Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI, USA.
Memory Impairment and Neurodegenerative Dementia (MIND) Center, University of Mississippi Medical Center, Jackson, MS, USA.
Clin Epigenetics. 2021 Mar 16;13(1):55. doi: 10.1186/s13148-021-01035-3.
Cardiovascular disease (CVD) is the leading cause of mortality among US adults. African Americans have higher burden of CVD morbidity and mortality compared to any other racial group. Identifying biomarkers for clinical risk prediction of CVD offers an opportunity for precision prevention and earlier intervention.
Using linear mixed models, we investigated the cross-sectional association between four measures of epigenetic age acceleration (intrinsic (IEAA), extrinsic (EEAA), PhenoAge (PhenoAA), and GrimAge (GrimAA)) and ten cardiometabolic markers of hypertension, insulin resistance, and dyslipidemia in 1,100 primarily hypertensive African Americans from sibships in the Genetic Epidemiology Network of Arteriopathy (GENOA). We then assessed the association between epigenetic age acceleration and time to self-reported incident CVD using frailty hazard models and investigated CVD risk prediction improvement compared to models with clinical risk scores (Framingham risk score (FRS) and the atherosclerotic cardiovascular disease (ASCVD) risk equation). After adjusting for sex and chronological age, increased epigenetic age acceleration was associated with higher systolic blood pressure (IEAA), higher pulse pressure (EEAA and GrimAA), higher fasting glucose (PhenoAA and GrimAA), higher fasting insulin (EEAA), lower low density cholesterol (GrimAA), and higher triglycerides (GrimAA). A five-year increase in GrimAA was associated with CVD incidence with a hazard ratio of 1.54 (95% CI 1.22-2.01) and remained significant after adjusting for CVD risk factors. The addition of GrimAA to risk score models improved model fit using likelihood ratio tests (P = 0.013 for FRS and P = 0.008 for ASCVD), but did not improve C statistics (P > 0.05). Net reclassification index (NRI) showed small but significant improvement in reassignment of risk categories with the addition of GrimAA to FRS (NRI: 0.055, 95% CI 0.040-0.071) and the ASCVD equation (NRI: 0.029, 95% CI 0.006-0.064).
Epigenetic age acceleration measures are associated with traditional CVD risk factors in an African-American cohort with a high prevalence of hypertension. GrimAA was associated with CVD incidence and slightly improved prediction of CVD events over clinical risk scores.
心血管疾病（CVD）是美国成年人死亡的主要原因。与其他任何种族群体相比，非裔美国人患 CVD 的发病率和死亡率更高。寻找 CVD 临床风险预测的生物标志物为精准预防和早期干预提供了机会。
我们使用线性混合模型，在 1100 名主要为高血压的非裔美国人中，调查了四种表观遗传年龄加速（内在（IEAA）、外在（EEAA）、PhenoAge（PhenoAA）和 GrimAge（GrimAA））与十种高血压、胰岛素抵抗和血脂异常的心血管代谢标志物之间的横断面关联，这些人来自遗传流行病学网络动脉粥样硬化（GENOA）的同胞。然后，我们使用脆弱性风险模型评估了表观遗传年龄加速与自我报告的 CVD 事件时间之间的关联，并与临床风险评分（Framingham 风险评分（FRS）和动脉粥样硬化性心血管疾病（ASCVD）风险方程）进行了比较，以评估其对 CVD 风险预测的改善。在调整了性别和年龄后，表观遗传年龄加速与收缩压升高（IEAA）、脉压升高（EEAA 和 GrimAA）、空腹血糖升高（PhenoAA 和 GrimAA）、空腹胰岛素升高（EEAA）、低密度胆固醇降低（GrimAA）和甘油三酯升高（GrimAA）有关。GrimAA 每增加 5 年，CVD 发病的风险比为 1.54（95%CI 1.22-2.01），在调整 CVD 危险因素后仍有统计学意义。似然比检验表明，将 GrimAA 添加到风险评分模型中可以改善模型拟合（P=0.013 用于 FRS，P=0.008 用于 ASCVD），但 C 统计量无显著改善（P>0.05）。净重新分类指数（NRI）显示，在将 GrimAA 添加到 FRS（NRI：0.055，95%CI 0.040-0.071）和 ASCVD 方程（NRI：0.029，95%CI 0.006-0.064）时，风险分类的重新分配有较小但有统计学意义的改善。
在高血压患病率较高的非裔美国人群中，表观遗传年龄加速与传统 CVD 危险因素相关。GrimAA 与 CVD 发病相关，且略微提高了对临床风险评分的 CVD 事件预测。
