To Assess Mediation Effects of BIN1 Risk Variant rs6733839 and APOE4 on Alzheimer’s disease Pathology

Abstract

AbstractBackgroundAlzheimer’s disease (AD) patients carrying T‐allele in the single nucleotide polymorphism rs744373 show higher levels of misfolded tau protein levels, as well as faster accumulation of neurofibrillary tangles (Franzmeier‐2019; Franzmeier‐2022), in addition to established risk gene apolipoprotein E (APOE4). Here we conduct pathway analysis to understand the direct and indirect effects, mediated through amyloid levels, of the rs6733839 T‐allele and APOE e4‐allele on tau.MethodAmyloid‐positive (florbetapir PET) placebo‐arm patients with a clinical diagnosis of AD dementia from four clinical trials (NCT01900665, NCT02791191, NCT02245737, NCT02016560, Table.1) underwent baseline (N = 547) and follow‐up (N = 230) flortaucipir PET scans. Tau burden was assessed using standardized uptake value ratio (SUVr) in AD‐specific region (MUBADA; Devous‐2018) with respect to cerebellum‐crus. We tested direct and indirect effects of BIN1 (rs6733839) and APOE4 on baseline SUVr and change‐from baseline (CFB) in tau‐PET mediated by baseline florbetapir signal. Unstandardized indirect effects were computed for each of 1,000 bootstrapped samples, and the 95% confidence interval was computed by determining the indirect effects at the 2.5th and 97.5th percentiles.ResultBIN1 and age showed significant association with the baseline tau‐PET level (BIN1: ßeta = 0.05, P = 0.05; Age: ßeta = ‐0.03, P<0.001), whereas APOE4 association with the baseline tau‐PET levels was not statistically significant (ßeta = 0.006, P = 0.81). The bootstrapped unstandardized indirect effect of BIN1, age, and APOE4 mediated via baseline amyloid levels was insignificant (BIN1: ßeta = 0.0002, P = 0.92; Age: ßeta = 3.82e−05, P = 0.74; APOE4: ßeta = ‐0.0003, P = 0.86). When evaluated on longitudinal tau measures, the direct effect of APOE4 on CFB in tau‐PET SUVr was significant (ßeta = 0.02; P = 0.01). The direct and indirect effects (mediated via baseline amyloid levels) of BIN1 (Direct: ßeta = ‐0.007, P = 0.215; Indirect: ßeta = ‐0.0007, P = 0.58) and Age (Direct: ßeta = ‐0.0009, P = 0.26; Indirect: ßeta = 0.0001, P = 0.32) were insignificant.ConclusionBIN1 showed significant association with the baseline tau‐PET level whereas APOE‐e4 allele directly affected longitudinal tau buildup independent of amyloid levels. These observations require additional validation using larger datasets and further optimization of the applied methodology is warranted.