Radiomic biomarkers of tumor immune biology and immunotherapy response

Abstract

Immunotherapies are leading to improved outcomes for many cancers, including those with devastating prognoses. As therapies like immune checkpoint inhibitors (ICI) become a mainstay in treatment regimens, many concurrent challenges have arisen -for instance, delineating clinical responders from non-responders. Predicting response has proven to be difficult given a lack of consistent and accurate biomarkers, heterogeneity of the tumor microenvironment (TME), and a poor understanding of resistance mechanisms. For the most part, imaging data have remained an untapped, yet abundant, resource to address these challenges. In recent years, quantitative image analyses have highlighted the utility of medical imaging in predicting tumor phenotypes, prognosis, and therapeutic response. These studies have been fueled by an explosion of resources in high-throughput mining of image features (i.e. radiomics) and artificial intelligence. In this review, we highlight current progress in radiomics to understand tumor immune biology and predict clinical responses to immunotherapies. We also discuss limitations in these studies and future directions for the field, particularly if high-dimensional imaging data are to play a larger role in precision medicine.

Additional Value of PET Radiomic Features for the Initial Staging of Prostate Cancer: A Systematic Review from the Literature

“…In recent years, many studies have been focused on the role of the tumor microenvironment (TME), which consists of tumor cells, tumor stromal cells, and immune cells, as well as the non-cellular components of the extracellular matrix such as collagen [ 33 ]. Understanding the interaction between cancer cells can be used to develop therapeutic strategies to predict and neutralize tactics deployed by cancer cells to survive and resist anti-cancer modalities [ 34 ]; in this scenario, the application of radiomic analysis to the tissue surrounding a tumor might provide further insight into the outcome prediction, as demonstrated by Tu [ 19 ] in PCa patients, or the response to treatment, as proved in other type of cancers [ 35 , 36 ].…”

Artificial Intelligence-based Radiomics in the Era of Immuno-oncology

“…However, applying radiomics alone will not suffice as feature reproducibility remains a major concern. 128 Radiomic features vary widely across different tumor types, imaging modalities, and even institutions, each with its own contrast enhancement protocols that may affect the radiological texture. Selected features across existing studies do not usually overlap owing to inconsistent feature definitions, extraction, interpretations, and calculations.…”

Immunotherapeutic Approaches for Glioblastoma Treatment

“…Some studies suggest that patients on immune checkpoint inhibitors (ICIs) or other targeted immunotherapies have been shown to develop radiation necrosis following SRS at a much higher rate than patients on chemotherapy or no systemic therapy [ 88 , 89 ]. Additionally, ongoing investigations are attempting to characterize tumor immune states by radiomic features, including sphericity, heterogeneity, and sharpness of borders with predicted response to ICIs [ 90 ]; and composite models using AI and deep learning to combine radiomic models with other variables, including clinical and molecular features, are promising. Immune-cell trafficking, as described above, is a critical issue for brain-tumor immunotherapy.…”