Abstract
PURPOSE: Novel molecular predictive biomarkers for chemotherapy have been screened and validated in non-small cell lung cancer (NSCLC). However, there was no report on the correlation of genome-wide DNA methylation with survival benefit from chemotherapy in NSCLC. METHODS: A sandwich enzyme-linked immunosorbent assay (ELISA) method was first established, optimized and validated. A total of 191 NSCLC samples were analyzed using the sandwich ELISA for the association between the relative genome-wide DNA methylation level and the survival outcomes from chemotherapy. RESULTS: The analytical performance of the sandwich ELISA method was satisfying and suitable for analysis. Using the sandwich ELISA method, we found that the genome-wide DNA methylation level in NSCLC cancer tissues was significantly lower than that in adjacent normal tissues, which further validated the assay. We found that there was no significant correlation between genome-wide DNA methylation level and patients' histology, stage and progression free survivals. However, in patients with high methylation level, those without chemotherapy had significantly better overall survival than those receiving chemotherapy. In patients receiving chemotherapy, those with low genome-wide DNA methylation level had significantly better overall survival than those with relatively high DNA methylation level. CONCLUSIONS: Genome-wide DNA hypomethylation as a sign of genomic instability may predict overall survival benefit from chemotherapy in NSCLC.