Associations of Accelerated Biological Aging With Dementia and the Mediation Role of Brain Structure: Findings From a Longitudinal Study.

BACKGROUND AND OBJECTIVES The association between biological aging and dementia, as well as the underlying mechanism, remains limited. The aim of this study was to investigate the relationships of biological age (BA) with incident dementia and examine the underlying neurobiological mechanisms. METHODS This study used data from the UK Biobank, a prospective longitudinal study. We included participants free of diagnosed dementia at baseline. BA was evaluated from clinical traits using the Klemera-Doubal method biological age (KDM-BA) and PhenoAge algorithms. Genetic risk of dementia was assessed using the apolipoprotein E (APOE) ε4 genotype and polygenic risk scores (PRSs). Cox proportional hazard regression models were used to estimate the associations of BA and the combined effect of genetic risk and BA with dementia. In addition, the potential roles of brain structures (gray matter volume [GMV], cortical mean thickness, and cortical surface area) in the associations between BA and dementia were evaluated using mediation analysis. RESULTS A total of 280,918 participants (mean age 56.80 years, 54.59% female) were enrolled in this study. Over a median follow-up of 13.58 years, 4,770 cases of dementia were recorded. Every SD increase in KDM-BA accelerations and PhenoAge accelerations was associated with a 14% (hazard ratio [HR] = 1.14; 95% CI 1.10-1.18) and 15% (HR = 1.15; 95% CI 1.12-1.19) higher incidence of dementia, respectively. Individuals with APOE ε4 and highest PhenoAge accelerations had the highest risk of dementia (HR = 4.20, 95% CI 3.69-4.78) compared with those with non-APOE ε4 and lowest PhenoAge accelerations, with significant interaction effect (Pinteraction < 0.001). We did not find significant modification effects of PRS on the associations between BA accelerations and dementia (Pinteraction = 0.347 for KDM-BA acceleration, Pinteraction = 0.279 for PhenoAge acceleration), as well as APOE ε4 on the association between KDM-BA accelerations and dementia (Pinteraction = 0.212). Mediation analysis showed that the identified GMV, cortical mean thickness, and cortical surface area partly mediated the association between BA accelerations and incident dementia, with proportion-mediated percentage ranging from 6.64% to 17.98%. DISCUSSION Advanced BA may be a potential risk factor of incident dementia. The risk is possibly mediated through the widespread reduction of brain structures.