A decade of DNA methylation profiling in cancer: What have we learned?

Investigators have studied DNA methylation in tumor cells for nearly 50 years. Differences in gene-specific methylation between cancer cells and their normal counterparts were described more than 30 years ago. From early techniques that measured overall DNA methylation levels to enzyme-associated techniques that interrogated methylation at a single CpG dinucleotide to present day assays that catalogue the methylation of every cytosine in the genome, technical advancement progressively has brought increasing clarity to our understanding of the complex epigenomes of normal and neoplastic cells. Over the past 10 years we have been witness to an explosion of investigation into the epigenetic basis of cancer, and application of the powerful genome-wide DNA methylation profiling techniques to be reviewed have yielded critical insights into the organization of the cancer methylome with its broad regions of hypomethylation and foci of hypermethylation resulting in critical differences in gene expression and chromosomal stability compared to normal cells. These insights, in turn, have prompted novel, testable hypotheses, to be discussed, pertaining to fundamental aspects of cancer biology including the potential stem cell/progenitor cell origins of cancer and the plasticity of gene expression that may underlie tumor heterogeneity and tumor progression. Finally to be discussed is the growing portfolio of epigenetic tools being provided by modern methylome profiling analyses. These biomarker tools are complementing and extending the current genomic tests that are improving cancer diagnosis and that increasingly will facilitate highly individualized cancer treatment in the upcoming decade.