Variability in the measurement of the methylation status of lung cancer-related genes in bronchial secretions

Assessment of the methylation status of genes related to the development of lung cancer (LC) in bronchial secretions has been proposed as a biomarker for early detection. Several techniques are available to detect gene methylation, and the method chosen may have an effect on the results. A cross-sectional study was conducted in which the methylation status of DAPK, CDKN2A (p16) and RASSF1A genes in sputum and bronchial washing (BW) from subjects at risk for LC was analyzed. The methylation results of both samples were compared, considering BW as the reference. Results obtained by methylation-sensitive PCR (MSP) were validated by methylation-sensitive high-resolution melting (MS-HRM). The methylation results obtained in sputum and BW samples did not show statistically significant differences for any of the three genes analyzed in 65 subjects (McNemar test >0.05). Concordant results between sputum and BW were found in 40 patients for DAPK (61%), in 52 patients for p16 (80%) and in 63 patients for RASSF1 (97%). More methylated samples were found in BW, however, and sputum sensitivities and specificities for the identification of methylation status were 44 and 72% for DAPK gene, 21 and 94% for p16 and 100 and 98% for RASSF1A, respectively. When MSP results were validated by MS-HRM, DAPK and p16 gene samples methylated by MSP appeared to be unmethylated by MS-HRM. One sample showing methylation of RASSF1A gene also showed methylation when tested following MS-HRM procedure. Sputum and BW samples may be considered equally valid for the identification of methylated genes in bronchial secretions. The low sensitivity of sputum for the assessment of the methylation status of DAPK and p16 genes, however, suggests that the analysis of two or more sputum samples, or of a BW obtained semi-invasively, would be needed to attain higher reliability, together with the use of confirmatory techniques for positive results.