Background: Psoriasis is an immune-mediated inflammatory skin disease. Psoriasis severity evaluation is important for clinicians in the assessment of disease severity and subsequent clinical decision making. However, no objective biomarker is available for accurately evaluating disease severity in psoriasis.
Objective: To define and compare biomarkers of disease severity and progression in psoriatic skin.
Methods: We performed proteome profiling to study the proteins circulating in the serum from patients with psoriasis, psoriatic arthritis and ankylosing spondylitis, and transcriptome sequencing to investigate the gene expression in skin from the same cohort. We then used machine learning approaches to evaluate different biomarker candidates across several independent cohorts. In order to reveal the cell-type specificity of different biomarkers, we also analyzed a single-cell dataset of skin samples. In-situ staining was applied for the validation of biomarker expression.
Results: We identified that the peptidase inhibitor 3 (PI3) was significantly correlated with the corresponding local skin gene expression, and was associated with disease severity. We applied machine learning methods to confirm that PI3 was an effective psoriasis classifier, Finally, we validated PI3 as psoriasis biomarker using in-situ staining and public datasets. Single-cell data and in-situ staining indicated that PI3 was specifically highly expressed in keratinocytes from psoriatic lesions.
Conclusion: Our results suggest that PI3 may be a psoriasis-specific biomarker for disease severity and hyper-keratinization.
Keywords: Biomarkers; Keratinocytes; Peptidase inhibitor 3; Psoriasis.
Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved.