体质指数对三阴型乳腺癌预后的影响分析

摘要:目的　探讨体质指数与三阴型乳腺癌预后的关系。方法　收集2003-2010年四川省肿瘤医院收治的三阴型乳腺癌患者101例，根据体质指数值（BMI）分为正常组（＜24 kg/m2）和超重肥胖组（≥24 kg/m2），采用χ2检验比较两组间的临床病理特征差异；用Cox比例风险回归分析BMI与总生存和无病生存风险比（HR）。合并2000-2014年关于肥胖与三阴型乳腺癌预后的研究文献6篇，共4446名病例，采用RevMan 5.0软件进行数据分析，计算合并风险比（HR合并）及其95%CI。结果　本次随访研究发现正常组和超重肥胖组患者的总生存和无病生存风险无统计学意义（HROS＝0.632，POS＝0.0.501；HRDFS＝0.934，PDFS＝0.873）；Meta分析结果显示，肥胖与三阴型乳腺癌的HROS合并＝1.02，95%CI:0.97～1.08；HRDFS合并＝1.01，95%CI:0.96～1.07。结论　现有的研究尚不支持BMI改变三阴型乳腺癌预后的假设，但尚存在随访期短，人群代表性局限的问题。

Impact of body mass index on survival outcome in triple-negative breast cancer

Abstract: Objective To investigate the correlation between body mass index (BMI) and the prognosis of patients in triple negative breast cancer (TNBC). Methods A total of 101 TNBC patients were collected during year 2003 to 2010 in The Second People's Hospital of Sichuan. Based on BMI 101 TNBC patients were divided into 2 group: normal weight group (<24kg/m2) and overweight or obese group (≥24kg/m2). The Chi-square test was used to compare the differences in clinicopathologic features between obese or overweight and normal patients. Cox proportional hazards regression analysis was performed to analyze hazard ratio between BMI and overall survival or disease-free survival. We also performed a meta-analysis of 6 studies including 4446 women between 2000 and 2014 to test the relationship between obesity and prognosis in TNBC patients. RevMan 5.0 software was used for data analysis and for calculating HR and its 95% CI. Results In the follow-up study, OS and DFS of normal group were not significantly different from those of overweight and obesity group (HROS=0.632, POS=0.0.501; HRDFS=0.934, PDFS=0.873). In the meta-analysis, no significant associations were observed between obesity and prognosis in TNBC patients (HROS=1.02, 95%CI: 0.97~1.08; HRDFS=1.01, 95%CI: 0.96~1.07). Conclusion Existing research does not support that BMI was associated with decreased overall or disease-free survival in patients with TNBC. But these are limited by the short follow-up period and the limited population representativeness problems.