Joslin Diabetes Center, One Joslin Place, Boston, MA 02215, USA.
Clin Chem. 2011 Feb;57(2):326-37. doi: 10.1373/clinchem.2010.156133. Epub 2010 Dec 13.
Biomarkers for estimating reduced glucose tolerance, insulin sensitivity, or impaired insulin secretion would be clinically useful, since these physiologic measures are important in the pathogenesis of type 2 diabetes mellitus.
We conducted a cross-sectional study in which 94 individuals, of whom 84 had 1 or more risk factors and 10 had no known risk factors for diabetes, underwent oral glucose tolerance testing. We measured 34 protein biomarkers associated with diabetes risk in 250-μL fasting serum samples. We applied multiple regression selection techniques to identify the most informative biomarkers and develop multivariate models to estimate glucose tolerance, insulin sensitivity, and insulin secretion. The ability of the glucose tolerance model to discriminate between diabetic individuals and those with impaired or normal glucose tolerance was evaluated by area under the ROC curve (AUC) analysis.
Of the at-risk participants, 25 (30%) were found to have impaired glucose tolerance, and 11 (13%) diabetes. Using molecular counting technology, we assessed multiple biomarkers with high accuracy in small volume samples. Multivariate biomarker models derived from fasting samples correlated strongly with 2-h postload glucose tolerance (R(2) = 0.45, P < 0.0001), composite insulin sensitivity index (R(2) = 0.91, P < 0.0001), and insulin secretion (R(2) = 0.45, P < 0.0001). Additionally, the glucose tolerance model provided strong discrimination between diabetes vs impaired or normal glucose tolerance (AUC 0.89) and between diabetes and impaired glucose tolerance vs normal tolerance (AUC 0.78).
Biomarkers in fasting blood samples may be useful in estimating glucose tolerance, insulin sensitivity, and insulin secretion.
能够评估葡萄糖耐量降低、胰岛素敏感性或胰岛素分泌受损的生物标志物在临床上可能会很有用，因为这些生理指标在 2 型糖尿病的发病机制中很重要。
我们进行了一项横断面研究，纳入了 94 名参与者，其中 84 名参与者有 1 种或多种糖尿病风险因素，10 名参与者无已知的糖尿病风险因素。所有参与者均接受了口服葡萄糖耐量试验。我们测量了 250μL 空腹血清样本中与糖尿病风险相关的 34 种蛋白质生物标志物。我们应用多元回归选择技术来识别最具信息量的生物标志物，并建立多变量模型来估计葡萄糖耐量、胰岛素敏感性和胰岛素分泌。通过 ROC 曲线（AUC）分析评估葡萄糖耐量模型区分糖尿病患者与葡萄糖耐量受损或正常患者的能力。
在有风险的参与者中，有 25 名（30%）参与者被发现葡萄糖耐量受损，有 11 名（13%）参与者患有糖尿病。我们使用分子计数技术在小体积样本中以高精度评估了多种生物标志物。从空腹样本中得出的多变量生物标志物模型与 2 小时餐后葡萄糖耐量高度相关（R²=0.45，P＜0.0001）、复合胰岛素敏感性指数（R²=0.91，P＜0.0001）和胰岛素分泌（R²=0.45，P＜0.0001）。此外，葡萄糖耐量模型在区分糖尿病与葡萄糖耐量受损或正常之间、糖尿病与葡萄糖耐量受损与正常之间提供了很好的判别能力（AUC 分别为 0.89 和 0.78）。
空腹血样中的生物标志物可能有助于评估葡萄糖耐量、胰岛素敏感性和胰岛素分泌。
