第一部分：负荷后1小时血浆血糖与糖尿病及其并发症发生的关系及对2型糖尿病发生风险预测能力的研究

中文摘要

目的：

分析口服葡萄糖耐量试验负荷后1小时血浆血糖（1hPG）水平升高人群进展为2型糖尿病及发生远期全因死亡和相关大、小血管并发症的风险及1hPG对未来糖尿病发生的预测能力。

方法：

纳入1986年原大庆研究中经75g口服葡萄糖耐量试验（OGTT）确定的共1026名非糖尿病受试者。纳入本研究的受试者根据基线OGTT后1hPG水平分为2个亚组，定义为研究 1hPG正常组（1hPG<8.6mmol/L）和1hPG升高组（1hPG≥8.6mmol/L）。对受试者进行了6年、20年和30年的随访，随访的终点事件包括2型糖尿病、全因死亡、心血管死亡、心血管事件(CVD)和微血管事件等。分析纳入的非糖尿病人群的基线1hPG与30年随访期间2型糖尿病及其相关并发症发生的关系。通过生存曲线分析统计累积事件发生率，并使用log-rank检验比较累积事件发生率。采用Cox回归分析评估1hPG与终点事件发生风险的关系。使用时间相依受试者工作特征曲线(Time-dependent ROC)分析方法，评估1hPG和2小时血浆血糖（2hPG）在6年、10年、20年和30年对2型糖尿病发病率的预测效能。ROC曲线下的面积(AUC)用于衡量连续变量在每个时间点预测2型糖尿病发生的效果。

结果：

在经过30年的随访后有578例（1hPG正常组154例，1hPG升高组424例）受试者进展为2型糖尿病。在调整年龄、性别、吸烟、收缩压、总胆固醇、体质指数（BMI）和生活方式干预等混杂因素后进行Cox回归分析，结果显示与 1hPG 正常组相比，1hPG 升高组的糖尿病风险增加（风险比 (HR)：4.45，95% CI：3.43-5.79）。糖尿病相关并发症包括全因死亡（HR：1.46，95% CI：1.07-2.01）、心血管死亡（HR：1.84，95% CI：1.16-2.95）、心血管事件（Cardiovascular disease, CVD）（HR：1.39，95% CI：1.03-1.86）和微血管事件（HR：1.70，95% CI：1.03-2.79）的发生风险均显著增加。在长期随访中，1hPG 预测糖尿病发生的AUC值高于 2hPG（AUC （1hPG vs 2hPG）：10年为0.86 vs 0.84，p=0.08；20 年为0.88 vs 0.87，p=0.04；30年为0.85 vs 0.82，p=0.009）。

结论：

非糖尿病1hPG升高(≥8.6mmol/L)人群中，与1hPG正常人群相比，其发生2型糖尿病及其长期并发症的风险增加相关。此外，1hPG可有效预测未来糖尿病发生，可被考虑为识别和筛查2型糖尿病高危个体的适宜指标。

第二部分：负荷后1小时血浆血糖水平升高人群的全基因组关联分析

中文摘要

目的：

   我们第一部分研究以及其他类似研究发现非糖尿病人群的负荷后1小时血浆血糖（1 hour plasma glucose，1hPG）进展为糖尿病及糖尿病相关并发症的风险升高，然而其发生的机制是否与基因相关未见报道。本研究旨在通过全基因组关联分析（Genome-wide Association study，GWAS）探索负荷后1小时血浆血糖升高人群单核苷酸多态性（single nucleotide polymorphism, SNP)以初步探究非糖尿病1hPG水平升高人群未来2型糖尿病发生风险升高的潜在分子机制和生物通路。

方法：

纳入原大庆糖尿病预防研究中采集且留存完好的513例非糖尿病受试者（包括NGT和IGT人群）的生物标本提取DNA样本后进行全基因组SNP位点检测。根据受试者基线1hPG水平分为1hPG正常组（1hPG<8.6mmol/L)和1hPG升高组（1hPG≥8.6mmol/L）。从受试者外周血细胞中采集和提取本研究中个体的基因组DNA后研究采用lllumina公司出品的ASAMD芯片对样本DNA进行全基因组分型。使用 Plink v1.9 软件对获得的SNP数据进行质量控制和表型关联分析，SNP位点达到显著性水平p < 10^-4为候选SNP位点并进一步将SNP位点定位注释到相应候选基因进行基因功能注释和通路分析。

结果：

本研究纳入了513名原大庆研究中非糖尿病患者（包括正常人和IGT人群），其中根据基线OGTT 1hPG分为1hPG升高组（n=220）和1hPG正常组（n=293），研究人群总体平均年龄为41.48±7.66岁，女性占比55.06%，平均FPG为5.03±0.7mmol/L。经过严格的数据质量控制后，共有513例DNA样本完成芯片检测，495567个SNP位点进入GWAS表型相关性分析。通过卡方检验，获得了55个达到显著性水平的 SNP位点，CHR1包含最多与1hPG升高人群相关的候选标记。此外，包含候选 SNP 的染色体都含有多个候选SNP。达到显著性水平p<10^-4的SNP位点中有34 个 SNP成功定位于26个基因。对这26个基因的 GO （Gene Ontology）分析中未有任何GO分类条目的调整P值达到显著性水平。KEGG（Kyoto Encyclopedia of Genes and Genomes）通路分析显示，这些SNP聚类到49条代谢通路，其中与内分泌系统相关的通路或生物过程共有6条，与内分泌与代谢相关的有3条，与碳水化合物代谢相关的有1条，与氨基酸代谢相关的有1条。通路相关基因包括SOCS3、PRDM2、UCP1以及GNG2。在校正p值后，统计学差异不显著（p>0.05）。

