Machine learning analysis for recurrence prediction of colorectal cancer liver metastases
European Journal of Surgical Oncology · 2026-01 · 52 (1) · 111168
DOI: 10.1016/j.ejso.2025.111168 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Journal of Surgical Oncology, 2026-01. Verfasst von Taijun Yi und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of General Surgery (Hepatobiliary, Pancreatic and Splenic Surgery), The Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guan…. Hauptthemen laut MEDLINE: Machine Learning, Liver Neoplasms, Colorectal Neoplasms, Neoplasm Recurrence, Local. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 34 Literaturangaben. Gefördert von: National Key Clinical Specialty Discipline Construction Program of China.
Abstract (Originalsprache)
OBJECTIVE: Recurrence of Colorectal Cancer Liver Metastases (CRLM) remains a major challenge impacting therapeutic outcomes and patient prognosis. This study aimed to evaluate the predictive value of liver metastases characteristics and other indicators for CRLM recurrence using machine learning models. METHODS: A total of 524 CRLM patients who underwent radical resection were stratified into recurrence and non-recurrence groups. Ten machine learning (ML) algorithms were employed to assess comprehensive predictive value of key variables selected via LASSO regression. SHapley Additive exPlanations (SHAP) were used for model interpretation and visualization. Additionally, a nomogram prediction model was developed based on the key variables selected. RESULTS: Factors including liver metastases characteristics were associated with CRLM recurrence risk. Machine learning algorithms demonstrated robust predictive performs. Among the ten models, the Logistic Regression (LR) exhibited the best discriminative ability (AUC = 0.762, F1_Score = 0.72, Accuracy = 0.686, MCC = 0.364). The constructed nomogram model provided moderate recurrence discrimination (1 year: AUC (95 %CI) = 0.692 (0.642-0.742); 3 years: AUC (95 %CI) = 0.732 (0.681-0.782); 5 years: AUC (95 %CI) = 0.771 (0.705-0.838)). We further developed machine learning models specifically for liver-only recurrence of CRLM, among which Naive Bayes (NB) showed the best performance (AUC = 0.767), and constructed an effective prediction nomogram. CONCLUSION: This study identified risk factors for overall recurrence and liver-only recurrence after radical resection for CRLM based on liver metastases characteristics and clinical-pathological factors, establishing predictive models that potentially aid in developing individualized treatment strategies.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of General Surgery (Hepatobiliary, Pancreatic and Splenic Surgery), The Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong Province, 510655, China; Biomedical Innovation Center, The Sixth…
- Department of General Surgery, Tongren Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, 200336, China.
Förderung
- National Key Clinical Specialty Discipline Construction Program of China
Zitieren
APA
Yi, T., Liang, Y., Li, X., Lin, Z., Yu, J., Wang, W., Huang, X., Wan, Y., Chen, Z., & Li, G. (2026). Machine learning analysis for recurrence prediction of colorectal cancer liver metastases. European Journal of Surgical Oncology, 52(1), 111168. https://doi.org/10.1016/j.ejso.2025.111168
Vancouver
Yi T, Liang Y, Li X, Lin Z, Yu J, Wang W, et al. Machine learning analysis for recurrence prediction of colorectal cancer liver metastases. European Journal of Surgical Oncology. 2026;52(1):111168. doi:10.1016/j.ejso.2025.111168
BibTeX
@article{Yi2026,
author = {Yi, Taijun and Liang, Yongling and Li, Xingyu and Lin, Zejin and Yu, Jiandong and Wang, Wandi and Huang, Xiaoyue and Wan, Yunle and Chen, Zhiping and Li, Guolin},
title = {Machine learning analysis for recurrence prediction of colorectal cancer liver metastases},
journal = {European Journal of Surgical Oncology},
year = {2026},
volume = {52},
number = {1},
pages = {111168},
doi = {10.1016/j.ejso.2025.111168}
}
EUVIMED