Development of A Novel Prediction Model for Assessing Malignant Transformation of Oral Epithelial Dysplasia

Introduction: Nomograms are effective statistical tools that may be integrated into medical practice to calculate treatment outcomes, diagnose diseases, and guide treatment decisions based on patient-specific variables. The current study aimed to develop a novel nomogram to stratify patients with oral epithelial dysplasia (OED) into different risk groups based on their clinic-pathological data. Materials and Methods: Clinico-pathological data of all cases of OED reported during a twelve-year period were retrospectively retrieved. The presence of malignant transformation (MT) among them was confirmed from the departmental cancer registry. Bivariate regression analyses and multivariate logistic regression were used to analyze the risk factors and the role of multiple independent variables, respectively. A p-value < 0.05 was considered statistically significant. An OED nomogram was constructed from the above data. Results: A novel OED nomogram was constructed using three variables - induration, texture, and clinical type. The Receiver operating characteristic (ROC) curve demonstrated an area under the curve (AUC) of 0.753, indicating fair predictive potential. MTR of our study population was 9.8%. Conclusion: The OED nomogram serves as a valuable tool for predicting MT in OED patients. Its application in dental science for stratifying patient risk enables clinicians to plan treatment and follow-up visits efficiently.