65 DNA methylation markers for endometrial cancer external validation could have implications for the reproducibility of our final marker selection. It is also worth noting that none of the studies performed a separate DNA methylation marker discovery screen on cytological material. Markers were either discovered using endometrial or cervical cancer tissue, or EC tissue data downloaded from The Cancer Genome Atlas Program. Although studies showed variable risk of bias scores, they were not excluded on the basis of their risk of bias. A high risk of bias was frequently introduced during patient selection due to their case-control design. Moreover, cases and controls were often not age-matched. Thresholds used to handle DNA methylation levels were not pre-specified in many studies or specified according to another cancer or sample type, which also introduced bias in the majority of the studies. CONCLUSIONS We selected the individual genes ADCYAP1, ASCL2, BHLHE22, CDH13, CDO1, CELF4, GALR1, HAND2, HS3ST2, HTR1B, MAGI2, MME, POU4F3, RASSF1, and ZNF662, with AUC values ranging from 0.80 to 0.96, as potential DNA methylation markers for the detection of EC using minimally invasive specimens. This approach could potentially guide the detection of EC in women presenting with postmenopausal bleeding, during cancer screening programs and potentially in women with Lynch syndrome. Validation in larger, prospective and unbiased cohorts is warranted to determine their true clinical diagnostic accuracy (45). FUTURE PERSPECTIVE Current literature indicates the feasibility of DNA methylation marker testing in minimally invasive samples for EC detection. Nevertheless, diagnostic methylation marker research is facing several challenges, with heterogeneity of study populations and the lack of standardized methylation assays as the main concerns. Instead of only contributing to the discovery of new diagnostic methylation markers, studies should also focus on the further validation of the methylation markers described so far. The performance of the selected DNA methylation markers with potential clinical value may be improved by combining DNA methylation markers with additional (epi)- genetic markers and using alternative sources of DNA. The most interesting and likely application of DNA methylation markers would be in diagnostics to guide two clinical problems: to discriminate the minority of women with underlying malignancy within the group of women presenting with postmenopausal bleeding symptoms, and to monitor women with increased EC risk. 3
RkJQdWJsaXNoZXIy MjY0ODMw