目的：探讨绝经后女性内源性性激素及血脂水平与乳腺癌的关系以及作为预测指标的可行性。方法：收集2011 年7 月—2014 年6 月期间手术治疗的绝经后女性乳腺癌患者274 例（乳腺癌组）与同期因其他疾病收治的非乳腺疾病的绝经后患者279 例（对照组）；比较两组间性激素与血脂水平的差异，分析可能的乳腺癌预测指标。结果：乳腺癌组的雌二醇（E2）与睾酮（T）水平明显高于对照组（均P<0.05），其他性激素与血脂指标两组间差异均无统计学意义（均P>0.05）。分层分析显示，随E2 与T 水平升高，乳腺癌发病率增加，其中E2 水平对乳腺癌的预测有较好的敏感性和特异性，受试者操作特征（ROC）曲线下面积（AUC）为0.730（95% CI=0.649~0.811），但T 水平的预测作用不明显。结论：E2 与T 水平与绝经后女性乳腺癌有关，其中E2 水平可作为预测因子之一，但尚未达到良好或者优秀标准。
Relations of levels of endogenous sex hormones and blood lipids with breast cancer in postmenopausal women
Objective: To investigate the relations of the levels of endogenous sex hormones and blood lipids with breast cancer in postmenopausal women and the feasibility of using them as predictors for breast cancer. Methods: Two-hundred and seventy-four postmenopausal women with breast cancer (breast cancer group) undergoing surgical treatment from July 2011 to June 2014, and 279 postmenopausal female patients admitted during the same period for other diseases who were excluded from breast diseases (control group) were enrolled. The levels of sex hormone and blood lipids between the two groups were compared, and the potential predictors for breast cancer were analyzed. Results: The levels of estrogens (E2) and testosterone (T) in breast cancer group were significantly higher than those in control group (both P<0.05), while levels of other sex hormones and blood lipids showed no significant difference between the two groups (all P>0.05). Stratification analysis demonstrated that the incidence of breast cancer was increased with either E2 level or T level, in which E2 level had a better sensitivity and specificity for prediction of breast cancer, and its area under curve (AUC) of the receiver operating characteristic (ROC) reached 0.730 (95% CI=0.649–0.811), but T level showed no significance in prediction of breast cancer. Conclusion: The E2 and T levels are associated with breast cancer in postmenopausal women, and E2 level can be used as one of the predictors of breast cancer, but has not yet become a satisfactory or excellent standard.