LASSO-derived nomogram prediction model for lymph node metastasis in colorectal cancer: a retrospective analysis
PeerJ · 2025
DOI: 10.7717/peerj.19148 ↗Kurz gefasst
Fachartikel in der Zeitschrift PeerJ, 2025. Verfasst von Xiyun Quan und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pathology, Zhuzhou Hospital Affiliated to Xiangya School of Medicine, Central South University, Zhuzhou, Hunan, China. Hauptthemen laut MEDLINE: Nomograms, Colorectal Neoplasms, Lymphatic Metastasis. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Objective
This study aims to develop a prediction model for lymph node metastasis (LNM) in colorectal cancer (CRC) patients using common clinicopathologic data and a nomogram. The model seeks to uncover correlations between LNM and clinical indicators, providing an effective tool to identify high-risk patients, aiding clinical decision-making, and enhancing patient prognosis.
Methods
We conducted a retrospective analysis of CRC patients diagnosed between January 2021 and December 2023 at Zhuzhou Hospital Affiliated to Xiangya School of Medicine, Central South University. Risk predictors for LNM were identified through comparative analysis and Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression. Nomograms were then utilized to predict the probability of metastasis, and their performance was assessed using calibration curves, receiver operating characteristic (ROC) curves, and decision curve analysis.
Results
The study comprised 869 CRC patients, with 435 cases allocated to the training set and 434 cases to the validation set. First, 12 potential risk factors for LNM in CRC patients were identified through comparative analysis in the training set. Next, nine independent predictors (T stage, vascular tumor thrombus, PMS2, MSH2, KRAS, BRAF, PIK3CA, leukocyte, and neutrophil) of LNM occurrence were refined using LASSO regression and multivariate logistic regression models. Subsequently, a clinical nomogram was developed based on these independent predictors of LNM. The nomogram exhibited a C-index of 0.751 (95% CI [0.728–0.774]), indicating its robust predictive value, which was further validated in the independent validation set.
Conclusion
T stage, vascular tumor thrombus, PMS2, MSH2, KRAS, BRAF, and neutrophil emerged as significant risk factors for LNM in CRC, while leukocytes appeared to be protective. These findings emphasize the importance of comprehensive risk assessment and personalized therapeutic strategies in CRC management.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Pathology, Zhuzhou Hospital Affiliated to Xiangya School of Medicine, Central South University, Zhuzhou, Hunan, China.
- Neurosurgery, Zhuzhou Hospital Affiliated to Xiangya School of Medicine, Central South University, Zhuzhou, Hunan, China.
- Health Management Center, Zhuzhou Hospital Affiliated to Xiangya School of Medicine, Central South University, Zhuzhou, Hunan, China.
Zitieren
APA
Quan, X., Deng, Y., Liu, Z., Gao, Z., Yi, H., & Li, M. (2025). LASSO-derived nomogram prediction model for lymph node metastasis in colorectal cancer: a retrospective analysis. PeerJ. https://doi.org/10.7717/peerj.19148
Vancouver
Quan X, Deng Y, Liu Z, Gao Z, Yi H, Li M. LASSO-derived nomogram prediction model for lymph node metastasis in colorectal cancer: a retrospective analysis. PeerJ. 2025. doi:10.7717/peerj.19148
BibTeX
@article{Quan2025,
author = {Quan, Xiyun and Deng, Yi and Liu, Zhimin and Gao, Zhenqin and Yi, Huimei and Li, Ming},
title = {LASSO-derived nomogram prediction model for lymph node metastasis in colorectal cancer: a retrospective analysis},
journal = {PeerJ},
year = {2025},
doi = {10.7717/peerj.19148}
}
EUVIMED