基于SEER数据库分析老年乳腺癌患者预后相关因素

目的:基于SEER数据库利用竞争风险模型分析老年女性乳腺癌患者的乳腺癌特异性死亡和非乳腺癌死亡的预后因素。方法:选择SEER数据库中2006年至2010年期间年龄≥70岁女性乳腺癌患者，利用R统计软件中的“cmprsk”包进行竞争风险模型的单因素及多因素预后分析。结果:共纳入31 823例老年乳腺癌患者，13 075例在随访中死亡，其中，3 735例因乳腺癌死亡，9 340例因非乳腺癌死亡。患者1年、3年及5年乳腺癌特异性死亡率分别为1.47%、5.39%及8.71%,非乳腺癌死亡率分别为2.17%、8.49%和15.79%。多因素分析结果显示，高龄、单身状态、黑种人、雌激素受体阳性、低分化、较晚T和N分期及未接受放疗是影响老年患者乳腺癌死亡的独立预后因素；高龄、单身状态、雌激素受体阳性、较晚的T分期等是影响老年患者非乳腺癌死亡的独立预后因素。结论:老年乳腺癌患者更多因非乳腺癌死亡而不是乳腺癌，其接受放疗可降低乳腺癌特异性死亡和非乳腺癌死亡。

Analysis of prognostic factors in elderly breast cancer patients based on SEER database

Objective: To analyze the prognostic factors of breast cancer-specific and non-breast cancer-specific deaths in elderly female breast cancer patients, using a competing risk model based on the SEER database. Methods: Female breast cancer patients aged ≥70 years were selected from 2006 to 2010 in the SEER database, and the “cmprsk” package in the R statistical software was used to perform univariate and multivariate analysis of the competing risk model. Results: A total of 31 823 elderly patients were included in the study, and 13 075 died during follow-up. Among them, 3 735 died from breast cancer, and 9 340 died from non-breast cancer causes. The 1-, 3-, and 5-year breast cancer-specific mortality rates were 1.47%, 5.39% and 8.71%, respectively; and the non-breast cancer mortality rates were 2.17%, 8.49% and 15.79%, respectively. The results of multivariate analysis showed that advanced age, single status, black race, estrogen receptor positivity, poorly differentiated, advanced T and N stages and no radiotherapy were independent poor prognostic factors affecting the cause of breast cancer death in elderly patients; advanced age, single status, estrogen receptor positivity, advanced T stage and other factors were independent poor prognostic factors affecting non-breast cancer death in elderly patients. Conclusion: Elderly breast cancer patients are more likely to die from non-breast cancer than from breast cancer, and their radiotherapy could reduce breast cancer-specific and non-specific breast cancer deaths.［Key words］elderly; breast cancer; SEER database; estrogen receptor; competing risk model; prognostic factors