Blostein Freida, Bakulski Kelly M, Fu Mingzhou, Wang Herong, Zawistowski Matthew, Ware Erin B
Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI, USA.
Department of Medical Informatics, University of California, Los Angeles, CA, USA.
J Alzheimers Dis. 2025 Jun;105(3):966-976. doi: 10.1177/13872877251333707. Epub 2025 May 4.
BackgroundDNA methylation clocks have emerged as promising biomarkers for cognitive impairment and dementia. Longitudinal studies exploring the association between DNA methylation clocks and cognitive decline have been constrained by limited sample sizes and a lack of diversity.ObjectiveOur study aimed to investigate associations between DNA methylation clocks and incident cognitive impairment using a larger and US nationally-representative sample from the Health and Retirement Study.MethodsWe measured DNA methylation age acceleration in 2016 by regressing the DNA methylation clocks, including GrimAge, against chronological age. Cognitive change over time was determined by Langa-Weir cognition status from 2016 to 2018. Multivariable logistic regression evaluated the association between DNA methylation age acceleration and cognitive change, adjusting for cell-type proportions, demographic, and health factors. We also applied inverse probability weighting to address potential selection bias from varying loss-to-follow-up rates.ResultsThe analytic sample (N = 2713) was 54% female, 8.4% Black/African American, 86% White, 7.5% Hispanic, and 68 years old at baseline. During the two years of follow-up, 12% experienced cognitive change and had higher baseline GrimAge (mean = 1.2 years) acceleration compared to those maintaining normal cognition (mean = -0.8 years). A one-year increase in GrimAge acceleration was associated with 1.05 times higher adjusted and survey-weighted odds of cognitive change during follow-up (95% CI: 1.01-1.10). This association was consistent after accounting for loss-to-follow-up (OR = 1.07, 95% CI: 1.04-1.11).ConclusionsOur study offers insights into DNA methylation age acceleration associated with cognitive change over time, suggesting avenues for improved prevention, diagnosis, and treatment.
背景
DNA甲基化时钟已成为认知障碍和痴呆症颇具前景的生物标志物。探索DNA甲基化时钟与认知衰退之间关联的纵向研究受到样本量有限和缺乏多样性的限制。
目的
我们的研究旨在使用来自健康与退休研究（Health and Retirement Study）的规模更大且具有美国全国代表性的样本，调查DNA甲基化时钟与新发认知障碍之间的关联。
方法
我们在2016年通过将包括GrimAge在内的DNA甲基化时钟与实际年龄进行回归分析，测量了DNA甲基化年龄加速情况。2016年至2018年期间的认知变化由兰加-韦尔认知状态确定。多变量逻辑回归评估了DNA甲基化年龄加速与认知变化之间的关联，并对细胞类型比例、人口统计学因素和健康因素进行了调整。我们还应用逆概率加权法来解决因随访失访率不同而可能产生的选择偏倚。
结果
分析样本（N = 2713）中，女性占54%，黑人/非裔美国人占8.4%，白人占86%，西班牙裔占7.5%，基线时的平均年龄为68岁。在两年的随访期间，12%的人经历了认知变化，与保持正常认知的人相比，他们的基线GrimAge加速更高（平均 = 1.2岁），而保持正常认知的人平均为-0.8岁。GrimAge加速每增加一岁与随访期间认知变化的校正和调查加权比值高1.05倍相关（95%CI：1.01 - 1.10）。在考虑随访失访情况后，这种关联仍然一致（OR = 1.07，95%CI：1.04 - 1.11）。
结论
我们的研究为随时间推移与认知变化相关的DNA甲基化年龄加速提供了见解，为改进预防、诊断和治疗方法指明了方向。
