Figure 1: Functional connectivity of the DLPFC and supramarginal gyrus predicting differential treatment response to sertraline and venlafaxine.

A: Responders to sertraline showed higher FC between the right DLPFC and supramarginal gyrus (t=3.809, p<0.001) whereas, in the venlafaxine-XR group, the opposite was true (t=−2.391, p=0.020). Higher coupling at baseline was associated with greater QIDS-SR₁₆ decrease after treatment in the sertraline group (rho=−0.566, p<0.001) and the opposite was true for venlafaxine-XR (rho=0.441, p=0.005). B: The coupling between right supramarginal gyrus and middle temporal gyrus was higher in responders to sertraline compared to non-responders (t=3.591, p=0.001), whereas in venlafaxine-XR responders connectivity was lower compared to non-responders (t=−2.770, p=0.009). Higher coupling at baseline was associated with greater QIDS-SR₁₆ decrease after treatment in the sertraline group (rho=−0.560, p<0.001) and the opposite was true in the venlafaxine-XR group (rho=0.457, p=0.003). Numbers indicate slice coordinates in MNI space. Stars mark significant comparisons between groups (t-test p<0.05). For correlations, a linear fit line calculated based on the data is shown. ROI coordinates in MNI space: DLPFC=(34 58 0), SMG=(52 −44 46). FC is expressed as difference of the parameter estimates of the hemodynamic response function (beta values) for the NoGo and Go conditions. Abbreviations: DLPFC=dorsolateral prefrontal cortex (MFG1 in Table 2), Esc=escitalopram, FC=functional connectivity, HC=healthy controls, MNI=Montreal neurological institute MTG=middle temporal gyrus, NR=non-responders, QIDS=Quick Inventory of Depressive Symptomatology scale, R=responders, ROI=region of interest, Ser=sertraline, SMG=supramarginal gyrus, Ven=venlafaxine, XR=extended-release, Δ=difference between follow-up and baseline.