Use of Radiomics in Characterizing Tumor Hypoxia

Abstract

Tumor hypoxia involves limited oxygen supply within the tumor microenvironment and is closely associated with aggressiveness, metastasis, and resistance to common cancer treatment modalities such as chemotherapy and radiotherapy. Traditional methodologies for hypoxia assessment, such as the use of invasive probes and clinical biomarkers, are generally not very suitable for routine clinical applications. Radiomics provides a non-invasive approach to hypoxia assessment by extracting quantitative features from medical images. Thus, radiomics is important in diagnosis and the formulation of a treatment strategy for tumor hypoxia. This article discusses the various imaging techniques used for the assessment of tumor hypoxia including magnetic resonance imaging (MRI), positron emission tomography (PET), and computed tomography (CT). It introduces the use of radiomics with machine learning and deep learning for extracting quantitative features, along with its possible clinical use in hypoxic tumors. This article further summarizes the key challenges hindering the clinical translation of radiomics, including the lack of imaging standardization and the limited availability of hypoxia-labeled datasets. It also highlights the potential of integrating radiomics with multi-omics to enhance hypoxia visualization and guide personalized cancer treatment.

Keywords: deep learning; machine learning; medical imaging; non-invasive assessment; radiomics; tumor hypoxia.

Figure 1

The distribution of the reviewed literature by research theme. This bar chart summarizes the distribution of reviewed articles according to five major research themes: tumor hypoxia mechanisms (38 studies, 31.1%), radiomics and imaging genomics (32 studies, 26.2%), therapy resistance and reversal strategies (24 studies, 19.7%), imaging technology advances (18 studies, 14.8%), and clinical translation/trials (10 studies, 8.2%).

Figure 2

Schematic illustration of hypoxia detection approaches, encompassing in vitro and in vivo strategies. Detection methods include endogenous markers, exogenous markers, direct pO₂ measurements, and imaging modalities such as PET and MRI [36,43,51].

Figure 3

The workflow of radiomics-based tumor hypoxia characterization. The process includes image acquisition (CT, MRI, PET), preprocessing (resampling, normalization, denoising), segmentation (manual, semi-automatic, or automatic), feature extraction (e.g., texture, shape, intensity), feature selection (e.g., LASSO, mRMR), model development (e.g., machine learning, deep learning), and clinical model validation.

Figure 4

Representative MRI and FMISO-PET images of a recurrent glioma patient before and after bevacizumab treatment. Gadolinium-enhanced MRI and FLAIR MRI show morphological changes, while FMISO-PET highlights hypoxic regions with decreased tracer uptake after treatment. Adapted from Hanley R et al. [55].