HR+/HER2-老年乳腺癌患者临床病理特征及预后分析

Clinicopathological features and prognostic analysis of elderly patients with HR+/HER2-breast cancer

目的:探讨激素受体阳性(hormone receptor-positive,HR+)/人表皮生长因子受体2阴性(human epidermal growth factor receptor-2 negative,HER2-)老年乳腺癌患者临床病理特征及预后相关情况,为临床治疗提供相关依据。方法:在监测、流行病学和结果(surveillance,epidemiology and end results,SEER)数据库中筛选出老年乳腺癌患者41972例,按照7∶3分成测试组和验证组,用Cox回归分析筛选出影响预后的相关因素并建立列线图,利用一致性指数(C-index)和ROC曲线及校准曲线评价模型的预测能力及一致性。结果:测试组与验证组的临床病理特征间差异无统计学意义(P>0.05)。Cox回归分析筛选出年龄、婚姻状况、种族、T分期、N分期、病理分级、手术、放疗、孕激素受体状态是60岁以上HR+/HER2-老年乳腺癌患者的独立影响因素(P<0.05)。测试组和验证组列线图的C-index为0.761和0.763,测试组1、3、5年的ROC曲线的AUC为0.810、0.791、0.784,验证组为0.811、0.799、0.793。结论:利用SEER数据库建立了HR+/HER2-老年乳腺癌患者预后模型,可为临床诊疗时提供更准确的参考依据。

Objective:To investigate the clinicopathological features and prognostic correlation of hormone receptor-positive(HR+)/human epidermal growth factor receptor-2 negative(HER2-)elderly patients with breast cancer,so as to provide relevant evidence for clinical treatment.Methods:A total of 41972 elderly patients with breast cancer were selected from the surveillance,epidemiology and end results(SEER)database and divided into test group and verification group according to 7∶3.Cox regression was used to screen out prognostic factors and establish a nomogram.C-index,ROC curve and calibration curve were used to evaluate the predictive power and consistency of the model.Results:There was no significant difference in clinicopathological features between test group and verification group(P>0.05).Cox regression analysis showed that age,marital status,race,T stage,N stage,tumor grade,surgery,radiotherapy and progesterone receptor status were independent influencing factors for HR+/HER2-breast cancer patients over 60 years old(P<0.05).The C-index of nomogram in test group and verification group were 0.761 and 0.763,and the AUC in test group were 0.810,0.791 and 0.784 at years 1,3 and 5,and those in verification group were 0.811,0.799 and 0.793.Conclusion:The prognostic model of elderly patients with HR+/HER2-breast cancer was established by using SEER database,which can provide a more accurate reference for clinical diagnosis and treatment.