Clinopathological Characteristics And Construction Of Prognostic Model Of HR-positive Breast Cancer With HER2-low Expression

Objective: : To study the clinicopathological features and prognosis of HER2-low HR positive breast cancer patients.Methods: Patients with HR-positive breast cancer who received radical surgery in the Cancer Hospital Affiliated to Xinjiang Medical University from January 1 to December 31,2017 were selected to collect clinicopathological data and prognostic information,and analyze the clinicopathological characteristics and prognosis of HR-positive breast cancer patients with different HER2 expression levels.Frequency and rate were used to describe the counting data,and chi-square test was used for comparison between groups.All measurement data were treated with normal distribution(central limit theorem).Mean ± Standard Deviation was used for description.wilcoxon rank sum test was used for comparison between groups.Survival curve was drawn by Kaplan-Meier method,and survival analysis was performed by Log-rank test.Univariate and multivariate analyses were performed using Cox hazard proportional models,and hazard ratios(HR)and corresponding 95%CI were calculated to determine independent prognostic factors.A nomogram prediction model was constructed,its prediction value was evaluated by C index,and then verified internally by correcting curves.Results: According to the expression level of HER2,711 patients were divided into three groups,HER2-0、 HER2-low and HER2-high,among which 500cases(70.32%)were HER2-low.By analyzing the clinical and pathological characteristics of patients,there were significant differences in T stage,histological grade,ER,PR,Ki67 and age among three groups(P<0.05).Prognostic analysis showed that there were significant differences in DFS and OS among the three groups(OS:P=0.0022,DFS:P=0.0013).Univariate analysis showed that patients’ DFS and OS were related to their HER2 status(P<0.001),while multivariate analysis showed that HER2 expression status was not an independent risk factor for patients’ DFS and OS(P >0.05).The results showed that the C index of 1-year,3-year and 5-year DFS prediction models were 0.636,0.815 and 0.822,which were statistically significant by Cox regression analysis.The C index of 1-year,3-year and 5-year OS prediction models were 0.930,0.827 and 0.880,indicating that the model had strong generalization ability.The correction curves of1-year,3-year and 5-year DFS and OS are all close to the ideal 45°dashed line,showing good consistency.Conclusions: The clinicopathological features and prognosis of HR-positive HER2-low breast cancer patients are different from those of HER2-0 and HER2-high patients,and the DFS and OS of patients are related to their HER2 status.The nomogram model established in this study has good prediction value.It can accurately evaluate the DFS and OS of HER2-low breast cancer at 1,3 and 5 years,which is conducive to the individualized treatment of HER2-low patients in later clinical work.