Xiangjie Chen,1– 3,* Bochun Cao,2,* Zhiren Tan,2 Xiaoping Li,4 Wenrong Xu,5 Ying Liu,2 Fang Gong1,2 

1Department of Laboratory Medicine, Jiangnan University Medical Center (Wuxi No. 2 People’s Hospital), Wuxi, Jiangsu, People’s Republic of China; 2Wuxi School of Medicine, Jiangnan University, Wuxi, Jiangsu, People’s Republic of China; 3State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi, Jiangsu, People’s Republic of China; 4Department of Laboratory Medicine, the First Affiliated Hospital of Soochow University, Soochow University, Suzhou, People’s Republic of China; 5Department of Dermatology, The Affiliated Wuxi People’ s Hospital of Nanjing Medical University, Wuxi, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Fang Gong; Ying Liu, Email 9862023157@jiangnan.edu.cn; 8202310001@jiangnan.edu.cn

Background: Atopic dermatitis (AD) is a chronic skin disorder that impacts patients’ physical and mental health. Diagnosing AD mainly depends on evaluating medical history and symptoms, as there are no universally accepted biomarkers for it. Identifying novel, reliable biomarkers is crucial to enhance diagnostic accuracy, reduce healthcare costs, and aid in developing new treatments.
Methods: Data from the Gene Expression Omnibus database were used to identify potential AD biomarkers through Weighted Gene Co-expression Network Analysis and machine-learning. External datasets confirmed these biomarkers’ diagnostic utility and their effectiveness in assessing clinical treatment. We also gathered peripheral blood mononuclear cells from healthy individuals and AD patients to validate these biomarkers’ diagnostic capability for AD. Correlation analyses linked these biomarkers to AD severity indicators. Euclidean distance clustering was employed to assess the ability of these biomarkers to distinguish between healthy individuals and AD patients. The study also examined their relationships with major inflammatory pathways in AD to understand their mechanisms.
Results: The study identified Ribonucleotide Reductase Regulatory Subunit M2 (RRM2), Late Cornified Envelope 3D (LCE3D), Cornifelin (CNFN), and Small Proline Rich Protein 2G (SPRR2G) as biomarkers with greater diagnostic value for AD than traditional biomarkers like eosinophil count and IgE levels. Treatment led to decreased expression of RRM2, LCE3D, and CNFN in AD patients’ skin, indicating their potential as markers for evaluating treatment efficacy. LCE3D, CNFN, and SPRR2G correlated with AD severity indicators such as the SCORAD score and serum IgE levels. Additionally, the overexpression of these biomarkers was linked to the activation of inflammatory pathways, suggesting their role in AD pathogenesis and progression.
Conclusion: Our study identifies RRM2, LCE3D, CNFN, and SPRR2G as novel biomarkers for diagnosing AD in peripheral blood and lesional tissues, with potential for assessing disease severity, evaluating treatment efficacy, and serving as targets for diagnosis and treatment.

Keywords: atopic dermatitis, AD, biomarker, transcriptome, bioinformatics