Related Experiment Video
Updated: Feb 20, 2026
11:27
Modeling Spontaneous Metastatic Renal Cell Carcinoma mRCC in Mice Following Nephrectomy
Published on: April 29, 2014
17.2K
MRI-Based Habitat Radiomics for Predicting Postoperative Recurrence in Nonmetastatic Clear Cell Renal Cell Carcinoma
Sicheng Yi1, Tongyu Jia2, Xu Bai3
1Department of Radiology, First Medical Center of the Chinese PLA General Hospital, 28 Fuxing Road, Haidian District, Beijing 100853, China (S.Y., H.K., B.L., C.L., X.W., H.X., X.N., M.C., S.Z., Y.M., H.W.).
Academic Radiology
|February 18, 2026
Summary
This study developed an MRI-based radiomics model to predict recurrence in clear cell renal cell carcinoma (ccRCC) patients. The combined model accurately predicts postoperative recurrence, aiding in personalized treatment strategies for ccRCC.
Area of Science:
- Oncology
- Radiology
- Medical Imaging
Background:
- Clear cell renal cell carcinoma (ccRCC) is the most common subtype of kidney cancer.
- Predicting recurrence-free survival (RFS) after surgical resection is crucial for managing nonmetastatic ccRCC.
- Current prediction methods may benefit from advanced imaging analysis.
Purpose of the Study:
- To develop and validate a magnetic resonance imaging (MRI)-based radiomics model.
- To predict recurrence-free survival (RFS) in patients with nonmetastatic ccRCC post-surgery.
- To integrate radiomics features with clinicopathological data for enhanced prediction.
Main Methods:
- Retrospective analysis of 630 nonmetastatic ccRCC patients.
- Utilized preoperative T2WI and contrast-enhanced corticomedullary phase (CP) MRI.
- Applied K-means clustering for tumor voxel segmentation into homogeneous habitats.
- Extracted radiomic features, selected recurrence-related features, and built a Cox regression model integrating clinicopathological indicators.
Main Results:
- Identified three distinct tumor habitat regions and 13 recurrence-related radiomic features, forming a Habitat Signature (HS).
- Age, male sex, and advanced pathological T stage were independent predictors of recurrence.
- The combined clinical-habitat model achieved high predictive performance (AUCs 0.80-0.85, C-indices 0.80-0.81) for 3- and 5-year recurrence.
Conclusions:
- The developed MRI-based Habitat Signature (HS) combined with clinical-pathological features demonstrates significant predictive value.
- This model can aid in predicting postoperative recurrence in nonmetastatic ccRCC patients.
- The findings support the use of radiomics in personalized cancer management.
