Healthy sleep, proteomic signatures, and incident depression: associations accounting for genetic risk in an adult cohort

OBJECTIVES: We examined associations between healthy sleep patterns and incident depression, accounting for genetic susceptibility, identified related plasma proteins and their mediating role, and explored associations with brain structure. METHODS: Among 39,702 depression-free UK Biobank participants, five healthy sleep factors, plasma proteomic profiles, polygenic risk scores (PRS) for depression, brain imaging, and covariates were assessed. Linear and LASSO regression identified sleep-associated proteins and constructed proteomic signatures. Cox proportional hazards models examined associations of sleep patterns and proteomic signatures with incident depression, adjusting for genetic susceptibility. Mediation analysis and brain structure correlations were also assessed. RESULTS: During a median follow-up of 14.5 years, 1,941 participants developed depression. After adjustment for covariates and PRS, higher overall healthy sleep pattern scores showed a dose-dependent reduced risk of depression (HR per 1-score increment = 0.81; 95% CI: 0.77-0.85). We identified 496 proteins linked to healthy sleep patterns and derived a composite signature of 181 proteins. This signature was inversely associated with depression risk across PRS groups, and mediated 15.8% of the sleep-depression association. The proteomic signature correlated with emotion regulation brain regions. CONCLUSIONS: Healthy sleep patterns and their plasma proteomic signature are associated with reduced depression risk, highlighting targets for prevention and mechanistic research.