Radiomics model based on multi-sequence MR images for predicting preoperative immunoscore in rectal cancer
Kaiming Xue1, Lin Liu1, Yunxia Liu1
1Department of Radiology, China-Japan Union Hospital of Jilin University, NO. 126 Xiantai Street, Changchun, 130033, China.
La Radiologia Medica
|July 13, 2022
Summary
This study developed a radiomics model using MRI scans to predict the immunoscore in rectal cancer patients before surgery. The model shows potential as a non-invasive tool for prognosis and personalized immunotherapy guidance.
Area of Science:
- Radiology
- Oncology
- Medical Imaging
Background:
- Rectal cancer prognosis and treatment depend on the tumor microenvironment, including immune cell infiltration.
- Accurate preoperative assessment of the tumor immune microenvironment, such as immunoscore, is crucial for guiding treatment strategies.
- Current methods for assessing immunoscore often require invasive biopsies, highlighting the need for non-invasive predictive tools.
Purpose of the Study:
- To develop and validate a radiomics model using multi-sequence magnetic resonance (MR) images for the preoperative prediction of immunoscore in rectal cancer.
- To assess the model's performance in distinguishing between different immunoscore levels.
- To evaluate the clinical utility of the radiomics model for non-invasive immunoscore assessment.
Main Methods:
- A retrospective study included 133 rectal cancer patients who underwent preoperative MR imaging.
- Radiomics features were extracted from T2-weighted images (T2WI) and apparent diffusion coefficient (ADC) images.
- A radiomics model (Integrated model A) was built using logistic regression and validated on independent cohorts, assessing performance with ROC curves, calibration curves, and DCA.
Main Results:
- Integrated model A, utilizing both T2WI and ADC images, demonstrated good predictive performance with an AUC of 0.770 in the training cohort and 0.768 in the validation cohort.
- The model's calibration curve indicated good agreement between predicted and actual outcomes.
- Decision curve analysis confirmed the clinical usefulness and robustness of the integrated model.
Conclusions:
- The developed radiomics model based on multi-sequence MR images shows potential as a non-invasive tool for preoperative immunoscore prediction in rectal cancer.
- This tool may aid in evaluating patient prognosis and guiding individualized immunotherapy strategies.
- Further validation in larger, diverse cohorts is warranted to confirm its clinical applicability.
