Biobehavioral Predictors of Illness Progression in Adolescent Depression

PROJECT SUMMARY / ABSTRACT Adolescent depression is a major public health concern associated with significant morbidity and mortality. We know that adolescent depression varies considerably in severity and course, but its underlying mechanisms remain unknown. This proposal addresses this concern, aiming to identify biobehavioral predictors of illness trajectory in adolescents to inform treatment influencing its course. Our overall hypothesis is that reward dysfunction and its underlying neuro-immunological mechanisms contribute to maintenance of depression in youth. We propose the following model: (1) peripheral inflammation (e.g., cytokines, kynurenines) is associated with anhedonia; (2) inflammation extends to the CNS, inducing oxidative stress [↓glutathione (GSH, antioxidant)] and GABA (inhibitory neurotransmitter) deficits; (3) such neuro-chemical changes alter the reward circuitry, leading to anhedonia, and contribute to depression progression. In support, we documented that of adolescent depression’s core symptoms, only anhedonia—reflecting deficits in reward processes—and not irritability was associated with worse outcome, including suicidality and chronicity. We also reported associations between circulatory cytokines and kynurenines with both anhedonia and reward neurocircuitry. Utilizing proton MR spectroscopy (1H MRS), we showed that decreased anterior cingulate cortex (ACC) GABA in adolescent depression was driven by anhedonia. Additionally, using fMRI, we mapped striatal-based connectivity in relation to anhedonia and documented distinct neural activation patterns of reward anticipation and attainment—both reward processes that contribute to anhedonia. In pilot studies, we found that ACC GABA and reward- anticipation brain activation predicted anhedonia severity at 2-year follow-up. Building upon our compelling findings to date, we propose a prospective study of reward processes and neuroimmunological mechanisms, with the goal to identify modifiable pathways predictive of depression course in adolescents. We will study 120 adolescents with depressive symptoms. Three in-laboratory comprehensive clinical evaluation and computerized reward task sessions will be done at baseline, 12-, and 24-month follow-ups. At baseline and 12 months, visits will include: a) immune blood collection for kynurenine metabolites, immune cell profiling, and cytokine secretion post immune stimulation (as a biological stressor); b) fMRI resting state and tasks examining distinct reward processes (e.g., anticipation, attainment, learning); and c) 1H MRS to probe ACC GABA and striatal GSH as indices for inflammatory neurometabolic consequences in reward-related brain regions. Outcome measures will include anhedonia severity (primary), along with depression severity, functioning, anxiety, and suicidality. Principled variable selection approach will address multiple comparison concerns. Dynamic causal modeling and machine learning classification methods will explore the relationships among all examined factors.