基于SEER数据库男性乳腺癌早期死亡预测模型的建立及验证  

Establishment and validation of the early death prediction model for male breast cancer based on SEER database

作　　者：侯艳妮 陈楠 王美霞 Hou Yanni;Chen Nan;Wang Meixia(Department of Mammary,ShaanXi Provincial Cancer Hospital,Xi′an 710061,China;Department of Health Examination,Shenmu Hospital,Shenmu 719300,China)

机构地区：[1]陕西省肿瘤医院乳腺中心,西安710061 [2]神木市医院(西北大学附属神木医院),神木719300

出　　处：《中华转移性肿瘤杂志》2024年第5期448-455,共8页Chinese Journal of Metastatic Cancer

基　　金：陕西省重点研发计划项目基金(2022SF-010);陕西省卫生健康科研项目基金(2022A010)。

摘　　要：目的探讨影响男性乳腺癌患者早期死亡的危险因素并构建预测模型。方法通过SEER数据库收集男性乳腺癌患者,采用Logistic模型单因素及多因素分析筛选出全因早期死亡和肿瘤特异性早期死亡的危险因素。利用筛选出的独立危险因素构建列线图预测模型,并评价模型的临床获益和应用价值。结果单因素分析显示全因早期死亡与其婚姻状况、组织学分级、病理分期、T分期、N分期、M分期、手术、化疗、分子分型均有关(均P<0.05),与年龄、种族、肿瘤偏向性、放疗、发病年份均无关(均P>0.05)。多因素分析显示婚姻状况、T分期、M分期、手术、化疗、分子分型(HR+/HER2+、三阴性)均为男性乳腺癌患者预后差的独立危险因素(均P<0.05)。模型验证结果表明建模组和验模组的校正曲线表现出良好的一致性。模型的临床净获益较高,临床效能较好。结论该列线图能准确预测男性乳腺癌患者的预后,为临床诊疗提供科学依据。Objective To explore the risk factors of the early death in male breast cancer patients and build a prediction model.Methods The risk factors of all-cause and tumor-specific early death were screened by univariate and multivariate Logistic model analysis in male breast cancer patients who were collected by SEER database.The nomogram prediction model was constructed by using the independent risk factors screened,and the clinical benefit and application value of the model were evaluated.Results The univariate analysis showed the early death was related to marital status,histological grade,pathological stage,T stage,N stage,M stage,surgery,chemotherapy and molecular classification(all P<0.05),there was no significant difference in age,race,tumor bias,radiotherapy and age(all P>0.05).The multivariate analysis showed analysis,marital status,T stage,M stage,surgery,chemotherapy,and molecular typing(HR+/HER2+,triple negative)were independent risk factors for poor prognosis in male patients with breast cancer(all P<0.05).The verification result of the model showed that the correction curves of the model group and the check group had good consistency.The net clinical benefit of the model was higher and the clinical efficacy was better.Conclusion The nomogram can accurately predict the prognosis of male breast cancer patients and provide scientific basis for clinical diagnosis and treatment.

关 键 词：男性乳腺癌 预测模型 列线图 肿瘤特异性生存 总生存 

分 类 号：R737.9[医药卫生—肿瘤]