Deconvolution of haematological cancer methylation patterns reveals a predominantly non-disease related proliferation signal and uncovers true disease associated methylation changes.

BACKGROUND: Cancers are associated with extensive reorganisation of epigenetic patterns, making identification of DNA methylation changes responsible for driving cancer development challenging. Here, we present a novel approach, integrative methylation mapping, which overcomes this, enabling identification of functionally relevant methylation-regulated genes in cancer. METHODS: Comparison of genome-wide DNA methylation across multiple B-lymphocyte derived malignant/normal samples (total n = 995), enabled delineation of changes related to normal or cancer cell functions. Chromatin structure profiling (SeSAMe) analysis delineated different properties characterising the different categories of methylation change and lentiviral based re-expression enabled functional assessment of identified candidate genes. RESULTS: This analysis determined that only 2-3% of DNA methylation changes in B-cell cancers are disease driven, with the overwhelming majority driven by normal processes, predominantly proliferation. Methylation changes associated with specific cancer or normal cell processes exhibited unique patterns of sequence context, chromatin structure and associated transcription factors. Furthermore, the low level of true disease-specific changes simplifies identification of functionally relevant methylation changes, illustrated here by identification and functional confirmation of SLC22A15 as a tumour suppressor in acute lymphoblastic leukaemia. CONCLUSIONS: This approach leads to a clearer understanding of the role of altered DNA methylation in haematological cancer, facilitates identification of cancer-relevant DNA methylation targeted genes and novel therapeutic targets.