Koos Boeve

157 Methylation biomarkers for the detection of tumor DNA in saliva further improve the diagnostic potential of detecting OSCC in saliva cells (100%) in an age matched analysis. Due to the limited size of our pilot group, the diagnostic potential of these biomarkers must be validated on a larger independent and prospective cohort. Similarly, a saliva database containing samples of 5-year-follow-up, pre- and post-operative as well as pre-malignant cases should be constructed for prospective studies and to assess the background methylation caused by non-tumor cells. The use of molecular markers for the early detection and monitoring treatment response and disease progression using body fluids like saliva, sputum, plasma, cerebrospinal fluid and urine [10,33] has limited clinical utility today [34] but has great promise to contribute to improved clinical care by early detection of OSCC or monitoring the treatment response. Since DNA methylation is important in carcinogenesis, occurs early in tumorigenesis and is detectable in patient saliva [35], DNA methylation markers could contribute to the early detection of local recurrences of OSCC. Additionally, aberrations in DNA methylation arise early in tumorigenesis [14]. Therefore, our results warrants further analysis in larger independent cohorts. Several methylation markers for the detection of cells in saliva of patients with OSCC were reported previous ( EDNRB, HOXA9, NID2 and TIMP3 ) [19,20,24]. As a comparison to our new markers, we analyzed these markers in parallel on the same samples using QMSP. In our cohort, methylation of HOXA9 and NID2 was detected in all saliva cells of health individuals. Methylation of EDNRB was observed in 50% of these saliva, but the difference between saliva of OSCC patients and of age-matched controls was not significant. An explanation for the frequent methylation in normal control, especially in the saliva of the “older” age- matched “healthy”(non-cancer) saliva cells is that methylation of many genomic sequences has been reported to increase with age [36]. Therefore, methylation of these reported genes are not suitable as methylation markers in the “older” age-matched OSCC cohort. Note that the four markers selected from literature showed significant methylation in our age-matched samples from normal saliva, which can also be an explanation why these markers were not present in our selected list of 2276 highest ranking methylation cores from the MethylCap-seq analysis of OSCC tissue samples. With thegenome-widemethylationanalysis, within themethylomeofmillions ofmethylated DNA fragments in 12 OSCC samples, we eventually identified and validated three of the six new candidate markers for OSCC (C11orf85, KCNA5 and SIPA1). The pathophysiology of the novel genes related to OSCC or other types of cancer is not yet fully clarified. KCNA5 is a member of the voltage-gated potassium (K v ) channel subfamily A [37]. In Ewing sarcoma cells methylation of the KCNA5 promoter region is correlated with cell survival and