Image Analysis as tool for Predicting Colorectal Cancer Molecular Alterations: A Scoping Review

Area of Science:

• Radiogenomics
• Medical Imaging
• Oncology

Background:

• Molecular alterations are key diagnostic indicators for colorectal cancer.
• Current methods for measuring molecular alterations are invasive and costly.
• There is a need for non-invasive methods to predict these alterations.

Purpose of the Study:

• To investigate the application of image processing and radiogenomics for predicting molecular alterations in colorectal cancer.
• To review existing literature on the use of imaging techniques to assess molecular markers in colorectal cancer.

Main Methods:

• A scoping review was conducted using Web of Science, Scopus, and PubMed.
• Studies were selected and reported following PRISMA guidelines.
• Study quality was assessed using the STROBE checklist.

Main Results:

• 41 high-quality studies were reviewed, focusing on predicting KRAS, NRAS, BRAF, TP53, and MSI.
• CT, MRI, and 18F-FDG PET/CT combined with radiomics and quantitative analysis were utilized.
• Significant relationships were found between imaging features and molecular alterations in 39 studies, with average AUCs for predicting KRAS, MSI, BRAF, and TP53 at 78%, 81%, 80%, and 71% respectively.

Conclusions:

• Radiogenomics holds significant potential for non-invasively predicting colorectal cancer molecular alterations.
• Imaging modalities like CT, MRI, and PET/CT, coupled with radiomics and texture analysis, can accurately predict key molecular markers.
• This approach offers a promising avenue for enhancing diagnostic accuracy and patient management in colorectal cancer.