Background: Systemic inflammation has been suggested to be associated with Alzheimer's-disease progression, although whether it is a cause or a downstream effect is still controversial. This study aims to assess the effect of systemic inflammatory regulators on Alzheimer's disease within a bidirectional Mendelian-randomization design.
Methods: Genetic associations with Alzheimer's disease were obtained from the largest and most up-to-date genome-wide association study (GWAS) (cases and proxy cases: 71 880; controls: 383 378) and with inflammatory regulators from two recent GWASs on the human proteome and cytokines. Estimates were obtained by inverse-variance weighting with sensitivity analyses using MR-Egger, weighted median and MR-PRESSO. Possible bias due to selective survival and competing risk was also considered.
Results: None of 41 systemic inflammatory regulators was associated with risk of Alzheimer's disease with consistent results in validation analysis. Conversely, Alzheimer's disease was suggestively associated with five systemic inflammatory regulators, i.e. basic fibroblast growth factor, granulocyte-colony-stimulating factor, interferon gamma, interleukin-13 and interleukin-7.
Conclusion: The systemic inflammatory regulators considered did not appear to be associated with the risk of Alzheimer's disease. Conversely, specific systemic inflammatory regulators may be downstream effects of Alzheimer's disease or consequences of common factors causing both inflammation and Alzheimer's disease.
Keywords: Alzheimer’s disease; Cytokines; Mendelian randomization; inflammation.
© The Author(s) 2020; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association.