Quantitative autofluorescence imaging for predicting malignant transformation of oral leukoplakia: a long-term
Chenxi Li1, Junting Li2, Tianhao Jin1
1Department of Oral Mucosal Diseases, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, China; College of Stomatology, Shanghai Jiao Tong University, China; National Center for Stomatology, China; National Clinical Research Center for Oral Diseases, China; Shanghai Key Laboratory of Stomatology, China; Shanghai Research Institute of Stomatology, China.
Objective:
This study aimed to evaluate the predictive value of a quantitative autofluorescence imaging (AFI) index, the Relative Mean Gray Value (RMGV), for the malignant transformation of oral leukoplakia (OLK).
Materials And Methods:
A prospective cohort of 184 OLK patients was followed for a median of 89 months (range: 6-105). Autofluorescence images were quantitatively analyzed using software to calculate RMGV, defined as the difference in mean gray value between negative (normal mucosa) and positive (lesion) regions. The cohort was stratified into RMGV+ and RMGV- groups based on the median RMGV (-1.02). Malignant transformation (MT) to oral squamous cell carcinoma was the primary outcome. Statistical analyses included Kaplan-Meier survival analysis, Cox proportional hazards models, time-dependent ROC curves, and decision curve analysis.
Results:
The MT rate was significantly higher in the RMGV+ group (24.73%, 23/93) than in the RMGV- group (7.69%, 7/91). RMGV was an independent predictor of MT in multivariate analysis (Hazard Ratio provided in main text). A predictive model combining RMGV and non-homogeneous lesion type showed good calibration and clinical utility.
Conclusion:
The quantitative AFI index RMGV is a robust, independent predictor for malignant transformation of OLK. A model incorporating RMGV and clinical features can effectively stratify patient risk, showing promise as a non-invasive tool for improving OLK management.
