Platelet count polygenic scores may modify the effect of aspirin in primary prevention of dementia.

INTRODUCTION: Aspirin is not recommended for dementia prevention due to limited benefit. We investigated whether genetic subgroups may benefit. METHODS: In the ASPirin in Reducing Events in the Elderly (ASPREE) trial (n = 13,541), we screened 5182 polygenic risk scores (PGSs) for aspirin interactions on dementia and major bleeding. After quality control, 1848 PGSs were analyzed using Cox models for aspirin×PGS interaction, adjusted for age, sex, apolipoprotein Estatus, and ancestry. Interactions were Bonferroni-corrected and examined by quintiles. RESULTS: Platelet-related PGS accounted for 18/112 nominally significant interactions (enrichment_OR = 54.7, p = 3.1 × 10- 1 8). Three highly correlated platelet-count PGSs passed multiple testing. For participants in the highest quintile of the platelet-count PGS003548, aspirin allocation was associated with a 3.5-fold reduction in incident dementia versus placebo (hazard ratio [HR]: 0.28, 95% confidence interval [CI]: 0.15 to 0.52). Major bleeding was also increased (HR: 2.13, 95% CI: 1.36 to 3.34). Similar interactions were observed using the highest PGS003548 tertile, decile, and 5%. DISCUSSION: Platelet count-related genetic variation may modify aspirin effects on dementia. These hypothesis-generating findings require validation.