0222 Greater Sleep Regularity Is Associated with Slower Pace of Epigenetic Aging

Abstract Introduction Sleep timing regularity is an important component of overall sleep health. People with more irregular sleep-wake schedules may more rapidly incur marks of biological aging, such as altered DNA methylation (DNAm). Here, we test whether sleep regularity index (SRI), a metric of the consistency of sleep-wake epochs, is associated with accelerated epigenetic aging in the Multi-Ethnic Study of Atherosclerosis (MESA). Methods SRI was calculated from scored sleep-wake epochs derived from actigraphy (wrist-worn ActiWatch Spectrum) collected over a week during the MESA Sleep Ancillary Study. Three epigenetic “clocks” representing markers of biological (Horvath DNAmAge 2018; developed by selecting DNAm sites associated with chronological age), phenotypic (Levine PhenoAge; developed by selecting DNAm sites associated with age-related biomarkers), and pace of aging (Belsky DunedinPACE; developed by selecting DNAm sites associated with individual trajectories of age-related biomarkers) were constructed from whole-blood DNAm data (EPIC Illumina Array) collected at MESA Exam 5, prior to actigraphy measurement. Associations between epigenetic age acceleration (outcome; the difference between chronological age and epigenetic age) and SRI (exposure, 10-unit increase) were evaluated with linear regression models adjusted for age, gender, study days between blood collection and actigraphy, self-reported race/ethnicity, study site, BMI, smoking status, depression score, and sleep duration. Results 431 MESA participants with valid actigraphy and DNAm data were included (mean chronological age=68 years, DNAm age=66 years, PhenoAge=56 years, pace of aging=1.01). Spearman correlations between chronological age and epigenetic aging measures were highest for biological aging (DNAmAge rho=0.89, PhenoAge rho=0.76, DunedinPACE rho=0.18). Results did not support an association between SRI and accelerated biological aging (DNAmAge; β=0.09, 95%CI:-0.23, 0.41) or phenotypic aging after additional adjustment for smoking (PhenoAge; β=-0.48, 95%CI:-1.0, 0.04). However, greater SRI was associated with slower pace of aging in fully adjusted models (10-unit SRI increase on DunedinPACE; β=-0.09, 95%CI:-0.02, -0.0004) and directionality was similar between PhenoAge and DunedinPACE. Conclusion Greater sleep regularity is associated with slower pace of aging in a cross-sectional analysis of older U.S. adults. Further work will assess epigenetic aging with other measures of sleep health. Support (if any) NIH-NHLBI K99HL166700, R35HL135818, R01HL098433, R01HL161012, Career Development Award from the Sleep Research Society Foundation.