Although immune-mediated diseases (IMDs) and major depressive disorder (MDD) commonly co-occur, the bidirectional relationship between them remains to be fully elucidated. Using data from the prospective UK Biobank cohort, we evaluated the bidirectional associations by time-varying Cox proportional hazards regression models and assessed shared genetic architecture using genome-wide association study summary statistics. Additionally, we employed collagen-induced arthritis (CIA) and chronic social defeat stress (CSDS) mouse models to investigate the relationship between rheumatoid arthritis (RA) and depression. Over 5,226,841 person-years of follow-up, 23,534 incident MDD cases were identified. The presence of any IMD was associated with higher MDD risk (hazard ratio HR: 1.95; 95% CI: 1.89–2.01). Conversely, 59,742 incident cases of IMD were documented. MDD was associated with increased IMD risk (HR: 1.47; 95% CI: 1.40–1.54). We observed significant global genetic correlations between IMDs and MDD (r g : 0.11–0.49) and identified 128 pleiotropic genes, including ZKSCAN4 , BTN3A2 , and HSPA1L . Clinically, RA patients exhibited systemic inflammation and decreased levels of brain-derived neurotrophic factor. In experimental models, CIA mice showed depressive-like behaviors and lesions in brain regions implicated in depression. Moreover, superimposing CSDS on CIA exacerbated depressive-like behavior and pain sensitivity, accelerated the onset and progression of arthritis, and heightened joint inflammation. Collectively, these population, genetic, and experimental findings support a bidirectional association and shared genetic susceptibility between IMDs and MDD, highlighting immune-neurobiological pathways, particularly those involving inflammation and neurotrophin dysregulation, as candidates for mechanistic dissection and therapeutic targets.
Xu et al. (2026) studied this question.