Generalizability of treatment outcome prediction in major depressive disorder using structural MRI: A NeuroPharm study

• DOI: 10.1016/j.nicl.2022.103224
• Publication Date: 2022-10-10T14:25:20Z
• Authors: Vincent Beliveau; Ella Hedeboe; Patrick M. Fisher; Vibeke H. Dam; Martin B. Jørgensen; Vibe G. Frokjaer; Gitte M. Knudsen; Melanie Ganz
• Keywords: Escitalopram; Sertraline

Brain morphology has been suggested to be predictive of drug treatment outcome in major depressive disorders (MDD). The current study aims at evaluating the performance pretreatment structural brain magnetic resonance imaging (MRI) measures predicting a MDD large single-site cohort, and, importantly, assess generalizability these findings an independent cohort. random forest, boosted trees, support vector machines and elastic net classifiers were evaluated response remission following eight week using derived with FastSurfer (FreeSurfer). Models trained tested within nested cross-validation framework NeuroPharm dataset (n = 79, treatment: escitalopram); their was assessed clinical dataset, EMBARC 64, sertraline). Prediction antidepressant Neuropharm cohort statistically significant for forest (p 0.048), whereas none models could significantly predict remission. Furthermore, entire dataset. Although our primary some, but limited value MRI MDD, did not generalize suggesting applicability. This emphasizes importance assessing model establishing utility.