The genetic and epigenetic alterations that underlie cancer pathogenesis are rapidly being identified. This provides novel insights in tumor biology as well as in potential cancer biomarkers. The somatic mutations in cancer genes that have been implemented in clinical practice are well defined and very specific. For epigenetic alterations, and more specifically aberrant methylation of promoter CpG islands, evidence is emerging that these markers could be used for the early detection of cancer as well as prediction of prognosis and response to therapy. However, the exact location of biologically and clinically relevant hypermethylation has not been identified for the majority of methylation markers. The most widely used approaches to analyze DNA methylation are based on primer- and probe-based assays that provide information for a limited number of CpG dinucleotides and thus for only part of the information available in a given CpG island. Validation of the current data and implementation of hypermethylation markers in clinical practice require a more comprehensive and critical evaluation of DNA methylation and limitations of the techniques currently used in methylation marker research. Here, we discuss the emerging evidence on the importance of the location of CpG dinucleotide hypermethylation in relation to gene expression and associations with clinicopathologic characteristics in cancer.