Epigenetic Aging and Rheumatoid Arthritis

Abstract

This is the first known comparative assessment of the associations of epigenetic age estimates with the prevalence of rheumatoid arthritis (RA). We used data available in Gene Expression Omnibus (GSE42861) from the Swedish Epidemiological Investigation of Rheumatoid Arthritis study. Information regarding RA diagnosis and 450K DNA methylation (DNAm) of 18- to 70-year-old participants was available. Utilizing Horvath's online DNAm Age Calculator, we determined the DNAm estimate of Telomere length (DNAmTL), Hannum's epigenetic age, Horvath's 2013 and 2018 epigenetic ages, PhenoAge, GrimAge, and the respective age-acceleration measures. The association of RA prevalence with epigenetic age measures was assessed using linear regression, adjusting for sex and smoking status. The p values were corrected for multiple testing using a false discovery rate. We identified statistically significant associations of RA with Horvath 2013 age acceleration (estimate: -1.34; FDR p value: 1.0 × 10-2), Horvath 2018 age acceleration (estimate: -1.32; FDR p value: 4.0 × 10-5), extrinsic age acceleration (estimate: 1.34; FDR p value: 1.0 × 10-2), PhenoAge acceleration (estimate: 2.31; FDR p value: 1.1 × 10-5), GrimAge (estimate: 2.54; FDR p value: 1.0 × 10-2), and GrimAge acceleration (estimate: 3.15; FDR p-value: 1.7 × 10-17). Of note, the raw and age-adjusted GrimAge surrogate DNAm protein components were significantly higher in RA cases than controls. Interestingly, the first-generation measures were associated only with women. No sex-specific effects were identified for PhenoAge or GrimAge accelerations. In this cross-sectional assessment, the second-generation clocks show promise as markers of biological aging, with higher epigenetic age acceleration observed in RA cases compared with healthy controls.

Keywords: Biological aging; GrimAge; Inflammation; PhenoAge.