Mixed cell adenocarcinoma of the endometrium: a population-based study

Abstract

Abstract Background Mixed cell adenocarcinoma of endometrium (MCAE) is a rare histological subtype of endometrial carcinoma. However, there are limited studies investigating the prognostic factors of MCAE. Therefore, the present study aimed to explore the independent predictors of MCEA and build a prognostic nomogram for the overall survival (OS) of MCAE patients. Methods Data on MCAE patients were screened from the Surveillance, Epidemiology, and End Results (SEER) database. The MCAE prognosis in patients with Federation of Gynecology and Obstetrics (FIGO) stage I/II and stage III/IV was explored. Next, the patients in stage I/II and stage III/IV were assigned into training and validation cohorts in the ratio of 7:3, respectively. Univariate and multivariate Cox regression was used to determine the independent risk factors for the OS, and nomograms were constructed based on these factors. The concordance index (C-index) and calibration plots were established to assess the discrimination ability and accuracy of the nomograms. Finally, the decision curve analysis (DCA), net reclassification index (NRI), and integrated discrimination improvement (IDI) were calculated to compare performance between the nomogram and the traditional FIGO stage model. Results A total of 6650 patients were enrolled in our study, among which 4635 patients were in stage I/II and 2015 in stage III/IV. Age, marital status, race, FIGO stage, and surgery were identified as the independent predictors for patients in FIGO stage I/II, while age, marital status, race, FIGO stage, surgery, radiation, and chemotherapy for patients in stage III/IV (p < 0.05). NomogramA for stage I/II and nomogramB for stage III/IV were constructed based on the independent predictors. The C-indexes in nomogramA were 0.730 (95%CI: 0.711 ~ 0.749) and 0.715 (95%CI: 0.685 ~ 0.745) for the training and validation cohorts, and 0.720 (95%CI: 0.702 ~ 0.738) and 0.724 (95%CI: 0.697 ~ 0.751) in nomogramB, respectively. Calibration curves for both nomograms followed the diagonal line, implying an excellent consistency between the predicted and actual survival. In addition, the DCA revealed that nomograms had a higher clinical benefit compared to the FIGO stages. In stage I/II patients, the NRI and IDI for 3, 5, and 10 years were more than 30% and 10%, respectively (p < 0.001). In stage III/IV patients, the NRI and IDI values were more than 20% and 6%, respectively (p < 0.001). Both NRI and IDI values were further tested in the validation cohort. Overall, the predictive performance of nomogramA and nomogramB was higher compared to the traditional FIGO stage model. Conclusions The predictive nomograms for MCAE patients in FIGO stage I/II and stage III/IV constructed and validated in this study could offer an effective tool for evaluating the treatment outcome in MCAE patients.