Department of Epidemiology and Public Health, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan.
Department of Medicine and Clinical Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan.
J Clin Endocrinol Metab. 2019 Aug 1;104(8):3213-3222. doi: 10.1210/jc.2018-01782.
Although recent genetic studies have identified many susceptibility loci associated with type 2 diabetes (T2D), the usefulness of such loci for precision medicine remains uncertain.
This study investigated the impact of genetic risk score (GRS) on the development of T2D in a general Japanese population.
The current study consists of 1465 subjects aged 40 to 79 years without diabetes who underwent a health examination in 2002.
The GRS was generated using the literature-based effect size for T2D of 84 susceptibility loci for the Japanese population, and the risk estimates of GRS on the incidence of T2D were computed by using a Cox proportional hazard model in a 10-year follow-up study. The influence of GRS on the predictive ability was estimated with Harrell C statistics, integrated discrimination improvement (IDI), and continuous net reclassification improvement (cNRI).
During the 10-year follow-up, 199 subjects experienced T2D. The risk of developing T2D increased significantly with elevating quintiles of GRS (multivariable-adjusted hazard ratio for the fifth vs first quintile, 2.85; 95% CI, 1.83 to 4.44). When incorporating GRS into the multivariable model comprising environmental risk factors, the Harrell C statistics (95% CI) increased from 0.681 (0.645 to 0.717) to 0.707 (0.672 to 0.742) and the predictive ability of T2D was significantly improved (IDI, 0.0376; 95% CI, 0.0284 to 0.0494; cNRI, 0.3565; 95% CI, 0.1278 to 0.5829). GRS was also associated with the risk of T2D independently of environmental risk factors.
These findings suggest the usefulness of GRS for identifying a high-risk population together with environmental risk factors in the Japanese population.
尽管最近的遗传研究已经确定了许多与 2 型糖尿病（T2D）相关的易感基因座，但这些基因座在精准医学中的应用仍不确定。
本研究旨在调查在一般日本人群中，遗传风险评分（GRS）对 T2D 发病的影响。
本研究纳入了 2002 年接受健康检查的 1465 名年龄在 40 至 79 岁之间、无糖尿病的个体。
使用基于文献的日本人群 T2D 易感性 84 个基因座的效应大小生成 GRS，在 10 年随访研究中，使用 Cox 比例风险模型计算 GRS 对 T2D 发病风险的风险估计。通过 Harrell C 统计量、综合判别改善（IDI）和连续净重新分类改善（cNRI）评估 GRS 对预测能力的影响。
在 10 年随访期间，199 名受试者发生了 T2D。随着 GRS 五分位的升高，发生 T2D 的风险显著增加（第五五分位与第一五分位相比，多变量调整后的风险比，2.85；95%置信区间，1.83 至 4.44）。当将 GRS 纳入包含环境危险因素的多变量模型中时，Harrell C 统计量（95%置信区间）从 0.681（0.645 至 0.717）增加到 0.707（0.672 至 0.742），并且 T2D 的预测能力显著改善（IDI，0.0376；95%置信区间，0.0284 至 0.0494；cNRI，0.3565；95%置信区间，0.1278 至 0.5829）。GRS 与环境危险因素独立相关，也与 T2D 发病风险相关。
这些发现表明，GRS 结合环境危险因素有助于在日本人群中识别高危人群。
