Predicting NRAS gene status in colorectal cancer using computed tomography (CT)-based radiomics

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• NRAS gene mutations are crucial in colorectal cancer (CRC) development and treatment.
• Accurate prediction of NRAS status is vital for personalized CRC therapy.
• Current methods for NRAS status determination can be invasive or time-consuming.

Purpose of the Study:

• To develop and validate a radiomics model using computed tomography (CT) images for predicting NRAS gene status in colorectal cancer (CRC) patients.
• To assess the efficacy of radiomics features extracted from CT scans in identifying NRAS mutation status.
• To establish a non-invasive tool for NRAS status prediction in CRC.

Main Methods:

• A cohort of 216 colorectal cancer patients from hospital A was analyzed.
• Radiomic features were extracted from enhanced venous phase CT images and analyzed using logistic regression.
• Model performance was validated using five-fold cross-validation, Hosmer-Lemeshow tests, and clinical decision curve analysis.

Main Results:

• No significant association was found between clinical features and NRAS gene status.
• The radiomics model achieved an area under the curve (AUC) of 0.803 in the training set and 0.766 in the test set.
• Excellent agreement between predicted and actual probabilities was observed, demonstrating the model's calibration and clinical utility.

Conclusions:

• A CT-based radiomics model demonstrates significant potential for effectively predicting NRAS gene status in colorectal cancer patients.
• This non-invasive radiomics approach offers a promising tool for personalized medicine in CRC management.
• Further validation in larger, diverse cohorts is warranted to solidify its clinical application.