Background. The present study investigated alteration of brain resting-state activity induced by antidepressant treatment and attempted to investigate whether treatment efficacy can be predicted at an early stage of pharmacological treatment. Method. Forty-eight first-episode medication-free patients diagnosed with major depression received treatment with escitalopram. Resting-state functional magnetic resonance imaging was administered prior to treatment, 5 h after the first dose, during the course of pharmacological treatment (week 4) and at endpoint (week 8). Resting-state activity was evaluated in the course of the 8-week treatment and in relation to clinical improvement. Results. Escitalopram dynamically modified resting-state activity in depression during the treatment. After 5 h the antidepressant induced a significant decrease in the signal in the occipital cortex and an increase in the dorsolateral and dorsomedial prefrontal cortices and middle cingulate cortex. Furthermore, while remitters demonstrated more obvious changes following treatment, these were more modest in non-responders suggesting possible tonic and dynamic differences in the serotonergic system. Changes after 5 h in the caudate, occipital and temporal cortices were the best predictor of clinical remission at endpoint. Conclusions. This study revealed the possibility of using the measurement of resting-state neural changes a few hours after acute administration of antidepressant to identify individuals likely to remit after a few weeks of treatment.