Evaluation of DNA Methylation in TAC1, SOX17, and RASSF1A for the Early Diagnosis of Lung Cancer

Background: Low-Dose Computed Tomography (LDCT) is commonly used to detect pulmonary nodules; however, it may also contribute to overdiagnosis. DNA methylation shows promise as an approach to discriminate early-stage lung cancer (LC) patients from individuals with benign nodules (hereafter referred to as benign pulmonary nodule group) and healthy individuals. Methods: This study investigated the performance of DNA methylation in three genes ( TAC1, SOX17 , and RASSF1A ) in plasma-derived cell-free DNA (cfDNA) for discriminating LC patients from benign and healthy individuals (collectively referred to as benign/healthy individuals). We enrolled 149 LC patients (96 with early-stage stage IA and 53 with advanced-stage non-IA), 54 benign pulmonary nodule group, and 75 healthy individuals. Results: Methylation-positive rates for all three genes were significantly higher in LC patients compared to benign/healthy individuals. A combined three-gene model based on the ΔCt values of the three genes demonstrated robust diagnostic performance, achieving a sensitivity of 97.7%, specificity of 96.6%, and an area under the curve (AUC) of 0.99 for discriminating LC patients from benign/healthy individuals. Furthermore, another combined three-gene model based on the ΔCt values of the same genes showed high diagnostic performance for discriminating IA-stage LC patients from benign/healthy individuals, with a sensitivity of 96.9%, specificity of 88.54%, and AUC of 0.95. Conclusion: This study highlights the robust diagnostic value of a combined three-gene ( TAC1, SOX17 , and RASSF1A ) methylation model for detecting LC, including early-stage disease, offering high sensitivity and specificity. Keywords: DNA methylation, diagnosis, lung cancer, early-stage