Type 2 diabetes and late-onset Alzheimer’s disease and related dementia: A longitudinal cohort study integrating polygenic risk score

Background The inherent genetic effects were not established between type 2 diabetes (T2DM) and Alzheimer's disease and related dementia (ADRD). Objective We aimed to investigate the association between T2DM and ADRD by integrating T2DM polygenic risk score (PRS) and applying matching in every subgroup. Methods We utilized UK Biobank First-occurrences datasets. T2DM were 1:1 matched to non-T2DM using propensity scores generated by 8 covariates; age at diagnosis, sex, cerebrovascular disease, ischemic heart disease, hypertensive disorders, lipid disorders, obesity, and mood disorders. T2DM PRS was additionally matched in T2DM PRS matched analysis. Subgroup analyses by age at diagnosis, sex, and APOE4 genotype were performed with the same matching criteria within each subgroup. Cox proportional hazard and Fine & Gray competing risk model were utilized. Results In T2DM PRS unmatched cohort, 24,583 T2DM were 1:1 matched to non-T2DM. The mean age at diagnosis was around 62 years old, with females constituting around 40%. Up to 25-year follow-up, ADRD rate/1000 person-years was 0.88 versus 1.52 (Non-T2DM versus T2DM); PRS unmatched (cHR: 1.72, 95% CI: 1.46–2.03) and matched (cHR:1.75, 95% CI: 1.47–2.09). Except for older age onset (≥75 years), the other subgroups demonstrated significantly increased ADRD risks in T2DM. T2DM PRS was higher in non-ADRD group across all subgroups. Contrarily, T2DM PRS was higher in ADRD in younger onset group (<55 years). Conclusions T2DM is one of the strong risk factors of ADRD but genetic T2DM effect does not contribute to ADRD risk. However, a genetic link might be present in younger age onset group.