Related Experiment Video
Updated: Aug 16, 2026
07:50
Genome-Wide Analysis of DNA Methylation in Gastrointestinal Cancer
Published on: September 18, 2020
Radiomics for Prediction of Microsatellite Instability Status in Gastric Cancer: A Systematic Review and
Qitao Gou1,2, Kangmeng Wang3, Xin He4
1Department of Oncology, Laboratory of Immunity, Inflammation & Cancer, The First Affiliated Hospital of Chongqing Medical University, Chongqing, 400016, China. 357937698@qq.com.
Journal of Gastrointestinal Cancer
|August 14, 2026
Summary
Radiomics shows moderate accuracy for predicting microsatellite instability (MSI) in gastric cancer (GC) before surgery. Combining radiomics with clinical data enhances specificity, improving patient selection for immunotherapy.
Area of Science:
- Oncology
- Medical Imaging
- Biomarkers
Background:
- Microsatellite instability (MSI) is crucial for immunotherapy response in gastric cancer (GC).
- Non-invasive preoperative prediction of MSI in GC is challenging.
- Radiomics shows potential but requires systematic evaluation of diagnostic performance and overfitting.
Purpose of the Study:
- To systematically evaluate the diagnostic performance of radiomics for preoperative MSI prediction in GC.
- To compare different radiomics models and identify factors influencing prediction accuracy.
- To assess the potential of radiomics in guiding patient selection for GC treatment.
Main Methods:
- Systematic literature search of PubMed, Embase, Web of Science, and Cochrane Library.
- Bivariate random-effects model to pool sensitivity, specificity, and diagnostic odds ratio (DOR).
- Subgroup analyses based on model type, data source, validation type, and algorithm.
Main Results:
- Thirteen studies including 2,447 patients were analyzed.
- Pooled AUC in validation sets was 0.82; sensitivity 0.79, specificity 0.72.
- Combined models (radiomics + clinical features) showed higher specificity (0.75) than radiomics alone (0.68).
Conclusions:
- Radiomics offers moderate accuracy for preoperative MSI prediction in GC.
- Combined models improve specificity, aiding in patient selection for immunotherapy.
- Rigorous external validation is essential, and prospective, multicenter studies are recommended.
