Identification of aberrant methylation regions in neuroblastoma by screening of tissue-specific differentially methylated regions

Background: The identification of tissue-specific differentially methylated regions (tDMRs) is key to our understanding of mammalian development. Research has indicated that tDMRs are aberrantly methylated in cancer and may affect the oncogenic process. Procedure: We used the MassARRAY EpiTYPER system to determine the quantitative methylation levels of seven neuroblastomas (NBs) and two control adrenal medullas at 12 conserved tDMRs. A second sample set of 19 NBs was also analyzed. Statistical analysis was carried out to determine the relationship of the quantitative methylation levels to other prognostic factors in these sample sets. Results: Screening of 12 tDMRs revealed 2 genomic regions (SLC16A5 and ZNF206) with frequent aberrant methylation patterns in NB. The methylation levels of SLC16A5 and ZNF206 were low compared to the control adrenal medullas. The SLC16A5 methylation level (cut-off point, 13.25%) was associated with age at diagnosis, disease stage, and Shimada classification but not with MYCN amplification. The ZNF206 methylation level (cut-off point, 68.80%) was associated with all of the prognostic factors analyzed. Although the methylation levels at these regions did not reach statistical significance in their association with prognosis in mono-variant analysis, patients with both hypomethylation of SLC16A5 and hypermethylation of ZNF206 had a significantly prolonged event-free survival, when these two variables were analyzed together. Conclusions: We demonstrated that two tDMRs frequently displayed altered methylation patterns in the NB genome, suggesting their distinct involvement in NB development/differentiation. The combined analysis of these two regions could serve as a diagnostic biomarker for poor clinical outcome.