Understanding the mechanisms of major depressive disorder (MDD) improvement is a key challenge to determining effective personalized treatments.

To identify a data-driven pattern of clinical improvement in MDD and to quantify neural-to-symptom relationships according to antidepressant treatment, we performed a secondary analysis of the publicly available dataset EMBARC (Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care). In EMBARC, participants with MDD were treated either by sertraline or placebo for 8 weeks (Stage 1), and then switched to bupropion according to clinical response (Stage 2). We computed a univariate measure of clinical improvement through a principal component (PC) analysis on the variations of individual items of four clinical scales measuring depression, anxiety, suicidal ideas, and manic-like symptoms. We then investigated how initial clinical and neural factors predicted this measure during Stage 1 by running a linear model for each brain parcel’s resting-state global brain connectivity (GBC) with individual improvement scores during Stage 1.

The first PC (PC1) was similar across treatment groups at stages 1 and 2, suggesting a shared pattern of symptom improvement. PC1 patients’ scores significantly differed according to treatment, whereas no difference in response was evidenced between groups with the Clinical Global Impressions Scale. Baseline GBC correlated with Stage 1 PC1 scores in the sertraline but not in the placebo group.

Using data-driven reduction of symptom scales, we identified a common profile of symptom improvement with distinct intensity between sertraline and placebo.

Mapping from data-driven symptom improvement onto neural circuits revealed treatment-responsive neural profiles that may aid in optimal patient selection for future trials.