Abstract Treatment for major depressive disorder (MDD) remains challenging as only 30–50% of patients respond to first-line antidepressant medications in primary care. Here we developed algorithms using predictors of response to sertraline and bupropion from a multisite study, and tested such markers in an independent, prospective clinical trial involving unmedicated individuals with MDD ( NCT05537584 ). Leave-one-out cross-validation models achieved good performance in the training sample (area under the curve of 0.66–0.86). In the preregistered clinical trial, no significant differences in treatment outcomes emerged for those assigned a drug consistent versus inconsistent with their biomarkers. However, significant differences emerged in symptom reduction trajectories for those with positive markers for both medications (response rate: 71.4%) or either drug (65.4%) compared with those with two negative markers (42.9%). This is the first study using biobehavioral markers to prospectively guide assignment to two widely used antidepressants, yielding a 66.8% boost in response rate and providing foundations for larger personalized treatment studies of MDD.
Zhukovsky et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: