Genetically proxied lean mass and risk of Alzheimer's disease: mendelian randomisation study.

OBJECTIVE

To examine whether genetically proxied lean mass is associated with risk of Alzheimer's disease.

DESIGN

Mendelian randomisation study.

SETTING

The UK Biobank study and genome wide association study meta-analyses of Alzheimer's disease and cognitive performance.

PARTICIPANTS

Summary level genetic data from: 450 243 UK Biobank participants with impedance measures of lean mass and fat mass; an independent sample of 21 982 patients with Alzheimer's disease and 41 944 controls without Alzheimer's disease; a replication sample of 7329 patients with Alzheimer's disease and 252 879 controls; and 269 867 individuals taking part in a genome wide association study of cognitive performance.

MAIN OUTCOME MEASURE

Effect of genetically proxied lean mass on the risk of Alzheimer's disease, and the related phenotype of cognitive performance.

RESULTS

An increase in genetically proxied appendicular lean mass of one standard deviation was associated with a 12% reduced risk of Alzheimer's disease (odds ratio 0.88, 95% confidence interval 0.82 to 0.95, P=0.001). This finding was replicated in an independent cohort of patients with Alzheimer's disease (0.91, 0.83 to 0.99, P=0.02) and was consistent in sensitivity analyses that are more robust to the inclusion of pleiotropic variants. Higher genetically proxied appendicular lean mass was also associated with increased cognitive performance (standard deviation increase in cognitive performance for each standard deviation increase in appendicular lean mass 0.09, 95% confidence interval 0.06 to 0.11, P=0.001), and adjusting for potential mediation through genetically proxied cognitive performance did not reduce the association between appendicular lean mass and risk of Alzheimer's disease. Similar results were found for the outcomes of Alzheimer's disease and cognitive performance when the risk factors of genetically proxied trunk lean mass and whole body lean mass were used, respectively, adjusted for genetically proxied fat mass.

CONCLUSIONS

These findings suggest that lean mass might be a possible modifiable protective factor for Alzheimer's disease. The mechanisms underlying this finding, as well as the clinical and public health implications, warrant further investigation.