Systematic Review on the Association of Radiomics with Tumor Biological Endpoints

Abstract

Radiomics supposes an alternative non-invasive tumor characterization tool, which has experienced increased interest with the advent of more powerful computers and more sophisticated machine learning algorithms. Nonetheless, the incorporation of radiomics in cancer clinical-decision support systems still necessitates a thorough analysis of its relationship with tumor biology. Herein, we present a systematic review focusing on the clinical evidence of radiomics as a surrogate method for tumor molecular profile characterization. An extensive literature review was conducted in PubMed, including papers on radiomics and a selected set of clinically relevant and commonly used tumor molecular markers. We summarized our findings based on different cancer entities, additionally evaluating the effect of different modalities for the prediction of biomarkers at each tumor site. Results suggest the existence of an association between the studied biomarkers and radiomics from different modalities and different tumor sites, even though a larger number of multi-center studies are required to further validate the reported outcomes.

Utilization of Radiomics Features Extracted From Preoperative Medical Images to Detect Metastatic Lymph Nodes in Cholangiocarcinoma and Gallbladder Cancer Patients: A Systemic Review and Meta-analysis

“…This is of particular importance, as, although numerous original contributions and meta-analyses have shown the necessity of combining radiomics models with other histologic, pathologic, immunologic, or other biomarkers, most of the currently available pieces of evidence do not include such markers in the development of their predictive models. 12 Although some methodologic aspects of the studies were problematic, there was a consistent adaptation of some others, including the implementation of feature selection (and reduction) algorithms, utilization of open source software for 3-dimensional segmentation and extraction of radiomics features, and utilization of optimized classifiers for making predictions. Hypothetically, manual segmentation could contribute to less reproducibility 42 ; thus, 6 of the 9 studies used multiple radiologists for the segmentation of lesions and determined the interreader reliability rate of features extracted from different contours, although none of the studies went as far as determining Dice similarity coefficient of the contours or considering the extraction of features from the intersection of 2 contours made by different readers.…”

“…Contrast-enhanced computed tomography (CT) imaging is the primary modality in the staging of biliary cancers and determining the appropriateness of surgical resectability; however, the sensitivity of detecting metastatic lymph nodes is potentially only 28% to 54%. [11][12][13] Magnetic resonance imaging of the abdomen is also shown to not consistently detect lymph nodes in some subvariants of cholangiocarcinoma. Overall, the current standard of care is not sufficiently sensitive in detecting lymph node metastasis in preoperative images.…”

“…However, there are considerable limitations in regard to detecting preoperative lymph node metastasis in biliary cancers. Contrast-enhanced computed tomography (CT) imaging is the primary modality in the staging of biliary cancers and determining the appropriateness of surgical resectability; however, the sensitivity of detecting metastatic lymph nodes is potentially only 28% to 54% 11–13 . Magnetic resonance imaging of the abdomen is also shown to not consistently detect lymph nodes in some subvariants of cholangiocarcinoma.…”

Multi-Omics Approaches for the Prediction of Clinical Endpoints after Immunotherapy in Non-Small Cell Lung Cancer: A Comprehensive Review

“…Several radiomics models were able to predict the level of expression of PD-L1 with reasonable performances [ 84 ]. For instance, using a small retrospective overall cohort of 72 patients and combining two features extracted from pre-treatment CTs, the model reached an AUC of 0.79 for the prediction of PD-L1 values ≥ 50% in the validation cohort [ 85 ].…”

Radiomics as an emerging tool in the management of brain metastases

“…For instance, prediction models can often accurately determine factors such as tumor mutation status and proliferative index (KI67 immunohistochemistry status). 48 Radiomic and quantitative image analysis have become increasingly relevant tools in neuro-oncology. Imaging is continually a part of patient care, and radiomic analyses exploit this existing data to provide further noninvasive predictions pertaining to patient survival, mutation status, and other factors that can help stratify patients into a novel and more personalized treatment plans ( Figure 2 ).…”