RISK ASSESSMENT OF ORAL LEUKOPLAKIA BY DNA CONTENT USING FLOW CYTOMETRY

Abstract

Risk assessment for oral leukoplakia (OL) for malignant transformations (MT) has evolved as innovative strategies, considering that DNA content measurements have been providing consistent and reproducible predictive values using archived material. Such strategies are based on costly equipment and expertise, often inaccessible to underprivileged populations. This study aimed to investigate DNA content as a predictive marker of MT adapting image-based cytometry to a conventional flow cytometry (FC) context. Cases were selected from 878 biopsies, including mild, moderate, and severe dysplasia, and lesions without dysplasia. Twenty-six of these cases had undergone MT. FC was used to establish DNA content based on propidium iodide fluorescent labeling of nuclear suspensions. Spearman correlation was used for associations between dysplasia and ploidy (p <0.05). The predictive values of each marker were calculated from Kaplan-Meier and the Log-rank tests (p <0.05). Multiple logistic regression was used to enhance DNA content classification criteria for FC. Enhanced aneuploidy classification showed up to a 75% positive predictive value and a 95% negative predictive value, which are among the highest reported in the literature. In conclusion, conventional FC can be used to detect both high-risk and low-risk OL with high sensitivity and specificity. Risk assessment for oral leukoplakia (OL) for malignant transformations (MT) has evolved as innovative strategies, considering that DNA content measurements have been providing consistent and reproducible predictive values using archived material. Such strategies are based on costly equipment and expertise, often inaccessible to underprivileged populations. This study aimed to investigate DNA content as a predictive marker of MT adapting image-based cytometry to a conventional flow cytometry (FC) context. Cases were selected from 878 biopsies, including mild, moderate, and severe dysplasia, and lesions without dysplasia. Twenty-six of these cases had undergone MT. FC was used to establish DNA content based on propidium iodide fluorescent labeling of nuclear suspensions. Spearman correlation was used for associations between dysplasia and ploidy (p <0.05). The predictive values of each marker were calculated from Kaplan-Meier and the Log-rank tests (p <0.05). Multiple logistic regression was used to enhance DNA content classification criteria for FC. Enhanced aneuploidy classification showed up to a 75% positive predictive value and a 95% negative predictive value, which are among the highest reported in the literature. In conclusion, conventional FC can be used to detect both high-risk and low-risk OL with high sensitivity and specificity.