Previous studies have implicated aberrant reward processing in the pathogenesis of adolescent depression. However, no study has used functional connectivity within a distributed reward network, assessed using resting-state functional MRI (fMRI), to predict the onset of depression in adolescents. This study used reward network-based functional connectivity at baseline to predict depressive disorder at follow-up in a community sample of adolescents.

A total of 637 children 6–12 years old underwent resting-state fMRI. Discovery and replication analyses tested intrinsic functional connectivity (iFC) among nodes of a putative reward network. Logistic regression tested whether striatal node strength, a measure of reward-related iFC, predicted onset of a depressive disorder at 3-year follow-up. Further analyses investigated the specificity of this prediction.

Increased left ventral striatum node strength predicted increased risk for future depressive disorder (odds ratio=1.54, 95% CI=1.09–2.18), even after excluding participants who had depressive disorders at baseline (odds ratio=1.52, 95% CI=1.05–2.20). Among 11 reward-network nodes, only the left ventral striatum significantly predicted depression. Striatal node strength did not predict other common adolescent psychopathology, such as anxiety, attention deficit hyperactivity disorder, and substance use.

Aberrant ventral striatum functional connectivity specifically predicts future risk for depressive disorder. This finding further emphasizes the need to understand how brain reward networks contribute to youth depression.

先前的研究表明异常的奖励处理与青少年抑郁症的发病机制有关。然而，还没有研究利用分布式奖励网络中的功能连接（通过静息态功能磁共振成像 (fMRI) 进行评估）来预测青少年抑郁症的发作。这项研究在基线时使用基于奖励网络的功能连接来预测社区青少年样本中随访时的抑郁症。

共有 637 名 6-12 岁儿童接受了静息态 fMRI。发现和复制分析测试了假定奖励网络节点之间的内在功能连接（iFC）。逻辑回归测试了纹状体节点强度（与奖励相关的 iFC 的测量）是否可以预测 3 年随访时抑郁症的发作。进一步的分析调查了这一预测的特异性。

左腹侧纹状体节点强度的增加预示着未来抑郁症的风险增加（比值比=1.54，95% CI=1.09-2.18），即使排除了基线时患有抑郁症的参与者（比值比=1.52，95% CI=1.05-） 2.20）。在 11 个奖励网络节点中，只有左腹侧纹状体显着预测抑郁症。纹状体结强度不能预测其他常见的青少年精神病理学，例如焦虑、注意力缺陷多动障碍和物质使用。

异常的腹侧纹状体功能连接可以明确预测未来患抑郁症的风险。这一发现进一步强调了了解大脑奖励网络如何导致青少年抑郁症的必要性。