survival outcome of the two groups.

Conclusions PLR can significantly improve the OS and BCSS, of stage Ⅳ TPBC patients except brain metastasis and unusual site metastasis. There is no significant effect on the survival outcome of stage Ⅳ TPBC patients.