**5. Conclusions**

To monitor the association between the efficacy of SSRIs and biomarker abundance, plasma samples were collected for 10 weeks during treatment of patients with MDD. Biomarkers have been identified through longitudinal measurements of protein concentrations, with some showing significant correlation with mental disease variables. These findings suggest that the liquid biopsy technique may solve unmet clinical problems.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2227-9059/8/11/455/s1: Figure S1. Box plots for protein abundances in each LC-MS/MS run. Protein abundances (a) before and (b) after endogenous protein-based normalization. (c) Two-dimensional global t-SNE map comparing the responders (blue) and non-responders (red) at each sampling time (1, 4, and 10 weeks) and 3 t-SNE parameters of perplexity. (d) Plasma protein log2 abundances (ng/mL) in the Plasma Proteome Database (bottom) and normalized protein abundance (right). Figure S2. Changes in responders over time to the amounts of 37 significant proteins belonging to six clusters, as determined by response/time interaction. Clusters 1 through 6 included 3, 5, 3, 5, 10 and 11 proteins, respectively. Figure S3. Calibration curves of 15 surrogate peptides relative to nine proteins based on heavy-to-light extracted ion chromatogram ratio. Figure S4. MRM results of five proteins at T0 vs. T1. Boxplot of five quantified proteins paired at T0 and T1 in responders and non-responders. Table S1. Demographic and clinical characteristics of the ten depressed patients at each of the four hospital visits. Table S2. The 1,159 plasma proteins identified in 104 LC-MS/MS measurements. The notation system is (sample name)\_(response)\_(time)\_(experiment number). Table S3. Log2 transformed normalized protein abundance of 316 proteins in 104 LC-MS/MS measurements. Table S4. MRM parameters optimized for 14 target peptides of nine proteins. Table S5. Spearman's rank correlation coefficients and adjusted *p-*values for the relationships between 316 protein abundance and six psychiatric symptom indices. Table S6. Estimation results in generalized estimation equation.

**Author Contributions:** Conceptualization, E.Y.K., M.Y.L., H.J.L., K.K. and Y.M.A.; methodology, H.M., J.Y. (Jiyoung Yu), J.Y. (Jeonghum Yeom), M.Y.L., and H.-S.A.; validation, H.-S.A.; formal analysis, H.J.; resources, E.Y.K., H.J.L., and Y.M.A.; data curation, H.-S.A.; writing—original draft preparation, H.-S.A., E.Y.K. and K.K.; writing—review and editing, H.-S.A., E.Y.K., M.Y.L., and K.K.; visualization, H.-S.A.; supervision, K.K.; funding acquisition, E.Y.K., K.K., and Y.M.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2017R1D1A1B03028787) and the Korean government (MSIT) (NRF-2019M3E5D3073106 and NRF-2019M3E5D30733690).

**Acknowledgments:** The authors gratefully acknowledge the participation of all patients and investigators involved in this trial.

**Conflicts of Interest:** Yong Min Ahn has received research support from or served as a speaker for Janssen Korea, Ltd., Lundbeck Korea Co., Ltd., and Korea Otsuka Pharmaceutical. Janssen Korea, Ltd., Lundbeck Korea Co., Ltd., and Korea Otsuka Pharmaceutical and the funder had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
