Abstract
INTRODUCTION: Therapy-resistant depression is associated with higher levels of systemic inflammation and increased odds of metabolic disorders. It is, therefore, crucial to identify the biomarkers of high-risk individuals and understand the key features of depression-immune-metabolic networks. METHODS: The multiethnic ≥ 50-year-old study population is a subset of the Health and Aging Brain Study: Health Disparities (HABS-HD) study. Spearman's rank correlation network analysis was performed between immunological, metabolic, and subscales of the Geriatric Depression Scale (GDS). Significant correlations were then evaluated using a multivariable linear regression analysis, including testing for non-linearity and clinical cutoffs. RESULTS: Two clusters were formed: the first included the immune-metabolic biomarkers, and the second included the different subscales of GDS. The two clusters were significantly correlated at six edges. IL-6 and HbA1c were significantly correlated with anhedonic and melancholic features. Abdominal circumference and BMI were significantly correlated with anhedonic features. In the subgroup without current depression, IL-6 and Abdominal circumference maintained a significant edge with anhedonic features. The observed correlations remained statistically significant in the confounder-adjusted regression analysis and followed specific patterns. CONCLUSIONS: Symptom clustering showed its superiority over relying on dichotomized depression diagnoses for identifying relevant immunometabolic biomarkers. This study is a first step toward understanding the particularities of immunometabolic depression for better risk stratification and to direct personalized preventive and therapeutic strategies in multiethnic aging populations.