Incorporation of DNA methylation profiling into the cytopathology laboratory

Abstract

Epigenetic modifications change gene expression without altering the DNA sequence. These modifications influence the accessibility of a gene to the transcription machinery and thus its ability to be expressed. Examples of epigenetic changes include DNA methylation, histone modifications, and noncoding RNA regulation.¹ Epigenetic studies have recently garnered significant attention in discussions related to tumor biomarker testing and translational research.^{1, 2}

DNA methylation is essential for DNA expression and repair, genome stability, cellular growth regulation, and overall normal development. Methylation testing can serve as a stable biomarker because of its relation to cell-fate differentiation and its ability to retain specific patterns throughout multiple cell divisions, akin to a form of genetic memory.^{1, 3} On a molecular basis, DNA methylation occurs when a methyl group is covalently added to the fifth carbon on a cytosine residue by the enzyme DNA methyltransferase.⁴ These cytosine residues are linked to guanine nucleotides, so called CpG nucleotides, that are unevenly distributed across our genome.⁴ DNA methylation generally results in silencing of gene expression by impairing the ability of transcriptional activators to bind DNA. For example, methylation of CpG islands in the promoter region of a gene can repress expression, whereas hypomethylation in the promoter region or hypermethylation in the gene body can enhance expression.⁴

Normal cells contain methylation profiles related to their cellular differentiation. In contrast, tumor methylomes can retain the molecular signature of their primary site while displaying aberrant DNA methylation patterns associated with oncogenesis. This characteristic can aid in determining the tumor's primary site of origin and help to predict its biologic activity.² Because of these discernable aberrant expression profiles, DNA methylation analysis is being increasingly recognized as a robust biomarker in the diagnostic laboratory and can help accurately classify tumors to help appropriately guide and monitor response to treatment.

Classification algorithms for tumors of the central nervous system (CNS) and sarcomas have been developed.^12-15 Distinct tumor types show unique and consistent methylation profiles that can be used as ancillary testing to improve tumor classification. These methylation classifiers are more advanced and routinely used in the clinical setting for CNS tumor classification. Methylation-based CNS tumor classifiers are being used in the clinical setting for diagnosing challenging cases. Moreover, methylation analysis has led to the identification of novel tumor types with distinct methylation profiles that are now incorporated into the World Health Organization classification of CNS tumors.¹⁶ Because the unique methylation profile is the defining feature of novel tumor types, some entities in the World Health Organization classification of CNS tumors can only be diagnosed with certainty after genome-wide methylation analysis. Similar efforts on the use of methylation profiling for tumor classification are ongoing for sarcomas and lymphomas.^{14, 17, 18}

The authors declared no conflicts of interest.