ABSTRACT
Objective@#The aim of this study was to establish criteria that would indicate whether fertility preservation therapy would likely be safe for patients aged 40 years or less with endometrioid endometrial cancer based on their DNA methylation profile. @*Methods@#Forty-nine fresh-frozen tissue samples from patients with endometrial cancer from an initial cohort and 31 formalin-fixed paraffin-embedded tissue samples from a second cohort were subjected to genome-wide DNA methylation analysis using the Infinium MethylationEPIC BeadChip. @*Results@#Epigenomic clustering of early-onset endometrial cancer was correlated with the widely used recurrence risk classification. Genes showing differences in DNA methylation levels between the low-recurrence-risk category and intermediate- and high-risk categories were accumulated in pathways related to fibroblast growth factor and nuclear factor-κB signaling. DNA hypomethylation and overexpression of ZBTB38 were frequently observed in the low-risk category. Eight hundred thirty-one marker CpG probes showed area under the curve values of >0.7 on the receiver operating characteristic curve for discrimination of patients belonging to the low-risk category. By combining marker CpG sites, seven panels for placing patients into the low-risk category with 91.3% or more sensitivity and specificity in both the initial and second cohorts were established. @*Conclusions@#DNA methylation diagnostics criteria using up to 6 of 8 CpG sites for LPP, FOXO1, RNF4, EXOC6B, CCPG1, RREB1 and ZBTB38 may be applicable to recurrence risk estimation for patients aged 40 years or less with endometrial cancer, regardless of tumor cell content, even if formalin-fixed paraffin-embedded biopsy or curettage materials are used.
ABSTRACT
Uterine tumors are the most common type of gynecologic neoplasm. Uterine leiomyosarcoma (LMS) is rare, accounting for 2% to 5% of tumors of the uterine body. Uterine LMS develops more often in the muscle tissue layer of the uterine body than in the uterine cervix. The development of gynecologic tumors is often correlated with female hormone secretion; however, the development of uterine LMS is not substantially correlated with hormonal conditions, and the risk factors are not yet known. Radiographic evaluation combined with PET/CT can be useless in the diagnosis and surveillance of uterine LMS. Importantly, a diagnostic biomarker, which distinguishes malignant LMS and benign tumor leiomyoma (LMA) is yet to be established. Accordingly, it is necessary to analyze risk factors associated with uterine LMS in order to establish a method of treatment. LMP2-deficient mice spontaneously develop uterine LMS, with a disease prevalence of ∼40% by 14 months of age. It is therefore of interest whether human uterine LMS shows a loss of LMP2 expression. We found LMP2 expression is absent in human LMS, but present in human LMA. Therefore, defective LMP2 expression may be one of the risk factors for LMS. LMP2 is potentially a diagnostic biomarker for uterine LMS, and gene therapy with LMP2-encording DNA may be a new therapeutic approach.