EUVIMEDEUVIMED entdecken
Multizentrische Studie

Machine learning model for recurrence-free survival in solitary resectable colorectal liver metastasis

Mingshuai Wang, Jianli Duan, Hongwei Wang, Yuhong Li, Baocai Xing

HPB · 2026-04 · 28 (4) · 497-507

DOI: 10.1016/j.hpb.2025.12.024 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift HPB, 2026-04. Verfasst von Mingshuai Wang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Hepatopancreatobiliary Surgery Department I, Peking Univ…. Hauptthemen laut MEDLINE: Machine Learning, Colorectal Neoplasms, Liver Neoplasms, Hepatectomy. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 36 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Centre for Future Health, University of York.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: Solitary colorectal liver metastasis (SCRLM) exhibits substantial heterogeneity in recurrence patterns after hepatic resection, yet individualized prediction tools for recurrence-free survival (RFS) are lacking. METHODS: In this multicenter retrospective study, 698 SCRLM patients undergoing hepatic resection were analyzed (training cohort: n=574; validation cohort: n=124). RFS was the primary endpoint. Three predictive models-random survival forest (RSF), Gradient Boosting Machine (GBM), and eXtreme Gradient Boosting (XGBoost)-were developed and compared. Model performance was assessed via concordance index (C-index), time-dependent area under the ROC curve (AUROC), and calibration plots. RESULTS: The XGBoost model achieved the best performance, with AUROCs of 0.93 and 0.87 at 1 year, and 0.89 and 0.86 at 3 years, in the training and validation cohorts, respectively. Compared with the modified Clinical Score (m-CS), the model demonstrated significantly higher discrimination at 1, 2, and 3 years (all P < 0.001), and also identified a subgroup of patients more likely to benefit from postoperative chemotherapy. A user-friendly online tool was developed for clinical application: https://scrlm.shinyapps.io/scrlmapp/. CONCLUSION: We developed and validated a machine learning-based model for SCRLM, enabling individualized recurrence risk prediction and guiding postoperative chemotherapy decisions. This approach may improve outcomes while reducing overtreatment.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningColorectal NeoplasmsLiver NeoplasmsHepatectomyDecision Support TechniquesHumansRetrospective StudiesFemaleMaleMiddle AgedAgedRisk AssessmentNeoplasm Recurrence, LocalPredictive Value of TestsRisk FactorsTime FactorsDisease-Free SurvivalTreatment Outcome

Autor:innen und Institutionen

  • Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Hepatopancreatobiliary Surgery Department I, Peking University Cancer Hospital & Institute, 52 Fucheng Road, Beijing, 100142,…
  • Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, 651 Dongfeng East Road, Guangzhou, China.…

Förderung

  • National Natural Science Foundation of China
  • Centre for Future Health, University of York

Zitieren

APA

Wang, M., Duan, J., Wang, H., Li, Y., & Xing, B. (2026). Machine learning model for recurrence-free survival in solitary resectable colorectal liver metastasis. HPB, 28(4), 497-507. https://doi.org/10.1016/j.hpb.2025.12.024

Vancouver

Wang M, Duan J, Wang H, Li Y, Xing B. Machine learning model for recurrence-free survival in solitary resectable colorectal liver metastasis. HPB. 2026;28(4):497-507. doi:10.1016/j.hpb.2025.12.024

BibTeX
@article{Wang2026,
  author = {Wang, Mingshuai and Duan, Jianli and Wang, Hongwei and Li, Yuhong and Xing, Baocai},
  title = {Machine learning model for recurrence-free survival in solitary resectable colorectal liver metastasis},
  journal = {HPB},
  year = {2026},
  volume = {28},
  number = {4},
  pages = {497--507},
  doi = {10.1016/j.hpb.2025.12.024}
}
Kostenlos starten →