第49届欧洲糖尿病研究协会年会(EASD2013)于9月23-27日在西班牙巴塞罗那召开。当地时间9月24日上午，在“Predictors of cardiovascular disease”专场上，英国邓迪大学Helen C Looker博士报道的一项研究证实，有6个生物标志物与2型糖尿病患者心血管疾病(CVD)的预测相关。
Looker博士在报告中说，“我们报告了我们认为的一个新发现，即白细胞介素-15可预测2型糖尿病CVD事件。此外，载脂蛋白C-III 和可溶性RAGE （sRAGE）与CVD事件也有关联，这个结果是出乎意料的，值得进一步去探讨。”
根据研究摘要，这项研究纳入了Go-DARTS（n=1,204）、斯堪尼亚糖尿病注册表组（n =666）、MONICA/KORA组（n=308）、IMPROVE （n=94）和 60岁斯德哥尔摩研究组（n=64）的2型糖尿病患者。
Looker博士和同事们检测了42个生物标志物，但只有16个标志物可进一步改善对心血管疾病的预测。
Looker博士说：“我们所测试的生物标志物许多与CVD事件有关，但在选择时，我们发现除了临床协变量，主要是6个标志物对预测有改善。”
与心血管疾病具有最显著关联的是：
• N -末端前B型利钠肽（ OR = 1.72 ，95％CI，1.49-1.99 ）;
• 载脂蛋白C-III （OR = 1.26 ，95％ CI , 1.10-1 .43）;
• sRAGE（OR = 0.82，95％ CI， 0.73-0.91）;
• 高敏
• IL -6 （OR = 1.16，95％ CI， 1.05-1.29 ）;
• IL -15 （ OR = 1.17，95％ CI， 1.06-1.29 ）。
“我们认为这些研究结果已经有力，因为多个研究都有发现，但我们还需要在其它队列中进行重复，这也是我们正在做的，”Looker博士说。
| Biomarkers for prediction of CVD in type 2 diabetes |
| Background and aims: Improving prediction of cardiovascular disease (CVD) in type 2 diabetes (T2D) is important for tailoring therapy and also for risk stratification into clinical trials. Here we examined a wide range of 42 potential serum biomarkers for incident CVD in a combined analysis of data from five European cohorts. Materials and methods: The study included T2D patients from 5 cohorts: Go-DARTS (n=1204), the Scania Diabetes Registry (n=666), MONICA/KORA (n=308), IMPROVE (n=94) and 60-year Old Stockholm Study (n=46). Baseline samples from incident cases of major CVD (MI or stroke) and an equal number of age and diabetes duration stratum-matched controls free of CVD at end of follow up were analysed. 9 markers were analysed by standard ELISAs or on automated Roche systems (Elecsys 2010 and Cobas Integra) and 33 were measured on a Luminex platform. Candidate biomarkers were chosen based on a literature mining search tool, previous reports of association with CVD in diabetic or non-diabetic cohorts or based on pathways reported as implicated in CVD. Data analysis was by logistic regression using 50-fold cross validation with forward selection and by 50-fold cross validated backward selection with LASSO penalty. Clinical covariates measured at baseline included in models were age, sex, diabetes duration, BMI, height, blood pressure, smoking, LDL, HDL, Triglycerides, HbA1c, eGFR, study centre, and medication (including antihypertensive, aspirin, lipid-lowering agent and insulin use). Results: Of 42 biomarkers analysed 16 were retained as improving the prediction of CVD beyond the clinical covariates. N-terminal pro B-type Natriuretic Peptide (NT-proBNP), Apolipoprotein CIII (ApoCIII), soluble RAGE (sRAGE), high sensitivity Troponin T (hsTnT), IL-6 and IL-15 were the most strongly associated with CVD (see table for ORs). However the increment in the area under the ROC curve was slight; from 0.67 for a model with clinical covariates only to 0.73 for a model including all clinical covariates and 16 associated biomarkers. Conclusion: We confirm the association of NT-proBNP and hsTnT with incident CVD in T2D. IL-6 is known to be associated with incident CVD in general populations and we confirm that finding now in T2D. We also report a novel association of IL-15 with incident CVD in T2D. The inverse associations between ApoCIII and sRAGE with incident CVD are unexpected and need further investigation. The improvement in prediction of CVD with these 16 biomarkers in this cross validation study indicates that validation of these markers is now warranted. |
点击进入>>EASD2013年会专题
我要投稿
