Establishment and Validation survival prediction models for T1 locally advanced breast cancer after breast conservation surgery versus mastectomy

Previous reports had shown that the survival rate of total mastectomy (TM) was better than that of breast-conserving surgery (BCS). This study established survival prediction models for T1 LABC mastectomy and BCS, and obtained the risk factors for OS of different surgical procedures, so as to provide a basis for clinicians to individualized treatment.Cases with pathologically confirmed T1 breast cancer (BC) between 2010 and 2015 in the Surveillance, Epidemiology, and End Results (SEER) database were identified. COX regression analysis was used to analyze the relationship between LABC TM, BCS and each factor, and the hazard ratio (HR) and 95% confidence interval (95%CI) were calculated to determine the possible influencing factors. The significant factors from multivariate COX regression were included in the model and then the nomogram was drawn. The receiver operating characteristic (ROC) curve of the model was drawn, and the area under the curve (AUC) and its 95% CI were calculated. Hosmer-Lemeshow goodness-of-fit test was performed. The results were validated in the validation group. The 5-year overall survival (OS) and breast cancer specific survival (BCSS) of BCS were higher than those of TM. Age, race, histological grade, N stage, molecular typing, chemotherapy and radiotherapy (RT) were correlated with 5-year OS of BCS, and age, race, pathological type, histological grade, human epidermal growth factor receptor-2 (Human epidermal growth factor receptor 2, HER2) status, N stage, molecular typing, chemotherapy and RT were related to 5-year OS of TM. The predictive nomogram was established using the above predictors, and the AUC of the modeling group was 0.743 (BCS 5-year OS) and 0.718 (TM 5-year OS), respectively. All models were well validated in the validation group. This study found that the survival rate of BCS group was better than that of TM group, and it indicated the effect of tumor size on BCS survival, while lymph node status was not a risk factor for BCS, BCS could be considered for LABC patients with small masses and more lymph node metastases. However, the risk of death after BCS in patients with N3, triple-negative and upper-inner quadrant primary tumors was higher than that in other groups, and BCS should be cautious in these patients.