Preoperative Noninvasive Prediction of Tertiary Lymphoid Structures in Rectal Cancer Using MRI Radiomics and
Fan Zhang1, Rong Li1, Ling Wang2
1Gansu University of Chinese Medicine, Lanzhou, China (F.Z., F.M., F.L., L.P., X.Z.).
Academic Radiology
|July 24, 2026
Summary
This study developed a noninvasive radiomics-clinical model to predict tertiary lymphoid structures (TLSs) status in rectal cancer patients before surgery. The model accurately predicts TLSs status and patient survival, aiding immunotherapy decisions.
Area of Science:
- Oncology
- Radiology
- Medical Imaging
Background:
- Noninvasive assessment of tertiary lymphoid structures (TLSs) is vital for rectal cancer immunotherapy.
- Current methods for TLSs assessment are invasive.
- Accurate preoperative prediction of TLSs status is needed.
Purpose of the Study:
- To develop and validate an interpretable model for noninvasive preoperative prediction of TLSs status in rectal cancer.
- To assess the model's performance in predicting overall survival.
Main Methods:
- Retrospective study of 177 rectal cancer patients.
- Development of radiomics, clinical, and combined models using preoperative MRI.
- Feature selection via ICC, RFE, and LASSO.
- Model interpretation using SHAP and LIME.
- Performance evaluation using ROC, calibration, DCA, NRI, and IDI analyses.
Main Results:
- The combined model achieved AUCs of 0.866 (training) and 0.851 (testing), outperforming single-modality models.
- NLR and age were identified as independent TLSs predictors.
- The RadScore was the primary positive contributor, while NLR and age had negative contributions.
- High-score patients demonstrated significantly longer overall survival.
Conclusions:
- An interpretable combined radiomics-clinical model enables accurate noninvasive preoperative prediction of TLSs status in rectal cancer.
- This model aids in prognostic stratification for rectal cancer patients.
- The findings support the use of this model for guiding immunotherapy decisions.