Corresponding author: Lamiae Grimaldi-Bensouda,
Diabetes Care. 2014;37(1):134-43. doi: 10.2337/dc13-0695. Epub 2013 Aug 15.
OBJECTIVE Several studies have been published in 2009 suggesting a possible association between insulin glargine and increased risk of malignancies, including breast cancer. The objective of this study was to assess the relation between the individual insulins (glargine, aspart, lispro, and human insulin) and development of breast cancer. RESEARCH DESIGN AND METHODS Seven hundred seventy-five incident cases of primary invasive or in situ carcinoma breast cancer occurring in women with diabetes from 92 centers in the U.K., Canada, and France were matched to a mean of 3.9 diabetic community control subjects (n = 3,050; recruited from 580 general practices) by country, age, recruitment date, and diabetes type and management. The main risk model was a multivariate conditional logistic regression model with case/control status as the dependent variable and individual insulin use, 8 years preceding the index date, as the independent variable, controlling for past use of any insulin, oral antidiabetes drugs, reproductive factors, lifestyle, education, hormone replacement therapy and history of contraceptive use, BMI, comorbidities, diabetes duration, and annual number of physician visits. Glargine was also compared with every other insulin by computing all ratios using the variance-covariance matrix of logistic model parameters. RESULTS Adjusted odds ratios of breast cancer for each type of insulin versus no use of that insulin were 1.04 (95% CI 0.76-1.44) for glargine, 1.23 (0.79-1.92) for lispro, 0.95 (0.64-1.40) for aspart, and 0.81 (0.55-1.20) for human insulin. Two-by-two comparisons found no difference between glargine and the different types of insulins. Insulin dosage or duration of use and tumor stage did not change the results. CONCLUSIONS This international study found no difference in the risk of developing breast cancer in patients with diabetes among the different types of insulin with short- to mid-term duration of use. Longer-term studies would be of interest.
2009 年发表的多项研究表明,甘精胰岛素与包括乳腺癌在内的恶性肿瘤风险增加之间可能存在关联。本研究的目的是评估各种胰岛素(甘精胰岛素、门冬胰岛素、赖脯胰岛素和人胰岛素)与乳腺癌发展之间的关系。
本研究共纳入来自英国、加拿大和法国 92 个中心的 775 例新发原发性浸润性或原位乳腺癌女性病例,以及每位病例平均 3.9 名糖尿病社区对照者(从 580 个普通诊所招募)。根据国家、年龄、入组日期、糖尿病类型和管理方式进行匹配。主要风险模型是多变量条件逻辑回归模型,病例/对照状态为因变量,索引日期前 8 年的个体胰岛素使用情况为自变量,同时还控制了过去任何胰岛素、口服抗糖尿病药物、生殖因素、生活方式、教育、激素替代疗法和避孕药使用史、体重指数、合并症、糖尿病病程和每年就诊次数的影响。甘精胰岛素还与其他胰岛素进行了比较,计算了所有比值,使用了逻辑模型参数的方差-协方差矩阵。
与未使用该胰岛素相比,每种胰岛素的乳腺癌调整后比值比分别为甘精胰岛素 1.04(95%CI 0.76-1.44)、赖脯胰岛素 1.23(0.79-1.92)、门冬胰岛素 0.95(0.64-1.40)和人胰岛素 0.81(0.55-1.20)。两两比较发现甘精胰岛素与其他胰岛素类型之间无差异。胰岛素剂量或使用时间和肿瘤分期均未改变结果。
本国际研究发现,在短期至中期使用期间,不同类型的胰岛素在糖尿病患者中发生乳腺癌的风险无差异。长期研究将更有意义。