Six Loci Support Shared Causal Variants

Colocalization analysis, which tests whether 2 traits share the same underlying variant, supported shared causal variants at 6 loci near APOE, HLA-B, NLGN1, AK5/DNAJB4, FAM114A2/GRIA1, and SATB1. The APOE signal showed near-complete colocalization (posterior probability, 0.9997). The APOE ε4 allele, which raises Alzheimer disease risk, was associated with lower insomnia risk and lower CRP. The investigators suggested this may reflect context-dependent roles in neuroimmune regulation, lipid metabolism, and glial biology.

Genes at the shared loci were enriched for immune and stress-response pathways, including interferon gamma response, tumor necrosis factor α signaling through NF-κB, and apoptosis. Chromatin organization and nucleosome assembly were also enriched. Enrichment appeared in several brain regions, including the hypothalamus, and in neuronal cell types in the midbrain and thalamus.

Mendelian Randomization Favors One Direction

Mendelian randomization, which uses inherited variants to test whether one trait influences another, showed that genetic liability to insomnia was associated with higher CRP (inverse variance weighted β = 0.299; P = 5.52 × 10⁻¹⁰). Weighted median, weighted mode, and simple mode estimates agreed, and there was no evidence of directional pleiotropy. One method, MR-Egger, did not reach significance, and heterogeneity across variants was present.

The reverse analysis was less consistent. CRP as the exposure produced a small positive estimate (β = 0.011; P = .0073), but other methods did not agree, and testing indicated directional pleiotropy. A sensitivity analysis using 23 variants from an independent 23andMe cohort, designed to address sample overlap, produced a significant inverse variance weighted estimate (P = .040). Other methods pointed in the same direction with less precision.

Limitations and Next Steps

Insomnia was self-reported, and residual pleiotropy and sample overlap cannot be fully excluded. Both datasets were limited to participants of European ancestry, and sex-stratified genetic analyses lacked 23andMe data and had reduced power. The study team called for longitudinal studies testing whether improved sleep lowers CRP and for functional work on the shared loci, particularly APOE.

“Targeting pathways involved in neuroimmune signaling, glial function, and synaptic regulation may offer strategies to mitigate the inflammatory consequences of sleep disruption,” the authors wrote. “Incorporating genetic information could further enable the development of individualized approaches aimed at improving sleep while reducing systemic inflammatory burden.”

References

Abou Daya F, Valliere J, Maher M, Saxena R. Genetic pleiotropy highlights immune mechanisms that link insomnia and C-reactive protein. Sleep. Published online 2026. doi:10.1093/sleep/zsag254Watanabe K, Jansen PR, Savage JE, et al. Genome-wide meta-analysis of insomnia prioritizes genes associated with metabolic and psychiatric pathways. Nat Genet. 2022;54(8):1125-1132. doi:10.1038/s41588-022-01124-w