0122 Genetic Factors and Sleep Traits as Predictors of Depressive Symptoms: Insights from a Longitudinal Follow-Up

Abstract Introduction Insomnia and major depressive disorder (MDD) are highly comorbid conditions with a multifactorial etiology. This study aims to investigate the genetic factors predictive of depressive symptoms and their overall clinical associations - beyond depression itself - and explores the longitudinal relationship between insomnia severity and the trajectory of depressive symptoms over an 8-year period. Methods We calculated a polygenic score for depression (depression-PGS) in the São Paulo Epidemiologic Sleep Study (EPISONO) cohort, an epidemiological sample with deep phenotyping, representative of the City of São Paulo. A phenome-wide association study (PheWAS) was performed to explore the relationship between the depression-PGS and 463 diverse phenotypic traits. We assessed the association between depression-PGS with Pittsburg Seep Quality Index (PSQI) and Insomnia Severity Index (ISI) scores, in 2007 (baseline, N=1,042) and 2015 (follow-up, N=712). Multinomial regression models evaluated longitudinal trajectories of depressive symptoms (defined by Beck Depression Inventory-BDI in baseline and follow-up), and investigated the predictive role of PSQI and ISI, as continuous variables, on these trajectories. Results The depression-PGS was significantly associated with both continuous and categorical BDI scores in baseline. The PheWAS analysis revealed significant associations between the depression-PGS and 37 traits, predominantly related to neuropsychiatric and sleep phenotypes, including polysomnography variables (total sleep efficiency and latency, REM sleep duration and latency). When BDI status (cases vs. controls) was used as covariate in the PheWAS, higher depression-PGS remained associated with polysomnography-derived sleep efficiency. Higher depression-PGS was associated with poorer sleep quality (PSQI) and greater insomnia severity (ISI) in both time points. The degree of baseline insomnia severity and its change in overtime were predictors for elevation of depressive symptoms in the eight-year period. Conclusion Genetic predisposition for depression is associated with a wide range of sleep traits, including polysomnography variables and - for sleep efficiency - this link in independent from symptoms manifestation. The degree of insomnia severity in the baseline, as continuous variable, is predictive of the longitudinal course of depressive symptoms. The usage of insomnia symptoms and genetic vulnerability to predict MDD onset might enable the implementation of more targeted preventive and therapeutic interventions. Support (if any) AFIP, FAPESP, CNPq.