Hippocampal and parahippocampal volume and function predict antidepressant response in patients with major depression: A multimodal neuroimaging study

Abstract

Background: For many patients with major depressive disorder (MDD) adequate treatment remains elusive. Neuroimaging techniques received attention for their potential use in guiding and predicting response, but were rarely investigated in real-world psychiatric settings. Aims: To identify structural and functional Magnetic Resonance Imaging (MRI) biomarkers associated with antidepressant response in a real-world clinical sample. Methods: We studied 100 MDD inpatients admitted to our psychiatric ward, treated with various antidepressants upon clinical need. Hamilton Depression Rating Scale percentage decrease from admission to discharge was used as a measure of response. All patients underwent 3.0 T MRI scanning. Grey matter (GM) volumes were investigated both in a voxel-based morphometry (VBM), and in a regions of interest (ROI) analysis. In a subsample of patients, functional resting-state connectivity patterns were also explored. Results: In the VBM analysis, worse response was associated to lower GM volumes in two clusters, encompassing the left hippocampus and parahippocampal gyrus, and the right superior and middle temporal gyrus. Investigating ROIs, lower bilateral hippocampi and amygdalae volumes predicted worse treatment outcomes. Functional connectivity in the right temporal and parahippocampal gyrus was also associated to response. Conclusion: Our results expand existing literature on the relationship between the structure and function of several brain regions and treatment response in MDD. While we are still far from routine use of MRI biomarkers in clinical practice, we confirm a possible role of these techniques in guiding treatment choices and predicting their efficacy.