Abstract
BACKGROUND: Major depressive disorder (MDD) is often comorbid with metabolic diseases such as obesity, cardiovascular disease, and type 2 diabetes. A potential link between these disorders is adiponectin, an adipocyte-derived circulating hormone with insulin-sensitizing, anti-inflammatory, and neuroplasticity effects. Reductions in plasma levels of adiponectin have been reported in both humans with depression and in the chronic-defeat mouse model of depression. However, the predictive value of adiponectin for treatment response to depression has not been determined. METHODS: We investigated the potential predictive effect of baseline adiponectin levels in patients who provided plasma and were undergoing one of three pharmacological treatments (escitalopram monotherapy; escitalopram plus bupropion; and venlafaxine plus mirtazapine) in the Combining Medications to Enhance Depression Outcomes clinical trial (n=160). Specifically, we assessed whether adiponectin moderates-that is, differentially predicts-treatment response among the treatment arms. Improvements with treatment were assessed using change in the clinician-rated 30-item Inventory of Depressive Symptomatology (IDS-C) from baseline through week 12. Moderator effects were tested using separate pairwise repeated measures mixed-effects models with a treatment-arm-by-adiponectin interaction. RESULTS: Baseline adiponectin levels moderated treatment outcome between two combination therapies. Specifically, low adiponectin predicted better response to escitalopram plus bupropion compared to venlafaxine plus mirtazapine, whereas high adiponectin predicted better response to venlafaxine plus mirtazapine compared to escitalopram plus bupropion (F=4.84, p=0.03). Adiponectin levels did not correlate with baseline depression severity (r=-0.03, p=.59). CONCLUSIONS: Antidepressant selection for patients with MDD can be personalized using pre-treatment blood-based biomarkers, such as adiponectin, thereby improving treatment outcomes.