Koos Boeve

19 General introduction and scope of this thesis key function in cell migration control, DNA repair and mitochondrial activity modulation [54]. Fas Associated Death Domein (FADD) is the protein encoded by the FADD gene and plays an important role in the apoptotic signalling, but has been related to cell cycle progression, innate immunity and autophagy more recently [48]. Recently, a study showed an increased rate of lymph node metastasis in head and neck squamous cell carcinoma patients with a high FADD expression [55]. Despite a clear association of expression of these genes such as CTTN/ cortactin, CCDN1/ cyclin D1 and FADD with lymph node status, and exploration of the underlying cellular processes involved in metastasis, none of these is currently implemented as a diagnostic tumour biomarker for lymph node status in early stage OSCC. One of the reasons is the lower predictive values of these tumour biomarkers for lymph node status compared to the SLNB procedure [30,56]. These lower predictive values might be due to the multistep character of metastasis with involvement of many different cellular processes (see above) [42]. Therefore predicting lymph node status using a single tumour marker only, most likely will not result in clinical suitable predictive values. Recently introduced laboratory techniques, such as DNA microarrays, enable the selection of genes and panels of genes using genome wide (epi)genetic approaches [56-58]. Despite that a gene-signature of 852 genes showed a high sensitivity of 86% and a NPV of 89% for detecting lymph node metastasis in a validation with early stage OSCC [56], such signatures for assessment of lymph node metastasis are not implemented to the clinical setting caused by the high costs, unfeasibility to use with formalin-fixed paraffin-embedded tissue and the availability of the high accurate SLNB procedure [59]. Moreover, comparing different signatures revealed hardly any similarity in selected genes and proteins between these expression signatures [60-63], that might display the heterogeneity in expression signatures among these tumours and the challenge of selecting genes or signatures with clinical applicable predictive values [57,63]. A promising method to select predictive molecular biomarkers for lymph node status is the analysis of the DNA hypermethylation status of certain cancer related genes. DNA methylation is an epigenetic process that regulates DNA transcription by adding a methyl group to a cytosine that precedes a guanine nucleotide, referred to as CpG sites [46]. A high density of these CpG sites are referred to as CpG islands and commonly observed within the regulatory regions of gene promotors [45]. Methylated CpG islands are associated with a lower gene expression (Figure 5) [45,64]. During cancer progression, gene promotor DNA methylation gradually increases (referred to as hypermethylation), while DNA methylation of repetitive sequences decreases (referred to as hypomethylation) (Figure 5) [65]. Abnormal expression of cancer-relevant genes by methylation has been reported to be at least as common as affecting transcription by genetic DNA alterations such as DNA amplification and mutations [66]. Because of hypermethylation is an independent mechanism of transcriptional regulation related to DNA sequence alterations, these mechanisms might