radiogenomics

Radiogenomics is an interdisciplinary field that studies the relationship between imaging features and the genomic properties of diseases, most notably cancer. It seeks to identify imaging biomarkers that reflect underlying molecular alterations, enabling non-invasive inference of tumor biology and potentially guiding therapy and prognosis.

Methods in radiogenomics typically rely on radiomics to extract quantitative features from imaging modalities such as MRI, CT, and PET. Features capture aspects of shape, texture, intensity, and spatial relationships. These features are then statistically correlated with genomic data, including gene mutations, expression profiles, methylation patterns, and copy number alterations. Analyses often involve machine learning or multivariate modeling and may incorporate multi-omics data integration. Public datasets combining imaging and genomics, such as The Cancer Genome Atlas and The Cancer Imaging Archive, have supported early discoveries, but external validation remains essential.

Applications are strongest in oncology, where radiogenomics aims to predict molecular subtypes, mutation status, or expression signatures from imaging to support non-invasive molecular profiling, prognosis, and personalized treatment planning. Examples include imaging correlates of MGMT promoter methylation and IDH mutation in gliomas, as well as associations between radiomic phenotypes and EGFR mutations or ALK rearrangements in lung cancer. Studies also explore breast, liver, and other cancers, as well as responses to targeted therapies and immunotherapies.

Limitations include reproducibility and standardization challenges for imaging protocols, variability across scanners and institutions, feature robustness concerns, and the risk of overfitting with small datasets. Robust validation, standardized pipelines, and prospective trials are needed before routine clinical adoption, alongside ethical and privacy considerations for data sharing.

Future directions involve standardized feature definitions, larger multi-center studies, and integration with other omics data to better inform precision medicine. See also radiomics, genomics, and precision medicine.