Causal associations of antioxidants with Alzheimer’s disease and cognitive function: a Mendelian randomisation study

Background Circulating antioxidants are associated with a lower risk of Alzheimer’s disease (AD) in observational studies, suggesting potential target areas for intervention. However, whether the associations are causal remains unclear. Here, we studied the causality between antioxidants and AD or cognitive function using two-sample Mendelian randomisation (MR). Methods Single nucleotide polymorphisms strongly (p<5×10−8) associated with antioxidants (vitamin A, vitamin C, zinc, selenium, β-carotene and urate) and outcomes (AD, cognitive performance and reaction time) were obtained from the largest and most recent genome-wide association studies (GWAS). MR inverse variance weighting (IVW) and MR pleiotropy residual sum and outlier test (MR-PRESSO) were used for data analysis. Results Higher genetically determined selenium level was associated with 5% higher risk of AD (OR 1.047, 95% CI 1.005 to 1.091, p=0.028) using IVW. Higher genetically determined urate level was associated with worse cognitive performance (β=−0.026, 95% CI −0.044 to −0.008, p=0.005) using MR-PRESSO. No association between the other antioxidants and AD, cognitive performance and reaction time was found. Similar results were found in the sensitivity analyses. Conclusion Our results suggest that lifelong exposure to higher selenium may be associated with a higher risk of AD, and higher urate levels could be associated with worse cognitive performance. Further analyses using larger GWAS of antioxidants are warranted to confirm these observations. Our results suggest that caution is needed in the interpretation of traditional observational evidence on the neuroprotective effects of antioxidants. Data are available in a public, open access repository. The data sets for MR analysis of this article are available from published genome-wide association studies: , , , , , , , , , .