Genetics
Reverse Mendelian Randomization of Alzheimer’s Disease Genetic Risk Score and Sensory impairments Minhyuk Choi* Minhyuk Choi Choi Choi Choi Choi Choi Choi Choi Choi Choi Choi University of California, San Francisco
Background: Identifying prodromal changes associated with incipient Alzheimer’s Disease (AD) is essential to guide prevention efforts. Sensory impairments have been hypothesized as early manifestations of AD pathophysiology. Using a reverse Mendelian Randomization (MR) design, we aimed to establish causal and temporal associations between AD genetic risk and time to sensory impairment diagnosis.
Methods: We used electronic health record data from the NIH All of Us Research Program (N=258,585). For each sensory impairment (vision impairment, cataract, glaucoma, age related macular degeneration (AMD), diabetic retinopathy, myopia, hearing loss, and severe hearing loss), outcome-free individuals were followed from 2014, until the earliest date of: sensory impairment diagnosis, death, or the data cut-off (Sep. 2023). We calculated a z-scored AD genetic risk score (AD-GRS) using 85 genetic instruments (with and without APOE-ε4). Cox models estimated the association between AD-GRS and time-to-first sensory impairment diagnosis. Positive controls (dementia, AD) and negative controls (pancreatic cancer, eczema) were included. Analyses were stratified by follow-up period: early-to-mid life (follow-up restricted to <65 years) and late-life (follow-up ≥65 years). Models adjusted for sex, race, baseline age, and genetic ancestry.
Results: In early-to-mid life, AMD (HR: 0.95 [0.91, 0.98]) and diabetic retinopathy (HR: 0.95 [0.92, 0.98]) were inversely associated with the AD-GRS with APOE-ε4 but not when excluding APOE-ε4 . In late-life, small positive associations were observed between AD-GRS (with APOE-ε4) and cataracts (HR: 1.02 [1.00, 1.04]) and blindness/vision impairment (HR: 1.05 [1.01, 1.09]). Associations with sensory impairments were minimal compared to the association with dementia (Figure 1).
Conclusion: Late-life analyses demonstrated small positive associations of AD genetic risk with cataracts and blindness, but overall null or inverse across other outcomes.
