Uses of DNA methylation in cancer diagnosis and risk assessment

Over the last decade, molecular markers have emerged as promising tools for early cancer detection, patient management, and assessment of prognosis. Research into nucleic acid-based markers has blossomed as their ability to allow for robust analysis of clinical samples (tissue or body fluids) in a high-throughput fashion [1] has become recognized. Among the DNA-based approaches, DNA methylation offers several advantages over other types of molecular markers. Promoter hypermethylation of key regulatory genes in human cancers is a frequent event and is often associated with transcription silencing [2]. In addition, most of the common human tumors appear to have one or more hypermethylated loci when several methylation markers are examined together [3]. Furthermore, a specific pattern of hypermethylation for each type of human cancer is emerging from the rapidly growing list of cancer-related methylated genes, eventually permitting the identification of the tissue of origin of a particular neoplasm [3]. Notably, promoter methylation occurs early in tumorigenesis and may be identified even in preneoplastic lesions [4-6]. From a methodological standpoint, the identification of methylated DNA is easier than RNA manipulation followed by reverse transcription-PCR, since DNA harboring methylation is more stable. Likewise, detection of methylated DNA is particularly suited for clinical samples because methylation is a positive signal that is less likely to be masked through contaminant normal DNA. Thus, methylation markers provide sensitive detection in samples where tumor DNA is scarce or diluted by excess normal DNA.