Identification of a MicroRNA Signature Associated With Lymph Node Metastasis in Endometrial Endometrioid Cancer

Kaiyou Fu; Yanrui Li; Jianyuan Song; Wangyu Cai; Wei Wu; Xiaohang Ye; Jian Xu

doi:10.3389/fgene.2021.650102

Identification of a MicroRNA Signature Associated With Lymph Node Metastasis in Endometrial Endometrioid Cancer

Front Genet. 2021 Apr 15:12:650102. doi: 10.3389/fgene.2021.650102. eCollection 2021.

Authors

Kaiyou Fu^{1

2}, Yanrui Li³, Jianyuan Song⁴, Wangyu Cai⁴, Wei Wu⁴, Xiaohang Ye¹, Jian Xu⁴

Affiliations

¹ School of Medicine, Zhejiang University, Hangzhou, China.
² Women's hospital, School of Medicine, Zhejiang University, Hangzhou, China.
³ School of Control Science and Engineering, Zhejiang University, Hangzhou, China.
⁴ Fourth Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China.

Abstract

Background: Lymph node metastasis (LNM) is an important prognostic factor in endometrial cancer. Anomalous microRNAs (miRNAs) are associated with cell functions and are becoming a powerful tool to characterize malignant transformation and metastasis. The aim of this study was to construct a miRNA signature to predict LNM in endometrial endometrioid carcinoma (EEC).

Method: Candidate target miRNAs related to LNM in EEC were screened by three methods including differentially expressed miRNAs (DEmiRs), weighted gene co-expression network analysis (WGCNA), and decision tree algorithms. Samples were randomly divided into the training and validation cohorts. A miRNA signature was built using a logistic regression model and was evaluated by the area under the curve (AUC) of receiver operating characteristic curve (ROC) and decision curve analysis (DCA). We also conducted pathway enrichment analysis and miRNA-gene regulatory network to look for potential genes and pathways engaged in LNM progression. Survival analysis was performed, and the miRNAs were tested whether they expressed differently in another independent GEO database.

Result: Thirty-one candidate miRNAs were screened and a final 15-miRNA signature was constructed by logistic regression. The model showed good calibration in the training and validation cohorts, with AUC of 0.824 (95% CI, 0.739-0.912) and 0.821 (95% CI, 0.691-0.925), respectively. The DCA demonstrated the miRNA signature was clinically useful. Hub miRNAs in signature seemed to contribute to EEC progression via mitotic cell cycle, cellular protein modification process, and molecular function. MiR-34c was statistically significant in survival that a higher expression of miR-34c indicated a higher survival time. MiR-34c-3p, miR-34c-5p, and miR-34b-5p were expressed differentially in GSE75968.

Conclusion: The miRNA signature could work as a noninvasive method to detect LNM in EEC with a high prediction accuracy. In addition, miR-34c cluster may be a key biomarker referring LNM in endometrial cancer.

Keywords: TCGA; endometrial cancer; lymph node metastasis; miRNA expression profile; molecular biomarker.