AUTHOR=Li Xi-ye , Pan Lei , Deng Yi-wen , Chen Jun-jun , Tian Zhen , Tang Guo-yao , Ge Shu-yun , Wang Yu-feng 

TITLE=Variable innate lymphoid cells predominancy in oral lichen planus latently led to diverse clinical outcomes: a proof-of-concept study

JOURNAL=Frontiers in Immunology

VOLUME=Volume 16 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2025.1551311

DOI=10.3389/fimmu.2025.1551311

ISSN=1664-3224

ABSTRACT=ObjectivesTo search for a new classification scheme for oral lichen planus (OLP) and oral lichenoid lesions (OLL) based on innate lymphoid cells (ILCs) and to evaluate the clinical significance of this classification for diagnosis and treatment.Subjects and methodsThis study was based on a clinical cohort and applied flow cytometry to prospectively analyze the ILC subgroups and proportions in OLP and OLL lesions using SPSS software (version 26.0) to attempt cluster analysis to classify diseases at the cellular level based on the phenotype and quantity of ILCs cells, analyze the correlation between the new classification of diseases and clinical risk factors based on the patient’s clinical background information and classification results, and evaluate the differences in therapeutic effects among patients in different groups in corresponding clinical cohorts.ResultsIn the OLP and OLL groups, the ILC compartment consisted mainly of ILC1 (75.02% ± 27.55% and 72.99% ± 25.23%, respectively), ILC2 (1.49% ± 4.12% and 1.72% ± 3.18%, respectively), and ILC3 (16.52% ± 19.47% and 18.77% ± 18.12%, respectively). Using k-means clustering and two-step clustering, patients could be clustered into three groups that did not respond equally to the same treatment. Using k-means clustering, there was a statistically significant difference in REU scores between the ILC1 advantage group and the OLL subgroup before and after treatment (P = 0.02), which was not observed in two-step clustering. This indicates that k-means clustering may have greater value in the clinical application of OLL. In the ILC1 absolute advantage group, using HCQ + TGP for one month could effectively treat the patients regardless of the use of k-means clustering or two-step clustering (P ≤0.001), whereas the other groups did not.ConclusionsThis study provides a preliminary OLP and OLL classification method based on ILC subgroups that can guide the cytological classification of diseases to a certain extent. Further clinical application values should be verified in subsequent cohort studies.