Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, 181 Longwood Ave 3rd Floor, Boston, MA, 02115 USA.
Breast Cancer Res. 2012 Mar 13;14(2):R44. doi: 10.1186/bcr3141.
Previous research in the Nurses' Health Study (NHS) and the NHSII observed that, among women diagnosed with benign breast disease (BBD), those with predominant type 1/no type 3 lobules (a marker of complete involution) versus other lobule types were at lower risk of subsequent breast cancer. Studies in animal models suggest that insulin-like growth factor-1 (IGF-1) may inhibit involution of lobules in the breast; however, this has not been studied in humans.
We conducted a cross-sectional study among 472 women in the NHSII who were diagnosed with biopsy-confirmed proliferative BBD between 1991 and 2002 and provided blood samples between 1996 and 1999. A pathologist, blinded to exposure status, classified lobule type in normal adjacent tissue on available biopsy slides according to the number of acini per lobule. For each participant, the pathologist determined the predominant lobule type (that is, type 1, type 2, or type 3) and whether any type 1 or any type 3 lobules were present. Lobule type was then classified as: predominant type 1/no type 3 lobules, which is suggestive of complete involution; or other lobule types. Multivariate logistic models were used to assess the associations between plasma IGF-1, insulin-like growth factor binding protein-3 (IGFBP-3), and the ratio of IGF-1:IGFBP-3 levels with lobule type.
In univariate analyses, greater age, higher body mass index, postmenopausal status, nulliparity, and lower IGF-1 levels were associated with predominant type 1/no type 3 lobules (P < 0.05). In multivariate models adjusting for age and assay batch, higher IGF-1 levels were associated with decreased odds of predominant type 1/no type 3 lobules (odds ratio quartile 4 vs. quartile 1 = 0.37, 95% confidence interval = 0.15 to 0.89). Greater ratios of IGF-1:IGFBP-3 levels were also associated with decreased odds of predominant type 1/no type 3 lobules (odds ratio quartile 4 vs. quartile 1 = 0.26, 95% confidence interval = 0.11 to 0.64). These results were slightly attenuated after adjustment for other potential predictors of lobule type.
Higher IGF-1 levels and a greater IGF-1:IGFBP-3 ratio were associated with decreased odds of having predominant type 1 lobules/no type 3 lobules among women with proliferative BBD in the NHSII. This study provides further evidence for the role of insulin-like growth factors in the structure of breast lobules and lobular involution.
之前的护士健康研究（NHS）和 NHSII 的研究观察到，在被诊断患有良性乳腺疾病（BBD）的女性中，那些具有占主导地位的 1 型/无 3 型小叶（完全退化的标志物）的女性，与其他小叶类型相比，随后发生乳腺癌的风险较低。动物模型研究表明，胰岛素样生长因子-1（IGF-1）可能抑制乳腺小叶的退化；然而，这在人类中尚未得到研究。
我们在 NHSII 中进行了一项横断面研究，该研究纳入了 472 名在 1991 年至 2002 年间被确诊为活检证实的增生性 BBD 的女性，她们在 1996 年至 1999 年间提供了血液样本。一位病理学家在不知道暴露状况的情况下，根据每个小叶中的腺泡数量，对现有活检切片上的正常相邻组织中的小叶类型进行分类。对于每个参与者，病理学家确定了主要小叶类型（即 1 型、2 型或 3 型），以及是否存在任何 1 型或任何 3 型小叶。然后将小叶类型分类为：主要 1 型/无 3 型小叶，提示完全退化；或其他小叶类型。多变量逻辑模型用于评估血浆 IGF-1、胰岛素样生长因子结合蛋白-3（IGFBP-3）和 IGF-1:IGFBP-3 比值与小叶类型之间的关系。
在单变量分析中，年龄较大、体重指数较高、绝经后状态、未婚和 IGF-1 水平较低与主要 1 型/无 3 型小叶有关（P < 0.05）。在调整年龄和检测批次的多变量模型中，较高的 IGF-1 水平与较低的主要 1 型/无 3 型小叶的可能性相关（四分位间距 4 与 1 四分位间距=0.37，95%置信区间=0.15 至 0.89）。更高的 IGF-1:IGFBP-3 比值也与主要 1 型/无 3 型小叶的可能性降低相关（四分位间距 4 与 1 四分位间距=0.26，95%置信区间=0.11 至 0.64）。这些结果在调整其他潜在的小叶类型预测因素后略有减弱。
在 NHSII 中，患有增生性 BBD 的女性中，较高的 IGF-1 水平和 IGF-1:IGFBP-3 比值与具有主要 1 型小叶/无 3 型小叶的可能性降低相关。这项研究为胰岛素样生长因子在乳腺小叶结构和小叶退化中的作用提供了进一步的证据。
