Introduction:
Radiomics refers to the qualitative and quantitative collection of data from clinical images and clinical details. It involves the methodology used to convert these features to support decision-making. Radiogenomics, a novel computational field, is an emerging discipline within radiomics. It is a combination of “radiology” and “genomics.”
With the increased advancements in new technologies like genome sequencing and artificial intelligence, radiogenomics is emerging as a field of personalized medicine. Radiogenomics integrates a large volume of quantitative data derived from medical images with individual genomic phenotypes. It develops a predictive model through deep learning to classify patients, guide therapeutic strategies, and calculate clinical outcomes. The workflow criteria and internationally agreed guidelines for statistical techniques must be confirmed; radiogenomics represents a cost-effective and repeatable approach for detecting continuous alterations. It is a promising alternative to invasive interventions.
What Is Radiogenomics?
Radiogenomics is a new direction in cancer research that focuses on the relationship between genomics and imaging phenotypes. The advancement in genomics and the far-reaching effects of precision medicine or personalized medicine (a novel approach for disease prevention and treatment that considers individual variability in genes, lifestyle, and environment for every patient). Precision medicine has accelerated interdependent research by carefully integrating the characteristics of patients. Compared with conventional medical treatment protocol, the concept of precision medicine uses a tailored therapeutic plan based on the genotypic and phenotypic data of individual patients.
Cancer is a disease that occurs as a result of genetic abnormalities caused by environmental or hereditary factors. Oncogenesis (formation of cancer) might occur if there is an error in the process of replication and alterations like:
-
Nucleotide substitution.
-
Insertions and deletions of chromosomes.
-
Chromosomal rearrangements.
-
Activation of oncogenes.
-
Loss of tumor suppressor genes.
What Are the Recent Advances in Cancer Research?
Recently, over the past decade, there have been great advancements in understanding the genetic changes involved in oncogenesis. For instance, in conventional clinical procedures, the mutations of access to genomic details are based primarily on the biopsy of focal tissue samples and genetic analysis.
Histopathological examination can be used to differentiate mutational signatures and genomic details. However, these data can only show the condition of a tumor at the time of biopsy. Technological progress in automated DNA (deoxyribonucleic acid) and RNA (ribonucleic acid) sequencing, microarrays, mass spectrometry, and comparative genomic hybridization are important for exploring tumor biomarkers and correct assessment of disease stages in patients.
Today, huge databases are apt for describing the relationship existing between gene expression and clinical features.
When it is used in combination with artificial intelligence, treatment options and survival rates can be predicted by the individual’s performance in models based on large data. Currently, it is possible to perform non-invasive detection and monitoring of diseases repeatedly without any harm, and is regarded as a hotspot in cancer research. The large variation in phenotypes can be demonstrated by non-invasive radiological imaging that shows numerous characteristics of tumors in both qualitative and subjective ways.
Recent advancements in image standardization, acquisition, and image analysis have helped in the identification of objective and precise imaging features, like prognostic and predictive biomarkers. The imaging examinations are often performed repeatedly; however, it is still impractical to obtain dynamic proteomic or genomic data.
This drawback can be solved by proper analysis of computer-processed images to look for underlying predictive and prognostic information. The advent of radiogenomics has caused a shift in the focus of pathology-radiology research from the complete anatomical level to the genetic level. Further, the goal of radiogenomics is to study the correlation between the integrated hierarchical analysis of numerous imaging characteristics and corresponding gene expression profiles. It further helps to identify optimal radiomic biomarkers for more reliable prediction of prognosis and treatment response.
What Is the Technology Behind Radiogenomics?
The huge amount of imaging data collected has led to an inventory of large-scale databases that are both complex and diverse. Therefore, to segregate significant and valuable radiomic data, advanced techniques, frameworks, algorithms, and analytics are required.
The growth and development of radiogenomics depend on high-computing output and machine learning. Both of these are good methods for managing and helping in the analysis of a huge number of variables for varying samples and procedures; for example, the image of a tumor region is usually examined by a radiologist by identifying functional or morphological features. The imaging usually shows more complicated information that can be extracted and processed efficiently using radiogenomic approaches.
What Is the Workflow of Radiogenomics?
The workflow of radiogenomics primarily includes the following:
-
Data acquisition and pre-processing.
-
Tumor segmentation.
-
Feature extraction.
-
Analysis.
-
Modeling.
The standard procedure of operating radiogenomics is basically maintaining the quality of studies to avoid errors, especially in multicenter studies.
A Radiogenomics Consortium (RGC) was established in 2009 to promote and facilitate the multi-center collaboration of researchers working on the linking of genetic variants with radiation therapy. The Radiogenomics Consortium is a Cancer Epidemiology Consortium supported by the Epidemiology and Genetics Research Program run by the National Cancer Institute. The researchers working under RGC have completed various clinical studies that showed genetic variants related to radiation toxicities in patients with breast, prostate, lung, head and neck, and other cancers.
Conclusion:
To conclude, radiogenomics can be summarized as an inevitable outcome of the currently trending precision medicine. A comprehensive radiogenomics approach can show a spatial variation and heterogeneity of voxel (3-dimensional computer graphics) intensities within a tumor and produce predictive and prognostic information. Further, it has a lower cost than conventional genome sequencing, provides complete tumor information, and increased spatial resolution.
Radiogenomics acts as a bridge between the phenotype and genotype of tumors as it applies a mass of automatic extraction of data algorithms and combines them with clinical information. The role of radiogenomics is expected to extend to every aspect of cancer management, ranging from diagnosis to predicting the therapeutic response and risk monitoring. In the future, it is anticipated that the data obtained from imaging examinations could be transformed into quantitative data and interfaced with existing databases to provide diagnostic and prognostic details for supporting clinical decisions.
