目的 探讨子宫内膜癌（EC）患者血浆细胞周期蛋白E1（CCNE1）、易洛魁人同源框2（IRX2）、异嗜性和多嗜性逆转录病毒受体1（XPR1）基因甲基化与临床病理特征及预后的关联性。 方法 选取2018年6月至2022年12月在医院进行手术治疗的EC患者120例（观察组），根据患者预后分为生存组（88例）、死亡组（32例）。另选取年龄相匹配的100例健康志愿者作为对照组。甲基化特异性实时定量聚合酶链反应检测血浆CCNE1、IRX2、XPR1基因甲基化水平；Kaplan-Meier法分析血浆CCNE1、IRX2、XPR1甲基化水平与EC患者预后的关系；Cox风险回归模型分析EC患者预后影响因素；受试者工作特征（ROC）曲线评估血浆CCNE1、IRX2、XPR1甲基化对EC患者预后不良的预测价值。 结果 相较于对照组，观察组血浆CCNE1、XPR1甲基化水平降低，IRX2甲基化水平升高（P < 0.05）。FIGO分期Ⅲ + Ⅳ期、组织学分级低分化、浸润肌壁深度 ≥ 1/2肌层、有淋巴结转移患者CCNE1、XPR1甲基化呈低表达，IRX2甲基化呈高表达（P < 0.05）。120例EC患者共随访5 ~ 36个月，中位随访时间为21个月，其中32例患者死亡，3年总生存率为73.33%（88/120）。血浆CCNE1、XPR1甲基化低表达患者3年总生存率低于其高表达组（χ² = 4.729、6.136，P = 0.030、0.013），血浆IRX2甲基化高表达患者3年总生存率低于其低表达组（χ² = 6.569，P = 0.010）。相较于生存组，死亡组CCNE1、XPR1甲基化水平降低，IRX2甲基化水平升高（P < 0.05）。IRX2高甲基化是EC患者预后不良的危险因素，CCNE1高甲基化、XPR1高甲基化则是保护因素（P < 0.05）。血浆CCNE1、IRX2、XPR1甲基化单独预测EC患者预后不良的AUC = 0.827、0.820、0.816，三者联合预测的AUC = 0.941，联合预测效能较优（Z = 3.427、2.610、3.068，P = 0.001、0.009、0.002）。 结论 EC患者血浆CCNE1、XPR1甲基化水平降低，IRX2甲基化水平升高，均与临床病理特征、预后不良密切相关，三者联合对EC患者预后不良具有较高的预测价值。

Objective To investigate the association between the methylation levels of plasma cyclin E1 （CCNE1）， iroquois homeobox protein 2 （IRX2）， and xenotropic and polytropic retrovirus receptor 1 （XPR1） genes， and the clinicopathological features and prognosis in patients with endometrial cancer （EC）. Methods A total of 120 patients with EC who underwent surgical treatment at the hospital between June 2018 and December 2022 （study group） were selected and divided into the survival group （88 cases） and the death group （32 cases） based on the patients' prognosis. Another 100 healthy volunteers of matching age were recruited as the control group. The methylation levels of plasma CCNE1， IRX2， and XPR1 genes were detected using methylation-specific real-time quantitative polymerase chain reaction. The relationship between the methylation levels of plasma CCNE1， IRX2， and XPR1 and the prognosis of EC patients was analyzed by the Kaplan-Meier method. The Cox risk regression model was employed to analyze the prognostic factors of EC patients. Receiver operating characteristic （ROC） curves were utilized to evaluate the predictive value of plasma CCNE1， IRX2， and XPR1 methylation for poor prognosis in patients with endometrial cancer. Results Compared with the control group， the methylation levels of plasma CCNE1 and XPR1 in the study group were lower， whereas the methylation level of IRX2 was higher （P < 0.05）. In patients with FIGO stage Ⅲ + Ⅳ， poorly differentiated histological grade， myometrial infiltration depth ≥ 1/2 myometrium， and lymph node metastasis， the methylation levels of CCNE1 and XPR1 were low， while the methylation level of IRX2 was high （P < 0.05）. A total of 120 EC patients were followed up for 5 - 36 months， with a median follow-up time of 21 months. Among them， 32 patients died， and the 3-year overall survival rate was 73.33% （88/120）. The 3-year overall survival rate of patients with low expression of plasma CCNE1 and XPR1 methylations was lower than that of the high-expression groups （χ² = 4.729， 6.136； P = 0.030， 0.013）， and the 3-year overall survival rate of patients with high expression of plasma IRX2 methylation was lower than that of the low-expression groups （χ² = 6.569； P = 0.010）. Compared with the survival group， the methylation levels of CCNE1 and XPR1 in the death group were lower， and the methylation level of IRX2 was higher （P < 0.05）. Hypermethylation of IRX2 was a risk factor for poor prognosis in patients with EC， while hypermethylation of CCNE1 and hypermethylation of XPR1 were protective factors （P < 0.05）. The AUC values for plasma CCNE1， IRX2， and XPR1 methylation as individual predictors of poor prognosis in EC patients were 0.827， 0.820， and 0.816， respectively； The AUC for the combined prediction was 0.941， indicating superior predictive performance （Z = 3.427， 2.610， 3.068； P = 0.001， 0.009， 0.002）. Conclusions Reduced levels of CCNE1 and XPR1 methylation， along with elevated levels of IRX2 methylation， in the plasma of endometrial cancer patients are closely associated with clinical and pathological characteristics as well as poor prognosis. The combination of these three markers holds high predictive value for poor prognosis in endometrial cancer patients.