Prediction of malignant transformation and recurrence of oral epithelial dysplasia using architectural and cytological feature specific prognostic models

Abstract

Oral epithelial dysplasia (OED) is a precursor state usually preceding oral squamous cell carcinoma (OSCC). Histological grading is the current gold standard for OED prognostication but is subjective and variable with unreliable outcome prediction. We explore if individual OED histological features can be used to develop and evaluate prognostic models for malignant transformation and recurrence prediction. Digitised tissue slides for a cohort of 109 OED cases were reviewed by three expert pathologists, where the prevalence and agreement of architectural and cytological histological features was assessed and association with clinical outcomes analysed using Cox proportional hazards regression and Kaplan–Meier curves. Within the cohort, the most prevalent features were basal cell hyperplasia (72%) and irregular surface keratin (60%), and least common were verrucous surface (26%), loss of epithelial cohesion (30%), lymphocytic band and dyskeratosis (34%). Several features were significant for transformation (p < 0.036) and recurrence (p < 0.015) including bulbous rete pegs, hyperchromatism, loss of epithelial cohesion, loss of stratification, suprabasal mitoses and nuclear pleomorphism. This led us to propose two prognostic scoring systems including a ‘6-point model’ using the six features showing a greater statistical association with transformation and recurrence (bulbous rete pegs, hyperchromatism, loss of epithelial cohesion, loss of stratification, suprabasal mitoses, nuclear pleomorphism) and a ‘two-point model’ using the two features with highest inter-pathologist agreement (loss of epithelial cohesion and bulbous rete pegs). Both the ‘six point’ and ‘two point’ models showed good predictive ability (AUROC ≥ 0.774 for transformation and 0.726 for recurrence) with further improvement when age, gender and histological grade were added. These results demonstrate a correlation between individual OED histological features and prognosis for the first time. The proposed models have the potential to simplify OED grading and aid patient management. Validation on larger multicentre cohorts with prospective analysis is needed to establish their usefulness in clinical practice.

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

“…They have stated in their work that the lack of CD8 Tcells in non-cytotoxic subtype and non-immune reactive subtype can lead to progression in moderate and severe dysplasia. It is also interesting to note that our study has found binary grading to be a significant indicator for malignant transformation whereas the study performed by Dost et al 30 has shown no association between grading and transformation whereas Mahmood et al 32 showed associate between nuclear features and transformation. However, the nuclear features used corresponds to OED grading e.g., bulbus rete pegs, loss of epithelial cohesion etc., and upon adding histological grades into the mix they observed improvements in their results.…”

“…We have demonstrated that deep learning based weakly supervised IDaRS can predict malignant transformation with an AUROC of ~0.78 (±0.07 SD) on stratified 5-fold cross-validation using three different random seeds. Mahmood et al 32 also reported the AUROC of 0.77 for transformation using a similar but smaller cohort with the nuclear features subjectively assessed by three pathologists. The higher performance of IDaRS is because it dynamically learns important feature representations from the patches internally, as compared to fixed feature representation of a patch as an input limiting the learning possibilities of the model.…”

“…Malignant transformation was also shown to have significant association with age (p = 0.034), clinical appearance (p = 0.030), lesion site (p = 0.007) and some other clinical features with a mean transformation time of 50 months (±32.5 SD). A recent study by Mahmood et al 32 examined the correlation between individual histological features and OED prognosis. They examined OED biopsies from 108 patients with a minimum of five-year follow-up to analyse histological features predictive of recurrence and malignant transformation.…”

Molecular Biomarkers of Malignant Transformation in Head and Neck Dysplasia

“…The use of these molecular markers in conjunction with other early clinical and histologic indicators may reliably predict disease progression [ 138 ]. The validation of prognostic models which significantly predict malignant transformation and recurrence based on specific histologic biomarkers (e.g., basal cell hyperplasia, loss of epithelial cohesion) has been successful and is ongoing [ 139 ]. Ultimately, validation of a prognostic scoring system incorporating molecular markers, clinical risk factors, and histologic grading will provide the greatest clinical utility.…”

“…Similarly, studies have shown that deficiencies in DNA mismatch repair predict a positive response to immune checkpoint inhibition as well as platinum-based chemotherapies in HNSCC [ 140 , 141 ]. This is also true of cell cycle regulators, as amplification of cyclin D1 predicts cisplatin resistance, TP53 mutations predict susceptibility to G2-M cell cycle inhibitors [ 142 ], and Bcl-2 overexpression predicts radiotherapy failure with 71% accuracy [ 139 ]. Various therapeutic biomarkers, ranging from viral oncoproteins (e.g., HPV, EBV) and receptor tyrosine kinases (e.g., EGFR, PIK3CA) to immune checkpoint markers (e.g., PD-L1, PD-L2) and tumor suppressor proteins (e.g., TP53) have been identified as therapeutic targets in HNSCC [ 142 ].…”