Though there is still much work to be done, experts believe combining radiomic data with genomic data could one day personalize treatment plans for lung cancer patients.
Combining radiomic and genomic data, known as radiogenomics, is achieved by extracting patients' imaging information and tumor biology data via artificial intelligence. In doing so, researchers can identify imaging biomarkers that may reflect potential mutations.
“In lung cancer, recent advances in whole-genome sequencing and the identification of actionable molecular alterations have led to an increased interest in understanding the complex relationships between imaging and genomic data, with the potential of guiding therapeutic strategies and predicting clinical outcomes,” corresponding author Maria Mayoral, with the Department of Radiology at Memorial Sloan Kettering Cancer Center, and co-authors explained.
The experts noted that more research regarding the implementation of radiogenomics is necessary using the technique in clinical practice. But recently in Clinical Imaging, they highlighted many of its potential benefits, particularly for personalizing cancer care.
Radiogenomics could open many doors for creating targeted lung cancer therapies, as the technique can identify unique tumor mutations. Typically, patients undergo biopsy or surgery to obtain samples of tissue for this, but radiogenomics can noninvasively predict mutation status with comparable results. And it does this with data from the entire tumor, rather than a small tissue sample.
The potential benefits also carry over treatment planning. Not only could radiogenomics be utilized to predict mutation status, but it could help providers understand how certain mutations may respond to various therapeutics.
“This information could also be integrated in more complex models to predict treatment response in order to select those patients who would be more likely to benefit from genotype-directed therapy, in addition to stratifying patients according to survival metrics to predict disease prognosis,” the experts wrote.
However, they added that achieving clinical value with radiogenomics does not come without its challenges. First, more multi-center studies that employ automatic segmentation, rather than manual, are needed. They also point out that more robust data with detailed endpoints are necessary to appropriately train deep learning algorithms, but current regulations inhibit researchers from constructing such large datasets.
“The standardization and improved robustness of radiogenomics models are still needed for this approach to be effectively integrated into the clinical practice,” the experts concluded.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She joined Innovate Healthcare in 2021 and has since put her unique expertise to use in her editorial role with Health Imaging.