Related Experiment Video
Updated: Jul 7, 2026
Investigating the Pathogenesis of MYH7 Mutation Gly823Glu in Familial Hypertrophic Cardiomyopathy using a Mouse Model
Published on: August 8, 2022
Identifying Malignant Transformation Risk of Dysplastic Oral Lesions Using the S100A7 Biomarker Signature-Based Assay
Audra Boehm1, Jayson Workman2, Krishna Kumar Kookal3
1Resident, Katz Department of Oral and Maxillofacial Surgery, The University of Texas Health and Science Center at Houston School of Dentistry, Houston, TX.
Background:
Despite the progression and acceptance of personalized tests in other disease indications, no clinically adopted, personalized molecular tests exist to manage and/or treat oral epithelial dysplasia. An emerging digital prognostic platform shows promise in estimating the risk of malignant transformation of dysplastic oral lesions.
Purpose:
This study aimed to evaluate a biomarker signature-based risk score (BRS) in identifying the risk of oral epithelial dysplasia transformation to oral squamous cell carcinoma (OSCC).
Study Design, Setting, And Sample:
A multi-center, retrospective cohort study was conducted on patients diagnosed with oral epithelial dysplasia between 2000 and 2021. Inclusion criteria were age ≥18 years, histopathological diagnosis of oral epithelial dysplasia, a minimum of 5-year clinical follow-up, absence of OSCC at initial biopsy, and available archived biopsy tissue. Exclusion criteria included concurrent diagnosis of oral epithelial dysplasia and OSCC or cancer transformation within 3 months of postbiopsy.
Predictor Variable:
The primary predictor variable was the BRS categories as normal or elevated.
Main Outcome Variable:
The main outcome was the transformation of oral epithelial dysplasia to OSCC within 5 years.
Covariates:
Covariates included smoking, alcohol use, age, sex, lesion site, biopsy type, and dysplasia grade.
Analyses:
The diagnostic parameters of the test-sensitivity, specificity, positive predictive value, and negative predictive value-were assessed for normal versus elevated BRS and patient subgroups. Furthermore, the area under the curve and Uno's C-index were estimated using the BRS. Kaplan-Meier curves and multivariable Cox regression models analyzed time to cancer transformation by dysplasia category.
Results:
The study set, consisting of 165 biopsies, was analyzed by a digital prognostic platform and showed a sensitivity of 1.00 (95% CI 1.00 to 1.00) and a specificity of 0.18 (95% CI 0.11 to 0.26). The PPV and NPV were 0.39 (95% CI 0.31 to 0.47) and 1.0 (95% CI 1.00 to 1.00), respectively. Kaplan-Meier curves showed a negative correlation between time to cancer transformation and higher dysplasia grade, with elevated Straticyte scores also showing this relationship (P = .0018). Receiver operator curves showed similar area under the curve as traditional dysplasia grading.
