62 Chapter 3 DISCUSSION In this review, we provide an overview of the reported DNA methylation markers for the detection of EC in minimally invasive specimens. EC-specific DNA methylation markers could be clinically relevant to guide the detection of EC in women presenting with postmenopausal bleeding, during cancer screening programs, and potentially in women with Lynch syndrome. We selected 15 DNA methylation markers with high potential that require further validation before drawing conclusions on their diagnostic accuracy in a clinical setting. The findings of studies included in this systematic review indicate the feasibility of EC detection in cytological specimens. Shedding of malignant endometrial cells into the cervical tract was already demonstrated by their detection in cervical scrapes by cytology (14). The use of sensitive DNA methylation markers broadly expands the use of cervical scrapes. Huang et al. (24) even reached a sensitivity and specificity of 91.8% and 95.5%, respectively, for EC detection when testing for a three-gene methylation marker panel in cervical scrapes. In addition to cervical scrapes, also endometrial brushes, vaginal swabs, and vaginal tampons seem promising sources of methylated DNA for EC detection. Wentzensen et al. (27) investigated the diagnostic potential of 8 genes frequently hypermethylated in EC tissue in DNA extracted from endometrial brushes. They validated their candidate markers in endometrial brush material with a combined AUC of 0.85. In a follow-up study, Bakkum-Gamez et al. (25) showed similar results in DNA extracted from vaginal tampons for all candidate genes, except for SOX1 due to technical issues. BakkumGamez et al. (25) also tested DNA methylation markers for EC detection established by others. Among these, RASSF1 methylation showed the second-highest AUC value of 0.75 in vaginal tampon specimens as compared to an AUC value of 0.82 found in paired endometrial brushes. Remarkably, for HTR1B the opposite effect was found, with an AUC value of 0.68 in endometrial brushes and 0.82 in tampons. Differences in marker performance may partly be explained by the use of different sample types. While endometrial brush samples are physician-taken and enable sampling of a wide area of the endometrial surface (31), vaginal tampons are self-collected specimens that indirectly obtain endometrial material. The varying presence of methylated background DNA in different sample types may also affect marker performance (32). The pioneering discovery studies included in this review all employed endometrial tissue samples to discover novel DNA methylation markers, of which the majority also
RkJQdWJsaXNoZXIy MjY0ODMw