Site Editors

Alexander Drilon, MD

Gregory J. Riely, MD, PhD


Applying Radiogenomics in Oncology: Another Step Toward Precision Medicine in the Clinic

By: Vanessa A. Carter, BS
Posted: Friday, January 20, 2023

Li and Zhou, of the First Hospital of China Medical University, Shenyang, China, published their review of cancer-related radiomic and genomics applications in Radiation Oncology. The study, which summarized recent research on radiogenomic applications in solid cancers, suggests that more standard guidelines are needed to normalize radiomics and develop it as a sophisticated field.

“Radiogenomics combines large volumes of radiomic features from medical digital images, genetic data from high-throughput sequencing, and clinical-epidemiological data into mathematical modeling,” the investigators stated. “The amalgamation of radiomics and genomics provides an approach to better study the molecular mechanism of tumor pathogenesis..., make clinical decisions by predicting prognosis, and improve the development of individualized treatment guidance.”

The process of radiogenomics begins with data collection of original medical images from CT, MRI, or PET scans, the authors noted. Images are then segmented to the region of interest, and quantitative radiomics features are extracted and divided into categories such as volume, surface area, shape, and statistical features. Developing a predictive model is the focus of radiomics and often employs machine learning, logistic regression, and Cox proportional hazards. To complete this process, performance assessment and external validation based on other data sets are required to prove the model’s credibility.

For example, a recent study established models to differentiate clear cell from non–clear cell renal carcinomas. By incorporating texture features extracted from CT images, this model significantly improved the predictive efficacy of clear cell renal cell carcinoma, appropriately discriminating non–clear cell from clear cell disease. Another study, which applied machine learning algorithms to predict gene mutations in 207 cases of clear cell renal cell carcinoma, demonstrated excellent performance in detecting mutations of genes SETD2, VHL, BAP1, and PBRM1. Combined, the results of these trials suggest that radiogenomics may be a noninvasive alternative to genetic testing, the study authors concluded.

Disclosure: The study authors reported no conflicts of interest.

By continuing to browse this site you permit us and our partners to place identification cookies on your browser and agree to our use of cookies to identify you for marketing. Read our Privacy Policy to learn more.