A digital score of peri-epithelial lymphocytic activity predicts malignant transformation in oral epithelial dysplasia

Oral squamous cell carcinoma (OSCC) is amongst the most common cancers worldwide, with more than 377,000 new cases worldwide each year. OSCC prognosis remains poor, related to cancer presentation at a late stage indicating the need for early detection to improve patient prognosis. OSCC is often preceded by a premalignant state known as oral epithelial dysplasia (OED), which is diagnosed and graded using subjective histological criteria leading to variability and prognostic unreliability. In this work, we propose a deep learning approach for the development of prognostic models for malignant transformation and their association with clinical outcomes in histology whole slide images (WSIs) of OED tissue sections. We train a weakly supervised method on OED (n= 137) cases with transformation (n= 50) status and mean malignant transformation time of 6.51 years (±5.35 SD). Performing stratified 5-fold cross-validation achieves an average AUROC of ~0.78 for predicting malignant transformations in OED. Hotspot analysis reveals various features from nuclei in the epithelium and peri-epithelial tissue to be significant prognostic factors for malignant transformation, including the count of peri-epithelial lymphocytes (PELs) (p < 0.05), epithelial layer nuclei count (NC) (p < 0.05) and basal layer NC (p < 0.05). Progression free survival using the Epithelial layer NC (p < 0.05, C-index = 0.73), Basal layer NC (p < 0.05, C-index = 0.70) and PEL count (p < 0.05, C-index = 0.73) shown association of these features with a high risk of malignant transformation. Our work shows the application of deep learning for prognostication and progression free survival (PFS) prediction of OED for the first time and has a significant potential to aid patient management. Further evaluation and testing on multi-centric data is required for validation and translation to clinical practice.