A disease capable of decimating and killing those affected, cancer involves cells in a specific part of the body growing and reproducing uncontrollably in a process known as proliferation. In a recent breakthrough, the tumor microenvironment (TME) has been established as a key driver for cancer progression, promoting resistance to therapeutics all the while enabling the disease to evade the immune system.
Specifically, myeloid-derived suppressor cells (MDSCs) have been shown to play a central role in maintaining the TME through the suppression of host immunity, the establishment of new vasculature, and the remodeling of connective tissue to support tumor growth.
Therefore, it is imperative to develop cancer immunotherapies able to promote the anti-oncological activity of the immune system with the dual ability to combat the highly detrimental effects of the TME. Unfortuately, while it is straightforward to assess the effect of new therapies on cancer cells, estimating the effectiveness of these novel therapies on the TME is challenging.
Tracking unseen cancer changes
Radiomics is a method possessing the unique ability to extract large amounts of features from medical images using artificial intelligence (AI), revealing tell-tell patterns and/or textures in the tumor. It does this by employing algorithms to extract hidden quantitative features from radiology images to make predictions on unseen data sets using big data, providing valuable information for personalized therapy.
Now, a study from researchers led by the Baylor College of Medicine develops a new approach dubbed AI-based nano-radiomics capable of assessing changes in the TME that cannot be detected using conventional imaging methods. The team states their new technique is a major step towards noninvasive imaging methods for the overall evaluation of cancer immunotherapies designed to debilitate the TME. The opensource study is published in the journal of Science Advances.
Inside a tumor microenvironment
Previous studies have indicated the analysis of how the TME responds to anti-cancer therapy is paramount, especially when the TME is inhibiting the anti-tumor effects of cancer immunotherapies. Unfortunately, imaging technologies such as computed tomography (CT) or magnetic resonance imaging provides information about the overall tumor response to therapy, but provides very little, if any, information about the TME.
This, married with the fact, immunotherapies targeting the TME result in variable and delayed responses, mean it is extremely difficult to gauge the early-stage efficacy of these therapies.
Therefore, a non-invasive method to assess the effect of cancer immunotherapies targetting the TME is desperately needed. The current study develops nano-radiomics, which uses AI and a nanoparticle contrast agent to computational mine for 3-D imaging data to evaluate a cancer-cell immunotherapy inhibiting MDSC in the TME.
Evaluating treatments in the TME
The current study uses mouse models of cancer, consisting of human solid tumors infiltrated by MDSC which were either treated with natural killer (NK) cells or a placebo. Both models were then imaged using both nano-based CT imaging and AI-based nano-radiomics to compare methods.
As MDSCs play a central role in tumor angiogenesis and reside in perivascular niches within the TME, it was hypothesized depletion of these immune cells could possibly alter tumor vascular architecture and, therefore, tumor texture.
Results show nano-radiomic analysis of both groups reveals 3D texture-based features within the TME distinguishing the cancer-positive and placebo groups while traditional CT scans did not. Data findings show these previously unseen features suggest immunotherapy-driven MDSC depletion results in a high degree of tumor texture heterogeneity.
The lab explains whereas conventional CT scans were unable to differentiate NK cell immunotherapy tumors from untreated ones, nano-radiomics revealed texture-based features capable of differentiating both treatment groups.
They go on to add it was demonstrated TME directed cellular immunotherapy causes subtle changes not effectively gauged by conventional imaging metrics that were revealed by nano-radiomics. They conclude their novel individualized technique provides a method for noninvasive assessment of TME-directed immunotherapy potentially applicable to numerous solid tumors.
New imaging system for the TME
The team surmises they have successfully demonstrated a method for noninvasive assessment of TME-directed immunotherapy using AI-based nano-radiomics. For the future, the researchers state this novel approach has the potential to enhance the ability of clinicians to noninvasively assess the effect of treatments directed at the TME, ultimately enhancing the impact of cancer treatment and management.
Source: Baylor College of Medicine
Don’t miss the latest discoveries from the health innovator community: