We here examine associations of a recently published polygenic risk score of antidepressant response (PRS-AR) with antidepressant treatment outcomes (remission and depression score change) in an independent clinical trial. We not only replicate the PRS-AR for escitalopram, but also find antidepressant interaction effects, suggesting drug-specificity of PRS-AR. We therefore also tested the utility of this PRS-AR to stratify between antidepressants and demonstrate a 14% increase in remission rate (from 43.6% to 49.7%), relative to the randomized remission rate.