PS209. Changes of cognitive function after escitalopram administration using single word-induced hemoglobin variation as an index

About 40% of patients treated with antidepressant do not show a response. However, the underlying mechanisms remain unclear and there is no objective test for the prediction of treatment response. The aim of this study was to investigate the proteomic differences between responder and non-responder in patients with major depressive disorder (MDD who were treated with escitalopram. Fasting blood samples were collected and Clinical Global Impression scale (CGI) was completed at baseline and at day 3, 7, 12, and 42. Five patients who completed the 42days study treatment were included in proteomic analysis. We performed liquid chromatography–mass spectrometry (LC-MS) based proteomics for quantitative profiling of the serum proteomes from seven patients with MDD. Data-independent acquisition (DIA) method was employed for the quantification of each serum proteome. In order to identify proteins that are correlated throughout the antidepressant treatment, the Pearson’s correlation coefficients between the protein profiles were calculated and the expression correlation networks (ExprsCorrNetwork) were constructed. The escitalopram network was constructed to identify functional associations between the serum proteins and the drug response. The genes associated with escitaloprm were retrieved from PharmGKB database and the 1st interactors of the seed genes were parsed from the protein–protein interaction network of BioGrid. Then, the overlaps between the ExprsCorrNetwork and the escitalopram network were computed. We tested functional associations of ExprsCorrNetwork with the known molecular and physiological evidences of MDD and escitalopram. We retrieved the genes identified in GWAS studies which enrolled patients treated with either citalopram or escitalopram. Between responders and nonresponders, there were differences in protein profile. The profiles of responders were closer at 6 weeks than the profiles of non-responders. R-specific correlations might be related with escitalopram response. Functional associations with MDD and escitalopram should be more investigated