Construction and validation of an IHC-based risk model predicting postoperative disease- free survival and overall survival in Chinese endometrial cancer patients

Abstract

Background: The objective of this study was to characterize the expression features base on three genes, including DUSP1, BUB1 and MCM7, and investigate their clinical significance in Chinese cohort with endometrial carcinoma. Materials and methods. Immunohistochemistry (IHC) was performed for assessing the expression of DUSP1, BUB1, and MCM7 in these patients. The associations between the expressions of three genes and the clinicopathological features were analyzed. The risk model was finally validated in TCGA cohort. Results: The expressions of DUSP1, BUB1, and MCM7 were associated with clinicopathological features (P＜0.05 for all). Patients with high expression level of DUSP1 and low expression level of BUB1 and MCM7 had better prognosis. The construction combining these three genes is an independent risk factor and had a high accuracy in predicting overall survival (OS, AUC=0.923) and disease-free survival (DFS, AUC=0.963). The three-gene risk signature was validated in TCGA-EC cohort, and the results showed a great precision for predicting OS and DFS (both AUC>0.7) in an external group. Conclusions: Our study revealed a three-gene prognostic signature, including DUSP1, BUB1, and MCM7, and its related nomogram could guide personalized strategy, providing a basis for the molecular analysis in EC, thereby improving their OS and DFS.