Objectives: DNA methylation has been shown to be a promising cancer biomarker. The aim of this study was to evaluate DNA methylation of three transcription factors, sex-determining region Y-box 1 (SOX1), paired box gene 1 (PAX1), and zinc-finger 582 (ZNF582), in detecting oral squamous cell carcinoma (OSCC). Materials and methods: A case-control study was conducted at Taipei Medical University Hospital in Taiwan with 31 cases of various oral cavity squamous cell carcinomas and 40 controls. Questionnaire data assessing environmental exposure, such as alcohol consumption, cigarette smoking, and betel nut chewing, were obtained from each participant. DNA from oral swabs were analyzed for methylation using quantitative methylation polymerase chain reaction with TaqMan probes. Methylation status was determined using a methylation index. Results: Methylation levels of SOX1, PAX1, and ZNF582 were significantly higher in cancer patients (p = 0.02, p = 0.02, and p = 0.03, respectively). Patients with highly methylated SOX1, PAX1, and ZNF582 had an increased cancer risk with odds ratios (ORs) of 16.50 (95 % CI = 2.85-96.65), 60.57 (95 % CI = 5.85-629.94), and 5.07 (95 % CI = 1.08-23.76), respectively. Area under the curve (AUC) values were 0.85, 0.78, and 0.78 for PAX1, SOX1, and ZNF582, respectively. When stratified based on environmental exposure, the AUC of PAX1 methylation (PAX1m) was 0.94 in environmental exposure-naïve subjects and 0.85 for SOX1 methylation in subjects who chewed betel nut. In general, the sensitivity and specificity of PAX1m were 87 and 80 % for OSCC detection. The sensitivity of PAX1m in subjects who chewed betel nut was 83 %, with a specificity of 75 %. Conclusions: Testing PAX1 DNA methylation using oral swabs is a promising method for oral cancer detection. Combined assessments regarding betel nut consumption and DNA methylation can improve OSCC screening. Clinical relevance: The double E (environmental and epigenetic) assessment is a potential strategy in OSCC screening.
ASJC Scopus subject areas