Height and intracranial volume as protective factors against Alzheimer’s disease: a Mendelian randomization analysis

Abstract

AbstractBackgroundMeta analyses have identified height as a risk factor inversely associated with Alzheimer’s disease. Height as a protective factor may be reflective of overlapping genetics with intracranial volume (ICV), which has also been shown to be inversely associated with AD, or height may reflect early life influences such as nutritional intake. Here we use Mendelian randomization (MR) analysis to further investigate these causal relationships.MethodTwo sample MR was used to obtain causal estimates of height and ICV on AD, using GWAS summary statistics from the GIANT consortium (n = 5,380,080), meta analyzed summary statistics from studies measuring ICV, brain volume, and head circumference, and the International Genomics of Alzheimer’s Project (n = 94,437). Independent (r2<0.001, 10mb window) genome‐wide significant single nucleotide polymorphisms (p<5×10−8) were included as genetic instruments (n = 960). MR methods included inverse variance weighted (IVW), weighted median, weighted mode, and MR Egger. Cochran’s Q test was used to test for heterogeneity; the MR‐Egger intercept for pleiotropy; and Radial MR to identify individual variants that were outliers.ResultGenetically predicted height was associated with a reduced risk of AD (IVW logOR [95% CI]: ‐0.11 [‐0.16, ‐0.06], p = 7.3e−5). There was, however, evidence of heterogeneity and pleiotropy. After outlier removal (n = 1), height remained significantly associated with reduced risk of AD (IVW OR [95% CI]: ‐0.12 [‐0.18, ‐0.07], p = 1.49e−5), with sensitivity analyses showing the same direction of effect. Genetically predicted ICV was significantly associated with AD (IVW logOR [95% CI]: 0.16 [0.003, 0.31], p = 0.046), though tests of heterogeneity and pleiotropy were significant while sensitivity analyses were not. After outlier removal (n = 1), ICV was no longer significantly associated with reduced risk of AD (IVW logOR [95% CI]: 0.13 [‐0.03, 0.30], p = 0.11).ConclusionGenetically predicted height was protective against AD, while ICV had a suggestive protective causal effect. Together, these findings may imply a link between these phenotypes and AD. Further research using multivariate MR is needed to fully evaluate the potential mediating role of ICV in the protective effect of height on AD, as well as to explore other potential mediators such as nutritional intake.