A Mendelian randomization analysis relying on genome-wide association studies (GWASs) found no causal association between gout and Alzheimer disease, according to a report published in the International Journal of Rheumatic Diseases.
Given its known antioxidant effects, uric acid has been proposed as potentially neuroprotective, particularly against AD, in which oxidative stress has been implicated. Despite research suggesting an inverse causal relationship between gout and Alzheimer disease, conflicting evidence leaves this association unclear. In a first of its kind study, an investigator sought to explore the possibility that gout may directly lower Alzheimer disease risk.
The NHGRI-EBI GWAS catalogue was used to identify 3 gout exposure GWAS datasets (7236 cases and 24,325 controls) and 4 Alzheimer disease outcome GWAS meta-analysis datasets (17,008 cases and 37,154 controls) that were included in the analysis. There were 7 independent gout-associated single nucleotide polymorphisms (SNPs) used as instrumental variables in the collective exposure dataset, and these were analyzed for their association with the Alzheimer disease outcome dataset, before being combined to estimate potential causal relationships between gout and Alzheimer disease. The 7 SNPs (and their respective allelic loci) used as instrumental variables to enhance inference were: rs10791821 (MAP3K11), rs4073582 (CNIH-2), rs1260326 (GCKR), rs11722228 and rs734553 (SLC2A9), and rs2231142 and rs2728125 (ABCG2). The only SNP showing a significant positive correlation with Alzheimer disease was rs11722228.
A 2-sample Mendelian randomization used the weighted median, inverse-variance weighted and MR-Egger regression methods to assess the likelihood of a causal association. The standardized beta coefficient (𝛽) and standard error (SE) were calculated and reported for each methodology.
A gout-Alzheimer disease causal association was not supported by the weighted median approach results (𝛽 = 0.004; SE = 0.022; P =.846), nor by those derived using the inverse-variance weighted method (𝛽 = 0.013; SE = 0.017; P =.439). Likewise, MR-Egger regression did not reveal a causal relationship (𝛽 = -0.013; SE = 0.076; P =.870) and indicated that results were not biased by directional pleiotropy (intercept = 0.002; P =.654). In addition, there was no apparent heterogeneity between instrumental variable SNP estimates, according to Cochran Q statistic (P = .236-.445), and a “leave one out” analysis showed that the inverse-variance weighted point estimates were not driven by any single SNP.
Study limitations included a small SNP total that may limit power to detect a relationship, the modest effect of most genetic variants on exposures, possible “weak instrument bias,” and restriction to European ancestry for participants with Alzheimer disease.
“Our study findings did not indicate that the inverse associations between gout and Alzheimer disease are causal,” noted the authors. They recommended that future research continue to seek confirmation or refutation of causality between the 2 conditions, using well-designed studies that involve more variants.
Lee YH. Gout and the risk of Alzheimer′s disease: a Mendelian randomization study [published online March 28, 2019]. Int J Rheum Dis. doi:10.1111/1756-185x.13548