Although discovering novel biomarkers for psoriasis is challenging, it may play an essential role in diagnosis, severity assessment, and prediction of treatment outcome and prognosis. The study was aimed to determine potential serum biomarkers of psoriasis via proteomic data analysis and clinical validity assessment. Thirty-one subjects manifested psoriasis and 19 subjects were healthy volunteers who were enrolled in the study. Protein expression was performed via two-dimensional gel electrophoresis (2-DE) using psoriasis patients’ sera before and after treatment and sera of patients without psoriasis. Image analysis was then performed. Nano-scale liquid chromatography-tandem mass spectrometry (LC-MS/MS) experiments subsequently identified points showing differential expression in 2-DE image analysis. To measure levels of candidate proteins to validate results obtained from 2-DE, enzyme linked immunosorbent assay (ELISA) was then conducted. Gelsolin was identified as a potential protein through LC-MS/MS analysis and database search. Serum gelsolin levels were lower in the groups of psoriasis patients before treatment than in the control group and the group of psoriasis patients after treatment. Additionally, in subgroup analysis, serum gelsolin level was correlated with various clinical severity scores. In conclusion, low serum gelsolin levels are associated with the severity of psoriasis, proposing the potential role of gelsolin as a biomarker for severity assessment and evaluation of treatment response of psoriasis.