结论：

综上所述，我们利用大庆研究保存完好的生物样本，在1hPG升高人群中发55个达到显著性水平（p < 10^-4）的SNP位点。此外，根据得到的SNP位点定位到4个与1hPG升高可能有潜在的关联的基因(SOCS3、PRDM2、UCP1以及GNG2)。本研究发现了一些新的对糖尿病发生发展可能有潜在作用的基因，需要在更大的样本中进一步研究，以提高我们对1hPG升高人群未来2型糖尿病发生风险升高的潜在遗传机制的理解。

Efficacy of one-hour postload plasma glucose as a suitable measurement in predicting type 2 diabetes and diabetes-related complications

Abstract

Objective:

 We aim to evaluate whether 1-hour plasma glucose (1hPG) can be a comparable measurement to 2-hour plasma glucose (2hPG) in identifying high risk individuals of developing diabetes.

Methods:

1026 non-diabetic subjects in the Da Qing IGT and Diabetes Study were included and classified according to the baseline postload 1hPG during the oral glucose tolerance test (OGTT). The participants were stratified into subgroups by their baseline 1hPG ≥ 8.6mmol/L (1hPG-high group) and 1hPG<8.6mmol/L (1hPG-normal group) and were followed at 6-, 20-, 30-year for event outcomes including type 2 diabetes, all-cause and cardiovascular mortality, cardiovascular disease (CVD) events and microvascular disease. We then conducted a Cox proportional hazard analysis in this post-hoc study to determine the risks of developing type 2 diabetes and its complications in 1hPG-normal group and 1hPG-high group. Cumulative incidences were calculated by Kaplan-Meier analysis and log-rank test were used to compare cumulative incidence between groups. Predictive value of 1hPG and 2hPG were evaluated by the time-dependent receiver operator characteristics (ROC) curve.

Results:

Overall, 578 study participants developed type 2 diabetes after 30-year follow-up. Compared with 1hPG-normal group, 1hPG-high group had increased risk of diabetes (Hazard Ratio (HR): 4.45, 95% CI: 3.43-5.79), all-cause mortality (HR: 1.46, 95% CI: 1.07-2.01), CVD mortality (HR: 1.84, 95% CI: 1.16-2.95), CVD events (HR: 1.39, 95% CI: 1.03-1.86) and microvascular disease (HR: 1.70, 95% CI: 1.03-2.79) after adjusting confounders. 1hPG exhibited higher area under the ROC curve (AUC) for predicting diabetes than 2hPG during the long-term follow-up (AUC（1hPG vs 2hPG）: 10-year: 0.86 vs 0.84, p=0.08; 20-year: 0.88 vs 0.87, p=0.04; 30-year: 0.85 vs 0.82, p=0.009).

Conclusion:

Elevated 1hPG level (≥ 8.6mmol/L) is associated with increased risk of developing type 2 diabetes and its long-term complications and could be considered as a suitable measurement in identifying high risk individuals of type 2 diabetes.

Genome-wide association analysis of p ovulation with elevated 1 hour plasma glucose

Abstract

Objective

To investigate single nucleotide polymorphism (SNP) in a population with elevated 1-hour plasma glucose to explore the genetic and molecular pathways associated with 1-hour glucose levels and the development of diabetes by genome-wide association study (GWAS).

Methods

Genome-wide SNP loci were examined in a sample of 514 NGT and IGT biological specimens from the former Daqing Diabetes Prevention Study. Subjects were divided into 1hPG normal group (1hPG<8.6mmol/) and 1hPG elevated group (1hPG≥8.6mmol/) according to their baseline 1hPG levels.

Results

A total of 513 DNA samples and 495,567 SNP loci were included into the GWAS phenotypic correlation analysis after rigorous data quality control. By chi-square test, 55 SNP loci that reached the suggestive significance level （p<10^-4） were obtained, and CHR1 contained the large of candidate markers associated with the population with elevated 1hPG. In addition, all of these chromosomes contain multiple candidate SNPs. 34 of the SNP loci that reached suggestive significance were successfully located in 26 genes, for which the GO analysis did not reach significance for any term after adjusting p value. Although KEGG pathway analyses were also statistically insignificant after correction, a total of six of the 49 metabolic pathways or processes clustered to were related to the endocrine system, three to endocrine and metabolism, one to carbohydrate metabolism, and one to amino acid metabolism. The locus genes associated with these pathways include four genes SOCS3, PRDM2, UCP1, and GNG2.   

Conclusion

In summary, we identified 55 SNP loci associated with 1hPG elevated people in the Chinese population. In addition, four genes (SOCS3, PRDM2, UCP1, and GNG2) may have a potential association with elevated 1hPG. This study has identified new genes that may have a potential role in the development of diabetes and need to be further investigated in larger samples to improve our understanding of the genetic basis of diabetes development.