AUTHOR=Liu Binhao , Shen Jiacheng , Li Jiayu , Tian Bowen , Zhou Bin , Gui Jiachen , Li Zhimin , Zhang Yue , Hu Wenzhi , Li Qiang 

TITLE=Candidate approaches for predicting vitiligo recurrence: an effective model and biomarkers

JOURNAL=Frontiers in Immunology

VOLUME=Volume 16 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2025.1468665

DOI=10.3389/fimmu.2025.1468665

ISSN=1664-3224

ABSTRACT=BackgroundVitiligo is a challenging chronic condition with unpredictable disease course and high propensity for relapse post-treatment. Recent studies have reported the biomarkers for disease activity, severity, and therapeutic response of vitiligo, yet very few have investigated cytokines as predictive biomarkers for disease recurrence in vitiligo. This study aims to explore cytokines that serve as biomarkers for disease recurrence and extend research on factors related to the disease’s activity.Methods92 patients and 40 healthy controls were recruited at the Air Force Medical Center from September 20, 2023, to November 30, 2023. Ultrasensitive multiplex cytokine array was used to measure plasma concentrations of cytokines, including IFN-γ, CXCL9, CXCL10, CXCL11, IL-6, and IL-15.ResultsIFN-γ, CXCL9, CXCL10, CXCL11, IL-6, and IL-15 were expressed at higher levels in the circulation of patients with both segmental and non-segmental vitiligo compared to healthy controls (p < 0.001). There were no significant differences in these cytokine levels between the two types of vitiligo. CXCL9 was associated with the activity of vitiligo (p = 0.027). Correlation analysis showed a positive relationship between IFN-γ, CXCL9, CXCL10, CXCL11, IL-6, and IL-15 in the plasma of patients with recurrent vitiligo. The expression of IFN-γ, CXCL9, CXCL10, CXCL11, and IL-6 was significantly higher in recurrent vitiligo than in cases of persistent stable vitiligo (p = 0.001, p = 0.003, p < 0.001, p = 0.002, p = 0.026, respectively), with ROC analysis demonstrating their predictive capability for vitiligo recurrence, with AUC values of 0.806, 0.773, 0.896, 0.785, and 0.709, respectively. Multivariate logistic regression model showed IFN-γ is an independent predictor for vitiligo recurrence [OR (95%CI) =1.051 (1.012~1.116)], with a prediction accuracy of 90.5% (38/42) on the training dataset and 88.9% (16/18) on the testing dataset.ConclusionPlasma IFN-γ, CXCL9, CXCL10, CXCL11 and IL-6 might be potential biomarkers for vitiligo recurrence, with CXCL9 also associated with disease activity. Additionally, multivariate logistic regression model demonstrated that IFN-γ is an independent predictor of vitiligo recurrence and the model could be a candidate approach for predicting vitiligo recurrence.