Related Experiment Video
Updated: Jan 7, 2026
06:46
Competing-Risk Nomogram for Predicting Cancer-Specific Survival in Multiple Primary Colorectal Cancer Patients after Surgery
Published on: September 27, 2024
616
Risk stratification for metachronous metastasis in colon cancer based on landscape ecological analysis
Yunxi Li1, Mingxin Chen2, Wei Wang1
1Department of Radiology, China-Japan Union Hospital, Jilin University, Changchun, China.
Scientific Reports
|December 20, 2025
Summary
This study introduces a new method using landscape pattern analysis to predict the risk of colon cancer spreading to other parts of the body. This helps stratify patients with stage II/III colon cancer for better risk assessment.
Area of Science:
- Oncology
- Radiology
- Spatial Analysis
Background:
- Metachronous metastasis is a significant concern in stage II/III colon cancer patients post-surgery.
- Effective risk stratification tools are crucial for personalized treatment strategies.
Purpose of the Study:
- To develop and validate a radiological risk stratification feature for predicting metachronous metastasis in colon cancer.
- To explore the utility of ecological spatial characteristics derived from landscape pattern analysis.
Main Methods:
- Retrospective analysis of stage II/III colon cancer patients (n=93) who underwent curative surgery.
- Landscape pattern analysis to extract tumor habitat spatial characteristics (Class-level and Landscape-level).
- Univariate and multivariate COX regression, construction of a landscape score, and time-dependent ROC curve analysis (AUC) for model evaluation.
Main Results:
- GYRATE-MD and CIRCLE-MD were significantly associated with metachronous distant metastasis (P < 0.05).
- The COX regression prediction model achieved an AUC of 0.78 in both training and validation sets at two years post-surgery.
- The model demonstrated effectiveness in risk stratification for metachronous metastasis.
Conclusions:
- Ecological spatial characteristics derived from landscape pattern analysis can serve as an effective radiological risk stratification feature for metachronous metastasis in stage II/III colon cancer.
- The developed prediction model aids in identifying patients at higher risk, facilitating tailored follow-up and treatment.
