Assessing Oral Epithelial Dysplasia Risk for Transformation to Cancer: Comparison Between Histologic Grading Systems Versus S100A7 Immunohistochemical Signature-based Grading

• Published in: Applied immunohistochemistry & molecular morphology Vol. 31; no. 6; pp. 399 - 405
• Main Authors: Darling, Mark Roger, Hwang, Jason T.K., Dickson, Benjamin J., Cutz, Jean-Claude, Salama, Samih, McCord, Christina, Pritzker, Kenneth P.H., Mock, David, Thompson, Lester D.R.
• Format: Journal Article
• Language: English
• Published: United States Lippincott Williams & Wilkins 01-07-2023
• ISSN: 1541-2016; 1533-4058
• DOI: 10.1097/PAI.0000000000001132

While a 3-tier oral epithelial dysplasia grading system has been utilized for decades, it is widely recognized as a suboptimal risk indicator for transformation to cancer. A 2-tier grading system has been proposed, although not yet validated. In this study, the 3-tier and 2-tier dysplasia grading systems, and an S100A7 immunohistochemical signature-based grading system were compared to assess prediction of risk of transformation to oral cancer. Formalin-fixed, paraffin-embedded biopsy specimens with known clinical outcomes were obtained retrospectively from a cohort of 48 patients. Hematoxylin and eosin-stained slides were used for the 2- and 3-tier dysplasia grading, while S100A7 for biomarker signature-based assessment was based on immunohistochemistry. Inter-observer variability was determined using Cohen’s kappa (
K
) statistic with Cox regression disease free survival analysis used to determine if any of the methods were a predictor of transformation to oral squamous cell carcinoma. Both the 2- and 3-tier dysplasia grading systems ranged from slight to substantial inter-observer agreement (
Kw
between 0.093 to 0.624), with neither system a good predictor of transformation to cancer (at least
P
=0.231; (
P
>>>0.05). In contrast, the S100A7 immunohistochemical signature-based grading system showed almost perfect inter-observer agreement (
Kw
=0.892) and was a good indicator of transformation to cancer (
P
=0.047 and 0.030). The inherent grading challenges with oral epithelial dysplasia grading systems and the lack of meaningful prediction of transformation to carcinoma highlights the significant need for a more objective, quantitative, and reproducible risk assessment tool such as the S100A7 immunohistochemical signature-based system.