BackgroundAlzheimer's disease (AD) is the leading cause of dementia worldwide, yet no disease-modifying therapy exists. Systemic inflammation has been proposed as a causal factor, supported by observational studies reporting elevated plasma interleukins in AD, though such studies cannot establish causality. Recent evidence instead highlights local neuroinflammation and microglial dysfunction as key mechanisms.ObjectiveThis study aimed to determine whether circulating interleukins and their receptors causally influence AD risk.MethodsWe conducted a two-sample Mendelian randomization (MR) analysis using genetic instruments to test whether genetically predicted levels of eleven circulating interleukins and receptor subunits influence AD risk. Three AD genome-wide association study (GWAS) datasets were analyzed: a primary cohort (n = 487,511) and two replication cohorts (n = 63,926; n = 55,134). Primary analyses used inverse-variance weighted models, with MR-Egger and weighted median as sensitivity methods. Additional analyses assessed heterogeneity, pleiotropy, directionality, and outlier influence. The study was preregistered and followed STROBE-MR guidelines.ResultsNo consistent evidence supported a causal effect of any circulating interleukin or receptor on AD risk. Nominal associations for IL-1 receptor antagonist (MR-Egger, p = 0.0054) and IL-6 receptor subunit alpha (weighted median, p = 0.014) did not survive false discovery rate correction and failed replication.ConclusionsThese findings do not support a causal role for circulating interleukins in AD risk. However, TNF-alpha and other non-interleukin mediators were not assessed, and stage-specific effects cannot be excluded due to limitations of summary-level GWAS data. Future studies including broader cytokine panels, disease-stage stratification, and individual-level analyses are warranted.
Ahmed et al. (2026) studied this question.