Depressive disorders contribute heavily to global disease burden; This is possibly because patients are often treated homogeneously, despite having heterogeneous symptoms with differing underlying neural mechanisms. A novel treatment that can directly influence the neural circuit relevant to an individual patient’s subset of symptoms might more precisely and thus effectively aid in the alleviation of their specific symptoms. We tested this hypothesis in a proof-of-concept study using fMRI functional connectivity neurofeedback. We targeted connectivity between the left dorsolateral prefrontal cortex/middle frontal gyrus and the left precuneus/posterior cingulate cortex, because this connection has been well-established as relating to a specific subset of depressive symptoms. Specifically, this connectivity has been shown in a data-driven manner to be less anticorrelated in patients with melancholic depression than in healthy controls. Furthermore, a posterior cingulate dominant state—which results in a loss of this anticorrelation—is expected to specifically relate to an increase in rumination symptoms such as brooding. In line with predictions, we found that, with neurofeedback training, the more a participant normalized this connectivity (restored the anticorrelation), the more related (depressive and brooding symptoms), but not unrelated (trait anxiety), symptoms were reduced. Because these results look promising, this paradigm next needs to be examined with a greater sample size and with better controls. Nonetheless, here we provide preliminary evidence for a correlation between the normalization of a neural network and a reduction in related symptoms. Showing their reproducibility, these results were found in two experiments that took place several years apart by different experimenters. Indicative of its potential clinical utility, effects of this treatment remained one-two months later.Clinical trial registration: Both experiments reported here were registered clinical trials (UMIN000015249, jRCTs052180169).

No abstract

Depressive disorders contribute heavily to global disease burden; This is possibly because patients are usually treated homogeneously, despite having heterogeneous symptoms with differing underlying neural mechanisms. On the contrary, treatment that directly influences the neural circuit relevant to an individual patient’s subset of symptoms might more precisely and thus effectively aid in the alleviation of their specific symptoms. We tested this hypothesis, using fMRI functional connectivity neurofeedback to target a neural biomarker that objectively relates to a specific subset (melancholic) of depressive symptoms and that is generalizable across independent cohorts of patients. The targeted biomarker was the functional connectivity between the left dorsolateral prefrontal cortex and left precuneus, which has been shown in a data-driven manner to be less anticorrelated in patients with melancholic depression than in healthy controls. We found that the more a participant normalized this biomarker, the more related (brooding and more general depressive), but not unrelated (trait anxiety), symptoms were reduced. Thus, one-to-one correspondence between a normalized neural network and decreased depressive symptoms was demonstrated. These results were found in two experiments that took place several years apart by different experimenters, indicating their reproducibility. Indicative of their potential clinical utility, effects remained one-two months later.