Lifestyle Factors and Genetic Variants on 2 Biological Age Measures: Evidence From 94 443 Taiwan Biobank Participants

Abstract Background Biological age (BA) can be estimated by phenotypes and is useful for predicting life span and health span. Levine et al. proposed a PhenoAge and a BioAge to measure BA. Although there have been studies investigating the genetic predisposition to BA acceleration in Europeans, little has been known regarding this topic in Asians. Methods I have estimated PhenoAgeAccel (age-adjusted PhenoAge) and BioAgeAccel (age-adjusted BioAge) of 94 443 Taiwan Biobank (TWB) participants, wherein 25 460 TWB1 participants formed a discovery cohort and 68 983 TWB2 individuals constructed a replication cohort. Lifestyle factors and genetic variants associated with PhenoAgeAccel and BioAgeAccel were investigated through regression analysis and a genome-wide association study. Results A unit (kg/m2) increase of body mass index was associated with a 0.177-year PhenoAgeAccel (95% confidence interval [CI] = 0.163–0.191, p = 6.0 × 10–129) and 0.171-year BioAgeAccel (95% CI = 0.165–0.177, p = 0). Smokers on average had a 1.134-year PhenoAgeAccel (95% CI = 0.966–1.303, p = 1.3 × 10–39) compared with nonsmokers. Drinkers on average had a 0.640-year PhenoAgeAccel (95% CI = 0.433–0.847, p = 1.3 × 10–9) and 0.193-year BioAgeAccel (95% CI = 0.107–0.279, p = 1.1 × 10–5) relative to nondrinkers. A total of 11 and 4 single-nucleotide polymorphisms (SNPs) were associated with PhenoAgeAccel and BioAgeAccel (p < 5 × 10–8 in both TWB1 and TWB2), respectively. Conclusions A PhenoAgeAccel-associated SNP (rs1260326 in GCKR) and 2 BioAgeAccel-associated SNPs (rs7412 in APOE; rs16998073 near FGF5) were consistent with the finding from the UK Biobank. The lifestyle analysis shows that prevention from obesity, cigarette smoking, and alcohol consumption is associated with a slower rate of biological aging.