Combining genome-wide methylation discovery and validation, several novel prognostic DNA methylation markers were identified in neuroblastoma (NB). Starting points are a microarray based re-expression study after treatment with 5-aza-2'-deoxycytidine (DAC) and a next-generation sequencing experiment using an enrichment strategy towards methylated DNA (methyl-CpG-binding domain (MBD) capture). Both were performed on the same panel of eight NB cell lines. Applying a bioinformatics and text-mining-based approach on the re-expression data, 120 candidate genes were selected and tested using an initial high-throughput methylation-specific PCR (MSP) screen. The MBD-seq data were combined with public mRNA expression studies to enrich for potential prognostic biomarkers. Using a rank-based scoring system, a final selection of 43 candidates was made, which were then tested using MSP on 89 primary NB samples (in the following subgroups: LR-SURV, low-risk patients with long follow-up; HR-DOD, high-risk patients that die of disease; HR-SURV, high-risk patients with long follow-up). Finally, mRNA expression levels of seven DNA methylation biomarkers were determined. qPCR, quantitative PCR.
Decock et al. Genome Biology 2012 13:R95 doi:10.1186/gb-2012-13-10-r95