Cognitive dysfunction in depression and bipolar disorder (BD) is insufficiently targeted by available treatments. Erythropoietin (EPO) increases neuroplasticity and may improve cognition in mood disorders, but the neuronal mechanisms of these effects are unknown. This functional magnetic resonance imaging (fMRI) study investigated the effects of EPO on neural circuitry activity during working memory (WM) performance.

Patients with treatment-resistant major depression, who were moderately depressed, or with BD in partial remission, were randomized to eight weekly infusions of EPO (40 000 IU) (N = 30) or saline (N = 26) in a double-blind, parallel-group design. Patients underwent fMRI, mood ratings and blood tests at baseline and week 14. During fMRI patients performed an n-back WM task.

EPO improved WM accuracy compared with saline (p = 0.045). Whole-brain analyses revealed that EPO increased WM load-related activity in the right superior frontal gyrus (SFG) compared with saline (p = 0.01). There was also enhanced WM load-related deactivation of the left hippocampus in EPO-treated compared to saline-treated patients (p = 0.03). Across the entire sample, baseline to follow-up changes in WM performance correlated positively with changes in WM-related SFG activity and negatively with hippocampal response (r = 0.28–0.30, p < 0.05). The effects of EPO were not associated with changes in mood or red blood cells (p ⩾0.08).

The present findings associate changes in WM-load related activity in the right SFG and left hippocampus with improved executive function in EPO-treated patients. Clinical trial registration: clinicaltrials.gov: NCT00916552.

Healthy first-degree relatives of patients with major depression (rMD+) show brain structure and functional response anomalies and have elevated risk for developing depression, a disorder linked to abnormal serotonergic neurotransmission and reward processing.

In a two-step functional magnetic resonance imaging (fMRI) investigation, we first evaluated whether positive and negative monetary outcomes were differentially processed by rMD+ individuals compared to healthy first-degree relatives of control probands (rMD−). Second, in a double-blinded placebo-controlled randomized trial we investigated whether a 4-week intervention with the selective serotonergic reuptake inhibitor (SSRI) escitalopram had a normalizing effect on behavior and brain responses of the rMD+ individuals.

Negative outcomes increased the probability of risk-averse choices in the subsequent trial in rMD+ but not in rMD− individuals. The orbitofrontal cortex (OFC) displayed a stronger neural response when subjects missed a large reward after a low-risk choice in the rMD+ group compared to the rMD− group. The enhanced orbitofrontal response to negative outcomes was reversed following escitalopram intervention compared to placebo. Conversely, for positive outcomes, the left hippocampus showed attenuated response to high wins in the rMD+ compared to the rMD− group. The SSRI intervention reinforced the hippocampal response to large wins. A subsequent structural analysis revealed that the abnormal neural responses were not accounted for by changes in gray matter density in rMD+ individuals.

Our study in first-degree relatives of depressive patients showed abnormal brain responses to aversive and rewarding outcomes in regions known to be dysfunctional in depression. We further confirmed the reversal of these aberrant activations with SSRI intervention.