Orbitofrontal Cortex Activity and Connectivity Predict Future Depression Symptoms in Adolescence

Abstract

Background Major depressive disorder is a leading cause of disability worldwide; however, little is known about pathological mechanisms involved in its development. Research in adolescent depression has focused on reward sensitivity and striatal mechanisms implementing it. The contribution of loss sensitivity to future depression, as well as the orbitofrontal cortex (OFC) mechanisms critical for processing losses and rewards, remain unexplored. Furthermore, it is unclear whether OFC functioning interacts with familial history in predicting future depression. Methods In this longitudinal study we recorded functional magnetic resonance imaging (fMRI) data while 229 adolescent females with or without parental history of depression completed a monetary gambling task. We examined if OFC blood-oxygen-level-dependent (BOLD) response and functional connectivity during loss and win feedback was associated with depression symptoms concurrently and prospectively (9 months later), and whether this relationship was moderated by parental history of depression. Results Reduced OFC response during loss was associated with higher depression symptoms concurrently and prospectively, even after controlling for concurrent depression, specifically in adolescents with parental history of depression. Similarly, increased OFC-posterior insula connectivity during loss was associated with future depression symptoms but this relationship was not moderated by parental history of depression. Conclusions This study provides the first evidence for loss-related alterations in OFC functioning and its interaction with familial history of depression as possible mechanisms in the development of depression. While the current fMRI literature has mainly focused on reward, the present findings underscore the need to include prefrontal loss processing in existing developmental models of depression.

Convergent neurobiological predictors of emergent psychopathology during adolescence

“…This body of work suggests that within limbic regions, smaller volumes and thinner cortices (Cheetham et al, ; Pagliaccio et al, ; Rao et al, ; Squeglia et al, ; Urosevic et al, ; Whittle et al, ) and reduced functional activation, particularly surrounding emotional (McClure et al, ) and rewarding stimuli (Buchel et al, ; Forbes et al, ; Hanson et al, ; Morgan et al, ; Straub et al, ; Stringaris et al, ; Telzer et al, ), appear to serve as a predictor for onset, escalation, and persistence of adolescent psychopathology. Additionally, smaller volumes and thinner prefrontal cortices (Brumback et al, ; Cheetham et al, ; Foland‐Ross et al, ; Kuhn et al, ; Squeglia et al, ) and reduced prefrontal activation during tasks involving rewarding and emotional stimuli (Buchel et al, ; Jin et al, ; Jones et al, ; Kujawa et al, ) and executive control (Heitzeg et al, ; Mahmood et al, ; Norman et al, ) also appear to be associated with the onset, escalation, and persistence of adolescent psychopathology. Lastly, reduced functional connectivity both within limbic regions (Connolly et al, ), and between limbic and prefrontal regions (Camchong et al, ; Scheuer et al, ; Strikwerda‐Brown et al, ), as well as smaller indices of white matter maturation (i.e., fractional anisotropy) in tracts serving limbic and frontal regions (Chung et al, ; Huang et al, ) may serve as a risk marker for greater psychopathological symptoms.…”

“…These findings are not surprising, given the developmental asynchrony of the frontolimbic system during the adolescent years (Mills et al, ; Simmonds et al, ), but they are by no means conclusive. Studies suggest that the predictive nature of these findings can differ based on sex (Foland‐Ross et al, ; Hanson et al, ; Jin et al, ; Whittle et al, ), environmental variables (e.g., maternal agression; Whittle et al, ), personality traits (e.g., novelty seeking; Buchel et al, ), and in the case of substance use, can vary based on the presence of substance‐induced consequences (e.g., blackouts; Wetherill et al, ) or the drug under investigation (e.g., marijuana vs alcohol; Cousijn et al, ; Jones et al, ). In order to make better predictions about emergent psychopathology and resolve inconsistencies in the existing literature, we need large, prospective, longitudinal studies that simultaneously examine a wide range of potential causal factors and their interactions.…”

“…Meanwhile, prefrontal activation during reward‐ and loss‐based decision making is also predictive of future depressive symptom severity. In healthy adolescents, greater ventromedial prefrontal cortex activation during reward receipt at baseline (males only) (Hanson et al, ) predicted a greater increase in depressive symptoms across a 2 year follow‐up, while reduced orbitofrontal cortex activation during loss at baseline (females only) predicted greater depressive symptoms at a 9 month follow‐up (Jin et al, ), with greater orbitofrontal‐posterior insula connectivity during loss also being associated with greater future depressive symptoms. In addition to studies looking at reward‐based decision making, another study used task‐based functional MRI during a peer exclusion task, and found that in healthy adolescents, greater subgenual anterior cingulate cortex activation during exclusion was associated with greater increases in parent‐reported adolescent depressive symptoms during the following year (Masten et al, ).…”

Intrinsic neural circuitry of depression in adolescent females

“…The fMRI sample ( N = 261) consisted of all participants who were willing to participate and met basic eligibility criteria for MRI (e.g., without metal implants, braces, and claustrophobia.). Difference in severity of depression symptoms between the fMRI subsample and participants who were not scanned were minimal suggesting little if any bias (Jin et al., ).…”

Great Expectations: A Critical Review of and Recommendations for the study of Reward Processing as a Cause and Predictor of Depression

“…Figure 4: Forest plot for random effects meta-analysis of observational fmri studies reporting a striatal effect for the correlation with change in depressive symptoms. Across these studies (23,28,55,59,62,66,69), predominantly conducted in adolescents, we found that the mean of the distribution effects size for similar studies was -0.10 [-0.17, -0.03]. 1 indicates statistics reported for the entire study population, not for the subgroup upon which displayed prediction is based.…”