Related Experiment Video
Updated: Aug 5, 2026
Competing-Risk Nomogram for Predicting Cancer-Specific Survival in Multiple Primary Colorectal Cancer Patients after Surgery
Published on: September 27, 2024
Non-contrast MRI Radiomics Nomogram for Predicting Recurrence and Progression-Free Survival in Cervical Cancer
Wei Wang1, Xiaoting Li2, Yueqi Jiang3
1Department of Gynecologic Oncology, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), Peking University Cancer Hospital & Institute, Hai Dian District, Beijing 100142, China.
This study developed a non-contrast MRI radiomics nomogram to predict cervical cancer recurrence risk and progression-free survival (PFS) after chemoradiotherapy. The model offers accessible, personalized risk stratification without contrast agents.
Area of Science:
- Oncology
- Radiology
- Medical Imaging
Background:
- Accurate prognosis for cervical cancer patients undergoing concurrent chemoradiotherapy (CCRT) is challenging.
- Contrast-enhanced MRI (CE-MRI) is limited by renal dysfunction and resource constraints.
- Development of non-contrast MRI-based models is crucial for broader applicability.
Purpose of the Study:
- To develop and validate a non-contrast MRI radiomics nomogram for predicting recurrence risk and 5-year progression-free survival (PFS) in cervical cancer patients.
- To integrate radiomics features with clinical predictors for enhanced prognostic accuracy.
- To provide a clinically applicable and accessible tool for personalized risk stratification.
Main Methods:
- Retrospective analysis of 145 cervical cancer patients from two centers.
- Extraction of radiomics features from pre-treatment non-contrast MRI.
- Development of an extreme gradient boosting classifier and integration with clinical predictors into a nomogram.
- External validation in an independent cohort of 31 patients.
Main Results:
- The nomogram effectively stratified patients into distinct low- and high-risk groups.
- High-risk patients demonstrated significantly worse 5-year PFS in both construction and validation cohorts (P < 0.001 and P=0.012, respectively).
- Prognostic performance was comparable to CE-MRI-based models, without the need for gadolinium contrast.
Conclusions:
- A non-contrast MRI radiomics nomogram provides robust prognostic information for cervical cancer patients post-CCRT.
- Combining imaging features and clinical data improves predictive accuracy.
- This approach enhances accessibility for patients with contraindications to contrast agents or in low-resource settings.
