Leveraging radiomics to predict glioblastoma multiforme microenvironment and personalise treatment planning

University of Edinburgh

nearmejobs.eu

Glioblastoma multiforme (GBM) is the most common primary brain tumour, with a dismal median survival. The tumour microenvironment (TME) is the microscopic-scale environment of tissue in which a tumour evolves and is a key component that controls GBM aggressiveness. The TME varies from patient to patient, but also varies spatially within tumours, and the heterogeneity of the TME is a barrier to therapeutic treatments as treatment response is dependent on it. For example, drug delivery to the tumour tissue or immune infiltration are both regulated by the TME. Precision medicine, the tailoring of therapies to individual patients, could be a solution by adapting the treatment plan of a patient to the specificities of their own TME. However, precision medicine is dependent on accurate biomarkers indicating which treatment is most suitable. Biomarkers obtained through tumour biopsies, a sample of a tumour, do not give a full picture of the TME due to the limited size of the sample, while clinical imaging modalities image the entire tumour, but currently do not directly measure the numerous aspects of the TME.

Radiomics is a more novel approach consisting of extracting quantitative features from clinical medical imaging modalities [1], such as positron emission tomography (PET), fluoromisonidazole-PET (FMISO-PET), computed tomography (CT) or magnetic resonance imaging (MRI), and using these features for prognosis or treatment planning. Recently, radiomic features from clinical imaging modalities have been shown to correlate with TME features, obtained from histology, such as hypoxic fraction, vascular density [2], or gene status [3]. These works have demonstrated the potential of radiomics that has yet to be explored to non-invasively predict TME features. As the features of the TME can predict treatment response, radiomics is increasingly seen as a tool to obtain non-invasive and clinically available biomarkers for precision medicine.

The supervisory team has experience in pre-clinical models [4, 5], medical imaging [6, 7] and the tumour microenvironment [4, 5, 8], and will leverage that expertise to use radiomics to predict GBM TME features from medical images, MRI and FMISO-PET, in a pre-clinical setting. In this project, we propose to build on previous studies [2, 3] and hypothesise that radiomics can be leveraged to predict biomarkers at a cellular level, obtained from spatial transcriptomics, to phenotype the TME and predict features such as immune infiltration. The ability to correlate radiomic features to cellular-level TME phenotypes that inform treatment response will progress treatment planning and precision medicine.  

Supervisory Team: Prof Miguel O. Bernabeu, Prof Val Brunton and Prof Adriana Tavares

Application procedure

Please provide a CV, a personal statement detailing your research interests and reasons for applying, degree certificate(s), and 2 written academic references. All documents should be in electronic format and sent via e-mail to: .

Please direct informal enquiries to Miguel Bernabeu – email:

The closing date for applications is: 15 October 2024

Interviews will be held during November 2024 for January 2025 start date, preferably.

To help us track our recruitment effort, please indicate in your email – cover/motivation letter where (nearmejobs.eu) you saw this posting.

Job Location