Abstract
PURPOSE: Oral epithelial dysplasia (OED) is a premalignant condition with a variable risk of malignant transformation (MT). This study investigates the recurrence and MT rates in an Iranian patient cohort with OED, evaluating associations with demographic factors, lifestyle habits, and histopathologic grading systems, with the aim of providing population-specific prognostic insights and enhancing disease prediction and patient outcomes. METHODS: This mixed cohort study included 89 surgically treated patients with OED, followed for 24-168 months. Demographic data, smoking and alcohol consumption history, recurrence, and MT status were collected from medical records. Histopathological grading was performed using both the WHO 2017 classification and a binary grading system. Kaplan-Meier survival analysis, Cox regression, and statistical tests were applied to identify prognostic factors. A p-value <0.05 was considered statistically significant. RESULTS: During follow-up, 22.5% (20/89) of patients experienced recurrence, and 30.3% (27/89) underwent MT. The tongue was the most common site for OED (48.3%), recurrence (55%), and MT (44.4%). Males were more frequently affected than females (60.7% vs. 39.3%,), but females tended to remain cancer-free more often. Smoking and alcohol consumption did not significantly influence cancer development. Kaplan-Meier analysis showed 50% of MT cases occurred after about 108 months (mean: 114.8 months). The combined mean survival for recurrence and MT was 92.8 months. Cox regression identified WHO grading as the only statistically significant predictor of MT and recurrence/MT (p = 0.001). CONCLUSIONS: The WHO grading system plays a crucial role in predicting the likelihood of OED recurrence and malignant transformation. Lesion location showed only a non-significant trend, suggesting that its effect may be mediated by grade. These findings highlight the importance of using grading-based risk assessments and ensuring close follow-up care to better manage OED and improve patient outcomes.