Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.
World J Gastrointest Oncol. Dec 15, 2026; 18(12): 124081
Published online Dec 15, 2026. doi: 10.4251/wjgo.124081
Development and validation of a computed tomography-based radiomics nomogram for predicting metachronous liver metastasis in colorectal cancer
Yang Li, Zi-Cheng Zhou, Heng Jiang, Mi Wang, Tian Tian, Xi-Ming Wang, Wei-Guo Zhang, Zhi-Hua Lu
Yang Li, Mi Wang, Tian Tian, Wei-Guo Zhang, Zhi-Hua Lu, Department of Radiology, The Fourth Affiliated Hospital of Soochow University, Suzhou Dushu Lake Hospital, Medical Center of Soochow University, Suzhou 215123, Jiangsu Province, China
Zi-Cheng Zhou, Suzhou Medical College of Soochow University, Suzhou 215123, Jiangsu Province, China
Heng Jiang, Department of Radiology, Changshu No. 1 People’s Hospital, Affiliated Changshu Hospital of Soochow University, Changshu 215500, Jiangsu Province, China
Xi-Ming Wang, Department of Radiology, The First Affiliated Hospital of Soochow University, Suzhou 215031, Jiangsu Province, China
Co-first authors: Yang Li and Zi-Cheng Zhou.
Co-corresponding authors: Wei-Guo Zhang and Zhi-Hua Lu.
Author contributions: Li Y, Zhou ZC, and Jiang H had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis; Li Y and Lu ZH were involved in the study concepts and design; all authors involved in the acquisition, analysis and interpretation of data; Wang XM and Zhang WG supervised the analysis; Li Y and Lu ZH involved in the draft of the manuscript; Li Y and Zhou ZC contributed equally to this study as co-first authors; Zhang WG and Lu ZH contributed equally as co-corresponding authors. All authors read, critically revised and approved the manuscript.
AI contribution statement: No AI tools were used in this study.
Supported by Suzhou Clinical Special Disease Diagnosis and Treatment Program, No. LCZX202346; and Junzheng Fund of Soochow University.
Institutional review board statement: The study protocol was approved by the Ethics Committee of The Fourth Affiliated Hospital of Soochow University, No. 251315.
Informed consent statement: The requirement for written informed consent was waived due to the retrospective nature of the study.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
STROBE statement: The authors have read the STROBE Statement-checklist of items, and the manuscript was prepared and revised according to the STROBE Statement-checklist of items.
Data sharing statement: The data supporting the findings of this study are available from the authors upon reasonable request.
Corresponding author: Zhi-Hua Lu, MD, Department of Radiology, The Fourth Affiliated Hospital of Soochow University, Suzhou Dushu Lake Hospital, Medical Center of Soochow University, No. 9 Chongwen Road, Suzhou Industrial Park, Suzhou 215123, Jiangsu Province, China.
luzhihua4100@suda.edu.cn
Received: June 8, 2026
Revised: July 27, 2026
Accepted: September 9, 2026
Published online: December 15, 2026
Processing time: 157 Days and 8.3 Hours
BACKGROUND
Metachronous liver metastasis (MLM) represents a major contributor to disease relapse and poor survival outcomes in patients with colorectal cancer (CRC). The prognostic value of combining intra- and peritumoral radiomics features for predicting MLM has not been fully explored.
AIM
To develop and validate a computed tomography-based intratumoral and peritumoral radiomics model for predicting MLM and a nomogram for predicting 5-year liver metastasis-free survival (LMFS) of CRC patients.
METHODS
This retrospective study included CRC patients who underwent preoperative contrast-enhanced computed tomography.
RESULTS
A total of 203 CRC patients were included (142 training cohort, 61 external validation cohort). Multivariable analysis confirmed N stage, tumor deposits, and pericolic fat infiltration pattern as independent liver metastasis-free survival (LMFS) predictors. The baseline clinical model showed C-indices of 0.766 [95% confidence interval (CI): 0.694-0.838] and 0.777 (95%CI: 0.683-0.872) in training and validation sets, respectively. Notably, the combined intra-peritumoral radiomics model achieved superior area under the curve values (0.898, 95%CI: 0.838-0.957; 0.768, 95%CI: 0.634-0.902) compared with standalone intratumoral or peritumoral models. The radiomics nomogram further improved performance, with C-indices of 0.822 and 0.836 for LMFS prediction in the two cohorts, and effectively stratified patients into high- and low-risk groups with significant 5-year survival differences.
CONCLUSION
The combined radiomics model and nomogram showed potential for MLM prediction, LMFS estimation, and risk stratification, which may help inform personalized postoperative surveillance. Further validation in larger multicenter cohorts is required.
Core Tip: This study is the first to integrate intra- and peritumoral radiomics features to predict colorectal cancer-related metachronous liver metastasis, outperforming single-region radiomics models. The established nomogram provides a novel non-invasive tool for liver metastasis-free survival prognostic evaluation and individualised clinical risk management.
