Development and Validation of a classifier for discriminating Esophageal Carcinoma patients in early stage from advanced stage
International Journal of Genetics and Genomic Science · 2025
DOI: 10.58489/2836-2306/008 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Genetics and Genomic Science, 2025. Verfasst von Jia Yu,PhD und 10 weiteren Autor:innen.
Abstract (Originalsprache)
Purpose
This study aimed to construct a model which consisted of miRNAs and individual risk factors to accurately classify esophageal cancer (EC) patients in early-stage from the advanced stage ones. Methods: miRNAs of differential expression (DE-miRNAs) were analyzed between EC cancerous and para-cancerous tissues. The prognosis value of these miRNAs was investigated by Kaplan-Meier (KM) survival curves. The prognosis miRNAs and personal factors were selected by the least absolute shrinkage and selection operator (LASSO) logistic regression to build the classifier. The support vector machine (SVM) algorithm verified and evaluated the classifier’s accuracy and reliability. Furthermore, the nomogram model was established based on those factors for predicting the stage status of EC and was assessed using Harrell’s concordance index (C-index), the area under the time-dependent receiver operating characteristics curve (ROC-AUC), decision curve analysis (DCA), and calibration curve.
Results
Among 23 DE-miRNAs between the cancerous (n =111) and para-cancerous samples (n = 9), 18 were prognosis-related. hsa-let-7g-5p, hsa-miR-452-5p, hsa-miR-92b-5p, hsa-miR-139-3p, gender, M status, N status, T status, and postoperative radiation which were identified by LASSO. The results of SVM showed that the classifier which consisted of these 9 factors could classify the early and advanced stage of EC patients precisely and exhibit satisfied classification value with favorable reliability. These 9 factors were entered into the nomogram which had favorable discrimination, clinical usage, and calibration. hsa-let-7g-5p was the most significant one for predicting the probability of early stage, followed by N status, T status, M status, and hsa-miR-139-3p.
Conclusion
We established and validated a model to discriminate the EC patients from early-stage to late-stage for diagnosing.
Zitieren
APA
Yu,PhD, J., Yu, Y., MD, Yu, H., MD, Jia, Y., MD, Xu, L., MD, Bi*, S., & MD (2025). Development and Validation of a classifier for discriminating Esophageal Carcinoma patients in early stage from advanced stage. International Journal of Genetics and Genomic Science. https://doi.org/10.58489/2836-2306/008
Vancouver
Yu,PhD J, Yu Y, MD, Yu H, MD, Jia Y, et al. Development and Validation of a classifier for discriminating Esophageal Carcinoma patients in early stage from advanced stage. International Journal of Genetics and Genomic Science. 2025. doi:10.58489/2836-2306/008
BibTeX
@article{YuPhD2025,
author = {Yu,PhD, Jia and Yu, Yang and MD and Yu, Huixia and MD and Jia, Yanfeng and MD and Xu, Lianjie and MD and Bi*, Shufeng and MD},
title = {Development and Validation of a classifier for discriminating Esophageal Carcinoma patients in early stage from advanced stage},
journal = {International Journal of Genetics and Genomic Science},
year = {2025},
doi = {10.58489/2836-2306/008}
}
EUVIMED